{ "best_metric": 2.20203996, "best_model_checkpoint": "/scratch/ms-swift-chatas/exp_output_paligemma_imgchat/v3-20250529-040720/checkpoint-105000", "epoch": 4.498521914228182, "eval_steps": 500, "global_step": 105000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 4.284306584979221e-05, "grad_norm": 24.429079055786133, "learning_rate": 9.999999998188407e-05, "loss": 4.934672832489014, "memory(GiB)": 14.51, "step": 1, "token_acc": 0.19298245614035087, "train_speed(iter/s)": 0.332266 }, { "epoch": 0.00021421532924896104, "grad_norm": 21.24069595336914, "learning_rate": 9.999999954710156e-05, "loss": 4.759311676025391, "memory(GiB)": 20.53, "step": 5, "token_acc": 0.2627450980392157, "train_speed(iter/s)": 0.918491 }, { "epoch": 0.0004284306584979221, "grad_norm": 9.067500114440918, "learning_rate": 9.99999981884062e-05, "loss": 3.620919036865234, "memory(GiB)": 29.98, "step": 10, "token_acc": 0.34177215189873417, "train_speed(iter/s)": 1.127471 }, { "epoch": 0.0006426459877468832, "grad_norm": 23.622352600097656, "learning_rate": 9.999999592391398e-05, "loss": 2.6890094757080076, "memory(GiB)": 29.98, "step": 15, "token_acc": 0.42168674698795183, "train_speed(iter/s)": 1.249265 }, { "epoch": 0.0008568613169958442, "grad_norm": 17.0115909576416, "learning_rate": 9.999999275362494e-05, "loss": 2.420244598388672, "memory(GiB)": 29.98, "step": 20, "token_acc": 0.45454545454545453, "train_speed(iter/s)": 1.309945 }, { "epoch": 0.0010710766462448053, "grad_norm": 16.966550827026367, "learning_rate": 9.999998867753912e-05, "loss": 2.4537260055541994, "memory(GiB)": 32.43, "step": 25, "token_acc": 0.4560810810810811, "train_speed(iter/s)": 1.352727 }, { "epoch": 0.0012852919754937663, "grad_norm": 25.70478057861328, "learning_rate": 9.999998369565659e-05, "loss": 1.8708484649658204, "memory(GiB)": 37.5, "step": 30, "token_acc": 0.516320474777448, "train_speed(iter/s)": 1.392101 }, { "epoch": 0.0014995073047427273, "grad_norm": 20.09503173828125, "learning_rate": 9.999997780797748e-05, "loss": 2.0371498107910155, "memory(GiB)": 37.5, "step": 35, "token_acc": 0.5399239543726235, "train_speed(iter/s)": 1.420257 }, { "epoch": 0.0017137226339916883, "grad_norm": 12.851018905639648, "learning_rate": 9.999997101450185e-05, "loss": 1.8779441833496093, "memory(GiB)": 37.5, "step": 40, "token_acc": 0.5800711743772242, "train_speed(iter/s)": 1.437215 }, { "epoch": 0.0019279379632406496, "grad_norm": 13.68235969543457, "learning_rate": 9.999996331522983e-05, "loss": 1.577826690673828, "memory(GiB)": 37.5, "step": 45, "token_acc": 0.6123188405797102, "train_speed(iter/s)": 1.494192 }, { "epoch": 0.0021421532924896106, "grad_norm": 17.75782585144043, "learning_rate": 9.99999547101616e-05, "loss": 1.5587465286254882, "memory(GiB)": 37.5, "step": 50, "token_acc": 0.6147540983606558, "train_speed(iter/s)": 1.505993 }, { "epoch": 0.002356368621738572, "grad_norm": 17.360958099365234, "learning_rate": 9.999994519929725e-05, "loss": 1.8693782806396484, "memory(GiB)": 37.5, "step": 55, "token_acc": 0.564935064935065, "train_speed(iter/s)": 1.541611 }, { "epoch": 0.0025705839509875326, "grad_norm": 13.06774616241455, "learning_rate": 9.9999934782637e-05, "loss": 1.2743507385253907, "memory(GiB)": 37.5, "step": 60, "token_acc": 0.6413043478260869, "train_speed(iter/s)": 1.539487 }, { "epoch": 0.002784799280236494, "grad_norm": 13.181123733520508, "learning_rate": 9.999992346018105e-05, "loss": 1.2007725715637207, "memory(GiB)": 37.5, "step": 65, "token_acc": 0.6468401486988847, "train_speed(iter/s)": 1.544209 }, { "epoch": 0.0029990146094854547, "grad_norm": 16.61612892150879, "learning_rate": 9.999991123192957e-05, "loss": 1.581649112701416, "memory(GiB)": 37.5, "step": 70, "token_acc": 0.6384364820846905, "train_speed(iter/s)": 1.572008 }, { "epoch": 0.003213229938734416, "grad_norm": 9.395951271057129, "learning_rate": 9.99998980978828e-05, "loss": 1.1370491981506348, "memory(GiB)": 37.5, "step": 75, "token_acc": 0.7173252279635258, "train_speed(iter/s)": 1.569566 }, { "epoch": 0.0034274452679833767, "grad_norm": 18.296876907348633, "learning_rate": 9.999988405804095e-05, "loss": 1.1459007263183594, "memory(GiB)": 37.5, "step": 80, "token_acc": 0.6875, "train_speed(iter/s)": 1.57181 }, { "epoch": 0.003641660597232338, "grad_norm": 14.201404571533203, "learning_rate": 9.999986911240431e-05, "loss": 0.8491975784301757, "memory(GiB)": 37.5, "step": 85, "token_acc": 0.7830508474576271, "train_speed(iter/s)": 1.569464 }, { "epoch": 0.003855875926481299, "grad_norm": 18.60801124572754, "learning_rate": 9.999985326097314e-05, "loss": 0.8592746734619141, "memory(GiB)": 37.5, "step": 90, "token_acc": 0.8047945205479452, "train_speed(iter/s)": 1.571062 }, { "epoch": 0.00407009125573026, "grad_norm": 12.978971481323242, "learning_rate": 9.999983650374773e-05, "loss": 0.620181941986084, "memory(GiB)": 37.5, "step": 95, "token_acc": 0.84375, "train_speed(iter/s)": 1.571153 }, { "epoch": 0.004284306584979221, "grad_norm": 11.887883186340332, "learning_rate": 9.999981884072838e-05, "loss": 0.9150229454040527, "memory(GiB)": 37.5, "step": 100, "token_acc": 0.7623456790123457, "train_speed(iter/s)": 1.573168 }, { "epoch": 0.004498521914228182, "grad_norm": 9.54543685913086, "learning_rate": 9.999980027191539e-05, "loss": 0.6089622497558593, "memory(GiB)": 37.5, "step": 105, "token_acc": 0.8269896193771626, "train_speed(iter/s)": 1.575867 }, { "epoch": 0.004712737243477144, "grad_norm": 8.852176666259766, "learning_rate": 9.999978079730912e-05, "loss": 0.77999906539917, "memory(GiB)": 37.5, "step": 110, "token_acc": 0.8046875, "train_speed(iter/s)": 1.574787 }, { "epoch": 0.0049269525727261045, "grad_norm": 10.61286735534668, "learning_rate": 9.999976041690993e-05, "loss": 0.4582165241241455, "memory(GiB)": 37.5, "step": 115, "token_acc": 0.87, "train_speed(iter/s)": 1.574832 }, { "epoch": 0.005141167901975065, "grad_norm": 14.49724006652832, "learning_rate": 9.999973913071817e-05, "loss": 0.7001579761505127, "memory(GiB)": 37.5, "step": 120, "token_acc": 0.8417721518987342, "train_speed(iter/s)": 1.57417 }, { "epoch": 0.005355383231224026, "grad_norm": 5.007709980010986, "learning_rate": 9.999971693873423e-05, "loss": 0.5804262161254883, "memory(GiB)": 37.5, "step": 125, "token_acc": 0.8686131386861314, "train_speed(iter/s)": 1.577141 }, { "epoch": 0.005569598560472988, "grad_norm": 5.54777717590332, "learning_rate": 9.999969384095851e-05, "loss": 0.793398094177246, "memory(GiB)": 37.5, "step": 130, "token_acc": 0.8489425981873112, "train_speed(iter/s)": 1.579617 }, { "epoch": 0.0057838138897219485, "grad_norm": 6.828816890716553, "learning_rate": 9.999966983739143e-05, "loss": 0.7011696815490722, "memory(GiB)": 37.5, "step": 135, "token_acc": 0.8466453674121406, "train_speed(iter/s)": 1.584269 }, { "epoch": 0.005998029218970909, "grad_norm": 3.7883553504943848, "learning_rate": 9.999964492803344e-05, "loss": 0.6213380336761475, "memory(GiB)": 37.5, "step": 140, "token_acc": 0.8313609467455622, "train_speed(iter/s)": 1.586497 }, { "epoch": 0.006212244548219871, "grad_norm": 3.573042392730713, "learning_rate": 9.999961911288497e-05, "loss": 0.7266541957855225, "memory(GiB)": 37.5, "step": 145, "token_acc": 0.8593272171253823, "train_speed(iter/s)": 1.586805 }, { "epoch": 0.006426459877468832, "grad_norm": 8.436943054199219, "learning_rate": 9.99995923919465e-05, "loss": 0.7316388607025146, "memory(GiB)": 37.5, "step": 150, "token_acc": 0.84375, "train_speed(iter/s)": 1.582906 }, { "epoch": 0.006640675206717793, "grad_norm": 6.530291557312012, "learning_rate": 9.99995647652185e-05, "loss": 0.97288818359375, "memory(GiB)": 37.5, "step": 155, "token_acc": 0.7899686520376176, "train_speed(iter/s)": 1.579621 }, { "epoch": 0.006854890535966753, "grad_norm": 12.907530784606934, "learning_rate": 9.99995362327015e-05, "loss": 1.5455241203308105, "memory(GiB)": 37.5, "step": 160, "token_acc": 0.7263843648208469, "train_speed(iter/s)": 1.587404 }, { "epoch": 0.007069105865215715, "grad_norm": 4.760413646697998, "learning_rate": 9.999950679439598e-05, "loss": 0.8227068901062011, "memory(GiB)": 37.5, "step": 165, "token_acc": 0.788135593220339, "train_speed(iter/s)": 1.583452 }, { "epoch": 0.007283321194464676, "grad_norm": 5.44415283203125, "learning_rate": 9.99994764503025e-05, "loss": 0.5133443355560303, "memory(GiB)": 37.5, "step": 170, "token_acc": 0.8970588235294118, "train_speed(iter/s)": 1.58748 }, { "epoch": 0.007497536523713637, "grad_norm": 6.975241184234619, "learning_rate": 9.99994452004216e-05, "loss": 0.5468311309814453, "memory(GiB)": 37.5, "step": 175, "token_acc": 0.8566433566433567, "train_speed(iter/s)": 1.588545 }, { "epoch": 0.007711751852962598, "grad_norm": 7.101792812347412, "learning_rate": 9.999941304475385e-05, "loss": 0.6864105224609375, "memory(GiB)": 37.5, "step": 180, "token_acc": 0.862876254180602, "train_speed(iter/s)": 1.585918 }, { "epoch": 0.007925967182211559, "grad_norm": 6.368824481964111, "learning_rate": 9.999937998329982e-05, "loss": 0.4765233039855957, "memory(GiB)": 37.5, "step": 185, "token_acc": 0.8905660377358491, "train_speed(iter/s)": 1.58731 }, { "epoch": 0.00814018251146052, "grad_norm": 7.348299503326416, "learning_rate": 9.999934601606014e-05, "loss": 0.5627351760864258, "memory(GiB)": 37.5, "step": 190, "token_acc": 0.90234375, "train_speed(iter/s)": 1.589881 }, { "epoch": 0.00835439784070948, "grad_norm": 1.9697582721710205, "learning_rate": 9.999931114303538e-05, "loss": 0.5596720695495605, "memory(GiB)": 37.5, "step": 195, "token_acc": 0.8664259927797834, "train_speed(iter/s)": 1.592077 }, { "epoch": 0.008568613169958442, "grad_norm": 10.599143028259277, "learning_rate": 9.99992753642262e-05, "loss": 0.606598949432373, "memory(GiB)": 37.5, "step": 200, "token_acc": 0.890625, "train_speed(iter/s)": 1.59393 }, { "epoch": 0.008782828499207404, "grad_norm": 12.356439590454102, "learning_rate": 9.999923867963326e-05, "loss": 0.6704184532165527, "memory(GiB)": 37.5, "step": 205, "token_acc": 0.8259493670886076, "train_speed(iter/s)": 1.594455 }, { "epoch": 0.008997043828456364, "grad_norm": 12.116998672485352, "learning_rate": 9.999920108925719e-05, "loss": 0.3740549564361572, "memory(GiB)": 37.5, "step": 210, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.593623 }, { "epoch": 0.009211259157705326, "grad_norm": 7.658473491668701, "learning_rate": 9.99991625930987e-05, "loss": 0.46704916954040526, "memory(GiB)": 37.5, "step": 215, "token_acc": 0.8715277777777778, "train_speed(iter/s)": 1.591522 }, { "epoch": 0.009425474486954287, "grad_norm": 7.820702075958252, "learning_rate": 9.999912319115848e-05, "loss": 0.5134871482849122, "memory(GiB)": 37.5, "step": 220, "token_acc": 0.889795918367347, "train_speed(iter/s)": 1.590246 }, { "epoch": 0.009639689816203247, "grad_norm": 10.563454627990723, "learning_rate": 9.999908288343722e-05, "loss": 0.36905412673950194, "memory(GiB)": 37.5, "step": 225, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.590638 }, { "epoch": 0.009853905145452209, "grad_norm": 7.489234447479248, "learning_rate": 9.999904166993568e-05, "loss": 0.6234542846679687, "memory(GiB)": 37.5, "step": 230, "token_acc": 0.8741721854304636, "train_speed(iter/s)": 1.592308 }, { "epoch": 0.010068120474701169, "grad_norm": 9.480616569519043, "learning_rate": 9.999899955065461e-05, "loss": 0.27893307209014895, "memory(GiB)": 37.5, "step": 235, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.593709 }, { "epoch": 0.01028233580395013, "grad_norm": 6.010523796081543, "learning_rate": 9.999895652559475e-05, "loss": 0.515814733505249, "memory(GiB)": 37.5, "step": 240, "token_acc": 0.8803680981595092, "train_speed(iter/s)": 1.593967 }, { "epoch": 0.010496551133199092, "grad_norm": 2.7071032524108887, "learning_rate": 9.999891259475688e-05, "loss": 0.5466588973999024, "memory(GiB)": 37.5, "step": 245, "token_acc": 0.8641114982578397, "train_speed(iter/s)": 1.597942 }, { "epoch": 0.010710766462448052, "grad_norm": 1.9350463151931763, "learning_rate": 9.999886775814182e-05, "loss": 0.5178678512573243, "memory(GiB)": 37.5, "step": 250, "token_acc": 0.8927536231884058, "train_speed(iter/s)": 1.59555 }, { "epoch": 0.010924981791697014, "grad_norm": 10.719970703125, "learning_rate": 9.999882201575036e-05, "loss": 0.7327447414398194, "memory(GiB)": 37.5, "step": 255, "token_acc": 0.840782122905028, "train_speed(iter/s)": 1.594301 }, { "epoch": 0.011139197120945975, "grad_norm": 5.915596008300781, "learning_rate": 9.999877536758334e-05, "loss": 0.35034244060516356, "memory(GiB)": 37.5, "step": 260, "token_acc": 0.9221789883268483, "train_speed(iter/s)": 1.594393 }, { "epoch": 0.011353412450194935, "grad_norm": 2.983351230621338, "learning_rate": 9.99987278136416e-05, "loss": 0.3717252969741821, "memory(GiB)": 37.5, "step": 265, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.594807 }, { "epoch": 0.011567627779443897, "grad_norm": 3.788642644882202, "learning_rate": 9.9998679353926e-05, "loss": 0.4846203327178955, "memory(GiB)": 37.5, "step": 270, "token_acc": 0.8902439024390244, "train_speed(iter/s)": 1.594427 }, { "epoch": 0.011781843108692859, "grad_norm": 18.387821197509766, "learning_rate": 9.999862998843743e-05, "loss": 0.48232064247131345, "memory(GiB)": 37.5, "step": 275, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.595423 }, { "epoch": 0.011996058437941819, "grad_norm": 6.788666248321533, "learning_rate": 9.999857971717678e-05, "loss": 0.42172679901123045, "memory(GiB)": 37.5, "step": 280, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.595964 }, { "epoch": 0.01221027376719078, "grad_norm": 4.882266521453857, "learning_rate": 9.999852854014495e-05, "loss": 0.5513773918151855, "memory(GiB)": 37.5, "step": 285, "token_acc": 0.9, "train_speed(iter/s)": 1.595335 }, { "epoch": 0.012424489096439742, "grad_norm": 3.1192901134490967, "learning_rate": 9.999847645734286e-05, "loss": 0.5741590023040771, "memory(GiB)": 37.5, "step": 290, "token_acc": 0.8801498127340824, "train_speed(iter/s)": 1.593588 }, { "epoch": 0.012638704425688702, "grad_norm": 1.6065020561218262, "learning_rate": 9.999842346877148e-05, "loss": 0.6212421894073487, "memory(GiB)": 37.5, "step": 295, "token_acc": 0.887719298245614, "train_speed(iter/s)": 1.595328 }, { "epoch": 0.012852919754937664, "grad_norm": 5.243360996246338, "learning_rate": 9.999836957443175e-05, "loss": 0.5061389923095703, "memory(GiB)": 37.5, "step": 300, "token_acc": 0.8674698795180723, "train_speed(iter/s)": 1.597761 }, { "epoch": 0.013067135084186625, "grad_norm": 2.6955361366271973, "learning_rate": 9.999831477432466e-05, "loss": 0.5978442192077636, "memory(GiB)": 37.5, "step": 305, "token_acc": 0.8909090909090909, "train_speed(iter/s)": 1.599338 }, { "epoch": 0.013281350413435585, "grad_norm": 2.0817008018493652, "learning_rate": 9.99982590684512e-05, "loss": 0.8117890357971191, "memory(GiB)": 37.5, "step": 310, "token_acc": 0.7814070351758794, "train_speed(iter/s)": 1.598153 }, { "epoch": 0.013495565742684547, "grad_norm": 4.403717517852783, "learning_rate": 9.999820245681236e-05, "loss": 0.3038931131362915, "memory(GiB)": 37.5, "step": 315, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.596641 }, { "epoch": 0.013709781071933507, "grad_norm": 4.085194110870361, "learning_rate": 9.999814493940918e-05, "loss": 0.46436023712158203, "memory(GiB)": 37.5, "step": 320, "token_acc": 0.8921933085501859, "train_speed(iter/s)": 1.597431 }, { "epoch": 0.013923996401182468, "grad_norm": 3.3823909759521484, "learning_rate": 9.999808651624271e-05, "loss": 0.6125343799591064, "memory(GiB)": 37.5, "step": 325, "token_acc": 0.897887323943662, "train_speed(iter/s)": 1.597719 }, { "epoch": 0.01413821173043143, "grad_norm": 2.7831838130950928, "learning_rate": 9.9998027187314e-05, "loss": 0.2069556474685669, "memory(GiB)": 37.5, "step": 330, "token_acc": 0.9566666666666667, "train_speed(iter/s)": 1.598621 }, { "epoch": 0.01435242705968039, "grad_norm": 2.0364551544189453, "learning_rate": 9.999796695262413e-05, "loss": 0.288651704788208, "memory(GiB)": 37.5, "step": 335, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.597348 }, { "epoch": 0.014566642388929352, "grad_norm": 2.626919984817505, "learning_rate": 9.999790581217418e-05, "loss": 0.41779317855834963, "memory(GiB)": 37.5, "step": 340, "token_acc": 0.8868613138686131, "train_speed(iter/s)": 1.596166 }, { "epoch": 0.014780857718178313, "grad_norm": 1.1496812105178833, "learning_rate": 9.999784376596526e-05, "loss": 0.8692158699035645, "memory(GiB)": 37.5, "step": 345, "token_acc": 0.8304498269896193, "train_speed(iter/s)": 1.59527 }, { "epoch": 0.014995073047427273, "grad_norm": 21.77144432067871, "learning_rate": 9.999778081399851e-05, "loss": 0.4736666679382324, "memory(GiB)": 37.5, "step": 350, "token_acc": 0.8617021276595744, "train_speed(iter/s)": 1.595339 }, { "epoch": 0.015209288376676235, "grad_norm": 2.4206700325012207, "learning_rate": 9.999771695627505e-05, "loss": 0.7434126377105713, "memory(GiB)": 37.5, "step": 355, "token_acc": 0.8715596330275229, "train_speed(iter/s)": 1.597642 }, { "epoch": 0.015423503705925197, "grad_norm": 1.0732156038284302, "learning_rate": 9.999765219279605e-05, "loss": 0.329630184173584, "memory(GiB)": 37.5, "step": 360, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.600512 }, { "epoch": 0.015637719035174157, "grad_norm": 3.207695484161377, "learning_rate": 9.999758652356268e-05, "loss": 0.5550663471221924, "memory(GiB)": 37.5, "step": 365, "token_acc": 0.8937728937728938, "train_speed(iter/s)": 1.601205 }, { "epoch": 0.015851934364423118, "grad_norm": 9.176240921020508, "learning_rate": 9.999751994857614e-05, "loss": 0.44768366813659666, "memory(GiB)": 37.5, "step": 370, "token_acc": 0.9, "train_speed(iter/s)": 1.600252 }, { "epoch": 0.01606614969367208, "grad_norm": 1.6933114528656006, "learning_rate": 9.999745246783761e-05, "loss": 0.30030553340911864, "memory(GiB)": 37.5, "step": 375, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.601716 }, { "epoch": 0.01628036502292104, "grad_norm": 3.441840171813965, "learning_rate": 9.999738408134832e-05, "loss": 0.557397985458374, "memory(GiB)": 37.5, "step": 380, "token_acc": 0.8806451612903226, "train_speed(iter/s)": 1.601425 }, { "epoch": 0.01649458035217, "grad_norm": 2.432704448699951, "learning_rate": 9.999731478910952e-05, "loss": 0.6375790596008301, "memory(GiB)": 37.5, "step": 385, "token_acc": 0.8679245283018868, "train_speed(iter/s)": 1.603461 }, { "epoch": 0.01670879568141896, "grad_norm": 21.01386833190918, "learning_rate": 9.999724459112247e-05, "loss": 0.8644254684448243, "memory(GiB)": 37.5, "step": 390, "token_acc": 0.8057324840764332, "train_speed(iter/s)": 1.602844 }, { "epoch": 0.016923011010667923, "grad_norm": 2.6111390590667725, "learning_rate": 9.999717348738843e-05, "loss": 0.3196715354919434, "memory(GiB)": 37.5, "step": 395, "token_acc": 0.9330708661417323, "train_speed(iter/s)": 1.602462 }, { "epoch": 0.017137226339916885, "grad_norm": 2.777621269226074, "learning_rate": 9.999710147790869e-05, "loss": 0.9574935913085938, "memory(GiB)": 37.5, "step": 400, "token_acc": 0.8109965635738832, "train_speed(iter/s)": 1.602791 }, { "epoch": 0.017351441669165846, "grad_norm": 3.066444158554077, "learning_rate": 9.999702856268456e-05, "loss": 0.29958043098449705, "memory(GiB)": 37.5, "step": 405, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.602091 }, { "epoch": 0.017565656998414808, "grad_norm": 1.603281855583191, "learning_rate": 9.999695474171735e-05, "loss": 0.27283363342285155, "memory(GiB)": 37.5, "step": 410, "token_acc": 0.9421487603305785, "train_speed(iter/s)": 1.600658 }, { "epoch": 0.017779872327663766, "grad_norm": 4.4891743659973145, "learning_rate": 9.99968800150084e-05, "loss": 0.5744076728820801, "memory(GiB)": 37.5, "step": 415, "token_acc": 0.8875502008032129, "train_speed(iter/s)": 1.600338 }, { "epoch": 0.017994087656912728, "grad_norm": 2.173382520675659, "learning_rate": 9.999680438255908e-05, "loss": 0.31710736751556395, "memory(GiB)": 37.5, "step": 420, "token_acc": 0.9172932330827067, "train_speed(iter/s)": 1.598889 }, { "epoch": 0.01820830298616169, "grad_norm": 4.699966907501221, "learning_rate": 9.999672784437075e-05, "loss": 0.4020712852478027, "memory(GiB)": 37.5, "step": 425, "token_acc": 0.896797153024911, "train_speed(iter/s)": 1.597389 }, { "epoch": 0.01842251831541065, "grad_norm": 2.7447617053985596, "learning_rate": 9.999665040044477e-05, "loss": 0.9783286094665528, "memory(GiB)": 37.5, "step": 430, "token_acc": 0.8021201413427562, "train_speed(iter/s)": 1.59801 }, { "epoch": 0.018636733644659613, "grad_norm": 2.9146900177001953, "learning_rate": 9.99965720507826e-05, "loss": 0.5752747535705567, "memory(GiB)": 37.5, "step": 435, "token_acc": 0.8657718120805369, "train_speed(iter/s)": 1.598449 }, { "epoch": 0.018850948973908575, "grad_norm": 3.7495129108428955, "learning_rate": 9.999649279538562e-05, "loss": 0.5166052341461181, "memory(GiB)": 37.5, "step": 440, "token_acc": 0.9022556390977443, "train_speed(iter/s)": 1.59966 }, { "epoch": 0.019065164303157533, "grad_norm": 8.264739990234375, "learning_rate": 9.999641263425524e-05, "loss": 0.5609297752380371, "memory(GiB)": 37.5, "step": 445, "token_acc": 0.881578947368421, "train_speed(iter/s)": 1.599844 }, { "epoch": 0.019279379632406494, "grad_norm": 3.1182284355163574, "learning_rate": 9.999633156739298e-05, "loss": 0.26049754619598386, "memory(GiB)": 37.5, "step": 450, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.598198 }, { "epoch": 0.019493594961655456, "grad_norm": 2.6130237579345703, "learning_rate": 9.999624959480025e-05, "loss": 0.3706441164016724, "memory(GiB)": 37.5, "step": 455, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.597391 }, { "epoch": 0.019707810290904418, "grad_norm": 3.1408162117004395, "learning_rate": 9.999616671647858e-05, "loss": 0.6261306762695312, "memory(GiB)": 37.5, "step": 460, "token_acc": 0.8765060240963856, "train_speed(iter/s)": 1.597834 }, { "epoch": 0.01992202562015338, "grad_norm": 1.5691791772842407, "learning_rate": 9.999608293242943e-05, "loss": 0.4467515468597412, "memory(GiB)": 37.5, "step": 465, "token_acc": 0.8958990536277602, "train_speed(iter/s)": 1.597067 }, { "epoch": 0.020136240949402338, "grad_norm": 0.5782378315925598, "learning_rate": 9.999599824265436e-05, "loss": 0.2356778860092163, "memory(GiB)": 37.5, "step": 470, "token_acc": 0.9591078066914498, "train_speed(iter/s)": 1.596383 }, { "epoch": 0.0203504562786513, "grad_norm": 2.6782076358795166, "learning_rate": 9.999591264715485e-05, "loss": 0.462082576751709, "memory(GiB)": 37.5, "step": 475, "token_acc": 0.8793650793650793, "train_speed(iter/s)": 1.596656 }, { "epoch": 0.02056467160790026, "grad_norm": 3.1960690021514893, "learning_rate": 9.99958261459325e-05, "loss": 0.5928387641906738, "memory(GiB)": 37.5, "step": 480, "token_acc": 0.8697183098591549, "train_speed(iter/s)": 1.597514 }, { "epoch": 0.020778886937149223, "grad_norm": 1.192889928817749, "learning_rate": 9.999573873898887e-05, "loss": 0.4984272956848145, "memory(GiB)": 37.5, "step": 485, "token_acc": 0.8830645161290323, "train_speed(iter/s)": 1.598017 }, { "epoch": 0.020993102266398184, "grad_norm": 2.858001470565796, "learning_rate": 9.999565042632553e-05, "loss": 0.405056095123291, "memory(GiB)": 37.5, "step": 490, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.598171 }, { "epoch": 0.021207317595647146, "grad_norm": 3.2187016010284424, "learning_rate": 9.999556120794407e-05, "loss": 0.5845155715942383, "memory(GiB)": 37.5, "step": 495, "token_acc": 0.875, "train_speed(iter/s)": 1.597507 }, { "epoch": 0.021421532924896104, "grad_norm": 2.782825469970703, "learning_rate": 9.999547108384612e-05, "loss": 0.3612308740615845, "memory(GiB)": 37.5, "step": 500, "token_acc": 0.9260450160771704, "train_speed(iter/s)": 1.596431 }, { "epoch": 0.021421532924896104, "eval_loss": 2.7975118160247803, "eval_runtime": 11.6902, "eval_samples_per_second": 8.554, "eval_steps_per_second": 8.554, "eval_token_acc": 0.41278295605858856, "step": 500 }, { "epoch": 0.021635748254145066, "grad_norm": 3.0929579734802246, "learning_rate": 9.999538005403332e-05, "loss": 0.2719956159591675, "memory(GiB)": 37.5, "step": 505, "token_acc": 0.5549076773566569, "train_speed(iter/s)": 1.535152 }, { "epoch": 0.021849963583394028, "grad_norm": 2.1490466594696045, "learning_rate": 9.999528811850732e-05, "loss": 0.2739660978317261, "memory(GiB)": 37.5, "step": 510, "token_acc": 0.9323308270676691, "train_speed(iter/s)": 1.534688 }, { "epoch": 0.02206417891264299, "grad_norm": 19.568870544433594, "learning_rate": 9.999519527726974e-05, "loss": 0.44692068099975585, "memory(GiB)": 37.5, "step": 515, "token_acc": 0.8656716417910447, "train_speed(iter/s)": 1.536063 }, { "epoch": 0.02227839424189195, "grad_norm": 1.5414575338363647, "learning_rate": 9.999510153032234e-05, "loss": 0.3300382375717163, "memory(GiB)": 37.5, "step": 520, "token_acc": 0.9042553191489362, "train_speed(iter/s)": 1.535326 }, { "epoch": 0.022492609571140913, "grad_norm": 2.7912769317626953, "learning_rate": 9.999500687766675e-05, "loss": 0.43247041702270506, "memory(GiB)": 37.5, "step": 525, "token_acc": 0.8990536277602523, "train_speed(iter/s)": 1.535081 }, { "epoch": 0.02270682490038987, "grad_norm": 3.2685458660125732, "learning_rate": 9.999491131930472e-05, "loss": 0.4273550510406494, "memory(GiB)": 37.5, "step": 530, "token_acc": 0.9026845637583892, "train_speed(iter/s)": 1.536094 }, { "epoch": 0.022921040229638832, "grad_norm": 9.24952507019043, "learning_rate": 9.999481485523798e-05, "loss": 0.4911740779876709, "memory(GiB)": 37.5, "step": 535, "token_acc": 0.8870967741935484, "train_speed(iter/s)": 1.535483 }, { "epoch": 0.023135255558887794, "grad_norm": 3.0415287017822266, "learning_rate": 9.999471748546826e-05, "loss": 0.51724271774292, "memory(GiB)": 37.5, "step": 540, "token_acc": 0.8731707317073171, "train_speed(iter/s)": 1.535609 }, { "epoch": 0.023349470888136756, "grad_norm": 3.4089932441711426, "learning_rate": 9.999461920999733e-05, "loss": 0.5261340141296387, "memory(GiB)": 37.5, "step": 545, "token_acc": 0.888135593220339, "train_speed(iter/s)": 1.535769 }, { "epoch": 0.023563686217385717, "grad_norm": 18.20499610900879, "learning_rate": 9.999452002882699e-05, "loss": 0.49789934158325194, "memory(GiB)": 37.5, "step": 550, "token_acc": 0.8901734104046243, "train_speed(iter/s)": 1.534946 }, { "epoch": 0.023777901546634676, "grad_norm": 1.6879369020462036, "learning_rate": 9.999441994195902e-05, "loss": 0.46434779167175294, "memory(GiB)": 37.5, "step": 555, "token_acc": 0.8896797153024911, "train_speed(iter/s)": 1.53443 }, { "epoch": 0.023992116875883637, "grad_norm": 4.219235420227051, "learning_rate": 9.999431894939522e-05, "loss": 0.455576229095459, "memory(GiB)": 37.5, "step": 560, "token_acc": 0.912, "train_speed(iter/s)": 1.534119 }, { "epoch": 0.0242063322051326, "grad_norm": 3.3723971843719482, "learning_rate": 9.999421705113743e-05, "loss": 0.31247141361236574, "memory(GiB)": 37.5, "step": 565, "token_acc": 0.9265536723163842, "train_speed(iter/s)": 1.533377 }, { "epoch": 0.02442054753438156, "grad_norm": 8.35122299194336, "learning_rate": 9.999411424718752e-05, "loss": 0.5269171714782714, "memory(GiB)": 37.5, "step": 570, "token_acc": 0.8896321070234113, "train_speed(iter/s)": 1.532915 }, { "epoch": 0.024634762863630522, "grad_norm": 4.387744426727295, "learning_rate": 9.999401053754731e-05, "loss": 0.5076206207275391, "memory(GiB)": 37.5, "step": 575, "token_acc": 0.872093023255814, "train_speed(iter/s)": 1.532767 }, { "epoch": 0.024848978192879484, "grad_norm": 7.58473014831543, "learning_rate": 9.999390592221871e-05, "loss": 0.5435320854187011, "memory(GiB)": 37.5, "step": 580, "token_acc": 0.8801261829652997, "train_speed(iter/s)": 1.532441 }, { "epoch": 0.025063193522128442, "grad_norm": 5.430418491363525, "learning_rate": 9.999380040120362e-05, "loss": 0.4074378967285156, "memory(GiB)": 37.5, "step": 585, "token_acc": 0.907051282051282, "train_speed(iter/s)": 1.533476 }, { "epoch": 0.025277408851377404, "grad_norm": 5.6270222663879395, "learning_rate": 9.999369397450392e-05, "loss": 0.29258089065551757, "memory(GiB)": 37.5, "step": 590, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.533386 }, { "epoch": 0.025491624180626365, "grad_norm": 3.47228741645813, "learning_rate": 9.999358664212158e-05, "loss": 0.46176977157592775, "memory(GiB)": 37.5, "step": 595, "token_acc": 0.9120879120879121, "train_speed(iter/s)": 1.532775 }, { "epoch": 0.025705839509875327, "grad_norm": 2.8095438480377197, "learning_rate": 9.999347840405848e-05, "loss": 0.5177341938018799, "memory(GiB)": 37.5, "step": 600, "token_acc": 0.8825622775800712, "train_speed(iter/s)": 1.532215 }, { "epoch": 0.02592005483912429, "grad_norm": 2.0089073181152344, "learning_rate": 9.999336926031666e-05, "loss": 0.5143534183502197, "memory(GiB)": 37.5, "step": 605, "token_acc": 0.8606271777003485, "train_speed(iter/s)": 1.531688 }, { "epoch": 0.02613427016837325, "grad_norm": 4.277317523956299, "learning_rate": 9.999325921089804e-05, "loss": 0.6927183151245118, "memory(GiB)": 37.5, "step": 610, "token_acc": 0.8675213675213675, "train_speed(iter/s)": 1.530854 }, { "epoch": 0.02634848549762221, "grad_norm": 1.7104228734970093, "learning_rate": 9.999314825580464e-05, "loss": 0.22567715644836425, "memory(GiB)": 37.5, "step": 615, "token_acc": 0.9471698113207547, "train_speed(iter/s)": 1.530962 }, { "epoch": 0.02656270082687117, "grad_norm": 1.8629549741744995, "learning_rate": 9.999303639503845e-05, "loss": 0.43591885566711425, "memory(GiB)": 37.5, "step": 620, "token_acc": 0.8933333333333333, "train_speed(iter/s)": 1.532513 }, { "epoch": 0.026776916156120132, "grad_norm": 5.931727886199951, "learning_rate": 9.999292362860151e-05, "loss": 0.48184781074523925, "memory(GiB)": 37.5, "step": 625, "token_acc": 0.8773006134969326, "train_speed(iter/s)": 1.532107 }, { "epoch": 0.026991131485369094, "grad_norm": 5.596036911010742, "learning_rate": 9.999280995649588e-05, "loss": 0.5929035663604736, "memory(GiB)": 37.5, "step": 630, "token_acc": 0.8419117647058824, "train_speed(iter/s)": 1.534377 }, { "epoch": 0.027205346814618055, "grad_norm": 1.9156562089920044, "learning_rate": 9.999269537872357e-05, "loss": 0.48219833374023435, "memory(GiB)": 37.5, "step": 635, "token_acc": 0.8817204301075269, "train_speed(iter/s)": 1.534226 }, { "epoch": 0.027419562143867014, "grad_norm": 4.1125078201293945, "learning_rate": 9.99925798952867e-05, "loss": 0.594445276260376, "memory(GiB)": 37.5, "step": 640, "token_acc": 0.8737864077669902, "train_speed(iter/s)": 1.535033 }, { "epoch": 0.027633777473115975, "grad_norm": 5.104694843292236, "learning_rate": 9.999246350618734e-05, "loss": 0.3508764266967773, "memory(GiB)": 37.5, "step": 645, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.535305 }, { "epoch": 0.027847992802364937, "grad_norm": 0.9887524843215942, "learning_rate": 9.999234621142764e-05, "loss": 0.26190521717071535, "memory(GiB)": 37.5, "step": 650, "token_acc": 0.952076677316294, "train_speed(iter/s)": 1.534936 }, { "epoch": 0.0280622081316139, "grad_norm": 1.7562607526779175, "learning_rate": 9.999222801100966e-05, "loss": 0.3767815351486206, "memory(GiB)": 37.5, "step": 655, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.536059 }, { "epoch": 0.02827642346086286, "grad_norm": 3.4479074478149414, "learning_rate": 9.999210890493558e-05, "loss": 0.5469394207000733, "memory(GiB)": 37.5, "step": 660, "token_acc": 0.8622950819672132, "train_speed(iter/s)": 1.536771 }, { "epoch": 0.028490638790111822, "grad_norm": 2.079695701599121, "learning_rate": 9.999198889320755e-05, "loss": 0.3911888599395752, "memory(GiB)": 37.5, "step": 665, "token_acc": 0.907051282051282, "train_speed(iter/s)": 1.537522 }, { "epoch": 0.02870485411936078, "grad_norm": 16.93535614013672, "learning_rate": 9.999186797582775e-05, "loss": 0.5500761032104492, "memory(GiB)": 37.5, "step": 670, "token_acc": 0.8861538461538462, "train_speed(iter/s)": 1.537979 }, { "epoch": 0.028919069448609742, "grad_norm": 0.9915966987609863, "learning_rate": 9.999174615279836e-05, "loss": 0.35157554149627684, "memory(GiB)": 37.5, "step": 675, "token_acc": 0.9375, "train_speed(iter/s)": 1.537608 }, { "epoch": 0.029133284777858703, "grad_norm": 4.492029190063477, "learning_rate": 9.999162342412161e-05, "loss": 0.7387548446655273, "memory(GiB)": 37.5, "step": 680, "token_acc": 0.8228782287822878, "train_speed(iter/s)": 1.539182 }, { "epoch": 0.029347500107107665, "grad_norm": 5.000621795654297, "learning_rate": 9.999149978979969e-05, "loss": 0.4824190616607666, "memory(GiB)": 37.5, "step": 685, "token_acc": 0.9118541033434651, "train_speed(iter/s)": 1.540438 }, { "epoch": 0.029561715436356627, "grad_norm": 11.169784545898438, "learning_rate": 9.999137524983485e-05, "loss": 0.43438305854797366, "memory(GiB)": 37.5, "step": 690, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.541368 }, { "epoch": 0.02977593076560559, "grad_norm": 1.4074715375900269, "learning_rate": 9.999124980422936e-05, "loss": 0.4303253173828125, "memory(GiB)": 37.5, "step": 695, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.541026 }, { "epoch": 0.029990146094854547, "grad_norm": 4.374685287475586, "learning_rate": 9.999112345298549e-05, "loss": 0.31561267375946045, "memory(GiB)": 37.5, "step": 700, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.540852 }, { "epoch": 0.03020436142410351, "grad_norm": 2.667980194091797, "learning_rate": 9.999099619610552e-05, "loss": 0.39325225353240967, "memory(GiB)": 37.5, "step": 705, "token_acc": 0.9301310043668122, "train_speed(iter/s)": 1.540221 }, { "epoch": 0.03041857675335247, "grad_norm": 3.627833366394043, "learning_rate": 9.999086803359174e-05, "loss": 0.40247907638549807, "memory(GiB)": 37.5, "step": 710, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.540041 }, { "epoch": 0.03063279208260143, "grad_norm": 9.585838317871094, "learning_rate": 9.99907389654465e-05, "loss": 0.3499692678451538, "memory(GiB)": 37.5, "step": 715, "token_acc": 0.9121338912133892, "train_speed(iter/s)": 1.540381 }, { "epoch": 0.030847007411850393, "grad_norm": 3.949024200439453, "learning_rate": 9.999060899167214e-05, "loss": 0.237078857421875, "memory(GiB)": 37.5, "step": 720, "token_acc": 0.93359375, "train_speed(iter/s)": 1.540724 }, { "epoch": 0.03106122274109935, "grad_norm": 1.7394932508468628, "learning_rate": 9.9990478112271e-05, "loss": 0.34479718208312987, "memory(GiB)": 37.5, "step": 725, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.540744 }, { "epoch": 0.03127543807034831, "grad_norm": 2.779735803604126, "learning_rate": 9.999034632724546e-05, "loss": 0.37788422107696534, "memory(GiB)": 37.5, "step": 730, "token_acc": 0.9292307692307692, "train_speed(iter/s)": 1.541058 }, { "epoch": 0.03148965339959728, "grad_norm": 1.852493166923523, "learning_rate": 9.999021363659788e-05, "loss": 0.6921148300170898, "memory(GiB)": 37.5, "step": 735, "token_acc": 0.8653198653198653, "train_speed(iter/s)": 1.541423 }, { "epoch": 0.031703868728846236, "grad_norm": 0.9349266886711121, "learning_rate": 9.99900800403307e-05, "loss": 0.21682896614074706, "memory(GiB)": 37.5, "step": 740, "token_acc": 0.9425675675675675, "train_speed(iter/s)": 1.540994 }, { "epoch": 0.031918084058095195, "grad_norm": 8.975934982299805, "learning_rate": 9.998994553844632e-05, "loss": 0.428836727142334, "memory(GiB)": 37.5, "step": 745, "token_acc": 0.9118541033434651, "train_speed(iter/s)": 1.540857 }, { "epoch": 0.03213229938734416, "grad_norm": 2.3629822731018066, "learning_rate": 9.99898101309472e-05, "loss": 0.33065619468688967, "memory(GiB)": 37.5, "step": 750, "token_acc": 0.9331103678929766, "train_speed(iter/s)": 1.540187 }, { "epoch": 0.03234651471659312, "grad_norm": 7.855225563049316, "learning_rate": 9.998967381783575e-05, "loss": 0.8223451614379883, "memory(GiB)": 37.5, "step": 755, "token_acc": 0.8458904109589042, "train_speed(iter/s)": 1.5409 }, { "epoch": 0.03256073004584208, "grad_norm": 3.2996723651885986, "learning_rate": 9.998953659911447e-05, "loss": 0.598546314239502, "memory(GiB)": 37.5, "step": 760, "token_acc": 0.8862876254180602, "train_speed(iter/s)": 1.541645 }, { "epoch": 0.03277494537509104, "grad_norm": 0.6902215480804443, "learning_rate": 9.998939847478585e-05, "loss": 0.3614722490310669, "memory(GiB)": 37.5, "step": 765, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.540245 }, { "epoch": 0.03298916070434, "grad_norm": 2.2645633220672607, "learning_rate": 9.998925944485238e-05, "loss": 0.48872909545898435, "memory(GiB)": 38.87, "step": 770, "token_acc": 0.8937007874015748, "train_speed(iter/s)": 1.541166 }, { "epoch": 0.033203376033588965, "grad_norm": 1.8801430463790894, "learning_rate": 9.998911950931658e-05, "loss": 0.5205110549926758, "memory(GiB)": 38.87, "step": 775, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.541245 }, { "epoch": 0.03341759136283792, "grad_norm": 3.595539093017578, "learning_rate": 9.998897866818099e-05, "loss": 0.36504881381988524, "memory(GiB)": 38.87, "step": 780, "token_acc": 0.9351851851851852, "train_speed(iter/s)": 1.541212 }, { "epoch": 0.03363180669208689, "grad_norm": 2.0644922256469727, "learning_rate": 9.998883692144817e-05, "loss": 0.21479194164276122, "memory(GiB)": 38.87, "step": 785, "token_acc": 0.945619335347432, "train_speed(iter/s)": 1.540907 }, { "epoch": 0.033846022021335846, "grad_norm": 1.4296272993087769, "learning_rate": 9.998869426912065e-05, "loss": 0.3351928234100342, "memory(GiB)": 38.87, "step": 790, "token_acc": 0.8934707903780069, "train_speed(iter/s)": 1.540714 }, { "epoch": 0.034060237350584804, "grad_norm": 2.193852186203003, "learning_rate": 9.998855071120104e-05, "loss": 0.603451156616211, "memory(GiB)": 43.84, "step": 795, "token_acc": 0.8564231738035264, "train_speed(iter/s)": 1.541129 }, { "epoch": 0.03427445267983377, "grad_norm": 1.8556686639785767, "learning_rate": 9.998840624769196e-05, "loss": 0.6407320022583007, "memory(GiB)": 43.84, "step": 800, "token_acc": 0.8717948717948718, "train_speed(iter/s)": 1.540613 }, { "epoch": 0.03448866800908273, "grad_norm": 5.794185638427734, "learning_rate": 9.9988260878596e-05, "loss": 0.506827974319458, "memory(GiB)": 43.84, "step": 805, "token_acc": 0.8806451612903226, "train_speed(iter/s)": 1.541607 }, { "epoch": 0.03470288333833169, "grad_norm": 2.377993106842041, "learning_rate": 9.99881146039158e-05, "loss": 0.45647807121276857, "memory(GiB)": 46.34, "step": 810, "token_acc": 0.9004149377593361, "train_speed(iter/s)": 1.541177 }, { "epoch": 0.03491709866758065, "grad_norm": 2.1459977626800537, "learning_rate": 9.998796742365402e-05, "loss": 0.7901912212371827, "memory(GiB)": 46.34, "step": 815, "token_acc": 0.8502994011976048, "train_speed(iter/s)": 1.540799 }, { "epoch": 0.035131313996829616, "grad_norm": 1.7076811790466309, "learning_rate": 9.99878193378133e-05, "loss": 0.4915444850921631, "memory(GiB)": 46.34, "step": 820, "token_acc": 0.8833333333333333, "train_speed(iter/s)": 1.540874 }, { "epoch": 0.035345529326078574, "grad_norm": 2.646634101867676, "learning_rate": 9.998767034639636e-05, "loss": 0.2828678131103516, "memory(GiB)": 46.34, "step": 825, "token_acc": 0.94, "train_speed(iter/s)": 1.541539 }, { "epoch": 0.03555974465532753, "grad_norm": 7.530813217163086, "learning_rate": 9.998752044940587e-05, "loss": 0.5640981197357178, "memory(GiB)": 46.34, "step": 830, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.541242 }, { "epoch": 0.0357739599845765, "grad_norm": 2.490334987640381, "learning_rate": 9.998736964684454e-05, "loss": 0.6337035179138184, "memory(GiB)": 46.34, "step": 835, "token_acc": 0.8539682539682539, "train_speed(iter/s)": 1.540477 }, { "epoch": 0.035988175313825456, "grad_norm": 3.5539307594299316, "learning_rate": 9.998721793871513e-05, "loss": 0.4594445705413818, "memory(GiB)": 46.34, "step": 840, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.540459 }, { "epoch": 0.03620239064307442, "grad_norm": 6.169102191925049, "learning_rate": 9.998706532502038e-05, "loss": 0.6339198112487793, "memory(GiB)": 46.34, "step": 845, "token_acc": 0.8452722063037249, "train_speed(iter/s)": 1.53969 }, { "epoch": 0.03641660597232338, "grad_norm": 1.6487029790878296, "learning_rate": 9.998691180576306e-05, "loss": 0.2965723514556885, "memory(GiB)": 46.34, "step": 850, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.539564 }, { "epoch": 0.03663082130157234, "grad_norm": 1.9045162200927734, "learning_rate": 9.998675738094591e-05, "loss": 0.3619127988815308, "memory(GiB)": 46.34, "step": 855, "token_acc": 0.9110169491525424, "train_speed(iter/s)": 1.539488 }, { "epoch": 0.0368450366308213, "grad_norm": 2.148482322692871, "learning_rate": 9.998660205057179e-05, "loss": 0.32315311431884763, "memory(GiB)": 46.34, "step": 860, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.539673 }, { "epoch": 0.03705925196007026, "grad_norm": 3.091083526611328, "learning_rate": 9.998644581464348e-05, "loss": 0.44289202690124513, "memory(GiB)": 51.38, "step": 865, "token_acc": 0.9039145907473309, "train_speed(iter/s)": 1.539281 }, { "epoch": 0.037273467289319226, "grad_norm": 8.895548820495605, "learning_rate": 9.998628867316378e-05, "loss": 0.38748114109039306, "memory(GiB)": 51.38, "step": 870, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.540028 }, { "epoch": 0.037487682618568184, "grad_norm": 5.906196594238281, "learning_rate": 9.99861306261356e-05, "loss": 0.20566120147705078, "memory(GiB)": 51.38, "step": 875, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.539577 }, { "epoch": 0.03770189794781715, "grad_norm": 1.3452590703964233, "learning_rate": 9.998597167356176e-05, "loss": 0.5080497741699219, "memory(GiB)": 51.38, "step": 880, "token_acc": 0.8972602739726028, "train_speed(iter/s)": 1.539294 }, { "epoch": 0.03791611327706611, "grad_norm": 0.6689194440841675, "learning_rate": 9.998581181544516e-05, "loss": 0.4085539817810059, "memory(GiB)": 51.38, "step": 885, "token_acc": 0.8862876254180602, "train_speed(iter/s)": 1.539055 }, { "epoch": 0.038130328606315066, "grad_norm": 2.1357405185699463, "learning_rate": 9.998565105178869e-05, "loss": 0.14531863927841188, "memory(GiB)": 51.38, "step": 890, "token_acc": 0.9611307420494699, "train_speed(iter/s)": 1.538501 }, { "epoch": 0.03834454393556403, "grad_norm": 1.4525089263916016, "learning_rate": 9.998548938259525e-05, "loss": 0.5426840305328369, "memory(GiB)": 51.38, "step": 895, "token_acc": 0.86, "train_speed(iter/s)": 1.539598 }, { "epoch": 0.03855875926481299, "grad_norm": 1.4722031354904175, "learning_rate": 9.998532680786778e-05, "loss": 0.2029109239578247, "memory(GiB)": 51.38, "step": 900, "token_acc": 0.9448051948051948, "train_speed(iter/s)": 1.540471 }, { "epoch": 0.038772974594061954, "grad_norm": 5.93183708190918, "learning_rate": 9.998516332760924e-05, "loss": 0.40140385627746583, "memory(GiB)": 51.38, "step": 905, "token_acc": 0.9105960264900662, "train_speed(iter/s)": 1.540066 }, { "epoch": 0.03898718992331091, "grad_norm": 3.162872076034546, "learning_rate": 9.998499894182255e-05, "loss": 0.23935751914978026, "memory(GiB)": 51.38, "step": 910, "token_acc": 0.9539007092198581, "train_speed(iter/s)": 1.540308 }, { "epoch": 0.03920140525255987, "grad_norm": 8.498069763183594, "learning_rate": 9.998483365051075e-05, "loss": 0.6205127239227295, "memory(GiB)": 51.38, "step": 915, "token_acc": 0.9044117647058824, "train_speed(iter/s)": 1.540324 }, { "epoch": 0.039415620581808836, "grad_norm": 0.29947859048843384, "learning_rate": 9.998466745367678e-05, "loss": 0.5096943855285645, "memory(GiB)": 51.38, "step": 920, "token_acc": 0.8939393939393939, "train_speed(iter/s)": 1.540293 }, { "epoch": 0.039629835911057794, "grad_norm": 2.7575831413269043, "learning_rate": 9.998450035132367e-05, "loss": 0.45034236907958985, "memory(GiB)": 51.38, "step": 925, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.539629 }, { "epoch": 0.03984405124030676, "grad_norm": 2.149522542953491, "learning_rate": 9.998433234345446e-05, "loss": 0.4027276039123535, "memory(GiB)": 51.38, "step": 930, "token_acc": 0.9046052631578947, "train_speed(iter/s)": 1.539632 }, { "epoch": 0.04005826656955572, "grad_norm": 9.035931587219238, "learning_rate": 9.998416343007215e-05, "loss": 0.7069473743438721, "memory(GiB)": 51.38, "step": 935, "token_acc": 0.8589341692789969, "train_speed(iter/s)": 1.539922 }, { "epoch": 0.040272481898804675, "grad_norm": 1.8420287370681763, "learning_rate": 9.998399361117985e-05, "loss": 0.48861284255981446, "memory(GiB)": 51.38, "step": 940, "token_acc": 0.8902077151335311, "train_speed(iter/s)": 1.539309 }, { "epoch": 0.04048669722805364, "grad_norm": 3.6801247596740723, "learning_rate": 9.998382288678062e-05, "loss": 0.9159446716308594, "memory(GiB)": 51.38, "step": 945, "token_acc": 0.832089552238806, "train_speed(iter/s)": 1.538811 }, { "epoch": 0.0407009125573026, "grad_norm": 1.0735516548156738, "learning_rate": 9.998365125687754e-05, "loss": 0.4056541919708252, "memory(GiB)": 51.38, "step": 950, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.538582 }, { "epoch": 0.040915127886551564, "grad_norm": 4.29098653793335, "learning_rate": 9.998347872147374e-05, "loss": 0.28139595985412597, "memory(GiB)": 51.38, "step": 955, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.538806 }, { "epoch": 0.04112934321580052, "grad_norm": 5.001364231109619, "learning_rate": 9.998330528057234e-05, "loss": 0.7093822479248046, "memory(GiB)": 51.38, "step": 960, "token_acc": 0.847682119205298, "train_speed(iter/s)": 1.538301 }, { "epoch": 0.04134355854504949, "grad_norm": 1.2526419162750244, "learning_rate": 9.998313093417646e-05, "loss": 0.3430963039398193, "memory(GiB)": 51.38, "step": 965, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.537642 }, { "epoch": 0.041557773874298445, "grad_norm": 1.0064009428024292, "learning_rate": 9.998295568228928e-05, "loss": 0.5640243053436279, "memory(GiB)": 51.38, "step": 970, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.538076 }, { "epoch": 0.041771989203547404, "grad_norm": 0.7051804661750793, "learning_rate": 9.998277952491397e-05, "loss": 0.22609176635742187, "memory(GiB)": 51.38, "step": 975, "token_acc": 0.9472295514511874, "train_speed(iter/s)": 1.538037 }, { "epoch": 0.04198620453279637, "grad_norm": 3.6526687145233154, "learning_rate": 9.998260246205372e-05, "loss": 0.43772711753845217, "memory(GiB)": 51.38, "step": 980, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.538141 }, { "epoch": 0.04220041986204533, "grad_norm": 2.3816516399383545, "learning_rate": 9.998242449371174e-05, "loss": 0.3657313346862793, "memory(GiB)": 51.38, "step": 985, "token_acc": 0.9190283400809717, "train_speed(iter/s)": 1.538226 }, { "epoch": 0.04241463519129429, "grad_norm": 2.511533737182617, "learning_rate": 9.998224561989124e-05, "loss": 0.5338665962219238, "memory(GiB)": 51.38, "step": 990, "token_acc": 0.8845070422535212, "train_speed(iter/s)": 1.537996 }, { "epoch": 0.04262885052054325, "grad_norm": 1.8468236923217773, "learning_rate": 9.99820658405955e-05, "loss": 0.32266232967376707, "memory(GiB)": 51.38, "step": 995, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.538164 }, { "epoch": 0.04284306584979221, "grad_norm": 3.6786274909973145, "learning_rate": 9.998188515582773e-05, "loss": 0.5549796104431153, "memory(GiB)": 51.38, "step": 1000, "token_acc": 0.8723404255319149, "train_speed(iter/s)": 1.538393 }, { "epoch": 0.04284306584979221, "eval_loss": 2.5639302730560303, "eval_runtime": 11.3485, "eval_samples_per_second": 8.812, "eval_steps_per_second": 8.812, "eval_token_acc": 0.41347626339969373, "step": 1000 }, { "epoch": 0.043057281179041174, "grad_norm": 3.444283962249756, "learning_rate": 9.998170356559125e-05, "loss": 0.5242456912994384, "memory(GiB)": 51.38, "step": 1005, "token_acc": 0.5530546623794212, "train_speed(iter/s)": 1.510343 }, { "epoch": 0.04327149650829013, "grad_norm": 2.8902993202209473, "learning_rate": 9.99815210698893e-05, "loss": 0.45065760612487793, "memory(GiB)": 51.38, "step": 1010, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.510987 }, { "epoch": 0.0434857118375391, "grad_norm": 1.7290685176849365, "learning_rate": 9.99813376687252e-05, "loss": 0.3726775884628296, "memory(GiB)": 51.38, "step": 1015, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.511306 }, { "epoch": 0.043699927166788055, "grad_norm": 10.609648704528809, "learning_rate": 9.998115336210229e-05, "loss": 0.37380251884460447, "memory(GiB)": 51.38, "step": 1020, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.512109 }, { "epoch": 0.04391414249603701, "grad_norm": 2.4873204231262207, "learning_rate": 9.998096815002391e-05, "loss": 0.5291335582733154, "memory(GiB)": 51.38, "step": 1025, "token_acc": 0.8952702702702703, "train_speed(iter/s)": 1.512907 }, { "epoch": 0.04412835782528598, "grad_norm": 9.017045974731445, "learning_rate": 9.998078203249341e-05, "loss": 0.5442335605621338, "memory(GiB)": 51.38, "step": 1030, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.513625 }, { "epoch": 0.04434257315453494, "grad_norm": 1.3961349725723267, "learning_rate": 9.998059500951415e-05, "loss": 0.45108256340026853, "memory(GiB)": 51.38, "step": 1035, "token_acc": 0.9116719242902208, "train_speed(iter/s)": 1.513298 }, { "epoch": 0.0445567884837839, "grad_norm": 0.7750174403190613, "learning_rate": 9.998040708108953e-05, "loss": 0.31025550365447996, "memory(GiB)": 51.38, "step": 1040, "token_acc": 0.9447852760736196, "train_speed(iter/s)": 1.51326 }, { "epoch": 0.04477100381303286, "grad_norm": 4.272421360015869, "learning_rate": 9.998021824722295e-05, "loss": 0.5295130252838135, "memory(GiB)": 51.38, "step": 1045, "token_acc": 0.874251497005988, "train_speed(iter/s)": 1.513447 }, { "epoch": 0.044985219142281825, "grad_norm": 2.995204210281372, "learning_rate": 9.998002850791782e-05, "loss": 0.5304511547088623, "memory(GiB)": 51.38, "step": 1050, "token_acc": 0.8664495114006515, "train_speed(iter/s)": 1.513135 }, { "epoch": 0.04519943447153078, "grad_norm": 1.2003620862960815, "learning_rate": 9.99798378631776e-05, "loss": 0.18754096031188966, "memory(GiB)": 51.38, "step": 1055, "token_acc": 0.9590443686006825, "train_speed(iter/s)": 1.513315 }, { "epoch": 0.04541364980077974, "grad_norm": 0.8188661932945251, "learning_rate": 9.997964631300573e-05, "loss": 0.3413902997970581, "memory(GiB)": 51.38, "step": 1060, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.513143 }, { "epoch": 0.04562786513002871, "grad_norm": 1.6876928806304932, "learning_rate": 9.997945385740568e-05, "loss": 0.3495121240615845, "memory(GiB)": 51.38, "step": 1065, "token_acc": 0.92, "train_speed(iter/s)": 1.513539 }, { "epoch": 0.045842080459277665, "grad_norm": 1.2105671167373657, "learning_rate": 9.997926049638094e-05, "loss": 0.23825652599334718, "memory(GiB)": 51.38, "step": 1070, "token_acc": 0.9543726235741445, "train_speed(iter/s)": 1.514577 }, { "epoch": 0.04605629578852663, "grad_norm": 7.211338043212891, "learning_rate": 9.9979066229935e-05, "loss": 0.42479524612426756, "memory(GiB)": 51.38, "step": 1075, "token_acc": 0.8953488372093024, "train_speed(iter/s)": 1.51458 }, { "epoch": 0.04627051111777559, "grad_norm": 2.2632038593292236, "learning_rate": 9.997887105807141e-05, "loss": 0.3877760171890259, "memory(GiB)": 51.38, "step": 1080, "token_acc": 0.9216300940438872, "train_speed(iter/s)": 1.514682 }, { "epoch": 0.046484726447024546, "grad_norm": 2.8610596656799316, "learning_rate": 9.997867498079368e-05, "loss": 0.28857877254486086, "memory(GiB)": 51.38, "step": 1085, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.514415 }, { "epoch": 0.04669894177627351, "grad_norm": 1.3112672567367554, "learning_rate": 9.997847799810537e-05, "loss": 0.4946000576019287, "memory(GiB)": 51.38, "step": 1090, "token_acc": 0.8867924528301887, "train_speed(iter/s)": 1.513892 }, { "epoch": 0.04691315710552247, "grad_norm": 2.83030104637146, "learning_rate": 9.997828011001005e-05, "loss": 0.38821485042572024, "memory(GiB)": 51.38, "step": 1095, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.513706 }, { "epoch": 0.047127372434771435, "grad_norm": 3.2246286869049072, "learning_rate": 9.99780813165113e-05, "loss": 0.2919719696044922, "memory(GiB)": 51.38, "step": 1100, "token_acc": 0.9438943894389439, "train_speed(iter/s)": 1.513807 }, { "epoch": 0.04734158776402039, "grad_norm": 1.0547010898590088, "learning_rate": 9.997788161761272e-05, "loss": 0.3507228374481201, "memory(GiB)": 51.38, "step": 1105, "token_acc": 0.9455782312925171, "train_speed(iter/s)": 1.51463 }, { "epoch": 0.04755580309326935, "grad_norm": 1.1477594375610352, "learning_rate": 9.997768101331793e-05, "loss": 0.2699378252029419, "memory(GiB)": 51.38, "step": 1110, "token_acc": 0.9387096774193548, "train_speed(iter/s)": 1.515017 }, { "epoch": 0.047770018422518316, "grad_norm": 1.6579521894454956, "learning_rate": 9.997747950363056e-05, "loss": 0.350109338760376, "memory(GiB)": 51.38, "step": 1115, "token_acc": 0.9196428571428571, "train_speed(iter/s)": 1.514926 }, { "epoch": 0.047984233751767275, "grad_norm": 2.149534225463867, "learning_rate": 9.997727708855429e-05, "loss": 0.3524376153945923, "memory(GiB)": 51.38, "step": 1120, "token_acc": 0.9132231404958677, "train_speed(iter/s)": 1.516448 }, { "epoch": 0.04819844908101624, "grad_norm": 2.5158627033233643, "learning_rate": 9.997707376809273e-05, "loss": 0.42693653106689455, "memory(GiB)": 51.38, "step": 1125, "token_acc": 0.8966789667896679, "train_speed(iter/s)": 1.517175 }, { "epoch": 0.0484126644102652, "grad_norm": 5.94671630859375, "learning_rate": 9.997686954224963e-05, "loss": 0.33208017349243163, "memory(GiB)": 51.38, "step": 1130, "token_acc": 0.8864468864468864, "train_speed(iter/s)": 1.517158 }, { "epoch": 0.04862687973951416, "grad_norm": 2.741412878036499, "learning_rate": 9.997666441102864e-05, "loss": 0.47496490478515624, "memory(GiB)": 51.38, "step": 1135, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.517172 }, { "epoch": 0.04884109506876312, "grad_norm": 2.6481494903564453, "learning_rate": 9.99764583744335e-05, "loss": 0.3166871309280396, "memory(GiB)": 51.38, "step": 1140, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.517129 }, { "epoch": 0.04905531039801208, "grad_norm": 2.980315685272217, "learning_rate": 9.997625143246791e-05, "loss": 0.44851255416870117, "memory(GiB)": 51.38, "step": 1145, "token_acc": 0.888268156424581, "train_speed(iter/s)": 1.517136 }, { "epoch": 0.049269525727261045, "grad_norm": 2.0442922115325928, "learning_rate": 9.997604358513566e-05, "loss": 0.3007690906524658, "memory(GiB)": 51.38, "step": 1150, "token_acc": 0.9176829268292683, "train_speed(iter/s)": 1.517127 }, { "epoch": 0.04948374105651, "grad_norm": 2.4869332313537598, "learning_rate": 9.997583483244051e-05, "loss": 0.36042160987854005, "memory(GiB)": 51.38, "step": 1155, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.516898 }, { "epoch": 0.04969795638575897, "grad_norm": 2.435574531555176, "learning_rate": 9.997562517438621e-05, "loss": 0.5172206878662109, "memory(GiB)": 51.38, "step": 1160, "token_acc": 0.8896797153024911, "train_speed(iter/s)": 1.516905 }, { "epoch": 0.049912171715007926, "grad_norm": 1.7027918100357056, "learning_rate": 9.997541461097659e-05, "loss": 0.2353591203689575, "memory(GiB)": 51.38, "step": 1165, "token_acc": 0.9543726235741445, "train_speed(iter/s)": 1.516868 }, { "epoch": 0.050126387044256884, "grad_norm": 3.160538673400879, "learning_rate": 9.997520314221547e-05, "loss": 0.7534217357635498, "memory(GiB)": 51.38, "step": 1170, "token_acc": 0.8269230769230769, "train_speed(iter/s)": 1.516644 }, { "epoch": 0.05034060237350585, "grad_norm": 1.4558066129684448, "learning_rate": 9.997499076810664e-05, "loss": 0.4312772750854492, "memory(GiB)": 51.38, "step": 1175, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.516551 }, { "epoch": 0.05055481770275481, "grad_norm": 1.4315602779388428, "learning_rate": 9.997477748865399e-05, "loss": 0.2778109788894653, "memory(GiB)": 51.38, "step": 1180, "token_acc": 0.9471698113207547, "train_speed(iter/s)": 1.516573 }, { "epoch": 0.05076903303200377, "grad_norm": 1.6092584133148193, "learning_rate": 9.997456330386135e-05, "loss": 0.20491538047790528, "memory(GiB)": 51.38, "step": 1185, "token_acc": 0.959349593495935, "train_speed(iter/s)": 1.516727 }, { "epoch": 0.05098324836125273, "grad_norm": 8.878623008728027, "learning_rate": 9.997434821373262e-05, "loss": 0.509013032913208, "memory(GiB)": 51.38, "step": 1190, "token_acc": 0.8806584362139918, "train_speed(iter/s)": 1.516521 }, { "epoch": 0.05119746369050169, "grad_norm": 3.344947338104248, "learning_rate": 9.997413221827169e-05, "loss": 0.4472821712493896, "memory(GiB)": 51.38, "step": 1195, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.516518 }, { "epoch": 0.051411679019750654, "grad_norm": 0.34666386246681213, "learning_rate": 9.997391531748248e-05, "loss": 0.4593306541442871, "memory(GiB)": 51.38, "step": 1200, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.517171 }, { "epoch": 0.05162589434899961, "grad_norm": 2.0713672637939453, "learning_rate": 9.997369751136891e-05, "loss": 0.5158166408538818, "memory(GiB)": 51.38, "step": 1205, "token_acc": 0.8837209302325582, "train_speed(iter/s)": 1.517103 }, { "epoch": 0.05184010967824858, "grad_norm": 2.8909451961517334, "learning_rate": 9.997347879993495e-05, "loss": 0.5558267116546631, "memory(GiB)": 51.38, "step": 1210, "token_acc": 0.8827160493827161, "train_speed(iter/s)": 1.516776 }, { "epoch": 0.052054325007497536, "grad_norm": 1.9511113166809082, "learning_rate": 9.997325918318452e-05, "loss": 0.3545339345932007, "memory(GiB)": 51.38, "step": 1215, "token_acc": 0.9346153846153846, "train_speed(iter/s)": 1.516667 }, { "epoch": 0.0522685403367465, "grad_norm": 2.092160224914551, "learning_rate": 9.997303866112163e-05, "loss": 0.35593552589416505, "memory(GiB)": 51.38, "step": 1220, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.516334 }, { "epoch": 0.05248275566599546, "grad_norm": 1.5525164604187012, "learning_rate": 9.997281723375027e-05, "loss": 0.3682552814483643, "memory(GiB)": 51.38, "step": 1225, "token_acc": 0.8964401294498382, "train_speed(iter/s)": 1.516233 }, { "epoch": 0.05269697099524442, "grad_norm": 2.2168424129486084, "learning_rate": 9.997259490107444e-05, "loss": 0.20818624496459961, "memory(GiB)": 51.38, "step": 1230, "token_acc": 0.9580152671755725, "train_speed(iter/s)": 1.516862 }, { "epoch": 0.05291118632449338, "grad_norm": 1.4821821451187134, "learning_rate": 9.997237166309818e-05, "loss": 0.330265736579895, "memory(GiB)": 51.38, "step": 1235, "token_acc": 0.9184782608695652, "train_speed(iter/s)": 1.516919 }, { "epoch": 0.05312540165374234, "grad_norm": 2.0231335163116455, "learning_rate": 9.997214751982552e-05, "loss": 0.36339459419250486, "memory(GiB)": 51.38, "step": 1240, "token_acc": 0.8934707903780069, "train_speed(iter/s)": 1.516675 }, { "epoch": 0.053339616982991306, "grad_norm": 1.8057479858398438, "learning_rate": 9.997192247126055e-05, "loss": 0.3743975877761841, "memory(GiB)": 51.38, "step": 1245, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.516694 }, { "epoch": 0.053553832312240264, "grad_norm": 1.090306043624878, "learning_rate": 9.99716965174073e-05, "loss": 0.36141095161437986, "memory(GiB)": 51.38, "step": 1250, "token_acc": 0.9114391143911439, "train_speed(iter/s)": 1.516947 }, { "epoch": 0.05376804764148922, "grad_norm": 7.8877058029174805, "learning_rate": 9.997146965826991e-05, "loss": 0.48181495666503904, "memory(GiB)": 51.38, "step": 1255, "token_acc": 0.8859060402684564, "train_speed(iter/s)": 1.517056 }, { "epoch": 0.05398226297073819, "grad_norm": 1.3813396692276, "learning_rate": 9.997124189385246e-05, "loss": 0.5706547737121582, "memory(GiB)": 51.38, "step": 1260, "token_acc": 0.8902077151335311, "train_speed(iter/s)": 1.517625 }, { "epoch": 0.054196478299987146, "grad_norm": 1.9278581142425537, "learning_rate": 9.99710132241591e-05, "loss": 0.35936579704284666, "memory(GiB)": 51.38, "step": 1265, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.517747 }, { "epoch": 0.05441069362923611, "grad_norm": 1.1481678485870361, "learning_rate": 9.997078364919395e-05, "loss": 0.2735938310623169, "memory(GiB)": 51.38, "step": 1270, "token_acc": 0.9376947040498442, "train_speed(iter/s)": 1.517946 }, { "epoch": 0.05462490895848507, "grad_norm": 4.700014591217041, "learning_rate": 9.997055316896118e-05, "loss": 0.6771193027496338, "memory(GiB)": 51.38, "step": 1275, "token_acc": 0.8317460317460318, "train_speed(iter/s)": 1.518431 }, { "epoch": 0.05483912428773403, "grad_norm": 3.1703174114227295, "learning_rate": 9.997032178346495e-05, "loss": 0.6109764099121093, "memory(GiB)": 51.38, "step": 1280, "token_acc": 0.8733333333333333, "train_speed(iter/s)": 1.518472 }, { "epoch": 0.05505333961698299, "grad_norm": 2.3404700756073, "learning_rate": 9.997008949270947e-05, "loss": 0.297015643119812, "memory(GiB)": 51.38, "step": 1285, "token_acc": 0.9522058823529411, "train_speed(iter/s)": 1.518602 }, { "epoch": 0.05526755494623195, "grad_norm": 4.2131147384643555, "learning_rate": 9.996985629669894e-05, "loss": 0.3699419736862183, "memory(GiB)": 51.38, "step": 1290, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.51856 }, { "epoch": 0.055481770275480916, "grad_norm": 2.8389170169830322, "learning_rate": 9.996962219543762e-05, "loss": 0.5456232070922852, "memory(GiB)": 51.38, "step": 1295, "token_acc": 0.8678678678678678, "train_speed(iter/s)": 1.518415 }, { "epoch": 0.055695985604729874, "grad_norm": 2.318523406982422, "learning_rate": 9.996938718892969e-05, "loss": 0.726338005065918, "memory(GiB)": 51.38, "step": 1300, "token_acc": 0.8176470588235294, "train_speed(iter/s)": 1.518156 }, { "epoch": 0.05591020093397884, "grad_norm": 1.733306884765625, "learning_rate": 9.996915127717944e-05, "loss": 0.32180898189544677, "memory(GiB)": 51.38, "step": 1305, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.517906 }, { "epoch": 0.0561244162632278, "grad_norm": 0.3320750892162323, "learning_rate": 9.996891446019114e-05, "loss": 0.2527446746826172, "memory(GiB)": 51.38, "step": 1310, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.517771 }, { "epoch": 0.056338631592476755, "grad_norm": 2.064291000366211, "learning_rate": 9.996867673796908e-05, "loss": 0.4167318820953369, "memory(GiB)": 51.38, "step": 1315, "token_acc": 0.9, "train_speed(iter/s)": 1.518003 }, { "epoch": 0.05655284692172572, "grad_norm": 3.971425771713257, "learning_rate": 9.996843811051757e-05, "loss": 0.5044966697692871, "memory(GiB)": 51.38, "step": 1320, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.517826 }, { "epoch": 0.05676706225097468, "grad_norm": 3.699244499206543, "learning_rate": 9.996819857784093e-05, "loss": 0.48551130294799805, "memory(GiB)": 51.38, "step": 1325, "token_acc": 0.8914728682170543, "train_speed(iter/s)": 1.51909 }, { "epoch": 0.056981277580223644, "grad_norm": 2.5205869674682617, "learning_rate": 9.996795813994348e-05, "loss": 0.30181887149810793, "memory(GiB)": 51.38, "step": 1330, "token_acc": 0.9291338582677166, "train_speed(iter/s)": 1.519103 }, { "epoch": 0.0571954929094726, "grad_norm": 1.52571439743042, "learning_rate": 9.996771679682961e-05, "loss": 0.3247175455093384, "memory(GiB)": 51.38, "step": 1335, "token_acc": 0.9299363057324841, "train_speed(iter/s)": 1.519235 }, { "epoch": 0.05740970823872156, "grad_norm": 2.56339693069458, "learning_rate": 9.996747454850368e-05, "loss": 0.20460102558135987, "memory(GiB)": 51.38, "step": 1340, "token_acc": 0.9638157894736842, "train_speed(iter/s)": 1.519522 }, { "epoch": 0.057623923567970525, "grad_norm": 2.294597625732422, "learning_rate": 9.996723139497008e-05, "loss": 0.3838940143585205, "memory(GiB)": 51.38, "step": 1345, "token_acc": 0.9134615384615384, "train_speed(iter/s)": 1.520011 }, { "epoch": 0.057838138897219483, "grad_norm": 1.980448842048645, "learning_rate": 9.99669873362332e-05, "loss": 0.5758762359619141, "memory(GiB)": 51.38, "step": 1350, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.5202 }, { "epoch": 0.05805235422646845, "grad_norm": 2.297365188598633, "learning_rate": 9.996674237229748e-05, "loss": 0.3962706565856934, "memory(GiB)": 51.38, "step": 1355, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.520016 }, { "epoch": 0.05826656955571741, "grad_norm": 1.7686997652053833, "learning_rate": 9.996649650316734e-05, "loss": 0.5702567577362061, "memory(GiB)": 51.38, "step": 1360, "token_acc": 0.8688524590163934, "train_speed(iter/s)": 1.521128 }, { "epoch": 0.058480784884966365, "grad_norm": 2.8843319416046143, "learning_rate": 9.996624972884725e-05, "loss": 0.2401970386505127, "memory(GiB)": 51.38, "step": 1365, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.521219 }, { "epoch": 0.05869500021421533, "grad_norm": 1.9975236654281616, "learning_rate": 9.996600204934165e-05, "loss": 0.9091359138488769, "memory(GiB)": 51.38, "step": 1370, "token_acc": 0.8115015974440895, "train_speed(iter/s)": 1.521695 }, { "epoch": 0.05890921554346429, "grad_norm": 0.9617891907691956, "learning_rate": 9.996575346465508e-05, "loss": 0.23971343040466309, "memory(GiB)": 51.38, "step": 1375, "token_acc": 0.9605734767025089, "train_speed(iter/s)": 1.521473 }, { "epoch": 0.059123430872713253, "grad_norm": 2.983628988265991, "learning_rate": 9.9965503974792e-05, "loss": 0.233282208442688, "memory(GiB)": 51.38, "step": 1380, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.52146 }, { "epoch": 0.05933764620196221, "grad_norm": 3.085155725479126, "learning_rate": 9.996525357975697e-05, "loss": 0.37925961017608645, "memory(GiB)": 51.38, "step": 1385, "token_acc": 0.9073482428115016, "train_speed(iter/s)": 1.521391 }, { "epoch": 0.05955186153121118, "grad_norm": 0.47211727499961853, "learning_rate": 9.996500227955447e-05, "loss": 0.19322605133056642, "memory(GiB)": 51.38, "step": 1390, "token_acc": 0.9606557377049181, "train_speed(iter/s)": 1.521499 }, { "epoch": 0.059766076860460135, "grad_norm": 1.674572467803955, "learning_rate": 9.99647500741891e-05, "loss": 0.4049801826477051, "memory(GiB)": 51.38, "step": 1395, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.521672 }, { "epoch": 0.05998029218970909, "grad_norm": 3.302664279937744, "learning_rate": 9.99644969636654e-05, "loss": 0.6448951244354248, "memory(GiB)": 51.38, "step": 1400, "token_acc": 0.8580645161290322, "train_speed(iter/s)": 1.521502 }, { "epoch": 0.06019450751895806, "grad_norm": 0.8046497702598572, "learning_rate": 9.996424294798796e-05, "loss": 0.37420654296875, "memory(GiB)": 51.38, "step": 1405, "token_acc": 0.9077490774907749, "train_speed(iter/s)": 1.52237 }, { "epoch": 0.06040872284820702, "grad_norm": 1.8491744995117188, "learning_rate": 9.99639880271614e-05, "loss": 0.4521156311035156, "memory(GiB)": 51.38, "step": 1410, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.522562 }, { "epoch": 0.06062293817745598, "grad_norm": 3.1976118087768555, "learning_rate": 9.996373220119034e-05, "loss": 0.454149055480957, "memory(GiB)": 51.38, "step": 1415, "token_acc": 0.898989898989899, "train_speed(iter/s)": 1.52216 }, { "epoch": 0.06083715350670494, "grad_norm": 4.897167682647705, "learning_rate": 9.996347547007938e-05, "loss": 0.5430358409881592, "memory(GiB)": 51.38, "step": 1420, "token_acc": 0.870722433460076, "train_speed(iter/s)": 1.522393 }, { "epoch": 0.0610513688359539, "grad_norm": 2.6525139808654785, "learning_rate": 9.99632178338332e-05, "loss": 0.6310051918029785, "memory(GiB)": 51.38, "step": 1425, "token_acc": 0.8692579505300353, "train_speed(iter/s)": 1.524007 }, { "epoch": 0.06126558416520286, "grad_norm": 4.990790367126465, "learning_rate": 9.996295929245647e-05, "loss": 0.42801456451416015, "memory(GiB)": 51.38, "step": 1430, "token_acc": 0.9212598425196851, "train_speed(iter/s)": 1.524911 }, { "epoch": 0.06147979949445182, "grad_norm": 15.687381744384766, "learning_rate": 9.996269984595386e-05, "loss": 0.4732830047607422, "memory(GiB)": 51.38, "step": 1435, "token_acc": 0.89419795221843, "train_speed(iter/s)": 1.525455 }, { "epoch": 0.06169401482370079, "grad_norm": 2.1707847118377686, "learning_rate": 9.996243949433006e-05, "loss": 0.4140774250030518, "memory(GiB)": 51.38, "step": 1440, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.525211 }, { "epoch": 0.061908230152949745, "grad_norm": 2.256479024887085, "learning_rate": 9.996217823758981e-05, "loss": 0.7792396545410156, "memory(GiB)": 51.38, "step": 1445, "token_acc": 0.8466257668711656, "train_speed(iter/s)": 1.524942 }, { "epoch": 0.0621224454821987, "grad_norm": 3.6348414421081543, "learning_rate": 9.996191607573782e-05, "loss": 0.4104060173034668, "memory(GiB)": 51.38, "step": 1450, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.524937 }, { "epoch": 0.06233666081144767, "grad_norm": 6.284909248352051, "learning_rate": 9.996165300877886e-05, "loss": 0.3679136514663696, "memory(GiB)": 51.38, "step": 1455, "token_acc": 0.8955696202531646, "train_speed(iter/s)": 1.52487 }, { "epoch": 0.06255087614069663, "grad_norm": 6.9890642166137695, "learning_rate": 9.99613890367177e-05, "loss": 0.3739237070083618, "memory(GiB)": 51.38, "step": 1460, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.524622 }, { "epoch": 0.06276509146994559, "grad_norm": 1.7847683429718018, "learning_rate": 9.99611241595591e-05, "loss": 0.36158716678619385, "memory(GiB)": 51.38, "step": 1465, "token_acc": 0.9154518950437318, "train_speed(iter/s)": 1.524604 }, { "epoch": 0.06297930679919456, "grad_norm": 1.2134177684783936, "learning_rate": 9.996085837730787e-05, "loss": 0.13551115989685059, "memory(GiB)": 51.38, "step": 1470, "token_acc": 0.975177304964539, "train_speed(iter/s)": 1.524672 }, { "epoch": 0.06319352212844351, "grad_norm": 1.8310520648956299, "learning_rate": 9.996059168996883e-05, "loss": 0.2702314853668213, "memory(GiB)": 51.38, "step": 1475, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.524726 }, { "epoch": 0.06340773745769247, "grad_norm": 1.1094238758087158, "learning_rate": 9.996032409754679e-05, "loss": 0.25986108779907224, "memory(GiB)": 51.38, "step": 1480, "token_acc": 0.9251968503937008, "train_speed(iter/s)": 1.5248 }, { "epoch": 0.06362195278694144, "grad_norm": 3.232971668243408, "learning_rate": 9.996005560004662e-05, "loss": 0.3610643148422241, "memory(GiB)": 51.38, "step": 1485, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.525003 }, { "epoch": 0.06383616811619039, "grad_norm": 1.139391303062439, "learning_rate": 9.995978619747317e-05, "loss": 0.3275965929031372, "memory(GiB)": 51.38, "step": 1490, "token_acc": 0.9052287581699346, "train_speed(iter/s)": 1.524837 }, { "epoch": 0.06405038344543935, "grad_norm": 5.002138614654541, "learning_rate": 9.995951588983135e-05, "loss": 0.3748301029205322, "memory(GiB)": 51.38, "step": 1495, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.524804 }, { "epoch": 0.06426459877468832, "grad_norm": 1.408638834953308, "learning_rate": 9.995924467712601e-05, "loss": 0.47899637222290037, "memory(GiB)": 51.38, "step": 1500, "token_acc": 0.8906882591093117, "train_speed(iter/s)": 1.525231 }, { "epoch": 0.06426459877468832, "eval_loss": 2.458785057067871, "eval_runtime": 11.466, "eval_samples_per_second": 8.721, "eval_steps_per_second": 8.721, "eval_token_acc": 0.4444444444444444, "step": 1500 }, { "epoch": 0.06447881410393727, "grad_norm": 5.9918212890625, "learning_rate": 9.995897255936209e-05, "loss": 0.4755388736724854, "memory(GiB)": 51.38, "step": 1505, "token_acc": 0.5711592836946278, "train_speed(iter/s)": 1.506281 }, { "epoch": 0.06469302943318624, "grad_norm": 1.8650624752044678, "learning_rate": 9.995869953654452e-05, "loss": 0.5633135318756104, "memory(GiB)": 51.38, "step": 1510, "token_acc": 0.8761904761904762, "train_speed(iter/s)": 1.506264 }, { "epoch": 0.0649072447624352, "grad_norm": 2.417308807373047, "learning_rate": 9.995842560867826e-05, "loss": 0.4304789066314697, "memory(GiB)": 51.38, "step": 1515, "token_acc": 0.8949044585987261, "train_speed(iter/s)": 1.506243 }, { "epoch": 0.06512146009168417, "grad_norm": 1.6686387062072754, "learning_rate": 9.995815077576822e-05, "loss": 0.5134516716003418, "memory(GiB)": 51.38, "step": 1520, "token_acc": 0.88, "train_speed(iter/s)": 1.506452 }, { "epoch": 0.06533567542093312, "grad_norm": 1.8646697998046875, "learning_rate": 9.995787503781944e-05, "loss": 0.46658010482788087, "memory(GiB)": 51.38, "step": 1525, "token_acc": 0.88671875, "train_speed(iter/s)": 1.506583 }, { "epoch": 0.06554989075018208, "grad_norm": 2.853902578353882, "learning_rate": 9.995759839483687e-05, "loss": 0.5853761672973633, "memory(GiB)": 51.38, "step": 1530, "token_acc": 0.8825757575757576, "train_speed(iter/s)": 1.506613 }, { "epoch": 0.06576410607943105, "grad_norm": 2.231388807296753, "learning_rate": 9.995732084682557e-05, "loss": 0.48647642135620117, "memory(GiB)": 51.38, "step": 1535, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.507245 }, { "epoch": 0.06597832140868, "grad_norm": 1.7234394550323486, "learning_rate": 9.995704239379051e-05, "loss": 0.30050714015960694, "memory(GiB)": 51.38, "step": 1540, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.507184 }, { "epoch": 0.06619253673792896, "grad_norm": 6.198657989501953, "learning_rate": 9.995676303573678e-05, "loss": 0.48572711944580077, "memory(GiB)": 51.38, "step": 1545, "token_acc": 0.9238095238095239, "train_speed(iter/s)": 1.507765 }, { "epoch": 0.06640675206717793, "grad_norm": 3.2689449787139893, "learning_rate": 9.995648277266942e-05, "loss": 0.44418816566467284, "memory(GiB)": 51.38, "step": 1550, "token_acc": 0.9064516129032258, "train_speed(iter/s)": 1.508001 }, { "epoch": 0.0666209673964269, "grad_norm": 3.0660290718078613, "learning_rate": 9.995620160459351e-05, "loss": 0.5430166244506835, "memory(GiB)": 51.38, "step": 1555, "token_acc": 0.8611111111111112, "train_speed(iter/s)": 1.508214 }, { "epoch": 0.06683518272567585, "grad_norm": 3.0595614910125732, "learning_rate": 9.995591953151415e-05, "loss": 0.26342718601226806, "memory(GiB)": 51.38, "step": 1560, "token_acc": 0.9495798319327731, "train_speed(iter/s)": 1.508262 }, { "epoch": 0.06704939805492481, "grad_norm": 1.744653344154358, "learning_rate": 9.995563655343645e-05, "loss": 0.20707738399505615, "memory(GiB)": 51.38, "step": 1565, "token_acc": 0.9407114624505929, "train_speed(iter/s)": 1.508451 }, { "epoch": 0.06726361338417378, "grad_norm": 1.7197892665863037, "learning_rate": 9.995535267036551e-05, "loss": 0.5075142860412598, "memory(GiB)": 51.38, "step": 1570, "token_acc": 0.9, "train_speed(iter/s)": 1.508742 }, { "epoch": 0.06747782871342273, "grad_norm": 1.5631802082061768, "learning_rate": 9.995506788230652e-05, "loss": 0.21766080856323242, "memory(GiB)": 51.38, "step": 1575, "token_acc": 0.9449838187702265, "train_speed(iter/s)": 1.509162 }, { "epoch": 0.06769204404267169, "grad_norm": 2.897325038909912, "learning_rate": 9.99547821892646e-05, "loss": 0.38132286071777344, "memory(GiB)": 51.38, "step": 1580, "token_acc": 0.8971061093247589, "train_speed(iter/s)": 1.50918 }, { "epoch": 0.06790625937192066, "grad_norm": 9.3455228805542, "learning_rate": 9.995449559124495e-05, "loss": 0.43044614791870117, "memory(GiB)": 51.38, "step": 1585, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.508915 }, { "epoch": 0.06812047470116961, "grad_norm": 1.4786064624786377, "learning_rate": 9.995420808825274e-05, "loss": 0.2306354522705078, "memory(GiB)": 51.38, "step": 1590, "token_acc": 0.9528985507246377, "train_speed(iter/s)": 1.508495 }, { "epoch": 0.06833469003041857, "grad_norm": 2.4360625743865967, "learning_rate": 9.995391968029318e-05, "loss": 0.5117662429809571, "memory(GiB)": 51.38, "step": 1595, "token_acc": 0.8714733542319749, "train_speed(iter/s)": 1.508606 }, { "epoch": 0.06854890535966754, "grad_norm": 3.7222166061401367, "learning_rate": 9.995363036737152e-05, "loss": 0.40784215927124023, "memory(GiB)": 51.38, "step": 1600, "token_acc": 0.8911290322580645, "train_speed(iter/s)": 1.509014 }, { "epoch": 0.0687631206889165, "grad_norm": 3.1575143337249756, "learning_rate": 9.995334014949297e-05, "loss": 0.5195985794067383, "memory(GiB)": 51.38, "step": 1605, "token_acc": 0.851145038167939, "train_speed(iter/s)": 1.509073 }, { "epoch": 0.06897733601816546, "grad_norm": 1.0886998176574707, "learning_rate": 9.995304902666283e-05, "loss": 0.3257907390594482, "memory(GiB)": 51.38, "step": 1610, "token_acc": 0.9296636085626911, "train_speed(iter/s)": 1.509478 }, { "epoch": 0.06919155134741442, "grad_norm": 2.6948862075805664, "learning_rate": 9.995275699888633e-05, "loss": 0.4741551876068115, "memory(GiB)": 51.38, "step": 1615, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.509345 }, { "epoch": 0.06940576667666339, "grad_norm": 8.57968807220459, "learning_rate": 9.995246406616879e-05, "loss": 0.4293004035949707, "memory(GiB)": 51.38, "step": 1620, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.50991 }, { "epoch": 0.06961998200591234, "grad_norm": 3.5558295249938965, "learning_rate": 9.995217022851548e-05, "loss": 0.5159621238708496, "memory(GiB)": 51.38, "step": 1625, "token_acc": 0.9045936395759717, "train_speed(iter/s)": 1.509928 }, { "epoch": 0.0698341973351613, "grad_norm": 1.4698773622512817, "learning_rate": 9.995187548593176e-05, "loss": 0.2708472013473511, "memory(GiB)": 51.38, "step": 1630, "token_acc": 0.937125748502994, "train_speed(iter/s)": 1.510065 }, { "epoch": 0.07004841266441027, "grad_norm": 5.997922420501709, "learning_rate": 9.995157983842297e-05, "loss": 0.2844212055206299, "memory(GiB)": 51.38, "step": 1635, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.510468 }, { "epoch": 0.07026262799365923, "grad_norm": 2.321852445602417, "learning_rate": 9.995128328599444e-05, "loss": 0.1962536096572876, "memory(GiB)": 51.38, "step": 1640, "token_acc": 0.9602649006622517, "train_speed(iter/s)": 1.510542 }, { "epoch": 0.07047684332290818, "grad_norm": 1.744938850402832, "learning_rate": 9.995098582865157e-05, "loss": 0.5735814094543457, "memory(GiB)": 51.38, "step": 1645, "token_acc": 0.9119718309859155, "train_speed(iter/s)": 1.510522 }, { "epoch": 0.07069105865215715, "grad_norm": 1.280989170074463, "learning_rate": 9.99506874663997e-05, "loss": 0.3865849018096924, "memory(GiB)": 51.38, "step": 1650, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.510791 }, { "epoch": 0.07090527398140611, "grad_norm": 1.4695428609848022, "learning_rate": 9.995038819924429e-05, "loss": 0.4433786392211914, "memory(GiB)": 51.38, "step": 1655, "token_acc": 0.925, "train_speed(iter/s)": 1.510613 }, { "epoch": 0.07111948931065507, "grad_norm": 3.5578935146331787, "learning_rate": 9.995008802719075e-05, "loss": 0.30309362411499025, "memory(GiB)": 51.38, "step": 1660, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.510605 }, { "epoch": 0.07133370463990403, "grad_norm": 1.125504970550537, "learning_rate": 9.99497869502445e-05, "loss": 0.7048148155212403, "memory(GiB)": 51.38, "step": 1665, "token_acc": 0.8827838827838828, "train_speed(iter/s)": 1.510553 }, { "epoch": 0.071547919969153, "grad_norm": 2.987349510192871, "learning_rate": 9.9949484968411e-05, "loss": 0.4896413326263428, "memory(GiB)": 51.38, "step": 1670, "token_acc": 0.9134615384615384, "train_speed(iter/s)": 1.510793 }, { "epoch": 0.07176213529840195, "grad_norm": 4.866359233856201, "learning_rate": 9.994918208169572e-05, "loss": 0.5303282737731934, "memory(GiB)": 51.38, "step": 1675, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.511138 }, { "epoch": 0.07197635062765091, "grad_norm": 1.8487114906311035, "learning_rate": 9.994887829010416e-05, "loss": 0.20537817478179932, "memory(GiB)": 51.38, "step": 1680, "token_acc": 0.9584905660377359, "train_speed(iter/s)": 1.511191 }, { "epoch": 0.07219056595689988, "grad_norm": 1.6131118535995483, "learning_rate": 9.994857359364181e-05, "loss": 0.5045605182647706, "memory(GiB)": 51.38, "step": 1685, "token_acc": 0.8971061093247589, "train_speed(iter/s)": 1.51108 }, { "epoch": 0.07240478128614884, "grad_norm": 5.356106758117676, "learning_rate": 9.99482679923142e-05, "loss": 0.18098180294036864, "memory(GiB)": 51.38, "step": 1690, "token_acc": 0.9589905362776026, "train_speed(iter/s)": 1.510889 }, { "epoch": 0.0726189966153978, "grad_norm": 2.7739832401275635, "learning_rate": 9.994796148612684e-05, "loss": 0.6482490539550781, "memory(GiB)": 51.38, "step": 1695, "token_acc": 0.8773584905660378, "train_speed(iter/s)": 1.510975 }, { "epoch": 0.07283321194464676, "grad_norm": 2.1476213932037354, "learning_rate": 9.994765407508532e-05, "loss": 0.2253375768661499, "memory(GiB)": 51.38, "step": 1700, "token_acc": 0.942652329749104, "train_speed(iter/s)": 1.51093 }, { "epoch": 0.07304742727389572, "grad_norm": 0.5031588673591614, "learning_rate": 9.99473457591952e-05, "loss": 0.5168910980224609, "memory(GiB)": 51.38, "step": 1705, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.510912 }, { "epoch": 0.07326164260314467, "grad_norm": 2.3974556922912598, "learning_rate": 9.994703653846203e-05, "loss": 0.5229966640472412, "memory(GiB)": 51.38, "step": 1710, "token_acc": 0.9006622516556292, "train_speed(iter/s)": 1.511488 }, { "epoch": 0.07347585793239364, "grad_norm": 2.3954434394836426, "learning_rate": 9.994672641289146e-05, "loss": 0.7623198986053467, "memory(GiB)": 51.38, "step": 1715, "token_acc": 0.8791208791208791, "train_speed(iter/s)": 1.512072 }, { "epoch": 0.0736900732616426, "grad_norm": 1.0660268068313599, "learning_rate": 9.994641538248907e-05, "loss": 0.5481812953948975, "memory(GiB)": 51.38, "step": 1720, "token_acc": 0.8758620689655172, "train_speed(iter/s)": 1.512295 }, { "epoch": 0.07390428859089157, "grad_norm": 1.665285587310791, "learning_rate": 9.994610344726055e-05, "loss": 0.2775076627731323, "memory(GiB)": 51.38, "step": 1725, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.512287 }, { "epoch": 0.07411850392014052, "grad_norm": 2.8702712059020996, "learning_rate": 9.994579060721147e-05, "loss": 0.3199134349822998, "memory(GiB)": 51.38, "step": 1730, "token_acc": 0.9051724137931034, "train_speed(iter/s)": 1.512112 }, { "epoch": 0.07433271924938949, "grad_norm": 2.889202356338501, "learning_rate": 9.994547686234755e-05, "loss": 0.5146369934082031, "memory(GiB)": 51.38, "step": 1735, "token_acc": 0.8949416342412452, "train_speed(iter/s)": 1.512193 }, { "epoch": 0.07454693457863845, "grad_norm": 2.725888729095459, "learning_rate": 9.994516221267446e-05, "loss": 0.5483423233032226, "memory(GiB)": 51.38, "step": 1740, "token_acc": 0.8862876254180602, "train_speed(iter/s)": 1.512653 }, { "epoch": 0.0747611499078874, "grad_norm": 0.8897895812988281, "learning_rate": 9.994484665819793e-05, "loss": 0.2682744026184082, "memory(GiB)": 51.38, "step": 1745, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.512607 }, { "epoch": 0.07497536523713637, "grad_norm": 2.2083475589752197, "learning_rate": 9.994453019892364e-05, "loss": 0.48996148109436033, "memory(GiB)": 51.38, "step": 1750, "token_acc": 0.8813559322033898, "train_speed(iter/s)": 1.512926 }, { "epoch": 0.07518958056638533, "grad_norm": 5.652266025543213, "learning_rate": 9.994421283485733e-05, "loss": 0.5376258850097656, "memory(GiB)": 52.69, "step": 1755, "token_acc": 0.891156462585034, "train_speed(iter/s)": 1.513313 }, { "epoch": 0.0754037958956343, "grad_norm": 2.5043323040008545, "learning_rate": 9.994389456600474e-05, "loss": 0.4817495346069336, "memory(GiB)": 52.69, "step": 1760, "token_acc": 0.8835341365461847, "train_speed(iter/s)": 1.513813 }, { "epoch": 0.07561801122488325, "grad_norm": 4.360466003417969, "learning_rate": 9.994357539237166e-05, "loss": 0.46340327262878417, "memory(GiB)": 52.69, "step": 1765, "token_acc": 0.9014598540145985, "train_speed(iter/s)": 1.514111 }, { "epoch": 0.07583222655413221, "grad_norm": 3.356398105621338, "learning_rate": 9.994325531396387e-05, "loss": 0.24489102363586426, "memory(GiB)": 52.69, "step": 1770, "token_acc": 0.9618055555555556, "train_speed(iter/s)": 1.514312 }, { "epoch": 0.07604644188338118, "grad_norm": 4.485963344573975, "learning_rate": 9.994293433078714e-05, "loss": 0.34725489616394045, "memory(GiB)": 52.69, "step": 1775, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.514456 }, { "epoch": 0.07626065721263013, "grad_norm": 3.3638997077941895, "learning_rate": 9.994261244284733e-05, "loss": 0.31113581657409667, "memory(GiB)": 52.69, "step": 1780, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.514429 }, { "epoch": 0.0764748725418791, "grad_norm": 3.9617857933044434, "learning_rate": 9.994228965015022e-05, "loss": 0.4347806453704834, "memory(GiB)": 57.83, "step": 1785, "token_acc": 0.8912280701754386, "train_speed(iter/s)": 1.514216 }, { "epoch": 0.07668908787112806, "grad_norm": 4.323101997375488, "learning_rate": 9.99419659527017e-05, "loss": 0.6595251560211182, "memory(GiB)": 57.83, "step": 1790, "token_acc": 0.867741935483871, "train_speed(iter/s)": 1.514202 }, { "epoch": 0.07690330320037701, "grad_norm": 1.7291053533554077, "learning_rate": 9.994164135050761e-05, "loss": 0.6059975624084473, "memory(GiB)": 57.83, "step": 1795, "token_acc": 0.8788659793814433, "train_speed(iter/s)": 1.514357 }, { "epoch": 0.07711751852962598, "grad_norm": 3.2331202030181885, "learning_rate": 9.994131584357384e-05, "loss": 0.6180335998535156, "memory(GiB)": 57.83, "step": 1800, "token_acc": 0.8861538461538462, "train_speed(iter/s)": 1.514862 }, { "epoch": 0.07733173385887494, "grad_norm": 2.0732340812683105, "learning_rate": 9.994098943190629e-05, "loss": 0.598891019821167, "memory(GiB)": 57.83, "step": 1805, "token_acc": 0.8802395209580839, "train_speed(iter/s)": 1.515055 }, { "epoch": 0.07754594918812391, "grad_norm": 1.7091871500015259, "learning_rate": 9.994066211551085e-05, "loss": 0.2506587266921997, "memory(GiB)": 57.83, "step": 1810, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.514987 }, { "epoch": 0.07776016451737286, "grad_norm": 3.40126633644104, "learning_rate": 9.994033389439348e-05, "loss": 0.4099709987640381, "memory(GiB)": 57.83, "step": 1815, "token_acc": 0.9098039215686274, "train_speed(iter/s)": 1.514911 }, { "epoch": 0.07797437984662182, "grad_norm": 2.9106945991516113, "learning_rate": 9.994000476856011e-05, "loss": 0.5257450580596924, "memory(GiB)": 57.83, "step": 1820, "token_acc": 0.886435331230284, "train_speed(iter/s)": 1.514607 }, { "epoch": 0.07818859517587079, "grad_norm": 5.167994976043701, "learning_rate": 9.993967473801671e-05, "loss": 0.49937734603881834, "memory(GiB)": 57.83, "step": 1825, "token_acc": 0.8858267716535433, "train_speed(iter/s)": 1.514787 }, { "epoch": 0.07840281050511974, "grad_norm": 5.245072364807129, "learning_rate": 9.993934380276926e-05, "loss": 0.4515682220458984, "memory(GiB)": 57.83, "step": 1830, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.51556 }, { "epoch": 0.0786170258343687, "grad_norm": 0.852872371673584, "learning_rate": 9.993901196282374e-05, "loss": 0.3152703046798706, "memory(GiB)": 57.83, "step": 1835, "token_acc": 0.9208333333333333, "train_speed(iter/s)": 1.515643 }, { "epoch": 0.07883124116361767, "grad_norm": 3.0528903007507324, "learning_rate": 9.993867921818619e-05, "loss": 0.4761196613311768, "memory(GiB)": 57.83, "step": 1840, "token_acc": 0.8937007874015748, "train_speed(iter/s)": 1.515642 }, { "epoch": 0.07904545649286664, "grad_norm": 3.452143430709839, "learning_rate": 9.993834556886259e-05, "loss": 0.5900424003601075, "memory(GiB)": 57.83, "step": 1845, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.515572 }, { "epoch": 0.07925967182211559, "grad_norm": 6.209501266479492, "learning_rate": 9.993801101485903e-05, "loss": 0.5946532726287842, "memory(GiB)": 57.83, "step": 1850, "token_acc": 0.8819444444444444, "train_speed(iter/s)": 1.516136 }, { "epoch": 0.07947388715136455, "grad_norm": 4.788073539733887, "learning_rate": 9.993767555618157e-05, "loss": 0.35467915534973143, "memory(GiB)": 57.83, "step": 1855, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.515836 }, { "epoch": 0.07968810248061352, "grad_norm": 7.519866466522217, "learning_rate": 9.993733919283624e-05, "loss": 0.6217827320098877, "memory(GiB)": 57.83, "step": 1860, "token_acc": 0.8855421686746988, "train_speed(iter/s)": 1.5158 }, { "epoch": 0.07990231780986247, "grad_norm": 1.1160533428192139, "learning_rate": 9.993700192482918e-05, "loss": 0.10757068395614625, "memory(GiB)": 57.83, "step": 1865, "token_acc": 0.9748201438848921, "train_speed(iter/s)": 1.516023 }, { "epoch": 0.08011653313911143, "grad_norm": 1.7273410558700562, "learning_rate": 9.99366637521665e-05, "loss": 0.43184590339660645, "memory(GiB)": 57.83, "step": 1870, "token_acc": 0.9030303030303031, "train_speed(iter/s)": 1.516274 }, { "epoch": 0.0803307484683604, "grad_norm": 2.937713861465454, "learning_rate": 9.99363246748543e-05, "loss": 0.32599639892578125, "memory(GiB)": 57.83, "step": 1875, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.516639 }, { "epoch": 0.08054496379760935, "grad_norm": 1.5094424486160278, "learning_rate": 9.993598469289874e-05, "loss": 0.5161242008209228, "memory(GiB)": 57.83, "step": 1880, "token_acc": 0.8850174216027874, "train_speed(iter/s)": 1.51654 }, { "epoch": 0.08075917912685832, "grad_norm": 3.188816785812378, "learning_rate": 9.993564380630595e-05, "loss": 0.49753561019897463, "memory(GiB)": 57.83, "step": 1885, "token_acc": 0.9042904290429042, "train_speed(iter/s)": 1.516695 }, { "epoch": 0.08097339445610728, "grad_norm": 0.977247416973114, "learning_rate": 9.993530201508216e-05, "loss": 0.16397618055343627, "memory(GiB)": 57.83, "step": 1890, "token_acc": 0.9607142857142857, "train_speed(iter/s)": 1.516867 }, { "epoch": 0.08118760978535625, "grad_norm": 3.4290144443511963, "learning_rate": 9.993495931923352e-05, "loss": 0.22274844646453856, "memory(GiB)": 57.83, "step": 1895, "token_acc": 0.9466192170818505, "train_speed(iter/s)": 1.516676 }, { "epoch": 0.0814018251146052, "grad_norm": 1.784302830696106, "learning_rate": 9.993461571876624e-05, "loss": 0.4341409206390381, "memory(GiB)": 57.83, "step": 1900, "token_acc": 0.9109792284866469, "train_speed(iter/s)": 1.516712 }, { "epoch": 0.08161604044385416, "grad_norm": 4.658522129058838, "learning_rate": 9.993427121368656e-05, "loss": 0.23212058544158937, "memory(GiB)": 57.83, "step": 1905, "token_acc": 0.9465408805031447, "train_speed(iter/s)": 1.516708 }, { "epoch": 0.08183025577310313, "grad_norm": 1.0936843156814575, "learning_rate": 9.993392580400072e-05, "loss": 0.47963547706604004, "memory(GiB)": 57.83, "step": 1910, "token_acc": 0.8757763975155279, "train_speed(iter/s)": 1.516817 }, { "epoch": 0.08204447110235208, "grad_norm": 2.5820348262786865, "learning_rate": 9.993357948971496e-05, "loss": 0.4552448749542236, "memory(GiB)": 57.83, "step": 1915, "token_acc": 0.8964285714285715, "train_speed(iter/s)": 1.516672 }, { "epoch": 0.08225868643160104, "grad_norm": 3.126863956451416, "learning_rate": 9.993323227083557e-05, "loss": 0.35478835105895995, "memory(GiB)": 57.83, "step": 1920, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.517165 }, { "epoch": 0.08247290176085001, "grad_norm": 1.8618804216384888, "learning_rate": 9.993288414736885e-05, "loss": 0.21129944324493408, "memory(GiB)": 57.83, "step": 1925, "token_acc": 0.9563492063492064, "train_speed(iter/s)": 1.517506 }, { "epoch": 0.08268711709009897, "grad_norm": 5.763302326202393, "learning_rate": 9.993253511932108e-05, "loss": 0.40717458724975586, "memory(GiB)": 57.83, "step": 1930, "token_acc": 0.8952702702702703, "train_speed(iter/s)": 1.517775 }, { "epoch": 0.08290133241934793, "grad_norm": 3.0122766494750977, "learning_rate": 9.99321851866986e-05, "loss": 0.47696518898010254, "memory(GiB)": 57.83, "step": 1935, "token_acc": 0.892, "train_speed(iter/s)": 1.518296 }, { "epoch": 0.08311554774859689, "grad_norm": 2.215914011001587, "learning_rate": 9.993183434950774e-05, "loss": 0.36210026741027834, "memory(GiB)": 57.83, "step": 1940, "token_acc": 0.926984126984127, "train_speed(iter/s)": 1.518298 }, { "epoch": 0.08332976307784586, "grad_norm": 2.2248759269714355, "learning_rate": 9.993148260775487e-05, "loss": 0.4623382091522217, "memory(GiB)": 57.83, "step": 1945, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.518437 }, { "epoch": 0.08354397840709481, "grad_norm": 2.2277722358703613, "learning_rate": 9.993112996144634e-05, "loss": 0.36342473030090333, "memory(GiB)": 57.83, "step": 1950, "token_acc": 0.9184952978056427, "train_speed(iter/s)": 1.518341 }, { "epoch": 0.08375819373634377, "grad_norm": 4.132850170135498, "learning_rate": 9.993077641058856e-05, "loss": 0.4066637992858887, "memory(GiB)": 57.83, "step": 1955, "token_acc": 0.9172185430463576, "train_speed(iter/s)": 1.518701 }, { "epoch": 0.08397240906559274, "grad_norm": 4.002188205718994, "learning_rate": 9.993042195518793e-05, "loss": 0.2413576602935791, "memory(GiB)": 57.83, "step": 1960, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.51881 }, { "epoch": 0.08418662439484169, "grad_norm": 1.2864103317260742, "learning_rate": 9.993006659525087e-05, "loss": 0.49903020858764646, "memory(GiB)": 57.83, "step": 1965, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.518998 }, { "epoch": 0.08440083972409065, "grad_norm": 1.430816411972046, "learning_rate": 9.99297103307838e-05, "loss": 0.19716956615447997, "memory(GiB)": 57.83, "step": 1970, "token_acc": 0.9629629629629629, "train_speed(iter/s)": 1.519337 }, { "epoch": 0.08461505505333962, "grad_norm": 4.550480365753174, "learning_rate": 9.99293531617932e-05, "loss": 0.2833199977874756, "memory(GiB)": 57.83, "step": 1975, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.519094 }, { "epoch": 0.08482927038258858, "grad_norm": 2.145718574523926, "learning_rate": 9.992899508828553e-05, "loss": 0.19301217794418335, "memory(GiB)": 57.83, "step": 1980, "token_acc": 0.950381679389313, "train_speed(iter/s)": 1.519171 }, { "epoch": 0.08504348571183754, "grad_norm": 2.4572529792785645, "learning_rate": 9.992863611026725e-05, "loss": 0.309356164932251, "memory(GiB)": 57.83, "step": 1985, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.519304 }, { "epoch": 0.0852577010410865, "grad_norm": 3.178694248199463, "learning_rate": 9.992827622774494e-05, "loss": 0.22644588947296143, "memory(GiB)": 57.83, "step": 1990, "token_acc": 0.9294871794871795, "train_speed(iter/s)": 1.519357 }, { "epoch": 0.08547191637033547, "grad_norm": 2.015859842300415, "learning_rate": 9.992791544072504e-05, "loss": 0.241288161277771, "memory(GiB)": 57.83, "step": 1995, "token_acc": 0.946360153256705, "train_speed(iter/s)": 1.51945 }, { "epoch": 0.08568613169958442, "grad_norm": 4.4971513748168945, "learning_rate": 9.992755374921412e-05, "loss": 0.3095445156097412, "memory(GiB)": 57.83, "step": 2000, "token_acc": 0.9416342412451362, "train_speed(iter/s)": 1.519401 }, { "epoch": 0.08568613169958442, "eval_loss": 2.613004684448242, "eval_runtime": 11.7941, "eval_samples_per_second": 8.479, "eval_steps_per_second": 8.479, "eval_token_acc": 0.4307692307692308, "step": 2000 }, { "epoch": 0.08590034702883338, "grad_norm": 3.23629093170166, "learning_rate": 9.992719115321872e-05, "loss": 0.3552790641784668, "memory(GiB)": 57.83, "step": 2005, "token_acc": 0.5737211634904714, "train_speed(iter/s)": 1.505128 }, { "epoch": 0.08611456235808235, "grad_norm": 6.062796115875244, "learning_rate": 9.992682765274543e-05, "loss": 0.38274221420288085, "memory(GiB)": 57.83, "step": 2010, "token_acc": 0.921875, "train_speed(iter/s)": 1.504998 }, { "epoch": 0.08632877768733131, "grad_norm": 3.6300177574157715, "learning_rate": 9.992646324780082e-05, "loss": 0.27340590953826904, "memory(GiB)": 57.83, "step": 2015, "token_acc": 0.9681274900398407, "train_speed(iter/s)": 1.505316 }, { "epoch": 0.08654299301658026, "grad_norm": 11.48849868774414, "learning_rate": 9.992609793839148e-05, "loss": 0.49669971466064455, "memory(GiB)": 57.83, "step": 2020, "token_acc": 0.9094827586206896, "train_speed(iter/s)": 1.505956 }, { "epoch": 0.08675720834582923, "grad_norm": 2.913541078567505, "learning_rate": 9.992573172452406e-05, "loss": 0.3669614553451538, "memory(GiB)": 57.83, "step": 2025, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.506047 }, { "epoch": 0.0869714236750782, "grad_norm": 1.4147032499313354, "learning_rate": 9.992536460620516e-05, "loss": 0.44008736610412597, "memory(GiB)": 57.83, "step": 2030, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.506004 }, { "epoch": 0.08718563900432715, "grad_norm": 0.9683845639228821, "learning_rate": 9.992499658344145e-05, "loss": 0.3149267673492432, "memory(GiB)": 57.83, "step": 2035, "token_acc": 0.9212827988338192, "train_speed(iter/s)": 1.506246 }, { "epoch": 0.08739985433357611, "grad_norm": 1.521776795387268, "learning_rate": 9.99246276562396e-05, "loss": 0.45855345726013186, "memory(GiB)": 57.83, "step": 2040, "token_acc": 0.8885017421602788, "train_speed(iter/s)": 1.506281 }, { "epoch": 0.08761406966282508, "grad_norm": 7.82253360748291, "learning_rate": 9.992425782460628e-05, "loss": 0.7940710544586181, "memory(GiB)": 57.83, "step": 2045, "token_acc": 0.7971014492753623, "train_speed(iter/s)": 1.506563 }, { "epoch": 0.08782828499207403, "grad_norm": 2.2983646392822266, "learning_rate": 9.99238870885482e-05, "loss": 0.37485313415527344, "memory(GiB)": 57.83, "step": 2050, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.506562 }, { "epoch": 0.08804250032132299, "grad_norm": 2.137162208557129, "learning_rate": 9.992351544807208e-05, "loss": 0.656770658493042, "memory(GiB)": 57.83, "step": 2055, "token_acc": 0.8585209003215434, "train_speed(iter/s)": 1.506955 }, { "epoch": 0.08825671565057196, "grad_norm": 0.5717587471008301, "learning_rate": 9.992314290318465e-05, "loss": 0.6279329776763916, "memory(GiB)": 57.83, "step": 2060, "token_acc": 0.8754863813229572, "train_speed(iter/s)": 1.506902 }, { "epoch": 0.08847093097982092, "grad_norm": 2.1988844871520996, "learning_rate": 9.992276945389263e-05, "loss": 0.3494650602340698, "memory(GiB)": 57.83, "step": 2065, "token_acc": 0.9271255060728745, "train_speed(iter/s)": 1.506966 }, { "epoch": 0.08868514630906987, "grad_norm": 2.193023920059204, "learning_rate": 9.992239510020282e-05, "loss": 0.37491090297698976, "memory(GiB)": 57.83, "step": 2070, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.507381 }, { "epoch": 0.08889936163831884, "grad_norm": 2.935596466064453, "learning_rate": 9.9922019842122e-05, "loss": 0.6336065769195557, "memory(GiB)": 57.83, "step": 2075, "token_acc": 0.8517350157728707, "train_speed(iter/s)": 1.507402 }, { "epoch": 0.0891135769675678, "grad_norm": 2.072000741958618, "learning_rate": 9.992164367965696e-05, "loss": 0.2197868824005127, "memory(GiB)": 57.83, "step": 2080, "token_acc": 0.9484978540772532, "train_speed(iter/s)": 1.508086 }, { "epoch": 0.08932779229681675, "grad_norm": 2.5668697357177734, "learning_rate": 9.992126661281452e-05, "loss": 0.5369738578796387, "memory(GiB)": 57.83, "step": 2085, "token_acc": 0.8838028169014085, "train_speed(iter/s)": 1.5082 }, { "epoch": 0.08954200762606572, "grad_norm": 3.954017162322998, "learning_rate": 9.99208886416015e-05, "loss": 0.3172943115234375, "memory(GiB)": 57.83, "step": 2090, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.508848 }, { "epoch": 0.08975622295531469, "grad_norm": 1.311085820198059, "learning_rate": 9.992050976602474e-05, "loss": 0.35313918590545657, "memory(GiB)": 57.83, "step": 2095, "token_acc": 0.9457364341085271, "train_speed(iter/s)": 1.50895 }, { "epoch": 0.08997043828456365, "grad_norm": 4.80161190032959, "learning_rate": 9.992012998609112e-05, "loss": 0.6552850723266601, "memory(GiB)": 57.83, "step": 2100, "token_acc": 0.8407407407407408, "train_speed(iter/s)": 1.509421 }, { "epoch": 0.0901846536138126, "grad_norm": 2.4349679946899414, "learning_rate": 9.991974930180752e-05, "loss": 0.31186347007751464, "memory(GiB)": 57.83, "step": 2105, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.509465 }, { "epoch": 0.09039886894306157, "grad_norm": 3.718183755874634, "learning_rate": 9.991936771318083e-05, "loss": 0.7494807243347168, "memory(GiB)": 57.83, "step": 2110, "token_acc": 0.85, "train_speed(iter/s)": 1.509634 }, { "epoch": 0.09061308427231053, "grad_norm": 2.127686023712158, "learning_rate": 9.991898522021798e-05, "loss": 0.4112898826599121, "memory(GiB)": 57.83, "step": 2115, "token_acc": 0.9051987767584098, "train_speed(iter/s)": 1.509704 }, { "epoch": 0.09082729960155948, "grad_norm": 1.349186897277832, "learning_rate": 9.991860182292587e-05, "loss": 0.29790692329406737, "memory(GiB)": 57.83, "step": 2120, "token_acc": 0.9356223175965666, "train_speed(iter/s)": 1.509641 }, { "epoch": 0.09104151493080845, "grad_norm": 4.534247398376465, "learning_rate": 9.991821752131146e-05, "loss": 0.5310953617095947, "memory(GiB)": 57.83, "step": 2125, "token_acc": 0.9054545454545454, "train_speed(iter/s)": 1.509527 }, { "epoch": 0.09125573026005741, "grad_norm": 0.1368785947561264, "learning_rate": 9.991783231538172e-05, "loss": 0.26440062522888186, "memory(GiB)": 57.83, "step": 2130, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.509433 }, { "epoch": 0.09146994558930636, "grad_norm": 9.457904815673828, "learning_rate": 9.991744620514363e-05, "loss": 0.3672290802001953, "memory(GiB)": 57.83, "step": 2135, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.50964 }, { "epoch": 0.09168416091855533, "grad_norm": 3.4838483333587646, "learning_rate": 9.991705919060416e-05, "loss": 0.3392383575439453, "memory(GiB)": 57.83, "step": 2140, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.509757 }, { "epoch": 0.0918983762478043, "grad_norm": 2.9616620540618896, "learning_rate": 9.991667127177033e-05, "loss": 0.4247230052947998, "memory(GiB)": 57.83, "step": 2145, "token_acc": 0.8885245901639345, "train_speed(iter/s)": 1.510005 }, { "epoch": 0.09211259157705326, "grad_norm": 2.5032119750976562, "learning_rate": 9.99162824486492e-05, "loss": 0.37349083423614504, "memory(GiB)": 57.83, "step": 2150, "token_acc": 0.9067524115755627, "train_speed(iter/s)": 1.510032 }, { "epoch": 0.09232680690630221, "grad_norm": 2.251166820526123, "learning_rate": 9.991589272124778e-05, "loss": 0.4075924396514893, "memory(GiB)": 57.83, "step": 2155, "token_acc": 0.91796875, "train_speed(iter/s)": 1.510054 }, { "epoch": 0.09254102223555118, "grad_norm": 2.7242372035980225, "learning_rate": 9.991550208957312e-05, "loss": 0.37511701583862306, "memory(GiB)": 57.83, "step": 2160, "token_acc": 0.9124579124579124, "train_speed(iter/s)": 1.510119 }, { "epoch": 0.09275523756480014, "grad_norm": 2.6500437259674072, "learning_rate": 9.991511055363232e-05, "loss": 0.28545854091644285, "memory(GiB)": 57.83, "step": 2165, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.510347 }, { "epoch": 0.09296945289404909, "grad_norm": 2.7835581302642822, "learning_rate": 9.991471811343248e-05, "loss": 0.583714485168457, "memory(GiB)": 57.83, "step": 2170, "token_acc": 0.8945147679324894, "train_speed(iter/s)": 1.510669 }, { "epoch": 0.09318366822329806, "grad_norm": 0.1971823126077652, "learning_rate": 9.991432476898069e-05, "loss": 0.3036702394485474, "memory(GiB)": 57.83, "step": 2175, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.510926 }, { "epoch": 0.09339788355254702, "grad_norm": 6.180114269256592, "learning_rate": 9.991393052028408e-05, "loss": 0.3885409116744995, "memory(GiB)": 57.83, "step": 2180, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.511205 }, { "epoch": 0.09361209888179599, "grad_norm": 3.63148832321167, "learning_rate": 9.991353536734981e-05, "loss": 0.18328100442886353, "memory(GiB)": 57.83, "step": 2185, "token_acc": 0.95, "train_speed(iter/s)": 1.511533 }, { "epoch": 0.09382631421104494, "grad_norm": 2.290264129638672, "learning_rate": 9.991313931018503e-05, "loss": 0.25804924964904785, "memory(GiB)": 57.83, "step": 2190, "token_acc": 0.9415384615384615, "train_speed(iter/s)": 1.511479 }, { "epoch": 0.0940405295402939, "grad_norm": 3.2631585597991943, "learning_rate": 9.991274234879688e-05, "loss": 0.6042183399200439, "memory(GiB)": 57.83, "step": 2195, "token_acc": 0.8566037735849057, "train_speed(iter/s)": 1.511622 }, { "epoch": 0.09425474486954287, "grad_norm": 3.987161874771118, "learning_rate": 9.99123444831926e-05, "loss": 0.2985223770141602, "memory(GiB)": 57.83, "step": 2200, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.511633 }, { "epoch": 0.09446896019879182, "grad_norm": 2.3610386848449707, "learning_rate": 9.991194571337937e-05, "loss": 0.496595573425293, "memory(GiB)": 57.83, "step": 2205, "token_acc": 0.8944099378881988, "train_speed(iter/s)": 1.511663 }, { "epoch": 0.09468317552804079, "grad_norm": 9.483367919921875, "learning_rate": 9.991154603936443e-05, "loss": 0.31942932605743407, "memory(GiB)": 57.83, "step": 2210, "token_acc": 0.9267399267399268, "train_speed(iter/s)": 1.512148 }, { "epoch": 0.09489739085728975, "grad_norm": 1.038737177848816, "learning_rate": 9.991114546115502e-05, "loss": 0.2663280725479126, "memory(GiB)": 57.83, "step": 2215, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.512615 }, { "epoch": 0.0951116061865387, "grad_norm": 0.5263956189155579, "learning_rate": 9.991074397875836e-05, "loss": 0.18030860424041747, "memory(GiB)": 57.83, "step": 2220, "token_acc": 0.9497041420118343, "train_speed(iter/s)": 1.512799 }, { "epoch": 0.09532582151578767, "grad_norm": 2.107821226119995, "learning_rate": 9.991034159218178e-05, "loss": 0.4009974479675293, "memory(GiB)": 57.83, "step": 2225, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.513203 }, { "epoch": 0.09554003684503663, "grad_norm": 3.1267387866973877, "learning_rate": 9.990993830143253e-05, "loss": 0.5532732963562011, "memory(GiB)": 57.83, "step": 2230, "token_acc": 0.8862068965517241, "train_speed(iter/s)": 1.513061 }, { "epoch": 0.0957542521742856, "grad_norm": 1.7846169471740723, "learning_rate": 9.990953410651793e-05, "loss": 0.388879132270813, "memory(GiB)": 57.83, "step": 2235, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.513063 }, { "epoch": 0.09596846750353455, "grad_norm": 0.23187610507011414, "learning_rate": 9.990912900744529e-05, "loss": 0.3028446674346924, "memory(GiB)": 57.83, "step": 2240, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.513164 }, { "epoch": 0.09618268283278351, "grad_norm": 2.04084849357605, "learning_rate": 9.990872300422198e-05, "loss": 0.42904019355773926, "memory(GiB)": 57.83, "step": 2245, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.513439 }, { "epoch": 0.09639689816203248, "grad_norm": 2.102917194366455, "learning_rate": 9.990831609685532e-05, "loss": 0.47067956924438475, "memory(GiB)": 57.83, "step": 2250, "token_acc": 0.8785046728971962, "train_speed(iter/s)": 1.513464 }, { "epoch": 0.09661111349128143, "grad_norm": 1.7556836605072021, "learning_rate": 9.990790828535271e-05, "loss": 0.35270354747772215, "memory(GiB)": 57.83, "step": 2255, "token_acc": 0.908745247148289, "train_speed(iter/s)": 1.513282 }, { "epoch": 0.0968253288205304, "grad_norm": 2.412956714630127, "learning_rate": 9.990749956972152e-05, "loss": 0.44744095802307127, "memory(GiB)": 57.83, "step": 2260, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.513102 }, { "epoch": 0.09703954414977936, "grad_norm": 2.0045385360717773, "learning_rate": 9.990708994996916e-05, "loss": 0.24230170249938965, "memory(GiB)": 57.83, "step": 2265, "token_acc": 0.9385964912280702, "train_speed(iter/s)": 1.513215 }, { "epoch": 0.09725375947902833, "grad_norm": 2.103858470916748, "learning_rate": 9.990667942610303e-05, "loss": 0.3797850370407104, "memory(GiB)": 57.83, "step": 2270, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.513399 }, { "epoch": 0.09746797480827728, "grad_norm": 3.922213315963745, "learning_rate": 9.990626799813061e-05, "loss": 0.4533543109893799, "memory(GiB)": 57.83, "step": 2275, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.513706 }, { "epoch": 0.09768219013752624, "grad_norm": 1.3530110120773315, "learning_rate": 9.990585566605932e-05, "loss": 0.5892696380615234, "memory(GiB)": 57.83, "step": 2280, "token_acc": 0.8724137931034482, "train_speed(iter/s)": 1.513912 }, { "epoch": 0.09789640546677521, "grad_norm": 4.482554912567139, "learning_rate": 9.990544242989663e-05, "loss": 0.4555466651916504, "memory(GiB)": 57.83, "step": 2285, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.51404 }, { "epoch": 0.09811062079602416, "grad_norm": 4.787440776824951, "learning_rate": 9.990502828965005e-05, "loss": 0.6266982078552246, "memory(GiB)": 57.83, "step": 2290, "token_acc": 0.8792569659442725, "train_speed(iter/s)": 1.514366 }, { "epoch": 0.09832483612527312, "grad_norm": 0.13243888318538666, "learning_rate": 9.990461324532705e-05, "loss": 0.5205819129943847, "memory(GiB)": 57.83, "step": 2295, "token_acc": 0.883495145631068, "train_speed(iter/s)": 1.514322 }, { "epoch": 0.09853905145452209, "grad_norm": 1.6306568384170532, "learning_rate": 9.990419729693519e-05, "loss": 0.43828411102294923, "memory(GiB)": 57.83, "step": 2300, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.514963 }, { "epoch": 0.09875326678377104, "grad_norm": 0.08481105417013168, "learning_rate": 9.990378044448197e-05, "loss": 0.27972941398620604, "memory(GiB)": 57.83, "step": 2305, "token_acc": 0.9288389513108615, "train_speed(iter/s)": 1.515411 }, { "epoch": 0.09896748211302, "grad_norm": 1.7977423667907715, "learning_rate": 9.990336268797494e-05, "loss": 0.41505069732666017, "memory(GiB)": 57.83, "step": 2310, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.515631 }, { "epoch": 0.09918169744226897, "grad_norm": 1.4308570623397827, "learning_rate": 9.990294402742171e-05, "loss": 0.28970272541046144, "memory(GiB)": 57.83, "step": 2315, "token_acc": 0.9484848484848485, "train_speed(iter/s)": 1.515668 }, { "epoch": 0.09939591277151794, "grad_norm": 2.57517147064209, "learning_rate": 9.990252446282982e-05, "loss": 0.25223736763000487, "memory(GiB)": 57.83, "step": 2320, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.515704 }, { "epoch": 0.09961012810076689, "grad_norm": 4.679688930511475, "learning_rate": 9.990210399420688e-05, "loss": 0.3138675928115845, "memory(GiB)": 57.83, "step": 2325, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.515768 }, { "epoch": 0.09982434343001585, "grad_norm": 1.648979902267456, "learning_rate": 9.990168262156052e-05, "loss": 0.41692361831665037, "memory(GiB)": 57.83, "step": 2330, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.515863 }, { "epoch": 0.10003855875926482, "grad_norm": 1.9181838035583496, "learning_rate": 9.990126034489837e-05, "loss": 0.2203444719314575, "memory(GiB)": 57.83, "step": 2335, "token_acc": 0.9495268138801262, "train_speed(iter/s)": 1.515816 }, { "epoch": 0.10025277408851377, "grad_norm": 2.8398234844207764, "learning_rate": 9.990083716422808e-05, "loss": 1.014545726776123, "memory(GiB)": 57.83, "step": 2340, "token_acc": 0.8084507042253521, "train_speed(iter/s)": 1.516224 }, { "epoch": 0.10046698941776273, "grad_norm": 1.7266123294830322, "learning_rate": 9.990041307955731e-05, "loss": 0.39326908588409426, "memory(GiB)": 57.83, "step": 2345, "token_acc": 0.8807017543859649, "train_speed(iter/s)": 1.516299 }, { "epoch": 0.1006812047470117, "grad_norm": 0.9288516640663147, "learning_rate": 9.989998809089376e-05, "loss": 0.3655564785003662, "memory(GiB)": 57.83, "step": 2350, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.516616 }, { "epoch": 0.10089542007626066, "grad_norm": 8.746625900268555, "learning_rate": 9.98995621982451e-05, "loss": 0.3313194990158081, "memory(GiB)": 57.83, "step": 2355, "token_acc": 0.9191489361702128, "train_speed(iter/s)": 1.516719 }, { "epoch": 0.10110963540550962, "grad_norm": 3.8279190063476562, "learning_rate": 9.989913540161906e-05, "loss": 0.4110274314880371, "memory(GiB)": 57.83, "step": 2360, "token_acc": 0.9039145907473309, "train_speed(iter/s)": 1.517016 }, { "epoch": 0.10132385073475858, "grad_norm": 2.5983002185821533, "learning_rate": 9.989870770102339e-05, "loss": 0.5118196487426758, "memory(GiB)": 57.83, "step": 2365, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.516827 }, { "epoch": 0.10153806606400755, "grad_norm": 3.447355270385742, "learning_rate": 9.989827909646581e-05, "loss": 0.365372896194458, "memory(GiB)": 57.83, "step": 2370, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.516718 }, { "epoch": 0.1017522813932565, "grad_norm": 2.768977403640747, "learning_rate": 9.989784958795409e-05, "loss": 0.3143187999725342, "memory(GiB)": 57.83, "step": 2375, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.516913 }, { "epoch": 0.10196649672250546, "grad_norm": 2.491548776626587, "learning_rate": 9.989741917549603e-05, "loss": 0.531307601928711, "memory(GiB)": 57.83, "step": 2380, "token_acc": 0.8753993610223643, "train_speed(iter/s)": 1.517235 }, { "epoch": 0.10218071205175443, "grad_norm": 2.7866947650909424, "learning_rate": 9.98969878590994e-05, "loss": 0.4757603645324707, "memory(GiB)": 57.83, "step": 2385, "token_acc": 0.8912280701754386, "train_speed(iter/s)": 1.517309 }, { "epoch": 0.10239492738100338, "grad_norm": 3.2842047214508057, "learning_rate": 9.989655563877203e-05, "loss": 0.5569895744323731, "memory(GiB)": 57.83, "step": 2390, "token_acc": 0.8838028169014085, "train_speed(iter/s)": 1.517266 }, { "epoch": 0.10260914271025234, "grad_norm": 1.5558867454528809, "learning_rate": 9.989612251452176e-05, "loss": 0.5476381778717041, "memory(GiB)": 57.83, "step": 2395, "token_acc": 0.8827838827838828, "train_speed(iter/s)": 1.517307 }, { "epoch": 0.10282335803950131, "grad_norm": 1.2142301797866821, "learning_rate": 9.98956884863564e-05, "loss": 0.44309329986572266, "memory(GiB)": 57.83, "step": 2400, "token_acc": 0.9152542372881356, "train_speed(iter/s)": 1.517484 }, { "epoch": 0.10303757336875027, "grad_norm": 2.1688730716705322, "learning_rate": 9.989525355428386e-05, "loss": 0.23651521205902098, "memory(GiB)": 57.83, "step": 2405, "token_acc": 0.9465408805031447, "train_speed(iter/s)": 1.51754 }, { "epoch": 0.10325178869799922, "grad_norm": 3.397451162338257, "learning_rate": 9.9894817718312e-05, "loss": 0.7424505233764649, "memory(GiB)": 57.83, "step": 2410, "token_acc": 0.8496732026143791, "train_speed(iter/s)": 1.517815 }, { "epoch": 0.10346600402724819, "grad_norm": 2.6040303707122803, "learning_rate": 9.98943809784487e-05, "loss": 0.27676370143890383, "memory(GiB)": 57.83, "step": 2415, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.517849 }, { "epoch": 0.10368021935649716, "grad_norm": 2.110495090484619, "learning_rate": 9.989394333470188e-05, "loss": 0.3664689064025879, "memory(GiB)": 57.83, "step": 2420, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.517859 }, { "epoch": 0.1038944346857461, "grad_norm": 3.028832197189331, "learning_rate": 9.989350478707949e-05, "loss": 0.535345458984375, "memory(GiB)": 57.83, "step": 2425, "token_acc": 0.8821656050955414, "train_speed(iter/s)": 1.518282 }, { "epoch": 0.10410865001499507, "grad_norm": 4.8028082847595215, "learning_rate": 9.989306533558944e-05, "loss": 0.45121183395385744, "memory(GiB)": 57.83, "step": 2430, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.518241 }, { "epoch": 0.10432286534424404, "grad_norm": 4.624881744384766, "learning_rate": 9.98926249802397e-05, "loss": 0.6293641090393066, "memory(GiB)": 57.83, "step": 2435, "token_acc": 0.8825214899713467, "train_speed(iter/s)": 1.518328 }, { "epoch": 0.104537080673493, "grad_norm": 1.427350401878357, "learning_rate": 9.989218372103829e-05, "loss": 0.42759904861450193, "memory(GiB)": 57.83, "step": 2440, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.518226 }, { "epoch": 0.10475129600274195, "grad_norm": 6.799141883850098, "learning_rate": 9.989174155799314e-05, "loss": 0.7371190071105957, "memory(GiB)": 57.83, "step": 2445, "token_acc": 0.8288973384030418, "train_speed(iter/s)": 1.518514 }, { "epoch": 0.10496551133199092, "grad_norm": 3.528456926345825, "learning_rate": 9.989129849111229e-05, "loss": 0.2748852729797363, "memory(GiB)": 57.83, "step": 2450, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.518335 }, { "epoch": 0.10517972666123988, "grad_norm": 3.6448042392730713, "learning_rate": 9.989085452040377e-05, "loss": 0.38316822052001953, "memory(GiB)": 57.83, "step": 2455, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.518441 }, { "epoch": 0.10539394199048883, "grad_norm": 2.464329719543457, "learning_rate": 9.989040964587563e-05, "loss": 0.4537045478820801, "memory(GiB)": 57.83, "step": 2460, "token_acc": 0.9150326797385621, "train_speed(iter/s)": 1.518517 }, { "epoch": 0.1056081573197378, "grad_norm": 3.271143913269043, "learning_rate": 9.988996386753591e-05, "loss": 0.7689284801483154, "memory(GiB)": 57.83, "step": 2465, "token_acc": 0.8699690402476781, "train_speed(iter/s)": 1.518747 }, { "epoch": 0.10582237264898676, "grad_norm": 1.68510901927948, "learning_rate": 9.988951718539269e-05, "loss": 0.26269633769989015, "memory(GiB)": 57.83, "step": 2470, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.518672 }, { "epoch": 0.10603658797823572, "grad_norm": 5.02957820892334, "learning_rate": 9.988906959945407e-05, "loss": 0.22794866561889648, "memory(GiB)": 57.83, "step": 2475, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.51892 }, { "epoch": 0.10625080330748468, "grad_norm": 4.044186592102051, "learning_rate": 9.988862110972816e-05, "loss": 0.4170974254608154, "memory(GiB)": 57.83, "step": 2480, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.518912 }, { "epoch": 0.10646501863673365, "grad_norm": 4.527347087860107, "learning_rate": 9.988817171622307e-05, "loss": 0.402128267288208, "memory(GiB)": 57.83, "step": 2485, "token_acc": 0.9196141479099679, "train_speed(iter/s)": 1.518822 }, { "epoch": 0.10667923396598261, "grad_norm": 1.041957974433899, "learning_rate": 9.988772141894696e-05, "loss": 0.4703659534454346, "memory(GiB)": 57.83, "step": 2490, "token_acc": 0.9, "train_speed(iter/s)": 1.518862 }, { "epoch": 0.10689344929523156, "grad_norm": 2.0295379161834717, "learning_rate": 9.988727021790796e-05, "loss": 0.31834211349487307, "memory(GiB)": 57.83, "step": 2495, "token_acc": 0.92, "train_speed(iter/s)": 1.51896 }, { "epoch": 0.10710766462448053, "grad_norm": 4.6844282150268555, "learning_rate": 9.988681811311428e-05, "loss": 0.7860642910003662, "memory(GiB)": 57.83, "step": 2500, "token_acc": 0.7908496732026143, "train_speed(iter/s)": 1.519281 }, { "epoch": 0.10710766462448053, "eval_loss": 2.6587765216827393, "eval_runtime": 10.3867, "eval_samples_per_second": 9.628, "eval_steps_per_second": 9.628, "eval_token_acc": 0.45966709346991036, "step": 2500 }, { "epoch": 0.1073218799537295, "grad_norm": 2.199474573135376, "learning_rate": 9.988636510457408e-05, "loss": 0.4809749126434326, "memory(GiB)": 57.83, "step": 2505, "token_acc": 0.5813528336380256, "train_speed(iter/s)": 1.50909 }, { "epoch": 0.10753609528297844, "grad_norm": 2.483931303024292, "learning_rate": 9.988591119229559e-05, "loss": 0.3802937030792236, "memory(GiB)": 57.83, "step": 2510, "token_acc": 0.9191616766467066, "train_speed(iter/s)": 1.509405 }, { "epoch": 0.10775031061222741, "grad_norm": 11.589332580566406, "learning_rate": 9.988545637628702e-05, "loss": 0.6338032245635986, "memory(GiB)": 57.83, "step": 2515, "token_acc": 0.8818181818181818, "train_speed(iter/s)": 1.509805 }, { "epoch": 0.10796452594147637, "grad_norm": 2.526390314102173, "learning_rate": 9.98850006565566e-05, "loss": 0.3199204921722412, "memory(GiB)": 57.83, "step": 2520, "token_acc": 0.936026936026936, "train_speed(iter/s)": 1.509653 }, { "epoch": 0.10817874127072534, "grad_norm": 1.5145788192749023, "learning_rate": 9.988454403311262e-05, "loss": 0.2501399040222168, "memory(GiB)": 57.83, "step": 2525, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.509578 }, { "epoch": 0.10839295659997429, "grad_norm": 4.9559855461120605, "learning_rate": 9.98840865059633e-05, "loss": 0.4948309898376465, "memory(GiB)": 57.83, "step": 2530, "token_acc": 0.897887323943662, "train_speed(iter/s)": 1.509829 }, { "epoch": 0.10860717192922326, "grad_norm": 2.4412269592285156, "learning_rate": 9.988362807511697e-05, "loss": 0.30352048873901366, "memory(GiB)": 57.83, "step": 2535, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.509779 }, { "epoch": 0.10882138725847222, "grad_norm": 2.288729667663574, "learning_rate": 9.988316874058195e-05, "loss": 0.28680422306060793, "memory(GiB)": 57.83, "step": 2540, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.509928 }, { "epoch": 0.10903560258772117, "grad_norm": 3.496809959411621, "learning_rate": 9.988270850236649e-05, "loss": 0.39579975605010986, "memory(GiB)": 57.83, "step": 2545, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.510211 }, { "epoch": 0.10924981791697014, "grad_norm": 4.863556861877441, "learning_rate": 9.9882247360479e-05, "loss": 0.2680971145629883, "memory(GiB)": 57.83, "step": 2550, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.510217 }, { "epoch": 0.1094640332462191, "grad_norm": 3.6153016090393066, "learning_rate": 9.98817853149278e-05, "loss": 0.33464913368225097, "memory(GiB)": 57.83, "step": 2555, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.510367 }, { "epoch": 0.10967824857546805, "grad_norm": 7.424304962158203, "learning_rate": 9.988132236572126e-05, "loss": 0.47231369018554686, "memory(GiB)": 57.83, "step": 2560, "token_acc": 0.8881789137380192, "train_speed(iter/s)": 1.510337 }, { "epoch": 0.10989246390471702, "grad_norm": 1.878862738609314, "learning_rate": 9.988085851286778e-05, "loss": 0.43070077896118164, "memory(GiB)": 57.83, "step": 2565, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.510741 }, { "epoch": 0.11010667923396598, "grad_norm": 2.123058557510376, "learning_rate": 9.988039375637574e-05, "loss": 0.473504638671875, "memory(GiB)": 57.83, "step": 2570, "token_acc": 0.8867924528301887, "train_speed(iter/s)": 1.510934 }, { "epoch": 0.11032089456321495, "grad_norm": 1.0594854354858398, "learning_rate": 9.98799280962536e-05, "loss": 0.44127693176269533, "memory(GiB)": 57.83, "step": 2575, "token_acc": 0.9, "train_speed(iter/s)": 1.510983 }, { "epoch": 0.1105351098924639, "grad_norm": 2.140557050704956, "learning_rate": 9.987946153250976e-05, "loss": 0.2431337833404541, "memory(GiB)": 57.83, "step": 2580, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.51098 }, { "epoch": 0.11074932522171287, "grad_norm": 2.583028554916382, "learning_rate": 9.987899406515268e-05, "loss": 0.3909148693084717, "memory(GiB)": 57.83, "step": 2585, "token_acc": 0.9149560117302052, "train_speed(iter/s)": 1.511033 }, { "epoch": 0.11096354055096183, "grad_norm": 6.058218002319336, "learning_rate": 9.987852569419084e-05, "loss": 0.3425605773925781, "memory(GiB)": 57.83, "step": 2590, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.511016 }, { "epoch": 0.11117775588021078, "grad_norm": 0.9075403809547424, "learning_rate": 9.98780564196327e-05, "loss": 0.23711061477661133, "memory(GiB)": 57.83, "step": 2595, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.511336 }, { "epoch": 0.11139197120945975, "grad_norm": 2.48140287399292, "learning_rate": 9.98775862414868e-05, "loss": 0.6635601997375489, "memory(GiB)": 57.83, "step": 2600, "token_acc": 0.8636363636363636, "train_speed(iter/s)": 1.511422 }, { "epoch": 0.11160618653870871, "grad_norm": 3.184706449508667, "learning_rate": 9.987711515976164e-05, "loss": 0.4201952934265137, "memory(GiB)": 57.83, "step": 2605, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.511642 }, { "epoch": 0.11182040186795768, "grad_norm": 0.8893657326698303, "learning_rate": 9.987664317446572e-05, "loss": 0.16378759145736693, "memory(GiB)": 57.83, "step": 2610, "token_acc": 0.9633699633699634, "train_speed(iter/s)": 1.511578 }, { "epoch": 0.11203461719720663, "grad_norm": 4.392936706542969, "learning_rate": 9.987617028560765e-05, "loss": 0.374851131439209, "memory(GiB)": 57.83, "step": 2615, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.511496 }, { "epoch": 0.1122488325264556, "grad_norm": 2.598947048187256, "learning_rate": 9.987569649319595e-05, "loss": 0.5154757976531983, "memory(GiB)": 57.83, "step": 2620, "token_acc": 0.8821656050955414, "train_speed(iter/s)": 1.511425 }, { "epoch": 0.11246304785570456, "grad_norm": 3.128789186477661, "learning_rate": 9.987522179723923e-05, "loss": 0.37300848960876465, "memory(GiB)": 57.83, "step": 2625, "token_acc": 0.9205776173285198, "train_speed(iter/s)": 1.511376 }, { "epoch": 0.11267726318495351, "grad_norm": 2.8604416847229004, "learning_rate": 9.987474619774609e-05, "loss": 0.35259432792663575, "memory(GiB)": 57.83, "step": 2630, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.51142 }, { "epoch": 0.11289147851420248, "grad_norm": 6.133558750152588, "learning_rate": 9.98742696947251e-05, "loss": 0.5564678192138672, "memory(GiB)": 57.83, "step": 2635, "token_acc": 0.8961038961038961, "train_speed(iter/s)": 1.511749 }, { "epoch": 0.11310569384345144, "grad_norm": 7.68245792388916, "learning_rate": 9.987379228818497e-05, "loss": 0.24802730083465577, "memory(GiB)": 57.83, "step": 2640, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.511871 }, { "epoch": 0.11331990917270039, "grad_norm": 0.728346586227417, "learning_rate": 9.987331397813429e-05, "loss": 0.16649070978164673, "memory(GiB)": 57.83, "step": 2645, "token_acc": 0.9652173913043478, "train_speed(iter/s)": 1.511845 }, { "epoch": 0.11353412450194936, "grad_norm": 3.652160167694092, "learning_rate": 9.987283476458174e-05, "loss": 0.5777617931365967, "memory(GiB)": 57.83, "step": 2650, "token_acc": 0.8829113924050633, "train_speed(iter/s)": 1.512073 }, { "epoch": 0.11374833983119832, "grad_norm": 1.190796971321106, "learning_rate": 9.9872354647536e-05, "loss": 0.3771846055984497, "memory(GiB)": 57.83, "step": 2655, "token_acc": 0.9075907590759076, "train_speed(iter/s)": 1.51207 }, { "epoch": 0.11396255516044729, "grad_norm": 4.272654056549072, "learning_rate": 9.987187362700579e-05, "loss": 0.24348440170288085, "memory(GiB)": 57.83, "step": 2660, "token_acc": 0.9357429718875502, "train_speed(iter/s)": 1.512311 }, { "epoch": 0.11417677048969624, "grad_norm": 2.6485626697540283, "learning_rate": 9.987139170299978e-05, "loss": 0.4488868713378906, "memory(GiB)": 57.83, "step": 2665, "token_acc": 0.8949044585987261, "train_speed(iter/s)": 1.512758 }, { "epoch": 0.1143909858189452, "grad_norm": 10.228973388671875, "learning_rate": 9.987090887552675e-05, "loss": 0.6419280529022217, "memory(GiB)": 57.83, "step": 2670, "token_acc": 0.8522727272727273, "train_speed(iter/s)": 1.513388 }, { "epoch": 0.11460520114819417, "grad_norm": 2.5250794887542725, "learning_rate": 9.987042514459541e-05, "loss": 0.46475987434387206, "memory(GiB)": 57.83, "step": 2675, "token_acc": 0.8971631205673759, "train_speed(iter/s)": 1.513474 }, { "epoch": 0.11481941647744312, "grad_norm": 2.5227034091949463, "learning_rate": 9.986994051021454e-05, "loss": 0.4130404472351074, "memory(GiB)": 57.83, "step": 2680, "token_acc": 0.912621359223301, "train_speed(iter/s)": 1.513586 }, { "epoch": 0.11503363180669209, "grad_norm": 1.5147368907928467, "learning_rate": 9.98694549723929e-05, "loss": 0.12846802473068236, "memory(GiB)": 57.83, "step": 2685, "token_acc": 0.9753086419753086, "train_speed(iter/s)": 1.513595 }, { "epoch": 0.11524784713594105, "grad_norm": 0.04803531616926193, "learning_rate": 9.986896853113932e-05, "loss": 0.3742664098739624, "memory(GiB)": 57.83, "step": 2690, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.513805 }, { "epoch": 0.11546206246519002, "grad_norm": 1.5037118196487427, "learning_rate": 9.98684811864626e-05, "loss": 0.17264052629470825, "memory(GiB)": 57.83, "step": 2695, "token_acc": 0.9528985507246377, "train_speed(iter/s)": 1.514064 }, { "epoch": 0.11567627779443897, "grad_norm": 0.8887499570846558, "learning_rate": 9.986799293837155e-05, "loss": 0.32582592964172363, "memory(GiB)": 57.83, "step": 2700, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.514342 }, { "epoch": 0.11589049312368793, "grad_norm": 0.673583984375, "learning_rate": 9.986750378687502e-05, "loss": 0.229567551612854, "memory(GiB)": 57.83, "step": 2705, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.514409 }, { "epoch": 0.1161047084529369, "grad_norm": 2.6209771633148193, "learning_rate": 9.98670137319819e-05, "loss": 0.30509541034698484, "memory(GiB)": 57.83, "step": 2710, "token_acc": 0.9452554744525548, "train_speed(iter/s)": 1.514312 }, { "epoch": 0.11631892378218585, "grad_norm": 4.081751823425293, "learning_rate": 9.986652277370103e-05, "loss": 0.5211978912353515, "memory(GiB)": 57.83, "step": 2715, "token_acc": 0.8708333333333333, "train_speed(iter/s)": 1.514676 }, { "epoch": 0.11653313911143481, "grad_norm": 1.4904735088348389, "learning_rate": 9.986603091204134e-05, "loss": 0.49604358673095705, "memory(GiB)": 57.83, "step": 2720, "token_acc": 0.9116465863453815, "train_speed(iter/s)": 1.514809 }, { "epoch": 0.11674735444068378, "grad_norm": 4.701450824737549, "learning_rate": 9.98655381470117e-05, "loss": 0.5582290649414062, "memory(GiB)": 57.83, "step": 2725, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.514966 }, { "epoch": 0.11696156976993273, "grad_norm": 2.8145864009857178, "learning_rate": 9.986504447862108e-05, "loss": 0.38521294593811034, "memory(GiB)": 57.83, "step": 2730, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.514848 }, { "epoch": 0.1171757850991817, "grad_norm": 4.379426956176758, "learning_rate": 9.986454990687839e-05, "loss": 0.2908231973648071, "memory(GiB)": 57.83, "step": 2735, "token_acc": 0.9551020408163265, "train_speed(iter/s)": 1.515 }, { "epoch": 0.11739000042843066, "grad_norm": 1.9115129709243774, "learning_rate": 9.986405443179261e-05, "loss": 0.19821332693099974, "memory(GiB)": 57.83, "step": 2740, "token_acc": 0.9563492063492064, "train_speed(iter/s)": 1.514957 }, { "epoch": 0.11760421575767963, "grad_norm": 2.380143880844116, "learning_rate": 9.986355805337268e-05, "loss": 0.6040264129638672, "memory(GiB)": 57.83, "step": 2745, "token_acc": 0.8637992831541219, "train_speed(iter/s)": 1.514922 }, { "epoch": 0.11781843108692858, "grad_norm": 3.1103219985961914, "learning_rate": 9.986306077162766e-05, "loss": 0.29142882823944094, "memory(GiB)": 57.83, "step": 2750, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.514967 }, { "epoch": 0.11803264641617754, "grad_norm": 7.3828277587890625, "learning_rate": 9.98625625865665e-05, "loss": 0.40814385414123533, "memory(GiB)": 57.83, "step": 2755, "token_acc": 0.898989898989899, "train_speed(iter/s)": 1.515064 }, { "epoch": 0.11824686174542651, "grad_norm": 5.592649936676025, "learning_rate": 9.986206349819825e-05, "loss": 0.5608318328857422, "memory(GiB)": 57.83, "step": 2760, "token_acc": 0.8951310861423221, "train_speed(iter/s)": 1.515271 }, { "epoch": 0.11846107707467546, "grad_norm": 6.260732173919678, "learning_rate": 9.986156350653193e-05, "loss": 0.4508352279663086, "memory(GiB)": 57.83, "step": 2765, "token_acc": 0.9072847682119205, "train_speed(iter/s)": 1.51528 }, { "epoch": 0.11867529240392442, "grad_norm": 2.339463233947754, "learning_rate": 9.986106261157662e-05, "loss": 0.37288522720336914, "memory(GiB)": 57.83, "step": 2770, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.515621 }, { "epoch": 0.11888950773317339, "grad_norm": 1.611448049545288, "learning_rate": 9.986056081334139e-05, "loss": 0.32435002326965334, "memory(GiB)": 57.83, "step": 2775, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.515605 }, { "epoch": 0.11910372306242235, "grad_norm": 2.3192176818847656, "learning_rate": 9.986005811183533e-05, "loss": 0.2152947425842285, "memory(GiB)": 57.83, "step": 2780, "token_acc": 0.9629629629629629, "train_speed(iter/s)": 1.515683 }, { "epoch": 0.1193179383916713, "grad_norm": 1.2520666122436523, "learning_rate": 9.985955450706755e-05, "loss": 0.18490495681762695, "memory(GiB)": 57.83, "step": 2785, "token_acc": 0.9562043795620438, "train_speed(iter/s)": 1.515841 }, { "epoch": 0.11953215372092027, "grad_norm": 2.7619619369506836, "learning_rate": 9.985904999904715e-05, "loss": 0.6955222129821778, "memory(GiB)": 57.83, "step": 2790, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.515858 }, { "epoch": 0.11974636905016924, "grad_norm": 2.0278267860412598, "learning_rate": 9.98585445877833e-05, "loss": 0.29397571086883545, "memory(GiB)": 57.83, "step": 2795, "token_acc": 0.9501661129568106, "train_speed(iter/s)": 1.515919 }, { "epoch": 0.11996058437941819, "grad_norm": 3.959850311279297, "learning_rate": 9.985803827328515e-05, "loss": 0.5551119804382324, "memory(GiB)": 57.83, "step": 2800, "token_acc": 0.8618181818181818, "train_speed(iter/s)": 1.516182 }, { "epoch": 0.12017479970866715, "grad_norm": 3.183290481567383, "learning_rate": 9.985753105556185e-05, "loss": 0.5672118186950683, "memory(GiB)": 57.83, "step": 2805, "token_acc": 0.8962962962962963, "train_speed(iter/s)": 1.516486 }, { "epoch": 0.12038901503791612, "grad_norm": 2.8598294258117676, "learning_rate": 9.985702293462261e-05, "loss": 0.23244738578796387, "memory(GiB)": 57.83, "step": 2810, "token_acc": 0.940809968847352, "train_speed(iter/s)": 1.516338 }, { "epoch": 0.12060323036716507, "grad_norm": 2.8113303184509277, "learning_rate": 9.985651391047663e-05, "loss": 0.35447187423706056, "memory(GiB)": 57.83, "step": 2815, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.516594 }, { "epoch": 0.12081744569641403, "grad_norm": 5.964366436004639, "learning_rate": 9.985600398313313e-05, "loss": 0.589550256729126, "memory(GiB)": 57.83, "step": 2820, "token_acc": 0.8715277777777778, "train_speed(iter/s)": 1.517069 }, { "epoch": 0.121031661025663, "grad_norm": 4.003259181976318, "learning_rate": 9.985549315260137e-05, "loss": 0.4359861373901367, "memory(GiB)": 57.83, "step": 2825, "token_acc": 0.8943396226415095, "train_speed(iter/s)": 1.517268 }, { "epoch": 0.12124587635491196, "grad_norm": 1.2504783868789673, "learning_rate": 9.985498141889056e-05, "loss": 0.3731968641281128, "memory(GiB)": 57.83, "step": 2830, "token_acc": 0.9316546762589928, "train_speed(iter/s)": 1.517497 }, { "epoch": 0.12146009168416091, "grad_norm": 3.0300586223602295, "learning_rate": 9.985446878201e-05, "loss": 0.33304557800292967, "memory(GiB)": 57.83, "step": 2835, "token_acc": 0.9264305177111717, "train_speed(iter/s)": 1.517645 }, { "epoch": 0.12167430701340988, "grad_norm": 2.5835306644439697, "learning_rate": 9.985395524196896e-05, "loss": 0.2838619232177734, "memory(GiB)": 57.83, "step": 2840, "token_acc": 0.9372384937238494, "train_speed(iter/s)": 1.517533 }, { "epoch": 0.12188852234265884, "grad_norm": 4.262026786804199, "learning_rate": 9.985344079877677e-05, "loss": 0.21465463638305665, "memory(GiB)": 57.83, "step": 2845, "token_acc": 0.9505703422053232, "train_speed(iter/s)": 1.517481 }, { "epoch": 0.1221027376719078, "grad_norm": 3.0932278633117676, "learning_rate": 9.985292545244274e-05, "loss": 0.5274602890014648, "memory(GiB)": 57.83, "step": 2850, "token_acc": 0.9100719424460432, "train_speed(iter/s)": 1.517462 }, { "epoch": 0.12231695300115676, "grad_norm": 3.7710769176483154, "learning_rate": 9.985240920297618e-05, "loss": 0.5049216270446777, "memory(GiB)": 57.83, "step": 2855, "token_acc": 0.8945783132530121, "train_speed(iter/s)": 1.517692 }, { "epoch": 0.12253116833040573, "grad_norm": 6.446800708770752, "learning_rate": 9.985189205038647e-05, "loss": 0.43161487579345703, "memory(GiB)": 57.83, "step": 2860, "token_acc": 0.8931297709923665, "train_speed(iter/s)": 1.517711 }, { "epoch": 0.12274538365965469, "grad_norm": 2.2979273796081543, "learning_rate": 9.985137399468298e-05, "loss": 0.2703059196472168, "memory(GiB)": 57.83, "step": 2865, "token_acc": 0.9429530201342282, "train_speed(iter/s)": 1.51774 }, { "epoch": 0.12295959898890364, "grad_norm": 2.292294979095459, "learning_rate": 9.985085503587507e-05, "loss": 0.20255646705627442, "memory(GiB)": 57.83, "step": 2870, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.517784 }, { "epoch": 0.12317381431815261, "grad_norm": 1.2914764881134033, "learning_rate": 9.985033517397217e-05, "loss": 0.37209596633911135, "memory(GiB)": 57.83, "step": 2875, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.517703 }, { "epoch": 0.12338802964740157, "grad_norm": 3.382758617401123, "learning_rate": 9.984981440898366e-05, "loss": 0.24977903366088866, "memory(GiB)": 57.83, "step": 2880, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.51789 }, { "epoch": 0.12360224497665052, "grad_norm": 2.42804217338562, "learning_rate": 9.984929274091903e-05, "loss": 0.23378007411956786, "memory(GiB)": 57.83, "step": 2885, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.517984 }, { "epoch": 0.12381646030589949, "grad_norm": 3.560086250305176, "learning_rate": 9.984877016978768e-05, "loss": 0.4749479293823242, "memory(GiB)": 57.83, "step": 2890, "token_acc": 0.8881987577639752, "train_speed(iter/s)": 1.518 }, { "epoch": 0.12403067563514845, "grad_norm": 8.214594841003418, "learning_rate": 9.984824669559911e-05, "loss": 0.5336884498596192, "memory(GiB)": 57.83, "step": 2895, "token_acc": 0.8757396449704142, "train_speed(iter/s)": 1.518337 }, { "epoch": 0.1242448909643974, "grad_norm": 2.159132242202759, "learning_rate": 9.984772231836279e-05, "loss": 0.4905550479888916, "memory(GiB)": 57.83, "step": 2900, "token_acc": 0.8881118881118881, "train_speed(iter/s)": 1.518725 }, { "epoch": 0.12445910629364637, "grad_norm": 5.26190710067749, "learning_rate": 9.98471970380882e-05, "loss": 0.4117736339569092, "memory(GiB)": 57.83, "step": 2905, "token_acc": 0.9033333333333333, "train_speed(iter/s)": 1.518704 }, { "epoch": 0.12467332162289534, "grad_norm": 1.8422603607177734, "learning_rate": 9.98466708547849e-05, "loss": 0.25607502460479736, "memory(GiB)": 57.83, "step": 2910, "token_acc": 0.94375, "train_speed(iter/s)": 1.518674 }, { "epoch": 0.1248875369521443, "grad_norm": 2.9567577838897705, "learning_rate": 9.984614376846238e-05, "loss": 0.5268765926361084, "memory(GiB)": 57.83, "step": 2915, "token_acc": 0.8813056379821959, "train_speed(iter/s)": 1.518821 }, { "epoch": 0.12510175228139325, "grad_norm": 0.905709445476532, "learning_rate": 9.98456157791302e-05, "loss": 0.36819746494293215, "memory(GiB)": 57.83, "step": 2920, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.518903 }, { "epoch": 0.1253159676106422, "grad_norm": 3.4309463500976562, "learning_rate": 9.984508688679796e-05, "loss": 0.7342646598815918, "memory(GiB)": 57.83, "step": 2925, "token_acc": 0.851063829787234, "train_speed(iter/s)": 1.518924 }, { "epoch": 0.12553018293989118, "grad_norm": 2.333954095840454, "learning_rate": 9.984455709147519e-05, "loss": 0.5871391773223877, "memory(GiB)": 57.83, "step": 2930, "token_acc": 0.8712121212121212, "train_speed(iter/s)": 1.519175 }, { "epoch": 0.12574439826914013, "grad_norm": 2.9398860931396484, "learning_rate": 9.984402639317152e-05, "loss": 0.3576620101928711, "memory(GiB)": 57.83, "step": 2935, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.519355 }, { "epoch": 0.1259586135983891, "grad_norm": 2.3254878520965576, "learning_rate": 9.984349479189654e-05, "loss": 0.26828794479370116, "memory(GiB)": 57.83, "step": 2940, "token_acc": 0.933852140077821, "train_speed(iter/s)": 1.519666 }, { "epoch": 0.12617282892763806, "grad_norm": 0.22604233026504517, "learning_rate": 9.984296228765991e-05, "loss": 0.5109295845031738, "memory(GiB)": 57.83, "step": 2945, "token_acc": 0.8976377952755905, "train_speed(iter/s)": 1.51985 }, { "epoch": 0.12638704425688702, "grad_norm": 6.084090232849121, "learning_rate": 9.984242888047127e-05, "loss": 0.3661962985992432, "memory(GiB)": 57.83, "step": 2950, "token_acc": 0.9084507042253521, "train_speed(iter/s)": 1.519827 }, { "epoch": 0.126601259586136, "grad_norm": 1.9140913486480713, "learning_rate": 9.984189457034026e-05, "loss": 0.40433497428894044, "memory(GiB)": 57.83, "step": 2955, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.519774 }, { "epoch": 0.12681547491538495, "grad_norm": 3.9491164684295654, "learning_rate": 9.984135935727657e-05, "loss": 0.623484992980957, "memory(GiB)": 57.83, "step": 2960, "token_acc": 0.8774703557312253, "train_speed(iter/s)": 1.520053 }, { "epoch": 0.1270296902446339, "grad_norm": 0.3098786771297455, "learning_rate": 9.984082324128991e-05, "loss": 0.43290061950683595, "memory(GiB)": 57.83, "step": 2965, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.520065 }, { "epoch": 0.12724390557388288, "grad_norm": 0.8239394426345825, "learning_rate": 9.984028622238997e-05, "loss": 0.31000533103942873, "memory(GiB)": 57.83, "step": 2970, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.520269 }, { "epoch": 0.12745812090313183, "grad_norm": 3.0224175453186035, "learning_rate": 9.98397483005865e-05, "loss": 0.4273125648498535, "memory(GiB)": 57.83, "step": 2975, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.520463 }, { "epoch": 0.12767233623238078, "grad_norm": 2.816049337387085, "learning_rate": 9.983920947588923e-05, "loss": 0.7120389461517334, "memory(GiB)": 57.83, "step": 2980, "token_acc": 0.8383838383838383, "train_speed(iter/s)": 1.520613 }, { "epoch": 0.12788655156162976, "grad_norm": 5.196239471435547, "learning_rate": 9.983866974830794e-05, "loss": 0.3701418399810791, "memory(GiB)": 57.83, "step": 2985, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.520629 }, { "epoch": 0.1281007668908787, "grad_norm": 1.2969125509262085, "learning_rate": 9.983812911785238e-05, "loss": 0.3047261476516724, "memory(GiB)": 57.83, "step": 2990, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.520538 }, { "epoch": 0.12831498222012766, "grad_norm": 2.4254791736602783, "learning_rate": 9.983758758453238e-05, "loss": 0.2889134645462036, "memory(GiB)": 57.83, "step": 2995, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.520495 }, { "epoch": 0.12852919754937664, "grad_norm": 4.898059368133545, "learning_rate": 9.983704514835771e-05, "loss": 0.6448541641235351, "memory(GiB)": 57.83, "step": 3000, "token_acc": 0.8770226537216829, "train_speed(iter/s)": 1.52089 }, { "epoch": 0.12852919754937664, "eval_loss": 2.43453311920166, "eval_runtime": 12.0073, "eval_samples_per_second": 8.328, "eval_steps_per_second": 8.328, "eval_token_acc": 0.44519621109607577, "step": 3000 }, { "epoch": 0.1287434128786256, "grad_norm": 1.915229082107544, "learning_rate": 9.983650180933822e-05, "loss": 0.31440489292144774, "memory(GiB)": 57.83, "step": 3005, "token_acc": 0.5787401574803149, "train_speed(iter/s)": 1.511373 }, { "epoch": 0.12895762820787454, "grad_norm": 2.875525712966919, "learning_rate": 9.983595756748376e-05, "loss": 0.264946722984314, "memory(GiB)": 57.83, "step": 3010, "token_acc": 0.9510869565217391, "train_speed(iter/s)": 1.511271 }, { "epoch": 0.12917184353712352, "grad_norm": 3.1868622303009033, "learning_rate": 9.983541242280417e-05, "loss": 0.3569046497344971, "memory(GiB)": 57.83, "step": 3015, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.511181 }, { "epoch": 0.12938605886637247, "grad_norm": 6.661249160766602, "learning_rate": 9.983486637530934e-05, "loss": 0.6695085048675538, "memory(GiB)": 57.83, "step": 3020, "token_acc": 0.8636363636363636, "train_speed(iter/s)": 1.511774 }, { "epoch": 0.12960027419562145, "grad_norm": 1.2484638690948486, "learning_rate": 9.983431942500915e-05, "loss": 0.2602912664413452, "memory(GiB)": 57.83, "step": 3025, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.51193 }, { "epoch": 0.1298144895248704, "grad_norm": 1.3934756517410278, "learning_rate": 9.983377157191352e-05, "loss": 0.33817861080169676, "memory(GiB)": 57.83, "step": 3030, "token_acc": 0.9326599326599326, "train_speed(iter/s)": 1.512141 }, { "epoch": 0.13002870485411935, "grad_norm": 1.4336600303649902, "learning_rate": 9.983322281603235e-05, "loss": 0.44223337173461913, "memory(GiB)": 57.83, "step": 3035, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.512295 }, { "epoch": 0.13024292018336833, "grad_norm": 3.2466001510620117, "learning_rate": 9.983267315737563e-05, "loss": 0.5539028644561768, "memory(GiB)": 57.83, "step": 3040, "token_acc": 0.8923611111111112, "train_speed(iter/s)": 1.512722 }, { "epoch": 0.13045713551261728, "grad_norm": 3.6008198261260986, "learning_rate": 9.983212259595328e-05, "loss": 0.4770808696746826, "memory(GiB)": 57.83, "step": 3045, "token_acc": 0.8866666666666667, "train_speed(iter/s)": 1.512837 }, { "epoch": 0.13067135084186624, "grad_norm": 2.5835886001586914, "learning_rate": 9.983157113177528e-05, "loss": 0.37056674957275393, "memory(GiB)": 57.83, "step": 3050, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.513113 }, { "epoch": 0.13088556617111521, "grad_norm": 3.0609233379364014, "learning_rate": 9.983101876485162e-05, "loss": 0.648771333694458, "memory(GiB)": 57.83, "step": 3055, "token_acc": 0.8781362007168458, "train_speed(iter/s)": 1.513379 }, { "epoch": 0.13109978150036417, "grad_norm": 3.6907601356506348, "learning_rate": 9.983046549519232e-05, "loss": 0.1563814878463745, "memory(GiB)": 57.83, "step": 3060, "token_acc": 0.9688715953307393, "train_speed(iter/s)": 1.513387 }, { "epoch": 0.13131399682961312, "grad_norm": 4.946545600891113, "learning_rate": 9.982991132280739e-05, "loss": 0.40527806282043455, "memory(GiB)": 57.83, "step": 3065, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.51337 }, { "epoch": 0.1315282121588621, "grad_norm": 3.003506660461426, "learning_rate": 9.982935624770687e-05, "loss": 0.3683738708496094, "memory(GiB)": 57.83, "step": 3070, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.513167 }, { "epoch": 0.13174242748811105, "grad_norm": 2.828397035598755, "learning_rate": 9.98288002699008e-05, "loss": 0.45945448875427247, "memory(GiB)": 57.83, "step": 3075, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.51324 }, { "epoch": 0.13195664281736, "grad_norm": 1.51767897605896, "learning_rate": 9.982824338939929e-05, "loss": 0.3669546127319336, "memory(GiB)": 57.83, "step": 3080, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.513253 }, { "epoch": 0.13217085814660898, "grad_norm": 3.191429615020752, "learning_rate": 9.98276856062124e-05, "loss": 0.402158784866333, "memory(GiB)": 57.83, "step": 3085, "token_acc": 0.9140625, "train_speed(iter/s)": 1.513174 }, { "epoch": 0.13238507347585793, "grad_norm": 3.9821395874023438, "learning_rate": 9.982712692035025e-05, "loss": 0.37962965965270995, "memory(GiB)": 57.83, "step": 3090, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.51325 }, { "epoch": 0.13259928880510688, "grad_norm": 2.065192222595215, "learning_rate": 9.982656733182296e-05, "loss": 0.3114050626754761, "memory(GiB)": 57.83, "step": 3095, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.51335 }, { "epoch": 0.13281350413435586, "grad_norm": 1.861728310585022, "learning_rate": 9.982600684064065e-05, "loss": 0.21206517219543458, "memory(GiB)": 57.83, "step": 3100, "token_acc": 0.9609929078014184, "train_speed(iter/s)": 1.513332 }, { "epoch": 0.1330277194636048, "grad_norm": 5.755735874176025, "learning_rate": 9.98254454468135e-05, "loss": 0.3546006441116333, "memory(GiB)": 57.83, "step": 3105, "token_acc": 0.8903225806451613, "train_speed(iter/s)": 1.51332 }, { "epoch": 0.1332419347928538, "grad_norm": 2.5391054153442383, "learning_rate": 9.982488315035166e-05, "loss": 0.30806798934936525, "memory(GiB)": 57.83, "step": 3110, "token_acc": 0.9183098591549296, "train_speed(iter/s)": 1.513344 }, { "epoch": 0.13345615012210274, "grad_norm": 2.985980272293091, "learning_rate": 9.982431995126531e-05, "loss": 0.4255964279174805, "memory(GiB)": 57.83, "step": 3115, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.513363 }, { "epoch": 0.1336703654513517, "grad_norm": 1.192822813987732, "learning_rate": 9.982375584956467e-05, "loss": 0.4519073009490967, "memory(GiB)": 57.83, "step": 3120, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.513699 }, { "epoch": 0.13388458078060067, "grad_norm": 2.5834970474243164, "learning_rate": 9.982319084525995e-05, "loss": 0.5573528289794922, "memory(GiB)": 57.83, "step": 3125, "token_acc": 0.8982456140350877, "train_speed(iter/s)": 1.513675 }, { "epoch": 0.13409879610984962, "grad_norm": 3.230583906173706, "learning_rate": 9.98226249383614e-05, "loss": 0.43615331649780276, "memory(GiB)": 57.83, "step": 3130, "token_acc": 0.9037267080745341, "train_speed(iter/s)": 1.513614 }, { "epoch": 0.13431301143909857, "grad_norm": 2.409909725189209, "learning_rate": 9.982205812887925e-05, "loss": 0.27284438610076905, "memory(GiB)": 57.83, "step": 3135, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.513567 }, { "epoch": 0.13452722676834755, "grad_norm": 3.253040075302124, "learning_rate": 9.98214904168238e-05, "loss": 0.4054275035858154, "memory(GiB)": 57.83, "step": 3140, "token_acc": 0.92, "train_speed(iter/s)": 1.51364 }, { "epoch": 0.1347414420975965, "grad_norm": 3.522026300430298, "learning_rate": 9.98209218022053e-05, "loss": 0.5198966026306152, "memory(GiB)": 57.83, "step": 3145, "token_acc": 0.8961038961038961, "train_speed(iter/s)": 1.513736 }, { "epoch": 0.13495565742684545, "grad_norm": 2.8954782485961914, "learning_rate": 9.982035228503407e-05, "loss": 0.3559462785720825, "memory(GiB)": 57.83, "step": 3150, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.513648 }, { "epoch": 0.13516987275609443, "grad_norm": 7.2858381271362305, "learning_rate": 9.981978186532041e-05, "loss": 0.757664155960083, "memory(GiB)": 57.83, "step": 3155, "token_acc": 0.8562874251497006, "train_speed(iter/s)": 1.513848 }, { "epoch": 0.13538408808534338, "grad_norm": 1.913256049156189, "learning_rate": 9.981921054307469e-05, "loss": 0.38319687843322753, "memory(GiB)": 57.83, "step": 3160, "token_acc": 0.9262820512820513, "train_speed(iter/s)": 1.513857 }, { "epoch": 0.13559830341459234, "grad_norm": 2.34148907661438, "learning_rate": 9.981863831830723e-05, "loss": 0.4485775947570801, "memory(GiB)": 57.83, "step": 3165, "token_acc": 0.90625, "train_speed(iter/s)": 1.514197 }, { "epoch": 0.13581251874384132, "grad_norm": 2.7270936965942383, "learning_rate": 9.98180651910284e-05, "loss": 0.2431943893432617, "memory(GiB)": 57.83, "step": 3170, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.514341 }, { "epoch": 0.13602673407309027, "grad_norm": 3.342893123626709, "learning_rate": 9.981749116124859e-05, "loss": 0.31734466552734375, "memory(GiB)": 57.83, "step": 3175, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.5143 }, { "epoch": 0.13624094940233922, "grad_norm": 1.6825791597366333, "learning_rate": 9.981691622897818e-05, "loss": 0.435393762588501, "memory(GiB)": 57.83, "step": 3180, "token_acc": 0.9140127388535032, "train_speed(iter/s)": 1.51417 }, { "epoch": 0.1364551647315882, "grad_norm": 2.4499685764312744, "learning_rate": 9.981634039422761e-05, "loss": 0.5115778923034668, "memory(GiB)": 57.83, "step": 3185, "token_acc": 0.87890625, "train_speed(iter/s)": 1.514137 }, { "epoch": 0.13666938006083715, "grad_norm": 7.534067153930664, "learning_rate": 9.98157636570073e-05, "loss": 0.4986077308654785, "memory(GiB)": 57.83, "step": 3190, "token_acc": 0.9033333333333333, "train_speed(iter/s)": 1.514348 }, { "epoch": 0.13688359539008613, "grad_norm": 3.804619550704956, "learning_rate": 9.981518601732771e-05, "loss": 0.4113102912902832, "memory(GiB)": 57.83, "step": 3195, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.51441 }, { "epoch": 0.13709781071933508, "grad_norm": 2.3475706577301025, "learning_rate": 9.981460747519928e-05, "loss": 0.6937938690185547, "memory(GiB)": 57.83, "step": 3200, "token_acc": 0.8595317725752508, "train_speed(iter/s)": 1.514437 }, { "epoch": 0.13731202604858403, "grad_norm": 8.811067581176758, "learning_rate": 9.98140280306325e-05, "loss": 0.47876739501953125, "memory(GiB)": 57.83, "step": 3205, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.514595 }, { "epoch": 0.137526241377833, "grad_norm": 3.351703405380249, "learning_rate": 9.981344768363791e-05, "loss": 0.6727480888366699, "memory(GiB)": 57.83, "step": 3210, "token_acc": 0.8567251461988304, "train_speed(iter/s)": 1.514361 }, { "epoch": 0.13774045670708196, "grad_norm": 1.8882508277893066, "learning_rate": 9.981286643422596e-05, "loss": 0.3654142379760742, "memory(GiB)": 57.83, "step": 3215, "token_acc": 0.9276729559748428, "train_speed(iter/s)": 1.5143 }, { "epoch": 0.1379546720363309, "grad_norm": 3.439082622528076, "learning_rate": 9.981228428240721e-05, "loss": 0.34621386528015136, "memory(GiB)": 57.83, "step": 3220, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.514092 }, { "epoch": 0.1381688873655799, "grad_norm": 2.723161458969116, "learning_rate": 9.98117012281922e-05, "loss": 0.37717175483703613, "memory(GiB)": 57.83, "step": 3225, "token_acc": 0.9067524115755627, "train_speed(iter/s)": 1.514362 }, { "epoch": 0.13838310269482884, "grad_norm": 2.974764347076416, "learning_rate": 9.98111172715915e-05, "loss": 0.39750161170959475, "memory(GiB)": 57.83, "step": 3230, "token_acc": 0.898360655737705, "train_speed(iter/s)": 1.514271 }, { "epoch": 0.1385973180240778, "grad_norm": 2.8027303218841553, "learning_rate": 9.981053241261567e-05, "loss": 0.470366096496582, "memory(GiB)": 57.83, "step": 3235, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.51437 }, { "epoch": 0.13881153335332677, "grad_norm": 2.6217477321624756, "learning_rate": 9.980994665127535e-05, "loss": 0.5539867877960205, "memory(GiB)": 57.83, "step": 3240, "token_acc": 0.900709219858156, "train_speed(iter/s)": 1.514249 }, { "epoch": 0.13902574868257572, "grad_norm": 2.035142660140991, "learning_rate": 9.980935998758109e-05, "loss": 0.34348011016845703, "memory(GiB)": 57.83, "step": 3245, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.514194 }, { "epoch": 0.13923996401182467, "grad_norm": 3.680720806121826, "learning_rate": 9.980877242154356e-05, "loss": 0.41028032302856443, "memory(GiB)": 57.83, "step": 3250, "token_acc": 0.9372822299651568, "train_speed(iter/s)": 1.514283 }, { "epoch": 0.13945417934107365, "grad_norm": 1.3275294303894043, "learning_rate": 9.980818395317339e-05, "loss": 0.2973785877227783, "memory(GiB)": 57.83, "step": 3255, "token_acc": 0.9370860927152318, "train_speed(iter/s)": 1.514404 }, { "epoch": 0.1396683946703226, "grad_norm": 9.028099060058594, "learning_rate": 9.980759458248125e-05, "loss": 0.38358306884765625, "memory(GiB)": 57.83, "step": 3260, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.514218 }, { "epoch": 0.13988260999957156, "grad_norm": 9.966137886047363, "learning_rate": 9.98070043094778e-05, "loss": 0.7445281982421875, "memory(GiB)": 57.83, "step": 3265, "token_acc": 0.8394648829431438, "train_speed(iter/s)": 1.514261 }, { "epoch": 0.14009682532882053, "grad_norm": 4.5387749671936035, "learning_rate": 9.980641313417376e-05, "loss": 0.6203285217285156, "memory(GiB)": 57.83, "step": 3270, "token_acc": 0.8692579505300353, "train_speed(iter/s)": 1.514222 }, { "epoch": 0.14031104065806949, "grad_norm": 2.8319175243377686, "learning_rate": 9.98058210565798e-05, "loss": 0.3461270570755005, "memory(GiB)": 57.83, "step": 3275, "token_acc": 0.9158878504672897, "train_speed(iter/s)": 1.514171 }, { "epoch": 0.14052525598731846, "grad_norm": 3.8023581504821777, "learning_rate": 9.980522807670669e-05, "loss": 0.2738952159881592, "memory(GiB)": 57.83, "step": 3280, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.514289 }, { "epoch": 0.14073947131656742, "grad_norm": 1.7247800827026367, "learning_rate": 9.980463419456516e-05, "loss": 0.4290821075439453, "memory(GiB)": 57.83, "step": 3285, "token_acc": 0.9131944444444444, "train_speed(iter/s)": 1.514252 }, { "epoch": 0.14095368664581637, "grad_norm": 1.1629656553268433, "learning_rate": 9.980403941016594e-05, "loss": 0.3153548717498779, "memory(GiB)": 57.83, "step": 3290, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.514302 }, { "epoch": 0.14116790197506535, "grad_norm": 0.44684675335884094, "learning_rate": 9.980344372351983e-05, "loss": 0.41106529235839845, "memory(GiB)": 57.83, "step": 3295, "token_acc": 0.9088471849865952, "train_speed(iter/s)": 1.514324 }, { "epoch": 0.1413821173043143, "grad_norm": 2.70548677444458, "learning_rate": 9.980284713463763e-05, "loss": 0.4863539218902588, "memory(GiB)": 57.83, "step": 3300, "token_acc": 0.8921933085501859, "train_speed(iter/s)": 1.514554 }, { "epoch": 0.14159633263356325, "grad_norm": 0.2103116363286972, "learning_rate": 9.980224964353014e-05, "loss": 0.3223509550094604, "memory(GiB)": 57.83, "step": 3305, "token_acc": 0.9389067524115756, "train_speed(iter/s)": 1.514584 }, { "epoch": 0.14181054796281223, "grad_norm": 2.129995584487915, "learning_rate": 9.980165125020816e-05, "loss": 0.2904852867126465, "memory(GiB)": 57.83, "step": 3310, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.51457 }, { "epoch": 0.14202476329206118, "grad_norm": 7.964287281036377, "learning_rate": 9.980105195468256e-05, "loss": 0.32898569107055664, "memory(GiB)": 57.83, "step": 3315, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.514493 }, { "epoch": 0.14223897862131013, "grad_norm": 2.3501155376434326, "learning_rate": 9.980045175696418e-05, "loss": 0.46753315925598143, "memory(GiB)": 57.83, "step": 3320, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.514614 }, { "epoch": 0.1424531939505591, "grad_norm": 2.2071940898895264, "learning_rate": 9.97998506570639e-05, "loss": 0.2081993579864502, "memory(GiB)": 57.83, "step": 3325, "token_acc": 0.9548611111111112, "train_speed(iter/s)": 1.514744 }, { "epoch": 0.14266740927980806, "grad_norm": 1.1213449239730835, "learning_rate": 9.979924865499262e-05, "loss": 0.2008230209350586, "memory(GiB)": 57.83, "step": 3330, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.514742 }, { "epoch": 0.142881624609057, "grad_norm": 2.3316073417663574, "learning_rate": 9.979864575076124e-05, "loss": 0.4677445411682129, "memory(GiB)": 57.83, "step": 3335, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.514755 }, { "epoch": 0.143095839938306, "grad_norm": 2.3299098014831543, "learning_rate": 9.979804194438065e-05, "loss": 0.28764963150024414, "memory(GiB)": 57.83, "step": 3340, "token_acc": 0.9401993355481728, "train_speed(iter/s)": 1.514756 }, { "epoch": 0.14331005526755494, "grad_norm": 2.3503761291503906, "learning_rate": 9.979743723586184e-05, "loss": 0.4128252029418945, "memory(GiB)": 57.83, "step": 3345, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.51473 }, { "epoch": 0.1435242705968039, "grad_norm": 2.582024335861206, "learning_rate": 9.979683162521573e-05, "loss": 0.26147899627685545, "memory(GiB)": 57.83, "step": 3350, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.51491 }, { "epoch": 0.14373848592605287, "grad_norm": 11.51387882232666, "learning_rate": 9.97962251124533e-05, "loss": 0.5287271499633789, "memory(GiB)": 57.83, "step": 3355, "token_acc": 0.8828828828828829, "train_speed(iter/s)": 1.514887 }, { "epoch": 0.14395270125530182, "grad_norm": 2.950669288635254, "learning_rate": 9.979561769758555e-05, "loss": 0.15497267246246338, "memory(GiB)": 57.83, "step": 3360, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.515113 }, { "epoch": 0.1441669165845508, "grad_norm": 2.5429463386535645, "learning_rate": 9.979500938062345e-05, "loss": 0.500808572769165, "memory(GiB)": 57.83, "step": 3365, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.515429 }, { "epoch": 0.14438113191379975, "grad_norm": 2.239438772201538, "learning_rate": 9.979440016157805e-05, "loss": 0.3618643760681152, "memory(GiB)": 57.83, "step": 3370, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.51538 }, { "epoch": 0.1445953472430487, "grad_norm": 2.0689358711242676, "learning_rate": 9.97937900404604e-05, "loss": 0.45372705459594725, "memory(GiB)": 57.83, "step": 3375, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.515361 }, { "epoch": 0.14480956257229768, "grad_norm": 3.4086170196533203, "learning_rate": 9.979317901728153e-05, "loss": 0.28501780033111573, "memory(GiB)": 57.83, "step": 3380, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.515391 }, { "epoch": 0.14502377790154664, "grad_norm": 1.8076057434082031, "learning_rate": 9.979256709205251e-05, "loss": 0.40592327117919924, "memory(GiB)": 57.83, "step": 3385, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.515383 }, { "epoch": 0.1452379932307956, "grad_norm": 0.5522910356521606, "learning_rate": 9.979195426478443e-05, "loss": 0.44352116584777834, "memory(GiB)": 57.83, "step": 3390, "token_acc": 0.9057971014492754, "train_speed(iter/s)": 1.51541 }, { "epoch": 0.14545220856004457, "grad_norm": 7.993175983428955, "learning_rate": 9.979134053548838e-05, "loss": 0.46170997619628906, "memory(GiB)": 57.83, "step": 3395, "token_acc": 0.8939393939393939, "train_speed(iter/s)": 1.515358 }, { "epoch": 0.14566642388929352, "grad_norm": 1.7108323574066162, "learning_rate": 9.979072590417549e-05, "loss": 0.3712242364883423, "memory(GiB)": 57.83, "step": 3400, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.515782 }, { "epoch": 0.14588063921854247, "grad_norm": 2.7083516120910645, "learning_rate": 9.97901103708569e-05, "loss": 0.341778564453125, "memory(GiB)": 57.83, "step": 3405, "token_acc": 0.9227642276422764, "train_speed(iter/s)": 1.515989 }, { "epoch": 0.14609485454779145, "grad_norm": 3.2763168811798096, "learning_rate": 9.978949393554374e-05, "loss": 0.3628240585327148, "memory(GiB)": 57.83, "step": 3410, "token_acc": 0.901060070671378, "train_speed(iter/s)": 1.516191 }, { "epoch": 0.1463090698770404, "grad_norm": 0.768027663230896, "learning_rate": 9.978887659824721e-05, "loss": 0.25958895683288574, "memory(GiB)": 57.83, "step": 3415, "token_acc": 0.9529411764705882, "train_speed(iter/s)": 1.5161 }, { "epoch": 0.14652328520628935, "grad_norm": 2.428173065185547, "learning_rate": 9.978825835897844e-05, "loss": 0.1693664312362671, "memory(GiB)": 57.83, "step": 3420, "token_acc": 0.9519230769230769, "train_speed(iter/s)": 1.516289 }, { "epoch": 0.14673750053553833, "grad_norm": 5.774468898773193, "learning_rate": 9.978763921774869e-05, "loss": 0.5363493919372558, "memory(GiB)": 57.83, "step": 3425, "token_acc": 0.8970588235294118, "train_speed(iter/s)": 1.516823 }, { "epoch": 0.14695171586478728, "grad_norm": 2.377998113632202, "learning_rate": 9.978701917456916e-05, "loss": 0.29163658618927, "memory(GiB)": 57.83, "step": 3430, "token_acc": 0.9363057324840764, "train_speed(iter/s)": 1.516737 }, { "epoch": 0.14716593119403626, "grad_norm": 0.927766740322113, "learning_rate": 9.978639822945107e-05, "loss": 0.38797180652618407, "memory(GiB)": 57.83, "step": 3435, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.51684 }, { "epoch": 0.1473801465232852, "grad_norm": 2.2448580265045166, "learning_rate": 9.978577638240567e-05, "loss": 0.6224373817443848, "memory(GiB)": 57.83, "step": 3440, "token_acc": 0.8932806324110671, "train_speed(iter/s)": 1.517118 }, { "epoch": 0.14759436185253416, "grad_norm": 1.470595359802246, "learning_rate": 9.978515363344422e-05, "loss": 0.2460150957107544, "memory(GiB)": 57.83, "step": 3445, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.517185 }, { "epoch": 0.14780857718178314, "grad_norm": 1.8487331867218018, "learning_rate": 9.978452998257801e-05, "loss": 0.3197047710418701, "memory(GiB)": 57.83, "step": 3450, "token_acc": 0.9301587301587302, "train_speed(iter/s)": 1.517197 }, { "epoch": 0.1480227925110321, "grad_norm": 2.2068471908569336, "learning_rate": 9.978390542981835e-05, "loss": 0.43758363723754884, "memory(GiB)": 57.83, "step": 3455, "token_acc": 0.9112903225806451, "train_speed(iter/s)": 1.517198 }, { "epoch": 0.14823700784028104, "grad_norm": 4.254453182220459, "learning_rate": 9.978327997517652e-05, "loss": 0.5796511173248291, "memory(GiB)": 57.83, "step": 3460, "token_acc": 0.8765432098765432, "train_speed(iter/s)": 1.517209 }, { "epoch": 0.14845122316953002, "grad_norm": 2.888956069946289, "learning_rate": 9.978265361866389e-05, "loss": 0.5181376934051514, "memory(GiB)": 57.83, "step": 3465, "token_acc": 0.8897338403041825, "train_speed(iter/s)": 1.517204 }, { "epoch": 0.14866543849877897, "grad_norm": 2.9730756282806396, "learning_rate": 9.978202636029179e-05, "loss": 0.4953627586364746, "memory(GiB)": 57.83, "step": 3470, "token_acc": 0.8871951219512195, "train_speed(iter/s)": 1.517293 }, { "epoch": 0.14887965382802792, "grad_norm": 0.9205811023712158, "learning_rate": 9.978139820007158e-05, "loss": 0.13839194774627686, "memory(GiB)": 57.83, "step": 3475, "token_acc": 0.9656488549618321, "train_speed(iter/s)": 1.517491 }, { "epoch": 0.1490938691572769, "grad_norm": 5.189857006072998, "learning_rate": 9.978076913801464e-05, "loss": 0.3019744873046875, "memory(GiB)": 57.83, "step": 3480, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.517644 }, { "epoch": 0.14930808448652585, "grad_norm": 1.995017170906067, "learning_rate": 9.978013917413237e-05, "loss": 0.30265583992004397, "memory(GiB)": 57.83, "step": 3485, "token_acc": 0.9172185430463576, "train_speed(iter/s)": 1.517819 }, { "epoch": 0.1495222998157748, "grad_norm": 2.0031471252441406, "learning_rate": 9.977950830843619e-05, "loss": 0.5474436283111572, "memory(GiB)": 57.83, "step": 3490, "token_acc": 0.895774647887324, "train_speed(iter/s)": 1.518198 }, { "epoch": 0.14973651514502379, "grad_norm": 5.6422343254089355, "learning_rate": 9.977887654093751e-05, "loss": 0.2796337604522705, "memory(GiB)": 57.83, "step": 3495, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.518339 }, { "epoch": 0.14995073047427274, "grad_norm": 1.4210344552993774, "learning_rate": 9.977824387164778e-05, "loss": 0.40498552322387693, "memory(GiB)": 57.83, "step": 3500, "token_acc": 0.8809523809523809, "train_speed(iter/s)": 1.518428 }, { "epoch": 0.14995073047427274, "eval_loss": 2.446537733078003, "eval_runtime": 11.7222, "eval_samples_per_second": 8.531, "eval_steps_per_second": 8.531, "eval_token_acc": 0.42618741976893454, "step": 3500 }, { "epoch": 0.1501649458035217, "grad_norm": 2.3842875957489014, "learning_rate": 9.977761030057847e-05, "loss": 0.3279412746429443, "memory(GiB)": 57.83, "step": 3505, "token_acc": 0.5557692307692308, "train_speed(iter/s)": 1.510172 }, { "epoch": 0.15037916113277067, "grad_norm": 3.1697916984558105, "learning_rate": 9.977697582774106e-05, "loss": 0.9414101600646972, "memory(GiB)": 57.83, "step": 3510, "token_acc": 0.8172043010752689, "train_speed(iter/s)": 1.510172 }, { "epoch": 0.15059337646201962, "grad_norm": 0.06750098615884781, "learning_rate": 9.977634045314703e-05, "loss": 0.3700360059738159, "memory(GiB)": 57.83, "step": 3515, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.510253 }, { "epoch": 0.1508075917912686, "grad_norm": 2.2560830116271973, "learning_rate": 9.977570417680791e-05, "loss": 0.358535623550415, "memory(GiB)": 57.83, "step": 3520, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.510265 }, { "epoch": 0.15102180712051755, "grad_norm": 1.8879475593566895, "learning_rate": 9.977506699873521e-05, "loss": 0.38306117057800293, "memory(GiB)": 57.83, "step": 3525, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.510338 }, { "epoch": 0.1512360224497665, "grad_norm": 0.23100729286670685, "learning_rate": 9.977442891894048e-05, "loss": 0.5975110054016113, "memory(GiB)": 57.83, "step": 3530, "token_acc": 0.8785714285714286, "train_speed(iter/s)": 1.51053 }, { "epoch": 0.15145023777901548, "grad_norm": 2.2452738285064697, "learning_rate": 9.977378993743528e-05, "loss": 0.28678576946258544, "memory(GiB)": 57.83, "step": 3535, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.510815 }, { "epoch": 0.15166445310826443, "grad_norm": 2.090545654296875, "learning_rate": 9.977315005423117e-05, "loss": 0.5024127960205078, "memory(GiB)": 57.83, "step": 3540, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.510948 }, { "epoch": 0.15187866843751338, "grad_norm": 5.876445293426514, "learning_rate": 9.977250926933977e-05, "loss": 0.17773880958557128, "memory(GiB)": 57.83, "step": 3545, "token_acc": 0.9554794520547946, "train_speed(iter/s)": 1.511203 }, { "epoch": 0.15209288376676236, "grad_norm": 2.387523651123047, "learning_rate": 9.977186758277268e-05, "loss": 0.5856736183166504, "memory(GiB)": 57.83, "step": 3550, "token_acc": 0.887240356083086, "train_speed(iter/s)": 1.511117 }, { "epoch": 0.1523070990960113, "grad_norm": 2.739640712738037, "learning_rate": 9.97712249945415e-05, "loss": 0.3147712707519531, "memory(GiB)": 57.83, "step": 3555, "token_acc": 0.9375, "train_speed(iter/s)": 1.511172 }, { "epoch": 0.15252131442526026, "grad_norm": 3.8068037033081055, "learning_rate": 9.97705815046579e-05, "loss": 0.5998228073120118, "memory(GiB)": 57.83, "step": 3560, "token_acc": 0.8706293706293706, "train_speed(iter/s)": 1.511133 }, { "epoch": 0.15273552975450924, "grad_norm": 3.3296754360198975, "learning_rate": 9.976993711313354e-05, "loss": 0.5841238021850585, "memory(GiB)": 57.83, "step": 3565, "token_acc": 0.8771929824561403, "train_speed(iter/s)": 1.511068 }, { "epoch": 0.1529497450837582, "grad_norm": 1.73323655128479, "learning_rate": 9.976929181998007e-05, "loss": 0.33115923404693604, "memory(GiB)": 57.83, "step": 3570, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.510977 }, { "epoch": 0.15316396041300714, "grad_norm": 8.753364562988281, "learning_rate": 9.976864562520918e-05, "loss": 0.44001312255859376, "memory(GiB)": 57.83, "step": 3575, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.510928 }, { "epoch": 0.15337817574225612, "grad_norm": 0.8948017954826355, "learning_rate": 9.976799852883261e-05, "loss": 0.23592422008514405, "memory(GiB)": 57.83, "step": 3580, "token_acc": 0.944, "train_speed(iter/s)": 1.511017 }, { "epoch": 0.15359239107150507, "grad_norm": 2.7720398902893066, "learning_rate": 9.976735053086204e-05, "loss": 0.36278018951416013, "memory(GiB)": 57.83, "step": 3585, "token_acc": 0.9022222222222223, "train_speed(iter/s)": 1.510987 }, { "epoch": 0.15380660640075403, "grad_norm": 2.427790403366089, "learning_rate": 9.976670163130924e-05, "loss": 0.31174402236938475, "memory(GiB)": 57.83, "step": 3590, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.511183 }, { "epoch": 0.154020821730003, "grad_norm": 3.367298126220703, "learning_rate": 9.976605183018594e-05, "loss": 0.33286137580871583, "memory(GiB)": 57.83, "step": 3595, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.511204 }, { "epoch": 0.15423503705925196, "grad_norm": 11.743346214294434, "learning_rate": 9.976540112750394e-05, "loss": 0.460723352432251, "memory(GiB)": 57.83, "step": 3600, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.511203 }, { "epoch": 0.15444925238850093, "grad_norm": 2.4169223308563232, "learning_rate": 9.976474952327502e-05, "loss": 0.6255918979644776, "memory(GiB)": 57.83, "step": 3605, "token_acc": 0.8586206896551725, "train_speed(iter/s)": 1.511118 }, { "epoch": 0.1546634677177499, "grad_norm": 1.7793593406677246, "learning_rate": 9.976409701751097e-05, "loss": 0.4173449993133545, "memory(GiB)": 57.83, "step": 3610, "token_acc": 0.9058823529411765, "train_speed(iter/s)": 1.511072 }, { "epoch": 0.15487768304699884, "grad_norm": 2.0201361179351807, "learning_rate": 9.976344361022363e-05, "loss": 0.3983342409133911, "memory(GiB)": 57.83, "step": 3615, "token_acc": 0.9107142857142857, "train_speed(iter/s)": 1.511149 }, { "epoch": 0.15509189837624782, "grad_norm": 3.886936664581299, "learning_rate": 9.976278930142482e-05, "loss": 0.4150503635406494, "memory(GiB)": 57.83, "step": 3620, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.511125 }, { "epoch": 0.15530611370549677, "grad_norm": 0.9845520853996277, "learning_rate": 9.97621340911264e-05, "loss": 0.4149161815643311, "memory(GiB)": 57.83, "step": 3625, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.511237 }, { "epoch": 0.15552032903474572, "grad_norm": 1.4967237710952759, "learning_rate": 9.976147797934024e-05, "loss": 0.41015195846557617, "memory(GiB)": 57.83, "step": 3630, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.51135 }, { "epoch": 0.1557345443639947, "grad_norm": 3.5723297595977783, "learning_rate": 9.976082096607823e-05, "loss": 0.3702225923538208, "memory(GiB)": 57.83, "step": 3635, "token_acc": 0.9158576051779935, "train_speed(iter/s)": 1.511261 }, { "epoch": 0.15594875969324365, "grad_norm": 4.845510482788086, "learning_rate": 9.976016305135228e-05, "loss": 0.4890450477600098, "memory(GiB)": 57.83, "step": 3640, "token_acc": 0.9093959731543624, "train_speed(iter/s)": 1.511308 }, { "epoch": 0.1561629750224926, "grad_norm": 1.578256607055664, "learning_rate": 9.975950423517426e-05, "loss": 0.17587542533874512, "memory(GiB)": 57.83, "step": 3645, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.511392 }, { "epoch": 0.15637719035174158, "grad_norm": 1.2948048114776611, "learning_rate": 9.975884451755618e-05, "loss": 0.35112183094024657, "memory(GiB)": 57.83, "step": 3650, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.511419 }, { "epoch": 0.15659140568099053, "grad_norm": 2.8147289752960205, "learning_rate": 9.975818389850993e-05, "loss": 0.5415332794189454, "memory(GiB)": 57.83, "step": 3655, "token_acc": 0.8849557522123894, "train_speed(iter/s)": 1.51138 }, { "epoch": 0.15680562101023948, "grad_norm": 1.5178312063217163, "learning_rate": 9.975752237804752e-05, "loss": 0.6440974712371826, "memory(GiB)": 57.83, "step": 3660, "token_acc": 0.8806584362139918, "train_speed(iter/s)": 1.511509 }, { "epoch": 0.15701983633948846, "grad_norm": 1.194933533668518, "learning_rate": 9.975685995618092e-05, "loss": 0.5662200927734375, "memory(GiB)": 57.83, "step": 3665, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.511504 }, { "epoch": 0.1572340516687374, "grad_norm": 4.714276313781738, "learning_rate": 9.97561966329221e-05, "loss": 0.29403371810913087, "memory(GiB)": 57.83, "step": 3670, "token_acc": 0.9367588932806324, "train_speed(iter/s)": 1.511465 }, { "epoch": 0.15744826699798636, "grad_norm": 1.2164779901504517, "learning_rate": 9.975553240828312e-05, "loss": 0.3515486240386963, "memory(GiB)": 57.83, "step": 3675, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.511469 }, { "epoch": 0.15766248232723534, "grad_norm": 4.283686637878418, "learning_rate": 9.975486728227599e-05, "loss": 0.48273282051086425, "memory(GiB)": 57.83, "step": 3680, "token_acc": 0.8970976253298153, "train_speed(iter/s)": 1.51157 }, { "epoch": 0.1578766976564843, "grad_norm": 3.567147970199585, "learning_rate": 9.975420125491277e-05, "loss": 0.32115557193756106, "memory(GiB)": 57.83, "step": 3685, "token_acc": 0.9241379310344827, "train_speed(iter/s)": 1.511696 }, { "epoch": 0.15809091298573327, "grad_norm": 3.0105631351470947, "learning_rate": 9.975353432620552e-05, "loss": 0.2973757743835449, "memory(GiB)": 57.83, "step": 3690, "token_acc": 0.9371069182389937, "train_speed(iter/s)": 1.511752 }, { "epoch": 0.15830512831498222, "grad_norm": 3.013645648956299, "learning_rate": 9.975286649616633e-05, "loss": 0.4880228042602539, "memory(GiB)": 57.83, "step": 3695, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.511694 }, { "epoch": 0.15851934364423118, "grad_norm": 2.7897558212280273, "learning_rate": 9.975219776480727e-05, "loss": 0.40114269256591795, "memory(GiB)": 57.83, "step": 3700, "token_acc": 0.9153846153846154, "train_speed(iter/s)": 1.511824 }, { "epoch": 0.15873355897348015, "grad_norm": 4.134271144866943, "learning_rate": 9.97515281321405e-05, "loss": 0.7008815765380859, "memory(GiB)": 57.83, "step": 3705, "token_acc": 0.88, "train_speed(iter/s)": 1.512143 }, { "epoch": 0.1589477743027291, "grad_norm": 1.9787652492523193, "learning_rate": 9.975085759817811e-05, "loss": 0.30160953998565676, "memory(GiB)": 57.83, "step": 3710, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.512193 }, { "epoch": 0.15916198963197806, "grad_norm": 1.9093937873840332, "learning_rate": 9.975018616293228e-05, "loss": 0.227117919921875, "memory(GiB)": 57.83, "step": 3715, "token_acc": 0.9561752988047809, "train_speed(iter/s)": 1.512415 }, { "epoch": 0.15937620496122704, "grad_norm": 3.28204607963562, "learning_rate": 9.974951382641516e-05, "loss": 0.43659658432006837, "memory(GiB)": 57.83, "step": 3720, "token_acc": 0.8904593639575972, "train_speed(iter/s)": 1.512471 }, { "epoch": 0.159590420290476, "grad_norm": 2.5080363750457764, "learning_rate": 9.97488405886389e-05, "loss": 0.31663401126861573, "memory(GiB)": 57.83, "step": 3725, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.512581 }, { "epoch": 0.15980463561972494, "grad_norm": 0.7743328213691711, "learning_rate": 9.974816644961575e-05, "loss": 0.21046299934387208, "memory(GiB)": 57.83, "step": 3730, "token_acc": 0.9429429429429429, "train_speed(iter/s)": 1.512683 }, { "epoch": 0.16001885094897392, "grad_norm": 0.660235583782196, "learning_rate": 9.974749140935789e-05, "loss": 0.52157621383667, "memory(GiB)": 57.83, "step": 3735, "token_acc": 0.8796992481203008, "train_speed(iter/s)": 1.512754 }, { "epoch": 0.16023306627822287, "grad_norm": 4.521226406097412, "learning_rate": 9.974681546787755e-05, "loss": 0.4067714214324951, "memory(GiB)": 57.83, "step": 3740, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.512751 }, { "epoch": 0.16044728160747182, "grad_norm": 1.949479579925537, "learning_rate": 9.974613862518699e-05, "loss": 0.2712433338165283, "memory(GiB)": 57.83, "step": 3745, "token_acc": 0.9527272727272728, "train_speed(iter/s)": 1.512745 }, { "epoch": 0.1606614969367208, "grad_norm": 1.6037927865982056, "learning_rate": 9.974546088129846e-05, "loss": 0.3681286096572876, "memory(GiB)": 57.83, "step": 3750, "token_acc": 0.9144736842105263, "train_speed(iter/s)": 1.512854 }, { "epoch": 0.16087571226596975, "grad_norm": 3.949721097946167, "learning_rate": 9.974478223622424e-05, "loss": 0.48051066398620607, "memory(GiB)": 57.83, "step": 3755, "token_acc": 0.8810810810810811, "train_speed(iter/s)": 1.513012 }, { "epoch": 0.1610899275952187, "grad_norm": 1.9430617094039917, "learning_rate": 9.974410268997662e-05, "loss": 0.663322114944458, "memory(GiB)": 57.83, "step": 3760, "token_acc": 0.8722627737226277, "train_speed(iter/s)": 1.513232 }, { "epoch": 0.16130414292446768, "grad_norm": 1.7300478219985962, "learning_rate": 9.974342224256795e-05, "loss": 0.38237872123718264, "memory(GiB)": 57.83, "step": 3765, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.513162 }, { "epoch": 0.16151835825371663, "grad_norm": 2.5673322677612305, "learning_rate": 9.974274089401048e-05, "loss": 0.2913210391998291, "memory(GiB)": 57.83, "step": 3770, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.513221 }, { "epoch": 0.1617325735829656, "grad_norm": 1.1327354907989502, "learning_rate": 9.974205864431661e-05, "loss": 0.33402233123779296, "memory(GiB)": 57.83, "step": 3775, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.513293 }, { "epoch": 0.16194678891221456, "grad_norm": 7.10791540145874, "learning_rate": 9.974137549349869e-05, "loss": 0.6685216903686524, "memory(GiB)": 57.83, "step": 3780, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.513495 }, { "epoch": 0.1621610042414635, "grad_norm": 1.6271249055862427, "learning_rate": 9.97406914415691e-05, "loss": 0.2500523567199707, "memory(GiB)": 57.83, "step": 3785, "token_acc": 0.9418604651162791, "train_speed(iter/s)": 1.513489 }, { "epoch": 0.1623752195707125, "grad_norm": 0.12042249739170074, "learning_rate": 9.97400064885402e-05, "loss": 0.5365549564361572, "memory(GiB)": 57.83, "step": 3790, "token_acc": 0.89937106918239, "train_speed(iter/s)": 1.51342 }, { "epoch": 0.16258943489996144, "grad_norm": 5.95877742767334, "learning_rate": 9.973932063442444e-05, "loss": 0.23064014911651612, "memory(GiB)": 57.83, "step": 3795, "token_acc": 0.9511400651465798, "train_speed(iter/s)": 1.513404 }, { "epoch": 0.1628036502292104, "grad_norm": 2.0255298614501953, "learning_rate": 9.973863387923423e-05, "loss": 0.2860123634338379, "memory(GiB)": 57.83, "step": 3800, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.513495 }, { "epoch": 0.16301786555845937, "grad_norm": 12.810190200805664, "learning_rate": 9.9737946222982e-05, "loss": 0.4587386131286621, "memory(GiB)": 57.83, "step": 3805, "token_acc": 0.901595744680851, "train_speed(iter/s)": 1.513465 }, { "epoch": 0.16323208088770833, "grad_norm": 1.8858866691589355, "learning_rate": 9.973725766568023e-05, "loss": 0.44912099838256836, "memory(GiB)": 57.83, "step": 3810, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.513431 }, { "epoch": 0.16344629621695728, "grad_norm": 2.2747585773468018, "learning_rate": 9.973656820734136e-05, "loss": 0.25021114349365237, "memory(GiB)": 57.83, "step": 3815, "token_acc": 0.9496124031007752, "train_speed(iter/s)": 1.513455 }, { "epoch": 0.16366051154620626, "grad_norm": 2.7039852142333984, "learning_rate": 9.973587784797791e-05, "loss": 0.5598416328430176, "memory(GiB)": 57.83, "step": 3820, "token_acc": 0.8706293706293706, "train_speed(iter/s)": 1.513363 }, { "epoch": 0.1638747268754552, "grad_norm": 1.99837327003479, "learning_rate": 9.973518658760236e-05, "loss": 0.35742576122283937, "memory(GiB)": 57.83, "step": 3825, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.513227 }, { "epoch": 0.16408894220470416, "grad_norm": 3.08686900138855, "learning_rate": 9.973449442622729e-05, "loss": 0.392643141746521, "memory(GiB)": 57.83, "step": 3830, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.513248 }, { "epoch": 0.16430315753395314, "grad_norm": 2.2556662559509277, "learning_rate": 9.973380136386516e-05, "loss": 0.48528685569763186, "memory(GiB)": 57.83, "step": 3835, "token_acc": 0.9087947882736156, "train_speed(iter/s)": 1.513416 }, { "epoch": 0.1645173728632021, "grad_norm": 11.768698692321777, "learning_rate": 9.973310740052859e-05, "loss": 0.754575252532959, "memory(GiB)": 57.83, "step": 3840, "token_acc": 0.8659420289855072, "train_speed(iter/s)": 1.51357 }, { "epoch": 0.16473158819245104, "grad_norm": 2.8291118144989014, "learning_rate": 9.973241253623011e-05, "loss": 0.4900012969970703, "memory(GiB)": 57.83, "step": 3845, "token_acc": 0.9077490774907749, "train_speed(iter/s)": 1.513732 }, { "epoch": 0.16494580352170002, "grad_norm": 2.513559579849243, "learning_rate": 9.973171677098233e-05, "loss": 0.27779397964477537, "memory(GiB)": 57.83, "step": 3850, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.513798 }, { "epoch": 0.16516001885094897, "grad_norm": 2.3493876457214355, "learning_rate": 9.973102010479784e-05, "loss": 0.5915249347686767, "memory(GiB)": 57.83, "step": 3855, "token_acc": 0.8430769230769231, "train_speed(iter/s)": 1.513863 }, { "epoch": 0.16537423418019795, "grad_norm": 5.322275161743164, "learning_rate": 9.973032253768927e-05, "loss": 0.22183570861816407, "memory(GiB)": 57.83, "step": 3860, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.514068 }, { "epoch": 0.1655884495094469, "grad_norm": 3.158512592315674, "learning_rate": 9.972962406966928e-05, "loss": 0.23666105270385743, "memory(GiB)": 57.83, "step": 3865, "token_acc": 0.9408602150537635, "train_speed(iter/s)": 1.514112 }, { "epoch": 0.16580266483869585, "grad_norm": 2.108837127685547, "learning_rate": 9.972892470075048e-05, "loss": 0.2621962547302246, "memory(GiB)": 57.83, "step": 3870, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.514254 }, { "epoch": 0.16601688016794483, "grad_norm": 5.688636779785156, "learning_rate": 9.972822443094557e-05, "loss": 0.679051685333252, "memory(GiB)": 57.83, "step": 3875, "token_acc": 0.8713826366559485, "train_speed(iter/s)": 1.514122 }, { "epoch": 0.16623109549719378, "grad_norm": 2.2759616374969482, "learning_rate": 9.972752326026722e-05, "loss": 0.3941002368927002, "memory(GiB)": 57.83, "step": 3880, "token_acc": 0.9232876712328767, "train_speed(iter/s)": 1.513998 }, { "epoch": 0.16644531082644273, "grad_norm": 2.3648221492767334, "learning_rate": 9.972682118872813e-05, "loss": 0.25297958850860597, "memory(GiB)": 57.83, "step": 3885, "token_acc": 0.9346938775510204, "train_speed(iter/s)": 1.51403 }, { "epoch": 0.1666595261556917, "grad_norm": 6.614083766937256, "learning_rate": 9.972611821634104e-05, "loss": 0.343418550491333, "memory(GiB)": 57.83, "step": 3890, "token_acc": 0.9224489795918367, "train_speed(iter/s)": 1.514102 }, { "epoch": 0.16687374148494066, "grad_norm": 1.608016848564148, "learning_rate": 9.972541434311866e-05, "loss": 0.3323129892349243, "memory(GiB)": 57.83, "step": 3895, "token_acc": 0.9331210191082803, "train_speed(iter/s)": 1.51414 }, { "epoch": 0.16708795681418961, "grad_norm": 1.1930209398269653, "learning_rate": 9.972470956907375e-05, "loss": 0.2424311876296997, "memory(GiB)": 57.83, "step": 3900, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.51436 }, { "epoch": 0.1673021721434386, "grad_norm": 1.5823094844818115, "learning_rate": 9.97240038942191e-05, "loss": 0.659721851348877, "memory(GiB)": 57.83, "step": 3905, "token_acc": 0.871875, "train_speed(iter/s)": 1.514245 }, { "epoch": 0.16751638747268754, "grad_norm": 5.76854944229126, "learning_rate": 9.972329731856745e-05, "loss": 0.37930200099945066, "memory(GiB)": 57.83, "step": 3910, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.514499 }, { "epoch": 0.1677306028019365, "grad_norm": 3.249678611755371, "learning_rate": 9.972258984213164e-05, "loss": 0.5554678916931153, "memory(GiB)": 57.83, "step": 3915, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.514611 }, { "epoch": 0.16794481813118547, "grad_norm": 3.475437641143799, "learning_rate": 9.972188146492446e-05, "loss": 0.39207236766815184, "memory(GiB)": 57.83, "step": 3920, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.514608 }, { "epoch": 0.16815903346043443, "grad_norm": 2.7875428199768066, "learning_rate": 9.972117218695877e-05, "loss": 0.6198910713195801, "memory(GiB)": 57.83, "step": 3925, "token_acc": 0.8743016759776536, "train_speed(iter/s)": 1.514699 }, { "epoch": 0.16837324878968338, "grad_norm": 4.385772705078125, "learning_rate": 9.97204620082474e-05, "loss": 0.7625330924987793, "memory(GiB)": 57.83, "step": 3930, "token_acc": 0.8453947368421053, "train_speed(iter/s)": 1.514645 }, { "epoch": 0.16858746411893236, "grad_norm": 0.49179938435554504, "learning_rate": 9.971975092880321e-05, "loss": 0.16669465303421022, "memory(GiB)": 57.83, "step": 3935, "token_acc": 0.9552845528455285, "train_speed(iter/s)": 1.514604 }, { "epoch": 0.1688016794481813, "grad_norm": 3.707549571990967, "learning_rate": 9.97190389486391e-05, "loss": 0.5366193771362304, "memory(GiB)": 57.83, "step": 3940, "token_acc": 0.8802395209580839, "train_speed(iter/s)": 1.514571 }, { "epoch": 0.1690158947774303, "grad_norm": 4.463767051696777, "learning_rate": 9.971832606776797e-05, "loss": 0.4810950756072998, "memory(GiB)": 57.83, "step": 3945, "token_acc": 0.9184397163120568, "train_speed(iter/s)": 1.514486 }, { "epoch": 0.16923011010667924, "grad_norm": 2.6431195735931396, "learning_rate": 9.97176122862027e-05, "loss": 0.3314813613891602, "memory(GiB)": 57.83, "step": 3950, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.51441 }, { "epoch": 0.1694443254359282, "grad_norm": 3.0546669960021973, "learning_rate": 9.971689760395626e-05, "loss": 0.34649157524108887, "memory(GiB)": 57.83, "step": 3955, "token_acc": 0.937984496124031, "train_speed(iter/s)": 1.514399 }, { "epoch": 0.16965854076517717, "grad_norm": 1.5945426225662231, "learning_rate": 9.971618202104157e-05, "loss": 0.501087760925293, "memory(GiB)": 57.83, "step": 3960, "token_acc": 0.9002493765586035, "train_speed(iter/s)": 1.514314 }, { "epoch": 0.16987275609442612, "grad_norm": 4.94426965713501, "learning_rate": 9.971546553747163e-05, "loss": 0.3618204116821289, "memory(GiB)": 57.83, "step": 3965, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.514295 }, { "epoch": 0.17008697142367507, "grad_norm": 2.4892890453338623, "learning_rate": 9.971474815325936e-05, "loss": 0.378057861328125, "memory(GiB)": 57.83, "step": 3970, "token_acc": 0.8939929328621908, "train_speed(iter/s)": 1.514401 }, { "epoch": 0.17030118675292405, "grad_norm": 4.393266677856445, "learning_rate": 9.97140298684178e-05, "loss": 0.5813199043273926, "memory(GiB)": 57.83, "step": 3975, "token_acc": 0.8737864077669902, "train_speed(iter/s)": 1.51456 }, { "epoch": 0.170515402082173, "grad_norm": 2.829508066177368, "learning_rate": 9.971331068295998e-05, "loss": 0.4677761077880859, "memory(GiB)": 57.83, "step": 3980, "token_acc": 0.8837209302325582, "train_speed(iter/s)": 1.51465 }, { "epoch": 0.17072961741142195, "grad_norm": 1.861405611038208, "learning_rate": 9.971259059689887e-05, "loss": 0.1899072051048279, "memory(GiB)": 57.83, "step": 3985, "token_acc": 0.9569230769230769, "train_speed(iter/s)": 1.514715 }, { "epoch": 0.17094383274067093, "grad_norm": 1.2325413227081299, "learning_rate": 9.971186961024756e-05, "loss": 0.3861642360687256, "memory(GiB)": 57.83, "step": 3990, "token_acc": 0.9044585987261147, "train_speed(iter/s)": 1.514873 }, { "epoch": 0.17115804806991988, "grad_norm": 0.7051987648010254, "learning_rate": 9.97111477230191e-05, "loss": 0.36649236679077146, "memory(GiB)": 57.83, "step": 3995, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.515001 }, { "epoch": 0.17137226339916883, "grad_norm": 6.859620094299316, "learning_rate": 9.971042493522655e-05, "loss": 0.5443126201629639, "memory(GiB)": 57.83, "step": 4000, "token_acc": 0.8884892086330936, "train_speed(iter/s)": 1.515298 }, { "epoch": 0.17137226339916883, "eval_loss": 2.59641170501709, "eval_runtime": 11.2737, "eval_samples_per_second": 8.87, "eval_steps_per_second": 8.87, "eval_token_acc": 0.41797432239657634, "step": 4000 }, { "epoch": 0.1715864787284178, "grad_norm": 1.7579963207244873, "learning_rate": 9.970970124688305e-05, "loss": 0.344852089881897, "memory(GiB)": 57.83, "step": 4005, "token_acc": 0.5631951466127402, "train_speed(iter/s)": 1.508328 }, { "epoch": 0.17180069405766676, "grad_norm": 2.772745132446289, "learning_rate": 9.970897665800167e-05, "loss": 0.8416284561157227, "memory(GiB)": 57.83, "step": 4010, "token_acc": 0.8092105263157895, "train_speed(iter/s)": 1.508294 }, { "epoch": 0.17201490938691572, "grad_norm": 1.6289725303649902, "learning_rate": 9.970825116859554e-05, "loss": 0.24476804733276367, "memory(GiB)": 57.83, "step": 4015, "token_acc": 0.9446254071661238, "train_speed(iter/s)": 1.508382 }, { "epoch": 0.1722291247161647, "grad_norm": 4.644321441650391, "learning_rate": 9.97075247786778e-05, "loss": 0.26848759651184084, "memory(GiB)": 57.83, "step": 4020, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.50868 }, { "epoch": 0.17244334004541365, "grad_norm": 1.5828018188476562, "learning_rate": 9.970679748826164e-05, "loss": 0.32279582023620607, "memory(GiB)": 57.83, "step": 4025, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.50879 }, { "epoch": 0.17265755537466262, "grad_norm": 2.312274217605591, "learning_rate": 9.97060692973602e-05, "loss": 0.38735003471374513, "memory(GiB)": 57.83, "step": 4030, "token_acc": 0.9144542772861357, "train_speed(iter/s)": 1.508841 }, { "epoch": 0.17287177070391158, "grad_norm": 7.7694830894470215, "learning_rate": 9.97053402059867e-05, "loss": 0.3685455799102783, "memory(GiB)": 57.83, "step": 4035, "token_acc": 0.927797833935018, "train_speed(iter/s)": 1.509093 }, { "epoch": 0.17308598603316053, "grad_norm": 10.439327239990234, "learning_rate": 9.970461021415433e-05, "loss": 0.42048869132995603, "memory(GiB)": 57.83, "step": 4040, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.509127 }, { "epoch": 0.1733002013624095, "grad_norm": 5.950136184692383, "learning_rate": 9.97038793218763e-05, "loss": 0.43625383377075194, "memory(GiB)": 57.83, "step": 4045, "token_acc": 0.9102564102564102, "train_speed(iter/s)": 1.509342 }, { "epoch": 0.17351441669165846, "grad_norm": 4.210340976715088, "learning_rate": 9.97031475291659e-05, "loss": 0.29321064949035647, "memory(GiB)": 57.83, "step": 4050, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.509316 }, { "epoch": 0.1737286320209074, "grad_norm": 3.2795848846435547, "learning_rate": 9.970241483603635e-05, "loss": 0.45979628562927244, "memory(GiB)": 57.83, "step": 4055, "token_acc": 0.8844984802431611, "train_speed(iter/s)": 1.509377 }, { "epoch": 0.1739428473501564, "grad_norm": 7.25145149230957, "learning_rate": 9.970168124250093e-05, "loss": 0.347960090637207, "memory(GiB)": 57.83, "step": 4060, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.509588 }, { "epoch": 0.17415706267940534, "grad_norm": 3.4471166133880615, "learning_rate": 9.970094674857291e-05, "loss": 0.2726967096328735, "memory(GiB)": 57.83, "step": 4065, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.509616 }, { "epoch": 0.1743712780086543, "grad_norm": 2.9118540287017822, "learning_rate": 9.970021135426564e-05, "loss": 0.27073240280151367, "memory(GiB)": 57.83, "step": 4070, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.509633 }, { "epoch": 0.17458549333790327, "grad_norm": 3.7712721824645996, "learning_rate": 9.969947505959241e-05, "loss": 0.6009779930114746, "memory(GiB)": 57.83, "step": 4075, "token_acc": 0.8643410852713178, "train_speed(iter/s)": 1.509627 }, { "epoch": 0.17479970866715222, "grad_norm": 2.1265647411346436, "learning_rate": 9.969873786456656e-05, "loss": 0.1461083173751831, "memory(GiB)": 57.83, "step": 4080, "token_acc": 0.9673202614379085, "train_speed(iter/s)": 1.509627 }, { "epoch": 0.17501392399640117, "grad_norm": 5.783075332641602, "learning_rate": 9.969799976920147e-05, "loss": 0.6032055854797364, "memory(GiB)": 57.83, "step": 4085, "token_acc": 0.8757961783439491, "train_speed(iter/s)": 1.509451 }, { "epoch": 0.17522813932565015, "grad_norm": 2.392221689224243, "learning_rate": 9.969726077351047e-05, "loss": 0.33434596061706545, "memory(GiB)": 57.83, "step": 4090, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.509466 }, { "epoch": 0.1754423546548991, "grad_norm": 3.594744920730591, "learning_rate": 9.969652087750698e-05, "loss": 0.28260555267333987, "memory(GiB)": 57.83, "step": 4095, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.509403 }, { "epoch": 0.17565656998414805, "grad_norm": 5.460442066192627, "learning_rate": 9.969578008120441e-05, "loss": 0.5577815055847168, "memory(GiB)": 57.83, "step": 4100, "token_acc": 0.8711864406779661, "train_speed(iter/s)": 1.509502 }, { "epoch": 0.17587078531339703, "grad_norm": 3.3070948123931885, "learning_rate": 9.969503838461615e-05, "loss": 0.4042186737060547, "memory(GiB)": 57.83, "step": 4105, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.509499 }, { "epoch": 0.17608500064264598, "grad_norm": 1.6151496171951294, "learning_rate": 9.969429578775567e-05, "loss": 0.31232290267944335, "memory(GiB)": 57.83, "step": 4110, "token_acc": 0.9391304347826087, "train_speed(iter/s)": 1.509523 }, { "epoch": 0.17629921597189496, "grad_norm": 0.3587826192378998, "learning_rate": 9.969355229063638e-05, "loss": 0.15982770919799805, "memory(GiB)": 57.83, "step": 4115, "token_acc": 0.959409594095941, "train_speed(iter/s)": 1.509521 }, { "epoch": 0.1765134313011439, "grad_norm": 2.4136569499969482, "learning_rate": 9.969280789327179e-05, "loss": 0.46406307220458987, "memory(GiB)": 57.83, "step": 4120, "token_acc": 0.913946587537092, "train_speed(iter/s)": 1.50962 }, { "epoch": 0.17672764663039287, "grad_norm": 2.0314478874206543, "learning_rate": 9.969206259567537e-05, "loss": 0.39525983333587644, "memory(GiB)": 57.83, "step": 4125, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.509641 }, { "epoch": 0.17694186195964184, "grad_norm": 0.7080110311508179, "learning_rate": 9.969131639786061e-05, "loss": 0.21221537590026857, "memory(GiB)": 57.83, "step": 4130, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.509516 }, { "epoch": 0.1771560772888908, "grad_norm": 1.65465247631073, "learning_rate": 9.969056929984105e-05, "loss": 0.47389492988586424, "memory(GiB)": 57.83, "step": 4135, "token_acc": 0.9027237354085603, "train_speed(iter/s)": 1.509454 }, { "epoch": 0.17737029261813975, "grad_norm": 0.24332356452941895, "learning_rate": 9.968982130163021e-05, "loss": 0.23363909721374512, "memory(GiB)": 57.83, "step": 4140, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.509566 }, { "epoch": 0.17758450794738873, "grad_norm": 3.212531328201294, "learning_rate": 9.968907240324165e-05, "loss": 0.6157785415649414, "memory(GiB)": 57.83, "step": 4145, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.509694 }, { "epoch": 0.17779872327663768, "grad_norm": 1.7138736248016357, "learning_rate": 9.968832260468892e-05, "loss": 0.33520939350128176, "memory(GiB)": 57.83, "step": 4150, "token_acc": 0.9213483146067416, "train_speed(iter/s)": 1.509703 }, { "epoch": 0.17801293860588663, "grad_norm": 2.3866288661956787, "learning_rate": 9.96875719059856e-05, "loss": 0.24670026302337647, "memory(GiB)": 57.83, "step": 4155, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.509696 }, { "epoch": 0.1782271539351356, "grad_norm": 2.5758602619171143, "learning_rate": 9.968682030714534e-05, "loss": 0.44811391830444336, "memory(GiB)": 57.83, "step": 4160, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.510023 }, { "epoch": 0.17844136926438456, "grad_norm": 3.3657913208007812, "learning_rate": 9.96860678081817e-05, "loss": 0.3874398708343506, "memory(GiB)": 57.83, "step": 4165, "token_acc": 0.9171428571428571, "train_speed(iter/s)": 1.510337 }, { "epoch": 0.1786555845936335, "grad_norm": 1.790096640586853, "learning_rate": 9.968531440910835e-05, "loss": 0.18798787593841554, "memory(GiB)": 57.83, "step": 4170, "token_acc": 0.9483870967741935, "train_speed(iter/s)": 1.51045 }, { "epoch": 0.1788697999228825, "grad_norm": 1.8477568626403809, "learning_rate": 9.96845601099389e-05, "loss": 0.5316804409027099, "memory(GiB)": 57.83, "step": 4175, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.510462 }, { "epoch": 0.17908401525213144, "grad_norm": 1.6212821006774902, "learning_rate": 9.968380491068705e-05, "loss": 0.48261322975158694, "memory(GiB)": 57.83, "step": 4180, "token_acc": 0.9003831417624522, "train_speed(iter/s)": 1.510616 }, { "epoch": 0.1792982305813804, "grad_norm": 2.676987409591675, "learning_rate": 9.968304881136645e-05, "loss": 0.3033310890197754, "memory(GiB)": 57.83, "step": 4185, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.510541 }, { "epoch": 0.17951244591062937, "grad_norm": 0.05654676631093025, "learning_rate": 9.968229181199083e-05, "loss": 0.36115853786468505, "memory(GiB)": 57.83, "step": 4190, "token_acc": 0.8838709677419355, "train_speed(iter/s)": 1.51045 }, { "epoch": 0.17972666123987832, "grad_norm": 2.575967788696289, "learning_rate": 9.968153391257388e-05, "loss": 0.8765428543090821, "memory(GiB)": 57.83, "step": 4195, "token_acc": 0.8383838383838383, "train_speed(iter/s)": 1.510466 }, { "epoch": 0.1799408765691273, "grad_norm": 1.4130916595458984, "learning_rate": 9.968077511312934e-05, "loss": 0.3437749147415161, "memory(GiB)": 57.83, "step": 4200, "token_acc": 0.9316770186335404, "train_speed(iter/s)": 1.510521 }, { "epoch": 0.18015509189837625, "grad_norm": 1.4364113807678223, "learning_rate": 9.968001541367096e-05, "loss": 0.25515530109405515, "memory(GiB)": 57.83, "step": 4205, "token_acc": 0.948948948948949, "train_speed(iter/s)": 1.51067 }, { "epoch": 0.1803693072276252, "grad_norm": 2.408681631088257, "learning_rate": 9.967925481421249e-05, "loss": 0.4146378517150879, "memory(GiB)": 57.83, "step": 4210, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.510749 }, { "epoch": 0.18058352255687418, "grad_norm": 1.3611465692520142, "learning_rate": 9.967849331476771e-05, "loss": 0.3729621171951294, "memory(GiB)": 57.83, "step": 4215, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.510768 }, { "epoch": 0.18079773788612313, "grad_norm": 2.954491376876831, "learning_rate": 9.967773091535045e-05, "loss": 0.3815258026123047, "memory(GiB)": 57.83, "step": 4220, "token_acc": 0.9105058365758755, "train_speed(iter/s)": 1.510768 }, { "epoch": 0.18101195321537208, "grad_norm": 1.939063549041748, "learning_rate": 9.967696761597446e-05, "loss": 0.4038283348083496, "memory(GiB)": 57.83, "step": 4225, "token_acc": 0.8952702702702703, "train_speed(iter/s)": 1.511113 }, { "epoch": 0.18122616854462106, "grad_norm": 1.4618935585021973, "learning_rate": 9.967620341665361e-05, "loss": 0.45621585845947266, "memory(GiB)": 57.83, "step": 4230, "token_acc": 0.8997289972899729, "train_speed(iter/s)": 1.511056 }, { "epoch": 0.18144038387387001, "grad_norm": 1.9109877347946167, "learning_rate": 9.967543831740174e-05, "loss": 0.37326204776763916, "memory(GiB)": 57.83, "step": 4235, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.511017 }, { "epoch": 0.18165459920311897, "grad_norm": 1.9764270782470703, "learning_rate": 9.967467231823271e-05, "loss": 0.35768589973449705, "memory(GiB)": 57.83, "step": 4240, "token_acc": 0.9149659863945578, "train_speed(iter/s)": 1.511046 }, { "epoch": 0.18186881453236795, "grad_norm": 0.5221970081329346, "learning_rate": 9.967390541916037e-05, "loss": 0.33529441356658934, "memory(GiB)": 57.83, "step": 4245, "token_acc": 0.9164086687306502, "train_speed(iter/s)": 1.511119 }, { "epoch": 0.1820830298616169, "grad_norm": 2.359494209289551, "learning_rate": 9.967313762019866e-05, "loss": 0.244231915473938, "memory(GiB)": 57.83, "step": 4250, "token_acc": 0.9407114624505929, "train_speed(iter/s)": 1.511386 }, { "epoch": 0.18229724519086585, "grad_norm": 4.730870723724365, "learning_rate": 9.967236892136145e-05, "loss": 0.34471845626831055, "memory(GiB)": 57.83, "step": 4255, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.511515 }, { "epoch": 0.18251146052011483, "grad_norm": 4.2270050048828125, "learning_rate": 9.967159932266271e-05, "loss": 0.6696194171905517, "memory(GiB)": 57.83, "step": 4260, "token_acc": 0.8442622950819673, "train_speed(iter/s)": 1.511571 }, { "epoch": 0.18272567584936378, "grad_norm": 2.962764024734497, "learning_rate": 9.967082882411631e-05, "loss": 0.3858340740203857, "memory(GiB)": 57.83, "step": 4265, "token_acc": 0.8984375, "train_speed(iter/s)": 1.511565 }, { "epoch": 0.18293989117861273, "grad_norm": 5.059712886810303, "learning_rate": 9.967005742573627e-05, "loss": 0.32372465133666994, "memory(GiB)": 57.83, "step": 4270, "token_acc": 0.9201183431952663, "train_speed(iter/s)": 1.511555 }, { "epoch": 0.1831541065078617, "grad_norm": 3.7124290466308594, "learning_rate": 9.966928512753656e-05, "loss": 0.2611952304840088, "memory(GiB)": 57.83, "step": 4275, "token_acc": 0.9498207885304659, "train_speed(iter/s)": 1.511728 }, { "epoch": 0.18336832183711066, "grad_norm": 3.342367649078369, "learning_rate": 9.966851192953114e-05, "loss": 0.28155856132507323, "memory(GiB)": 57.83, "step": 4280, "token_acc": 0.94, "train_speed(iter/s)": 1.511842 }, { "epoch": 0.18358253716635964, "grad_norm": 3.497875452041626, "learning_rate": 9.966773783173403e-05, "loss": 0.48347039222717286, "memory(GiB)": 57.83, "step": 4285, "token_acc": 0.8900709219858156, "train_speed(iter/s)": 1.511998 }, { "epoch": 0.1837967524956086, "grad_norm": 4.6944732666015625, "learning_rate": 9.966696283415926e-05, "loss": 0.400393009185791, "memory(GiB)": 57.83, "step": 4290, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.51215 }, { "epoch": 0.18401096782485754, "grad_norm": 1.9652750492095947, "learning_rate": 9.966618693682089e-05, "loss": 0.16524970531463623, "memory(GiB)": 57.83, "step": 4295, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.512194 }, { "epoch": 0.18422518315410652, "grad_norm": 3.7286672592163086, "learning_rate": 9.966541013973291e-05, "loss": 0.41719279289245603, "memory(GiB)": 57.83, "step": 4300, "token_acc": 0.9025974025974026, "train_speed(iter/s)": 1.512149 }, { "epoch": 0.18443939848335547, "grad_norm": 1.9621535539627075, "learning_rate": 9.966463244290947e-05, "loss": 0.5357668399810791, "memory(GiB)": 57.83, "step": 4305, "token_acc": 0.9026548672566371, "train_speed(iter/s)": 1.512323 }, { "epoch": 0.18465361381260442, "grad_norm": 4.496447563171387, "learning_rate": 9.966385384636461e-05, "loss": 0.5068263530731201, "memory(GiB)": 57.83, "step": 4310, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.512469 }, { "epoch": 0.1848678291418534, "grad_norm": 2.1366496086120605, "learning_rate": 9.966307435011245e-05, "loss": 0.4369688034057617, "memory(GiB)": 57.83, "step": 4315, "token_acc": 0.9133858267716536, "train_speed(iter/s)": 1.51237 }, { "epoch": 0.18508204447110235, "grad_norm": 2.3534796237945557, "learning_rate": 9.96622939541671e-05, "loss": 0.1094194769859314, "memory(GiB)": 57.83, "step": 4320, "token_acc": 0.9757085020242915, "train_speed(iter/s)": 1.512324 }, { "epoch": 0.1852962598003513, "grad_norm": 1.7149937152862549, "learning_rate": 9.966151265854273e-05, "loss": 0.42652311325073244, "memory(GiB)": 57.83, "step": 4325, "token_acc": 0.9079497907949791, "train_speed(iter/s)": 1.512297 }, { "epoch": 0.18551047512960028, "grad_norm": 1.6824856996536255, "learning_rate": 9.966073046325346e-05, "loss": 0.24296181201934813, "memory(GiB)": 57.83, "step": 4330, "token_acc": 0.9484126984126984, "train_speed(iter/s)": 1.512204 }, { "epoch": 0.18572469045884923, "grad_norm": 2.950157403945923, "learning_rate": 9.965994736831348e-05, "loss": 0.4841159820556641, "memory(GiB)": 57.83, "step": 4335, "token_acc": 0.9125, "train_speed(iter/s)": 1.512251 }, { "epoch": 0.18593890578809819, "grad_norm": 3.585350751876831, "learning_rate": 9.965916337373696e-05, "loss": 0.34171342849731445, "memory(GiB)": 57.83, "step": 4340, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.512273 }, { "epoch": 0.18615312111734716, "grad_norm": 1.9751213788986206, "learning_rate": 9.965837847953812e-05, "loss": 0.423966646194458, "memory(GiB)": 57.83, "step": 4345, "token_acc": 0.8934169278996865, "train_speed(iter/s)": 1.512222 }, { "epoch": 0.18636733644659612, "grad_norm": 3.2218172550201416, "learning_rate": 9.965759268573117e-05, "loss": 0.5220558166503906, "memory(GiB)": 57.83, "step": 4350, "token_acc": 0.8962264150943396, "train_speed(iter/s)": 1.512138 }, { "epoch": 0.18658155177584507, "grad_norm": 0.8246796131134033, "learning_rate": 9.965680599233034e-05, "loss": 0.34073150157928467, "memory(GiB)": 57.83, "step": 4355, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.512107 }, { "epoch": 0.18679576710509405, "grad_norm": 2.5992727279663086, "learning_rate": 9.965601839934988e-05, "loss": 0.24677376747131347, "memory(GiB)": 57.83, "step": 4360, "token_acc": 0.9331306990881459, "train_speed(iter/s)": 1.512095 }, { "epoch": 0.187009982434343, "grad_norm": 3.915160894393921, "learning_rate": 9.965522990680406e-05, "loss": 0.4797336578369141, "memory(GiB)": 57.83, "step": 4365, "token_acc": 0.8859060402684564, "train_speed(iter/s)": 1.512548 }, { "epoch": 0.18722419776359198, "grad_norm": 3.1434991359710693, "learning_rate": 9.965444051470721e-05, "loss": 0.38097543716430665, "memory(GiB)": 57.83, "step": 4370, "token_acc": 0.9194029850746268, "train_speed(iter/s)": 1.512669 }, { "epoch": 0.18743841309284093, "grad_norm": 2.0081093311309814, "learning_rate": 9.965365022307356e-05, "loss": 0.3304772853851318, "memory(GiB)": 57.83, "step": 4375, "token_acc": 0.943609022556391, "train_speed(iter/s)": 1.512568 }, { "epoch": 0.18765262842208988, "grad_norm": 1.8937528133392334, "learning_rate": 9.965285903191746e-05, "loss": 0.16415760517120362, "memory(GiB)": 57.83, "step": 4380, "token_acc": 0.9572368421052632, "train_speed(iter/s)": 1.512557 }, { "epoch": 0.18786684375133886, "grad_norm": 2.0360870361328125, "learning_rate": 9.965206694125324e-05, "loss": 0.40878992080688475, "memory(GiB)": 57.83, "step": 4385, "token_acc": 0.9346590909090909, "train_speed(iter/s)": 1.512573 }, { "epoch": 0.1880810590805878, "grad_norm": 3.3475382328033447, "learning_rate": 9.965127395109525e-05, "loss": 0.461670446395874, "memory(GiB)": 57.83, "step": 4390, "token_acc": 0.8873720136518771, "train_speed(iter/s)": 1.512474 }, { "epoch": 0.18829527440983676, "grad_norm": 1.9903428554534912, "learning_rate": 9.965048006145785e-05, "loss": 0.6634902954101562, "memory(GiB)": 57.83, "step": 4395, "token_acc": 0.8322147651006712, "train_speed(iter/s)": 1.512648 }, { "epoch": 0.18850948973908574, "grad_norm": 4.307553291320801, "learning_rate": 9.964968527235544e-05, "loss": 0.40163116455078124, "memory(GiB)": 57.83, "step": 4400, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.512711 }, { "epoch": 0.1887237050683347, "grad_norm": 2.3854851722717285, "learning_rate": 9.96488895838024e-05, "loss": 0.20213549137115477, "memory(GiB)": 57.83, "step": 4405, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.512822 }, { "epoch": 0.18893792039758364, "grad_norm": 3.276015520095825, "learning_rate": 9.964809299581315e-05, "loss": 0.4857827663421631, "memory(GiB)": 57.83, "step": 4410, "token_acc": 0.9053497942386831, "train_speed(iter/s)": 1.512917 }, { "epoch": 0.18915213572683262, "grad_norm": 3.4937686920166016, "learning_rate": 9.964729550840212e-05, "loss": 0.32454349994659426, "memory(GiB)": 57.83, "step": 4415, "token_acc": 0.9394812680115274, "train_speed(iter/s)": 1.51286 }, { "epoch": 0.18936635105608157, "grad_norm": 4.493664264678955, "learning_rate": 9.964649712158377e-05, "loss": 0.545006799697876, "memory(GiB)": 57.83, "step": 4420, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.512908 }, { "epoch": 0.18958056638533052, "grad_norm": 5.973928928375244, "learning_rate": 9.964569783537255e-05, "loss": 0.31019368171691897, "memory(GiB)": 57.83, "step": 4425, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.513157 }, { "epoch": 0.1897947817145795, "grad_norm": 2.399397373199463, "learning_rate": 9.964489764978293e-05, "loss": 0.4737192153930664, "memory(GiB)": 57.83, "step": 4430, "token_acc": 0.8904494382022472, "train_speed(iter/s)": 1.513106 }, { "epoch": 0.19000899704382845, "grad_norm": 3.1527812480926514, "learning_rate": 9.964409656482943e-05, "loss": 0.2630627632141113, "memory(GiB)": 57.83, "step": 4435, "token_acc": 0.9301587301587302, "train_speed(iter/s)": 1.513105 }, { "epoch": 0.1902232123730774, "grad_norm": 5.364129543304443, "learning_rate": 9.964329458052655e-05, "loss": 0.572676706314087, "memory(GiB)": 57.83, "step": 4440, "token_acc": 0.8729096989966555, "train_speed(iter/s)": 1.513282 }, { "epoch": 0.19043742770232638, "grad_norm": 3.5646073818206787, "learning_rate": 9.964249169688882e-05, "loss": 0.5340888977050782, "memory(GiB)": 57.83, "step": 4445, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.513351 }, { "epoch": 0.19065164303157534, "grad_norm": 0.7731857895851135, "learning_rate": 9.964168791393077e-05, "loss": 0.3916285037994385, "memory(GiB)": 57.83, "step": 4450, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.513714 }, { "epoch": 0.19086585836082431, "grad_norm": 1.2108426094055176, "learning_rate": 9.964088323166698e-05, "loss": 0.5063121795654297, "memory(GiB)": 57.83, "step": 4455, "token_acc": 0.9002695417789758, "train_speed(iter/s)": 1.513657 }, { "epoch": 0.19108007369007327, "grad_norm": 2.339223861694336, "learning_rate": 9.964007765011204e-05, "loss": 0.24514691829681395, "memory(GiB)": 57.83, "step": 4460, "token_acc": 0.9442379182156134, "train_speed(iter/s)": 1.513899 }, { "epoch": 0.19129428901932222, "grad_norm": 1.554818868637085, "learning_rate": 9.963927116928051e-05, "loss": 0.36054420471191406, "memory(GiB)": 57.83, "step": 4465, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.513883 }, { "epoch": 0.1915085043485712, "grad_norm": 0.6397516131401062, "learning_rate": 9.963846378918702e-05, "loss": 0.23652188777923583, "memory(GiB)": 57.83, "step": 4470, "token_acc": 0.9521739130434783, "train_speed(iter/s)": 1.513856 }, { "epoch": 0.19172271967782015, "grad_norm": 2.72127628326416, "learning_rate": 9.96376555098462e-05, "loss": 0.2857382774353027, "memory(GiB)": 57.83, "step": 4475, "token_acc": 0.9522184300341296, "train_speed(iter/s)": 1.513792 }, { "epoch": 0.1919369350070691, "grad_norm": 4.367252349853516, "learning_rate": 9.963684633127269e-05, "loss": 0.38074870109558107, "memory(GiB)": 57.83, "step": 4480, "token_acc": 0.9136904761904762, "train_speed(iter/s)": 1.513707 }, { "epoch": 0.19215115033631808, "grad_norm": 1.1047741174697876, "learning_rate": 9.963603625348114e-05, "loss": 0.38932323455810547, "memory(GiB)": 57.83, "step": 4485, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.513659 }, { "epoch": 0.19236536566556703, "grad_norm": 2.3582239151000977, "learning_rate": 9.963522527648623e-05, "loss": 0.35733709335327146, "memory(GiB)": 57.83, "step": 4490, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.513815 }, { "epoch": 0.19257958099481598, "grad_norm": 3.3467960357666016, "learning_rate": 9.963441340030267e-05, "loss": 0.5948925971984863, "memory(GiB)": 57.83, "step": 4495, "token_acc": 0.8597560975609756, "train_speed(iter/s)": 1.514086 }, { "epoch": 0.19279379632406496, "grad_norm": 3.4974379539489746, "learning_rate": 9.963360062494512e-05, "loss": 0.5752200603485107, "memory(GiB)": 57.83, "step": 4500, "token_acc": 0.8830188679245283, "train_speed(iter/s)": 1.514203 }, { "epoch": 0.19279379632406496, "eval_loss": 2.4901604652404785, "eval_runtime": 11.5758, "eval_samples_per_second": 8.639, "eval_steps_per_second": 8.639, "eval_token_acc": 0.430327868852459, "step": 4500 }, { "epoch": 0.1930080116533139, "grad_norm": 1.142996072769165, "learning_rate": 9.963278695042835e-05, "loss": 0.2755196809768677, "memory(GiB)": 57.83, "step": 4505, "token_acc": 0.5657764589515332, "train_speed(iter/s)": 1.508263 }, { "epoch": 0.19322222698256286, "grad_norm": 3.735818386077881, "learning_rate": 9.963197237676709e-05, "loss": 0.33631291389465334, "memory(GiB)": 57.83, "step": 4510, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.508322 }, { "epoch": 0.19343644231181184, "grad_norm": 0.45004796981811523, "learning_rate": 9.963115690397608e-05, "loss": 0.4286465644836426, "memory(GiB)": 57.83, "step": 4515, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.508647 }, { "epoch": 0.1936506576410608, "grad_norm": 2.5010716915130615, "learning_rate": 9.96303405320701e-05, "loss": 0.38583898544311523, "memory(GiB)": 57.83, "step": 4520, "token_acc": 0.903125, "train_speed(iter/s)": 1.508699 }, { "epoch": 0.19386487297030974, "grad_norm": 2.5468995571136475, "learning_rate": 9.962952326106396e-05, "loss": 0.2166111946105957, "memory(GiB)": 57.83, "step": 4525, "token_acc": 0.9577039274924471, "train_speed(iter/s)": 1.50876 }, { "epoch": 0.19407908829955872, "grad_norm": 1.5576945543289185, "learning_rate": 9.962870509097245e-05, "loss": 0.23021812438964845, "memory(GiB)": 57.83, "step": 4530, "token_acc": 0.9467680608365019, "train_speed(iter/s)": 1.50875 }, { "epoch": 0.19429330362880767, "grad_norm": 5.2786335945129395, "learning_rate": 9.962788602181039e-05, "loss": 0.4434218883514404, "memory(GiB)": 57.83, "step": 4535, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.50877 }, { "epoch": 0.19450751895805665, "grad_norm": 4.071503162384033, "learning_rate": 9.962706605359262e-05, "loss": 0.4134981155395508, "memory(GiB)": 57.83, "step": 4540, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.509098 }, { "epoch": 0.1947217342873056, "grad_norm": 1.4005829095840454, "learning_rate": 9.962624518633399e-05, "loss": 0.4932573318481445, "memory(GiB)": 57.83, "step": 4545, "token_acc": 0.9071428571428571, "train_speed(iter/s)": 1.50929 }, { "epoch": 0.19493594961655455, "grad_norm": 0.8051711916923523, "learning_rate": 9.962542342004937e-05, "loss": 0.25854485034942626, "memory(GiB)": 57.83, "step": 4550, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.509286 }, { "epoch": 0.19515016494580353, "grad_norm": 2.243042230606079, "learning_rate": 9.962460075475366e-05, "loss": 0.33190934658050536, "memory(GiB)": 57.83, "step": 4555, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.509285 }, { "epoch": 0.19536438027505248, "grad_norm": 2.6498992443084717, "learning_rate": 9.962377719046176e-05, "loss": 0.3754370450973511, "memory(GiB)": 57.83, "step": 4560, "token_acc": 0.9013605442176871, "train_speed(iter/s)": 1.509405 }, { "epoch": 0.19557859560430144, "grad_norm": 3.259633779525757, "learning_rate": 9.962295272718858e-05, "loss": 0.4798471927642822, "memory(GiB)": 57.83, "step": 4565, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.509416 }, { "epoch": 0.19579281093355042, "grad_norm": 5.368512153625488, "learning_rate": 9.962212736494905e-05, "loss": 0.5335381031036377, "memory(GiB)": 57.83, "step": 4570, "token_acc": 0.8977272727272727, "train_speed(iter/s)": 1.509427 }, { "epoch": 0.19600702626279937, "grad_norm": 5.794067859649658, "learning_rate": 9.962130110375815e-05, "loss": 0.42315058708190917, "memory(GiB)": 57.83, "step": 4575, "token_acc": 0.9161290322580645, "train_speed(iter/s)": 1.509623 }, { "epoch": 0.19622124159204832, "grad_norm": 1.5571739673614502, "learning_rate": 9.962047394363083e-05, "loss": 0.5238777160644531, "memory(GiB)": 57.83, "step": 4580, "token_acc": 0.8931750741839762, "train_speed(iter/s)": 1.5097 }, { "epoch": 0.1964354569212973, "grad_norm": 2.372591257095337, "learning_rate": 9.961964588458207e-05, "loss": 0.8912220954895019, "memory(GiB)": 57.83, "step": 4585, "token_acc": 0.8402777777777778, "train_speed(iter/s)": 1.509843 }, { "epoch": 0.19664967225054625, "grad_norm": 4.191328048706055, "learning_rate": 9.961881692662688e-05, "loss": 0.5701568603515625, "memory(GiB)": 57.83, "step": 4590, "token_acc": 0.9152542372881356, "train_speed(iter/s)": 1.509976 }, { "epoch": 0.1968638875797952, "grad_norm": 1.2514963150024414, "learning_rate": 9.961798706978027e-05, "loss": 0.36083695888519285, "memory(GiB)": 57.83, "step": 4595, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.510002 }, { "epoch": 0.19707810290904418, "grad_norm": 4.413768291473389, "learning_rate": 9.96171563140573e-05, "loss": 0.24695179462432862, "memory(GiB)": 57.83, "step": 4600, "token_acc": 0.946031746031746, "train_speed(iter/s)": 1.509966 }, { "epoch": 0.19729231823829313, "grad_norm": 2.3286967277526855, "learning_rate": 9.961632465947297e-05, "loss": 0.3880555868148804, "memory(GiB)": 57.83, "step": 4605, "token_acc": 0.9337539432176656, "train_speed(iter/s)": 1.510053 }, { "epoch": 0.19750653356754208, "grad_norm": 2.0209157466888428, "learning_rate": 9.96154921060424e-05, "loss": 0.1740952253341675, "memory(GiB)": 57.83, "step": 4610, "token_acc": 0.9608938547486033, "train_speed(iter/s)": 1.510102 }, { "epoch": 0.19772074889679106, "grad_norm": 1.3671507835388184, "learning_rate": 9.961465865378063e-05, "loss": 0.43624320030212405, "memory(GiB)": 57.83, "step": 4615, "token_acc": 0.900398406374502, "train_speed(iter/s)": 1.510211 }, { "epoch": 0.19793496422604, "grad_norm": 4.369647026062012, "learning_rate": 9.961382430270278e-05, "loss": 0.4634563446044922, "memory(GiB)": 57.83, "step": 4620, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.510368 }, { "epoch": 0.198149179555289, "grad_norm": 0.9397644400596619, "learning_rate": 9.961298905282397e-05, "loss": 0.1763014554977417, "memory(GiB)": 57.83, "step": 4625, "token_acc": 0.9580838323353293, "train_speed(iter/s)": 1.510456 }, { "epoch": 0.19836339488453794, "grad_norm": 2.5887327194213867, "learning_rate": 9.961215290415931e-05, "loss": 0.31793875694274903, "memory(GiB)": 57.83, "step": 4630, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.510442 }, { "epoch": 0.1985776102137869, "grad_norm": 1.5717778205871582, "learning_rate": 9.961131585672396e-05, "loss": 0.29057703018188474, "memory(GiB)": 57.83, "step": 4635, "token_acc": 0.9432624113475178, "train_speed(iter/s)": 1.510487 }, { "epoch": 0.19879182554303587, "grad_norm": 2.337211847305298, "learning_rate": 9.961047791053309e-05, "loss": 0.3953556060791016, "memory(GiB)": 57.83, "step": 4640, "token_acc": 0.9105691056910569, "train_speed(iter/s)": 1.510506 }, { "epoch": 0.19900604087228482, "grad_norm": 3.594511032104492, "learning_rate": 9.960963906560188e-05, "loss": 0.331571364402771, "memory(GiB)": 57.83, "step": 4645, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.510401 }, { "epoch": 0.19922025620153377, "grad_norm": 2.704123020172119, "learning_rate": 9.96087993219455e-05, "loss": 0.29836723804473875, "memory(GiB)": 57.83, "step": 4650, "token_acc": 0.9398496240601504, "train_speed(iter/s)": 1.510532 }, { "epoch": 0.19943447153078275, "grad_norm": 5.459751605987549, "learning_rate": 9.960795867957921e-05, "loss": 0.24799885749816894, "memory(GiB)": 57.83, "step": 4655, "token_acc": 0.944206008583691, "train_speed(iter/s)": 1.510529 }, { "epoch": 0.1996486868600317, "grad_norm": 1.7926923036575317, "learning_rate": 9.96071171385182e-05, "loss": 0.3452512502670288, "memory(GiB)": 57.83, "step": 4660, "token_acc": 0.9313725490196079, "train_speed(iter/s)": 1.51073 }, { "epoch": 0.19986290218928066, "grad_norm": 2.4634172916412354, "learning_rate": 9.960627469877773e-05, "loss": 0.3703289985656738, "memory(GiB)": 57.83, "step": 4665, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.5108 }, { "epoch": 0.20007711751852963, "grad_norm": 1.9648911952972412, "learning_rate": 9.960543136037306e-05, "loss": 0.237589693069458, "memory(GiB)": 57.83, "step": 4670, "token_acc": 0.9400749063670412, "train_speed(iter/s)": 1.510747 }, { "epoch": 0.20029133284777859, "grad_norm": 6.241596221923828, "learning_rate": 9.960458712331946e-05, "loss": 0.3864278316497803, "memory(GiB)": 57.83, "step": 4675, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.510836 }, { "epoch": 0.20050554817702754, "grad_norm": 7.256152153015137, "learning_rate": 9.960374198763225e-05, "loss": 0.33722898960113523, "memory(GiB)": 57.83, "step": 4680, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.510875 }, { "epoch": 0.20071976350627652, "grad_norm": 2.9519288539886475, "learning_rate": 9.96028959533267e-05, "loss": 0.5873964309692383, "memory(GiB)": 57.83, "step": 4685, "token_acc": 0.8827361563517915, "train_speed(iter/s)": 1.511071 }, { "epoch": 0.20093397883552547, "grad_norm": 1.1075732707977295, "learning_rate": 9.960204902041816e-05, "loss": 0.42109146118164065, "memory(GiB)": 57.83, "step": 4690, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.511159 }, { "epoch": 0.20114819416477442, "grad_norm": 4.078517913818359, "learning_rate": 9.960120118892199e-05, "loss": 0.4327414035797119, "memory(GiB)": 57.83, "step": 4695, "token_acc": 0.9, "train_speed(iter/s)": 1.511233 }, { "epoch": 0.2013624094940234, "grad_norm": 2.834219455718994, "learning_rate": 9.960035245885352e-05, "loss": 0.48975229263305664, "memory(GiB)": 57.83, "step": 4700, "token_acc": 0.8828571428571429, "train_speed(iter/s)": 1.511251 }, { "epoch": 0.20157662482327235, "grad_norm": 2.5123965740203857, "learning_rate": 9.959950283022813e-05, "loss": 0.41951642036437986, "memory(GiB)": 57.83, "step": 4705, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.511223 }, { "epoch": 0.20179084015252133, "grad_norm": 2.2336857318878174, "learning_rate": 9.959865230306122e-05, "loss": 0.4983494281768799, "memory(GiB)": 57.83, "step": 4710, "token_acc": 0.9034749034749034, "train_speed(iter/s)": 1.511242 }, { "epoch": 0.20200505548177028, "grad_norm": 2.172847032546997, "learning_rate": 9.959780087736821e-05, "loss": 0.8044857978820801, "memory(GiB)": 57.83, "step": 4715, "token_acc": 0.8202614379084967, "train_speed(iter/s)": 1.511299 }, { "epoch": 0.20221927081101923, "grad_norm": 5.335090637207031, "learning_rate": 9.959694855316451e-05, "loss": 0.7707884788513184, "memory(GiB)": 57.83, "step": 4720, "token_acc": 0.8698630136986302, "train_speed(iter/s)": 1.511195 }, { "epoch": 0.2024334861402682, "grad_norm": 1.963152289390564, "learning_rate": 9.959609533046554e-05, "loss": 0.4504702568054199, "memory(GiB)": 57.83, "step": 4725, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.511203 }, { "epoch": 0.20264770146951716, "grad_norm": 2.7646634578704834, "learning_rate": 9.959524120928678e-05, "loss": 0.43719005584716797, "memory(GiB)": 57.83, "step": 4730, "token_acc": 0.9174917491749175, "train_speed(iter/s)": 1.511441 }, { "epoch": 0.2028619167987661, "grad_norm": 0.7673521637916565, "learning_rate": 9.959438618964372e-05, "loss": 0.32702615261077883, "memory(GiB)": 57.83, "step": 4735, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.511454 }, { "epoch": 0.2030761321280151, "grad_norm": 3.1478071212768555, "learning_rate": 9.959353027155183e-05, "loss": 0.31312952041625974, "memory(GiB)": 57.83, "step": 4740, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.511531 }, { "epoch": 0.20329034745726404, "grad_norm": 3.342362403869629, "learning_rate": 9.959267345502661e-05, "loss": 0.4484687805175781, "memory(GiB)": 57.83, "step": 4745, "token_acc": 0.8779661016949153, "train_speed(iter/s)": 1.511565 }, { "epoch": 0.203504562786513, "grad_norm": 1.811107873916626, "learning_rate": 9.959181574008358e-05, "loss": 0.5537642478942871, "memory(GiB)": 57.83, "step": 4750, "token_acc": 0.8811188811188811, "train_speed(iter/s)": 1.511681 }, { "epoch": 0.20371877811576197, "grad_norm": 2.0825393199920654, "learning_rate": 9.95909571267383e-05, "loss": 0.3292291879653931, "memory(GiB)": 57.83, "step": 4755, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.511602 }, { "epoch": 0.20393299344501092, "grad_norm": 2.1439127922058105, "learning_rate": 9.95900976150063e-05, "loss": 0.286649489402771, "memory(GiB)": 57.83, "step": 4760, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.511617 }, { "epoch": 0.20414720877425988, "grad_norm": 3.171640396118164, "learning_rate": 9.958923720490317e-05, "loss": 0.4476619243621826, "memory(GiB)": 57.83, "step": 4765, "token_acc": 0.9112903225806451, "train_speed(iter/s)": 1.511584 }, { "epoch": 0.20436142410350885, "grad_norm": 2.0208709239959717, "learning_rate": 9.958837589644449e-05, "loss": 0.3627376317977905, "memory(GiB)": 57.83, "step": 4770, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.511479 }, { "epoch": 0.2045756394327578, "grad_norm": 11.23732852935791, "learning_rate": 9.958751368964585e-05, "loss": 0.5284453392028808, "memory(GiB)": 57.83, "step": 4775, "token_acc": 0.8814229249011858, "train_speed(iter/s)": 1.511446 }, { "epoch": 0.20478985476200676, "grad_norm": 4.156768798828125, "learning_rate": 9.95866505845229e-05, "loss": 0.43798327445983887, "memory(GiB)": 57.83, "step": 4780, "token_acc": 0.9296187683284457, "train_speed(iter/s)": 1.511389 }, { "epoch": 0.20500407009125574, "grad_norm": 7.37565803527832, "learning_rate": 9.958578658109125e-05, "loss": 0.4032639980316162, "memory(GiB)": 57.83, "step": 4785, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.511643 }, { "epoch": 0.2052182854205047, "grad_norm": 1.2982243299484253, "learning_rate": 9.958492167936658e-05, "loss": 0.49030346870422364, "memory(GiB)": 57.83, "step": 4790, "token_acc": 0.9, "train_speed(iter/s)": 1.51209 }, { "epoch": 0.20543250074975367, "grad_norm": 1.5245417356491089, "learning_rate": 9.958405587936452e-05, "loss": 0.43039278984069823, "memory(GiB)": 57.83, "step": 4795, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.512257 }, { "epoch": 0.20564671607900262, "grad_norm": 1.8226889371871948, "learning_rate": 9.958318918110077e-05, "loss": 0.3904107570648193, "memory(GiB)": 57.83, "step": 4800, "token_acc": 0.921875, "train_speed(iter/s)": 1.512323 }, { "epoch": 0.20586093140825157, "grad_norm": 1.377678394317627, "learning_rate": 9.958232158459104e-05, "loss": 0.2371608018875122, "memory(GiB)": 57.83, "step": 4805, "token_acc": 0.9590163934426229, "train_speed(iter/s)": 1.512485 }, { "epoch": 0.20607514673750055, "grad_norm": 1.0462480783462524, "learning_rate": 9.958145308985105e-05, "loss": 0.24427549839019774, "memory(GiB)": 57.83, "step": 4810, "token_acc": 0.9614243323442137, "train_speed(iter/s)": 1.512482 }, { "epoch": 0.2062893620667495, "grad_norm": 2.586125135421753, "learning_rate": 9.958058369689651e-05, "loss": 0.39127578735351565, "memory(GiB)": 57.83, "step": 4815, "token_acc": 0.9105691056910569, "train_speed(iter/s)": 1.512711 }, { "epoch": 0.20650357739599845, "grad_norm": 1.9469267129898071, "learning_rate": 9.95797134057432e-05, "loss": 0.2736664772033691, "memory(GiB)": 57.83, "step": 4820, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.512761 }, { "epoch": 0.20671779272524743, "grad_norm": 2.6577377319335938, "learning_rate": 9.957884221640687e-05, "loss": 0.470093822479248, "memory(GiB)": 57.83, "step": 4825, "token_acc": 0.9042553191489362, "train_speed(iter/s)": 1.51272 }, { "epoch": 0.20693200805449638, "grad_norm": 2.445228338241577, "learning_rate": 9.95779701289033e-05, "loss": 0.35083160400390623, "memory(GiB)": 57.83, "step": 4830, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.512789 }, { "epoch": 0.20714622338374533, "grad_norm": 1.2371408939361572, "learning_rate": 9.95770971432483e-05, "loss": 0.27249836921691895, "memory(GiB)": 57.83, "step": 4835, "token_acc": 0.9566666666666667, "train_speed(iter/s)": 1.512879 }, { "epoch": 0.2073604387129943, "grad_norm": 1.713184118270874, "learning_rate": 9.957622325945768e-05, "loss": 0.45069570541381837, "memory(GiB)": 57.83, "step": 4840, "token_acc": 0.8892405063291139, "train_speed(iter/s)": 1.512984 }, { "epoch": 0.20757465404224326, "grad_norm": 5.630221366882324, "learning_rate": 9.957534847754726e-05, "loss": 0.4888094425201416, "memory(GiB)": 57.83, "step": 4845, "token_acc": 0.8812154696132597, "train_speed(iter/s)": 1.512977 }, { "epoch": 0.2077888693714922, "grad_norm": 5.215254306793213, "learning_rate": 9.95744727975329e-05, "loss": 0.2060638427734375, "memory(GiB)": 57.83, "step": 4850, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.512908 }, { "epoch": 0.2080030847007412, "grad_norm": 1.721095085144043, "learning_rate": 9.957359621943045e-05, "loss": 0.43081250190734866, "memory(GiB)": 57.83, "step": 4855, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.513036 }, { "epoch": 0.20821730002999014, "grad_norm": 2.5146608352661133, "learning_rate": 9.957271874325582e-05, "loss": 0.3351567506790161, "memory(GiB)": 57.83, "step": 4860, "token_acc": 0.9346938775510204, "train_speed(iter/s)": 1.513021 }, { "epoch": 0.2084315153592391, "grad_norm": 0.07776860147714615, "learning_rate": 9.957184036902488e-05, "loss": 0.16938194036483764, "memory(GiB)": 57.83, "step": 4865, "token_acc": 0.966789667896679, "train_speed(iter/s)": 1.513206 }, { "epoch": 0.20864573068848807, "grad_norm": 2.8645334243774414, "learning_rate": 9.957096109675354e-05, "loss": 0.49625649452209475, "memory(GiB)": 57.83, "step": 4870, "token_acc": 0.8961424332344213, "train_speed(iter/s)": 1.513254 }, { "epoch": 0.20885994601773702, "grad_norm": 5.428793907165527, "learning_rate": 9.957008092645777e-05, "loss": 0.5637722969055176, "memory(GiB)": 57.83, "step": 4875, "token_acc": 0.8603174603174604, "train_speed(iter/s)": 1.51325 }, { "epoch": 0.209074161346986, "grad_norm": 1.8897464275360107, "learning_rate": 9.956919985815345e-05, "loss": 0.4287055015563965, "memory(GiB)": 57.83, "step": 4880, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.513222 }, { "epoch": 0.20928837667623496, "grad_norm": 3.3509104251861572, "learning_rate": 9.956831789185659e-05, "loss": 0.41711721420288084, "memory(GiB)": 57.83, "step": 4885, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.513123 }, { "epoch": 0.2095025920054839, "grad_norm": 10.413986206054688, "learning_rate": 9.956743502758315e-05, "loss": 0.6830052375793457, "memory(GiB)": 57.83, "step": 4890, "token_acc": 0.8621908127208481, "train_speed(iter/s)": 1.513262 }, { "epoch": 0.20971680733473289, "grad_norm": 5.843488693237305, "learning_rate": 9.956655126534911e-05, "loss": 0.30419929027557374, "memory(GiB)": 57.83, "step": 4895, "token_acc": 0.9386792452830188, "train_speed(iter/s)": 1.513247 }, { "epoch": 0.20993102266398184, "grad_norm": 3.5841634273529053, "learning_rate": 9.956566660517053e-05, "loss": 0.2566602945327759, "memory(GiB)": 57.83, "step": 4900, "token_acc": 0.9375, "train_speed(iter/s)": 1.513151 }, { "epoch": 0.2101452379932308, "grad_norm": 2.064887285232544, "learning_rate": 9.956478104706337e-05, "loss": 0.25555408000946045, "memory(GiB)": 57.83, "step": 4905, "token_acc": 0.9501915708812261, "train_speed(iter/s)": 1.513169 }, { "epoch": 0.21035945332247977, "grad_norm": 2.448301315307617, "learning_rate": 9.956389459104372e-05, "loss": 0.5230181694030762, "memory(GiB)": 57.83, "step": 4910, "token_acc": 0.8971061093247589, "train_speed(iter/s)": 1.51335 }, { "epoch": 0.21057366865172872, "grad_norm": 2.210487127304077, "learning_rate": 9.956300723712764e-05, "loss": 0.21928803920745848, "memory(GiB)": 57.83, "step": 4915, "token_acc": 0.9515151515151515, "train_speed(iter/s)": 1.513353 }, { "epoch": 0.21078788398097767, "grad_norm": 2.2405714988708496, "learning_rate": 9.956211898533117e-05, "loss": 0.2661418914794922, "memory(GiB)": 57.83, "step": 4920, "token_acc": 0.9489795918367347, "train_speed(iter/s)": 1.513458 }, { "epoch": 0.21100209931022665, "grad_norm": 0.6359110474586487, "learning_rate": 9.956122983567042e-05, "loss": 0.3111840009689331, "memory(GiB)": 57.83, "step": 4925, "token_acc": 0.9446064139941691, "train_speed(iter/s)": 1.513396 }, { "epoch": 0.2112163146394756, "grad_norm": 3.654067039489746, "learning_rate": 9.95603397881615e-05, "loss": 0.31097049713134767, "memory(GiB)": 57.83, "step": 4930, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.513354 }, { "epoch": 0.21143052996872455, "grad_norm": 1.3887428045272827, "learning_rate": 9.955944884282055e-05, "loss": 0.37289867401123045, "memory(GiB)": 57.83, "step": 4935, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.513319 }, { "epoch": 0.21164474529797353, "grad_norm": 2.798880100250244, "learning_rate": 9.955855699966368e-05, "loss": 0.20896704196929933, "memory(GiB)": 57.83, "step": 4940, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.513308 }, { "epoch": 0.21185896062722248, "grad_norm": 4.069769859313965, "learning_rate": 9.955766425870706e-05, "loss": 0.22878079414367675, "memory(GiB)": 57.83, "step": 4945, "token_acc": 0.9543726235741445, "train_speed(iter/s)": 1.513286 }, { "epoch": 0.21207317595647143, "grad_norm": 2.969782590866089, "learning_rate": 9.955677061996687e-05, "loss": 0.597745132446289, "memory(GiB)": 57.83, "step": 4950, "token_acc": 0.8760330578512396, "train_speed(iter/s)": 1.513297 }, { "epoch": 0.2122873912857204, "grad_norm": 4.079213619232178, "learning_rate": 9.955587608345928e-05, "loss": 0.15140869617462158, "memory(GiB)": 57.83, "step": 4955, "token_acc": 0.9736070381231672, "train_speed(iter/s)": 1.513389 }, { "epoch": 0.21250160661496936, "grad_norm": 0.27629417181015015, "learning_rate": 9.955498064920052e-05, "loss": 0.3577702283859253, "memory(GiB)": 57.83, "step": 4960, "token_acc": 0.9391304347826087, "train_speed(iter/s)": 1.513457 }, { "epoch": 0.21271582194421834, "grad_norm": 1.8692045211791992, "learning_rate": 9.955408431720681e-05, "loss": 0.16478021144866944, "memory(GiB)": 57.83, "step": 4965, "token_acc": 0.9614035087719298, "train_speed(iter/s)": 1.513468 }, { "epoch": 0.2129300372734673, "grad_norm": 4.027073860168457, "learning_rate": 9.955318708749435e-05, "loss": 0.39575934410095215, "memory(GiB)": 57.83, "step": 4970, "token_acc": 0.9256198347107438, "train_speed(iter/s)": 1.51353 }, { "epoch": 0.21314425260271624, "grad_norm": 1.9805190563201904, "learning_rate": 9.955228896007944e-05, "loss": 0.27673914432525637, "memory(GiB)": 57.83, "step": 4975, "token_acc": 0.9330543933054394, "train_speed(iter/s)": 1.513666 }, { "epoch": 0.21335846793196522, "grad_norm": 2.718724489212036, "learning_rate": 9.955138993497832e-05, "loss": 0.5144206523895264, "memory(GiB)": 57.83, "step": 4980, "token_acc": 0.8870431893687708, "train_speed(iter/s)": 1.513723 }, { "epoch": 0.21357268326121417, "grad_norm": 2.1398208141326904, "learning_rate": 9.95504900122073e-05, "loss": 0.5672962665557861, "memory(GiB)": 57.83, "step": 4985, "token_acc": 0.8917910447761194, "train_speed(iter/s)": 1.513904 }, { "epoch": 0.21378689859046313, "grad_norm": 4.935085296630859, "learning_rate": 9.954958919178265e-05, "loss": 0.2340653657913208, "memory(GiB)": 57.83, "step": 4990, "token_acc": 0.947565543071161, "train_speed(iter/s)": 1.51379 }, { "epoch": 0.2140011139197121, "grad_norm": 2.009493589401245, "learning_rate": 9.954868747372074e-05, "loss": 0.33717391490936277, "memory(GiB)": 57.83, "step": 4995, "token_acc": 0.9307692307692308, "train_speed(iter/s)": 1.513802 }, { "epoch": 0.21421532924896106, "grad_norm": 6.309597492218018, "learning_rate": 9.954778485803787e-05, "loss": 0.8703402519226074, "memory(GiB)": 57.83, "step": 5000, "token_acc": 0.8315018315018315, "train_speed(iter/s)": 1.513713 }, { "epoch": 0.21421532924896106, "eval_loss": 2.675294876098633, "eval_runtime": 10.0949, "eval_samples_per_second": 9.906, "eval_steps_per_second": 9.906, "eval_token_acc": 0.43375174337517436, "step": 5000 }, { "epoch": 0.21442954457821, "grad_norm": 5.3748250007629395, "learning_rate": 9.954688134475038e-05, "loss": 0.6445850372314453, "memory(GiB)": 57.83, "step": 5005, "token_acc": 0.5448136958710977, "train_speed(iter/s)": 1.508916 }, { "epoch": 0.214643759907459, "grad_norm": 2.3254969120025635, "learning_rate": 9.954597693387467e-05, "loss": 0.3609497547149658, "memory(GiB)": 57.83, "step": 5010, "token_acc": 0.9370860927152318, "train_speed(iter/s)": 1.508905 }, { "epoch": 0.21485797523670794, "grad_norm": 2.387509822845459, "learning_rate": 9.95450716254271e-05, "loss": 0.32059264183044434, "memory(GiB)": 57.83, "step": 5015, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.509035 }, { "epoch": 0.2150721905659569, "grad_norm": 3.7847301959991455, "learning_rate": 9.954416541942408e-05, "loss": 0.6047274589538574, "memory(GiB)": 57.83, "step": 5020, "token_acc": 0.8809523809523809, "train_speed(iter/s)": 1.509218 }, { "epoch": 0.21528640589520587, "grad_norm": 0.8641358613967896, "learning_rate": 9.954325831588204e-05, "loss": 0.338655686378479, "memory(GiB)": 57.83, "step": 5025, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.509236 }, { "epoch": 0.21550062122445482, "grad_norm": 4.458874225616455, "learning_rate": 9.954235031481739e-05, "loss": 0.4009979724884033, "memory(GiB)": 57.83, "step": 5030, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.509388 }, { "epoch": 0.21571483655370377, "grad_norm": 3.2322330474853516, "learning_rate": 9.95414414162466e-05, "loss": 0.3125793218612671, "memory(GiB)": 57.83, "step": 5035, "token_acc": 0.9368131868131868, "train_speed(iter/s)": 1.509367 }, { "epoch": 0.21592905188295275, "grad_norm": 7.080336093902588, "learning_rate": 9.95405316201861e-05, "loss": 0.16540060043334961, "memory(GiB)": 57.83, "step": 5040, "token_acc": 0.956140350877193, "train_speed(iter/s)": 1.509439 }, { "epoch": 0.2161432672122017, "grad_norm": 0.4881705641746521, "learning_rate": 9.953962092665243e-05, "loss": 0.7544567108154296, "memory(GiB)": 57.83, "step": 5045, "token_acc": 0.8542372881355932, "train_speed(iter/s)": 1.509627 }, { "epoch": 0.21635748254145068, "grad_norm": 4.409151077270508, "learning_rate": 9.953870933566203e-05, "loss": 0.3059594392776489, "memory(GiB)": 57.83, "step": 5050, "token_acc": 0.9375, "train_speed(iter/s)": 1.5097 }, { "epoch": 0.21657169787069963, "grad_norm": 2.3916854858398438, "learning_rate": 9.953779684723146e-05, "loss": 0.3484335899353027, "memory(GiB)": 57.83, "step": 5055, "token_acc": 0.9176029962546817, "train_speed(iter/s)": 1.509796 }, { "epoch": 0.21678591319994858, "grad_norm": 4.1899309158325195, "learning_rate": 9.953688346137722e-05, "loss": 0.1644739031791687, "memory(GiB)": 57.83, "step": 5060, "token_acc": 0.9551724137931035, "train_speed(iter/s)": 1.509893 }, { "epoch": 0.21700012852919756, "grad_norm": 2.385327100753784, "learning_rate": 9.953596917811586e-05, "loss": 0.3313437461853027, "memory(GiB)": 57.83, "step": 5065, "token_acc": 0.9184952978056427, "train_speed(iter/s)": 1.509873 }, { "epoch": 0.2172143438584465, "grad_norm": 0.4071959853172302, "learning_rate": 9.953505399746395e-05, "loss": 0.17530163526535034, "memory(GiB)": 57.83, "step": 5070, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.509903 }, { "epoch": 0.21742855918769546, "grad_norm": 2.0987300872802734, "learning_rate": 9.953413791943808e-05, "loss": 0.3404728412628174, "memory(GiB)": 57.83, "step": 5075, "token_acc": 0.9245901639344263, "train_speed(iter/s)": 1.509921 }, { "epoch": 0.21764277451694444, "grad_norm": 2.720486640930176, "learning_rate": 9.953322094405482e-05, "loss": 0.20528907775878907, "memory(GiB)": 57.83, "step": 5080, "token_acc": 0.9646302250803859, "train_speed(iter/s)": 1.509986 }, { "epoch": 0.2178569898461934, "grad_norm": 2.70168399810791, "learning_rate": 9.953230307133082e-05, "loss": 0.31954638957977294, "memory(GiB)": 57.83, "step": 5085, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.51007 }, { "epoch": 0.21807120517544235, "grad_norm": 1.6465981006622314, "learning_rate": 9.953138430128266e-05, "loss": 0.2926447868347168, "memory(GiB)": 57.83, "step": 5090, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.51009 }, { "epoch": 0.21828542050469132, "grad_norm": 2.1916592121124268, "learning_rate": 9.953046463392703e-05, "loss": 0.5407407760620118, "memory(GiB)": 57.83, "step": 5095, "token_acc": 0.890625, "train_speed(iter/s)": 1.510061 }, { "epoch": 0.21849963583394028, "grad_norm": 2.6684317588806152, "learning_rate": 9.952954406928056e-05, "loss": 0.548837947845459, "memory(GiB)": 57.83, "step": 5100, "token_acc": 0.8914956011730205, "train_speed(iter/s)": 1.509996 }, { "epoch": 0.21871385116318923, "grad_norm": 5.903489112854004, "learning_rate": 9.952862260735993e-05, "loss": 0.3151813268661499, "memory(GiB)": 57.83, "step": 5105, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.510063 }, { "epoch": 0.2189280664924382, "grad_norm": 1.8895301818847656, "learning_rate": 9.952770024818185e-05, "loss": 0.349971866607666, "memory(GiB)": 57.83, "step": 5110, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.510125 }, { "epoch": 0.21914228182168716, "grad_norm": 4.979547023773193, "learning_rate": 9.952677699176301e-05, "loss": 0.47161126136779785, "memory(GiB)": 57.83, "step": 5115, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.510141 }, { "epoch": 0.2193564971509361, "grad_norm": 2.4540181159973145, "learning_rate": 9.952585283812015e-05, "loss": 0.5155468940734863, "memory(GiB)": 57.83, "step": 5120, "token_acc": 0.891156462585034, "train_speed(iter/s)": 1.510277 }, { "epoch": 0.2195707124801851, "grad_norm": 3.4255449771881104, "learning_rate": 9.952492778727e-05, "loss": 0.5414937496185303, "memory(GiB)": 57.83, "step": 5125, "token_acc": 0.8741258741258742, "train_speed(iter/s)": 1.510302 }, { "epoch": 0.21978492780943404, "grad_norm": 1.277795672416687, "learning_rate": 9.952400183922933e-05, "loss": 0.2674806356430054, "memory(GiB)": 57.83, "step": 5130, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.510252 }, { "epoch": 0.21999914313868302, "grad_norm": 3.1024203300476074, "learning_rate": 9.952307499401492e-05, "loss": 0.4587207317352295, "memory(GiB)": 57.83, "step": 5135, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.510196 }, { "epoch": 0.22021335846793197, "grad_norm": 1.8915486335754395, "learning_rate": 9.952214725164355e-05, "loss": 0.3985797882080078, "memory(GiB)": 57.83, "step": 5140, "token_acc": 0.9084507042253521, "train_speed(iter/s)": 1.510218 }, { "epoch": 0.22042757379718092, "grad_norm": 4.540233612060547, "learning_rate": 9.952121861213202e-05, "loss": 0.3215314865112305, "memory(GiB)": 57.83, "step": 5145, "token_acc": 0.9295302013422819, "train_speed(iter/s)": 1.510308 }, { "epoch": 0.2206417891264299, "grad_norm": 3.0935866832733154, "learning_rate": 9.952028907549715e-05, "loss": 0.4107221126556396, "memory(GiB)": 57.83, "step": 5150, "token_acc": 0.9233870967741935, "train_speed(iter/s)": 1.510337 }, { "epoch": 0.22085600445567885, "grad_norm": 3.1413145065307617, "learning_rate": 9.951935864175581e-05, "loss": 0.5816514015197753, "memory(GiB)": 57.83, "step": 5155, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.510519 }, { "epoch": 0.2210702197849278, "grad_norm": 3.121201515197754, "learning_rate": 9.951842731092482e-05, "loss": 0.39647445678710935, "memory(GiB)": 57.83, "step": 5160, "token_acc": 0.9063545150501672, "train_speed(iter/s)": 1.510542 }, { "epoch": 0.22128443511417678, "grad_norm": 2.1440978050231934, "learning_rate": 9.951749508302106e-05, "loss": 0.30507454872131345, "memory(GiB)": 57.83, "step": 5165, "token_acc": 0.9299191374663073, "train_speed(iter/s)": 1.510545 }, { "epoch": 0.22149865044342573, "grad_norm": 3.7923879623413086, "learning_rate": 9.951656195806145e-05, "loss": 0.28185625076293946, "memory(GiB)": 57.83, "step": 5170, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.510549 }, { "epoch": 0.22171286577267468, "grad_norm": 2.7128372192382812, "learning_rate": 9.951562793606286e-05, "loss": 0.29074954986572266, "memory(GiB)": 57.83, "step": 5175, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.510602 }, { "epoch": 0.22192708110192366, "grad_norm": 1.5737231969833374, "learning_rate": 9.951469301704221e-05, "loss": 0.29572305679321287, "memory(GiB)": 57.83, "step": 5180, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.51059 }, { "epoch": 0.2221412964311726, "grad_norm": 3.7051002979278564, "learning_rate": 9.951375720101645e-05, "loss": 0.43220534324646, "memory(GiB)": 57.83, "step": 5185, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.510546 }, { "epoch": 0.22235551176042156, "grad_norm": 4.30851411819458, "learning_rate": 9.951282048800255e-05, "loss": 0.3695226669311523, "memory(GiB)": 57.83, "step": 5190, "token_acc": 0.9176954732510288, "train_speed(iter/s)": 1.510835 }, { "epoch": 0.22256972708967054, "grad_norm": 4.588947296142578, "learning_rate": 9.951188287801744e-05, "loss": 0.16680512428283692, "memory(GiB)": 57.83, "step": 5195, "token_acc": 0.961038961038961, "train_speed(iter/s)": 1.51088 }, { "epoch": 0.2227839424189195, "grad_norm": 2.491497278213501, "learning_rate": 9.951094437107812e-05, "loss": 0.373577356338501, "memory(GiB)": 57.83, "step": 5200, "token_acc": 0.9291338582677166, "train_speed(iter/s)": 1.510948 }, { "epoch": 0.22299815774816845, "grad_norm": 1.0047444105148315, "learning_rate": 9.951000496720162e-05, "loss": 0.2987224817276001, "memory(GiB)": 57.83, "step": 5205, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.510943 }, { "epoch": 0.22321237307741743, "grad_norm": 0.6407073140144348, "learning_rate": 9.950906466640493e-05, "loss": 0.08725958466529846, "memory(GiB)": 57.83, "step": 5210, "token_acc": 0.9703389830508474, "train_speed(iter/s)": 1.510841 }, { "epoch": 0.22342658840666638, "grad_norm": 3.1565215587615967, "learning_rate": 9.950812346870508e-05, "loss": 0.34076144695281985, "memory(GiB)": 57.83, "step": 5215, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.510908 }, { "epoch": 0.22364080373591536, "grad_norm": 1.9211819171905518, "learning_rate": 9.950718137411913e-05, "loss": 0.31551053524017336, "memory(GiB)": 57.83, "step": 5220, "token_acc": 0.9425587467362925, "train_speed(iter/s)": 1.511015 }, { "epoch": 0.2238550190651643, "grad_norm": 4.720902919769287, "learning_rate": 9.950623838266415e-05, "loss": 0.457856559753418, "memory(GiB)": 57.83, "step": 5225, "token_acc": 0.8590785907859079, "train_speed(iter/s)": 1.511158 }, { "epoch": 0.22406923439441326, "grad_norm": 4.120320796966553, "learning_rate": 9.950529449435722e-05, "loss": 0.34027581214904784, "memory(GiB)": 57.83, "step": 5230, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.511135 }, { "epoch": 0.22428344972366224, "grad_norm": 4.801386833190918, "learning_rate": 9.950434970921544e-05, "loss": 0.6660527229309082, "memory(GiB)": 57.83, "step": 5235, "token_acc": 0.8771331058020477, "train_speed(iter/s)": 1.511102 }, { "epoch": 0.2244976650529112, "grad_norm": 0.04696160554885864, "learning_rate": 9.950340402725596e-05, "loss": 0.35281646251678467, "memory(GiB)": 57.83, "step": 5240, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.511111 }, { "epoch": 0.22471188038216014, "grad_norm": 1.169893503189087, "learning_rate": 9.950245744849583e-05, "loss": 0.31919116973876954, "memory(GiB)": 57.83, "step": 5245, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.511167 }, { "epoch": 0.22492609571140912, "grad_norm": 2.2136363983154297, "learning_rate": 9.950150997295226e-05, "loss": 0.35060093402862547, "memory(GiB)": 57.83, "step": 5250, "token_acc": 0.9409937888198758, "train_speed(iter/s)": 1.51112 }, { "epoch": 0.22514031104065807, "grad_norm": 1.971913456916809, "learning_rate": 9.950056160064242e-05, "loss": 0.26299481391906737, "memory(GiB)": 57.83, "step": 5255, "token_acc": 0.9346153846153846, "train_speed(iter/s)": 1.511047 }, { "epoch": 0.22535452636990702, "grad_norm": 3.196977376937866, "learning_rate": 9.949961233158346e-05, "loss": 0.40016889572143555, "memory(GiB)": 57.83, "step": 5260, "token_acc": 0.899641577060932, "train_speed(iter/s)": 1.511109 }, { "epoch": 0.225568741699156, "grad_norm": 4.53477668762207, "learning_rate": 9.949866216579258e-05, "loss": 0.4731039524078369, "memory(GiB)": 57.83, "step": 5265, "token_acc": 0.8763636363636363, "train_speed(iter/s)": 1.511188 }, { "epoch": 0.22578295702840495, "grad_norm": 1.6615723371505737, "learning_rate": 9.949771110328701e-05, "loss": 0.30172224044799806, "memory(GiB)": 57.83, "step": 5270, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.511325 }, { "epoch": 0.2259971723576539, "grad_norm": 0.7929633855819702, "learning_rate": 9.949675914408396e-05, "loss": 0.22967841625213622, "memory(GiB)": 57.83, "step": 5275, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.511481 }, { "epoch": 0.22621138768690288, "grad_norm": 2.9471487998962402, "learning_rate": 9.94958062882007e-05, "loss": 0.6572547912597656, "memory(GiB)": 57.83, "step": 5280, "token_acc": 0.8546712802768166, "train_speed(iter/s)": 1.511526 }, { "epoch": 0.22642560301615183, "grad_norm": 3.272223711013794, "learning_rate": 9.949485253565446e-05, "loss": 0.34925568103790283, "memory(GiB)": 57.83, "step": 5285, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.511583 }, { "epoch": 0.22663981834540078, "grad_norm": 4.095811367034912, "learning_rate": 9.949389788646257e-05, "loss": 0.5195927619934082, "memory(GiB)": 57.83, "step": 5290, "token_acc": 0.8765822784810127, "train_speed(iter/s)": 1.511747 }, { "epoch": 0.22685403367464976, "grad_norm": 5.455345153808594, "learning_rate": 9.949294234064226e-05, "loss": 0.34976768493652344, "memory(GiB)": 57.83, "step": 5295, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.511749 }, { "epoch": 0.22706824900389871, "grad_norm": 10.6483793258667, "learning_rate": 9.949198589821089e-05, "loss": 0.41425271034240724, "memory(GiB)": 57.83, "step": 5300, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.511684 }, { "epoch": 0.2272824643331477, "grad_norm": 6.114230155944824, "learning_rate": 9.949102855918575e-05, "loss": 0.560057258605957, "memory(GiB)": 57.83, "step": 5305, "token_acc": 0.8844984802431611, "train_speed(iter/s)": 1.511763 }, { "epoch": 0.22749667966239664, "grad_norm": 6.928066730499268, "learning_rate": 9.949007032358422e-05, "loss": 0.7222353458404541, "memory(GiB)": 57.83, "step": 5310, "token_acc": 0.8270676691729323, "train_speed(iter/s)": 1.511987 }, { "epoch": 0.2277108949916456, "grad_norm": 3.095252275466919, "learning_rate": 9.948911119142363e-05, "loss": 0.1847246527671814, "memory(GiB)": 57.83, "step": 5315, "token_acc": 0.964516129032258, "train_speed(iter/s)": 1.511976 }, { "epoch": 0.22792511032089458, "grad_norm": 0.5363704562187195, "learning_rate": 9.948815116272137e-05, "loss": 0.3391617774963379, "memory(GiB)": 57.83, "step": 5320, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.512032 }, { "epoch": 0.22813932565014353, "grad_norm": 1.4267855882644653, "learning_rate": 9.948719023749482e-05, "loss": 0.31775732040405275, "memory(GiB)": 57.83, "step": 5325, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.51207 }, { "epoch": 0.22835354097939248, "grad_norm": 4.1315765380859375, "learning_rate": 9.94862284157614e-05, "loss": 0.39707896709442136, "memory(GiB)": 57.83, "step": 5330, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.512011 }, { "epoch": 0.22856775630864146, "grad_norm": 2.8201379776000977, "learning_rate": 9.948526569753853e-05, "loss": 0.32093186378479005, "memory(GiB)": 57.83, "step": 5335, "token_acc": 0.9136690647482014, "train_speed(iter/s)": 1.51203 }, { "epoch": 0.2287819716378904, "grad_norm": 3.2525758743286133, "learning_rate": 9.948430208284366e-05, "loss": 0.5865938663482666, "memory(GiB)": 57.83, "step": 5340, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.512157 }, { "epoch": 0.22899618696713936, "grad_norm": 8.997231483459473, "learning_rate": 9.948333757169424e-05, "loss": 0.3157021999359131, "memory(GiB)": 57.83, "step": 5345, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.512288 }, { "epoch": 0.22921040229638834, "grad_norm": 2.3136560916900635, "learning_rate": 9.948237216410771e-05, "loss": 0.42507014274597166, "memory(GiB)": 57.83, "step": 5350, "token_acc": 0.9305135951661632, "train_speed(iter/s)": 1.512369 }, { "epoch": 0.2294246176256373, "grad_norm": 4.124917984008789, "learning_rate": 9.948140586010162e-05, "loss": 0.31001982688903806, "memory(GiB)": 57.83, "step": 5355, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.512448 }, { "epoch": 0.22963883295488624, "grad_norm": 3.143082857131958, "learning_rate": 9.948043865969344e-05, "loss": 0.7840003967285156, "memory(GiB)": 57.83, "step": 5360, "token_acc": 0.8561151079136691, "train_speed(iter/s)": 1.512538 }, { "epoch": 0.22985304828413522, "grad_norm": 4.400677680969238, "learning_rate": 9.94794705629007e-05, "loss": 0.38356771469116213, "memory(GiB)": 57.83, "step": 5365, "token_acc": 0.8971631205673759, "train_speed(iter/s)": 1.512624 }, { "epoch": 0.23006726361338417, "grad_norm": 2.106562376022339, "learning_rate": 9.947850156974093e-05, "loss": 0.2474466323852539, "memory(GiB)": 57.83, "step": 5370, "token_acc": 0.9358974358974359, "train_speed(iter/s)": 1.512742 }, { "epoch": 0.23028147894263312, "grad_norm": 2.5826547145843506, "learning_rate": 9.947753168023168e-05, "loss": 0.1397019624710083, "memory(GiB)": 57.83, "step": 5375, "token_acc": 0.9624060150375939, "train_speed(iter/s)": 1.512711 }, { "epoch": 0.2304956942718821, "grad_norm": 2.5432941913604736, "learning_rate": 9.947656089439055e-05, "loss": 0.44531831741333006, "memory(GiB)": 57.83, "step": 5380, "token_acc": 0.8992248062015504, "train_speed(iter/s)": 1.512806 }, { "epoch": 0.23070990960113105, "grad_norm": 1.2170251607894897, "learning_rate": 9.947558921223509e-05, "loss": 0.2659919023513794, "memory(GiB)": 57.83, "step": 5385, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.512882 }, { "epoch": 0.23092412493038003, "grad_norm": 2.260990619659424, "learning_rate": 9.947461663378292e-05, "loss": 0.268111252784729, "memory(GiB)": 57.83, "step": 5390, "token_acc": 0.9421221864951769, "train_speed(iter/s)": 1.512845 }, { "epoch": 0.23113834025962898, "grad_norm": 1.892275094985962, "learning_rate": 9.947364315905168e-05, "loss": 0.3372787952423096, "memory(GiB)": 57.83, "step": 5395, "token_acc": 0.9398496240601504, "train_speed(iter/s)": 1.512899 }, { "epoch": 0.23135255558887793, "grad_norm": 3.1225664615631104, "learning_rate": 9.947266878805896e-05, "loss": 0.487364387512207, "memory(GiB)": 57.83, "step": 5400, "token_acc": 0.8860294117647058, "train_speed(iter/s)": 1.512891 }, { "epoch": 0.2315667709181269, "grad_norm": 5.693371295928955, "learning_rate": 9.947169352082245e-05, "loss": 0.3013789176940918, "memory(GiB)": 57.83, "step": 5405, "token_acc": 0.9385964912280702, "train_speed(iter/s)": 1.512806 }, { "epoch": 0.23178098624737586, "grad_norm": 3.890791416168213, "learning_rate": 9.94707173573598e-05, "loss": 0.41579318046569824, "memory(GiB)": 57.83, "step": 5410, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.512745 }, { "epoch": 0.23199520157662482, "grad_norm": 4.252648830413818, "learning_rate": 9.946974029768869e-05, "loss": 0.36588201522827146, "memory(GiB)": 57.83, "step": 5415, "token_acc": 0.9381107491856677, "train_speed(iter/s)": 1.512677 }, { "epoch": 0.2322094169058738, "grad_norm": 1.296302318572998, "learning_rate": 9.946876234182682e-05, "loss": 0.3545071601867676, "memory(GiB)": 57.83, "step": 5420, "token_acc": 0.9197707736389685, "train_speed(iter/s)": 1.512705 }, { "epoch": 0.23242363223512275, "grad_norm": 2.104524612426758, "learning_rate": 9.946778348979193e-05, "loss": 0.30637402534484864, "memory(GiB)": 57.83, "step": 5425, "token_acc": 0.9405204460966543, "train_speed(iter/s)": 1.512849 }, { "epoch": 0.2326378475643717, "grad_norm": 2.365469217300415, "learning_rate": 9.946680374160174e-05, "loss": 0.24333884716033935, "memory(GiB)": 57.83, "step": 5430, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.513084 }, { "epoch": 0.23285206289362068, "grad_norm": 3.718229055404663, "learning_rate": 9.9465823097274e-05, "loss": 0.6619626045227051, "memory(GiB)": 57.83, "step": 5435, "token_acc": 0.878125, "train_speed(iter/s)": 1.513219 }, { "epoch": 0.23306627822286963, "grad_norm": 0.0720333531498909, "learning_rate": 9.946484155682646e-05, "loss": 0.15011850595474244, "memory(GiB)": 57.83, "step": 5440, "token_acc": 0.9644268774703557, "train_speed(iter/s)": 1.513348 }, { "epoch": 0.23328049355211858, "grad_norm": 2.354560613632202, "learning_rate": 9.946385912027692e-05, "loss": 0.4197212219238281, "memory(GiB)": 57.83, "step": 5445, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.513425 }, { "epoch": 0.23349470888136756, "grad_norm": 2.438321590423584, "learning_rate": 9.946287578764318e-05, "loss": 0.2757858753204346, "memory(GiB)": 57.83, "step": 5450, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.51355 }, { "epoch": 0.2337089242106165, "grad_norm": 4.068017959594727, "learning_rate": 9.946189155894303e-05, "loss": 0.5945128917694091, "memory(GiB)": 57.83, "step": 5455, "token_acc": 0.8908450704225352, "train_speed(iter/s)": 1.513794 }, { "epoch": 0.23392313953986546, "grad_norm": 8.71953010559082, "learning_rate": 9.946090643419432e-05, "loss": 0.5110372066497803, "memory(GiB)": 57.83, "step": 5460, "token_acc": 0.8885245901639345, "train_speed(iter/s)": 1.513743 }, { "epoch": 0.23413735486911444, "grad_norm": 3.1455283164978027, "learning_rate": 9.945992041341489e-05, "loss": 0.2318859577178955, "memory(GiB)": 57.83, "step": 5465, "token_acc": 0.9438943894389439, "train_speed(iter/s)": 1.513782 }, { "epoch": 0.2343515701983634, "grad_norm": 5.344240188598633, "learning_rate": 9.945893349662261e-05, "loss": 0.29605422019958494, "memory(GiB)": 57.83, "step": 5470, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.513706 }, { "epoch": 0.23456578552761237, "grad_norm": 2.3667197227478027, "learning_rate": 9.945794568383534e-05, "loss": 0.43529467582702636, "memory(GiB)": 57.83, "step": 5475, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.513873 }, { "epoch": 0.23478000085686132, "grad_norm": 2.797313690185547, "learning_rate": 9.945695697507101e-05, "loss": 0.3729333639144897, "memory(GiB)": 57.83, "step": 5480, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.514044 }, { "epoch": 0.23499421618611027, "grad_norm": 0.5955561995506287, "learning_rate": 9.945596737034748e-05, "loss": 0.1907772421836853, "memory(GiB)": 57.83, "step": 5485, "token_acc": 0.9518518518518518, "train_speed(iter/s)": 1.514244 }, { "epoch": 0.23520843151535925, "grad_norm": 4.504206657409668, "learning_rate": 9.945497686968273e-05, "loss": 0.25456700325012205, "memory(GiB)": 57.83, "step": 5490, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.514363 }, { "epoch": 0.2354226468446082, "grad_norm": 2.924765110015869, "learning_rate": 9.945398547309467e-05, "loss": 0.5754400253295898, "memory(GiB)": 57.83, "step": 5495, "token_acc": 0.8952095808383234, "train_speed(iter/s)": 1.514374 }, { "epoch": 0.23563686217385715, "grad_norm": 2.1713807582855225, "learning_rate": 9.945299318060129e-05, "loss": 0.4750159740447998, "memory(GiB)": 57.83, "step": 5500, "token_acc": 0.8837209302325582, "train_speed(iter/s)": 1.514276 }, { "epoch": 0.23563686217385715, "eval_loss": 2.562443494796753, "eval_runtime": 12.8283, "eval_samples_per_second": 7.795, "eval_steps_per_second": 7.795, "eval_token_acc": 0.43073047858942065, "step": 5500 }, { "epoch": 0.23585107750310613, "grad_norm": 2.902458906173706, "learning_rate": 9.945199999222053e-05, "loss": 0.48052301406860354, "memory(GiB)": 57.83, "step": 5505, "token_acc": 0.5678793256433008, "train_speed(iter/s)": 1.508463 }, { "epoch": 0.23606529283235508, "grad_norm": 2.8268961906433105, "learning_rate": 9.945100590797041e-05, "loss": 0.17994810342788697, "memory(GiB)": 57.83, "step": 5510, "token_acc": 0.9567901234567902, "train_speed(iter/s)": 1.508501 }, { "epoch": 0.23627950816160403, "grad_norm": 0.816490888595581, "learning_rate": 9.945001092786893e-05, "loss": 0.2698091506958008, "memory(GiB)": 57.83, "step": 5515, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.508505 }, { "epoch": 0.23649372349085301, "grad_norm": 0.8489622473716736, "learning_rate": 9.944901505193411e-05, "loss": 0.661684513092041, "memory(GiB)": 57.83, "step": 5520, "token_acc": 0.8794326241134752, "train_speed(iter/s)": 1.508536 }, { "epoch": 0.23670793882010197, "grad_norm": 0.5180311799049377, "learning_rate": 9.9448018280184e-05, "loss": 0.35913102626800536, "memory(GiB)": 57.83, "step": 5525, "token_acc": 0.9320754716981132, "train_speed(iter/s)": 1.50853 }, { "epoch": 0.23692215414935092, "grad_norm": 3.2442588806152344, "learning_rate": 9.944702061263664e-05, "loss": 0.4688821792602539, "memory(GiB)": 57.83, "step": 5530, "token_acc": 0.89, "train_speed(iter/s)": 1.508531 }, { "epoch": 0.2371363694785999, "grad_norm": 4.331479072570801, "learning_rate": 9.944602204931013e-05, "loss": 0.2763098955154419, "memory(GiB)": 57.83, "step": 5535, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.508649 }, { "epoch": 0.23735058480784885, "grad_norm": 2.188426971435547, "learning_rate": 9.944502259022255e-05, "loss": 0.3549040794372559, "memory(GiB)": 57.83, "step": 5540, "token_acc": 0.9258064516129032, "train_speed(iter/s)": 1.508717 }, { "epoch": 0.2375648001370978, "grad_norm": 2.1943702697753906, "learning_rate": 9.944402223539199e-05, "loss": 0.2968748092651367, "memory(GiB)": 57.83, "step": 5545, "token_acc": 0.9288537549407114, "train_speed(iter/s)": 1.508711 }, { "epoch": 0.23777901546634678, "grad_norm": 3.432704448699951, "learning_rate": 9.944302098483659e-05, "loss": 0.6237911224365235, "memory(GiB)": 57.83, "step": 5550, "token_acc": 0.9, "train_speed(iter/s)": 1.508806 }, { "epoch": 0.23799323079559573, "grad_norm": 2.660513401031494, "learning_rate": 9.944201883857449e-05, "loss": 0.4590492248535156, "memory(GiB)": 57.83, "step": 5555, "token_acc": 0.9133858267716536, "train_speed(iter/s)": 1.508811 }, { "epoch": 0.2382074461248447, "grad_norm": 2.025660276412964, "learning_rate": 9.944101579662381e-05, "loss": 0.3621434211730957, "memory(GiB)": 57.83, "step": 5560, "token_acc": 0.9177631578947368, "train_speed(iter/s)": 1.508813 }, { "epoch": 0.23842166145409366, "grad_norm": 3.6760523319244385, "learning_rate": 9.944001185900278e-05, "loss": 0.4041323184967041, "memory(GiB)": 57.83, "step": 5565, "token_acc": 0.9129032258064517, "train_speed(iter/s)": 1.50883 }, { "epoch": 0.2386358767833426, "grad_norm": 3.9829273223876953, "learning_rate": 9.943900702572955e-05, "loss": 0.46795082092285156, "memory(GiB)": 57.83, "step": 5570, "token_acc": 0.9133858267716536, "train_speed(iter/s)": 1.508845 }, { "epoch": 0.2388500921125916, "grad_norm": 2.142476797103882, "learning_rate": 9.943800129682233e-05, "loss": 0.18385369777679444, "memory(GiB)": 57.83, "step": 5575, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.508919 }, { "epoch": 0.23906430744184054, "grad_norm": 3.0535292625427246, "learning_rate": 9.943699467229935e-05, "loss": 0.43838043212890626, "memory(GiB)": 57.83, "step": 5580, "token_acc": 0.8892307692307693, "train_speed(iter/s)": 1.508961 }, { "epoch": 0.2392785227710895, "grad_norm": 1.9316390752792358, "learning_rate": 9.94359871521788e-05, "loss": 0.296180272102356, "memory(GiB)": 57.83, "step": 5585, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.50894 }, { "epoch": 0.23949273810033847, "grad_norm": 3.25744366645813, "learning_rate": 9.9434978736479e-05, "loss": 0.5372891426086426, "memory(GiB)": 57.83, "step": 5590, "token_acc": 0.8873994638069705, "train_speed(iter/s)": 1.508991 }, { "epoch": 0.23970695342958742, "grad_norm": 3.0241265296936035, "learning_rate": 9.943396942521818e-05, "loss": 0.6532359600067139, "memory(GiB)": 57.83, "step": 5595, "token_acc": 0.8585858585858586, "train_speed(iter/s)": 1.508983 }, { "epoch": 0.23992116875883637, "grad_norm": 2.477348804473877, "learning_rate": 9.943295921841462e-05, "loss": 0.3205082654953003, "memory(GiB)": 57.83, "step": 5600, "token_acc": 0.9319148936170213, "train_speed(iter/s)": 1.509086 }, { "epoch": 0.24013538408808535, "grad_norm": 5.338733196258545, "learning_rate": 9.943194811608665e-05, "loss": 0.39242305755615237, "memory(GiB)": 57.83, "step": 5605, "token_acc": 0.8969072164948454, "train_speed(iter/s)": 1.509217 }, { "epoch": 0.2403495994173343, "grad_norm": 1.4055964946746826, "learning_rate": 9.943093611825256e-05, "loss": 0.25687711238861083, "memory(GiB)": 57.83, "step": 5610, "token_acc": 0.9283489096573209, "train_speed(iter/s)": 1.509254 }, { "epoch": 0.24056381474658325, "grad_norm": 1.6302130222320557, "learning_rate": 9.942992322493068e-05, "loss": 0.28458738327026367, "memory(GiB)": 57.83, "step": 5615, "token_acc": 0.934984520123839, "train_speed(iter/s)": 1.509267 }, { "epoch": 0.24077803007583223, "grad_norm": 1.93070387840271, "learning_rate": 9.942890943613939e-05, "loss": 0.4355536937713623, "memory(GiB)": 57.83, "step": 5620, "token_acc": 0.8926174496644296, "train_speed(iter/s)": 1.509262 }, { "epoch": 0.24099224540508118, "grad_norm": 2.4056849479675293, "learning_rate": 9.942789475189704e-05, "loss": 0.36538283824920653, "memory(GiB)": 57.83, "step": 5625, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.509423 }, { "epoch": 0.24120646073433014, "grad_norm": 1.8821650743484497, "learning_rate": 9.9426879172222e-05, "loss": 0.295369029045105, "memory(GiB)": 57.83, "step": 5630, "token_acc": 0.9550561797752809, "train_speed(iter/s)": 1.509475 }, { "epoch": 0.24142067606357911, "grad_norm": 2.8725759983062744, "learning_rate": 9.942586269713268e-05, "loss": 0.3583981037139893, "memory(GiB)": 57.83, "step": 5635, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.509592 }, { "epoch": 0.24163489139282807, "grad_norm": 2.384573221206665, "learning_rate": 9.942484532664748e-05, "loss": 0.24067537784576415, "memory(GiB)": 57.83, "step": 5640, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.509604 }, { "epoch": 0.24184910672207705, "grad_norm": 0.7550927996635437, "learning_rate": 9.942382706078486e-05, "loss": 0.39318921566009524, "memory(GiB)": 57.83, "step": 5645, "token_acc": 0.9228723404255319, "train_speed(iter/s)": 1.509697 }, { "epoch": 0.242063322051326, "grad_norm": 5.150145053863525, "learning_rate": 9.942280789956325e-05, "loss": 0.6348095893859863, "memory(GiB)": 57.83, "step": 5650, "token_acc": 0.8562300319488818, "train_speed(iter/s)": 1.510071 }, { "epoch": 0.24227753738057495, "grad_norm": 2.3230161666870117, "learning_rate": 9.942178784300109e-05, "loss": 0.4688426494598389, "memory(GiB)": 57.83, "step": 5655, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.510062 }, { "epoch": 0.24249175270982393, "grad_norm": 1.9243894815444946, "learning_rate": 9.94207668911169e-05, "loss": 0.3555765151977539, "memory(GiB)": 57.83, "step": 5660, "token_acc": 0.925, "train_speed(iter/s)": 1.510038 }, { "epoch": 0.24270596803907288, "grad_norm": 1.6827868223190308, "learning_rate": 9.941974504392916e-05, "loss": 0.5181779861450195, "memory(GiB)": 57.83, "step": 5665, "token_acc": 0.884375, "train_speed(iter/s)": 1.510111 }, { "epoch": 0.24292018336832183, "grad_norm": 4.821323871612549, "learning_rate": 9.941872230145637e-05, "loss": 0.2714211463928223, "memory(GiB)": 57.83, "step": 5670, "token_acc": 0.933852140077821, "train_speed(iter/s)": 1.510096 }, { "epoch": 0.2431343986975708, "grad_norm": 1.8279590606689453, "learning_rate": 9.941769866371708e-05, "loss": 0.4186239719390869, "memory(GiB)": 57.83, "step": 5675, "token_acc": 0.9097744360902256, "train_speed(iter/s)": 1.510038 }, { "epoch": 0.24334861402681976, "grad_norm": 3.8753745555877686, "learning_rate": 9.94166741307298e-05, "loss": 0.4857377529144287, "memory(GiB)": 57.83, "step": 5680, "token_acc": 0.8932384341637011, "train_speed(iter/s)": 1.510033 }, { "epoch": 0.2435628293560687, "grad_norm": 2.4755523204803467, "learning_rate": 9.941564870251312e-05, "loss": 0.37761640548706055, "memory(GiB)": 57.83, "step": 5685, "token_acc": 0.9104938271604939, "train_speed(iter/s)": 1.50998 }, { "epoch": 0.2437770446853177, "grad_norm": 0.06999822705984116, "learning_rate": 9.941462237908561e-05, "loss": 0.5130293369293213, "memory(GiB)": 57.83, "step": 5690, "token_acc": 0.8519553072625698, "train_speed(iter/s)": 1.510115 }, { "epoch": 0.24399126001456664, "grad_norm": 0.5665782690048218, "learning_rate": 9.941359516046586e-05, "loss": 0.23040213584899902, "memory(GiB)": 57.83, "step": 5695, "token_acc": 0.9658119658119658, "train_speed(iter/s)": 1.510301 }, { "epoch": 0.2442054753438156, "grad_norm": 3.069899797439575, "learning_rate": 9.941256704667249e-05, "loss": 0.4131608009338379, "memory(GiB)": 57.83, "step": 5700, "token_acc": 0.8982456140350877, "train_speed(iter/s)": 1.51041 }, { "epoch": 0.24441969067306457, "grad_norm": 0.2548567056655884, "learning_rate": 9.941153803772412e-05, "loss": 0.43128304481506347, "memory(GiB)": 57.83, "step": 5705, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.510359 }, { "epoch": 0.24463390600231352, "grad_norm": 1.8311127424240112, "learning_rate": 9.941050813363937e-05, "loss": 0.2588664054870605, "memory(GiB)": 57.83, "step": 5710, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.510323 }, { "epoch": 0.24484812133156247, "grad_norm": 1.2560371160507202, "learning_rate": 9.94094773344369e-05, "loss": 0.24485232830047607, "memory(GiB)": 57.83, "step": 5715, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.510252 }, { "epoch": 0.24506233666081145, "grad_norm": 2.676442861557007, "learning_rate": 9.940844564013542e-05, "loss": 0.5758266448974609, "memory(GiB)": 57.83, "step": 5720, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.510476 }, { "epoch": 0.2452765519900604, "grad_norm": 2.0763230323791504, "learning_rate": 9.94074130507536e-05, "loss": 0.28690686225891116, "memory(GiB)": 57.83, "step": 5725, "token_acc": 0.9337748344370861, "train_speed(iter/s)": 1.510511 }, { "epoch": 0.24549076731930938, "grad_norm": 2.919618606567383, "learning_rate": 9.940637956631013e-05, "loss": 0.3124751329421997, "memory(GiB)": 57.83, "step": 5730, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.510559 }, { "epoch": 0.24570498264855833, "grad_norm": 2.847921848297119, "learning_rate": 9.940534518682376e-05, "loss": 0.516752052307129, "memory(GiB)": 57.83, "step": 5735, "token_acc": 0.8755020080321285, "train_speed(iter/s)": 1.51061 }, { "epoch": 0.24591919797780729, "grad_norm": 2.555818557739258, "learning_rate": 9.940430991231322e-05, "loss": 0.5540193557739258, "memory(GiB)": 57.83, "step": 5740, "token_acc": 0.8907849829351536, "train_speed(iter/s)": 1.510695 }, { "epoch": 0.24613341330705626, "grad_norm": 1.632690191268921, "learning_rate": 9.940327374279725e-05, "loss": 0.4879893779754639, "memory(GiB)": 57.83, "step": 5745, "token_acc": 0.8832116788321168, "train_speed(iter/s)": 1.510805 }, { "epoch": 0.24634762863630522, "grad_norm": 1.7316380739212036, "learning_rate": 9.940223667829465e-05, "loss": 0.530971622467041, "memory(GiB)": 57.83, "step": 5750, "token_acc": 0.8951310861423221, "train_speed(iter/s)": 1.510739 }, { "epoch": 0.24656184396555417, "grad_norm": 2.4903275966644287, "learning_rate": 9.940119871882416e-05, "loss": 0.14352777004241943, "memory(GiB)": 57.83, "step": 5755, "token_acc": 0.9579288025889967, "train_speed(iter/s)": 1.510707 }, { "epoch": 0.24677605929480315, "grad_norm": 4.270592212677002, "learning_rate": 9.940015986440464e-05, "loss": 0.6464215755462647, "memory(GiB)": 57.83, "step": 5760, "token_acc": 0.8766233766233766, "train_speed(iter/s)": 1.510813 }, { "epoch": 0.2469902746240521, "grad_norm": 2.738633155822754, "learning_rate": 9.939912011505487e-05, "loss": 0.472660493850708, "memory(GiB)": 57.83, "step": 5765, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.51089 }, { "epoch": 0.24720448995330105, "grad_norm": 3.218891143798828, "learning_rate": 9.939807947079369e-05, "loss": 0.43903179168701173, "memory(GiB)": 57.83, "step": 5770, "token_acc": 0.8935483870967742, "train_speed(iter/s)": 1.510977 }, { "epoch": 0.24741870528255003, "grad_norm": 0.3646969795227051, "learning_rate": 9.939703793163998e-05, "loss": 0.3428455352783203, "memory(GiB)": 57.83, "step": 5775, "token_acc": 0.9043824701195219, "train_speed(iter/s)": 1.511012 }, { "epoch": 0.24763292061179898, "grad_norm": 0.6202086806297302, "learning_rate": 9.939599549761259e-05, "loss": 0.3049603462219238, "memory(GiB)": 57.83, "step": 5780, "token_acc": 0.9407114624505929, "train_speed(iter/s)": 1.511018 }, { "epoch": 0.24784713594104793, "grad_norm": 4.795821189880371, "learning_rate": 9.939495216873038e-05, "loss": 0.511199951171875, "memory(GiB)": 57.83, "step": 5785, "token_acc": 0.8851963746223565, "train_speed(iter/s)": 1.511197 }, { "epoch": 0.2480613512702969, "grad_norm": 2.1931283473968506, "learning_rate": 9.93939079450123e-05, "loss": 0.2502848386764526, "memory(GiB)": 57.83, "step": 5790, "token_acc": 0.9423728813559322, "train_speed(iter/s)": 1.511291 }, { "epoch": 0.24827556659954586, "grad_norm": 6.448992729187012, "learning_rate": 9.939286282647723e-05, "loss": 0.7362441539764404, "memory(GiB)": 57.83, "step": 5795, "token_acc": 0.8734567901234568, "train_speed(iter/s)": 1.511247 }, { "epoch": 0.2484897819287948, "grad_norm": 1.588218092918396, "learning_rate": 9.939181681314411e-05, "loss": 0.34906442165374757, "memory(GiB)": 57.83, "step": 5800, "token_acc": 0.909375, "train_speed(iter/s)": 1.511251 }, { "epoch": 0.2487039972580438, "grad_norm": 1.735824465751648, "learning_rate": 9.939076990503191e-05, "loss": 0.3238396167755127, "memory(GiB)": 57.83, "step": 5805, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.511411 }, { "epoch": 0.24891821258729274, "grad_norm": 3.8043572902679443, "learning_rate": 9.938972210215958e-05, "loss": 0.37131457328796386, "memory(GiB)": 57.83, "step": 5810, "token_acc": 0.9214876033057852, "train_speed(iter/s)": 1.511541 }, { "epoch": 0.24913242791654172, "grad_norm": 2.871668577194214, "learning_rate": 9.938867340454609e-05, "loss": 0.513247013092041, "memory(GiB)": 57.83, "step": 5815, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.511612 }, { "epoch": 0.24934664324579067, "grad_norm": 0.849934458732605, "learning_rate": 9.938762381221047e-05, "loss": 0.2646330833435059, "memory(GiB)": 57.83, "step": 5820, "token_acc": 0.9308176100628931, "train_speed(iter/s)": 1.511641 }, { "epoch": 0.24956085857503962, "grad_norm": 3.6498281955718994, "learning_rate": 9.93865733251717e-05, "loss": 0.33288071155548093, "memory(GiB)": 57.83, "step": 5825, "token_acc": 0.9466192170818505, "train_speed(iter/s)": 1.511772 }, { "epoch": 0.2497750739042886, "grad_norm": 3.247913360595703, "learning_rate": 9.938552194344883e-05, "loss": 0.4855414867401123, "memory(GiB)": 57.83, "step": 5830, "token_acc": 0.8885448916408669, "train_speed(iter/s)": 1.51171 }, { "epoch": 0.24998928923353755, "grad_norm": 2.1371631622314453, "learning_rate": 9.93844696670609e-05, "loss": 0.5209426879882812, "memory(GiB)": 57.83, "step": 5835, "token_acc": 0.887459807073955, "train_speed(iter/s)": 1.511713 }, { "epoch": 0.2502035045627865, "grad_norm": 1.6553810834884644, "learning_rate": 9.938341649602698e-05, "loss": 0.43607077598571775, "memory(GiB)": 57.83, "step": 5840, "token_acc": 0.8872180451127819, "train_speed(iter/s)": 1.511652 }, { "epoch": 0.2504177198920355, "grad_norm": 2.712770938873291, "learning_rate": 9.938236243036613e-05, "loss": 0.3960700988769531, "memory(GiB)": 57.83, "step": 5845, "token_acc": 0.9174917491749175, "train_speed(iter/s)": 1.51167 }, { "epoch": 0.2506319352212844, "grad_norm": 1.7924548387527466, "learning_rate": 9.938130747009748e-05, "loss": 0.37431812286376953, "memory(GiB)": 57.83, "step": 5850, "token_acc": 0.92, "train_speed(iter/s)": 1.511694 }, { "epoch": 0.2508461505505334, "grad_norm": 2.274765968322754, "learning_rate": 9.938025161524012e-05, "loss": 0.6086924076080322, "memory(GiB)": 57.83, "step": 5855, "token_acc": 0.85, "train_speed(iter/s)": 1.511657 }, { "epoch": 0.25106036587978237, "grad_norm": 2.8473243713378906, "learning_rate": 9.937919486581317e-05, "loss": 0.4715022087097168, "memory(GiB)": 57.83, "step": 5860, "token_acc": 0.9158249158249159, "train_speed(iter/s)": 1.511643 }, { "epoch": 0.25127458120903134, "grad_norm": 4.891002655029297, "learning_rate": 9.937813722183579e-05, "loss": 0.3254450559616089, "memory(GiB)": 57.83, "step": 5865, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.511665 }, { "epoch": 0.25148879653828027, "grad_norm": 1.9175771474838257, "learning_rate": 9.937707868332713e-05, "loss": 0.21567115783691407, "memory(GiB)": 57.83, "step": 5870, "token_acc": 0.9463722397476341, "train_speed(iter/s)": 1.511761 }, { "epoch": 0.25170301186752925, "grad_norm": 3.0330631732940674, "learning_rate": 9.937601925030638e-05, "loss": 0.4073945999145508, "memory(GiB)": 57.83, "step": 5875, "token_acc": 0.907051282051282, "train_speed(iter/s)": 1.511758 }, { "epoch": 0.2519172271967782, "grad_norm": 3.6874682903289795, "learning_rate": 9.937495892279272e-05, "loss": 0.4708213806152344, "memory(GiB)": 57.83, "step": 5880, "token_acc": 0.897196261682243, "train_speed(iter/s)": 1.51192 }, { "epoch": 0.25213144252602715, "grad_norm": 7.353977680206299, "learning_rate": 9.937389770080535e-05, "loss": 0.26984405517578125, "memory(GiB)": 57.83, "step": 5885, "token_acc": 0.9361022364217252, "train_speed(iter/s)": 1.512107 }, { "epoch": 0.25234565785527613, "grad_norm": 4.260404586791992, "learning_rate": 9.937283558436352e-05, "loss": 0.515605878829956, "memory(GiB)": 57.83, "step": 5890, "token_acc": 0.8945578231292517, "train_speed(iter/s)": 1.512329 }, { "epoch": 0.2525598731845251, "grad_norm": 1.5599665641784668, "learning_rate": 9.937177257348645e-05, "loss": 0.21643147468566895, "memory(GiB)": 57.83, "step": 5895, "token_acc": 0.9513888888888888, "train_speed(iter/s)": 1.512413 }, { "epoch": 0.25277408851377403, "grad_norm": 1.393804907798767, "learning_rate": 9.93707086681934e-05, "loss": 0.47597761154174806, "memory(GiB)": 57.83, "step": 5900, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.512418 }, { "epoch": 0.252988303843023, "grad_norm": 7.8237833976745605, "learning_rate": 9.936964386850366e-05, "loss": 0.29127166271209715, "memory(GiB)": 57.83, "step": 5905, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.512579 }, { "epoch": 0.253202519172272, "grad_norm": 1.7640070915222168, "learning_rate": 9.93685781744365e-05, "loss": 0.4950842380523682, "memory(GiB)": 57.83, "step": 5910, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.512603 }, { "epoch": 0.2534167345015209, "grad_norm": 4.689111232757568, "learning_rate": 9.936751158601124e-05, "loss": 0.4903552532196045, "memory(GiB)": 57.83, "step": 5915, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.51284 }, { "epoch": 0.2536309498307699, "grad_norm": 3.578521251678467, "learning_rate": 9.93664441032472e-05, "loss": 0.5357282638549805, "memory(GiB)": 57.83, "step": 5920, "token_acc": 0.890625, "train_speed(iter/s)": 1.512975 }, { "epoch": 0.25384516516001887, "grad_norm": 1.4946907758712769, "learning_rate": 9.936537572616372e-05, "loss": 0.1460087776184082, "memory(GiB)": 57.83, "step": 5925, "token_acc": 0.9646643109540636, "train_speed(iter/s)": 1.512918 }, { "epoch": 0.2540593804892678, "grad_norm": 1.525254487991333, "learning_rate": 9.936430645478014e-05, "loss": 0.21395883560180665, "memory(GiB)": 57.83, "step": 5930, "token_acc": 0.9504950495049505, "train_speed(iter/s)": 1.512901 }, { "epoch": 0.2542735958185168, "grad_norm": 2.55198073387146, "learning_rate": 9.936323628911584e-05, "loss": 0.2551476001739502, "memory(GiB)": 57.83, "step": 5935, "token_acc": 0.9533333333333334, "train_speed(iter/s)": 1.512888 }, { "epoch": 0.25448781114776575, "grad_norm": 2.390024423599243, "learning_rate": 9.936216522919021e-05, "loss": 0.30000245571136475, "memory(GiB)": 57.83, "step": 5940, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.513093 }, { "epoch": 0.2547020264770147, "grad_norm": 2.3630917072296143, "learning_rate": 9.936109327502266e-05, "loss": 0.2939441680908203, "memory(GiB)": 57.83, "step": 5945, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.51313 }, { "epoch": 0.25491624180626365, "grad_norm": 2.0397026538848877, "learning_rate": 9.936002042663258e-05, "loss": 0.4676835536956787, "memory(GiB)": 57.83, "step": 5950, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.513206 }, { "epoch": 0.25513045713551263, "grad_norm": 2.654291868209839, "learning_rate": 9.935894668403945e-05, "loss": 0.5101615905761718, "memory(GiB)": 57.83, "step": 5955, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.51333 }, { "epoch": 0.25534467246476156, "grad_norm": 0.5669079422950745, "learning_rate": 9.935787204726268e-05, "loss": 0.4804722309112549, "memory(GiB)": 57.83, "step": 5960, "token_acc": 0.8924050632911392, "train_speed(iter/s)": 1.513354 }, { "epoch": 0.25555888779401054, "grad_norm": 7.910981178283691, "learning_rate": 9.935679651632177e-05, "loss": 0.4848268985748291, "memory(GiB)": 57.83, "step": 5965, "token_acc": 0.8923076923076924, "train_speed(iter/s)": 1.513461 }, { "epoch": 0.2557731031232595, "grad_norm": 3.056152820587158, "learning_rate": 9.935572009123619e-05, "loss": 0.22553761005401612, "memory(GiB)": 57.83, "step": 5970, "token_acc": 0.945859872611465, "train_speed(iter/s)": 1.513468 }, { "epoch": 0.25598731845250844, "grad_norm": 0.06077791005373001, "learning_rate": 9.935464277202544e-05, "loss": 0.6827641010284424, "memory(GiB)": 57.83, "step": 5975, "token_acc": 0.8713235294117647, "train_speed(iter/s)": 1.513566 }, { "epoch": 0.2562015337817574, "grad_norm": 1.8649221658706665, "learning_rate": 9.935356455870904e-05, "loss": 0.26709082126617434, "memory(GiB)": 57.83, "step": 5980, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.513551 }, { "epoch": 0.2564157491110064, "grad_norm": 2.91180419921875, "learning_rate": 9.93524854513065e-05, "loss": 0.41860089302062986, "memory(GiB)": 57.83, "step": 5985, "token_acc": 0.9027355623100304, "train_speed(iter/s)": 1.513538 }, { "epoch": 0.2566299644402553, "grad_norm": 2.233454942703247, "learning_rate": 9.93514054498374e-05, "loss": 0.4784512519836426, "memory(GiB)": 57.83, "step": 5990, "token_acc": 0.9105431309904153, "train_speed(iter/s)": 1.513587 }, { "epoch": 0.2568441797695043, "grad_norm": 2.9003942012786865, "learning_rate": 9.93503245543213e-05, "loss": 0.3319584369659424, "memory(GiB)": 57.83, "step": 5995, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.513656 }, { "epoch": 0.2570583950987533, "grad_norm": 6.065582752227783, "learning_rate": 9.934924276477779e-05, "loss": 0.9388570785522461, "memory(GiB)": 57.83, "step": 6000, "token_acc": 0.8288973384030418, "train_speed(iter/s)": 1.513667 }, { "epoch": 0.2570583950987533, "eval_loss": 2.597562551498413, "eval_runtime": 12.842, "eval_samples_per_second": 7.787, "eval_steps_per_second": 7.787, "eval_token_acc": 0.4169054441260745, "step": 6000 }, { "epoch": 0.2572726104280022, "grad_norm": 3.860740900039673, "learning_rate": 9.934816008122643e-05, "loss": 0.6344261646270752, "memory(GiB)": 57.83, "step": 6005, "token_acc": 0.548618219037871, "train_speed(iter/s)": 1.508699 }, { "epoch": 0.2574868257572512, "grad_norm": 2.9173896312713623, "learning_rate": 9.934707650368686e-05, "loss": 0.4706707954406738, "memory(GiB)": 57.83, "step": 6010, "token_acc": 0.887719298245614, "train_speed(iter/s)": 1.508721 }, { "epoch": 0.25770104108650016, "grad_norm": 3.3867757320404053, "learning_rate": 9.934599203217874e-05, "loss": 0.32073657512664794, "memory(GiB)": 57.83, "step": 6015, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.508777 }, { "epoch": 0.2579152564157491, "grad_norm": 1.2818666696548462, "learning_rate": 9.934490666672164e-05, "loss": 0.3143477916717529, "memory(GiB)": 57.83, "step": 6020, "token_acc": 0.9240506329113924, "train_speed(iter/s)": 1.508893 }, { "epoch": 0.25812947174499806, "grad_norm": 2.0671682357788086, "learning_rate": 9.93438204073353e-05, "loss": 0.42448949813842773, "memory(GiB)": 57.83, "step": 6025, "token_acc": 0.9113573407202216, "train_speed(iter/s)": 1.508865 }, { "epoch": 0.25834368707424704, "grad_norm": 0.9335662722587585, "learning_rate": 9.934273325403935e-05, "loss": 0.17360650300979613, "memory(GiB)": 57.83, "step": 6030, "token_acc": 0.9641693811074918, "train_speed(iter/s)": 1.508945 }, { "epoch": 0.258557902403496, "grad_norm": 3.428192138671875, "learning_rate": 9.934164520685349e-05, "loss": 0.45486106872558596, "memory(GiB)": 57.83, "step": 6035, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.509053 }, { "epoch": 0.25877211773274494, "grad_norm": 1.124332070350647, "learning_rate": 9.934055626579746e-05, "loss": 0.3286736965179443, "memory(GiB)": 57.83, "step": 6040, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.509181 }, { "epoch": 0.2589863330619939, "grad_norm": 5.32781982421875, "learning_rate": 9.933946643089096e-05, "loss": 0.2620370626449585, "memory(GiB)": 57.83, "step": 6045, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.509122 }, { "epoch": 0.2592005483912429, "grad_norm": 3.7868382930755615, "learning_rate": 9.933837570215374e-05, "loss": 0.4194647789001465, "memory(GiB)": 57.83, "step": 6050, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.509117 }, { "epoch": 0.2594147637204918, "grad_norm": 0.42373430728912354, "learning_rate": 9.933728407960556e-05, "loss": 0.3305581331253052, "memory(GiB)": 57.83, "step": 6055, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.509175 }, { "epoch": 0.2596289790497408, "grad_norm": 1.3747516870498657, "learning_rate": 9.933619156326621e-05, "loss": 0.21721489429473878, "memory(GiB)": 57.83, "step": 6060, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.509214 }, { "epoch": 0.2598431943789898, "grad_norm": 3.8711090087890625, "learning_rate": 9.933509815315545e-05, "loss": 0.2547018527984619, "memory(GiB)": 57.83, "step": 6065, "token_acc": 0.9528619528619529, "train_speed(iter/s)": 1.509209 }, { "epoch": 0.2600574097082387, "grad_norm": 4.243188381195068, "learning_rate": 9.933400384929313e-05, "loss": 0.5954276561737061, "memory(GiB)": 57.83, "step": 6070, "token_acc": 0.8631921824104235, "train_speed(iter/s)": 1.509226 }, { "epoch": 0.2602716250374877, "grad_norm": 2.7614622116088867, "learning_rate": 9.933290865169903e-05, "loss": 0.42738494873046873, "memory(GiB)": 57.83, "step": 6075, "token_acc": 0.9127725856697819, "train_speed(iter/s)": 1.509352 }, { "epoch": 0.26048584036673667, "grad_norm": 3.5418944358825684, "learning_rate": 9.933181256039301e-05, "loss": 0.4812305450439453, "memory(GiB)": 57.83, "step": 6080, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.509262 }, { "epoch": 0.2607000556959856, "grad_norm": 2.369384288787842, "learning_rate": 9.933071557539494e-05, "loss": 0.2539195537567139, "memory(GiB)": 57.83, "step": 6085, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.509278 }, { "epoch": 0.26091427102523457, "grad_norm": 1.8879045248031616, "learning_rate": 9.932961769672469e-05, "loss": 0.5910563945770264, "memory(GiB)": 57.83, "step": 6090, "token_acc": 0.890728476821192, "train_speed(iter/s)": 1.509334 }, { "epoch": 0.26112848635448355, "grad_norm": 1.497489333152771, "learning_rate": 9.932851892440211e-05, "loss": 0.2868366241455078, "memory(GiB)": 57.83, "step": 6095, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.509366 }, { "epoch": 0.26134270168373247, "grad_norm": 1.8995040655136108, "learning_rate": 9.932741925844717e-05, "loss": 0.27905290126800536, "memory(GiB)": 57.83, "step": 6100, "token_acc": 0.9312714776632303, "train_speed(iter/s)": 1.509373 }, { "epoch": 0.26155691701298145, "grad_norm": 3.1291537284851074, "learning_rate": 9.932631869887974e-05, "loss": 0.4369076728820801, "memory(GiB)": 57.83, "step": 6105, "token_acc": 0.8892307692307693, "train_speed(iter/s)": 1.509506 }, { "epoch": 0.26177113234223043, "grad_norm": 4.167664527893066, "learning_rate": 9.932521724571977e-05, "loss": 0.5211634635925293, "memory(GiB)": 57.83, "step": 6110, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.509572 }, { "epoch": 0.26198534767147935, "grad_norm": 4.496867656707764, "learning_rate": 9.932411489898723e-05, "loss": 0.4579469203948975, "memory(GiB)": 57.83, "step": 6115, "token_acc": 0.9218106995884774, "train_speed(iter/s)": 1.50963 }, { "epoch": 0.26219956300072833, "grad_norm": 1.0329607725143433, "learning_rate": 9.932301165870206e-05, "loss": 0.222471284866333, "memory(GiB)": 57.83, "step": 6120, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.509676 }, { "epoch": 0.2624137783299773, "grad_norm": 1.2141727209091187, "learning_rate": 9.932190752488428e-05, "loss": 0.2282252073287964, "memory(GiB)": 57.83, "step": 6125, "token_acc": 0.943609022556391, "train_speed(iter/s)": 1.509799 }, { "epoch": 0.26262799365922623, "grad_norm": 3.278120517730713, "learning_rate": 9.932080249755389e-05, "loss": 0.2538751602172852, "memory(GiB)": 57.83, "step": 6130, "token_acc": 0.9528985507246377, "train_speed(iter/s)": 1.509809 }, { "epoch": 0.2628422089884752, "grad_norm": 4.585535526275635, "learning_rate": 9.931969657673088e-05, "loss": 0.49903693199157717, "memory(GiB)": 57.83, "step": 6135, "token_acc": 0.8708609271523179, "train_speed(iter/s)": 1.510063 }, { "epoch": 0.2630564243177242, "grad_norm": 1.9522123336791992, "learning_rate": 9.931858976243531e-05, "loss": 0.3062258720397949, "memory(GiB)": 57.83, "step": 6140, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.510221 }, { "epoch": 0.2632706396469731, "grad_norm": 3.4891018867492676, "learning_rate": 9.931748205468721e-05, "loss": 0.42903151512146, "memory(GiB)": 57.83, "step": 6145, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.510277 }, { "epoch": 0.2634848549762221, "grad_norm": 1.9862321615219116, "learning_rate": 9.931637345350667e-05, "loss": 0.38887991905212405, "memory(GiB)": 57.83, "step": 6150, "token_acc": 0.9211956521739131, "train_speed(iter/s)": 1.510299 }, { "epoch": 0.2636990703054711, "grad_norm": 1.3124617338180542, "learning_rate": 9.931526395891375e-05, "loss": 0.2441765308380127, "memory(GiB)": 57.83, "step": 6155, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.510319 }, { "epoch": 0.26391328563472, "grad_norm": 5.413501262664795, "learning_rate": 9.931415357092858e-05, "loss": 0.306878399848938, "memory(GiB)": 57.83, "step": 6160, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.510337 }, { "epoch": 0.264127500963969, "grad_norm": 1.8066558837890625, "learning_rate": 9.931304228957123e-05, "loss": 0.45495071411132815, "memory(GiB)": 57.83, "step": 6165, "token_acc": 0.9157894736842105, "train_speed(iter/s)": 1.510431 }, { "epoch": 0.26434171629321795, "grad_norm": 1.5612115859985352, "learning_rate": 9.931193011486188e-05, "loss": 0.41306471824645996, "memory(GiB)": 57.83, "step": 6170, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.510632 }, { "epoch": 0.2645559316224669, "grad_norm": 2.812962293624878, "learning_rate": 9.931081704682066e-05, "loss": 0.3414760589599609, "memory(GiB)": 57.83, "step": 6175, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.510661 }, { "epoch": 0.26477014695171586, "grad_norm": 2.1002464294433594, "learning_rate": 9.930970308546772e-05, "loss": 0.3947620153427124, "memory(GiB)": 57.83, "step": 6180, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.510635 }, { "epoch": 0.26498436228096484, "grad_norm": 1.0055547952651978, "learning_rate": 9.930858823082327e-05, "loss": 0.3768031597137451, "memory(GiB)": 57.83, "step": 6185, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.510568 }, { "epoch": 0.26519857761021376, "grad_norm": 3.471759557723999, "learning_rate": 9.930747248290747e-05, "loss": 0.2893956184387207, "memory(GiB)": 57.83, "step": 6190, "token_acc": 0.9400749063670412, "train_speed(iter/s)": 1.510581 }, { "epoch": 0.26541279293946274, "grad_norm": 3.5775911808013916, "learning_rate": 9.930635584174056e-05, "loss": 0.5177459716796875, "memory(GiB)": 57.83, "step": 6195, "token_acc": 0.8931750741839762, "train_speed(iter/s)": 1.510705 }, { "epoch": 0.2656270082687117, "grad_norm": 2.467583179473877, "learning_rate": 9.930523830734276e-05, "loss": 0.49843549728393555, "memory(GiB)": 57.83, "step": 6200, "token_acc": 0.8991097922848664, "train_speed(iter/s)": 1.510667 }, { "epoch": 0.2658412235979607, "grad_norm": 1.0195355415344238, "learning_rate": 9.930411987973431e-05, "loss": 0.39285480976104736, "memory(GiB)": 57.83, "step": 6205, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.510672 }, { "epoch": 0.2660554389272096, "grad_norm": 1.5361075401306152, "learning_rate": 9.930300055893549e-05, "loss": 0.21763620376586915, "memory(GiB)": 57.83, "step": 6210, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.51071 }, { "epoch": 0.2662696542564586, "grad_norm": 1.9134093523025513, "learning_rate": 9.930188034496655e-05, "loss": 0.42278389930725097, "memory(GiB)": 57.83, "step": 6215, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.510785 }, { "epoch": 0.2664838695857076, "grad_norm": 2.883085012435913, "learning_rate": 9.93007592378478e-05, "loss": 0.6707213878631592, "memory(GiB)": 57.83, "step": 6220, "token_acc": 0.875, "train_speed(iter/s)": 1.510694 }, { "epoch": 0.2666980849149565, "grad_norm": 4.380506992340088, "learning_rate": 9.929963723759956e-05, "loss": 0.32514955997467043, "memory(GiB)": 57.83, "step": 6225, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.510762 }, { "epoch": 0.2669123002442055, "grad_norm": 2.6944563388824463, "learning_rate": 9.929851434424216e-05, "loss": 0.4875223159790039, "memory(GiB)": 57.83, "step": 6230, "token_acc": 0.8797250859106529, "train_speed(iter/s)": 1.510765 }, { "epoch": 0.26712651557345446, "grad_norm": 3.106567859649658, "learning_rate": 9.92973905577959e-05, "loss": 0.5267004013061524, "memory(GiB)": 57.83, "step": 6235, "token_acc": 0.8723404255319149, "train_speed(iter/s)": 1.510713 }, { "epoch": 0.2673407309027034, "grad_norm": 0.2307433784008026, "learning_rate": 9.929626587828118e-05, "loss": 0.3483583450317383, "memory(GiB)": 57.83, "step": 6240, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.510668 }, { "epoch": 0.26755494623195236, "grad_norm": 3.830824851989746, "learning_rate": 9.929514030571834e-05, "loss": 0.4847075939178467, "memory(GiB)": 57.83, "step": 6245, "token_acc": 0.8730769230769231, "train_speed(iter/s)": 1.510762 }, { "epoch": 0.26776916156120134, "grad_norm": 0.7896789908409119, "learning_rate": 9.92940138401278e-05, "loss": 0.35685760974884034, "memory(GiB)": 57.83, "step": 6250, "token_acc": 0.9301587301587302, "train_speed(iter/s)": 1.510818 }, { "epoch": 0.26798337689045026, "grad_norm": 2.6878812313079834, "learning_rate": 9.929288648152997e-05, "loss": 0.5847362995147705, "memory(GiB)": 57.83, "step": 6255, "token_acc": 0.8881789137380192, "train_speed(iter/s)": 1.51081 }, { "epoch": 0.26819759221969924, "grad_norm": 1.2500518560409546, "learning_rate": 9.929175822994526e-05, "loss": 0.40888237953186035, "memory(GiB)": 57.83, "step": 6260, "token_acc": 0.8892215568862275, "train_speed(iter/s)": 1.51079 }, { "epoch": 0.2684118075489482, "grad_norm": 1.586540937423706, "learning_rate": 9.929062908539408e-05, "loss": 0.3394345760345459, "memory(GiB)": 57.83, "step": 6265, "token_acc": 0.9301587301587302, "train_speed(iter/s)": 1.510814 }, { "epoch": 0.26862602287819715, "grad_norm": 0.17603528499603271, "learning_rate": 9.928949904789695e-05, "loss": 0.31188762187957764, "memory(GiB)": 57.83, "step": 6270, "token_acc": 0.9296875, "train_speed(iter/s)": 1.510825 }, { "epoch": 0.2688402382074461, "grad_norm": 8.228421211242676, "learning_rate": 9.928836811747429e-05, "loss": 0.4547892093658447, "memory(GiB)": 57.83, "step": 6275, "token_acc": 0.8892857142857142, "train_speed(iter/s)": 1.511019 }, { "epoch": 0.2690544535366951, "grad_norm": 0.07260742783546448, "learning_rate": 9.928723629414662e-05, "loss": 0.13691699504852295, "memory(GiB)": 57.83, "step": 6280, "token_acc": 0.9588014981273408, "train_speed(iter/s)": 1.511016 }, { "epoch": 0.269268668865944, "grad_norm": 4.772095680236816, "learning_rate": 9.928610357793441e-05, "loss": 0.6448341846466065, "memory(GiB)": 57.83, "step": 6285, "token_acc": 0.8759398496240601, "train_speed(iter/s)": 1.511196 }, { "epoch": 0.269482884195193, "grad_norm": 1.185380458831787, "learning_rate": 9.928496996885821e-05, "loss": 0.29379274845123293, "memory(GiB)": 57.83, "step": 6290, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.511241 }, { "epoch": 0.269697099524442, "grad_norm": 6.321060657501221, "learning_rate": 9.928383546693854e-05, "loss": 0.3794532775878906, "memory(GiB)": 57.83, "step": 6295, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.511215 }, { "epoch": 0.2699113148536909, "grad_norm": 2.030837059020996, "learning_rate": 9.928270007219598e-05, "loss": 0.4416524410247803, "memory(GiB)": 57.83, "step": 6300, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.511289 }, { "epoch": 0.2701255301829399, "grad_norm": 2.938039541244507, "learning_rate": 9.928156378465106e-05, "loss": 0.43562803268432615, "memory(GiB)": 57.83, "step": 6305, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.511539 }, { "epoch": 0.27033974551218887, "grad_norm": 2.070509910583496, "learning_rate": 9.928042660432437e-05, "loss": 0.22926650047302247, "memory(GiB)": 57.83, "step": 6310, "token_acc": 0.943217665615142, "train_speed(iter/s)": 1.511577 }, { "epoch": 0.2705539608414378, "grad_norm": 1.3242130279541016, "learning_rate": 9.927928853123654e-05, "loss": 0.34583144187927245, "memory(GiB)": 57.83, "step": 6315, "token_acc": 0.9365558912386707, "train_speed(iter/s)": 1.511685 }, { "epoch": 0.27076817617068677, "grad_norm": 2.14395809173584, "learning_rate": 9.927814956540818e-05, "loss": 0.3411599397659302, "memory(GiB)": 57.83, "step": 6320, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.511917 }, { "epoch": 0.27098239149993575, "grad_norm": 0.6943260431289673, "learning_rate": 9.927700970685989e-05, "loss": 0.5581017017364502, "memory(GiB)": 57.83, "step": 6325, "token_acc": 0.8691860465116279, "train_speed(iter/s)": 1.511982 }, { "epoch": 0.27119660682918467, "grad_norm": 2.602654218673706, "learning_rate": 9.927586895561237e-05, "loss": 0.3490911960601807, "memory(GiB)": 57.83, "step": 6330, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.512171 }, { "epoch": 0.27141082215843365, "grad_norm": 4.910849571228027, "learning_rate": 9.927472731168623e-05, "loss": 0.25747361183166506, "memory(GiB)": 57.83, "step": 6335, "token_acc": 0.9498069498069498, "train_speed(iter/s)": 1.512208 }, { "epoch": 0.27162503748768263, "grad_norm": 2.9110817909240723, "learning_rate": 9.92735847751022e-05, "loss": 0.6300963401794434, "memory(GiB)": 57.83, "step": 6340, "token_acc": 0.861198738170347, "train_speed(iter/s)": 1.512111 }, { "epoch": 0.27183925281693155, "grad_norm": 4.162040710449219, "learning_rate": 9.927244134588095e-05, "loss": 0.7952901363372803, "memory(GiB)": 57.83, "step": 6345, "token_acc": 0.8501742160278746, "train_speed(iter/s)": 1.512383 }, { "epoch": 0.27205346814618053, "grad_norm": 3.174696445465088, "learning_rate": 9.927129702404321e-05, "loss": 0.5402266502380371, "memory(GiB)": 57.83, "step": 6350, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.512338 }, { "epoch": 0.2722676834754295, "grad_norm": 2.1442363262176514, "learning_rate": 9.92701518096097e-05, "loss": 0.4409471035003662, "memory(GiB)": 57.83, "step": 6355, "token_acc": 0.8938356164383562, "train_speed(iter/s)": 1.512288 }, { "epoch": 0.27248189880467844, "grad_norm": 1.0602281093597412, "learning_rate": 9.926900570260118e-05, "loss": 0.14982073307037352, "memory(GiB)": 57.83, "step": 6360, "token_acc": 0.9550561797752809, "train_speed(iter/s)": 1.512277 }, { "epoch": 0.2726961141339274, "grad_norm": 0.8592976331710815, "learning_rate": 9.926785870303839e-05, "loss": 0.5739680767059326, "memory(GiB)": 57.83, "step": 6365, "token_acc": 0.8682170542635659, "train_speed(iter/s)": 1.512276 }, { "epoch": 0.2729103294631764, "grad_norm": 7.0117011070251465, "learning_rate": 9.926671081094214e-05, "loss": 0.34189066886901853, "memory(GiB)": 57.83, "step": 6370, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.512264 }, { "epoch": 0.27312454479242537, "grad_norm": 3.1320619583129883, "learning_rate": 9.92655620263332e-05, "loss": 0.33453075885772704, "memory(GiB)": 57.83, "step": 6375, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.512203 }, { "epoch": 0.2733387601216743, "grad_norm": 3.2054800987243652, "learning_rate": 9.926441234923239e-05, "loss": 0.5331183433532715, "memory(GiB)": 57.83, "step": 6380, "token_acc": 0.8962962962962963, "train_speed(iter/s)": 1.512156 }, { "epoch": 0.2735529754509233, "grad_norm": 2.2337279319763184, "learning_rate": 9.926326177966052e-05, "loss": 0.3796213388442993, "memory(GiB)": 57.83, "step": 6385, "token_acc": 0.9348659003831418, "train_speed(iter/s)": 1.512191 }, { "epoch": 0.27376719078017225, "grad_norm": 2.0649187564849854, "learning_rate": 9.926211031763846e-05, "loss": 0.24826536178588868, "memory(GiB)": 57.83, "step": 6390, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.512172 }, { "epoch": 0.2739814061094212, "grad_norm": 4.460148334503174, "learning_rate": 9.926095796318706e-05, "loss": 0.5327774524688721, "memory(GiB)": 57.83, "step": 6395, "token_acc": 0.8683274021352313, "train_speed(iter/s)": 1.512427 }, { "epoch": 0.27419562143867016, "grad_norm": 4.966132640838623, "learning_rate": 9.92598047163272e-05, "loss": 0.29757394790649416, "memory(GiB)": 57.83, "step": 6400, "token_acc": 0.9296875, "train_speed(iter/s)": 1.512397 }, { "epoch": 0.27440983676791914, "grad_norm": 2.0085175037384033, "learning_rate": 9.925865057707977e-05, "loss": 0.711307144165039, "memory(GiB)": 57.83, "step": 6405, "token_acc": 0.8515901060070671, "train_speed(iter/s)": 1.512531 }, { "epoch": 0.27462405209716806, "grad_norm": 4.277284622192383, "learning_rate": 9.925749554546568e-05, "loss": 0.6434778213500977, "memory(GiB)": 57.83, "step": 6410, "token_acc": 0.8485804416403786, "train_speed(iter/s)": 1.512558 }, { "epoch": 0.27483826742641704, "grad_norm": 2.425518751144409, "learning_rate": 9.925633962150584e-05, "loss": 0.2835421562194824, "memory(GiB)": 57.83, "step": 6415, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.512474 }, { "epoch": 0.275052482755666, "grad_norm": 2.1404807567596436, "learning_rate": 9.925518280522121e-05, "loss": 0.2600611686706543, "memory(GiB)": 57.83, "step": 6420, "token_acc": 0.9439655172413793, "train_speed(iter/s)": 1.512418 }, { "epoch": 0.27526669808491494, "grad_norm": 2.3433375358581543, "learning_rate": 9.925402509663273e-05, "loss": 0.21788403987884522, "memory(GiB)": 57.83, "step": 6425, "token_acc": 0.936026936026936, "train_speed(iter/s)": 1.512482 }, { "epoch": 0.2754809134141639, "grad_norm": 4.9804558753967285, "learning_rate": 9.92528664957614e-05, "loss": 0.6327625274658203, "memory(GiB)": 57.83, "step": 6430, "token_acc": 0.8775510204081632, "train_speed(iter/s)": 1.512379 }, { "epoch": 0.2756951287434129, "grad_norm": 1.603032112121582, "learning_rate": 9.925170700262817e-05, "loss": 0.3081965923309326, "memory(GiB)": 57.83, "step": 6435, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.512395 }, { "epoch": 0.2759093440726618, "grad_norm": 2.740758180618286, "learning_rate": 9.925054661725406e-05, "loss": 0.38141565322875975, "memory(GiB)": 57.83, "step": 6440, "token_acc": 0.9122137404580153, "train_speed(iter/s)": 1.512389 }, { "epoch": 0.2761235594019108, "grad_norm": 3.2003042697906494, "learning_rate": 9.924938533966012e-05, "loss": 0.2652620553970337, "memory(GiB)": 57.83, "step": 6445, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.512399 }, { "epoch": 0.2763377747311598, "grad_norm": 0.058249182999134064, "learning_rate": 9.924822316986735e-05, "loss": 0.3652892351150513, "memory(GiB)": 57.83, "step": 6450, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.512448 }, { "epoch": 0.2765519900604087, "grad_norm": 2.380427598953247, "learning_rate": 9.924706010789683e-05, "loss": 0.36388554573059084, "memory(GiB)": 57.83, "step": 6455, "token_acc": 0.9098837209302325, "train_speed(iter/s)": 1.512474 }, { "epoch": 0.2767662053896577, "grad_norm": 8.194302558898926, "learning_rate": 9.924589615376962e-05, "loss": 0.34850521087646485, "memory(GiB)": 57.83, "step": 6460, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.512479 }, { "epoch": 0.27698042071890666, "grad_norm": 0.3888047933578491, "learning_rate": 9.92447313075068e-05, "loss": 0.486391019821167, "memory(GiB)": 57.83, "step": 6465, "token_acc": 0.888030888030888, "train_speed(iter/s)": 1.512472 }, { "epoch": 0.2771946360481556, "grad_norm": 4.803922176361084, "learning_rate": 9.924356556912946e-05, "loss": 0.6917881011962891, "memory(GiB)": 57.83, "step": 6470, "token_acc": 0.8661710037174721, "train_speed(iter/s)": 1.512746 }, { "epoch": 0.27740885137740456, "grad_norm": 0.0627179816365242, "learning_rate": 9.924239893865874e-05, "loss": 0.3671936750411987, "memory(GiB)": 57.83, "step": 6475, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.51273 }, { "epoch": 0.27762306670665354, "grad_norm": 7.361741542816162, "learning_rate": 9.924123141611578e-05, "loss": 0.4177116394042969, "memory(GiB)": 57.83, "step": 6480, "token_acc": 0.9, "train_speed(iter/s)": 1.512884 }, { "epoch": 0.27783728203590247, "grad_norm": 2.2144763469696045, "learning_rate": 9.924006300152173e-05, "loss": 0.4577498435974121, "memory(GiB)": 57.83, "step": 6485, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.512934 }, { "epoch": 0.27805149736515145, "grad_norm": 1.6593632698059082, "learning_rate": 9.923889369489774e-05, "loss": 0.3942890167236328, "memory(GiB)": 57.83, "step": 6490, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.512977 }, { "epoch": 0.2782657126944004, "grad_norm": 1.4027423858642578, "learning_rate": 9.9237723496265e-05, "loss": 0.24048173427581787, "memory(GiB)": 57.83, "step": 6495, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.512993 }, { "epoch": 0.27847992802364935, "grad_norm": 1.80897057056427, "learning_rate": 9.923655240564472e-05, "loss": 0.27468180656433105, "memory(GiB)": 57.83, "step": 6500, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.512977 }, { "epoch": 0.27847992802364935, "eval_loss": 2.5823864936828613, "eval_runtime": 10.4329, "eval_samples_per_second": 9.585, "eval_steps_per_second": 9.585, "eval_token_acc": 0.44571428571428573, "step": 6500 }, { "epoch": 0.2786941433528983, "grad_norm": 2.5597188472747803, "learning_rate": 9.92353804230581e-05, "loss": 0.23364367485046386, "memory(GiB)": 57.83, "step": 6505, "token_acc": 0.599802371541502, "train_speed(iter/s)": 1.509 }, { "epoch": 0.2789083586821473, "grad_norm": 3.2958898544311523, "learning_rate": 9.923420754852634e-05, "loss": 0.44205989837646487, "memory(GiB)": 57.83, "step": 6510, "token_acc": 0.8881118881118881, "train_speed(iter/s)": 1.509165 }, { "epoch": 0.27912257401139623, "grad_norm": 3.260833263397217, "learning_rate": 9.923303378207077e-05, "loss": 0.3547123908996582, "memory(GiB)": 57.83, "step": 6515, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.509282 }, { "epoch": 0.2793367893406452, "grad_norm": 2.1002116203308105, "learning_rate": 9.92318591237126e-05, "loss": 0.5263885974884033, "memory(GiB)": 57.83, "step": 6520, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.509232 }, { "epoch": 0.2795510046698942, "grad_norm": 9.666547775268555, "learning_rate": 9.923068357347312e-05, "loss": 0.3767200469970703, "memory(GiB)": 57.83, "step": 6525, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.509369 }, { "epoch": 0.2797652199991431, "grad_norm": 8.11375617980957, "learning_rate": 9.92295071313736e-05, "loss": 0.4401705741882324, "memory(GiB)": 57.83, "step": 6530, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.509484 }, { "epoch": 0.2799794353283921, "grad_norm": 1.5643974542617798, "learning_rate": 9.922832979743542e-05, "loss": 0.32225751876831055, "memory(GiB)": 57.83, "step": 6535, "token_acc": 0.9129129129129129, "train_speed(iter/s)": 1.509641 }, { "epoch": 0.28019365065764107, "grad_norm": 1.9318766593933105, "learning_rate": 9.922715157167984e-05, "loss": 0.2171966552734375, "memory(GiB)": 57.83, "step": 6540, "token_acc": 0.9427480916030534, "train_speed(iter/s)": 1.50983 }, { "epoch": 0.28040786598689005, "grad_norm": 3.45180082321167, "learning_rate": 9.922597245412822e-05, "loss": 0.49033565521240235, "memory(GiB)": 57.83, "step": 6545, "token_acc": 0.9013157894736842, "train_speed(iter/s)": 1.509808 }, { "epoch": 0.28062208131613897, "grad_norm": 0.616552472114563, "learning_rate": 9.922479244480194e-05, "loss": 0.4805943489074707, "memory(GiB)": 57.83, "step": 6550, "token_acc": 0.8892508143322475, "train_speed(iter/s)": 1.509935 }, { "epoch": 0.28083629664538795, "grad_norm": 2.7647953033447266, "learning_rate": 9.922361154372237e-05, "loss": 0.4357913017272949, "memory(GiB)": 57.83, "step": 6555, "token_acc": 0.9007633587786259, "train_speed(iter/s)": 1.50997 }, { "epoch": 0.28105051197463693, "grad_norm": 3.669450283050537, "learning_rate": 9.922242975091092e-05, "loss": 0.2386949062347412, "memory(GiB)": 57.83, "step": 6560, "token_acc": 0.9401197604790419, "train_speed(iter/s)": 1.509952 }, { "epoch": 0.28126472730388585, "grad_norm": 3.3747401237487793, "learning_rate": 9.922124706638896e-05, "loss": 0.343776273727417, "memory(GiB)": 57.83, "step": 6565, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.509892 }, { "epoch": 0.28147894263313483, "grad_norm": 2.7108800411224365, "learning_rate": 9.922006349017793e-05, "loss": 0.5734564781188964, "memory(GiB)": 57.83, "step": 6570, "token_acc": 0.8728813559322034, "train_speed(iter/s)": 1.509839 }, { "epoch": 0.2816931579623838, "grad_norm": 4.904617786407471, "learning_rate": 9.92188790222993e-05, "loss": 0.42230963706970215, "memory(GiB)": 57.83, "step": 6575, "token_acc": 0.8990228013029316, "train_speed(iter/s)": 1.50994 }, { "epoch": 0.28190737329163273, "grad_norm": 2.8900465965270996, "learning_rate": 9.921769366277449e-05, "loss": 0.4845584869384766, "memory(GiB)": 57.83, "step": 6580, "token_acc": 0.8843283582089553, "train_speed(iter/s)": 1.51003 }, { "epoch": 0.2821215886208817, "grad_norm": 3.6472396850585938, "learning_rate": 9.9216507411625e-05, "loss": 0.25217835903167723, "memory(GiB)": 57.83, "step": 6585, "token_acc": 0.9490196078431372, "train_speed(iter/s)": 1.510149 }, { "epoch": 0.2823358039501307, "grad_norm": 3.904006242752075, "learning_rate": 9.92153202688723e-05, "loss": 0.15459029674530028, "memory(GiB)": 57.83, "step": 6590, "token_acc": 0.9739776951672863, "train_speed(iter/s)": 1.510188 }, { "epoch": 0.2825500192793796, "grad_norm": 7.50876522064209, "learning_rate": 9.921413223453791e-05, "loss": 0.3593508243560791, "memory(GiB)": 57.83, "step": 6595, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.510095 }, { "epoch": 0.2827642346086286, "grad_norm": 0.4729389250278473, "learning_rate": 9.921294330864334e-05, "loss": 0.31595792770385744, "memory(GiB)": 57.83, "step": 6600, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.510229 }, { "epoch": 0.2829784499378776, "grad_norm": 0.17370688915252686, "learning_rate": 9.921175349121015e-05, "loss": 0.3152081251144409, "memory(GiB)": 57.83, "step": 6605, "token_acc": 0.9551282051282052, "train_speed(iter/s)": 1.510297 }, { "epoch": 0.2831926652671265, "grad_norm": 2.7992656230926514, "learning_rate": 9.921056278225986e-05, "loss": 0.46057796478271484, "memory(GiB)": 57.83, "step": 6610, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.510469 }, { "epoch": 0.2834068805963755, "grad_norm": 3.308286428451538, "learning_rate": 9.920937118181408e-05, "loss": 0.4269710540771484, "memory(GiB)": 57.83, "step": 6615, "token_acc": 0.900990099009901, "train_speed(iter/s)": 1.510544 }, { "epoch": 0.28362109592562446, "grad_norm": 2.072024345397949, "learning_rate": 9.920817868989439e-05, "loss": 0.3146365642547607, "memory(GiB)": 57.83, "step": 6620, "token_acc": 0.9161073825503355, "train_speed(iter/s)": 1.510524 }, { "epoch": 0.2838353112548734, "grad_norm": 1.5315922498703003, "learning_rate": 9.920698530652235e-05, "loss": 0.4002521991729736, "memory(GiB)": 57.83, "step": 6625, "token_acc": 0.9184397163120568, "train_speed(iter/s)": 1.510552 }, { "epoch": 0.28404952658412236, "grad_norm": 5.095638275146484, "learning_rate": 9.920579103171963e-05, "loss": 0.45280728340148924, "memory(GiB)": 57.83, "step": 6630, "token_acc": 0.9053497942386831, "train_speed(iter/s)": 1.510563 }, { "epoch": 0.28426374191337134, "grad_norm": 1.6966438293457031, "learning_rate": 9.920459586550785e-05, "loss": 0.508296537399292, "memory(GiB)": 57.83, "step": 6635, "token_acc": 0.8810975609756098, "train_speed(iter/s)": 1.510545 }, { "epoch": 0.28447795724262026, "grad_norm": 1.189117431640625, "learning_rate": 9.920339980790864e-05, "loss": 0.40941715240478516, "memory(GiB)": 57.83, "step": 6640, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.51058 }, { "epoch": 0.28469217257186924, "grad_norm": 2.3083910942077637, "learning_rate": 9.92022028589437e-05, "loss": 0.44415531158447263, "memory(GiB)": 57.83, "step": 6645, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.510579 }, { "epoch": 0.2849063879011182, "grad_norm": 1.6124197244644165, "learning_rate": 9.920100501863472e-05, "loss": 0.3650284051895142, "memory(GiB)": 57.83, "step": 6650, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.510626 }, { "epoch": 0.28512060323036714, "grad_norm": 8.5571928024292, "learning_rate": 9.919980628700335e-05, "loss": 0.5643478393554687, "memory(GiB)": 57.83, "step": 6655, "token_acc": 0.8436578171091446, "train_speed(iter/s)": 1.510659 }, { "epoch": 0.2853348185596161, "grad_norm": 3.511624813079834, "learning_rate": 9.919860666407135e-05, "loss": 0.6758069038391114, "memory(GiB)": 57.83, "step": 6660, "token_acc": 0.8715596330275229, "train_speed(iter/s)": 1.510655 }, { "epoch": 0.2855490338888651, "grad_norm": 3.285041093826294, "learning_rate": 9.919740614986043e-05, "loss": 0.3759666919708252, "memory(GiB)": 57.83, "step": 6665, "token_acc": 0.919732441471572, "train_speed(iter/s)": 1.510674 }, { "epoch": 0.285763249218114, "grad_norm": 2.7371280193328857, "learning_rate": 9.919620474439236e-05, "loss": 0.4539473533630371, "memory(GiB)": 57.83, "step": 6670, "token_acc": 0.8918032786885246, "train_speed(iter/s)": 1.510676 }, { "epoch": 0.285977464547363, "grad_norm": 2.984240770339966, "learning_rate": 9.919500244768886e-05, "loss": 0.5049794673919678, "memory(GiB)": 57.83, "step": 6675, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.510665 }, { "epoch": 0.286191679876612, "grad_norm": 2.604233503341675, "learning_rate": 9.919379925977177e-05, "loss": 0.4652516841888428, "memory(GiB)": 57.83, "step": 6680, "token_acc": 0.886986301369863, "train_speed(iter/s)": 1.510782 }, { "epoch": 0.2864058952058609, "grad_norm": 2.2249717712402344, "learning_rate": 9.919259518066285e-05, "loss": 0.4353911876678467, "memory(GiB)": 57.83, "step": 6685, "token_acc": 0.8873720136518771, "train_speed(iter/s)": 1.510773 }, { "epoch": 0.2866201105351099, "grad_norm": 4.0739216804504395, "learning_rate": 9.919139021038392e-05, "loss": 0.4287562847137451, "memory(GiB)": 57.83, "step": 6690, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.510803 }, { "epoch": 0.28683432586435886, "grad_norm": 1.6508581638336182, "learning_rate": 9.919018434895681e-05, "loss": 0.4799161911010742, "memory(GiB)": 57.83, "step": 6695, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.510899 }, { "epoch": 0.2870485411936078, "grad_norm": 3.239684581756592, "learning_rate": 9.918897759640338e-05, "loss": 0.557736873626709, "memory(GiB)": 57.83, "step": 6700, "token_acc": 0.889261744966443, "train_speed(iter/s)": 1.510913 }, { "epoch": 0.28726275652285677, "grad_norm": 0.48320579528808594, "learning_rate": 9.918776995274547e-05, "loss": 0.3017580986022949, "memory(GiB)": 57.83, "step": 6705, "token_acc": 0.9290123456790124, "train_speed(iter/s)": 1.510928 }, { "epoch": 0.28747697185210574, "grad_norm": 3.4178154468536377, "learning_rate": 9.918656141800496e-05, "loss": 0.3725148916244507, "memory(GiB)": 57.83, "step": 6710, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.511012 }, { "epoch": 0.2876911871813547, "grad_norm": 2.7487685680389404, "learning_rate": 9.918535199220376e-05, "loss": 0.46137566566467286, "memory(GiB)": 57.83, "step": 6715, "token_acc": 0.8955223880597015, "train_speed(iter/s)": 1.51118 }, { "epoch": 0.28790540251060365, "grad_norm": 0.9815278649330139, "learning_rate": 9.918414167536376e-05, "loss": 0.25387632846832275, "memory(GiB)": 57.83, "step": 6720, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.511176 }, { "epoch": 0.2881196178398526, "grad_norm": 3.529553174972534, "learning_rate": 9.918293046750689e-05, "loss": 0.19895347356796264, "memory(GiB)": 57.83, "step": 6725, "token_acc": 0.9587301587301588, "train_speed(iter/s)": 1.511334 }, { "epoch": 0.2883338331691016, "grad_norm": 5.436715602874756, "learning_rate": 9.918171836865511e-05, "loss": 0.7510737419128418, "memory(GiB)": 57.83, "step": 6730, "token_acc": 0.8239202657807309, "train_speed(iter/s)": 1.511271 }, { "epoch": 0.28854804849835053, "grad_norm": 3.512310028076172, "learning_rate": 9.918050537883037e-05, "loss": 0.32025809288024903, "memory(GiB)": 57.83, "step": 6735, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.511346 }, { "epoch": 0.2887622638275995, "grad_norm": 3.33569073677063, "learning_rate": 9.917929149805462e-05, "loss": 0.30508835315704347, "memory(GiB)": 57.83, "step": 6740, "token_acc": 0.9314285714285714, "train_speed(iter/s)": 1.511338 }, { "epoch": 0.2889764791568485, "grad_norm": 1.867612361907959, "learning_rate": 9.917807672634989e-05, "loss": 0.5618594646453857, "memory(GiB)": 57.83, "step": 6745, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.51151 }, { "epoch": 0.2891906944860974, "grad_norm": 3.2929255962371826, "learning_rate": 9.917686106373816e-05, "loss": 0.2535130977630615, "memory(GiB)": 57.83, "step": 6750, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.511519 }, { "epoch": 0.2894049098153464, "grad_norm": 4.31369161605835, "learning_rate": 9.917564451024147e-05, "loss": 0.3781198740005493, "memory(GiB)": 57.83, "step": 6755, "token_acc": 0.9240506329113924, "train_speed(iter/s)": 1.511568 }, { "epoch": 0.28961912514459537, "grad_norm": 4.760101795196533, "learning_rate": 9.917442706588183e-05, "loss": 0.27748737335205076, "memory(GiB)": 57.83, "step": 6760, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.511599 }, { "epoch": 0.2898333404738443, "grad_norm": 2.5961623191833496, "learning_rate": 9.917320873068132e-05, "loss": 0.2841786861419678, "memory(GiB)": 57.83, "step": 6765, "token_acc": 0.9292035398230089, "train_speed(iter/s)": 1.511546 }, { "epoch": 0.29004755580309327, "grad_norm": 2.423522472381592, "learning_rate": 9.9171989504662e-05, "loss": 0.43582978248596194, "memory(GiB)": 57.83, "step": 6770, "token_acc": 0.9093851132686084, "train_speed(iter/s)": 1.51169 }, { "epoch": 0.29026177113234225, "grad_norm": 2.969874858856201, "learning_rate": 9.917076938784597e-05, "loss": 0.31996402740478513, "memory(GiB)": 57.83, "step": 6775, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.511681 }, { "epoch": 0.2904759864615912, "grad_norm": 2.9201338291168213, "learning_rate": 9.916954838025533e-05, "loss": 0.53292875289917, "memory(GiB)": 57.83, "step": 6780, "token_acc": 0.8827838827838828, "train_speed(iter/s)": 1.511883 }, { "epoch": 0.29069020179084015, "grad_norm": 2.5437862873077393, "learning_rate": 9.91683264819122e-05, "loss": 0.47618885040283204, "memory(GiB)": 57.83, "step": 6785, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.51199 }, { "epoch": 0.29090441712008913, "grad_norm": 2.312044620513916, "learning_rate": 9.91671036928387e-05, "loss": 0.18593474626541137, "memory(GiB)": 57.83, "step": 6790, "token_acc": 0.965034965034965, "train_speed(iter/s)": 1.512089 }, { "epoch": 0.29111863244933806, "grad_norm": 1.845430850982666, "learning_rate": 9.9165880013057e-05, "loss": 0.26776275634765623, "memory(GiB)": 57.83, "step": 6795, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.512067 }, { "epoch": 0.29133284777858703, "grad_norm": 1.6939797401428223, "learning_rate": 9.916465544258926e-05, "loss": 0.2662634372711182, "memory(GiB)": 57.83, "step": 6800, "token_acc": 0.9490196078431372, "train_speed(iter/s)": 1.512121 }, { "epoch": 0.291547063107836, "grad_norm": 1.1998353004455566, "learning_rate": 9.916342998145766e-05, "loss": 0.19747133255004884, "memory(GiB)": 57.83, "step": 6805, "token_acc": 0.9413793103448276, "train_speed(iter/s)": 1.512216 }, { "epoch": 0.29176127843708494, "grad_norm": 2.131963014602661, "learning_rate": 9.916220362968443e-05, "loss": 0.41002440452575684, "memory(GiB)": 57.83, "step": 6810, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.512256 }, { "epoch": 0.2919754937663339, "grad_norm": 2.257479429244995, "learning_rate": 9.916097638729174e-05, "loss": 0.25909056663513186, "memory(GiB)": 57.83, "step": 6815, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.512202 }, { "epoch": 0.2921897090955829, "grad_norm": 7.505083084106445, "learning_rate": 9.915974825430187e-05, "loss": 0.43703441619873046, "memory(GiB)": 57.83, "step": 6820, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.512205 }, { "epoch": 0.2924039244248318, "grad_norm": 2.9049503803253174, "learning_rate": 9.915851923073702e-05, "loss": 0.23073127269744872, "memory(GiB)": 57.83, "step": 6825, "token_acc": 0.956140350877193, "train_speed(iter/s)": 1.51215 }, { "epoch": 0.2926181397540808, "grad_norm": 2.8844106197357178, "learning_rate": 9.915728931661949e-05, "loss": 0.7423871040344239, "memory(GiB)": 57.83, "step": 6830, "token_acc": 0.7993079584775087, "train_speed(iter/s)": 1.51224 }, { "epoch": 0.2928323550833298, "grad_norm": 1.5598363876342773, "learning_rate": 9.915605851197156e-05, "loss": 0.4712527275085449, "memory(GiB)": 57.83, "step": 6835, "token_acc": 0.8950819672131147, "train_speed(iter/s)": 1.512329 }, { "epoch": 0.2930465704125787, "grad_norm": 1.703057885169983, "learning_rate": 9.91548268168155e-05, "loss": 0.28084487915039064, "memory(GiB)": 57.83, "step": 6840, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.512308 }, { "epoch": 0.2932607857418277, "grad_norm": 4.292952537536621, "learning_rate": 9.915359423117366e-05, "loss": 0.2443718433380127, "memory(GiB)": 57.83, "step": 6845, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.51242 }, { "epoch": 0.29347500107107666, "grad_norm": 2.707275152206421, "learning_rate": 9.915236075506833e-05, "loss": 0.3532017946243286, "memory(GiB)": 57.83, "step": 6850, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.512508 }, { "epoch": 0.2936892164003256, "grad_norm": 0.422793984413147, "learning_rate": 9.915112638852188e-05, "loss": 0.4425339698791504, "memory(GiB)": 57.83, "step": 6855, "token_acc": 0.9243027888446215, "train_speed(iter/s)": 1.512556 }, { "epoch": 0.29390343172957456, "grad_norm": 2.2368855476379395, "learning_rate": 9.914989113155668e-05, "loss": 0.2529210090637207, "memory(GiB)": 57.83, "step": 6860, "token_acc": 0.9432624113475178, "train_speed(iter/s)": 1.512556 }, { "epoch": 0.29411764705882354, "grad_norm": 1.9608335494995117, "learning_rate": 9.91486549841951e-05, "loss": 0.3926483869552612, "memory(GiB)": 57.83, "step": 6865, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.512569 }, { "epoch": 0.2943318623880725, "grad_norm": 4.714572906494141, "learning_rate": 9.914741794645952e-05, "loss": 0.42986578941345216, "memory(GiB)": 57.83, "step": 6870, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.512636 }, { "epoch": 0.29454607771732144, "grad_norm": 0.22730474174022675, "learning_rate": 9.914618001837235e-05, "loss": 0.5279127597808838, "memory(GiB)": 57.83, "step": 6875, "token_acc": 0.8885017421602788, "train_speed(iter/s)": 1.51257 }, { "epoch": 0.2947602930465704, "grad_norm": 3.1628425121307373, "learning_rate": 9.914494119995604e-05, "loss": 0.41266465187072754, "memory(GiB)": 57.83, "step": 6880, "token_acc": 0.8885350318471338, "train_speed(iter/s)": 1.512656 }, { "epoch": 0.2949745083758194, "grad_norm": 6.228361129760742, "learning_rate": 9.914370149123302e-05, "loss": 0.3760586500167847, "memory(GiB)": 57.83, "step": 6885, "token_acc": 0.9114391143911439, "train_speed(iter/s)": 1.51271 }, { "epoch": 0.2951887237050683, "grad_norm": 4.047192573547363, "learning_rate": 9.914246089222575e-05, "loss": 0.24162516593933106, "memory(GiB)": 57.83, "step": 6890, "token_acc": 0.9554896142433235, "train_speed(iter/s)": 1.512843 }, { "epoch": 0.2954029390343173, "grad_norm": 3.0303304195404053, "learning_rate": 9.914121940295669e-05, "loss": 0.4208077430725098, "memory(GiB)": 57.83, "step": 6895, "token_acc": 0.9122257053291536, "train_speed(iter/s)": 1.512982 }, { "epoch": 0.2956171543635663, "grad_norm": 1.7910605669021606, "learning_rate": 9.913997702344834e-05, "loss": 0.46535773277282716, "memory(GiB)": 57.83, "step": 6900, "token_acc": 0.888283378746594, "train_speed(iter/s)": 1.5131 }, { "epoch": 0.2958313696928152, "grad_norm": 4.656360149383545, "learning_rate": 9.913873375372321e-05, "loss": 0.65853910446167, "memory(GiB)": 57.83, "step": 6905, "token_acc": 0.8867924528301887, "train_speed(iter/s)": 1.513087 }, { "epoch": 0.2960455850220642, "grad_norm": 1.7705928087234497, "learning_rate": 9.913748959380382e-05, "loss": 0.22304284572601318, "memory(GiB)": 57.83, "step": 6910, "token_acc": 0.9515570934256056, "train_speed(iter/s)": 1.513227 }, { "epoch": 0.29625980035131316, "grad_norm": 4.257490634918213, "learning_rate": 9.913624454371273e-05, "loss": 0.44872655868530276, "memory(GiB)": 57.83, "step": 6915, "token_acc": 0.8984126984126984, "train_speed(iter/s)": 1.513312 }, { "epoch": 0.2964740156805621, "grad_norm": 3.3854780197143555, "learning_rate": 9.913499860347246e-05, "loss": 0.46138315200805663, "memory(GiB)": 57.83, "step": 6920, "token_acc": 0.9045936395759717, "train_speed(iter/s)": 1.513393 }, { "epoch": 0.29668823100981107, "grad_norm": 3.9084274768829346, "learning_rate": 9.91337517731056e-05, "loss": 0.23434827327728272, "memory(GiB)": 57.83, "step": 6925, "token_acc": 0.9540229885057471, "train_speed(iter/s)": 1.513354 }, { "epoch": 0.29690244633906004, "grad_norm": 2.3918936252593994, "learning_rate": 9.913250405263474e-05, "loss": 0.7487889766693115, "memory(GiB)": 57.83, "step": 6930, "token_acc": 0.8745874587458746, "train_speed(iter/s)": 1.513454 }, { "epoch": 0.29711666166830897, "grad_norm": 2.6662282943725586, "learning_rate": 9.913125544208248e-05, "loss": 0.4085972785949707, "memory(GiB)": 57.83, "step": 6935, "token_acc": 0.916083916083916, "train_speed(iter/s)": 1.513533 }, { "epoch": 0.29733087699755795, "grad_norm": 2.9608867168426514, "learning_rate": 9.913000594147144e-05, "loss": 0.30474166870117186, "memory(GiB)": 57.83, "step": 6940, "token_acc": 0.9288135593220339, "train_speed(iter/s)": 1.513592 }, { "epoch": 0.2975450923268069, "grad_norm": 1.96047842502594, "learning_rate": 9.912875555082425e-05, "loss": 0.4579944610595703, "memory(GiB)": 57.83, "step": 6945, "token_acc": 0.9368770764119602, "train_speed(iter/s)": 1.513525 }, { "epoch": 0.29775930765605585, "grad_norm": 1.3485209941864014, "learning_rate": 9.912750427016356e-05, "loss": 0.3267801284790039, "memory(GiB)": 57.83, "step": 6950, "token_acc": 0.9216216216216216, "train_speed(iter/s)": 1.513519 }, { "epoch": 0.29797352298530483, "grad_norm": 1.4287772178649902, "learning_rate": 9.912625209951206e-05, "loss": 0.5012242794036865, "memory(GiB)": 57.83, "step": 6955, "token_acc": 0.9052132701421801, "train_speed(iter/s)": 1.513531 }, { "epoch": 0.2981877383145538, "grad_norm": 8.136691093444824, "learning_rate": 9.91249990388924e-05, "loss": 0.33779704570770264, "memory(GiB)": 57.83, "step": 6960, "token_acc": 0.9241379310344827, "train_speed(iter/s)": 1.513459 }, { "epoch": 0.29840195364380273, "grad_norm": 2.9507246017456055, "learning_rate": 9.912374508832732e-05, "loss": 0.41786603927612304, "memory(GiB)": 57.83, "step": 6965, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.513463 }, { "epoch": 0.2986161689730517, "grad_norm": 3.5575520992279053, "learning_rate": 9.912249024783951e-05, "loss": 0.4341097354888916, "memory(GiB)": 57.83, "step": 6970, "token_acc": 0.915129151291513, "train_speed(iter/s)": 1.513458 }, { "epoch": 0.2988303843023007, "grad_norm": 2.142886161804199, "learning_rate": 9.91212345174517e-05, "loss": 0.5189872741699219, "memory(GiB)": 57.83, "step": 6975, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.513461 }, { "epoch": 0.2990445996315496, "grad_norm": 2.6330983638763428, "learning_rate": 9.911997789718666e-05, "loss": 0.5798794746398925, "memory(GiB)": 57.83, "step": 6980, "token_acc": 0.8673139158576052, "train_speed(iter/s)": 1.513426 }, { "epoch": 0.2992588149607986, "grad_norm": 4.009677410125732, "learning_rate": 9.911872038706713e-05, "loss": 0.3306786298751831, "memory(GiB)": 57.83, "step": 6985, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.513512 }, { "epoch": 0.29947303029004757, "grad_norm": 2.3541738986968994, "learning_rate": 9.911746198711591e-05, "loss": 0.19779937267303466, "memory(GiB)": 57.83, "step": 6990, "token_acc": 0.9575289575289575, "train_speed(iter/s)": 1.513516 }, { "epoch": 0.2996872456192965, "grad_norm": 3.373408794403076, "learning_rate": 9.911620269735578e-05, "loss": 0.4963990211486816, "memory(GiB)": 57.83, "step": 6995, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.513512 }, { "epoch": 0.2999014609485455, "grad_norm": 6.596762180328369, "learning_rate": 9.911494251780957e-05, "loss": 0.39969258308410643, "memory(GiB)": 57.83, "step": 7000, "token_acc": 0.9124293785310734, "train_speed(iter/s)": 1.513533 }, { "epoch": 0.2999014609485455, "eval_loss": 2.6168062686920166, "eval_runtime": 12.2294, "eval_samples_per_second": 8.177, "eval_steps_per_second": 8.177, "eval_token_acc": 0.4351145038167939, "step": 7000 }, { "epoch": 0.30011567627779445, "grad_norm": 2.481531858444214, "learning_rate": 9.911368144850011e-05, "loss": 0.4692357063293457, "memory(GiB)": 57.83, "step": 7005, "token_acc": 0.5747747747747748, "train_speed(iter/s)": 1.509252 }, { "epoch": 0.3003298916070434, "grad_norm": 4.111051559448242, "learning_rate": 9.911241948945022e-05, "loss": 0.3280489921569824, "memory(GiB)": 57.83, "step": 7010, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.509267 }, { "epoch": 0.30054410693629235, "grad_norm": 1.2875235080718994, "learning_rate": 9.91111566406828e-05, "loss": 0.4860054016113281, "memory(GiB)": 57.83, "step": 7015, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.509267 }, { "epoch": 0.30075832226554133, "grad_norm": 4.2224225997924805, "learning_rate": 9.91098929022207e-05, "loss": 0.280168342590332, "memory(GiB)": 57.83, "step": 7020, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.509257 }, { "epoch": 0.30097253759479026, "grad_norm": 3.7367804050445557, "learning_rate": 9.910862827408682e-05, "loss": 0.420220947265625, "memory(GiB)": 57.83, "step": 7025, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.509311 }, { "epoch": 0.30118675292403924, "grad_norm": 6.199322700500488, "learning_rate": 9.910736275630408e-05, "loss": 0.23392961025238038, "memory(GiB)": 57.83, "step": 7030, "token_acc": 0.9400749063670412, "train_speed(iter/s)": 1.509367 }, { "epoch": 0.3014009682532882, "grad_norm": 3.7858235836029053, "learning_rate": 9.910609634889538e-05, "loss": 0.3799464225769043, "memory(GiB)": 57.83, "step": 7035, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.509381 }, { "epoch": 0.3016151835825372, "grad_norm": 3.575618267059326, "learning_rate": 9.91048290518837e-05, "loss": 0.5931829452514649, "memory(GiB)": 57.83, "step": 7040, "token_acc": 0.8847262247838616, "train_speed(iter/s)": 1.509484 }, { "epoch": 0.3018293989117861, "grad_norm": 2.242368698120117, "learning_rate": 9.910356086529196e-05, "loss": 0.39293580055236815, "memory(GiB)": 57.83, "step": 7045, "token_acc": 0.8931297709923665, "train_speed(iter/s)": 1.509537 }, { "epoch": 0.3020436142410351, "grad_norm": 5.320410251617432, "learning_rate": 9.910229178914317e-05, "loss": 0.5389390468597413, "memory(GiB)": 57.83, "step": 7050, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.509554 }, { "epoch": 0.3022578295702841, "grad_norm": 0.04408201575279236, "learning_rate": 9.910102182346029e-05, "loss": 0.2133228063583374, "memory(GiB)": 57.83, "step": 7055, "token_acc": 0.9533333333333334, "train_speed(iter/s)": 1.509536 }, { "epoch": 0.302472044899533, "grad_norm": 2.8032453060150146, "learning_rate": 9.909975096826634e-05, "loss": 0.30033714771270753, "memory(GiB)": 57.83, "step": 7060, "token_acc": 0.9395973154362416, "train_speed(iter/s)": 1.509598 }, { "epoch": 0.302686260228782, "grad_norm": 1.5181622505187988, "learning_rate": 9.909847922358432e-05, "loss": 0.3096545934677124, "memory(GiB)": 57.83, "step": 7065, "token_acc": 0.9294871794871795, "train_speed(iter/s)": 1.509599 }, { "epoch": 0.30290047555803096, "grad_norm": 1.9028234481811523, "learning_rate": 9.909720658943733e-05, "loss": 0.3219677686691284, "memory(GiB)": 57.83, "step": 7070, "token_acc": 0.9307692307692308, "train_speed(iter/s)": 1.50974 }, { "epoch": 0.3031146908872799, "grad_norm": 2.1404778957366943, "learning_rate": 9.909593306584837e-05, "loss": 0.3770033597946167, "memory(GiB)": 57.83, "step": 7075, "token_acc": 0.9204892966360856, "train_speed(iter/s)": 1.509829 }, { "epoch": 0.30332890621652886, "grad_norm": 2.360537052154541, "learning_rate": 9.909465865284052e-05, "loss": 0.36577568054199217, "memory(GiB)": 57.83, "step": 7080, "token_acc": 0.9265175718849841, "train_speed(iter/s)": 1.509862 }, { "epoch": 0.30354312154577784, "grad_norm": 5.3821492195129395, "learning_rate": 9.909338335043688e-05, "loss": 0.41852707862854005, "memory(GiB)": 57.83, "step": 7085, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.510056 }, { "epoch": 0.30375733687502676, "grad_norm": 1.135039210319519, "learning_rate": 9.909210715866055e-05, "loss": 0.31890828609466554, "memory(GiB)": 57.83, "step": 7090, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.510078 }, { "epoch": 0.30397155220427574, "grad_norm": 2.247605085372925, "learning_rate": 9.909083007753464e-05, "loss": 0.4199732780456543, "memory(GiB)": 57.83, "step": 7095, "token_acc": 0.9020979020979021, "train_speed(iter/s)": 1.510076 }, { "epoch": 0.3041857675335247, "grad_norm": 3.257331609725952, "learning_rate": 9.90895521070823e-05, "loss": 0.3356150150299072, "memory(GiB)": 57.83, "step": 7100, "token_acc": 0.9396551724137931, "train_speed(iter/s)": 1.510097 }, { "epoch": 0.30439998286277364, "grad_norm": 1.700039029121399, "learning_rate": 9.908827324732667e-05, "loss": 0.2570678949356079, "memory(GiB)": 57.83, "step": 7105, "token_acc": 0.9446153846153846, "train_speed(iter/s)": 1.510109 }, { "epoch": 0.3046141981920226, "grad_norm": 1.92388117313385, "learning_rate": 9.908699349829091e-05, "loss": 0.2944466590881348, "memory(GiB)": 57.83, "step": 7110, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.510128 }, { "epoch": 0.3048284135212716, "grad_norm": 3.6306886672973633, "learning_rate": 9.908571285999824e-05, "loss": 0.344616174697876, "memory(GiB)": 57.83, "step": 7115, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.51033 }, { "epoch": 0.3050426288505205, "grad_norm": 2.640101671218872, "learning_rate": 9.908443133247182e-05, "loss": 0.23415727615356446, "memory(GiB)": 57.83, "step": 7120, "token_acc": 0.9636363636363636, "train_speed(iter/s)": 1.510335 }, { "epoch": 0.3052568441797695, "grad_norm": 1.8156806230545044, "learning_rate": 9.908314891573489e-05, "loss": 0.4750521183013916, "memory(GiB)": 57.83, "step": 7125, "token_acc": 0.9, "train_speed(iter/s)": 1.51035 }, { "epoch": 0.3054710595090185, "grad_norm": 2.5328686237335205, "learning_rate": 9.908186560981066e-05, "loss": 0.42148470878601074, "memory(GiB)": 57.83, "step": 7130, "token_acc": 0.9213483146067416, "train_speed(iter/s)": 1.510329 }, { "epoch": 0.3056852748382674, "grad_norm": 2.3980135917663574, "learning_rate": 9.908058141472239e-05, "loss": 0.3529754638671875, "memory(GiB)": 57.83, "step": 7135, "token_acc": 0.9071428571428571, "train_speed(iter/s)": 1.51031 }, { "epoch": 0.3058994901675164, "grad_norm": 2.0531702041625977, "learning_rate": 9.907929633049336e-05, "loss": 0.399121618270874, "memory(GiB)": 57.83, "step": 7140, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.510471 }, { "epoch": 0.30611370549676536, "grad_norm": 7.756535053253174, "learning_rate": 9.907801035714684e-05, "loss": 0.3605761766433716, "memory(GiB)": 57.83, "step": 7145, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.510448 }, { "epoch": 0.3063279208260143, "grad_norm": 1.69660222530365, "learning_rate": 9.907672349470612e-05, "loss": 0.43435087203979494, "memory(GiB)": 57.83, "step": 7150, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.510435 }, { "epoch": 0.30654213615526327, "grad_norm": 1.186453104019165, "learning_rate": 9.90754357431945e-05, "loss": 0.2823386907577515, "memory(GiB)": 57.83, "step": 7155, "token_acc": 0.9257294429708223, "train_speed(iter/s)": 1.51039 }, { "epoch": 0.30675635148451225, "grad_norm": 0.7937273979187012, "learning_rate": 9.907414710263534e-05, "loss": 0.3387059926986694, "memory(GiB)": 57.83, "step": 7160, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.510632 }, { "epoch": 0.30697056681376117, "grad_norm": 2.9408700466156006, "learning_rate": 9.907285757305198e-05, "loss": 0.3308942079544067, "memory(GiB)": 57.83, "step": 7165, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.510635 }, { "epoch": 0.30718478214301015, "grad_norm": 1.8256409168243408, "learning_rate": 9.907156715446775e-05, "loss": 0.37187657356262205, "memory(GiB)": 57.83, "step": 7170, "token_acc": 0.9267015706806283, "train_speed(iter/s)": 1.510746 }, { "epoch": 0.30739899747225913, "grad_norm": 2.4910104274749756, "learning_rate": 9.907027584690605e-05, "loss": 0.2627954244613647, "memory(GiB)": 57.83, "step": 7175, "token_acc": 0.9343629343629344, "train_speed(iter/s)": 1.510785 }, { "epoch": 0.30761321280150805, "grad_norm": 2.7703402042388916, "learning_rate": 9.906898365039027e-05, "loss": 0.4567392826080322, "memory(GiB)": 57.83, "step": 7180, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.510854 }, { "epoch": 0.30782742813075703, "grad_norm": 2.7235138416290283, "learning_rate": 9.906769056494384e-05, "loss": 0.4511093616485596, "memory(GiB)": 57.83, "step": 7185, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.510887 }, { "epoch": 0.308041643460006, "grad_norm": 1.822920560836792, "learning_rate": 9.906639659059015e-05, "loss": 0.49080400466918944, "memory(GiB)": 57.83, "step": 7190, "token_acc": 0.8855421686746988, "train_speed(iter/s)": 1.510892 }, { "epoch": 0.30825585878925493, "grad_norm": 3.0704030990600586, "learning_rate": 9.906510172735266e-05, "loss": 0.39001994132995604, "memory(GiB)": 57.83, "step": 7195, "token_acc": 0.8955223880597015, "train_speed(iter/s)": 1.510879 }, { "epoch": 0.3084700741185039, "grad_norm": 1.502203106880188, "learning_rate": 9.906380597525484e-05, "loss": 0.38588905334472656, "memory(GiB)": 57.83, "step": 7200, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.511061 }, { "epoch": 0.3086842894477529, "grad_norm": 7.2122697830200195, "learning_rate": 9.906250933432013e-05, "loss": 0.326758599281311, "memory(GiB)": 57.83, "step": 7205, "token_acc": 0.9442379182156134, "train_speed(iter/s)": 1.511154 }, { "epoch": 0.30889850477700187, "grad_norm": 3.6200544834136963, "learning_rate": 9.906121180457204e-05, "loss": 0.32402992248535156, "memory(GiB)": 57.83, "step": 7210, "token_acc": 0.9064516129032258, "train_speed(iter/s)": 1.511188 }, { "epoch": 0.3091127201062508, "grad_norm": 2.5188660621643066, "learning_rate": 9.905991338603409e-05, "loss": 0.30301337242126464, "memory(GiB)": 57.83, "step": 7215, "token_acc": 0.9272030651340997, "train_speed(iter/s)": 1.511148 }, { "epoch": 0.3093269354354998, "grad_norm": 3.2398176193237305, "learning_rate": 9.905861407872977e-05, "loss": 0.5865688800811768, "memory(GiB)": 57.83, "step": 7220, "token_acc": 0.8838709677419355, "train_speed(iter/s)": 1.511133 }, { "epoch": 0.30954115076474875, "grad_norm": 1.9867908954620361, "learning_rate": 9.905731388268265e-05, "loss": 0.48798527717590334, "memory(GiB)": 57.83, "step": 7225, "token_acc": 0.9107142857142857, "train_speed(iter/s)": 1.511117 }, { "epoch": 0.3097553660939977, "grad_norm": 1.8459936380386353, "learning_rate": 9.905601279791626e-05, "loss": 0.2745154142379761, "memory(GiB)": 57.83, "step": 7230, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.511229 }, { "epoch": 0.30996958142324665, "grad_norm": 1.4962657690048218, "learning_rate": 9.905471082445419e-05, "loss": 0.38381710052490237, "memory(GiB)": 57.83, "step": 7235, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.511319 }, { "epoch": 0.31018379675249563, "grad_norm": 1.8625282049179077, "learning_rate": 9.905340796232e-05, "loss": 0.3996737957000732, "memory(GiB)": 57.83, "step": 7240, "token_acc": 0.9027355623100304, "train_speed(iter/s)": 1.511356 }, { "epoch": 0.31039801208174456, "grad_norm": 3.514888286590576, "learning_rate": 9.905210421153732e-05, "loss": 0.39510197639465333, "memory(GiB)": 57.83, "step": 7245, "token_acc": 0.9135446685878963, "train_speed(iter/s)": 1.51133 }, { "epoch": 0.31061222741099354, "grad_norm": 3.0047366619110107, "learning_rate": 9.905079957212975e-05, "loss": 0.29152426719665525, "memory(GiB)": 57.83, "step": 7250, "token_acc": 0.9323308270676691, "train_speed(iter/s)": 1.511435 }, { "epoch": 0.3108264427402425, "grad_norm": 2.310354709625244, "learning_rate": 9.904949404412094e-05, "loss": 0.22519874572753906, "memory(GiB)": 57.83, "step": 7255, "token_acc": 0.9372937293729373, "train_speed(iter/s)": 1.51153 }, { "epoch": 0.31104065806949144, "grad_norm": 3.532041549682617, "learning_rate": 9.904818762753454e-05, "loss": 0.4787618637084961, "memory(GiB)": 57.83, "step": 7260, "token_acc": 0.9148264984227129, "train_speed(iter/s)": 1.511536 }, { "epoch": 0.3112548733987404, "grad_norm": 0.8043096661567688, "learning_rate": 9.904688032239419e-05, "loss": 0.556431245803833, "memory(GiB)": 57.83, "step": 7265, "token_acc": 0.883495145631068, "train_speed(iter/s)": 1.511569 }, { "epoch": 0.3114690887279894, "grad_norm": 4.572041988372803, "learning_rate": 9.904557212872361e-05, "loss": 0.4755897521972656, "memory(GiB)": 57.83, "step": 7270, "token_acc": 0.9011299435028248, "train_speed(iter/s)": 1.511649 }, { "epoch": 0.3116833040572383, "grad_norm": 3.7162179946899414, "learning_rate": 9.904426304654648e-05, "loss": 0.6100828170776367, "memory(GiB)": 57.83, "step": 7275, "token_acc": 0.8745387453874539, "train_speed(iter/s)": 1.511665 }, { "epoch": 0.3118975193864873, "grad_norm": 1.6004219055175781, "learning_rate": 9.904295307588651e-05, "loss": 0.28150181770324706, "memory(GiB)": 57.83, "step": 7280, "token_acc": 0.9498069498069498, "train_speed(iter/s)": 1.511663 }, { "epoch": 0.3121117347157363, "grad_norm": 3.929323434829712, "learning_rate": 9.904164221676745e-05, "loss": 0.38715271949768065, "memory(GiB)": 57.83, "step": 7285, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.511704 }, { "epoch": 0.3123259500449852, "grad_norm": 1.162683367729187, "learning_rate": 9.904033046921303e-05, "loss": 0.4963186740875244, "memory(GiB)": 57.83, "step": 7290, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.511725 }, { "epoch": 0.3125401653742342, "grad_norm": 4.510733127593994, "learning_rate": 9.903901783324702e-05, "loss": 0.3640536546707153, "memory(GiB)": 57.83, "step": 7295, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.511888 }, { "epoch": 0.31275438070348316, "grad_norm": 3.521667718887329, "learning_rate": 9.90377043088932e-05, "loss": 0.5091758251190186, "memory(GiB)": 57.83, "step": 7300, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.512013 }, { "epoch": 0.3129685960327321, "grad_norm": 2.503905773162842, "learning_rate": 9.903638989617537e-05, "loss": 0.4140665531158447, "memory(GiB)": 57.83, "step": 7305, "token_acc": 0.9116719242902208, "train_speed(iter/s)": 1.512093 }, { "epoch": 0.31318281136198106, "grad_norm": 5.485773086547852, "learning_rate": 9.903507459511733e-05, "loss": 0.7755597114562989, "memory(GiB)": 57.83, "step": 7310, "token_acc": 0.8536585365853658, "train_speed(iter/s)": 1.512175 }, { "epoch": 0.31339702669123004, "grad_norm": 4.57386589050293, "learning_rate": 9.903375840574291e-05, "loss": 0.7655128479003906, "memory(GiB)": 57.83, "step": 7315, "token_acc": 0.8305647840531561, "train_speed(iter/s)": 1.51223 }, { "epoch": 0.31361124202047896, "grad_norm": 0.18039590120315552, "learning_rate": 9.903244132807597e-05, "loss": 0.25089752674102783, "memory(GiB)": 57.83, "step": 7320, "token_acc": 0.948339483394834, "train_speed(iter/s)": 1.512267 }, { "epoch": 0.31382545734972794, "grad_norm": 5.29689359664917, "learning_rate": 9.903112336214035e-05, "loss": 0.42615246772766113, "memory(GiB)": 57.83, "step": 7325, "token_acc": 0.9087947882736156, "train_speed(iter/s)": 1.512264 }, { "epoch": 0.3140396726789769, "grad_norm": 3.888011932373047, "learning_rate": 9.902980450795996e-05, "loss": 0.35603621006011965, "memory(GiB)": 57.83, "step": 7330, "token_acc": 0.9127725856697819, "train_speed(iter/s)": 1.512413 }, { "epoch": 0.31425388800822585, "grad_norm": 4.163918972015381, "learning_rate": 9.902848476555864e-05, "loss": 0.6894698143005371, "memory(GiB)": 57.83, "step": 7335, "token_acc": 0.8617363344051447, "train_speed(iter/s)": 1.512513 }, { "epoch": 0.3144681033374748, "grad_norm": 5.216299533843994, "learning_rate": 9.902716413496034e-05, "loss": 0.5508815765380859, "memory(GiB)": 57.83, "step": 7340, "token_acc": 0.8741496598639455, "train_speed(iter/s)": 1.512526 }, { "epoch": 0.3146823186667238, "grad_norm": 3.7649664878845215, "learning_rate": 9.902584261618896e-05, "loss": 0.33192620277404783, "memory(GiB)": 57.83, "step": 7345, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.512612 }, { "epoch": 0.3148965339959727, "grad_norm": 0.5429190397262573, "learning_rate": 9.902452020926845e-05, "loss": 0.24877243041992186, "memory(GiB)": 57.83, "step": 7350, "token_acc": 0.9496124031007752, "train_speed(iter/s)": 1.512744 }, { "epoch": 0.3151107493252217, "grad_norm": 2.708301067352295, "learning_rate": 9.902319691422277e-05, "loss": 0.335737419128418, "memory(GiB)": 57.83, "step": 7355, "token_acc": 0.9164086687306502, "train_speed(iter/s)": 1.512727 }, { "epoch": 0.3153249646544707, "grad_norm": 1.7026225328445435, "learning_rate": 9.902187273107591e-05, "loss": 0.23086197376251222, "memory(GiB)": 57.83, "step": 7360, "token_acc": 0.9535714285714286, "train_speed(iter/s)": 1.512824 }, { "epoch": 0.3155391799837196, "grad_norm": 2.1995391845703125, "learning_rate": 9.902054765985182e-05, "loss": 0.6104345321655273, "memory(GiB)": 57.83, "step": 7365, "token_acc": 0.8770491803278688, "train_speed(iter/s)": 1.51292 }, { "epoch": 0.3157533953129686, "grad_norm": 3.046618938446045, "learning_rate": 9.901922170057452e-05, "loss": 0.42105765342712403, "memory(GiB)": 57.83, "step": 7370, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.51293 }, { "epoch": 0.31596761064221757, "grad_norm": 3.6870734691619873, "learning_rate": 9.901789485326804e-05, "loss": 0.619783878326416, "memory(GiB)": 57.83, "step": 7375, "token_acc": 0.8854166666666666, "train_speed(iter/s)": 1.513035 }, { "epoch": 0.31618182597146655, "grad_norm": 2.9537124633789062, "learning_rate": 9.901656711795641e-05, "loss": 0.5222084045410156, "memory(GiB)": 57.83, "step": 7380, "token_acc": 0.8859060402684564, "train_speed(iter/s)": 1.513194 }, { "epoch": 0.31639604130071547, "grad_norm": 2.0172665119171143, "learning_rate": 9.90152384946637e-05, "loss": 0.32829673290252687, "memory(GiB)": 57.83, "step": 7385, "token_acc": 0.9249146757679181, "train_speed(iter/s)": 1.513381 }, { "epoch": 0.31661025662996445, "grad_norm": 3.6691906452178955, "learning_rate": 9.901390898341397e-05, "loss": 0.2679779052734375, "memory(GiB)": 57.83, "step": 7390, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.51355 }, { "epoch": 0.3168244719592134, "grad_norm": 3.602094888687134, "learning_rate": 9.901257858423127e-05, "loss": 0.6269121170043945, "memory(GiB)": 57.83, "step": 7395, "token_acc": 0.8669064748201439, "train_speed(iter/s)": 1.513526 }, { "epoch": 0.31703868728846235, "grad_norm": 6.41495943069458, "learning_rate": 9.901124729713975e-05, "loss": 0.5300108909606933, "memory(GiB)": 57.83, "step": 7400, "token_acc": 0.889261744966443, "train_speed(iter/s)": 1.513503 }, { "epoch": 0.31725290261771133, "grad_norm": 1.362703800201416, "learning_rate": 9.900991512216351e-05, "loss": 0.30472090244293215, "memory(GiB)": 57.83, "step": 7405, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.513474 }, { "epoch": 0.3174671179469603, "grad_norm": 0.4710277318954468, "learning_rate": 9.900858205932668e-05, "loss": 0.34242470264434816, "memory(GiB)": 57.83, "step": 7410, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.513474 }, { "epoch": 0.31768133327620923, "grad_norm": 2.573176383972168, "learning_rate": 9.900724810865341e-05, "loss": 0.44007582664489747, "memory(GiB)": 57.83, "step": 7415, "token_acc": 0.9164086687306502, "train_speed(iter/s)": 1.51335 }, { "epoch": 0.3178955486054582, "grad_norm": 3.1057937145233154, "learning_rate": 9.900591327016786e-05, "loss": 0.24982075691223143, "memory(GiB)": 57.83, "step": 7420, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.513283 }, { "epoch": 0.3181097639347072, "grad_norm": 2.5903825759887695, "learning_rate": 9.900457754389422e-05, "loss": 0.39055862426757815, "memory(GiB)": 57.83, "step": 7425, "token_acc": 0.9121813031161473, "train_speed(iter/s)": 1.513255 }, { "epoch": 0.3183239792639561, "grad_norm": 2.4176464080810547, "learning_rate": 9.900324092985671e-05, "loss": 0.22842936515808104, "memory(GiB)": 57.83, "step": 7430, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.513277 }, { "epoch": 0.3185381945932051, "grad_norm": 6.054247856140137, "learning_rate": 9.900190342807951e-05, "loss": 0.35565962791442873, "memory(GiB)": 57.83, "step": 7435, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.51343 }, { "epoch": 0.31875240992245407, "grad_norm": 2.6251511573791504, "learning_rate": 9.900056503858685e-05, "loss": 0.30728271007537844, "memory(GiB)": 57.83, "step": 7440, "token_acc": 0.9168975069252078, "train_speed(iter/s)": 1.513492 }, { "epoch": 0.318966625251703, "grad_norm": 2.477504253387451, "learning_rate": 9.8999225761403e-05, "loss": 0.22523784637451172, "memory(GiB)": 57.83, "step": 7445, "token_acc": 0.9501661129568106, "train_speed(iter/s)": 1.513571 }, { "epoch": 0.319180840580952, "grad_norm": 3.2621123790740967, "learning_rate": 9.899788559655221e-05, "loss": 0.41948490142822265, "memory(GiB)": 57.83, "step": 7450, "token_acc": 0.8904899135446686, "train_speed(iter/s)": 1.513587 }, { "epoch": 0.31939505591020095, "grad_norm": 4.438309192657471, "learning_rate": 9.899654454405876e-05, "loss": 0.4035494804382324, "memory(GiB)": 57.83, "step": 7455, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.513566 }, { "epoch": 0.3196092712394499, "grad_norm": 1.571084976196289, "learning_rate": 9.899520260394695e-05, "loss": 0.5077800750732422, "memory(GiB)": 57.83, "step": 7460, "token_acc": 0.9063444108761329, "train_speed(iter/s)": 1.51363 }, { "epoch": 0.31982348656869886, "grad_norm": 12.723091125488281, "learning_rate": 9.899385977624107e-05, "loss": 0.6054377555847168, "memory(GiB)": 57.83, "step": 7465, "token_acc": 0.8781094527363185, "train_speed(iter/s)": 1.513839 }, { "epoch": 0.32003770189794783, "grad_norm": 4.132344722747803, "learning_rate": 9.899251606096546e-05, "loss": 0.47092862129211427, "memory(GiB)": 57.83, "step": 7470, "token_acc": 0.8962655601659751, "train_speed(iter/s)": 1.513811 }, { "epoch": 0.32025191722719676, "grad_norm": 1.435442566871643, "learning_rate": 9.899117145814448e-05, "loss": 0.4270926952362061, "memory(GiB)": 57.83, "step": 7475, "token_acc": 0.9100719424460432, "train_speed(iter/s)": 1.513902 }, { "epoch": 0.32046613255644574, "grad_norm": 8.602392196655273, "learning_rate": 9.898982596780244e-05, "loss": 0.4475921630859375, "memory(GiB)": 57.83, "step": 7480, "token_acc": 0.9134615384615384, "train_speed(iter/s)": 1.514004 }, { "epoch": 0.3206803478856947, "grad_norm": 5.675102233886719, "learning_rate": 9.898847958996377e-05, "loss": 0.3992893934249878, "memory(GiB)": 57.83, "step": 7485, "token_acc": 0.8829787234042553, "train_speed(iter/s)": 1.514034 }, { "epoch": 0.32089456321494364, "grad_norm": 1.5683780908584595, "learning_rate": 9.898713232465283e-05, "loss": 0.2874259948730469, "memory(GiB)": 57.83, "step": 7490, "token_acc": 0.9252336448598131, "train_speed(iter/s)": 1.514212 }, { "epoch": 0.3211087785441926, "grad_norm": 4.750735282897949, "learning_rate": 9.898578417189403e-05, "loss": 0.6235380649566651, "memory(GiB)": 57.83, "step": 7495, "token_acc": 0.86328125, "train_speed(iter/s)": 1.514294 }, { "epoch": 0.3213229938734416, "grad_norm": 4.05375862121582, "learning_rate": 9.89844351317118e-05, "loss": 0.628300952911377, "memory(GiB)": 61.91, "step": 7500, "token_acc": 0.8819188191881919, "train_speed(iter/s)": 1.514305 }, { "epoch": 0.3213229938734416, "eval_loss": 2.471130132675171, "eval_runtime": 11.2317, "eval_samples_per_second": 8.903, "eval_steps_per_second": 8.903, "eval_token_acc": 0.43293492695883135, "step": 7500 }, { "epoch": 0.3215372092026905, "grad_norm": 2.394857883453369, "learning_rate": 9.89830852041306e-05, "loss": 0.37443411350250244, "memory(GiB)": 61.91, "step": 7505, "token_acc": 0.5771375464684015, "train_speed(iter/s)": 1.510706 }, { "epoch": 0.3217514245319395, "grad_norm": 3.833411455154419, "learning_rate": 9.898173438917485e-05, "loss": 0.2604188919067383, "memory(GiB)": 61.91, "step": 7510, "token_acc": 0.9430604982206405, "train_speed(iter/s)": 1.510912 }, { "epoch": 0.3219656398611885, "grad_norm": 0.5899659395217896, "learning_rate": 9.8980382686869e-05, "loss": 0.19144421815872192, "memory(GiB)": 61.91, "step": 7515, "token_acc": 0.9614147909967846, "train_speed(iter/s)": 1.510866 }, { "epoch": 0.3221798551904374, "grad_norm": 5.379552364349365, "learning_rate": 9.89790300972376e-05, "loss": 0.6795905113220215, "memory(GiB)": 61.91, "step": 7520, "token_acc": 0.8622950819672132, "train_speed(iter/s)": 1.510819 }, { "epoch": 0.3223940705196864, "grad_norm": 2.046382427215576, "learning_rate": 9.897767662030512e-05, "loss": 0.18946577310562135, "memory(GiB)": 61.91, "step": 7525, "token_acc": 0.9609929078014184, "train_speed(iter/s)": 1.510886 }, { "epoch": 0.32260828584893536, "grad_norm": 0.10184159874916077, "learning_rate": 9.897632225609607e-05, "loss": 0.4604065418243408, "memory(GiB)": 61.91, "step": 7530, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.510987 }, { "epoch": 0.3228225011781843, "grad_norm": 3.41998028755188, "learning_rate": 9.897496700463502e-05, "loss": 0.48907318115234377, "memory(GiB)": 61.91, "step": 7535, "token_acc": 0.9003115264797508, "train_speed(iter/s)": 1.511196 }, { "epoch": 0.32303671650743326, "grad_norm": 1.2632415294647217, "learning_rate": 9.897361086594649e-05, "loss": 0.3889720916748047, "memory(GiB)": 61.91, "step": 7540, "token_acc": 0.9161849710982659, "train_speed(iter/s)": 1.511251 }, { "epoch": 0.32325093183668224, "grad_norm": 3.7106964588165283, "learning_rate": 9.897225384005507e-05, "loss": 0.5398503303527832, "memory(GiB)": 61.91, "step": 7545, "token_acc": 0.9060402684563759, "train_speed(iter/s)": 1.51128 }, { "epoch": 0.3234651471659312, "grad_norm": 0.9939740896224976, "learning_rate": 9.897089592698532e-05, "loss": 0.2735555171966553, "memory(GiB)": 61.91, "step": 7550, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.511345 }, { "epoch": 0.32367936249518015, "grad_norm": 2.3408849239349365, "learning_rate": 9.896953712676184e-05, "loss": 0.12352042198181153, "memory(GiB)": 61.91, "step": 7555, "token_acc": 0.9666666666666667, "train_speed(iter/s)": 1.511299 }, { "epoch": 0.3238935778244291, "grad_norm": 1.8687247037887573, "learning_rate": 9.896817743940928e-05, "loss": 0.3166090965270996, "memory(GiB)": 61.91, "step": 7560, "token_acc": 0.9350282485875706, "train_speed(iter/s)": 1.511336 }, { "epoch": 0.3241077931536781, "grad_norm": 4.465818881988525, "learning_rate": 9.896681686495224e-05, "loss": 0.5729311943054199, "memory(GiB)": 61.91, "step": 7565, "token_acc": 0.8711111111111111, "train_speed(iter/s)": 1.511368 }, { "epoch": 0.324322008482927, "grad_norm": 1.1936551332473755, "learning_rate": 9.896545540341538e-05, "loss": 0.2624319553375244, "memory(GiB)": 61.91, "step": 7570, "token_acc": 0.9387096774193548, "train_speed(iter/s)": 1.511386 }, { "epoch": 0.324536223812176, "grad_norm": 5.345091819763184, "learning_rate": 9.896409305482336e-05, "loss": 0.5879055023193359, "memory(GiB)": 61.91, "step": 7575, "token_acc": 0.8701754385964913, "train_speed(iter/s)": 1.511341 }, { "epoch": 0.324750439141425, "grad_norm": 5.5520758628845215, "learning_rate": 9.896272981920087e-05, "loss": 0.3642292499542236, "memory(GiB)": 61.91, "step": 7580, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.511333 }, { "epoch": 0.3249646544706739, "grad_norm": 4.101814270019531, "learning_rate": 9.89613656965726e-05, "loss": 0.7646390914916992, "memory(GiB)": 61.91, "step": 7585, "token_acc": 0.8488372093023255, "train_speed(iter/s)": 1.511417 }, { "epoch": 0.3251788697999229, "grad_norm": 1.3303775787353516, "learning_rate": 9.896000068696325e-05, "loss": 0.3317232847213745, "memory(GiB)": 61.91, "step": 7590, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.511425 }, { "epoch": 0.32539308512917187, "grad_norm": 3.1817715167999268, "learning_rate": 9.895863479039756e-05, "loss": 0.4084768772125244, "memory(GiB)": 61.91, "step": 7595, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.511553 }, { "epoch": 0.3256073004584208, "grad_norm": 4.652787208557129, "learning_rate": 9.895726800690028e-05, "loss": 0.3986428022384644, "memory(GiB)": 61.91, "step": 7600, "token_acc": 0.8893129770992366, "train_speed(iter/s)": 1.511789 }, { "epoch": 0.32582151578766977, "grad_norm": 10.184610366821289, "learning_rate": 9.895590033649616e-05, "loss": 0.5533396244049072, "memory(GiB)": 61.91, "step": 7605, "token_acc": 0.8941176470588236, "train_speed(iter/s)": 1.511941 }, { "epoch": 0.32603573111691875, "grad_norm": 5.7659220695495605, "learning_rate": 9.895453177920997e-05, "loss": 0.5086452484130859, "memory(GiB)": 61.91, "step": 7610, "token_acc": 0.898360655737705, "train_speed(iter/s)": 1.512105 }, { "epoch": 0.32624994644616767, "grad_norm": 4.5104265213012695, "learning_rate": 9.895316233506653e-05, "loss": 0.5718632698059082, "memory(GiB)": 61.91, "step": 7615, "token_acc": 0.8966789667896679, "train_speed(iter/s)": 1.512118 }, { "epoch": 0.32646416177541665, "grad_norm": 2.036647319793701, "learning_rate": 9.895179200409063e-05, "loss": 0.33007044792175294, "memory(GiB)": 61.91, "step": 7620, "token_acc": 0.9208333333333333, "train_speed(iter/s)": 1.5121 }, { "epoch": 0.32667837710466563, "grad_norm": 4.815316200256348, "learning_rate": 9.895042078630709e-05, "loss": 0.48027191162109373, "memory(GiB)": 61.91, "step": 7625, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.512128 }, { "epoch": 0.32689259243391455, "grad_norm": 4.029754638671875, "learning_rate": 9.894904868174076e-05, "loss": 0.40860648155212403, "memory(GiB)": 61.91, "step": 7630, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.512328 }, { "epoch": 0.32710680776316353, "grad_norm": 3.0286965370178223, "learning_rate": 9.89476756904165e-05, "loss": 0.5275236606597901, "memory(GiB)": 61.91, "step": 7635, "token_acc": 0.8834586466165414, "train_speed(iter/s)": 1.512375 }, { "epoch": 0.3273210230924125, "grad_norm": 3.2249832153320312, "learning_rate": 9.894630181235917e-05, "loss": 0.663758373260498, "memory(GiB)": 61.91, "step": 7640, "token_acc": 0.8501529051987767, "train_speed(iter/s)": 1.512375 }, { "epoch": 0.32753523842166143, "grad_norm": 1.9380398988723755, "learning_rate": 9.894492704759369e-05, "loss": 0.34820919036865233, "memory(GiB)": 61.91, "step": 7645, "token_acc": 0.9400630914826499, "train_speed(iter/s)": 1.512455 }, { "epoch": 0.3277494537509104, "grad_norm": 2.604832410812378, "learning_rate": 9.894355139614493e-05, "loss": 0.7080265045166015, "memory(GiB)": 61.91, "step": 7650, "token_acc": 0.805111821086262, "train_speed(iter/s)": 1.512485 }, { "epoch": 0.3279636690801594, "grad_norm": 4.896474838256836, "learning_rate": 9.89421748580378e-05, "loss": 0.335146427154541, "memory(GiB)": 61.91, "step": 7655, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.512457 }, { "epoch": 0.3281778844094083, "grad_norm": 0.37768590450286865, "learning_rate": 9.894079743329729e-05, "loss": 0.2550898551940918, "memory(GiB)": 61.91, "step": 7660, "token_acc": 0.9429657794676806, "train_speed(iter/s)": 1.512499 }, { "epoch": 0.3283920997386573, "grad_norm": 2.009614944458008, "learning_rate": 9.893941912194831e-05, "loss": 0.5242821216583252, "memory(GiB)": 61.91, "step": 7665, "token_acc": 0.8877887788778878, "train_speed(iter/s)": 1.512506 }, { "epoch": 0.3286063150679063, "grad_norm": 0.8282000422477722, "learning_rate": 9.893803992401586e-05, "loss": 0.3551580190658569, "memory(GiB)": 61.91, "step": 7670, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.51245 }, { "epoch": 0.3288205303971552, "grad_norm": 5.967987060546875, "learning_rate": 9.893665983952489e-05, "loss": 0.501586627960205, "memory(GiB)": 61.91, "step": 7675, "token_acc": 0.8964285714285715, "train_speed(iter/s)": 1.512462 }, { "epoch": 0.3290347457264042, "grad_norm": 0.7959259152412415, "learning_rate": 9.893527886850044e-05, "loss": 0.27297139167785645, "memory(GiB)": 61.91, "step": 7680, "token_acc": 0.9111969111969112, "train_speed(iter/s)": 1.512429 }, { "epoch": 0.32924896105565316, "grad_norm": 2.402778387069702, "learning_rate": 9.89338970109675e-05, "loss": 0.4628042697906494, "memory(GiB)": 61.91, "step": 7685, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.512426 }, { "epoch": 0.3294631763849021, "grad_norm": 0.6705512404441833, "learning_rate": 9.893251426695111e-05, "loss": 0.28538012504577637, "memory(GiB)": 61.91, "step": 7690, "token_acc": 0.9349593495934959, "train_speed(iter/s)": 1.512433 }, { "epoch": 0.32967739171415106, "grad_norm": 2.4664878845214844, "learning_rate": 9.893113063647632e-05, "loss": 0.18037619590759277, "memory(GiB)": 61.91, "step": 7695, "token_acc": 0.9628252788104089, "train_speed(iter/s)": 1.512407 }, { "epoch": 0.32989160704340004, "grad_norm": 3.625056028366089, "learning_rate": 9.89297461195682e-05, "loss": 0.6730573654174805, "memory(GiB)": 61.91, "step": 7700, "token_acc": 0.8779761904761905, "train_speed(iter/s)": 1.512404 }, { "epoch": 0.33010582237264896, "grad_norm": 4.593713283538818, "learning_rate": 9.892836071625182e-05, "loss": 0.9273441314697266, "memory(GiB)": 61.91, "step": 7705, "token_acc": 0.7448453608247423, "train_speed(iter/s)": 1.512364 }, { "epoch": 0.33032003770189794, "grad_norm": 6.467137336730957, "learning_rate": 9.89269744265523e-05, "loss": 0.4195289134979248, "memory(GiB)": 61.91, "step": 7710, "token_acc": 0.9129032258064517, "train_speed(iter/s)": 1.512386 }, { "epoch": 0.3305342530311469, "grad_norm": 5.038894176483154, "learning_rate": 9.892558725049474e-05, "loss": 0.29413418769836425, "memory(GiB)": 61.91, "step": 7715, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.512523 }, { "epoch": 0.3307484683603959, "grad_norm": 2.5692009925842285, "learning_rate": 9.892419918810426e-05, "loss": 0.1654599905014038, "memory(GiB)": 61.91, "step": 7720, "token_acc": 0.963855421686747, "train_speed(iter/s)": 1.512513 }, { "epoch": 0.3309626836896448, "grad_norm": 3.5114898681640625, "learning_rate": 9.892281023940602e-05, "loss": 0.4527285575866699, "memory(GiB)": 61.91, "step": 7725, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.512574 }, { "epoch": 0.3311768990188938, "grad_norm": 1.7316348552703857, "learning_rate": 9.892142040442518e-05, "loss": 0.18726415634155275, "memory(GiB)": 61.91, "step": 7730, "token_acc": 0.9664429530201343, "train_speed(iter/s)": 1.512535 }, { "epoch": 0.3313911143481428, "grad_norm": 3.220822811126709, "learning_rate": 9.892002968318692e-05, "loss": 0.6320162773132324, "memory(GiB)": 61.91, "step": 7735, "token_acc": 0.8689138576779026, "train_speed(iter/s)": 1.51246 }, { "epoch": 0.3316053296773917, "grad_norm": 3.338132381439209, "learning_rate": 9.891863807571644e-05, "loss": 0.5521392822265625, "memory(GiB)": 61.91, "step": 7740, "token_acc": 0.8805031446540881, "train_speed(iter/s)": 1.512405 }, { "epoch": 0.3318195450066407, "grad_norm": 3.192582845687866, "learning_rate": 9.891724558203893e-05, "loss": 0.24721839427947997, "memory(GiB)": 61.91, "step": 7745, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.512391 }, { "epoch": 0.33203376033588966, "grad_norm": 2.6268222332000732, "learning_rate": 9.891585220217964e-05, "loss": 0.4913632869720459, "memory(GiB)": 61.91, "step": 7750, "token_acc": 0.9061371841155235, "train_speed(iter/s)": 1.51237 }, { "epoch": 0.3322479756651386, "grad_norm": 1.774531364440918, "learning_rate": 9.891445793616378e-05, "loss": 0.6036566734313965, "memory(GiB)": 61.91, "step": 7755, "token_acc": 0.8618618618618619, "train_speed(iter/s)": 1.512465 }, { "epoch": 0.33246219099438756, "grad_norm": 2.4534406661987305, "learning_rate": 9.891306278401665e-05, "loss": 0.3423898220062256, "memory(GiB)": 61.91, "step": 7760, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.512425 }, { "epoch": 0.33267640632363654, "grad_norm": 2.692946195602417, "learning_rate": 9.891166674576349e-05, "loss": 0.5008926391601562, "memory(GiB)": 61.91, "step": 7765, "token_acc": 0.8996763754045307, "train_speed(iter/s)": 1.512325 }, { "epoch": 0.33289062165288547, "grad_norm": 3.4724202156066895, "learning_rate": 9.891026982142962e-05, "loss": 0.5043286323547364, "memory(GiB)": 61.91, "step": 7770, "token_acc": 0.875, "train_speed(iter/s)": 1.512252 }, { "epoch": 0.33310483698213444, "grad_norm": 3.0407371520996094, "learning_rate": 9.890887201104032e-05, "loss": 0.747948169708252, "memory(GiB)": 61.91, "step": 7775, "token_acc": 0.8230088495575221, "train_speed(iter/s)": 1.512373 }, { "epoch": 0.3333190523113834, "grad_norm": 3.656493902206421, "learning_rate": 9.890747331462092e-05, "loss": 0.4628904819488525, "memory(GiB)": 61.91, "step": 7780, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.512486 }, { "epoch": 0.33353326764063235, "grad_norm": 0.5777880549430847, "learning_rate": 9.890607373219676e-05, "loss": 0.4038088798522949, "memory(GiB)": 61.91, "step": 7785, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.512467 }, { "epoch": 0.3337474829698813, "grad_norm": 3.3492591381073, "learning_rate": 9.89046732637932e-05, "loss": 0.415972375869751, "memory(GiB)": 61.91, "step": 7790, "token_acc": 0.9071729957805907, "train_speed(iter/s)": 1.512473 }, { "epoch": 0.3339616982991303, "grad_norm": 3.8986597061157227, "learning_rate": 9.890327190943561e-05, "loss": 0.5687922477722168, "memory(GiB)": 61.91, "step": 7795, "token_acc": 0.8832807570977917, "train_speed(iter/s)": 1.512481 }, { "epoch": 0.33417591362837923, "grad_norm": 3.045149803161621, "learning_rate": 9.890186966914938e-05, "loss": 0.16700277328491211, "memory(GiB)": 61.91, "step": 7800, "token_acc": 0.9566929133858267, "train_speed(iter/s)": 1.512548 }, { "epoch": 0.3343901289576282, "grad_norm": 1.0379130840301514, "learning_rate": 9.89004665429599e-05, "loss": 0.23621225357055664, "memory(GiB)": 61.91, "step": 7805, "token_acc": 0.9316546762589928, "train_speed(iter/s)": 1.51252 }, { "epoch": 0.3346043442868772, "grad_norm": 4.811845779418945, "learning_rate": 9.88990625308926e-05, "loss": 0.32094731330871584, "memory(GiB)": 61.91, "step": 7810, "token_acc": 0.9348659003831418, "train_speed(iter/s)": 1.512562 }, { "epoch": 0.3348185596161261, "grad_norm": 3.1429548263549805, "learning_rate": 9.889765763297291e-05, "loss": 0.30712404251098635, "memory(GiB)": 61.91, "step": 7815, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.512546 }, { "epoch": 0.3350327749453751, "grad_norm": 0.26470065116882324, "learning_rate": 9.889625184922628e-05, "loss": 0.26955225467681887, "memory(GiB)": 61.91, "step": 7820, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.512526 }, { "epoch": 0.33524699027462407, "grad_norm": 5.7473249435424805, "learning_rate": 9.889484517967818e-05, "loss": 0.5988773345947266, "memory(GiB)": 61.91, "step": 7825, "token_acc": 0.8581081081081081, "train_speed(iter/s)": 1.512617 }, { "epoch": 0.335461205603873, "grad_norm": 1.860182285308838, "learning_rate": 9.889343762435409e-05, "loss": 0.6269731998443604, "memory(GiB)": 61.91, "step": 7830, "token_acc": 0.8953488372093024, "train_speed(iter/s)": 1.512655 }, { "epoch": 0.33567542093312197, "grad_norm": 3.147463321685791, "learning_rate": 9.88920291832795e-05, "loss": 0.5964645862579345, "memory(GiB)": 61.91, "step": 7835, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.512698 }, { "epoch": 0.33588963626237095, "grad_norm": 3.220151424407959, "learning_rate": 9.889061985647996e-05, "loss": 0.3388816356658936, "memory(GiB)": 61.91, "step": 7840, "token_acc": 0.92, "train_speed(iter/s)": 1.512711 }, { "epoch": 0.3361038515916199, "grad_norm": 4.865410327911377, "learning_rate": 9.888920964398099e-05, "loss": 0.5054830551147461, "memory(GiB)": 61.91, "step": 7845, "token_acc": 0.8869257950530035, "train_speed(iter/s)": 1.512784 }, { "epoch": 0.33631806692086885, "grad_norm": 5.422826290130615, "learning_rate": 9.88877985458081e-05, "loss": 0.6081606864929199, "memory(GiB)": 61.91, "step": 7850, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.512771 }, { "epoch": 0.33653228225011783, "grad_norm": 2.2120232582092285, "learning_rate": 9.888638656198688e-05, "loss": 0.3049129009246826, "memory(GiB)": 61.91, "step": 7855, "token_acc": 0.9435215946843853, "train_speed(iter/s)": 1.512785 }, { "epoch": 0.33674649757936675, "grad_norm": 2.3543052673339844, "learning_rate": 9.88849736925429e-05, "loss": 0.33651270866394045, "memory(GiB)": 61.91, "step": 7860, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.512862 }, { "epoch": 0.33696071290861573, "grad_norm": 3.1480846405029297, "learning_rate": 9.888355993750178e-05, "loss": 0.2654085874557495, "memory(GiB)": 61.91, "step": 7865, "token_acc": 0.9496124031007752, "train_speed(iter/s)": 1.512836 }, { "epoch": 0.3371749282378647, "grad_norm": 2.259981155395508, "learning_rate": 9.888214529688912e-05, "loss": 0.5286569118499755, "memory(GiB)": 61.91, "step": 7870, "token_acc": 0.8699690402476781, "train_speed(iter/s)": 1.512937 }, { "epoch": 0.33738914356711364, "grad_norm": 2.343120574951172, "learning_rate": 9.888072977073053e-05, "loss": 0.5320161342620849, "memory(GiB)": 61.91, "step": 7875, "token_acc": 0.8793650793650793, "train_speed(iter/s)": 1.51294 }, { "epoch": 0.3376033588963626, "grad_norm": 1.5522041320800781, "learning_rate": 9.887931335905168e-05, "loss": 0.2479764461517334, "memory(GiB)": 61.91, "step": 7880, "token_acc": 0.944078947368421, "train_speed(iter/s)": 1.512902 }, { "epoch": 0.3378175742256116, "grad_norm": 3.3239758014678955, "learning_rate": 9.887789606187819e-05, "loss": 0.4302541732788086, "memory(GiB)": 61.91, "step": 7885, "token_acc": 0.9226361031518625, "train_speed(iter/s)": 1.512798 }, { "epoch": 0.3380317895548606, "grad_norm": 2.633512020111084, "learning_rate": 9.887647787923578e-05, "loss": 0.4566482067108154, "memory(GiB)": 61.91, "step": 7890, "token_acc": 0.9066265060240963, "train_speed(iter/s)": 1.512796 }, { "epoch": 0.3382460048841095, "grad_norm": 2.6229753494262695, "learning_rate": 9.887505881115013e-05, "loss": 0.43462309837341306, "memory(GiB)": 61.91, "step": 7895, "token_acc": 0.9225806451612903, "train_speed(iter/s)": 1.512846 }, { "epoch": 0.3384602202133585, "grad_norm": 4.167659282684326, "learning_rate": 9.887363885764693e-05, "loss": 0.24906535148620607, "memory(GiB)": 61.91, "step": 7900, "token_acc": 0.9530201342281879, "train_speed(iter/s)": 1.512864 }, { "epoch": 0.33867443554260745, "grad_norm": 5.141737461090088, "learning_rate": 9.887221801875192e-05, "loss": 0.5086905479431152, "memory(GiB)": 61.91, "step": 7905, "token_acc": 0.8951310861423221, "train_speed(iter/s)": 1.512873 }, { "epoch": 0.3388886508718564, "grad_norm": 1.3504524230957031, "learning_rate": 9.887079629449083e-05, "loss": 0.20787546634674073, "memory(GiB)": 61.91, "step": 7910, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.512919 }, { "epoch": 0.33910286620110536, "grad_norm": 3.4897971153259277, "learning_rate": 9.886937368488942e-05, "loss": 0.29567530155181887, "memory(GiB)": 61.91, "step": 7915, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.512884 }, { "epoch": 0.33931708153035434, "grad_norm": 2.843583583831787, "learning_rate": 9.886795018997347e-05, "loss": 0.4375617980957031, "memory(GiB)": 61.91, "step": 7920, "token_acc": 0.9039735099337748, "train_speed(iter/s)": 1.512811 }, { "epoch": 0.33953129685960326, "grad_norm": 3.080043077468872, "learning_rate": 9.886652580976876e-05, "loss": 0.44490795135498046, "memory(GiB)": 61.91, "step": 7925, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.512832 }, { "epoch": 0.33974551218885224, "grad_norm": 2.547186851501465, "learning_rate": 9.886510054430108e-05, "loss": 0.29417855739593507, "memory(GiB)": 61.91, "step": 7930, "token_acc": 0.9427480916030534, "train_speed(iter/s)": 1.512843 }, { "epoch": 0.3399597275181012, "grad_norm": 8.5975923538208, "learning_rate": 9.886367439359627e-05, "loss": 0.3139829635620117, "memory(GiB)": 61.91, "step": 7935, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.512853 }, { "epoch": 0.34017394284735014, "grad_norm": 3.0203027725219727, "learning_rate": 9.886224735768017e-05, "loss": 0.5224937915802002, "memory(GiB)": 61.91, "step": 7940, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.512953 }, { "epoch": 0.3403881581765991, "grad_norm": 4.282827854156494, "learning_rate": 9.886081943657862e-05, "loss": 0.4737374782562256, "memory(GiB)": 61.91, "step": 7945, "token_acc": 0.8921933085501859, "train_speed(iter/s)": 1.512919 }, { "epoch": 0.3406023735058481, "grad_norm": 0.5158848166465759, "learning_rate": 9.885939063031748e-05, "loss": 0.3342712163925171, "memory(GiB)": 61.91, "step": 7950, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.512945 }, { "epoch": 0.340816588835097, "grad_norm": 3.530867099761963, "learning_rate": 9.885796093892266e-05, "loss": 0.6008416652679444, "memory(GiB)": 61.91, "step": 7955, "token_acc": 0.868421052631579, "train_speed(iter/s)": 1.513011 }, { "epoch": 0.341030804164346, "grad_norm": 1.5691142082214355, "learning_rate": 9.885653036242004e-05, "loss": 0.49411940574645996, "memory(GiB)": 61.91, "step": 7960, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.513109 }, { "epoch": 0.341245019493595, "grad_norm": 2.442148447036743, "learning_rate": 9.885509890083555e-05, "loss": 0.4743558406829834, "memory(GiB)": 61.91, "step": 7965, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.513095 }, { "epoch": 0.3414592348228439, "grad_norm": 3.512605667114258, "learning_rate": 9.88536665541951e-05, "loss": 0.4941071033477783, "memory(GiB)": 61.91, "step": 7970, "token_acc": 0.8849206349206349, "train_speed(iter/s)": 1.51315 }, { "epoch": 0.3416734501520929, "grad_norm": 1.1643120050430298, "learning_rate": 9.885223332252464e-05, "loss": 0.3719114542007446, "memory(GiB)": 61.91, "step": 7975, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.513268 }, { "epoch": 0.34188766548134186, "grad_norm": 5.690678119659424, "learning_rate": 9.885079920585017e-05, "loss": 0.2568089962005615, "memory(GiB)": 61.91, "step": 7980, "token_acc": 0.9409448818897638, "train_speed(iter/s)": 1.513298 }, { "epoch": 0.3421018808105908, "grad_norm": 1.0471134185791016, "learning_rate": 9.884936420419763e-05, "loss": 0.30522263050079346, "memory(GiB)": 61.91, "step": 7985, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.513354 }, { "epoch": 0.34231609613983977, "grad_norm": 1.7572581768035889, "learning_rate": 9.884792831759305e-05, "loss": 0.49479265213012696, "memory(GiB)": 61.91, "step": 7990, "token_acc": 0.8945454545454545, "train_speed(iter/s)": 1.513384 }, { "epoch": 0.34253031146908874, "grad_norm": 5.182215213775635, "learning_rate": 9.884649154606242e-05, "loss": 0.3236077070236206, "memory(GiB)": 61.91, "step": 7995, "token_acc": 0.9184952978056427, "train_speed(iter/s)": 1.513386 }, { "epoch": 0.34274452679833767, "grad_norm": 2.6816155910491943, "learning_rate": 9.884505388963176e-05, "loss": 0.157029128074646, "memory(GiB)": 61.91, "step": 8000, "token_acc": 0.9741935483870968, "train_speed(iter/s)": 1.513428 }, { "epoch": 0.34274452679833767, "eval_loss": 2.548301935195923, "eval_runtime": 11.9686, "eval_samples_per_second": 8.355, "eval_steps_per_second": 8.355, "eval_token_acc": 0.41893644617380027, "step": 8000 }, { "epoch": 0.34295874212758665, "grad_norm": 1.4683775901794434, "learning_rate": 9.884361534832716e-05, "loss": 0.4339284896850586, "memory(GiB)": 61.91, "step": 8005, "token_acc": 0.5536881419234361, "train_speed(iter/s)": 1.509888 }, { "epoch": 0.3431729574568356, "grad_norm": 1.051316499710083, "learning_rate": 9.884217592217461e-05, "loss": 0.2653156280517578, "memory(GiB)": 61.91, "step": 8010, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.509974 }, { "epoch": 0.34338717278608455, "grad_norm": 0.9830781817436218, "learning_rate": 9.884073561120026e-05, "loss": 0.15035663843154906, "memory(GiB)": 61.91, "step": 8015, "token_acc": 0.9711191335740073, "train_speed(iter/s)": 1.510085 }, { "epoch": 0.34360138811533353, "grad_norm": 1.0296674966812134, "learning_rate": 9.883929441543014e-05, "loss": 0.21684775352478028, "memory(GiB)": 61.91, "step": 8020, "token_acc": 0.9504643962848297, "train_speed(iter/s)": 1.510043 }, { "epoch": 0.3438156034445825, "grad_norm": 1.4799716472625732, "learning_rate": 9.88378523348904e-05, "loss": 0.4440103054046631, "memory(GiB)": 61.91, "step": 8025, "token_acc": 0.9120879120879121, "train_speed(iter/s)": 1.509965 }, { "epoch": 0.34402981877383143, "grad_norm": 1.81011164188385, "learning_rate": 9.883640936960716e-05, "loss": 0.2558621406555176, "memory(GiB)": 61.91, "step": 8030, "token_acc": 0.943609022556391, "train_speed(iter/s)": 1.50995 }, { "epoch": 0.3442440341030804, "grad_norm": 2.9986572265625, "learning_rate": 9.883496551960654e-05, "loss": 0.6187628746032715, "memory(GiB)": 61.91, "step": 8035, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.50998 }, { "epoch": 0.3444582494323294, "grad_norm": 6.73447322845459, "learning_rate": 9.88335207849147e-05, "loss": 0.49090075492858887, "memory(GiB)": 61.91, "step": 8040, "token_acc": 0.9162162162162162, "train_speed(iter/s)": 1.510016 }, { "epoch": 0.3446724647615783, "grad_norm": 1.6483649015426636, "learning_rate": 9.883207516555784e-05, "loss": 0.3171398639678955, "memory(GiB)": 61.91, "step": 8045, "token_acc": 0.9102902374670184, "train_speed(iter/s)": 1.510086 }, { "epoch": 0.3448866800908273, "grad_norm": 4.812012672424316, "learning_rate": 9.883062866156213e-05, "loss": 0.825739860534668, "memory(GiB)": 61.91, "step": 8050, "token_acc": 0.8264705882352941, "train_speed(iter/s)": 1.510217 }, { "epoch": 0.34510089542007627, "grad_norm": 2.5436317920684814, "learning_rate": 9.882918127295376e-05, "loss": 0.23584940433502197, "memory(GiB)": 61.91, "step": 8055, "token_acc": 0.9577922077922078, "train_speed(iter/s)": 1.51023 }, { "epoch": 0.34531511074932525, "grad_norm": 2.9797515869140625, "learning_rate": 9.882773299975897e-05, "loss": 0.258160924911499, "memory(GiB)": 61.91, "step": 8060, "token_acc": 0.9425675675675675, "train_speed(iter/s)": 1.510272 }, { "epoch": 0.3455293260785742, "grad_norm": 1.9484457969665527, "learning_rate": 9.8826283842004e-05, "loss": 0.29850473403930666, "memory(GiB)": 61.91, "step": 8065, "token_acc": 0.9312977099236641, "train_speed(iter/s)": 1.510248 }, { "epoch": 0.34574354140782315, "grad_norm": 0.6624013185501099, "learning_rate": 9.882483379971509e-05, "loss": 0.20448989868164064, "memory(GiB)": 61.91, "step": 8070, "token_acc": 0.9467084639498433, "train_speed(iter/s)": 1.510224 }, { "epoch": 0.34595775673707213, "grad_norm": 3.270261526107788, "learning_rate": 9.882338287291851e-05, "loss": 0.4455327033996582, "memory(GiB)": 61.91, "step": 8075, "token_acc": 0.889967637540453, "train_speed(iter/s)": 1.510207 }, { "epoch": 0.34617197206632105, "grad_norm": 1.2199444770812988, "learning_rate": 9.882193106164055e-05, "loss": 0.14718282222747803, "memory(GiB)": 61.91, "step": 8080, "token_acc": 0.9606299212598425, "train_speed(iter/s)": 1.510202 }, { "epoch": 0.34638618739557003, "grad_norm": 6.157723903656006, "learning_rate": 9.882047836590752e-05, "loss": 0.5213945388793946, "memory(GiB)": 61.91, "step": 8085, "token_acc": 0.8988326848249028, "train_speed(iter/s)": 1.510292 }, { "epoch": 0.346600402724819, "grad_norm": 1.4888267517089844, "learning_rate": 9.881902478574571e-05, "loss": 0.219823956489563, "memory(GiB)": 61.91, "step": 8090, "token_acc": 0.952076677316294, "train_speed(iter/s)": 1.510318 }, { "epoch": 0.34681461805406794, "grad_norm": 3.5085060596466064, "learning_rate": 9.88175703211815e-05, "loss": 0.4283764362335205, "memory(GiB)": 61.91, "step": 8095, "token_acc": 0.8996282527881041, "train_speed(iter/s)": 1.510275 }, { "epoch": 0.3470288333833169, "grad_norm": 2.08805251121521, "learning_rate": 9.88161149722412e-05, "loss": 0.4971491813659668, "memory(GiB)": 61.91, "step": 8100, "token_acc": 0.8903225806451613, "train_speed(iter/s)": 1.510266 }, { "epoch": 0.3472430487125659, "grad_norm": 1.02228581905365, "learning_rate": 9.881465873895116e-05, "loss": 0.34533441066741943, "memory(GiB)": 61.91, "step": 8105, "token_acc": 0.916, "train_speed(iter/s)": 1.510353 }, { "epoch": 0.3474572640418148, "grad_norm": 3.541059732437134, "learning_rate": 9.881320162133781e-05, "loss": 0.609687089920044, "memory(GiB)": 61.91, "step": 8110, "token_acc": 0.8532423208191127, "train_speed(iter/s)": 1.5103 }, { "epoch": 0.3476714793710638, "grad_norm": 3.559436559677124, "learning_rate": 9.881174361942751e-05, "loss": 0.4117569446563721, "memory(GiB)": 61.91, "step": 8115, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.510332 }, { "epoch": 0.3478856947003128, "grad_norm": 2.2064106464385986, "learning_rate": 9.881028473324669e-05, "loss": 0.4159360408782959, "memory(GiB)": 61.91, "step": 8120, "token_acc": 0.906832298136646, "train_speed(iter/s)": 1.51036 }, { "epoch": 0.3480999100295617, "grad_norm": 2.1873438358306885, "learning_rate": 9.880882496282176e-05, "loss": 0.24685986042022706, "memory(GiB)": 61.91, "step": 8125, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.510391 }, { "epoch": 0.3483141253588107, "grad_norm": 3.090034008026123, "learning_rate": 9.88073643081792e-05, "loss": 0.38810784816741944, "memory(GiB)": 61.91, "step": 8130, "token_acc": 0.9148148148148149, "train_speed(iter/s)": 1.510366 }, { "epoch": 0.34852834068805966, "grad_norm": 2.0415897369384766, "learning_rate": 9.880590276934543e-05, "loss": 0.3362722635269165, "memory(GiB)": 61.91, "step": 8135, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.510332 }, { "epoch": 0.3487425560173086, "grad_norm": 2.252575397491455, "learning_rate": 9.880444034634698e-05, "loss": 0.25062129497528074, "memory(GiB)": 61.91, "step": 8140, "token_acc": 0.9323076923076923, "train_speed(iter/s)": 1.51037 }, { "epoch": 0.34895677134655756, "grad_norm": 3.299809455871582, "learning_rate": 9.880297703921027e-05, "loss": 0.39873974323272704, "memory(GiB)": 61.91, "step": 8145, "token_acc": 0.8819188191881919, "train_speed(iter/s)": 1.510499 }, { "epoch": 0.34917098667580654, "grad_norm": 0.6998719573020935, "learning_rate": 9.880151284796187e-05, "loss": 0.43930683135986326, "memory(GiB)": 61.91, "step": 8150, "token_acc": 0.9046052631578947, "train_speed(iter/s)": 1.510582 }, { "epoch": 0.34938520200505546, "grad_norm": 1.9232062101364136, "learning_rate": 9.880004777262829e-05, "loss": 0.5072129726409912, "memory(GiB)": 61.91, "step": 8155, "token_acc": 0.9184952978056427, "train_speed(iter/s)": 1.510551 }, { "epoch": 0.34959941733430444, "grad_norm": 2.6348376274108887, "learning_rate": 9.879858181323607e-05, "loss": 0.45771026611328125, "memory(GiB)": 61.91, "step": 8160, "token_acc": 0.8862876254180602, "train_speed(iter/s)": 1.510615 }, { "epoch": 0.3498136326635534, "grad_norm": 5.187633991241455, "learning_rate": 9.879711496981174e-05, "loss": 0.27827725410461424, "memory(GiB)": 61.91, "step": 8165, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.510687 }, { "epoch": 0.35002784799280234, "grad_norm": 3.302506923675537, "learning_rate": 9.879564724238193e-05, "loss": 1.103024959564209, "memory(GiB)": 61.91, "step": 8170, "token_acc": 0.8239202657807309, "train_speed(iter/s)": 1.510789 }, { "epoch": 0.3502420633220513, "grad_norm": 3.8758699893951416, "learning_rate": 9.879417863097318e-05, "loss": 0.2220090389251709, "memory(GiB)": 61.91, "step": 8175, "token_acc": 0.9625, "train_speed(iter/s)": 1.51079 }, { "epoch": 0.3504562786513003, "grad_norm": 2.136594295501709, "learning_rate": 9.879270913561209e-05, "loss": 0.3327904224395752, "memory(GiB)": 61.91, "step": 8180, "token_acc": 0.9274447949526814, "train_speed(iter/s)": 1.510787 }, { "epoch": 0.3506704939805492, "grad_norm": 2.158339023590088, "learning_rate": 9.879123875632534e-05, "loss": 0.16064339876174927, "memory(GiB)": 61.91, "step": 8185, "token_acc": 0.9631901840490797, "train_speed(iter/s)": 1.510757 }, { "epoch": 0.3508847093097982, "grad_norm": 1.9888887405395508, "learning_rate": 9.878976749313951e-05, "loss": 0.4534247398376465, "memory(GiB)": 61.91, "step": 8190, "token_acc": 0.8867313915857605, "train_speed(iter/s)": 1.510675 }, { "epoch": 0.3510989246390472, "grad_norm": 3.444143533706665, "learning_rate": 9.878829534608127e-05, "loss": 0.46887855529785155, "memory(GiB)": 61.91, "step": 8195, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.510635 }, { "epoch": 0.3513131399682961, "grad_norm": 2.3185007572174072, "learning_rate": 9.878682231517731e-05, "loss": 0.4610457420349121, "memory(GiB)": 61.91, "step": 8200, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.510627 }, { "epoch": 0.3515273552975451, "grad_norm": 1.0078370571136475, "learning_rate": 9.878534840045428e-05, "loss": 0.3914927959442139, "memory(GiB)": 61.91, "step": 8205, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.510606 }, { "epoch": 0.35174157062679406, "grad_norm": 4.583042621612549, "learning_rate": 9.878387360193891e-05, "loss": 0.3980301856994629, "memory(GiB)": 61.91, "step": 8210, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.510649 }, { "epoch": 0.351955785956043, "grad_norm": 1.251065731048584, "learning_rate": 9.87823979196579e-05, "loss": 0.4213674068450928, "memory(GiB)": 61.91, "step": 8215, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.510732 }, { "epoch": 0.35217000128529197, "grad_norm": 0.09758938103914261, "learning_rate": 9.8780921353638e-05, "loss": 0.23020036220550538, "memory(GiB)": 61.91, "step": 8220, "token_acc": 0.9339622641509434, "train_speed(iter/s)": 1.51076 }, { "epoch": 0.35238421661454095, "grad_norm": 1.883832335472107, "learning_rate": 9.877944390390594e-05, "loss": 0.34256079196929934, "memory(GiB)": 61.91, "step": 8225, "token_acc": 0.939873417721519, "train_speed(iter/s)": 1.510751 }, { "epoch": 0.3525984319437899, "grad_norm": 2.5986592769622803, "learning_rate": 9.87779655704885e-05, "loss": 0.5016443729400635, "memory(GiB)": 61.91, "step": 8230, "token_acc": 0.8869565217391304, "train_speed(iter/s)": 1.51076 }, { "epoch": 0.35281264727303885, "grad_norm": 2.5744216442108154, "learning_rate": 9.877648635341245e-05, "loss": 0.39737393856048586, "memory(GiB)": 61.91, "step": 8235, "token_acc": 0.9, "train_speed(iter/s)": 1.510754 }, { "epoch": 0.3530268626022878, "grad_norm": 2.31666898727417, "learning_rate": 9.877500625270459e-05, "loss": 0.2819164752960205, "memory(GiB)": 61.91, "step": 8240, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.510727 }, { "epoch": 0.3532410779315368, "grad_norm": 3.0742480754852295, "learning_rate": 9.877352526839174e-05, "loss": 0.4782444953918457, "memory(GiB)": 61.91, "step": 8245, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.51068 }, { "epoch": 0.35345529326078573, "grad_norm": 2.3592817783355713, "learning_rate": 9.877204340050075e-05, "loss": 0.3943544864654541, "memory(GiB)": 61.91, "step": 8250, "token_acc": 0.9067524115755627, "train_speed(iter/s)": 1.510723 }, { "epoch": 0.3536695085900347, "grad_norm": 7.544619083404541, "learning_rate": 9.87705606490584e-05, "loss": 0.5255274295806884, "memory(GiB)": 61.91, "step": 8255, "token_acc": 0.8753623188405797, "train_speed(iter/s)": 1.510766 }, { "epoch": 0.3538837239192837, "grad_norm": 2.65968656539917, "learning_rate": 9.876907701409164e-05, "loss": 0.5254755973815918, "memory(GiB)": 61.91, "step": 8260, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.510797 }, { "epoch": 0.3540979392485326, "grad_norm": 4.306763172149658, "learning_rate": 9.876759249562727e-05, "loss": 0.9895835876464844, "memory(GiB)": 61.91, "step": 8265, "token_acc": 0.8, "train_speed(iter/s)": 1.510752 }, { "epoch": 0.3543121545777816, "grad_norm": 5.25244140625, "learning_rate": 9.876610709369221e-05, "loss": 0.6747769355773926, "memory(GiB)": 61.91, "step": 8270, "token_acc": 0.8655737704918033, "train_speed(iter/s)": 1.510732 }, { "epoch": 0.35452636990703057, "grad_norm": 3.238536834716797, "learning_rate": 9.876462080831338e-05, "loss": 0.4111518859863281, "memory(GiB)": 61.91, "step": 8275, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.510708 }, { "epoch": 0.3547405852362795, "grad_norm": 1.3382694721221924, "learning_rate": 9.876313363951772e-05, "loss": 0.17621071338653566, "memory(GiB)": 61.91, "step": 8280, "token_acc": 0.9612676056338029, "train_speed(iter/s)": 1.510735 }, { "epoch": 0.35495480056552847, "grad_norm": 2.601794958114624, "learning_rate": 9.876164558733213e-05, "loss": 0.3946081638336182, "memory(GiB)": 61.91, "step": 8285, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.510716 }, { "epoch": 0.35516901589477745, "grad_norm": 2.5786194801330566, "learning_rate": 9.87601566517836e-05, "loss": 0.33294243812561036, "memory(GiB)": 61.91, "step": 8290, "token_acc": 0.9246376811594202, "train_speed(iter/s)": 1.510709 }, { "epoch": 0.3553832312240264, "grad_norm": 2.289860725402832, "learning_rate": 9.875866683289907e-05, "loss": 0.3189689159393311, "memory(GiB)": 61.91, "step": 8295, "token_acc": 0.926509186351706, "train_speed(iter/s)": 1.51064 }, { "epoch": 0.35559744655327535, "grad_norm": 2.7072980403900146, "learning_rate": 9.875717613070558e-05, "loss": 0.34669952392578124, "memory(GiB)": 61.91, "step": 8300, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.510673 }, { "epoch": 0.35581166188252433, "grad_norm": 1.0098376274108887, "learning_rate": 9.875568454523008e-05, "loss": 0.281189489364624, "memory(GiB)": 61.91, "step": 8305, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.510731 }, { "epoch": 0.35602587721177326, "grad_norm": 5.978972911834717, "learning_rate": 9.875419207649963e-05, "loss": 0.3442498207092285, "memory(GiB)": 61.91, "step": 8310, "token_acc": 0.9157088122605364, "train_speed(iter/s)": 1.510722 }, { "epoch": 0.35624009254102224, "grad_norm": 3.1284725666046143, "learning_rate": 9.875269872454127e-05, "loss": 0.8159701347351074, "memory(GiB)": 61.91, "step": 8315, "token_acc": 0.8440677966101695, "train_speed(iter/s)": 1.510873 }, { "epoch": 0.3564543078702712, "grad_norm": 1.1218222379684448, "learning_rate": 9.875120448938201e-05, "loss": 0.3257554054260254, "memory(GiB)": 61.91, "step": 8320, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.51088 }, { "epoch": 0.35666852319952014, "grad_norm": 1.0046412944793701, "learning_rate": 9.874970937104897e-05, "loss": 0.4218743801116943, "memory(GiB)": 61.91, "step": 8325, "token_acc": 0.916083916083916, "train_speed(iter/s)": 1.510883 }, { "epoch": 0.3568827385287691, "grad_norm": 3.202280282974243, "learning_rate": 9.87482133695692e-05, "loss": 0.6752405166625977, "memory(GiB)": 61.91, "step": 8330, "token_acc": 0.8395061728395061, "train_speed(iter/s)": 1.510925 }, { "epoch": 0.3570969538580181, "grad_norm": 2.449254035949707, "learning_rate": 9.874671648496983e-05, "loss": 0.3161851167678833, "memory(GiB)": 61.91, "step": 8335, "token_acc": 0.9226006191950464, "train_speed(iter/s)": 1.510896 }, { "epoch": 0.357311169187267, "grad_norm": 2.987575054168701, "learning_rate": 9.874521871727795e-05, "loss": 0.2683391094207764, "memory(GiB)": 61.91, "step": 8340, "token_acc": 0.9383116883116883, "train_speed(iter/s)": 1.510861 }, { "epoch": 0.357525384516516, "grad_norm": 6.7677836418151855, "learning_rate": 9.87437200665207e-05, "loss": 0.539757776260376, "memory(GiB)": 61.91, "step": 8345, "token_acc": 0.8681948424068768, "train_speed(iter/s)": 1.510809 }, { "epoch": 0.357739599845765, "grad_norm": 3.3542592525482178, "learning_rate": 9.874222053272526e-05, "loss": 0.47335238456726075, "memory(GiB)": 61.91, "step": 8350, "token_acc": 0.9042904290429042, "train_speed(iter/s)": 1.510776 }, { "epoch": 0.3579538151750139, "grad_norm": 2.0968735218048096, "learning_rate": 9.874072011591875e-05, "loss": 0.3348551750183105, "memory(GiB)": 61.91, "step": 8355, "token_acc": 0.9182389937106918, "train_speed(iter/s)": 1.510868 }, { "epoch": 0.3581680305042629, "grad_norm": 2.8707685470581055, "learning_rate": 9.873921881612839e-05, "loss": 0.22373542785644532, "memory(GiB)": 61.91, "step": 8360, "token_acc": 0.9596412556053812, "train_speed(iter/s)": 1.51087 }, { "epoch": 0.35838224583351186, "grad_norm": 2.783865451812744, "learning_rate": 9.873771663338134e-05, "loss": 0.23436853885650635, "memory(GiB)": 61.91, "step": 8365, "token_acc": 0.9488054607508533, "train_speed(iter/s)": 1.510881 }, { "epoch": 0.3585964611627608, "grad_norm": 5.039813995361328, "learning_rate": 9.873621356770485e-05, "loss": 0.37659153938293455, "memory(GiB)": 61.91, "step": 8370, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.510836 }, { "epoch": 0.35881067649200976, "grad_norm": 4.267390251159668, "learning_rate": 9.873470961912612e-05, "loss": 0.10687878131866455, "memory(GiB)": 61.91, "step": 8375, "token_acc": 0.9703703703703703, "train_speed(iter/s)": 1.510858 }, { "epoch": 0.35902489182125874, "grad_norm": 1.7893646955490112, "learning_rate": 9.873320478767242e-05, "loss": 0.30151369571685793, "memory(GiB)": 61.91, "step": 8380, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.511002 }, { "epoch": 0.35923910715050766, "grad_norm": 3.6432948112487793, "learning_rate": 9.873169907337098e-05, "loss": 0.35772995948791503, "memory(GiB)": 61.91, "step": 8385, "token_acc": 0.9385245901639344, "train_speed(iter/s)": 1.510989 }, { "epoch": 0.35945332247975664, "grad_norm": 6.112993240356445, "learning_rate": 9.873019247624913e-05, "loss": 0.5271028041839599, "memory(GiB)": 61.91, "step": 8390, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.510975 }, { "epoch": 0.3596675378090056, "grad_norm": 2.8249731063842773, "learning_rate": 9.87286849963341e-05, "loss": 0.42355051040649416, "memory(GiB)": 61.91, "step": 8395, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.511035 }, { "epoch": 0.3598817531382546, "grad_norm": 9.644018173217773, "learning_rate": 9.872717663365325e-05, "loss": 0.8424102783203125, "memory(GiB)": 61.91, "step": 8400, "token_acc": 0.8312883435582822, "train_speed(iter/s)": 1.511131 }, { "epoch": 0.3600959684675035, "grad_norm": 3.3169281482696533, "learning_rate": 9.872566738823388e-05, "loss": 0.47966642379760743, "memory(GiB)": 61.91, "step": 8405, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.511244 }, { "epoch": 0.3603101837967525, "grad_norm": 4.327241897583008, "learning_rate": 9.872415726010334e-05, "loss": 0.5246102809906006, "memory(GiB)": 61.91, "step": 8410, "token_acc": 0.8801261829652997, "train_speed(iter/s)": 1.511211 }, { "epoch": 0.3605243991260015, "grad_norm": 3.983896255493164, "learning_rate": 9.872264624928898e-05, "loss": 0.27153425216674804, "memory(GiB)": 61.91, "step": 8415, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.511219 }, { "epoch": 0.3607386144552504, "grad_norm": 0.9352177381515503, "learning_rate": 9.872113435581818e-05, "loss": 0.4178792476654053, "memory(GiB)": 61.91, "step": 8420, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.511189 }, { "epoch": 0.3609528297844994, "grad_norm": 3.2495973110198975, "learning_rate": 9.871962157971832e-05, "loss": 0.5291205883026123, "memory(GiB)": 61.91, "step": 8425, "token_acc": 0.887240356083086, "train_speed(iter/s)": 1.511153 }, { "epoch": 0.36116704511374836, "grad_norm": 2.9210622310638428, "learning_rate": 9.871810792101681e-05, "loss": 0.6802117824554443, "memory(GiB)": 61.91, "step": 8430, "token_acc": 0.875, "train_speed(iter/s)": 1.511269 }, { "epoch": 0.3613812604429973, "grad_norm": 2.3755571842193604, "learning_rate": 9.871659337974109e-05, "loss": 0.5982009410858155, "memory(GiB)": 61.91, "step": 8435, "token_acc": 0.8973509933774835, "train_speed(iter/s)": 1.511261 }, { "epoch": 0.36159547577224627, "grad_norm": 1.5956039428710938, "learning_rate": 9.871507795591857e-05, "loss": 0.3831546068191528, "memory(GiB)": 61.91, "step": 8440, "token_acc": 0.9135802469135802, "train_speed(iter/s)": 1.511235 }, { "epoch": 0.36180969110149525, "grad_norm": 3.2444026470184326, "learning_rate": 9.87135616495767e-05, "loss": 0.48974223136901857, "memory(GiB)": 61.91, "step": 8445, "token_acc": 0.901060070671378, "train_speed(iter/s)": 1.511247 }, { "epoch": 0.36202390643074417, "grad_norm": 3.5096280574798584, "learning_rate": 9.871204446074298e-05, "loss": 0.5237321853637695, "memory(GiB)": 61.91, "step": 8450, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.51124 }, { "epoch": 0.36223812175999315, "grad_norm": 4.675867557525635, "learning_rate": 9.871052638944489e-05, "loss": 0.44454364776611327, "memory(GiB)": 61.91, "step": 8455, "token_acc": 0.8754448398576512, "train_speed(iter/s)": 1.51122 }, { "epoch": 0.3624523370892421, "grad_norm": 3.3041279315948486, "learning_rate": 9.87090074357099e-05, "loss": 0.4129662036895752, "memory(GiB)": 61.91, "step": 8460, "token_acc": 0.910958904109589, "train_speed(iter/s)": 1.511396 }, { "epoch": 0.36266655241849105, "grad_norm": 1.6807105541229248, "learning_rate": 9.870748759956556e-05, "loss": 0.18156312704086303, "memory(GiB)": 61.91, "step": 8465, "token_acc": 0.9440559440559441, "train_speed(iter/s)": 1.511466 }, { "epoch": 0.36288076774774003, "grad_norm": 2.544825553894043, "learning_rate": 9.87059668810394e-05, "loss": 0.5279191970825196, "memory(GiB)": 61.91, "step": 8470, "token_acc": 0.8896797153024911, "train_speed(iter/s)": 1.511453 }, { "epoch": 0.363094983076989, "grad_norm": 3.9574735164642334, "learning_rate": 9.870444528015895e-05, "loss": 0.3346928834915161, "memory(GiB)": 61.91, "step": 8475, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.511495 }, { "epoch": 0.36330919840623793, "grad_norm": 3.2448995113372803, "learning_rate": 9.870292279695177e-05, "loss": 0.4908749580383301, "memory(GiB)": 61.91, "step": 8480, "token_acc": 0.8710801393728222, "train_speed(iter/s)": 1.511579 }, { "epoch": 0.3635234137354869, "grad_norm": 3.2215542793273926, "learning_rate": 9.870139943144547e-05, "loss": 0.5620937824249268, "memory(GiB)": 61.91, "step": 8485, "token_acc": 0.8802588996763754, "train_speed(iter/s)": 1.511511 }, { "epoch": 0.3637376290647359, "grad_norm": 1.8716936111450195, "learning_rate": 9.869987518366763e-05, "loss": 0.3343369722366333, "memory(GiB)": 61.91, "step": 8490, "token_acc": 0.9390243902439024, "train_speed(iter/s)": 1.511486 }, { "epoch": 0.3639518443939848, "grad_norm": 4.0952229499816895, "learning_rate": 9.869835005364587e-05, "loss": 0.8576143264770508, "memory(GiB)": 61.91, "step": 8495, "token_acc": 0.7667731629392971, "train_speed(iter/s)": 1.511549 }, { "epoch": 0.3641660597232338, "grad_norm": 1.5087676048278809, "learning_rate": 9.869682404140781e-05, "loss": 0.2970644474029541, "memory(GiB)": 61.91, "step": 8500, "token_acc": 0.9326599326599326, "train_speed(iter/s)": 1.511607 }, { "epoch": 0.3641660597232338, "eval_loss": 2.5182292461395264, "eval_runtime": 11.5797, "eval_samples_per_second": 8.636, "eval_steps_per_second": 8.636, "eval_token_acc": 0.4419525065963061, "step": 8500 }, { "epoch": 0.36438027505248277, "grad_norm": 2.8486454486846924, "learning_rate": 9.869529714698111e-05, "loss": 0.49498896598815917, "memory(GiB)": 61.91, "step": 8505, "token_acc": 0.5596153846153846, "train_speed(iter/s)": 1.508319 }, { "epoch": 0.3645944903817317, "grad_norm": 0.8331350684165955, "learning_rate": 9.869376937039342e-05, "loss": 0.5306161403656006, "memory(GiB)": 61.91, "step": 8510, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.508284 }, { "epoch": 0.3648087057109807, "grad_norm": 2.1671202182769775, "learning_rate": 9.869224071167242e-05, "loss": 0.36143980026245115, "memory(GiB)": 61.91, "step": 8515, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.508288 }, { "epoch": 0.36502292104022965, "grad_norm": 1.772369146347046, "learning_rate": 9.869071117084581e-05, "loss": 0.23359076976776122, "memory(GiB)": 61.91, "step": 8520, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.508227 }, { "epoch": 0.3652371363694786, "grad_norm": 3.719489812850952, "learning_rate": 9.868918074794126e-05, "loss": 0.5181020259857178, "memory(GiB)": 61.91, "step": 8525, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.508372 }, { "epoch": 0.36545135169872756, "grad_norm": 2.3016295433044434, "learning_rate": 9.868764944298656e-05, "loss": 0.6089282989501953, "memory(GiB)": 61.91, "step": 8530, "token_acc": 0.8633093525179856, "train_speed(iter/s)": 1.508318 }, { "epoch": 0.36566556702797653, "grad_norm": 3.6251890659332275, "learning_rate": 9.86861172560094e-05, "loss": 0.5702113628387451, "memory(GiB)": 61.91, "step": 8535, "token_acc": 0.9022801302931596, "train_speed(iter/s)": 1.508319 }, { "epoch": 0.36587978235722546, "grad_norm": 3.182731866836548, "learning_rate": 9.868458418703756e-05, "loss": 0.5166587352752685, "memory(GiB)": 61.91, "step": 8540, "token_acc": 0.8778135048231511, "train_speed(iter/s)": 1.508293 }, { "epoch": 0.36609399768647444, "grad_norm": 2.6207497119903564, "learning_rate": 9.868305023609881e-05, "loss": 0.19877753257751465, "memory(GiB)": 61.91, "step": 8545, "token_acc": 0.9423868312757202, "train_speed(iter/s)": 1.508357 }, { "epoch": 0.3663082130157234, "grad_norm": 3.0303564071655273, "learning_rate": 9.868151540322094e-05, "loss": 0.6614083766937255, "memory(GiB)": 61.91, "step": 8550, "token_acc": 0.8714285714285714, "train_speed(iter/s)": 1.508362 }, { "epoch": 0.36652242834497234, "grad_norm": 1.4050486087799072, "learning_rate": 9.867997968843175e-05, "loss": 0.36556806564331057, "memory(GiB)": 61.91, "step": 8555, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.508454 }, { "epoch": 0.3667366436742213, "grad_norm": 0.5342100262641907, "learning_rate": 9.867844309175906e-05, "loss": 0.3324640989303589, "memory(GiB)": 61.91, "step": 8560, "token_acc": 0.9171270718232044, "train_speed(iter/s)": 1.508449 }, { "epoch": 0.3669508590034703, "grad_norm": 3.052577018737793, "learning_rate": 9.86769056132307e-05, "loss": 0.19841744899749755, "memory(GiB)": 61.91, "step": 8565, "token_acc": 0.9547038327526133, "train_speed(iter/s)": 1.508365 }, { "epoch": 0.3671650743327193, "grad_norm": 3.786212921142578, "learning_rate": 9.867536725287455e-05, "loss": 0.4685386657714844, "memory(GiB)": 61.91, "step": 8570, "token_acc": 0.8697183098591549, "train_speed(iter/s)": 1.508343 }, { "epoch": 0.3673792896619682, "grad_norm": 1.6783490180969238, "learning_rate": 9.867382801071844e-05, "loss": 0.28863441944122314, "memory(GiB)": 61.91, "step": 8575, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.508382 }, { "epoch": 0.3675935049912172, "grad_norm": 2.4865825176239014, "learning_rate": 9.86722878867903e-05, "loss": 0.30562381744384765, "memory(GiB)": 61.91, "step": 8580, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.508383 }, { "epoch": 0.36780772032046616, "grad_norm": 4.11130952835083, "learning_rate": 9.867074688111799e-05, "loss": 0.41965465545654296, "memory(GiB)": 61.91, "step": 8585, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.508407 }, { "epoch": 0.3680219356497151, "grad_norm": 2.2702999114990234, "learning_rate": 9.866920499372944e-05, "loss": 0.3968060493469238, "memory(GiB)": 61.91, "step": 8590, "token_acc": 0.9110320284697508, "train_speed(iter/s)": 1.508435 }, { "epoch": 0.36823615097896406, "grad_norm": 3.3359034061431885, "learning_rate": 9.86676622246526e-05, "loss": 0.49109683036804197, "memory(GiB)": 61.91, "step": 8595, "token_acc": 0.9094827586206896, "train_speed(iter/s)": 1.508482 }, { "epoch": 0.36845036630821304, "grad_norm": 4.422955513000488, "learning_rate": 9.866611857391541e-05, "loss": 0.32813491821289065, "memory(GiB)": 61.91, "step": 8600, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.508479 }, { "epoch": 0.36866458163746196, "grad_norm": 0.7508419752120972, "learning_rate": 9.866457404154581e-05, "loss": 0.5170837879180908, "memory(GiB)": 61.91, "step": 8605, "token_acc": 0.89, "train_speed(iter/s)": 1.508564 }, { "epoch": 0.36887879696671094, "grad_norm": 3.0916597843170166, "learning_rate": 9.86630286275718e-05, "loss": 0.4983029842376709, "memory(GiB)": 61.91, "step": 8610, "token_acc": 0.8885448916408669, "train_speed(iter/s)": 1.508589 }, { "epoch": 0.3690930122959599, "grad_norm": 2.3805835247039795, "learning_rate": 9.866148233202139e-05, "loss": 0.33534576892852785, "memory(GiB)": 61.91, "step": 8615, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.508667 }, { "epoch": 0.36930722762520884, "grad_norm": 4.513883590698242, "learning_rate": 9.865993515492258e-05, "loss": 0.5090461730957031, "memory(GiB)": 61.91, "step": 8620, "token_acc": 0.8940809968847352, "train_speed(iter/s)": 1.508649 }, { "epoch": 0.3695214429544578, "grad_norm": 4.873726844787598, "learning_rate": 9.865838709630339e-05, "loss": 0.5730202674865723, "memory(GiB)": 61.91, "step": 8625, "token_acc": 0.8971428571428571, "train_speed(iter/s)": 1.508673 }, { "epoch": 0.3697356582837068, "grad_norm": 2.3736422061920166, "learning_rate": 9.865683815619188e-05, "loss": 0.4917450904846191, "memory(GiB)": 61.91, "step": 8630, "token_acc": 0.9067524115755627, "train_speed(iter/s)": 1.508748 }, { "epoch": 0.3699498736129557, "grad_norm": 1.5080140829086304, "learning_rate": 9.865528833461611e-05, "loss": 0.4045114517211914, "memory(GiB)": 61.91, "step": 8635, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.50884 }, { "epoch": 0.3701640889422047, "grad_norm": 4.8142805099487305, "learning_rate": 9.865373763160413e-05, "loss": 0.5263293743133545, "memory(GiB)": 61.91, "step": 8640, "token_acc": 0.8906752411575563, "train_speed(iter/s)": 1.508905 }, { "epoch": 0.3703783042714537, "grad_norm": 3.0230648517608643, "learning_rate": 9.865218604718405e-05, "loss": 0.22323050498962402, "memory(GiB)": 61.91, "step": 8645, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.508954 }, { "epoch": 0.3705925196007026, "grad_norm": 1.6971248388290405, "learning_rate": 9.8650633581384e-05, "loss": 0.2324012041091919, "memory(GiB)": 61.91, "step": 8650, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.508954 }, { "epoch": 0.3708067349299516, "grad_norm": 2.856060266494751, "learning_rate": 9.864908023423207e-05, "loss": 0.39144277572631836, "memory(GiB)": 61.91, "step": 8655, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.509068 }, { "epoch": 0.37102095025920057, "grad_norm": 3.013428211212158, "learning_rate": 9.864752600575641e-05, "loss": 0.336905837059021, "memory(GiB)": 61.91, "step": 8660, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.509109 }, { "epoch": 0.3712351655884495, "grad_norm": 8.562774658203125, "learning_rate": 9.864597089598519e-05, "loss": 0.5130837917327881, "memory(GiB)": 61.91, "step": 8665, "token_acc": 0.8963210702341137, "train_speed(iter/s)": 1.509015 }, { "epoch": 0.37144938091769847, "grad_norm": 3.221346855163574, "learning_rate": 9.864441490494658e-05, "loss": 0.5477745056152343, "memory(GiB)": 61.91, "step": 8670, "token_acc": 0.8867313915857605, "train_speed(iter/s)": 1.508979 }, { "epoch": 0.37166359624694745, "grad_norm": 3.490896463394165, "learning_rate": 9.864285803266876e-05, "loss": 0.2877952098846436, "memory(GiB)": 61.91, "step": 8675, "token_acc": 0.9226006191950464, "train_speed(iter/s)": 1.508971 }, { "epoch": 0.37187781157619637, "grad_norm": 4.183293342590332, "learning_rate": 9.864130027917993e-05, "loss": 0.3909464359283447, "memory(GiB)": 61.91, "step": 8680, "token_acc": 0.9171779141104295, "train_speed(iter/s)": 1.508984 }, { "epoch": 0.37209202690544535, "grad_norm": 1.3292711973190308, "learning_rate": 9.863974164450833e-05, "loss": 0.13097889423370362, "memory(GiB)": 61.91, "step": 8685, "token_acc": 0.9754385964912281, "train_speed(iter/s)": 1.508967 }, { "epoch": 0.37230624223469433, "grad_norm": 2.6631674766540527, "learning_rate": 9.863818212868217e-05, "loss": 0.6474672317504883, "memory(GiB)": 61.91, "step": 8690, "token_acc": 0.8528528528528528, "train_speed(iter/s)": 1.508984 }, { "epoch": 0.37252045756394325, "grad_norm": 4.954692363739014, "learning_rate": 9.863662173172971e-05, "loss": 0.5688843727111816, "memory(GiB)": 61.91, "step": 8695, "token_acc": 0.8545454545454545, "train_speed(iter/s)": 1.509054 }, { "epoch": 0.37273467289319223, "grad_norm": 3.2487399578094482, "learning_rate": 9.863506045367923e-05, "loss": 0.4373831272125244, "memory(GiB)": 61.91, "step": 8700, "token_acc": 0.8905109489051095, "train_speed(iter/s)": 1.509096 }, { "epoch": 0.3729488882224412, "grad_norm": 6.788768291473389, "learning_rate": 9.863349829455899e-05, "loss": 0.34687998294830324, "memory(GiB)": 61.91, "step": 8705, "token_acc": 0.9187279151943463, "train_speed(iter/s)": 1.509093 }, { "epoch": 0.37316310355169013, "grad_norm": 3.353257656097412, "learning_rate": 9.863193525439734e-05, "loss": 0.4005410194396973, "memory(GiB)": 61.91, "step": 8710, "token_acc": 0.8961038961038961, "train_speed(iter/s)": 1.509095 }, { "epoch": 0.3733773188809391, "grad_norm": 3.125143527984619, "learning_rate": 9.863037133322252e-05, "loss": 0.4396620273590088, "memory(GiB)": 61.91, "step": 8715, "token_acc": 0.9065743944636678, "train_speed(iter/s)": 1.509062 }, { "epoch": 0.3735915342101881, "grad_norm": 1.795769453048706, "learning_rate": 9.862880653106294e-05, "loss": 0.3604635238647461, "memory(GiB)": 61.91, "step": 8720, "token_acc": 0.8884758364312267, "train_speed(iter/s)": 1.509115 }, { "epoch": 0.373805749539437, "grad_norm": 0.6067296266555786, "learning_rate": 9.86272408479469e-05, "loss": 0.546636962890625, "memory(GiB)": 61.91, "step": 8725, "token_acc": 0.8633093525179856, "train_speed(iter/s)": 1.509112 }, { "epoch": 0.374019964868686, "grad_norm": 0.27054882049560547, "learning_rate": 9.862567428390277e-05, "loss": 0.4152828693389893, "memory(GiB)": 61.91, "step": 8730, "token_acc": 0.9022082018927445, "train_speed(iter/s)": 1.509045 }, { "epoch": 0.374234180197935, "grad_norm": 1.43099844455719, "learning_rate": 9.862410683895895e-05, "loss": 0.24917030334472656, "memory(GiB)": 61.91, "step": 8735, "token_acc": 0.9323943661971831, "train_speed(iter/s)": 1.509083 }, { "epoch": 0.37444839552718395, "grad_norm": 3.4434139728546143, "learning_rate": 9.86225385131438e-05, "loss": 0.37369704246520996, "memory(GiB)": 61.91, "step": 8740, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.50919 }, { "epoch": 0.3746626108564329, "grad_norm": 1.5163078308105469, "learning_rate": 9.862096930648577e-05, "loss": 0.7606154918670655, "memory(GiB)": 61.91, "step": 8745, "token_acc": 0.7806451612903226, "train_speed(iter/s)": 1.50911 }, { "epoch": 0.37487682618568186, "grad_norm": 3.1859614849090576, "learning_rate": 9.861939921901326e-05, "loss": 0.6229585647583008, "memory(GiB)": 61.91, "step": 8750, "token_acc": 0.8914728682170543, "train_speed(iter/s)": 1.509094 }, { "epoch": 0.37509104151493083, "grad_norm": 2.189007520675659, "learning_rate": 9.861782825075475e-05, "loss": 0.4816583633422852, "memory(GiB)": 61.91, "step": 8755, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.509191 }, { "epoch": 0.37530525684417976, "grad_norm": 2.8951022624969482, "learning_rate": 9.861625640173865e-05, "loss": 0.3797547101974487, "memory(GiB)": 61.91, "step": 8760, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.509255 }, { "epoch": 0.37551947217342874, "grad_norm": 3.9653022289276123, "learning_rate": 9.861468367199346e-05, "loss": 0.504998779296875, "memory(GiB)": 61.91, "step": 8765, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.509315 }, { "epoch": 0.3757336875026777, "grad_norm": 1.3600616455078125, "learning_rate": 9.861311006154767e-05, "loss": 0.4426393985748291, "memory(GiB)": 61.91, "step": 8770, "token_acc": 0.9054054054054054, "train_speed(iter/s)": 1.509286 }, { "epoch": 0.37594790283192664, "grad_norm": 0.5868422985076904, "learning_rate": 9.86115355704298e-05, "loss": 0.3044178247451782, "memory(GiB)": 61.91, "step": 8775, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.509279 }, { "epoch": 0.3761621181611756, "grad_norm": 0.8829467296600342, "learning_rate": 9.860996019866836e-05, "loss": 0.6389955043792724, "memory(GiB)": 61.91, "step": 8780, "token_acc": 0.8696883852691218, "train_speed(iter/s)": 1.509371 }, { "epoch": 0.3763763334904246, "grad_norm": 2.661935567855835, "learning_rate": 9.860838394629188e-05, "loss": 0.4936662197113037, "memory(GiB)": 61.91, "step": 8785, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.509398 }, { "epoch": 0.3765905488196735, "grad_norm": 5.699188232421875, "learning_rate": 9.860680681332894e-05, "loss": 1.0407577514648438, "memory(GiB)": 61.91, "step": 8790, "token_acc": 0.7862068965517242, "train_speed(iter/s)": 1.509432 }, { "epoch": 0.3768047641489225, "grad_norm": 2.2169668674468994, "learning_rate": 9.860522879980809e-05, "loss": 0.49826393127441404, "memory(GiB)": 61.91, "step": 8795, "token_acc": 0.9121338912133892, "train_speed(iter/s)": 1.509482 }, { "epoch": 0.3770189794781715, "grad_norm": 2.9898324012756348, "learning_rate": 9.860364990575792e-05, "loss": 0.22864351272583008, "memory(GiB)": 61.91, "step": 8800, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.509478 }, { "epoch": 0.3772331948074204, "grad_norm": 7.7009406089782715, "learning_rate": 9.860207013120706e-05, "loss": 0.3538163423538208, "memory(GiB)": 61.91, "step": 8805, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.509428 }, { "epoch": 0.3774474101366694, "grad_norm": 1.1530338525772095, "learning_rate": 9.860048947618408e-05, "loss": 0.3381653308868408, "memory(GiB)": 61.91, "step": 8810, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.509492 }, { "epoch": 0.37766162546591836, "grad_norm": 3.4894630908966064, "learning_rate": 9.859890794071767e-05, "loss": 0.7393455982208252, "memory(GiB)": 61.91, "step": 8815, "token_acc": 0.8550185873605948, "train_speed(iter/s)": 1.509594 }, { "epoch": 0.3778758407951673, "grad_norm": 2.5774238109588623, "learning_rate": 9.859732552483642e-05, "loss": 0.33259222507476804, "memory(GiB)": 61.91, "step": 8820, "token_acc": 0.9404388714733543, "train_speed(iter/s)": 1.50965 }, { "epoch": 0.37809005612441626, "grad_norm": 2.599008798599243, "learning_rate": 9.859574222856905e-05, "loss": 0.6008551597595215, "memory(GiB)": 61.91, "step": 8825, "token_acc": 0.8633720930232558, "train_speed(iter/s)": 1.509661 }, { "epoch": 0.37830427145366524, "grad_norm": 6.7987895011901855, "learning_rate": 9.859415805194422e-05, "loss": 0.5541271209716797, "memory(GiB)": 61.91, "step": 8830, "token_acc": 0.8909774436090225, "train_speed(iter/s)": 1.509744 }, { "epoch": 0.37851848678291417, "grad_norm": 2.970623731613159, "learning_rate": 9.859257299499064e-05, "loss": 0.5560859680175781, "memory(GiB)": 61.91, "step": 8835, "token_acc": 0.8996539792387543, "train_speed(iter/s)": 1.50981 }, { "epoch": 0.37873270211216314, "grad_norm": 3.2410740852355957, "learning_rate": 9.859098705773701e-05, "loss": 0.7154887199401856, "memory(GiB)": 61.91, "step": 8840, "token_acc": 0.865814696485623, "train_speed(iter/s)": 1.509869 }, { "epoch": 0.3789469174414121, "grad_norm": 3.1667914390563965, "learning_rate": 9.858940024021205e-05, "loss": 0.6933043479919434, "memory(GiB)": 61.91, "step": 8845, "token_acc": 0.8764478764478765, "train_speed(iter/s)": 1.510022 }, { "epoch": 0.37916113277066105, "grad_norm": 2.7921695709228516, "learning_rate": 9.858781254244455e-05, "loss": 0.4047614574432373, "memory(GiB)": 61.91, "step": 8850, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.510042 }, { "epoch": 0.37937534809991, "grad_norm": 2.2751972675323486, "learning_rate": 9.858622396446325e-05, "loss": 0.3717304229736328, "memory(GiB)": 61.91, "step": 8855, "token_acc": 0.9226190476190477, "train_speed(iter/s)": 1.510011 }, { "epoch": 0.379589563429159, "grad_norm": 5.262077808380127, "learning_rate": 9.85846345062969e-05, "loss": 0.36046175956726073, "memory(GiB)": 61.91, "step": 8860, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.510122 }, { "epoch": 0.37980377875840793, "grad_norm": 3.5558032989501953, "learning_rate": 9.858304416797433e-05, "loss": 0.33611431121826174, "memory(GiB)": 61.91, "step": 8865, "token_acc": 0.925531914893617, "train_speed(iter/s)": 1.510117 }, { "epoch": 0.3800179940876569, "grad_norm": 2.180654764175415, "learning_rate": 9.858145294952435e-05, "loss": 0.3827423095703125, "memory(GiB)": 61.91, "step": 8870, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.510101 }, { "epoch": 0.3802322094169059, "grad_norm": 13.460418701171875, "learning_rate": 9.857986085097577e-05, "loss": 0.3861852645874023, "memory(GiB)": 61.91, "step": 8875, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.510144 }, { "epoch": 0.3804464247461548, "grad_norm": 1.0547109842300415, "learning_rate": 9.857826787235744e-05, "loss": 0.3148952960968018, "memory(GiB)": 61.91, "step": 8880, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.510225 }, { "epoch": 0.3806606400754038, "grad_norm": 2.995220184326172, "learning_rate": 9.85766740136982e-05, "loss": 0.48574090003967285, "memory(GiB)": 61.91, "step": 8885, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.510197 }, { "epoch": 0.38087485540465277, "grad_norm": 2.216392755508423, "learning_rate": 9.857507927502697e-05, "loss": 0.16846795082092286, "memory(GiB)": 61.91, "step": 8890, "token_acc": 0.9611307420494699, "train_speed(iter/s)": 1.510181 }, { "epoch": 0.3810890707339017, "grad_norm": 1.1120972633361816, "learning_rate": 9.85734836563726e-05, "loss": 0.44349985122680663, "memory(GiB)": 61.91, "step": 8895, "token_acc": 0.9123711340206185, "train_speed(iter/s)": 1.51014 }, { "epoch": 0.38130328606315067, "grad_norm": 6.4700517654418945, "learning_rate": 9.857188715776401e-05, "loss": 0.38845529556274416, "memory(GiB)": 61.91, "step": 8900, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.510101 }, { "epoch": 0.38151750139239965, "grad_norm": 3.369652032852173, "learning_rate": 9.857028977923011e-05, "loss": 0.5520412445068359, "memory(GiB)": 61.91, "step": 8905, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.510167 }, { "epoch": 0.38173171672164863, "grad_norm": 0.8413757085800171, "learning_rate": 9.856869152079986e-05, "loss": 0.4356602668762207, "memory(GiB)": 61.91, "step": 8910, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.510154 }, { "epoch": 0.38194593205089755, "grad_norm": 2.648301839828491, "learning_rate": 9.85670923825022e-05, "loss": 0.20371561050415038, "memory(GiB)": 61.91, "step": 8915, "token_acc": 0.9539007092198581, "train_speed(iter/s)": 1.510164 }, { "epoch": 0.38216014738014653, "grad_norm": 3.766246795654297, "learning_rate": 9.856549236436609e-05, "loss": 0.6960416793823242, "memory(GiB)": 61.91, "step": 8920, "token_acc": 0.862876254180602, "train_speed(iter/s)": 1.510116 }, { "epoch": 0.3823743627093955, "grad_norm": 3.9631288051605225, "learning_rate": 9.856389146642054e-05, "loss": 0.16741652488708497, "memory(GiB)": 61.91, "step": 8925, "token_acc": 0.9552238805970149, "train_speed(iter/s)": 1.510097 }, { "epoch": 0.38258857803864443, "grad_norm": 4.577905654907227, "learning_rate": 9.856228968869454e-05, "loss": 0.10362929105758667, "memory(GiB)": 61.91, "step": 8930, "token_acc": 0.9779735682819384, "train_speed(iter/s)": 1.510074 }, { "epoch": 0.3828027933678934, "grad_norm": 2.8035566806793213, "learning_rate": 9.85606870312171e-05, "loss": 0.40097761154174805, "memory(GiB)": 61.91, "step": 8935, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.510052 }, { "epoch": 0.3830170086971424, "grad_norm": 4.0206475257873535, "learning_rate": 9.855908349401727e-05, "loss": 0.3893759727478027, "memory(GiB)": 61.91, "step": 8940, "token_acc": 0.9096385542168675, "train_speed(iter/s)": 1.510041 }, { "epoch": 0.3832312240263913, "grad_norm": 2.1264829635620117, "learning_rate": 9.855747907712408e-05, "loss": 0.24885969161987304, "memory(GiB)": 61.91, "step": 8945, "token_acc": 0.9501915708812261, "train_speed(iter/s)": 1.510058 }, { "epoch": 0.3834454393556403, "grad_norm": 2.350576639175415, "learning_rate": 9.855587378056661e-05, "loss": 0.3490028142929077, "memory(GiB)": 61.91, "step": 8950, "token_acc": 0.9313186813186813, "train_speed(iter/s)": 1.510138 }, { "epoch": 0.3836596546848893, "grad_norm": 4.246707439422607, "learning_rate": 9.855426760437394e-05, "loss": 0.5204609394073486, "memory(GiB)": 61.91, "step": 8955, "token_acc": 0.9072847682119205, "train_speed(iter/s)": 1.510208 }, { "epoch": 0.3838738700141382, "grad_norm": 0.48214349150657654, "learning_rate": 9.855266054857518e-05, "loss": 0.2790213584899902, "memory(GiB)": 61.91, "step": 8960, "token_acc": 0.9509803921568627, "train_speed(iter/s)": 1.510267 }, { "epoch": 0.3840880853433872, "grad_norm": 2.4446237087249756, "learning_rate": 9.855105261319939e-05, "loss": 0.268202543258667, "memory(GiB)": 61.91, "step": 8965, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.510358 }, { "epoch": 0.38430230067263615, "grad_norm": 2.3376877307891846, "learning_rate": 9.854944379827577e-05, "loss": 0.5844768524169922, "memory(GiB)": 61.91, "step": 8970, "token_acc": 0.8697183098591549, "train_speed(iter/s)": 1.510401 }, { "epoch": 0.3845165160018851, "grad_norm": 4.177484035491943, "learning_rate": 9.854783410383341e-05, "loss": 0.21230506896972656, "memory(GiB)": 61.91, "step": 8975, "token_acc": 0.940809968847352, "train_speed(iter/s)": 1.510427 }, { "epoch": 0.38473073133113406, "grad_norm": 5.195810794830322, "learning_rate": 9.85462235299015e-05, "loss": 0.3816333293914795, "memory(GiB)": 61.91, "step": 8980, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.510401 }, { "epoch": 0.38494494666038304, "grad_norm": 1.755827784538269, "learning_rate": 9.854461207650922e-05, "loss": 0.3592185974121094, "memory(GiB)": 61.91, "step": 8985, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.510397 }, { "epoch": 0.38515916198963196, "grad_norm": 7.18878173828125, "learning_rate": 9.854299974368575e-05, "loss": 0.38291101455688475, "memory(GiB)": 61.91, "step": 8990, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.510556 }, { "epoch": 0.38537337731888094, "grad_norm": 2.348217010498047, "learning_rate": 9.85413865314603e-05, "loss": 0.3003303050994873, "memory(GiB)": 61.91, "step": 8995, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.510581 }, { "epoch": 0.3855875926481299, "grad_norm": 7.68508243560791, "learning_rate": 9.85397724398621e-05, "loss": 0.28995671272277834, "memory(GiB)": 61.91, "step": 9000, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.510707 }, { "epoch": 0.3855875926481299, "eval_loss": 2.493332862854004, "eval_runtime": 11.624, "eval_samples_per_second": 8.603, "eval_steps_per_second": 8.603, "eval_token_acc": 0.42686170212765956, "step": 9000 }, { "epoch": 0.38580180797737884, "grad_norm": 3.865699291229248, "learning_rate": 9.853815746892039e-05, "loss": 0.3241037607192993, "memory(GiB)": 61.91, "step": 9005, "token_acc": 0.564299424184261, "train_speed(iter/s)": 1.507536 }, { "epoch": 0.3860160233066278, "grad_norm": 2.1083109378814697, "learning_rate": 9.853654161866442e-05, "loss": 0.4097916126251221, "memory(GiB)": 61.91, "step": 9010, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.507543 }, { "epoch": 0.3862302386358768, "grad_norm": 2.979717254638672, "learning_rate": 9.853492488912347e-05, "loss": 0.4312241554260254, "memory(GiB)": 61.91, "step": 9015, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.507539 }, { "epoch": 0.3864444539651257, "grad_norm": 5.858214855194092, "learning_rate": 9.853330728032682e-05, "loss": 0.7904645442962647, "memory(GiB)": 61.91, "step": 9020, "token_acc": 0.859504132231405, "train_speed(iter/s)": 1.507493 }, { "epoch": 0.3866586692943747, "grad_norm": 3.635982036590576, "learning_rate": 9.853168879230379e-05, "loss": 0.526326322555542, "memory(GiB)": 61.91, "step": 9025, "token_acc": 0.9076433121019108, "train_speed(iter/s)": 1.507534 }, { "epoch": 0.3868728846236237, "grad_norm": 3.7731411457061768, "learning_rate": 9.853006942508369e-05, "loss": 0.4724577903747559, "memory(GiB)": 61.91, "step": 9030, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.507685 }, { "epoch": 0.3870870999528726, "grad_norm": 8.06865406036377, "learning_rate": 9.852844917869585e-05, "loss": 0.526760482788086, "memory(GiB)": 61.91, "step": 9035, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.507691 }, { "epoch": 0.3873013152821216, "grad_norm": 0.09817177802324295, "learning_rate": 9.852682805316964e-05, "loss": 0.2205878973007202, "memory(GiB)": 61.91, "step": 9040, "token_acc": 0.9449152542372882, "train_speed(iter/s)": 1.507704 }, { "epoch": 0.38751553061137056, "grad_norm": 3.4237284660339355, "learning_rate": 9.852520604853442e-05, "loss": 0.42710075378417967, "memory(GiB)": 61.91, "step": 9045, "token_acc": 0.8875502008032129, "train_speed(iter/s)": 1.507786 }, { "epoch": 0.3877297459406195, "grad_norm": 1.3031506538391113, "learning_rate": 9.852358316481955e-05, "loss": 0.27853851318359374, "memory(GiB)": 61.91, "step": 9050, "token_acc": 0.9329268292682927, "train_speed(iter/s)": 1.507872 }, { "epoch": 0.38794396126986846, "grad_norm": 1.7132227420806885, "learning_rate": 9.852195940205448e-05, "loss": 0.3284109115600586, "memory(GiB)": 61.91, "step": 9055, "token_acc": 0.9212121212121213, "train_speed(iter/s)": 1.507969 }, { "epoch": 0.38815817659911744, "grad_norm": 0.02764258161187172, "learning_rate": 9.852033476026859e-05, "loss": 0.21910562515258789, "memory(GiB)": 61.91, "step": 9060, "token_acc": 0.9435736677115988, "train_speed(iter/s)": 1.507965 }, { "epoch": 0.38837239192836637, "grad_norm": 4.636297225952148, "learning_rate": 9.851870923949131e-05, "loss": 0.4800222396850586, "memory(GiB)": 61.91, "step": 9065, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.507979 }, { "epoch": 0.38858660725761535, "grad_norm": 3.297797918319702, "learning_rate": 9.85170828397521e-05, "loss": 0.45043220520019533, "memory(GiB)": 61.91, "step": 9070, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.507983 }, { "epoch": 0.3888008225868643, "grad_norm": 8.735084533691406, "learning_rate": 9.851545556108042e-05, "loss": 0.2469538688659668, "memory(GiB)": 61.91, "step": 9075, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.508073 }, { "epoch": 0.3890150379161133, "grad_norm": 6.224905014038086, "learning_rate": 9.851382740350576e-05, "loss": 0.4206997394561768, "memory(GiB)": 61.91, "step": 9080, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.508185 }, { "epoch": 0.38922925324536223, "grad_norm": 5.104848384857178, "learning_rate": 9.851219836705761e-05, "loss": 0.36280670166015627, "memory(GiB)": 61.91, "step": 9085, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.508234 }, { "epoch": 0.3894434685746112, "grad_norm": 1.6219736337661743, "learning_rate": 9.851056845176547e-05, "loss": 0.26813039779663084, "memory(GiB)": 61.91, "step": 9090, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.508247 }, { "epoch": 0.3896576839038602, "grad_norm": 2.1730425357818604, "learning_rate": 9.850893765765887e-05, "loss": 0.2971574068069458, "memory(GiB)": 61.91, "step": 9095, "token_acc": 0.9315589353612167, "train_speed(iter/s)": 1.508345 }, { "epoch": 0.3898718992331091, "grad_norm": 2.4576165676116943, "learning_rate": 9.850730598476737e-05, "loss": 0.30891373157501223, "memory(GiB)": 61.91, "step": 9100, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.508328 }, { "epoch": 0.3900861145623581, "grad_norm": 3.0185914039611816, "learning_rate": 9.850567343312051e-05, "loss": 0.3441664218902588, "memory(GiB)": 61.91, "step": 9105, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.508309 }, { "epoch": 0.39030032989160707, "grad_norm": 2.554255247116089, "learning_rate": 9.850404000274789e-05, "loss": 0.7735501766204834, "memory(GiB)": 61.91, "step": 9110, "token_acc": 0.8586956521739131, "train_speed(iter/s)": 1.508368 }, { "epoch": 0.390514545220856, "grad_norm": 1.8763355016708374, "learning_rate": 9.850240569367908e-05, "loss": 0.5874289989471435, "memory(GiB)": 61.91, "step": 9115, "token_acc": 0.8695652173913043, "train_speed(iter/s)": 1.508441 }, { "epoch": 0.39072876055010497, "grad_norm": 3.8574509620666504, "learning_rate": 9.850077050594368e-05, "loss": 0.29838132858276367, "memory(GiB)": 61.91, "step": 9120, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.50845 }, { "epoch": 0.39094297587935395, "grad_norm": 3.053551197052002, "learning_rate": 9.849913443957133e-05, "loss": 0.5314243793487549, "memory(GiB)": 61.91, "step": 9125, "token_acc": 0.9003831417624522, "train_speed(iter/s)": 1.508431 }, { "epoch": 0.3911571912086029, "grad_norm": 6.160475254058838, "learning_rate": 9.849749749459167e-05, "loss": 0.34102027416229247, "memory(GiB)": 61.91, "step": 9130, "token_acc": 0.9254658385093167, "train_speed(iter/s)": 1.508443 }, { "epoch": 0.39137140653785185, "grad_norm": 0.933570384979248, "learning_rate": 9.849585967103434e-05, "loss": 0.49098052978515627, "memory(GiB)": 61.91, "step": 9135, "token_acc": 0.8966565349544073, "train_speed(iter/s)": 1.508521 }, { "epoch": 0.39158562186710083, "grad_norm": 2.8948614597320557, "learning_rate": 9.849422096892902e-05, "loss": 0.6587590217590332, "memory(GiB)": 61.91, "step": 9140, "token_acc": 0.8675496688741722, "train_speed(iter/s)": 1.5085 }, { "epoch": 0.39179983719634975, "grad_norm": 3.9750287532806396, "learning_rate": 9.84925813883054e-05, "loss": 0.4761202812194824, "memory(GiB)": 61.91, "step": 9145, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.508516 }, { "epoch": 0.39201405252559873, "grad_norm": 2.58135724067688, "learning_rate": 9.849094092919318e-05, "loss": 0.5301731109619141, "memory(GiB)": 61.91, "step": 9150, "token_acc": 0.908284023668639, "train_speed(iter/s)": 1.508668 }, { "epoch": 0.3922282678548477, "grad_norm": 2.325699806213379, "learning_rate": 9.848929959162207e-05, "loss": 0.32035117149353026, "memory(GiB)": 61.91, "step": 9155, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.508714 }, { "epoch": 0.39244248318409664, "grad_norm": 5.315793991088867, "learning_rate": 9.848765737562183e-05, "loss": 0.3063088893890381, "memory(GiB)": 61.91, "step": 9160, "token_acc": 0.9054054054054054, "train_speed(iter/s)": 1.508721 }, { "epoch": 0.3926566985133456, "grad_norm": 0.09928171336650848, "learning_rate": 9.848601428122217e-05, "loss": 0.237510347366333, "memory(GiB)": 61.91, "step": 9165, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.508745 }, { "epoch": 0.3928709138425946, "grad_norm": 6.997530460357666, "learning_rate": 9.848437030845288e-05, "loss": 0.5109744071960449, "memory(GiB)": 61.91, "step": 9170, "token_acc": 0.8945454545454545, "train_speed(iter/s)": 1.508893 }, { "epoch": 0.3930851291718435, "grad_norm": 2.8731443881988525, "learning_rate": 9.848272545734374e-05, "loss": 0.2373945951461792, "memory(GiB)": 61.91, "step": 9175, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.508912 }, { "epoch": 0.3932993445010925, "grad_norm": 3.1404454708099365, "learning_rate": 9.848107972792455e-05, "loss": 0.19891676902770997, "memory(GiB)": 61.91, "step": 9180, "token_acc": 0.9576271186440678, "train_speed(iter/s)": 1.508946 }, { "epoch": 0.3935135598303415, "grad_norm": 2.1783807277679443, "learning_rate": 9.847943312022511e-05, "loss": 0.6232982158660889, "memory(GiB)": 61.91, "step": 9185, "token_acc": 0.8580441640378549, "train_speed(iter/s)": 1.508909 }, { "epoch": 0.3937277751595904, "grad_norm": 1.9073419570922852, "learning_rate": 9.847778563427529e-05, "loss": 0.5501774787902832, "memory(GiB)": 61.91, "step": 9190, "token_acc": 0.8926553672316384, "train_speed(iter/s)": 1.508892 }, { "epoch": 0.3939419904888394, "grad_norm": 5.263160705566406, "learning_rate": 9.84761372701049e-05, "loss": 0.35663886070251466, "memory(GiB)": 61.91, "step": 9195, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.50899 }, { "epoch": 0.39415620581808836, "grad_norm": 4.348228454589844, "learning_rate": 9.847448802774379e-05, "loss": 0.23365044593811035, "memory(GiB)": 61.91, "step": 9200, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.509034 }, { "epoch": 0.3943704211473373, "grad_norm": 3.4054782390594482, "learning_rate": 9.847283790722187e-05, "loss": 0.3324253082275391, "memory(GiB)": 61.91, "step": 9205, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.509002 }, { "epoch": 0.39458463647658626, "grad_norm": 1.1170982122421265, "learning_rate": 9.847118690856903e-05, "loss": 0.58001708984375, "memory(GiB)": 61.91, "step": 9210, "token_acc": 0.8781163434903048, "train_speed(iter/s)": 1.509164 }, { "epoch": 0.39479885180583524, "grad_norm": 2.8606367111206055, "learning_rate": 9.846953503181515e-05, "loss": 0.4028620719909668, "memory(GiB)": 61.91, "step": 9215, "token_acc": 0.8934426229508197, "train_speed(iter/s)": 1.509209 }, { "epoch": 0.39501306713508416, "grad_norm": 3.958568811416626, "learning_rate": 9.846788227699017e-05, "loss": 0.3659388542175293, "memory(GiB)": 61.91, "step": 9220, "token_acc": 0.9181818181818182, "train_speed(iter/s)": 1.509214 }, { "epoch": 0.39522728246433314, "grad_norm": 2.0989201068878174, "learning_rate": 9.846622864412406e-05, "loss": 0.29769277572631836, "memory(GiB)": 61.91, "step": 9225, "token_acc": 0.9388379204892966, "train_speed(iter/s)": 1.509199 }, { "epoch": 0.3954414977935821, "grad_norm": 3.152843713760376, "learning_rate": 9.846457413324675e-05, "loss": 0.4384924411773682, "memory(GiB)": 61.91, "step": 9230, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.509192 }, { "epoch": 0.39565571312283104, "grad_norm": 1.5631520748138428, "learning_rate": 9.84629187443882e-05, "loss": 0.24437155723571777, "memory(GiB)": 61.91, "step": 9235, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.50921 }, { "epoch": 0.39586992845208, "grad_norm": 2.2578494548797607, "learning_rate": 9.846126247757843e-05, "loss": 0.2634784936904907, "memory(GiB)": 61.91, "step": 9240, "token_acc": 0.9632352941176471, "train_speed(iter/s)": 1.509293 }, { "epoch": 0.396084143781329, "grad_norm": 1.0716866254806519, "learning_rate": 9.845960533284742e-05, "loss": 0.3505596399307251, "memory(GiB)": 61.91, "step": 9245, "token_acc": 0.9212328767123288, "train_speed(iter/s)": 1.509311 }, { "epoch": 0.396298359110578, "grad_norm": 1.7266114950180054, "learning_rate": 9.84579473102252e-05, "loss": 0.11918718814849853, "memory(GiB)": 61.91, "step": 9250, "token_acc": 0.97265625, "train_speed(iter/s)": 1.509328 }, { "epoch": 0.3965125744398269, "grad_norm": 2.1381402015686035, "learning_rate": 9.845628840974182e-05, "loss": 0.3591198444366455, "memory(GiB)": 61.91, "step": 9255, "token_acc": 0.9217391304347826, "train_speed(iter/s)": 1.50933 }, { "epoch": 0.3967267897690759, "grad_norm": 4.002885341644287, "learning_rate": 9.84546286314273e-05, "loss": 0.317641019821167, "memory(GiB)": 61.91, "step": 9260, "token_acc": 0.9351032448377581, "train_speed(iter/s)": 1.509308 }, { "epoch": 0.39694100509832486, "grad_norm": 2.29093599319458, "learning_rate": 9.845296797531173e-05, "loss": 0.3493596076965332, "memory(GiB)": 61.91, "step": 9265, "token_acc": 0.9260450160771704, "train_speed(iter/s)": 1.50929 }, { "epoch": 0.3971552204275738, "grad_norm": 1.6279643774032593, "learning_rate": 9.845130644142522e-05, "loss": 0.3175642728805542, "memory(GiB)": 61.91, "step": 9270, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.509261 }, { "epoch": 0.39736943575682276, "grad_norm": 4.411620616912842, "learning_rate": 9.844964402979781e-05, "loss": 0.4816420555114746, "memory(GiB)": 61.91, "step": 9275, "token_acc": 0.8907849829351536, "train_speed(iter/s)": 1.509321 }, { "epoch": 0.39758365108607174, "grad_norm": 3.202716588973999, "learning_rate": 9.844798074045967e-05, "loss": 0.38599624633789065, "memory(GiB)": 61.91, "step": 9280, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.509251 }, { "epoch": 0.39779786641532067, "grad_norm": 4.817872047424316, "learning_rate": 9.84463165734409e-05, "loss": 0.35334599018096924, "memory(GiB)": 61.91, "step": 9285, "token_acc": 0.9313432835820895, "train_speed(iter/s)": 1.509218 }, { "epoch": 0.39801208174456965, "grad_norm": 2.0214006900787354, "learning_rate": 9.844465152877167e-05, "loss": 0.3602099895477295, "memory(GiB)": 61.91, "step": 9290, "token_acc": 0.9415584415584416, "train_speed(iter/s)": 1.509196 }, { "epoch": 0.3982262970738186, "grad_norm": 10.941797256469727, "learning_rate": 9.844298560648213e-05, "loss": 0.41784276962280276, "memory(GiB)": 61.91, "step": 9295, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.509146 }, { "epoch": 0.39844051240306755, "grad_norm": 3.914381265640259, "learning_rate": 9.844131880660246e-05, "loss": 0.8123879432678223, "memory(GiB)": 61.91, "step": 9300, "token_acc": 0.8472222222222222, "train_speed(iter/s)": 1.509227 }, { "epoch": 0.3986547277323165, "grad_norm": 4.4207563400268555, "learning_rate": 9.843965112916285e-05, "loss": 0.43529157638549804, "memory(GiB)": 61.91, "step": 9305, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.509331 }, { "epoch": 0.3988689430615655, "grad_norm": 3.0782039165496826, "learning_rate": 9.843798257419352e-05, "loss": 0.47614479064941406, "memory(GiB)": 61.91, "step": 9310, "token_acc": 0.8982456140350877, "train_speed(iter/s)": 1.509386 }, { "epoch": 0.39908315839081443, "grad_norm": 2.103883743286133, "learning_rate": 9.843631314172471e-05, "loss": 0.4043313503265381, "memory(GiB)": 61.91, "step": 9315, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.50939 }, { "epoch": 0.3992973737200634, "grad_norm": 2.76021671295166, "learning_rate": 9.843464283178665e-05, "loss": 0.3106995105743408, "memory(GiB)": 61.91, "step": 9320, "token_acc": 0.9399141630901288, "train_speed(iter/s)": 1.509364 }, { "epoch": 0.3995115890493124, "grad_norm": 0.18753162026405334, "learning_rate": 9.843297164440959e-05, "loss": 0.3293175458908081, "memory(GiB)": 61.91, "step": 9325, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.509443 }, { "epoch": 0.3997258043785613, "grad_norm": 3.9525043964385986, "learning_rate": 9.843129957962381e-05, "loss": 0.556052303314209, "memory(GiB)": 61.91, "step": 9330, "token_acc": 0.8711484593837535, "train_speed(iter/s)": 1.509362 }, { "epoch": 0.3999400197078103, "grad_norm": 2.609650135040283, "learning_rate": 9.842962663745963e-05, "loss": 0.5731803894042968, "memory(GiB)": 61.91, "step": 9335, "token_acc": 0.8828828828828829, "train_speed(iter/s)": 1.509343 }, { "epoch": 0.40015423503705927, "grad_norm": 2.8130507469177246, "learning_rate": 9.842795281794732e-05, "loss": 0.3710965633392334, "memory(GiB)": 61.91, "step": 9340, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.509435 }, { "epoch": 0.4003684503663082, "grad_norm": 1.3778797388076782, "learning_rate": 9.84262781211172e-05, "loss": 0.22917454242706298, "memory(GiB)": 61.91, "step": 9345, "token_acc": 0.9442622950819672, "train_speed(iter/s)": 1.509437 }, { "epoch": 0.40058266569555717, "grad_norm": 2.0346546173095703, "learning_rate": 9.842460254699963e-05, "loss": 0.2748678207397461, "memory(GiB)": 61.91, "step": 9350, "token_acc": 0.9147286821705426, "train_speed(iter/s)": 1.50944 }, { "epoch": 0.40079688102480615, "grad_norm": 4.485888481140137, "learning_rate": 9.842292609562498e-05, "loss": 0.4910168170928955, "memory(GiB)": 61.91, "step": 9355, "token_acc": 0.875, "train_speed(iter/s)": 1.509518 }, { "epoch": 0.4010110963540551, "grad_norm": 0.9089673757553101, "learning_rate": 9.84212487670236e-05, "loss": 0.11922661066055298, "memory(GiB)": 61.91, "step": 9360, "token_acc": 0.9761904761904762, "train_speed(iter/s)": 1.509534 }, { "epoch": 0.40122531168330405, "grad_norm": 3.0014424324035645, "learning_rate": 9.841957056122584e-05, "loss": 0.5776897430419922, "memory(GiB)": 61.91, "step": 9365, "token_acc": 0.8885869565217391, "train_speed(iter/s)": 1.509524 }, { "epoch": 0.40143952701255303, "grad_norm": 2.6283209323883057, "learning_rate": 9.841789147826217e-05, "loss": 0.3444781541824341, "memory(GiB)": 61.91, "step": 9370, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.509571 }, { "epoch": 0.40165374234180196, "grad_norm": 4.2410101890563965, "learning_rate": 9.841621151816296e-05, "loss": 0.33422231674194336, "memory(GiB)": 61.91, "step": 9375, "token_acc": 0.9216300940438872, "train_speed(iter/s)": 1.509637 }, { "epoch": 0.40186795767105093, "grad_norm": 4.932287693023682, "learning_rate": 9.841453068095867e-05, "loss": 0.6745366096496582, "memory(GiB)": 61.91, "step": 9380, "token_acc": 0.868421052631579, "train_speed(iter/s)": 1.509579 }, { "epoch": 0.4020821730002999, "grad_norm": 2.076151132583618, "learning_rate": 9.841284896667973e-05, "loss": 0.26653265953063965, "memory(GiB)": 61.91, "step": 9385, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.509617 }, { "epoch": 0.40229638832954884, "grad_norm": 2.144948720932007, "learning_rate": 9.841116637535662e-05, "loss": 0.35812427997589114, "memory(GiB)": 61.91, "step": 9390, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.509731 }, { "epoch": 0.4025106036587978, "grad_norm": 5.576577663421631, "learning_rate": 9.840948290701982e-05, "loss": 0.2969074010848999, "memory(GiB)": 61.91, "step": 9395, "token_acc": 0.934375, "train_speed(iter/s)": 1.509772 }, { "epoch": 0.4027248189880468, "grad_norm": 0.7447547316551208, "learning_rate": 9.840779856169982e-05, "loss": 0.30115158557891847, "memory(GiB)": 61.91, "step": 9400, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.509731 }, { "epoch": 0.4029390343172957, "grad_norm": 1.771230936050415, "learning_rate": 9.840611333942715e-05, "loss": 0.373026442527771, "memory(GiB)": 61.91, "step": 9405, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.509797 }, { "epoch": 0.4031532496465447, "grad_norm": 6.167954444885254, "learning_rate": 9.840442724023232e-05, "loss": 0.215767502784729, "memory(GiB)": 61.91, "step": 9410, "token_acc": 0.9488188976377953, "train_speed(iter/s)": 1.509812 }, { "epoch": 0.4033674649757937, "grad_norm": 2.287842273712158, "learning_rate": 9.840274026414589e-05, "loss": 0.4309383392333984, "memory(GiB)": 61.91, "step": 9415, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.509826 }, { "epoch": 0.40358168030504266, "grad_norm": 2.981257438659668, "learning_rate": 9.840105241119841e-05, "loss": 0.41526012420654296, "memory(GiB)": 61.91, "step": 9420, "token_acc": 0.9120521172638436, "train_speed(iter/s)": 1.509932 }, { "epoch": 0.4037958956342916, "grad_norm": 3.222212314605713, "learning_rate": 9.839936368142046e-05, "loss": 0.6014633655548096, "memory(GiB)": 61.91, "step": 9425, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.509914 }, { "epoch": 0.40401011096354056, "grad_norm": 5.132739067077637, "learning_rate": 9.839767407484264e-05, "loss": 0.4919823169708252, "memory(GiB)": 61.91, "step": 9430, "token_acc": 0.9077380952380952, "train_speed(iter/s)": 1.509985 }, { "epoch": 0.40422432629278954, "grad_norm": 3.8185617923736572, "learning_rate": 9.839598359149556e-05, "loss": 0.29718289375305174, "memory(GiB)": 61.91, "step": 9435, "token_acc": 0.9423728813559322, "train_speed(iter/s)": 1.509979 }, { "epoch": 0.40443854162203846, "grad_norm": 1.8861027956008911, "learning_rate": 9.839429223140981e-05, "loss": 0.3563275098800659, "memory(GiB)": 61.91, "step": 9440, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.510085 }, { "epoch": 0.40465275695128744, "grad_norm": 3.9010159969329834, "learning_rate": 9.839259999461609e-05, "loss": 0.45616750717163085, "memory(GiB)": 61.91, "step": 9445, "token_acc": 0.9050131926121372, "train_speed(iter/s)": 1.510084 }, { "epoch": 0.4048669722805364, "grad_norm": 1.5324698686599731, "learning_rate": 9.839090688114501e-05, "loss": 0.5001577377319336, "memory(GiB)": 61.91, "step": 9450, "token_acc": 0.8903225806451613, "train_speed(iter/s)": 1.510085 }, { "epoch": 0.40508118760978534, "grad_norm": 2.967945098876953, "learning_rate": 9.838921289102726e-05, "loss": 0.6575433731079101, "memory(GiB)": 61.91, "step": 9455, "token_acc": 0.8686440677966102, "train_speed(iter/s)": 1.510098 }, { "epoch": 0.4052954029390343, "grad_norm": 3.6217236518859863, "learning_rate": 9.838751802429352e-05, "loss": 0.4155773639678955, "memory(GiB)": 61.91, "step": 9460, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.510165 }, { "epoch": 0.4055096182682833, "grad_norm": 3.0473792552948, "learning_rate": 9.83858222809745e-05, "loss": 0.481168270111084, "memory(GiB)": 61.91, "step": 9465, "token_acc": 0.9148264984227129, "train_speed(iter/s)": 1.510219 }, { "epoch": 0.4057238335975322, "grad_norm": 5.155152320861816, "learning_rate": 9.838412566110094e-05, "loss": 0.41246652603149414, "memory(GiB)": 61.91, "step": 9470, "token_acc": 0.903114186851211, "train_speed(iter/s)": 1.51031 }, { "epoch": 0.4059380489267812, "grad_norm": 1.9429383277893066, "learning_rate": 9.838242816470353e-05, "loss": 0.2831919193267822, "memory(GiB)": 61.91, "step": 9475, "token_acc": 0.9315068493150684, "train_speed(iter/s)": 1.510322 }, { "epoch": 0.4061522642560302, "grad_norm": 2.4607765674591064, "learning_rate": 9.838072979181306e-05, "loss": 0.46988525390625, "memory(GiB)": 61.91, "step": 9480, "token_acc": 0.8880866425992779, "train_speed(iter/s)": 1.510449 }, { "epoch": 0.4063664795852791, "grad_norm": 2.2878575325012207, "learning_rate": 9.83790305424603e-05, "loss": 0.38763017654418946, "memory(GiB)": 61.91, "step": 9485, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.510441 }, { "epoch": 0.4065806949145281, "grad_norm": 3.168149709701538, "learning_rate": 9.837733041667598e-05, "loss": 0.7441182136535645, "memory(GiB)": 61.91, "step": 9490, "token_acc": 0.847682119205298, "train_speed(iter/s)": 1.510458 }, { "epoch": 0.40679491024377706, "grad_norm": 1.9657580852508545, "learning_rate": 9.837562941449096e-05, "loss": 0.4221211910247803, "memory(GiB)": 61.91, "step": 9495, "token_acc": 0.9147540983606557, "train_speed(iter/s)": 1.510511 }, { "epoch": 0.407009125573026, "grad_norm": 2.6780476570129395, "learning_rate": 9.837392753593604e-05, "loss": 0.4598477840423584, "memory(GiB)": 61.91, "step": 9500, "token_acc": 0.9060402684563759, "train_speed(iter/s)": 1.510512 }, { "epoch": 0.407009125573026, "eval_loss": 2.56831431388855, "eval_runtime": 11.2006, "eval_samples_per_second": 8.928, "eval_steps_per_second": 8.928, "eval_token_acc": 0.45892351274787535, "step": 9500 }, { "epoch": 0.40722334090227497, "grad_norm": 4.5664801597595215, "learning_rate": 9.837222478104205e-05, "loss": 0.680894422531128, "memory(GiB)": 61.91, "step": 9505, "token_acc": 0.5866797257590598, "train_speed(iter/s)": 1.50767 }, { "epoch": 0.40743755623152395, "grad_norm": 1.9718036651611328, "learning_rate": 9.837052114983982e-05, "loss": 0.5184160232543945, "memory(GiB)": 61.91, "step": 9510, "token_acc": 0.8877005347593583, "train_speed(iter/s)": 1.507733 }, { "epoch": 0.40765177156077287, "grad_norm": 4.973840236663818, "learning_rate": 9.836881664236021e-05, "loss": 0.5727438449859619, "memory(GiB)": 61.91, "step": 9515, "token_acc": 0.8632218844984803, "train_speed(iter/s)": 1.507702 }, { "epoch": 0.40786598689002185, "grad_norm": 2.8805673122406006, "learning_rate": 9.836711125863413e-05, "loss": 0.46134214401245116, "memory(GiB)": 61.91, "step": 9520, "token_acc": 0.8969072164948454, "train_speed(iter/s)": 1.507709 }, { "epoch": 0.4080802022192708, "grad_norm": 3.229316234588623, "learning_rate": 9.836540499869244e-05, "loss": 0.4668818473815918, "memory(GiB)": 61.91, "step": 9525, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.507699 }, { "epoch": 0.40829441754851975, "grad_norm": 4.1765031814575195, "learning_rate": 9.83636978625661e-05, "loss": 0.30263590812683105, "memory(GiB)": 61.91, "step": 9530, "token_acc": 0.9282868525896414, "train_speed(iter/s)": 1.507712 }, { "epoch": 0.40850863287776873, "grad_norm": 2.182018756866455, "learning_rate": 9.836198985028597e-05, "loss": 0.27615444660186766, "memory(GiB)": 61.91, "step": 9535, "token_acc": 0.9446254071661238, "train_speed(iter/s)": 1.507783 }, { "epoch": 0.4087228482070177, "grad_norm": 3.8643085956573486, "learning_rate": 9.836028096188306e-05, "loss": 0.3966536045074463, "memory(GiB)": 61.91, "step": 9540, "token_acc": 0.9173913043478261, "train_speed(iter/s)": 1.50783 }, { "epoch": 0.40893706353626663, "grad_norm": 3.702930212020874, "learning_rate": 9.835857119738827e-05, "loss": 0.41988568305969237, "memory(GiB)": 61.91, "step": 9545, "token_acc": 0.900398406374502, "train_speed(iter/s)": 1.507886 }, { "epoch": 0.4091512788655156, "grad_norm": 2.091282844543457, "learning_rate": 9.83568605568326e-05, "loss": 0.4098775863647461, "memory(GiB)": 61.91, "step": 9550, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.50789 }, { "epoch": 0.4093654941947646, "grad_norm": 1.894609808921814, "learning_rate": 9.835514904024705e-05, "loss": 0.5212301254272461, "memory(GiB)": 61.91, "step": 9555, "token_acc": 0.8698224852071006, "train_speed(iter/s)": 1.507954 }, { "epoch": 0.4095797095240135, "grad_norm": 3.5703845024108887, "learning_rate": 9.83534366476626e-05, "loss": 0.23590166568756105, "memory(GiB)": 61.91, "step": 9560, "token_acc": 0.9539007092198581, "train_speed(iter/s)": 1.50794 }, { "epoch": 0.4097939248532625, "grad_norm": 1.3079437017440796, "learning_rate": 9.83517233791103e-05, "loss": 0.3714072465896606, "memory(GiB)": 61.91, "step": 9565, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.508033 }, { "epoch": 0.41000814018251147, "grad_norm": 1.1767436265945435, "learning_rate": 9.835000923462117e-05, "loss": 0.28921120166778563, "memory(GiB)": 61.91, "step": 9570, "token_acc": 0.9409722222222222, "train_speed(iter/s)": 1.507982 }, { "epoch": 0.4102223555117604, "grad_norm": 2.228224515914917, "learning_rate": 9.834829421422627e-05, "loss": 0.5152900695800782, "memory(GiB)": 61.91, "step": 9575, "token_acc": 0.893687707641196, "train_speed(iter/s)": 1.507987 }, { "epoch": 0.4104365708410094, "grad_norm": 3.4651308059692383, "learning_rate": 9.834657831795666e-05, "loss": 0.44086122512817383, "memory(GiB)": 61.91, "step": 9580, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.508041 }, { "epoch": 0.41065078617025835, "grad_norm": 2.4249050617218018, "learning_rate": 9.834486154584342e-05, "loss": 0.4613369941711426, "memory(GiB)": 61.91, "step": 9585, "token_acc": 0.9, "train_speed(iter/s)": 1.508089 }, { "epoch": 0.41086500149950733, "grad_norm": 6.373779296875, "learning_rate": 9.834314389791767e-05, "loss": 0.49302358627319337, "memory(GiB)": 61.91, "step": 9590, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.508116 }, { "epoch": 0.41107921682875626, "grad_norm": 0.9531047344207764, "learning_rate": 9.834142537421053e-05, "loss": 0.3093680143356323, "memory(GiB)": 61.91, "step": 9595, "token_acc": 0.952076677316294, "train_speed(iter/s)": 1.508204 }, { "epoch": 0.41129343215800523, "grad_norm": 2.042576313018799, "learning_rate": 9.833970597475311e-05, "loss": 0.45992116928100585, "memory(GiB)": 61.91, "step": 9600, "token_acc": 0.8996763754045307, "train_speed(iter/s)": 1.508206 }, { "epoch": 0.4115076474872542, "grad_norm": 3.035285472869873, "learning_rate": 9.833798569957657e-05, "loss": 0.341129469871521, "memory(GiB)": 61.91, "step": 9605, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.508203 }, { "epoch": 0.41172186281650314, "grad_norm": 0.32897499203681946, "learning_rate": 9.833626454871208e-05, "loss": 0.5371445655822754, "memory(GiB)": 61.91, "step": 9610, "token_acc": 0.8498659517426274, "train_speed(iter/s)": 1.508274 }, { "epoch": 0.4119360781457521, "grad_norm": 6.9121270179748535, "learning_rate": 9.833454252219082e-05, "loss": 0.545198917388916, "memory(GiB)": 61.91, "step": 9615, "token_acc": 0.8375451263537906, "train_speed(iter/s)": 1.508295 }, { "epoch": 0.4121502934750011, "grad_norm": 3.3451621532440186, "learning_rate": 9.833281962004397e-05, "loss": 0.5942502498626709, "memory(GiB)": 61.91, "step": 9620, "token_acc": 0.9026217228464419, "train_speed(iter/s)": 1.508361 }, { "epoch": 0.41236450880425, "grad_norm": 3.7237138748168945, "learning_rate": 9.833109584230275e-05, "loss": 0.5255822658538818, "memory(GiB)": 61.91, "step": 9625, "token_acc": 0.8896551724137931, "train_speed(iter/s)": 1.508381 }, { "epoch": 0.412578724133499, "grad_norm": 1.8957005739212036, "learning_rate": 9.832937118899842e-05, "loss": 0.2867961645126343, "memory(GiB)": 61.91, "step": 9630, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.508603 }, { "epoch": 0.412792939462748, "grad_norm": 3.5014421939849854, "learning_rate": 9.832764566016216e-05, "loss": 0.6240625381469727, "memory(GiB)": 61.91, "step": 9635, "token_acc": 0.8766233766233766, "train_speed(iter/s)": 1.508592 }, { "epoch": 0.4130071547919969, "grad_norm": 2.6530356407165527, "learning_rate": 9.832591925582527e-05, "loss": 0.3201970100402832, "memory(GiB)": 61.91, "step": 9640, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.508678 }, { "epoch": 0.4132213701212459, "grad_norm": 1.9383749961853027, "learning_rate": 9.832419197601903e-05, "loss": 0.4156783580780029, "memory(GiB)": 61.91, "step": 9645, "token_acc": 0.9073359073359073, "train_speed(iter/s)": 1.5087 }, { "epoch": 0.41343558545049486, "grad_norm": 3.4899020195007324, "learning_rate": 9.832246382077471e-05, "loss": 0.4324374198913574, "memory(GiB)": 61.91, "step": 9650, "token_acc": 0.925, "train_speed(iter/s)": 1.508708 }, { "epoch": 0.4136498007797438, "grad_norm": 1.7662739753723145, "learning_rate": 9.832073479012364e-05, "loss": 0.46340503692626955, "memory(GiB)": 61.91, "step": 9655, "token_acc": 0.904, "train_speed(iter/s)": 1.508802 }, { "epoch": 0.41386401610899276, "grad_norm": 2.75042462348938, "learning_rate": 9.83190048840971e-05, "loss": 0.40130152702331545, "memory(GiB)": 61.91, "step": 9660, "token_acc": 0.9182879377431906, "train_speed(iter/s)": 1.508799 }, { "epoch": 0.41407823143824174, "grad_norm": 2.522662401199341, "learning_rate": 9.83172741027265e-05, "loss": 0.31501569747924807, "memory(GiB)": 61.91, "step": 9665, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.50879 }, { "epoch": 0.41429244676749066, "grad_norm": 3.5338311195373535, "learning_rate": 9.831554244604313e-05, "loss": 0.6635523796081543, "memory(GiB)": 61.91, "step": 9670, "token_acc": 0.8753894080996885, "train_speed(iter/s)": 1.508807 }, { "epoch": 0.41450666209673964, "grad_norm": 3.26469087600708, "learning_rate": 9.831380991407841e-05, "loss": 0.21499178409576417, "memory(GiB)": 61.91, "step": 9675, "token_acc": 0.9598214285714286, "train_speed(iter/s)": 1.508809 }, { "epoch": 0.4147208774259886, "grad_norm": 7.517244338989258, "learning_rate": 9.831207650686367e-05, "loss": 0.6489170074462891, "memory(GiB)": 61.91, "step": 9680, "token_acc": 0.8685015290519877, "train_speed(iter/s)": 1.508867 }, { "epoch": 0.41493509275523754, "grad_norm": 1.9291375875473022, "learning_rate": 9.831034222443037e-05, "loss": 0.30274648666381837, "memory(GiB)": 61.91, "step": 9685, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.508854 }, { "epoch": 0.4151493080844865, "grad_norm": 1.9034607410430908, "learning_rate": 9.830860706680989e-05, "loss": 0.26431980133056643, "memory(GiB)": 61.91, "step": 9690, "token_acc": 0.9418960244648318, "train_speed(iter/s)": 1.50893 }, { "epoch": 0.4153635234137355, "grad_norm": 11.554343223571777, "learning_rate": 9.830687103403367e-05, "loss": 0.38609585762023924, "memory(GiB)": 61.91, "step": 9695, "token_acc": 0.9375, "train_speed(iter/s)": 1.508966 }, { "epoch": 0.4155777387429844, "grad_norm": 0.8724965453147888, "learning_rate": 9.830513412613318e-05, "loss": 0.3114032745361328, "memory(GiB)": 61.91, "step": 9700, "token_acc": 0.927797833935018, "train_speed(iter/s)": 1.508989 }, { "epoch": 0.4157919540722334, "grad_norm": 3.7835073471069336, "learning_rate": 9.830339634313985e-05, "loss": 0.4040060043334961, "memory(GiB)": 61.91, "step": 9705, "token_acc": 0.9123867069486404, "train_speed(iter/s)": 1.509057 }, { "epoch": 0.4160061694014824, "grad_norm": 3.159074306488037, "learning_rate": 9.830165768508519e-05, "loss": 0.2844813585281372, "memory(GiB)": 61.91, "step": 9710, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.509037 }, { "epoch": 0.4162203847307313, "grad_norm": 1.0627632141113281, "learning_rate": 9.82999181520007e-05, "loss": 0.3560017108917236, "memory(GiB)": 61.91, "step": 9715, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.509033 }, { "epoch": 0.4164346000599803, "grad_norm": 3.1724600791931152, "learning_rate": 9.829817774391788e-05, "loss": 0.41548972129821776, "memory(GiB)": 61.91, "step": 9720, "token_acc": 0.9273743016759777, "train_speed(iter/s)": 1.509043 }, { "epoch": 0.41664881538922927, "grad_norm": 2.3089375495910645, "learning_rate": 9.829643646086826e-05, "loss": 0.2558934450149536, "memory(GiB)": 61.91, "step": 9725, "token_acc": 0.940625, "train_speed(iter/s)": 1.509074 }, { "epoch": 0.4168630307184782, "grad_norm": 3.5954220294952393, "learning_rate": 9.829469430288338e-05, "loss": 0.5200096130371094, "memory(GiB)": 61.91, "step": 9730, "token_acc": 0.8814814814814815, "train_speed(iter/s)": 1.509162 }, { "epoch": 0.41707724604772717, "grad_norm": 2.938887357711792, "learning_rate": 9.829295126999482e-05, "loss": 0.5172693729400635, "memory(GiB)": 61.91, "step": 9735, "token_acc": 0.8812949640287769, "train_speed(iter/s)": 1.509157 }, { "epoch": 0.41729146137697615, "grad_norm": 3.556504487991333, "learning_rate": 9.829120736223417e-05, "loss": 0.44468984603881834, "memory(GiB)": 61.91, "step": 9740, "token_acc": 0.8789808917197452, "train_speed(iter/s)": 1.50924 }, { "epoch": 0.41750567670622507, "grad_norm": 1.7452746629714966, "learning_rate": 9.828946257963296e-05, "loss": 0.2712139844894409, "memory(GiB)": 61.91, "step": 9745, "token_acc": 0.9426751592356688, "train_speed(iter/s)": 1.509223 }, { "epoch": 0.41771989203547405, "grad_norm": 3.7137451171875, "learning_rate": 9.828771692222284e-05, "loss": 0.2725858688354492, "memory(GiB)": 61.91, "step": 9750, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.509202 }, { "epoch": 0.41793410736472303, "grad_norm": 1.2882426977157593, "learning_rate": 9.828597039003544e-05, "loss": 0.45036468505859373, "memory(GiB)": 61.91, "step": 9755, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.509224 }, { "epoch": 0.418148322693972, "grad_norm": 6.886036396026611, "learning_rate": 9.82842229831024e-05, "loss": 0.903559398651123, "memory(GiB)": 61.91, "step": 9760, "token_acc": 0.825, "train_speed(iter/s)": 1.509253 }, { "epoch": 0.41836253802322093, "grad_norm": 8.46239185333252, "learning_rate": 9.828247470145535e-05, "loss": 0.7579894542694092, "memory(GiB)": 61.91, "step": 9765, "token_acc": 0.8267477203647416, "train_speed(iter/s)": 1.509204 }, { "epoch": 0.4185767533524699, "grad_norm": 1.8277407884597778, "learning_rate": 9.828072554512597e-05, "loss": 0.35467545986175536, "memory(GiB)": 61.91, "step": 9770, "token_acc": 0.9240506329113924, "train_speed(iter/s)": 1.509175 }, { "epoch": 0.4187909686817189, "grad_norm": 3.2110085487365723, "learning_rate": 9.827897551414598e-05, "loss": 0.36563384532928467, "memory(GiB)": 61.91, "step": 9775, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.509261 }, { "epoch": 0.4190051840109678, "grad_norm": 1.0927865505218506, "learning_rate": 9.827722460854705e-05, "loss": 0.4125990867614746, "memory(GiB)": 61.91, "step": 9780, "token_acc": 0.8996539792387543, "train_speed(iter/s)": 1.509302 }, { "epoch": 0.4192193993402168, "grad_norm": 3.430699586868286, "learning_rate": 9.82754728283609e-05, "loss": 0.4949087142944336, "memory(GiB)": 61.91, "step": 9785, "token_acc": 0.9151515151515152, "train_speed(iter/s)": 1.509314 }, { "epoch": 0.41943361466946577, "grad_norm": 1.2635524272918701, "learning_rate": 9.827372017361929e-05, "loss": 0.24602158069610597, "memory(GiB)": 61.91, "step": 9790, "token_acc": 0.9202898550724637, "train_speed(iter/s)": 1.509288 }, { "epoch": 0.4196478299987147, "grad_norm": 2.2388784885406494, "learning_rate": 9.827196664435394e-05, "loss": 0.3816941738128662, "memory(GiB)": 61.91, "step": 9795, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.509306 }, { "epoch": 0.4198620453279637, "grad_norm": 3.044820785522461, "learning_rate": 9.827021224059663e-05, "loss": 0.3346749544143677, "memory(GiB)": 61.91, "step": 9800, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.509317 }, { "epoch": 0.42007626065721265, "grad_norm": 5.654035568237305, "learning_rate": 9.826845696237917e-05, "loss": 0.3811981678009033, "memory(GiB)": 61.91, "step": 9805, "token_acc": 0.9105431309904153, "train_speed(iter/s)": 1.509247 }, { "epoch": 0.4202904759864616, "grad_norm": 6.331050395965576, "learning_rate": 9.826670080973331e-05, "loss": 0.38805584907531737, "memory(GiB)": 61.91, "step": 9810, "token_acc": 0.9136690647482014, "train_speed(iter/s)": 1.509334 }, { "epoch": 0.42050469131571055, "grad_norm": 1.1610158681869507, "learning_rate": 9.826494378269088e-05, "loss": 0.526564073562622, "memory(GiB)": 61.91, "step": 9815, "token_acc": 0.8783783783783784, "train_speed(iter/s)": 1.509404 }, { "epoch": 0.42071890664495953, "grad_norm": 1.2386915683746338, "learning_rate": 9.826318588128373e-05, "loss": 0.17634134292602538, "memory(GiB)": 61.91, "step": 9820, "token_acc": 0.958904109589041, "train_speed(iter/s)": 1.509409 }, { "epoch": 0.42093312197420846, "grad_norm": 2.638221025466919, "learning_rate": 9.82614271055437e-05, "loss": 0.22216193675994872, "memory(GiB)": 61.91, "step": 9825, "token_acc": 0.9466192170818505, "train_speed(iter/s)": 1.509384 }, { "epoch": 0.42114733730345744, "grad_norm": 1.9915095567703247, "learning_rate": 9.825966745550262e-05, "loss": 0.3601616144180298, "memory(GiB)": 61.91, "step": 9830, "token_acc": 0.921875, "train_speed(iter/s)": 1.509378 }, { "epoch": 0.4213615526327064, "grad_norm": 5.364609241485596, "learning_rate": 9.825790693119241e-05, "loss": 0.5846957206726074, "memory(GiB)": 61.91, "step": 9835, "token_acc": 0.8781362007168458, "train_speed(iter/s)": 1.509456 }, { "epoch": 0.42157576796195534, "grad_norm": 2.3219292163848877, "learning_rate": 9.825614553264495e-05, "loss": 0.30610995292663573, "memory(GiB)": 61.91, "step": 9840, "token_acc": 0.946875, "train_speed(iter/s)": 1.509445 }, { "epoch": 0.4217899832912043, "grad_norm": 4.356470108032227, "learning_rate": 9.825438325989214e-05, "loss": 0.45871548652648925, "memory(GiB)": 61.91, "step": 9845, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.509389 }, { "epoch": 0.4220041986204533, "grad_norm": 2.4539968967437744, "learning_rate": 9.825262011296591e-05, "loss": 0.4549094200134277, "memory(GiB)": 61.91, "step": 9850, "token_acc": 0.9375, "train_speed(iter/s)": 1.509459 }, { "epoch": 0.4222184139497022, "grad_norm": 4.017759799957275, "learning_rate": 9.825085609189821e-05, "loss": 0.3656139850616455, "memory(GiB)": 61.91, "step": 9855, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.509411 }, { "epoch": 0.4224326292789512, "grad_norm": 0.27963870763778687, "learning_rate": 9.824909119672098e-05, "loss": 0.2725200653076172, "memory(GiB)": 61.91, "step": 9860, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.509408 }, { "epoch": 0.4226468446082002, "grad_norm": 5.129289150238037, "learning_rate": 9.82473254274662e-05, "loss": 0.3880021333694458, "memory(GiB)": 61.91, "step": 9865, "token_acc": 0.900709219858156, "train_speed(iter/s)": 1.509388 }, { "epoch": 0.4228610599374491, "grad_norm": 0.6168048977851868, "learning_rate": 9.824555878416586e-05, "loss": 0.3714810848236084, "memory(GiB)": 61.91, "step": 9870, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.509439 }, { "epoch": 0.4230752752666981, "grad_norm": 1.9886531829833984, "learning_rate": 9.824379126685197e-05, "loss": 0.40896944999694823, "memory(GiB)": 61.91, "step": 9875, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.509397 }, { "epoch": 0.42328949059594706, "grad_norm": 6.798762798309326, "learning_rate": 9.824202287555655e-05, "loss": 0.7304451942443848, "memory(GiB)": 61.91, "step": 9880, "token_acc": 0.8343558282208589, "train_speed(iter/s)": 1.509423 }, { "epoch": 0.423503705925196, "grad_norm": 2.0493037700653076, "learning_rate": 9.824025361031162e-05, "loss": 0.41022186279296874, "memory(GiB)": 61.91, "step": 9885, "token_acc": 0.9157088122605364, "train_speed(iter/s)": 1.509486 }, { "epoch": 0.42371792125444496, "grad_norm": 3.376983642578125, "learning_rate": 9.823848347114925e-05, "loss": 0.15344417095184326, "memory(GiB)": 61.91, "step": 9890, "token_acc": 0.9661538461538461, "train_speed(iter/s)": 1.509481 }, { "epoch": 0.42393213658369394, "grad_norm": 3.84713077545166, "learning_rate": 9.82367124581015e-05, "loss": 0.5537586212158203, "memory(GiB)": 61.91, "step": 9895, "token_acc": 0.890728476821192, "train_speed(iter/s)": 1.509483 }, { "epoch": 0.42414635191294286, "grad_norm": 3.265084743499756, "learning_rate": 9.823494057120046e-05, "loss": 0.4434359550476074, "memory(GiB)": 61.91, "step": 9900, "token_acc": 0.9, "train_speed(iter/s)": 1.509469 }, { "epoch": 0.42436056724219184, "grad_norm": 3.947221040725708, "learning_rate": 9.82331678104782e-05, "loss": 0.3803727388381958, "memory(GiB)": 61.91, "step": 9905, "token_acc": 0.9307958477508651, "train_speed(iter/s)": 1.509456 }, { "epoch": 0.4245747825714408, "grad_norm": 1.523737907409668, "learning_rate": 9.823139417596688e-05, "loss": 0.40721569061279295, "memory(GiB)": 61.91, "step": 9910, "token_acc": 0.90625, "train_speed(iter/s)": 1.509407 }, { "epoch": 0.42478899790068975, "grad_norm": 1.0422919988632202, "learning_rate": 9.822961966769861e-05, "loss": 0.567678689956665, "memory(GiB)": 61.91, "step": 9915, "token_acc": 0.8703703703703703, "train_speed(iter/s)": 1.509365 }, { "epoch": 0.4250032132299387, "grad_norm": 7.92480993270874, "learning_rate": 9.822784428570552e-05, "loss": 0.31767802238464354, "memory(GiB)": 61.91, "step": 9920, "token_acc": 0.9073359073359073, "train_speed(iter/s)": 1.509318 }, { "epoch": 0.4252174285591877, "grad_norm": 2.9975814819335938, "learning_rate": 9.822606803001981e-05, "loss": 0.4217508792877197, "memory(GiB)": 61.91, "step": 9925, "token_acc": 0.9044368600682594, "train_speed(iter/s)": 1.509257 }, { "epoch": 0.4254316438884367, "grad_norm": 0.7514002323150635, "learning_rate": 9.822429090067363e-05, "loss": 0.42324223518371584, "memory(GiB)": 61.91, "step": 9930, "token_acc": 0.9196141479099679, "train_speed(iter/s)": 1.509256 }, { "epoch": 0.4256458592176856, "grad_norm": 3.9050991535186768, "learning_rate": 9.822251289769917e-05, "loss": 0.4803775787353516, "memory(GiB)": 61.91, "step": 9935, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.50942 }, { "epoch": 0.4258600745469346, "grad_norm": 1.1282000541687012, "learning_rate": 9.822073402112867e-05, "loss": 0.36203899383544924, "memory(GiB)": 61.91, "step": 9940, "token_acc": 0.9140625, "train_speed(iter/s)": 1.509399 }, { "epoch": 0.42607428987618357, "grad_norm": 5.544679641723633, "learning_rate": 9.821895427099434e-05, "loss": 0.5556458473205567, "memory(GiB)": 61.91, "step": 9945, "token_acc": 0.8977635782747604, "train_speed(iter/s)": 1.509476 }, { "epoch": 0.4262885052054325, "grad_norm": 1.1807975769042969, "learning_rate": 9.821717364732841e-05, "loss": 0.2382129430770874, "memory(GiB)": 61.91, "step": 9950, "token_acc": 0.9483282674772037, "train_speed(iter/s)": 1.509422 }, { "epoch": 0.42650272053468147, "grad_norm": 1.265651822090149, "learning_rate": 9.821539215016314e-05, "loss": 0.4865591049194336, "memory(GiB)": 61.91, "step": 9955, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.50939 }, { "epoch": 0.42671693586393045, "grad_norm": 1.8016465902328491, "learning_rate": 9.821360977953083e-05, "loss": 0.5596308231353759, "memory(GiB)": 61.91, "step": 9960, "token_acc": 0.8794117647058823, "train_speed(iter/s)": 1.509387 }, { "epoch": 0.42693115119317937, "grad_norm": 3.535043239593506, "learning_rate": 9.821182653546374e-05, "loss": 0.5865505695343017, "memory(GiB)": 61.91, "step": 9965, "token_acc": 0.8861538461538462, "train_speed(iter/s)": 1.509323 }, { "epoch": 0.42714536652242835, "grad_norm": 2.6049630641937256, "learning_rate": 9.821004241799419e-05, "loss": 0.5108200550079346, "memory(GiB)": 61.91, "step": 9970, "token_acc": 0.8616352201257862, "train_speed(iter/s)": 1.509299 }, { "epoch": 0.42735958185167733, "grad_norm": 2.7770256996154785, "learning_rate": 9.82082574271545e-05, "loss": 0.5152698516845703, "memory(GiB)": 61.91, "step": 9975, "token_acc": 0.88671875, "train_speed(iter/s)": 1.509293 }, { "epoch": 0.42757379718092625, "grad_norm": 2.4061505794525146, "learning_rate": 9.8206471562977e-05, "loss": 0.47693510055541993, "memory(GiB)": 61.91, "step": 9980, "token_acc": 0.8921161825726142, "train_speed(iter/s)": 1.509336 }, { "epoch": 0.42778801251017523, "grad_norm": 1.4498419761657715, "learning_rate": 9.820468482549403e-05, "loss": 0.4340983390808105, "memory(GiB)": 61.91, "step": 9985, "token_acc": 0.8982035928143712, "train_speed(iter/s)": 1.50938 }, { "epoch": 0.4280022278394242, "grad_norm": 1.6837610006332397, "learning_rate": 9.8202897214738e-05, "loss": 0.452302885055542, "memory(GiB)": 61.91, "step": 9990, "token_acc": 0.8897338403041825, "train_speed(iter/s)": 1.509383 }, { "epoch": 0.42821644316867313, "grad_norm": 3.5070812702178955, "learning_rate": 9.820110873074127e-05, "loss": 0.6498747825622558, "memory(GiB)": 61.91, "step": 9995, "token_acc": 0.872791519434629, "train_speed(iter/s)": 1.509408 }, { "epoch": 0.4284306584979221, "grad_norm": 3.213179111480713, "learning_rate": 9.819931937353622e-05, "loss": 0.4940351963043213, "memory(GiB)": 61.91, "step": 10000, "token_acc": 0.891156462585034, "train_speed(iter/s)": 1.509432 }, { "epoch": 0.4284306584979221, "eval_loss": 2.435257911682129, "eval_runtime": 13.258, "eval_samples_per_second": 7.543, "eval_steps_per_second": 7.543, "eval_token_acc": 0.4369747899159664, "step": 10000 }, { "epoch": 0.4286448738271711, "grad_norm": 1.8666521310806274, "learning_rate": 9.81975291431553e-05, "loss": 0.5318268299102783, "memory(GiB)": 61.91, "step": 10005, "token_acc": 0.5501043841336117, "train_speed(iter/s)": 1.506265 }, { "epoch": 0.42885908915642, "grad_norm": 2.5225818157196045, "learning_rate": 9.819573803963092e-05, "loss": 0.17265326976776124, "memory(GiB)": 61.91, "step": 10010, "token_acc": 0.9634146341463414, "train_speed(iter/s)": 1.506268 }, { "epoch": 0.429073304485669, "grad_norm": 2.483388662338257, "learning_rate": 9.819394606299552e-05, "loss": 0.49025769233703614, "memory(GiB)": 61.91, "step": 10015, "token_acc": 0.8741258741258742, "train_speed(iter/s)": 1.506254 }, { "epoch": 0.429287519814918, "grad_norm": 3.1599504947662354, "learning_rate": 9.819215321328159e-05, "loss": 0.3379845380783081, "memory(GiB)": 61.91, "step": 10020, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.506343 }, { "epoch": 0.4295017351441669, "grad_norm": 3.399993419647217, "learning_rate": 9.81903594905216e-05, "loss": 0.5720295429229736, "memory(GiB)": 61.91, "step": 10025, "token_acc": 0.8945578231292517, "train_speed(iter/s)": 1.506318 }, { "epoch": 0.4297159504734159, "grad_norm": 2.936277151107788, "learning_rate": 9.818856489474803e-05, "loss": 0.2937993049621582, "memory(GiB)": 61.91, "step": 10030, "token_acc": 0.9256965944272446, "train_speed(iter/s)": 1.506304 }, { "epoch": 0.42993016580266485, "grad_norm": 1.603687047958374, "learning_rate": 9.818676942599343e-05, "loss": 0.3212160587310791, "memory(GiB)": 61.91, "step": 10035, "token_acc": 0.9172185430463576, "train_speed(iter/s)": 1.506284 }, { "epoch": 0.4301443811319138, "grad_norm": 2.743325710296631, "learning_rate": 9.818497308429028e-05, "loss": 0.484254789352417, "memory(GiB)": 61.91, "step": 10040, "token_acc": 0.9014084507042254, "train_speed(iter/s)": 1.506264 }, { "epoch": 0.43035859646116276, "grad_norm": 4.657649993896484, "learning_rate": 9.818317586967114e-05, "loss": 0.3377114772796631, "memory(GiB)": 61.91, "step": 10045, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.506268 }, { "epoch": 0.43057281179041174, "grad_norm": 3.3420567512512207, "learning_rate": 9.818137778216857e-05, "loss": 0.21889607906341552, "memory(GiB)": 61.91, "step": 10050, "token_acc": 0.9433198380566802, "train_speed(iter/s)": 1.506272 }, { "epoch": 0.43078702711966066, "grad_norm": 2.5307488441467285, "learning_rate": 9.817957882181514e-05, "loss": 0.2744532585144043, "memory(GiB)": 61.91, "step": 10055, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.506305 }, { "epoch": 0.43100124244890964, "grad_norm": 2.3786802291870117, "learning_rate": 9.817777898864345e-05, "loss": 0.2986316442489624, "memory(GiB)": 61.91, "step": 10060, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.506375 }, { "epoch": 0.4312154577781586, "grad_norm": 1.7379429340362549, "learning_rate": 9.81759782826861e-05, "loss": 0.28394713401794436, "memory(GiB)": 61.91, "step": 10065, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.506409 }, { "epoch": 0.43142967310740754, "grad_norm": 0.3592197597026825, "learning_rate": 9.817417670397571e-05, "loss": 0.7175737380981445, "memory(GiB)": 61.91, "step": 10070, "token_acc": 0.8445945945945946, "train_speed(iter/s)": 1.506406 }, { "epoch": 0.4316438884366565, "grad_norm": 3.4017586708068848, "learning_rate": 9.817237425254492e-05, "loss": 0.46547822952270507, "memory(GiB)": 61.91, "step": 10075, "token_acc": 0.8960244648318043, "train_speed(iter/s)": 1.506355 }, { "epoch": 0.4318581037659055, "grad_norm": 2.834467649459839, "learning_rate": 9.817057092842639e-05, "loss": 0.5085031986236572, "memory(GiB)": 61.91, "step": 10080, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.506347 }, { "epoch": 0.4320723190951544, "grad_norm": 2.967838764190674, "learning_rate": 9.816876673165276e-05, "loss": 0.2504267692565918, "memory(GiB)": 61.91, "step": 10085, "token_acc": 0.919831223628692, "train_speed(iter/s)": 1.506324 }, { "epoch": 0.4322865344244034, "grad_norm": 8.58832836151123, "learning_rate": 9.816696166225674e-05, "loss": 0.2683766603469849, "memory(GiB)": 61.91, "step": 10090, "token_acc": 0.936026936026936, "train_speed(iter/s)": 1.506374 }, { "epoch": 0.4325007497536524, "grad_norm": 2.149585723876953, "learning_rate": 9.816515572027103e-05, "loss": 0.46616153717041015, "memory(GiB)": 61.91, "step": 10095, "token_acc": 0.8862745098039215, "train_speed(iter/s)": 1.50639 }, { "epoch": 0.43271496508290136, "grad_norm": 5.421866416931152, "learning_rate": 9.816334890572834e-05, "loss": 0.5225001335144043, "memory(GiB)": 61.91, "step": 10100, "token_acc": 0.8789808917197452, "train_speed(iter/s)": 1.506529 }, { "epoch": 0.4329291804121503, "grad_norm": 8.108086585998535, "learning_rate": 9.81615412186614e-05, "loss": 0.42041769027709963, "memory(GiB)": 61.91, "step": 10105, "token_acc": 0.9218106995884774, "train_speed(iter/s)": 1.506498 }, { "epoch": 0.43314339574139926, "grad_norm": 4.096609592437744, "learning_rate": 9.815973265910296e-05, "loss": 0.45472373962402346, "memory(GiB)": 61.91, "step": 10110, "token_acc": 0.9116465863453815, "train_speed(iter/s)": 1.50648 }, { "epoch": 0.43335761107064824, "grad_norm": 4.647186279296875, "learning_rate": 9.815792322708579e-05, "loss": 0.26973814964294435, "memory(GiB)": 61.91, "step": 10115, "token_acc": 0.9267399267399268, "train_speed(iter/s)": 1.506465 }, { "epoch": 0.43357182639989716, "grad_norm": 2.0985701084136963, "learning_rate": 9.815611292264267e-05, "loss": 0.3024921894073486, "memory(GiB)": 61.91, "step": 10120, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.506466 }, { "epoch": 0.43378604172914614, "grad_norm": 5.445104598999023, "learning_rate": 9.815430174580638e-05, "loss": 0.35630154609680176, "memory(GiB)": 61.91, "step": 10125, "token_acc": 0.92, "train_speed(iter/s)": 1.506544 }, { "epoch": 0.4340002570583951, "grad_norm": 4.421397686004639, "learning_rate": 9.815248969660975e-05, "loss": 0.702651834487915, "memory(GiB)": 61.91, "step": 10130, "token_acc": 0.8037383177570093, "train_speed(iter/s)": 1.506501 }, { "epoch": 0.43421447238764405, "grad_norm": 1.0670167207717896, "learning_rate": 9.81506767750856e-05, "loss": 0.3192596435546875, "memory(GiB)": 61.91, "step": 10135, "token_acc": 0.9322709163346613, "train_speed(iter/s)": 1.506503 }, { "epoch": 0.434428687716893, "grad_norm": 2.919353485107422, "learning_rate": 9.814886298126678e-05, "loss": 0.4952259063720703, "memory(GiB)": 61.91, "step": 10140, "token_acc": 0.9075342465753424, "train_speed(iter/s)": 1.506504 }, { "epoch": 0.434642903046142, "grad_norm": 2.230673313140869, "learning_rate": 9.814704831518612e-05, "loss": 0.34040088653564454, "memory(GiB)": 61.91, "step": 10145, "token_acc": 0.9273504273504274, "train_speed(iter/s)": 1.506488 }, { "epoch": 0.4348571183753909, "grad_norm": 2.02935791015625, "learning_rate": 9.814523277687651e-05, "loss": 0.2873332977294922, "memory(GiB)": 61.91, "step": 10150, "token_acc": 0.9400630914826499, "train_speed(iter/s)": 1.506489 }, { "epoch": 0.4350713337046399, "grad_norm": 0.7565959095954895, "learning_rate": 9.814341636637085e-05, "loss": 0.18743935823440552, "memory(GiB)": 61.91, "step": 10155, "token_acc": 0.9581881533101045, "train_speed(iter/s)": 1.5065 }, { "epoch": 0.4352855490338889, "grad_norm": 1.0379352569580078, "learning_rate": 9.814159908370206e-05, "loss": 0.37758305072784426, "memory(GiB)": 61.91, "step": 10160, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.506441 }, { "epoch": 0.4354997643631378, "grad_norm": 1.4163607358932495, "learning_rate": 9.813978092890302e-05, "loss": 0.30896997451782227, "memory(GiB)": 61.91, "step": 10165, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.506399 }, { "epoch": 0.4357139796923868, "grad_norm": 1.7334789037704468, "learning_rate": 9.813796190200671e-05, "loss": 0.43746585845947267, "memory(GiB)": 61.91, "step": 10170, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.506396 }, { "epoch": 0.43592819502163577, "grad_norm": 1.3903981447219849, "learning_rate": 9.813614200304604e-05, "loss": 0.5260143280029297, "memory(GiB)": 61.91, "step": 10175, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.506368 }, { "epoch": 0.4361424103508847, "grad_norm": 4.0374650955200195, "learning_rate": 9.813432123205401e-05, "loss": 0.4306884765625, "memory(GiB)": 61.91, "step": 10180, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.506386 }, { "epoch": 0.43635662568013367, "grad_norm": 3.235056161880493, "learning_rate": 9.813249958906362e-05, "loss": 0.5119011878967286, "memory(GiB)": 61.91, "step": 10185, "token_acc": 0.891640866873065, "train_speed(iter/s)": 1.506315 }, { "epoch": 0.43657084100938265, "grad_norm": 1.0276715755462646, "learning_rate": 9.813067707410781e-05, "loss": 0.3745177507400513, "memory(GiB)": 61.91, "step": 10190, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.506349 }, { "epoch": 0.43678505633863157, "grad_norm": 2.0960819721221924, "learning_rate": 9.812885368721966e-05, "loss": 0.42377629280090334, "memory(GiB)": 61.91, "step": 10195, "token_acc": 0.9027355623100304, "train_speed(iter/s)": 1.506338 }, { "epoch": 0.43699927166788055, "grad_norm": 3.2238616943359375, "learning_rate": 9.812702942843218e-05, "loss": 0.8300793647766114, "memory(GiB)": 61.91, "step": 10200, "token_acc": 0.7937853107344632, "train_speed(iter/s)": 1.506286 }, { "epoch": 0.43721348699712953, "grad_norm": 2.4800775051116943, "learning_rate": 9.812520429777839e-05, "loss": 0.4394670009613037, "memory(GiB)": 61.91, "step": 10205, "token_acc": 0.9171597633136095, "train_speed(iter/s)": 1.506346 }, { "epoch": 0.43742770232637845, "grad_norm": 2.0256688594818115, "learning_rate": 9.81233782952914e-05, "loss": 0.3772881507873535, "memory(GiB)": 61.91, "step": 10210, "token_acc": 0.9142091152815014, "train_speed(iter/s)": 1.506329 }, { "epoch": 0.43764191765562743, "grad_norm": 8.301093101501465, "learning_rate": 9.812155142100425e-05, "loss": 0.3515757083892822, "memory(GiB)": 61.91, "step": 10215, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.50633 }, { "epoch": 0.4378561329848764, "grad_norm": 2.299423933029175, "learning_rate": 9.811972367495008e-05, "loss": 0.3472987174987793, "memory(GiB)": 61.91, "step": 10220, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.506379 }, { "epoch": 0.43807034831412534, "grad_norm": 0.8623754382133484, "learning_rate": 9.811789505716195e-05, "loss": 0.26234433650970457, "memory(GiB)": 61.91, "step": 10225, "token_acc": 0.9571984435797666, "train_speed(iter/s)": 1.506349 }, { "epoch": 0.4382845636433743, "grad_norm": 2.46502423286438, "learning_rate": 9.811606556767303e-05, "loss": 0.44746694564819334, "memory(GiB)": 61.91, "step": 10230, "token_acc": 0.9018567639257294, "train_speed(iter/s)": 1.506327 }, { "epoch": 0.4384987789726233, "grad_norm": 3.469413995742798, "learning_rate": 9.811423520651644e-05, "loss": 0.3888083457946777, "memory(GiB)": 61.91, "step": 10235, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.506324 }, { "epoch": 0.4387129943018722, "grad_norm": 2.617112636566162, "learning_rate": 9.811240397372535e-05, "loss": 0.5537193298339844, "memory(GiB)": 61.91, "step": 10240, "token_acc": 0.8682432432432432, "train_speed(iter/s)": 1.506396 }, { "epoch": 0.4389272096311212, "grad_norm": 3.364072799682617, "learning_rate": 9.811057186933293e-05, "loss": 0.30262365341186526, "memory(GiB)": 61.91, "step": 10245, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.506456 }, { "epoch": 0.4391414249603702, "grad_norm": 13.93198299407959, "learning_rate": 9.810873889337235e-05, "loss": 0.46721415519714354, "memory(GiB)": 61.91, "step": 10250, "token_acc": 0.8909774436090225, "train_speed(iter/s)": 1.506508 }, { "epoch": 0.4393556402896191, "grad_norm": 4.195378303527832, "learning_rate": 9.810690504587685e-05, "loss": 0.28487725257873536, "memory(GiB)": 61.91, "step": 10255, "token_acc": 0.9588607594936709, "train_speed(iter/s)": 1.506504 }, { "epoch": 0.4395698556188681, "grad_norm": 3.779521942138672, "learning_rate": 9.810507032687964e-05, "loss": 0.5166985034942627, "memory(GiB)": 61.91, "step": 10260, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.506463 }, { "epoch": 0.43978407094811706, "grad_norm": 1.2961277961730957, "learning_rate": 9.810323473641395e-05, "loss": 0.10728344917297364, "memory(GiB)": 61.91, "step": 10265, "token_acc": 0.9899328859060402, "train_speed(iter/s)": 1.50643 }, { "epoch": 0.43999828627736604, "grad_norm": 2.037714958190918, "learning_rate": 9.810139827451305e-05, "loss": 0.2773106098175049, "memory(GiB)": 61.91, "step": 10270, "token_acc": 0.9368770764119602, "train_speed(iter/s)": 1.506424 }, { "epoch": 0.44021250160661496, "grad_norm": 2.567735433578491, "learning_rate": 9.809956094121017e-05, "loss": 0.3679696559906006, "memory(GiB)": 61.91, "step": 10275, "token_acc": 0.9228395061728395, "train_speed(iter/s)": 1.506503 }, { "epoch": 0.44042671693586394, "grad_norm": 3.379956007003784, "learning_rate": 9.809772273653866e-05, "loss": 0.4297483444213867, "memory(GiB)": 61.91, "step": 10280, "token_acc": 0.921875, "train_speed(iter/s)": 1.506455 }, { "epoch": 0.4406409322651129, "grad_norm": 2.2658281326293945, "learning_rate": 9.809588366053175e-05, "loss": 0.4789607524871826, "memory(GiB)": 61.91, "step": 10285, "token_acc": 0.9049295774647887, "train_speed(iter/s)": 1.50653 }, { "epoch": 0.44085514759436184, "grad_norm": 2.6683194637298584, "learning_rate": 9.80940437132228e-05, "loss": 0.410296106338501, "memory(GiB)": 61.91, "step": 10290, "token_acc": 0.9176829268292683, "train_speed(iter/s)": 1.506518 }, { "epoch": 0.4410693629236108, "grad_norm": 1.707667589187622, "learning_rate": 9.809220289464513e-05, "loss": 0.7021047115325928, "memory(GiB)": 61.91, "step": 10295, "token_acc": 0.856, "train_speed(iter/s)": 1.506463 }, { "epoch": 0.4412835782528598, "grad_norm": 0.33640891313552856, "learning_rate": 9.809036120483211e-05, "loss": 0.320736289024353, "memory(GiB)": 61.91, "step": 10300, "token_acc": 0.9084967320261438, "train_speed(iter/s)": 1.506529 }, { "epoch": 0.4414977935821087, "grad_norm": 6.469146251678467, "learning_rate": 9.808851864381706e-05, "loss": 0.4157205581665039, "memory(GiB)": 61.91, "step": 10305, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.506593 }, { "epoch": 0.4417120089113577, "grad_norm": 7.494106769561768, "learning_rate": 9.80866752116334e-05, "loss": 0.36469104290008547, "memory(GiB)": 61.91, "step": 10310, "token_acc": 0.8884758364312267, "train_speed(iter/s)": 1.506544 }, { "epoch": 0.4419262242406067, "grad_norm": 0.9967930912971497, "learning_rate": 9.80848309083145e-05, "loss": 0.435056734085083, "memory(GiB)": 61.91, "step": 10315, "token_acc": 0.9, "train_speed(iter/s)": 1.506517 }, { "epoch": 0.4421404395698556, "grad_norm": 3.689375162124634, "learning_rate": 9.808298573389379e-05, "loss": 0.3180785417556763, "memory(GiB)": 61.91, "step": 10320, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.506519 }, { "epoch": 0.4423546548991046, "grad_norm": 6.299117088317871, "learning_rate": 9.808113968840468e-05, "loss": 0.5036754608154297, "memory(GiB)": 61.91, "step": 10325, "token_acc": 0.8952702702702703, "train_speed(iter/s)": 1.506592 }, { "epoch": 0.44256887022835356, "grad_norm": 5.919503211975098, "learning_rate": 9.807929277188061e-05, "loss": 0.2616488218307495, "memory(GiB)": 61.91, "step": 10330, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.506611 }, { "epoch": 0.4427830855576025, "grad_norm": 3.2891223430633545, "learning_rate": 9.807744498435507e-05, "loss": 0.20905542373657227, "memory(GiB)": 61.91, "step": 10335, "token_acc": 0.9632352941176471, "train_speed(iter/s)": 1.506615 }, { "epoch": 0.44299730088685146, "grad_norm": 1.2426469326019287, "learning_rate": 9.80755963258615e-05, "loss": 0.5568090438842773, "memory(GiB)": 61.91, "step": 10340, "token_acc": 0.9, "train_speed(iter/s)": 1.506631 }, { "epoch": 0.44321151621610044, "grad_norm": 3.02644419670105, "learning_rate": 9.807374679643342e-05, "loss": 0.4243612289428711, "memory(GiB)": 61.91, "step": 10345, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.506628 }, { "epoch": 0.44342573154534937, "grad_norm": 2.3068771362304688, "learning_rate": 9.80718963961043e-05, "loss": 0.3067525625228882, "memory(GiB)": 61.91, "step": 10350, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.506639 }, { "epoch": 0.44363994687459835, "grad_norm": 3.1874725818634033, "learning_rate": 9.807004512490769e-05, "loss": 0.38964645862579345, "memory(GiB)": 61.91, "step": 10355, "token_acc": 0.9057971014492754, "train_speed(iter/s)": 1.506636 }, { "epoch": 0.4438541622038473, "grad_norm": 2.043302536010742, "learning_rate": 9.806819298287713e-05, "loss": 0.9457289695739746, "memory(GiB)": 61.91, "step": 10360, "token_acc": 0.7928176795580111, "train_speed(iter/s)": 1.506811 }, { "epoch": 0.44406837753309625, "grad_norm": 2.440476179122925, "learning_rate": 9.806633997004615e-05, "loss": 0.35591745376586914, "memory(GiB)": 61.91, "step": 10365, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.506765 }, { "epoch": 0.4442825928623452, "grad_norm": 7.384896278381348, "learning_rate": 9.806448608644834e-05, "loss": 0.29206008911132814, "memory(GiB)": 61.91, "step": 10370, "token_acc": 0.9369369369369369, "train_speed(iter/s)": 1.506853 }, { "epoch": 0.4444968081915942, "grad_norm": 3.2284624576568604, "learning_rate": 9.806263133211728e-05, "loss": 0.10596487522125245, "memory(GiB)": 61.91, "step": 10375, "token_acc": 0.9857142857142858, "train_speed(iter/s)": 1.506813 }, { "epoch": 0.44471102352084313, "grad_norm": 2.0987548828125, "learning_rate": 9.806077570708654e-05, "loss": 0.28821663856506347, "memory(GiB)": 61.91, "step": 10380, "token_acc": 0.9471698113207547, "train_speed(iter/s)": 1.506818 }, { "epoch": 0.4449252388500921, "grad_norm": 5.4336137771606445, "learning_rate": 9.805891921138979e-05, "loss": 0.6694289207458496, "memory(GiB)": 61.91, "step": 10385, "token_acc": 0.8689024390243902, "train_speed(iter/s)": 1.50684 }, { "epoch": 0.4451394541793411, "grad_norm": 0.22378452122211456, "learning_rate": 9.805706184506062e-05, "loss": 0.6008909702301025, "memory(GiB)": 61.91, "step": 10390, "token_acc": 0.8923076923076924, "train_speed(iter/s)": 1.506794 }, { "epoch": 0.44535366950859, "grad_norm": 2.9686124324798584, "learning_rate": 9.805520360813272e-05, "loss": 0.4844522476196289, "memory(GiB)": 61.91, "step": 10395, "token_acc": 0.8896551724137931, "train_speed(iter/s)": 1.506818 }, { "epoch": 0.445567884837839, "grad_norm": 2.4190595149993896, "learning_rate": 9.80533445006397e-05, "loss": 0.16563622951507567, "memory(GiB)": 61.91, "step": 10400, "token_acc": 0.958041958041958, "train_speed(iter/s)": 1.506797 }, { "epoch": 0.44578210016708797, "grad_norm": 2.4079484939575195, "learning_rate": 9.805148452261528e-05, "loss": 0.5499263763427734, "memory(GiB)": 61.91, "step": 10405, "token_acc": 0.8937007874015748, "train_speed(iter/s)": 1.506776 }, { "epoch": 0.4459963154963369, "grad_norm": 2.3879342079162598, "learning_rate": 9.804962367409313e-05, "loss": 0.38434929847717286, "memory(GiB)": 61.91, "step": 10410, "token_acc": 0.9148264984227129, "train_speed(iter/s)": 1.506823 }, { "epoch": 0.44621053082558587, "grad_norm": 8.252806663513184, "learning_rate": 9.804776195510699e-05, "loss": 0.24905743598937988, "memory(GiB)": 61.91, "step": 10415, "token_acc": 0.9297124600638977, "train_speed(iter/s)": 1.506741 }, { "epoch": 0.44642474615483485, "grad_norm": 0.3555503785610199, "learning_rate": 9.804589936569055e-05, "loss": 0.4679290294647217, "memory(GiB)": 61.91, "step": 10420, "token_acc": 0.9037037037037037, "train_speed(iter/s)": 1.506768 }, { "epoch": 0.4466389614840838, "grad_norm": 2.41595721244812, "learning_rate": 9.804403590587758e-05, "loss": 0.25032830238342285, "memory(GiB)": 61.91, "step": 10425, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.506727 }, { "epoch": 0.44685317681333275, "grad_norm": 2.582745313644409, "learning_rate": 9.804217157570184e-05, "loss": 0.3428165912628174, "memory(GiB)": 61.91, "step": 10430, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.506694 }, { "epoch": 0.44706739214258173, "grad_norm": 6.5446014404296875, "learning_rate": 9.804030637519708e-05, "loss": 0.23604507446289064, "memory(GiB)": 61.91, "step": 10435, "token_acc": 0.9391534391534392, "train_speed(iter/s)": 1.506655 }, { "epoch": 0.4472816074718307, "grad_norm": 3.8948042392730713, "learning_rate": 9.803844030439711e-05, "loss": 0.4295806884765625, "memory(GiB)": 61.91, "step": 10440, "token_acc": 0.9022082018927445, "train_speed(iter/s)": 1.506786 }, { "epoch": 0.44749582280107963, "grad_norm": 1.1619501113891602, "learning_rate": 9.803657336333574e-05, "loss": 0.3608191251754761, "memory(GiB)": 61.91, "step": 10445, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.506719 }, { "epoch": 0.4477100381303286, "grad_norm": 1.391656517982483, "learning_rate": 9.803470555204676e-05, "loss": 0.4493855953216553, "memory(GiB)": 61.91, "step": 10450, "token_acc": 0.9014492753623189, "train_speed(iter/s)": 1.50671 }, { "epoch": 0.4479242534595776, "grad_norm": 3.7088119983673096, "learning_rate": 9.803283687056404e-05, "loss": 0.4485030651092529, "memory(GiB)": 61.91, "step": 10455, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.506641 }, { "epoch": 0.4481384687888265, "grad_norm": 3.627711534500122, "learning_rate": 9.803096731892142e-05, "loss": 0.3340435028076172, "memory(GiB)": 61.91, "step": 10460, "token_acc": 0.921875, "train_speed(iter/s)": 1.506665 }, { "epoch": 0.4483526841180755, "grad_norm": 4.415846824645996, "learning_rate": 9.802909689715278e-05, "loss": 0.437451696395874, "memory(GiB)": 61.91, "step": 10465, "token_acc": 0.9206896551724137, "train_speed(iter/s)": 1.506807 }, { "epoch": 0.4485668994473245, "grad_norm": 2.820772886276245, "learning_rate": 9.802722560529199e-05, "loss": 0.45885496139526366, "memory(GiB)": 61.91, "step": 10470, "token_acc": 0.9128630705394191, "train_speed(iter/s)": 1.506773 }, { "epoch": 0.4487811147765734, "grad_norm": 2.9628384113311768, "learning_rate": 9.802535344337296e-05, "loss": 0.20114028453826904, "memory(GiB)": 61.91, "step": 10475, "token_acc": 0.9511278195488722, "train_speed(iter/s)": 1.506768 }, { "epoch": 0.4489953301058224, "grad_norm": 3.156007766723633, "learning_rate": 9.80234804114296e-05, "loss": 0.5338842391967773, "memory(GiB)": 61.91, "step": 10480, "token_acc": 0.8885017421602788, "train_speed(iter/s)": 1.506801 }, { "epoch": 0.44920954543507136, "grad_norm": 2.763643264770508, "learning_rate": 9.802160650949584e-05, "loss": 0.3767467260360718, "memory(GiB)": 61.91, "step": 10485, "token_acc": 0.9122257053291536, "train_speed(iter/s)": 1.506803 }, { "epoch": 0.4494237607643203, "grad_norm": 4.43600606918335, "learning_rate": 9.801973173760562e-05, "loss": 0.46334428787231446, "memory(GiB)": 61.91, "step": 10490, "token_acc": 0.8898809523809523, "train_speed(iter/s)": 1.50686 }, { "epoch": 0.44963797609356926, "grad_norm": 1.094356894493103, "learning_rate": 9.801785609579292e-05, "loss": 0.2921835660934448, "memory(GiB)": 61.91, "step": 10495, "token_acc": 0.9283387622149837, "train_speed(iter/s)": 1.506847 }, { "epoch": 0.44985219142281824, "grad_norm": 3.7235398292541504, "learning_rate": 9.801597958409172e-05, "loss": 0.31828458309173585, "memory(GiB)": 61.91, "step": 10500, "token_acc": 0.9274193548387096, "train_speed(iter/s)": 1.506816 }, { "epoch": 0.44985219142281824, "eval_loss": 2.4952213764190674, "eval_runtime": 12.456, "eval_samples_per_second": 8.028, "eval_steps_per_second": 8.028, "eval_token_acc": 0.4251412429378531, "step": 10500 }, { "epoch": 0.45006640675206716, "grad_norm": 4.398873805999756, "learning_rate": 9.8014102202536e-05, "loss": 0.26111912727355957, "memory(GiB)": 61.91, "step": 10505, "token_acc": 0.581441263573544, "train_speed(iter/s)": 1.504005 }, { "epoch": 0.45028062208131614, "grad_norm": 2.1144871711730957, "learning_rate": 9.801222395115976e-05, "loss": 0.3501842498779297, "memory(GiB)": 61.91, "step": 10510, "token_acc": 0.9151515151515152, "train_speed(iter/s)": 1.503997 }, { "epoch": 0.4504948374105651, "grad_norm": 2.0323758125305176, "learning_rate": 9.801034482999707e-05, "loss": 0.39953956604003904, "memory(GiB)": 61.91, "step": 10515, "token_acc": 0.889967637540453, "train_speed(iter/s)": 1.504068 }, { "epoch": 0.45070905273981404, "grad_norm": 1.715956449508667, "learning_rate": 9.800846483908195e-05, "loss": 0.3058905124664307, "memory(GiB)": 61.91, "step": 10520, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.504062 }, { "epoch": 0.450923268069063, "grad_norm": 3.091902256011963, "learning_rate": 9.800658397844844e-05, "loss": 0.4938746452331543, "memory(GiB)": 61.91, "step": 10525, "token_acc": 0.8986486486486487, "train_speed(iter/s)": 1.504172 }, { "epoch": 0.451137483398312, "grad_norm": 3.273632287979126, "learning_rate": 9.800470224813064e-05, "loss": 0.5177756786346436, "memory(GiB)": 61.91, "step": 10530, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.504315 }, { "epoch": 0.4513516987275609, "grad_norm": 0.5889707207679749, "learning_rate": 9.80028196481626e-05, "loss": 0.1466607093811035, "memory(GiB)": 61.91, "step": 10535, "token_acc": 0.959375, "train_speed(iter/s)": 1.504286 }, { "epoch": 0.4515659140568099, "grad_norm": 1.571751356124878, "learning_rate": 9.800093617857846e-05, "loss": 0.2921424627304077, "memory(GiB)": 61.91, "step": 10540, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.504292 }, { "epoch": 0.4517801293860589, "grad_norm": 1.7647569179534912, "learning_rate": 9.799905183941236e-05, "loss": 0.3393446922302246, "memory(GiB)": 61.91, "step": 10545, "token_acc": 0.9399293286219081, "train_speed(iter/s)": 1.504337 }, { "epoch": 0.4519943447153078, "grad_norm": 4.799156188964844, "learning_rate": 9.799716663069838e-05, "loss": 0.4347689151763916, "memory(GiB)": 61.91, "step": 10550, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.504405 }, { "epoch": 0.4522085600445568, "grad_norm": 2.6664297580718994, "learning_rate": 9.799528055247071e-05, "loss": 0.2719453811645508, "memory(GiB)": 61.91, "step": 10555, "token_acc": 0.9566563467492261, "train_speed(iter/s)": 1.504409 }, { "epoch": 0.45242277537380576, "grad_norm": 1.509096622467041, "learning_rate": 9.799339360476352e-05, "loss": 0.38424179553985593, "memory(GiB)": 61.91, "step": 10560, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.50447 }, { "epoch": 0.4526369907030547, "grad_norm": 3.2496020793914795, "learning_rate": 9.799150578761098e-05, "loss": 0.3456047534942627, "memory(GiB)": 61.91, "step": 10565, "token_acc": 0.9243027888446215, "train_speed(iter/s)": 1.504448 }, { "epoch": 0.45285120603230367, "grad_norm": 1.9822664260864258, "learning_rate": 9.798961710104728e-05, "loss": 0.3626154661178589, "memory(GiB)": 61.91, "step": 10570, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.504515 }, { "epoch": 0.45306542136155264, "grad_norm": 4.577199459075928, "learning_rate": 9.798772754510666e-05, "loss": 0.39915966987609863, "memory(GiB)": 61.91, "step": 10575, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.504465 }, { "epoch": 0.45327963669080157, "grad_norm": 4.124240875244141, "learning_rate": 9.798583711982332e-05, "loss": 0.39151668548583984, "memory(GiB)": 61.91, "step": 10580, "token_acc": 0.9075342465753424, "train_speed(iter/s)": 1.50452 }, { "epoch": 0.45349385202005055, "grad_norm": 2.9387259483337402, "learning_rate": 9.798394582523154e-05, "loss": 0.4705965042114258, "memory(GiB)": 61.91, "step": 10585, "token_acc": 0.9235474006116208, "train_speed(iter/s)": 1.504472 }, { "epoch": 0.4537080673492995, "grad_norm": 1.9763673543930054, "learning_rate": 9.798205366136558e-05, "loss": 0.5479199409484863, "memory(GiB)": 61.91, "step": 10590, "token_acc": 0.8930817610062893, "train_speed(iter/s)": 1.504585 }, { "epoch": 0.45392228267854845, "grad_norm": 1.0732959508895874, "learning_rate": 9.79801606282597e-05, "loss": 0.423629903793335, "memory(GiB)": 61.91, "step": 10595, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.50458 }, { "epoch": 0.45413649800779743, "grad_norm": 1.7126439809799194, "learning_rate": 9.797826672594819e-05, "loss": 0.22223351001739503, "memory(GiB)": 61.91, "step": 10600, "token_acc": 0.9504132231404959, "train_speed(iter/s)": 1.504603 }, { "epoch": 0.4543507133370464, "grad_norm": 2.4809999465942383, "learning_rate": 9.797637195446538e-05, "loss": 0.3199489116668701, "memory(GiB)": 61.91, "step": 10605, "token_acc": 0.9297752808988764, "train_speed(iter/s)": 1.504577 }, { "epoch": 0.4545649286662954, "grad_norm": 2.924457311630249, "learning_rate": 9.797447631384559e-05, "loss": 0.23085687160491944, "memory(GiB)": 61.91, "step": 10610, "token_acc": 0.9553264604810997, "train_speed(iter/s)": 1.504554 }, { "epoch": 0.4547791439955443, "grad_norm": 3.415520429611206, "learning_rate": 9.797257980412316e-05, "loss": 0.19014124870300292, "memory(GiB)": 61.91, "step": 10615, "token_acc": 0.9548872180451128, "train_speed(iter/s)": 1.504516 }, { "epoch": 0.4549933593247933, "grad_norm": 1.4875752925872803, "learning_rate": 9.797068242533243e-05, "loss": 0.4415329933166504, "memory(GiB)": 61.91, "step": 10620, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.504438 }, { "epoch": 0.45520757465404227, "grad_norm": 5.121063232421875, "learning_rate": 9.79687841775078e-05, "loss": 0.4747471809387207, "memory(GiB)": 61.91, "step": 10625, "token_acc": 0.8873720136518771, "train_speed(iter/s)": 1.504483 }, { "epoch": 0.4554217899832912, "grad_norm": 2.759798288345337, "learning_rate": 9.796688506068364e-05, "loss": 0.4551729679107666, "memory(GiB)": 61.91, "step": 10630, "token_acc": 0.8989071038251366, "train_speed(iter/s)": 1.504527 }, { "epoch": 0.45563600531254017, "grad_norm": 1.0758121013641357, "learning_rate": 9.796498507489436e-05, "loss": 0.1609519124031067, "memory(GiB)": 61.91, "step": 10635, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.504554 }, { "epoch": 0.45585022064178915, "grad_norm": 4.402721881866455, "learning_rate": 9.79630842201744e-05, "loss": 0.4950451374053955, "memory(GiB)": 61.91, "step": 10640, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.504592 }, { "epoch": 0.4560644359710381, "grad_norm": 3.6649341583251953, "learning_rate": 9.796118249655814e-05, "loss": 0.3835118055343628, "memory(GiB)": 61.91, "step": 10645, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.504562 }, { "epoch": 0.45627865130028705, "grad_norm": 1.6279388666152954, "learning_rate": 9.795927990408009e-05, "loss": 0.5677229881286621, "memory(GiB)": 61.91, "step": 10650, "token_acc": 0.8914956011730205, "train_speed(iter/s)": 1.504517 }, { "epoch": 0.45649286662953603, "grad_norm": 1.250881552696228, "learning_rate": 9.79573764427747e-05, "loss": 0.3676487922668457, "memory(GiB)": 61.91, "step": 10655, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.504513 }, { "epoch": 0.45670708195878496, "grad_norm": 4.570855140686035, "learning_rate": 9.795547211267643e-05, "loss": 0.19411447048187255, "memory(GiB)": 61.91, "step": 10660, "token_acc": 0.9595588235294118, "train_speed(iter/s)": 1.504504 }, { "epoch": 0.45692129728803393, "grad_norm": 1.0910184383392334, "learning_rate": 9.795356691381983e-05, "loss": 0.3593562364578247, "memory(GiB)": 61.91, "step": 10665, "token_acc": 0.9363057324840764, "train_speed(iter/s)": 1.504475 }, { "epoch": 0.4571355126172829, "grad_norm": 1.1544795036315918, "learning_rate": 9.795166084623934e-05, "loss": 0.1543000817298889, "memory(GiB)": 61.91, "step": 10670, "token_acc": 0.9704918032786886, "train_speed(iter/s)": 1.504476 }, { "epoch": 0.45734972794653184, "grad_norm": 2.202618360519409, "learning_rate": 9.794975390996956e-05, "loss": 0.4586010932922363, "memory(GiB)": 61.91, "step": 10675, "token_acc": 0.9119496855345912, "train_speed(iter/s)": 1.504516 }, { "epoch": 0.4575639432757808, "grad_norm": 4.257575511932373, "learning_rate": 9.7947846105045e-05, "loss": 0.5711541175842285, "memory(GiB)": 61.91, "step": 10680, "token_acc": 0.8825910931174089, "train_speed(iter/s)": 1.504524 }, { "epoch": 0.4577781586050298, "grad_norm": 1.591488003730774, "learning_rate": 9.794593743150022e-05, "loss": 0.33921175003051757, "memory(GiB)": 61.91, "step": 10685, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.504562 }, { "epoch": 0.4579923739342787, "grad_norm": 3.952836751937866, "learning_rate": 9.794402788936983e-05, "loss": 0.5442875385284424, "memory(GiB)": 61.91, "step": 10690, "token_acc": 0.8875502008032129, "train_speed(iter/s)": 1.50455 }, { "epoch": 0.4582065892635277, "grad_norm": 4.114651679992676, "learning_rate": 9.794211747868838e-05, "loss": 0.9469118118286133, "memory(GiB)": 61.91, "step": 10695, "token_acc": 0.8166189111747851, "train_speed(iter/s)": 1.504526 }, { "epoch": 0.4584208045927767, "grad_norm": 9.232556343078613, "learning_rate": 9.79402061994905e-05, "loss": 0.5873137474060058, "memory(GiB)": 61.91, "step": 10700, "token_acc": 0.8844765342960289, "train_speed(iter/s)": 1.504506 }, { "epoch": 0.4586350199220256, "grad_norm": 3.37101149559021, "learning_rate": 9.793829405181081e-05, "loss": 0.2145254373550415, "memory(GiB)": 61.91, "step": 10705, "token_acc": 0.97, "train_speed(iter/s)": 1.50445 }, { "epoch": 0.4588492352512746, "grad_norm": 1.2563836574554443, "learning_rate": 9.793638103568397e-05, "loss": 0.493220853805542, "memory(GiB)": 61.91, "step": 10710, "token_acc": 0.909375, "train_speed(iter/s)": 1.504435 }, { "epoch": 0.45906345058052356, "grad_norm": 4.284435749053955, "learning_rate": 9.79344671511446e-05, "loss": 0.3350714683532715, "memory(GiB)": 61.91, "step": 10715, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.504416 }, { "epoch": 0.4592776659097725, "grad_norm": 3.2404279708862305, "learning_rate": 9.79325523982274e-05, "loss": 0.19627162218093872, "memory(GiB)": 61.91, "step": 10720, "token_acc": 0.9551282051282052, "train_speed(iter/s)": 1.504378 }, { "epoch": 0.45949188123902146, "grad_norm": 4.492916584014893, "learning_rate": 9.793063677696706e-05, "loss": 0.3719411134719849, "memory(GiB)": 61.91, "step": 10725, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.504372 }, { "epoch": 0.45970609656827044, "grad_norm": 3.3882462978363037, "learning_rate": 9.792872028739826e-05, "loss": 0.542146110534668, "memory(GiB)": 61.91, "step": 10730, "token_acc": 0.8885245901639345, "train_speed(iter/s)": 1.504299 }, { "epoch": 0.45992031189751936, "grad_norm": 5.002381324768066, "learning_rate": 9.792680292955571e-05, "loss": 0.2555576801300049, "memory(GiB)": 61.91, "step": 10735, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.50424 }, { "epoch": 0.46013452722676834, "grad_norm": 5.066009998321533, "learning_rate": 9.792488470347421e-05, "loss": 0.4372603416442871, "memory(GiB)": 61.91, "step": 10740, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.50423 }, { "epoch": 0.4603487425560173, "grad_norm": 3.823007345199585, "learning_rate": 9.792296560918844e-05, "loss": 0.43454594612121583, "memory(GiB)": 61.91, "step": 10745, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.504244 }, { "epoch": 0.46056295788526624, "grad_norm": 4.247657775878906, "learning_rate": 9.792104564673319e-05, "loss": 0.4295774459838867, "memory(GiB)": 61.91, "step": 10750, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.504248 }, { "epoch": 0.4607771732145152, "grad_norm": 4.388004779815674, "learning_rate": 9.791912481614324e-05, "loss": 0.1938075065612793, "memory(GiB)": 61.91, "step": 10755, "token_acc": 0.956, "train_speed(iter/s)": 1.504241 }, { "epoch": 0.4609913885437642, "grad_norm": 5.581959247589111, "learning_rate": 9.791720311745342e-05, "loss": 0.173551344871521, "memory(GiB)": 61.91, "step": 10760, "token_acc": 0.9644670050761421, "train_speed(iter/s)": 1.50431 }, { "epoch": 0.4612056038730131, "grad_norm": 7.675710678100586, "learning_rate": 9.791528055069849e-05, "loss": 0.47142734527587893, "memory(GiB)": 61.91, "step": 10765, "token_acc": 0.8860759493670886, "train_speed(iter/s)": 1.504317 }, { "epoch": 0.4614198192022621, "grad_norm": 1.1534159183502197, "learning_rate": 9.791335711591332e-05, "loss": 0.29250483512878417, "memory(GiB)": 61.91, "step": 10770, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.504372 }, { "epoch": 0.4616340345315111, "grad_norm": 3.021414279937744, "learning_rate": 9.791143281313274e-05, "loss": 0.3430425405502319, "memory(GiB)": 61.91, "step": 10775, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.504361 }, { "epoch": 0.46184824986076006, "grad_norm": 3.635462522506714, "learning_rate": 9.79095076423916e-05, "loss": 0.5166188240051269, "memory(GiB)": 61.91, "step": 10780, "token_acc": 0.8721311475409836, "train_speed(iter/s)": 1.504316 }, { "epoch": 0.462062465190009, "grad_norm": 0.1785021424293518, "learning_rate": 9.790758160372479e-05, "loss": 0.209753155708313, "memory(GiB)": 61.91, "step": 10785, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.504315 }, { "epoch": 0.46227668051925797, "grad_norm": 3.907963752746582, "learning_rate": 9.79056546971672e-05, "loss": 0.31595139503479003, "memory(GiB)": 61.91, "step": 10790, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.504304 }, { "epoch": 0.46249089584850694, "grad_norm": 1.8917311429977417, "learning_rate": 9.790372692275374e-05, "loss": 0.16410307884216307, "memory(GiB)": 61.91, "step": 10795, "token_acc": 0.9691119691119691, "train_speed(iter/s)": 1.504277 }, { "epoch": 0.46270511117775587, "grad_norm": 2.0877163410186768, "learning_rate": 9.790179828051931e-05, "loss": 0.17526519298553467, "memory(GiB)": 61.91, "step": 10800, "token_acc": 0.95578231292517, "train_speed(iter/s)": 1.504239 }, { "epoch": 0.46291932650700485, "grad_norm": 1.8776946067810059, "learning_rate": 9.789986877049888e-05, "loss": 0.4406127452850342, "memory(GiB)": 61.91, "step": 10805, "token_acc": 0.9100719424460432, "train_speed(iter/s)": 1.504283 }, { "epoch": 0.4631335418362538, "grad_norm": 2.41778564453125, "learning_rate": 9.78979383927274e-05, "loss": 0.19430677890777587, "memory(GiB)": 61.91, "step": 10810, "token_acc": 0.9678571428571429, "train_speed(iter/s)": 1.504279 }, { "epoch": 0.46334775716550275, "grad_norm": 8.704602241516113, "learning_rate": 9.789600714723983e-05, "loss": 0.5961216449737549, "memory(GiB)": 61.91, "step": 10815, "token_acc": 0.8925925925925926, "train_speed(iter/s)": 1.504245 }, { "epoch": 0.46356197249475173, "grad_norm": 2.2799153327941895, "learning_rate": 9.789407503407115e-05, "loss": 0.2476491689682007, "memory(GiB)": 61.91, "step": 10820, "token_acc": 0.941747572815534, "train_speed(iter/s)": 1.504233 }, { "epoch": 0.4637761878240007, "grad_norm": 1.5535821914672852, "learning_rate": 9.789214205325638e-05, "loss": 0.45483851432800293, "memory(GiB)": 61.91, "step": 10825, "token_acc": 0.9153094462540716, "train_speed(iter/s)": 1.504232 }, { "epoch": 0.46399040315324963, "grad_norm": 3.8337602615356445, "learning_rate": 9.789020820483055e-05, "loss": 0.2660163640975952, "memory(GiB)": 61.91, "step": 10830, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.50422 }, { "epoch": 0.4642046184824986, "grad_norm": 1.8263661861419678, "learning_rate": 9.788827348882865e-05, "loss": 0.3584278106689453, "memory(GiB)": 61.91, "step": 10835, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.504234 }, { "epoch": 0.4644188338117476, "grad_norm": 1.652527093887329, "learning_rate": 9.788633790528576e-05, "loss": 0.27062201499938965, "memory(GiB)": 61.91, "step": 10840, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.504239 }, { "epoch": 0.4646330491409965, "grad_norm": 1.4574462175369263, "learning_rate": 9.788440145423695e-05, "loss": 0.2535589933395386, "memory(GiB)": 61.91, "step": 10845, "token_acc": 0.9528985507246377, "train_speed(iter/s)": 1.504259 }, { "epoch": 0.4648472644702455, "grad_norm": 4.577454566955566, "learning_rate": 9.788246413571727e-05, "loss": 0.7070561408996582, "memory(GiB)": 61.91, "step": 10850, "token_acc": 0.8576512455516014, "train_speed(iter/s)": 1.504308 }, { "epoch": 0.46506147979949447, "grad_norm": 2.851728916168213, "learning_rate": 9.788052594976184e-05, "loss": 0.4174467086791992, "memory(GiB)": 61.91, "step": 10855, "token_acc": 0.881619937694704, "train_speed(iter/s)": 1.504243 }, { "epoch": 0.4652756951287434, "grad_norm": 0.41548022627830505, "learning_rate": 9.787858689640577e-05, "loss": 0.30965867042541506, "memory(GiB)": 61.91, "step": 10860, "token_acc": 0.9430379746835443, "train_speed(iter/s)": 1.504231 }, { "epoch": 0.4654899104579924, "grad_norm": 2.009511947631836, "learning_rate": 9.787664697568418e-05, "loss": 0.37451269626617434, "memory(GiB)": 61.91, "step": 10865, "token_acc": 0.9108527131782945, "train_speed(iter/s)": 1.504223 }, { "epoch": 0.46570412578724135, "grad_norm": 2.6593475341796875, "learning_rate": 9.787470618763222e-05, "loss": 0.2738472938537598, "memory(GiB)": 61.91, "step": 10870, "token_acc": 0.9492537313432836, "train_speed(iter/s)": 1.504205 }, { "epoch": 0.4659183411164903, "grad_norm": 5.651359558105469, "learning_rate": 9.787276453228504e-05, "loss": 0.26958889961242677, "memory(GiB)": 61.91, "step": 10875, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.504219 }, { "epoch": 0.46613255644573925, "grad_norm": 2.792630434036255, "learning_rate": 9.787082200967784e-05, "loss": 0.6035027027130127, "memory(GiB)": 61.91, "step": 10880, "token_acc": 0.8541033434650456, "train_speed(iter/s)": 1.504306 }, { "epoch": 0.46634677177498823, "grad_norm": 2.221820592880249, "learning_rate": 9.786887861984578e-05, "loss": 0.3846782684326172, "memory(GiB)": 61.91, "step": 10885, "token_acc": 0.93, "train_speed(iter/s)": 1.504268 }, { "epoch": 0.46656098710423716, "grad_norm": 7.356494903564453, "learning_rate": 9.786693436282408e-05, "loss": 0.31231353282928465, "memory(GiB)": 61.91, "step": 10890, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.50425 }, { "epoch": 0.46677520243348614, "grad_norm": 0.09517529606819153, "learning_rate": 9.786498923864796e-05, "loss": 0.31556365489959715, "memory(GiB)": 61.91, "step": 10895, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.504273 }, { "epoch": 0.4669894177627351, "grad_norm": 2.6622560024261475, "learning_rate": 9.786304324735267e-05, "loss": 0.27003846168518064, "memory(GiB)": 61.91, "step": 10900, "token_acc": 0.9391891891891891, "train_speed(iter/s)": 1.504238 }, { "epoch": 0.46720363309198404, "grad_norm": 3.372133731842041, "learning_rate": 9.786109638897344e-05, "loss": 0.8242011070251465, "memory(GiB)": 61.91, "step": 10905, "token_acc": 0.8409785932721713, "train_speed(iter/s)": 1.504254 }, { "epoch": 0.467417848421233, "grad_norm": 1.6255995035171509, "learning_rate": 9.785914866354556e-05, "loss": 0.5057673454284668, "memory(GiB)": 61.91, "step": 10910, "token_acc": 0.8931750741839762, "train_speed(iter/s)": 1.504227 }, { "epoch": 0.467632063750482, "grad_norm": 3.3229823112487793, "learning_rate": 9.78572000711043e-05, "loss": 0.17411328554153443, "memory(GiB)": 61.91, "step": 10915, "token_acc": 0.9572368421052632, "train_speed(iter/s)": 1.504286 }, { "epoch": 0.4678462790797309, "grad_norm": 0.8872343301773071, "learning_rate": 9.785525061168497e-05, "loss": 0.3975203514099121, "memory(GiB)": 61.91, "step": 10920, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.504331 }, { "epoch": 0.4680604944089799, "grad_norm": 0.7639049887657166, "learning_rate": 9.785330028532288e-05, "loss": 0.13306140899658203, "memory(GiB)": 61.91, "step": 10925, "token_acc": 0.9651162790697675, "train_speed(iter/s)": 1.50431 }, { "epoch": 0.4682747097382289, "grad_norm": 1.3761647939682007, "learning_rate": 9.785134909205337e-05, "loss": 0.49283771514892577, "memory(GiB)": 61.91, "step": 10930, "token_acc": 0.8781163434903048, "train_speed(iter/s)": 1.504325 }, { "epoch": 0.4684889250674778, "grad_norm": 1.8667527437210083, "learning_rate": 9.784939703191179e-05, "loss": 0.3466392993927002, "memory(GiB)": 61.91, "step": 10935, "token_acc": 0.9331210191082803, "train_speed(iter/s)": 1.504403 }, { "epoch": 0.4687031403967268, "grad_norm": 2.496950149536133, "learning_rate": 9.784744410493348e-05, "loss": 0.3044441699981689, "memory(GiB)": 61.91, "step": 10940, "token_acc": 0.9216867469879518, "train_speed(iter/s)": 1.504469 }, { "epoch": 0.46891735572597576, "grad_norm": 3.2126963138580322, "learning_rate": 9.784549031115384e-05, "loss": 0.5388558387756348, "memory(GiB)": 61.91, "step": 10945, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.504447 }, { "epoch": 0.46913157105522474, "grad_norm": 3.8872692584991455, "learning_rate": 9.784353565060826e-05, "loss": 0.6361990928649902, "memory(GiB)": 61.91, "step": 10950, "token_acc": 0.8758169934640523, "train_speed(iter/s)": 1.504507 }, { "epoch": 0.46934578638447366, "grad_norm": 4.456762313842773, "learning_rate": 9.784158012333216e-05, "loss": 0.624901294708252, "memory(GiB)": 61.91, "step": 10955, "token_acc": 0.8597014925373134, "train_speed(iter/s)": 1.504471 }, { "epoch": 0.46956000171372264, "grad_norm": 2.2355990409851074, "learning_rate": 9.783962372936095e-05, "loss": 0.36996748447418215, "memory(GiB)": 61.91, "step": 10960, "token_acc": 0.928125, "train_speed(iter/s)": 1.504468 }, { "epoch": 0.4697742170429716, "grad_norm": 7.941888809204102, "learning_rate": 9.783766646873008e-05, "loss": 0.48934016227722166, "memory(GiB)": 61.91, "step": 10965, "token_acc": 0.9237536656891495, "train_speed(iter/s)": 1.504517 }, { "epoch": 0.46998843237222054, "grad_norm": 4.916469097137451, "learning_rate": 9.7835708341475e-05, "loss": 0.3824016571044922, "memory(GiB)": 61.91, "step": 10970, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.504543 }, { "epoch": 0.4702026477014695, "grad_norm": 6.496818542480469, "learning_rate": 9.78337493476312e-05, "loss": 0.46321806907653806, "memory(GiB)": 61.91, "step": 10975, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.504554 }, { "epoch": 0.4704168630307185, "grad_norm": 6.6948347091674805, "learning_rate": 9.783178948723415e-05, "loss": 0.40388102531433107, "memory(GiB)": 61.91, "step": 10980, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.504521 }, { "epoch": 0.4706310783599674, "grad_norm": 4.823043346405029, "learning_rate": 9.782982876031938e-05, "loss": 0.648367166519165, "memory(GiB)": 61.91, "step": 10985, "token_acc": 0.8562300319488818, "train_speed(iter/s)": 1.504522 }, { "epoch": 0.4708452936892164, "grad_norm": 3.0667011737823486, "learning_rate": 9.782786716692239e-05, "loss": 0.27856969833374023, "memory(GiB)": 61.91, "step": 10990, "token_acc": 0.940251572327044, "train_speed(iter/s)": 1.504516 }, { "epoch": 0.4710595090184654, "grad_norm": 5.844751834869385, "learning_rate": 9.782590470707871e-05, "loss": 0.7939065456390381, "memory(GiB)": 61.91, "step": 10995, "token_acc": 0.8211382113821138, "train_speed(iter/s)": 1.504481 }, { "epoch": 0.4712737243477143, "grad_norm": 1.5980963706970215, "learning_rate": 9.78239413808239e-05, "loss": 0.26999223232269287, "memory(GiB)": 61.91, "step": 11000, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.504477 }, { "epoch": 0.4712737243477143, "eval_loss": 2.4674136638641357, "eval_runtime": 13.2051, "eval_samples_per_second": 7.573, "eval_steps_per_second": 7.573, "eval_token_acc": 0.45590062111801244, "step": 11000 }, { "epoch": 0.4714879396769633, "grad_norm": 3.831713914871216, "learning_rate": 9.782197718819352e-05, "loss": 0.8251922607421875, "memory(GiB)": 61.91, "step": 11005, "token_acc": 0.557347670250896, "train_speed(iter/s)": 1.50151 }, { "epoch": 0.47170215500621226, "grad_norm": 2.620090961456299, "learning_rate": 9.782001212922319e-05, "loss": 0.40419855117797854, "memory(GiB)": 61.91, "step": 11010, "token_acc": 0.91, "train_speed(iter/s)": 1.501491 }, { "epoch": 0.4719163703354612, "grad_norm": 4.481198310852051, "learning_rate": 9.781804620394847e-05, "loss": 0.6506732940673828, "memory(GiB)": 61.91, "step": 11015, "token_acc": 0.8674698795180723, "train_speed(iter/s)": 1.501554 }, { "epoch": 0.47213058566471017, "grad_norm": 3.915330648422241, "learning_rate": 9.781607941240498e-05, "loss": 0.5186079502105713, "memory(GiB)": 61.91, "step": 11020, "token_acc": 0.8885630498533724, "train_speed(iter/s)": 1.501534 }, { "epoch": 0.47234480099395915, "grad_norm": 2.1781411170959473, "learning_rate": 9.781411175462836e-05, "loss": 0.6258825778961181, "memory(GiB)": 61.91, "step": 11025, "token_acc": 0.8714285714285714, "train_speed(iter/s)": 1.501529 }, { "epoch": 0.47255901632320807, "grad_norm": 2.6776087284088135, "learning_rate": 9.781214323065426e-05, "loss": 0.33896970748901367, "memory(GiB)": 61.91, "step": 11030, "token_acc": 0.9390681003584229, "train_speed(iter/s)": 1.50157 }, { "epoch": 0.47277323165245705, "grad_norm": 3.3133506774902344, "learning_rate": 9.781017384051832e-05, "loss": 0.4567694664001465, "memory(GiB)": 61.91, "step": 11035, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.501559 }, { "epoch": 0.47298744698170603, "grad_norm": 2.230670928955078, "learning_rate": 9.780820358425625e-05, "loss": 0.6396428108215332, "memory(GiB)": 61.91, "step": 11040, "token_acc": 0.8892857142857142, "train_speed(iter/s)": 1.501595 }, { "epoch": 0.47320166231095495, "grad_norm": 3.7043404579162598, "learning_rate": 9.780623246190371e-05, "loss": 0.41153564453125, "memory(GiB)": 61.91, "step": 11045, "token_acc": 0.9, "train_speed(iter/s)": 1.501578 }, { "epoch": 0.47341587764020393, "grad_norm": 5.23051118850708, "learning_rate": 9.780426047349642e-05, "loss": 0.6211215019226074, "memory(GiB)": 61.91, "step": 11050, "token_acc": 0.87, "train_speed(iter/s)": 1.501652 }, { "epoch": 0.4736300929694529, "grad_norm": 1.116700291633606, "learning_rate": 9.78022876190701e-05, "loss": 0.4593935489654541, "memory(GiB)": 61.91, "step": 11055, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.501608 }, { "epoch": 0.47384430829870183, "grad_norm": 3.602081060409546, "learning_rate": 9.780031389866053e-05, "loss": 0.45543899536132815, "memory(GiB)": 61.91, "step": 11060, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.501599 }, { "epoch": 0.4740585236279508, "grad_norm": 5.0756306648254395, "learning_rate": 9.77983393123034e-05, "loss": 0.6820398330688476, "memory(GiB)": 61.91, "step": 11065, "token_acc": 0.8614457831325302, "train_speed(iter/s)": 1.501597 }, { "epoch": 0.4742727389571998, "grad_norm": 2.0996296405792236, "learning_rate": 9.779636386003453e-05, "loss": 0.3295464992523193, "memory(GiB)": 61.91, "step": 11070, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.501586 }, { "epoch": 0.4744869542864487, "grad_norm": 1.3629323244094849, "learning_rate": 9.77943875418897e-05, "loss": 0.28213043212890626, "memory(GiB)": 61.91, "step": 11075, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.501556 }, { "epoch": 0.4747011696156977, "grad_norm": 1.9819992780685425, "learning_rate": 9.77924103579047e-05, "loss": 0.4748707294464111, "memory(GiB)": 61.91, "step": 11080, "token_acc": 0.9078498293515358, "train_speed(iter/s)": 1.501541 }, { "epoch": 0.4749153849449467, "grad_norm": 1.5666199922561646, "learning_rate": 9.779043230811534e-05, "loss": 0.3651329278945923, "memory(GiB)": 61.91, "step": 11085, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.50165 }, { "epoch": 0.4751296002741956, "grad_norm": 2.9183390140533447, "learning_rate": 9.778845339255749e-05, "loss": 0.6057643413543701, "memory(GiB)": 61.91, "step": 11090, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.501702 }, { "epoch": 0.4753438156034446, "grad_norm": 2.299323081970215, "learning_rate": 9.778647361126696e-05, "loss": 0.5313835144042969, "memory(GiB)": 61.91, "step": 11095, "token_acc": 0.8923611111111112, "train_speed(iter/s)": 1.501677 }, { "epoch": 0.47555803093269355, "grad_norm": 1.3721915483474731, "learning_rate": 9.778449296427962e-05, "loss": 0.2994457006454468, "memory(GiB)": 61.91, "step": 11100, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.501658 }, { "epoch": 0.4757722462619425, "grad_norm": 4.524265766143799, "learning_rate": 9.778251145163139e-05, "loss": 0.6106863021850586, "memory(GiB)": 61.91, "step": 11105, "token_acc": 0.864406779661017, "train_speed(iter/s)": 1.501635 }, { "epoch": 0.47598646159119146, "grad_norm": 1.2149649858474731, "learning_rate": 9.778052907335814e-05, "loss": 0.5066669464111329, "memory(GiB)": 61.91, "step": 11110, "token_acc": 0.8871473354231975, "train_speed(iter/s)": 1.501617 }, { "epoch": 0.47620067692044044, "grad_norm": 2.8715381622314453, "learning_rate": 9.777854582949578e-05, "loss": 0.5982160568237305, "memory(GiB)": 61.91, "step": 11115, "token_acc": 0.8776119402985074, "train_speed(iter/s)": 1.501609 }, { "epoch": 0.4764148922496894, "grad_norm": 2.941422700881958, "learning_rate": 9.777656172008023e-05, "loss": 0.5240360260009765, "memory(GiB)": 61.91, "step": 11120, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.501573 }, { "epoch": 0.47662910757893834, "grad_norm": 2.3349595069885254, "learning_rate": 9.777457674514748e-05, "loss": 0.3661135911941528, "memory(GiB)": 61.91, "step": 11125, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.501546 }, { "epoch": 0.4768433229081873, "grad_norm": 3.7229886054992676, "learning_rate": 9.777259090473341e-05, "loss": 0.6139730453491211, "memory(GiB)": 61.91, "step": 11130, "token_acc": 0.8793650793650793, "train_speed(iter/s)": 1.501619 }, { "epoch": 0.4770575382374363, "grad_norm": 3.4951014518737793, "learning_rate": 9.777060419887407e-05, "loss": 0.4089240074157715, "memory(GiB)": 61.91, "step": 11135, "token_acc": 0.9014598540145985, "train_speed(iter/s)": 1.501607 }, { "epoch": 0.4772717535666852, "grad_norm": 1.2809094190597534, "learning_rate": 9.776861662760541e-05, "loss": 0.36595497131347654, "memory(GiB)": 61.91, "step": 11140, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.50159 }, { "epoch": 0.4774859688959342, "grad_norm": 3.5160350799560547, "learning_rate": 9.776662819096347e-05, "loss": 0.39074931144714353, "memory(GiB)": 61.91, "step": 11145, "token_acc": 0.9097744360902256, "train_speed(iter/s)": 1.501536 }, { "epoch": 0.4777001842251832, "grad_norm": 4.839613437652588, "learning_rate": 9.776463888898423e-05, "loss": 0.37480626106262205, "memory(GiB)": 61.91, "step": 11150, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.50155 }, { "epoch": 0.4779143995544321, "grad_norm": 2.159552574157715, "learning_rate": 9.776264872170376e-05, "loss": 0.6248004913330079, "memory(GiB)": 61.91, "step": 11155, "token_acc": 0.8778877887788779, "train_speed(iter/s)": 1.501585 }, { "epoch": 0.4781286148836811, "grad_norm": 1.5152941942214966, "learning_rate": 9.77606576891581e-05, "loss": 0.29906795024871824, "memory(GiB)": 61.91, "step": 11160, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.501549 }, { "epoch": 0.47834283021293006, "grad_norm": 2.5016281604766846, "learning_rate": 9.775866579138332e-05, "loss": 0.716776704788208, "memory(GiB)": 61.91, "step": 11165, "token_acc": 0.8807692307692307, "train_speed(iter/s)": 1.501552 }, { "epoch": 0.478557045542179, "grad_norm": 1.8258247375488281, "learning_rate": 9.775667302841551e-05, "loss": 0.24568920135498046, "memory(GiB)": 61.91, "step": 11170, "token_acc": 0.9395973154362416, "train_speed(iter/s)": 1.501528 }, { "epoch": 0.47877126087142796, "grad_norm": 3.6775875091552734, "learning_rate": 9.775467940029077e-05, "loss": 0.27192375659942625, "memory(GiB)": 61.91, "step": 11175, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.501478 }, { "epoch": 0.47898547620067694, "grad_norm": 3.0330405235290527, "learning_rate": 9.775268490704522e-05, "loss": 0.45546665191650393, "memory(GiB)": 61.91, "step": 11180, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.501427 }, { "epoch": 0.47919969152992586, "grad_norm": 4.627752780914307, "learning_rate": 9.775068954871498e-05, "loss": 0.39276323318481443, "memory(GiB)": 61.91, "step": 11185, "token_acc": 0.9037656903765691, "train_speed(iter/s)": 1.501367 }, { "epoch": 0.47941390685917484, "grad_norm": 1.9896824359893799, "learning_rate": 9.774869332533622e-05, "loss": 0.42889938354492185, "memory(GiB)": 61.91, "step": 11190, "token_acc": 0.9129032258064517, "train_speed(iter/s)": 1.501352 }, { "epoch": 0.4796281221884238, "grad_norm": 2.8766486644744873, "learning_rate": 9.774669623694507e-05, "loss": 0.27031474113464354, "memory(GiB)": 61.91, "step": 11195, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.501358 }, { "epoch": 0.47984233751767275, "grad_norm": 2.4889278411865234, "learning_rate": 9.774469828357773e-05, "loss": 0.58183274269104, "memory(GiB)": 61.91, "step": 11200, "token_acc": 0.8654434250764526, "train_speed(iter/s)": 1.501341 }, { "epoch": 0.4800565528469217, "grad_norm": 6.558652400970459, "learning_rate": 9.77426994652704e-05, "loss": 0.14588444232940673, "memory(GiB)": 61.91, "step": 11205, "token_acc": 0.9525691699604744, "train_speed(iter/s)": 1.501325 }, { "epoch": 0.4802707681761707, "grad_norm": 4.634832382202148, "learning_rate": 9.774069978205928e-05, "loss": 0.49342708587646483, "memory(GiB)": 61.91, "step": 11210, "token_acc": 0.8712121212121212, "train_speed(iter/s)": 1.501326 }, { "epoch": 0.4804849835054196, "grad_norm": 1.5361520051956177, "learning_rate": 9.77386992339806e-05, "loss": 0.21704471111297607, "memory(GiB)": 61.91, "step": 11215, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.501332 }, { "epoch": 0.4806991988346686, "grad_norm": 3.0456817150115967, "learning_rate": 9.77366978210706e-05, "loss": 0.5614039421081543, "memory(GiB)": 61.91, "step": 11220, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.501316 }, { "epoch": 0.4809134141639176, "grad_norm": 1.2195398807525635, "learning_rate": 9.773469554336553e-05, "loss": 0.14305754899978637, "memory(GiB)": 61.91, "step": 11225, "token_acc": 0.9569230769230769, "train_speed(iter/s)": 1.501327 }, { "epoch": 0.4811276294931665, "grad_norm": 1.1317216157913208, "learning_rate": 9.773269240090169e-05, "loss": 0.40273032188415525, "memory(GiB)": 61.91, "step": 11230, "token_acc": 0.9119804400977995, "train_speed(iter/s)": 1.50132 }, { "epoch": 0.4813418448224155, "grad_norm": 2.9501614570617676, "learning_rate": 9.773068839371534e-05, "loss": 0.421356201171875, "memory(GiB)": 61.91, "step": 11235, "token_acc": 0.9007352941176471, "train_speed(iter/s)": 1.501316 }, { "epoch": 0.48155606015166447, "grad_norm": 3.7089927196502686, "learning_rate": 9.772868352184279e-05, "loss": 0.504508638381958, "memory(GiB)": 61.91, "step": 11240, "token_acc": 0.8653846153846154, "train_speed(iter/s)": 1.501308 }, { "epoch": 0.4817702754809134, "grad_norm": 2.667515993118286, "learning_rate": 9.772667778532036e-05, "loss": 0.3164717435836792, "memory(GiB)": 61.91, "step": 11245, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.501317 }, { "epoch": 0.48198449081016237, "grad_norm": 5.4000935554504395, "learning_rate": 9.77246711841844e-05, "loss": 0.25612828731536863, "memory(GiB)": 61.91, "step": 11250, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.501294 }, { "epoch": 0.48219870613941135, "grad_norm": 0.6427754163742065, "learning_rate": 9.772266371847125e-05, "loss": 0.26568450927734377, "memory(GiB)": 61.91, "step": 11255, "token_acc": 0.94140625, "train_speed(iter/s)": 1.501303 }, { "epoch": 0.48241292146866027, "grad_norm": 6.328140735626221, "learning_rate": 9.772065538821728e-05, "loss": 0.3022072553634644, "memory(GiB)": 61.91, "step": 11260, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.501358 }, { "epoch": 0.48262713679790925, "grad_norm": 4.575869083404541, "learning_rate": 9.771864619345888e-05, "loss": 0.43770418167114256, "memory(GiB)": 61.91, "step": 11265, "token_acc": 0.8966789667896679, "train_speed(iter/s)": 1.501333 }, { "epoch": 0.48284135212715823, "grad_norm": 2.764084577560425, "learning_rate": 9.771663613423243e-05, "loss": 0.4116383075714111, "memory(GiB)": 61.91, "step": 11270, "token_acc": 0.9127725856697819, "train_speed(iter/s)": 1.501307 }, { "epoch": 0.48305556745640715, "grad_norm": 2.286790609359741, "learning_rate": 9.771462521057436e-05, "loss": 0.43422880172729494, "memory(GiB)": 61.91, "step": 11275, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.501302 }, { "epoch": 0.48326978278565613, "grad_norm": 0.3930443525314331, "learning_rate": 9.771261342252109e-05, "loss": 0.23178510665893554, "memory(GiB)": 61.91, "step": 11280, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.501268 }, { "epoch": 0.4834839981149051, "grad_norm": 1.7516249418258667, "learning_rate": 9.771060077010907e-05, "loss": 0.2492055892944336, "memory(GiB)": 61.91, "step": 11285, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.501275 }, { "epoch": 0.4836982134441541, "grad_norm": 3.484316110610962, "learning_rate": 9.770858725337477e-05, "loss": 0.7568736553192139, "memory(GiB)": 61.91, "step": 11290, "token_acc": 0.8764705882352941, "train_speed(iter/s)": 1.501325 }, { "epoch": 0.483912428773403, "grad_norm": 5.073870658874512, "learning_rate": 9.770657287235465e-05, "loss": 0.4803306579589844, "memory(GiB)": 61.91, "step": 11295, "token_acc": 0.9171779141104295, "train_speed(iter/s)": 1.501281 }, { "epoch": 0.484126644102652, "grad_norm": 1.4493873119354248, "learning_rate": 9.770455762708521e-05, "loss": 0.44701132774353025, "memory(GiB)": 61.91, "step": 11300, "token_acc": 0.8714285714285714, "train_speed(iter/s)": 1.501259 }, { "epoch": 0.48434085943190097, "grad_norm": 3.717771053314209, "learning_rate": 9.770254151760297e-05, "loss": 0.432966947555542, "memory(GiB)": 61.91, "step": 11305, "token_acc": 0.8939929328621908, "train_speed(iter/s)": 1.501231 }, { "epoch": 0.4845550747611499, "grad_norm": 2.8678789138793945, "learning_rate": 9.770052454394443e-05, "loss": 0.29010250568389895, "memory(GiB)": 61.91, "step": 11310, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.501303 }, { "epoch": 0.4847692900903989, "grad_norm": 1.9964799880981445, "learning_rate": 9.769850670614613e-05, "loss": 0.2826425075531006, "memory(GiB)": 61.91, "step": 11315, "token_acc": 0.944, "train_speed(iter/s)": 1.501295 }, { "epoch": 0.48498350541964785, "grad_norm": 2.8159830570220947, "learning_rate": 9.769648800424465e-05, "loss": 0.323545503616333, "memory(GiB)": 61.91, "step": 11320, "token_acc": 0.9315589353612167, "train_speed(iter/s)": 1.501331 }, { "epoch": 0.4851977207488968, "grad_norm": 2.7975292205810547, "learning_rate": 9.769446843827655e-05, "loss": 0.3327798366546631, "memory(GiB)": 61.91, "step": 11325, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.501312 }, { "epoch": 0.48541193607814576, "grad_norm": 2.9015631675720215, "learning_rate": 9.769244800827841e-05, "loss": 0.5726801872253418, "memory(GiB)": 61.91, "step": 11330, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.501279 }, { "epoch": 0.48562615140739473, "grad_norm": 2.39996075630188, "learning_rate": 9.769042671428682e-05, "loss": 0.30933775901794436, "memory(GiB)": 61.91, "step": 11335, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.501317 }, { "epoch": 0.48584036673664366, "grad_norm": 3.1042895317077637, "learning_rate": 9.768840455633842e-05, "loss": 0.5047928810119628, "memory(GiB)": 61.91, "step": 11340, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.501366 }, { "epoch": 0.48605458206589264, "grad_norm": 2.3007922172546387, "learning_rate": 9.768638153446986e-05, "loss": 0.38799097537994387, "memory(GiB)": 61.91, "step": 11345, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.501339 }, { "epoch": 0.4862687973951416, "grad_norm": 3.1974494457244873, "learning_rate": 9.768435764871775e-05, "loss": 0.2912858724594116, "memory(GiB)": 61.91, "step": 11350, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.501339 }, { "epoch": 0.48648301272439054, "grad_norm": 3.434542179107666, "learning_rate": 9.768233289911877e-05, "loss": 0.32773020267486574, "memory(GiB)": 61.91, "step": 11355, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.501331 }, { "epoch": 0.4866972280536395, "grad_norm": 2.8062796592712402, "learning_rate": 9.768030728570959e-05, "loss": 0.6167914867401123, "memory(GiB)": 61.91, "step": 11360, "token_acc": 0.8681318681318682, "train_speed(iter/s)": 1.501459 }, { "epoch": 0.4869114433828885, "grad_norm": 1.3174638748168945, "learning_rate": 9.767828080852693e-05, "loss": 0.36702895164489746, "memory(GiB)": 61.91, "step": 11365, "token_acc": 0.9261538461538461, "train_speed(iter/s)": 1.501541 }, { "epoch": 0.4871256587121374, "grad_norm": 4.275399208068848, "learning_rate": 9.767625346760748e-05, "loss": 0.846738052368164, "memory(GiB)": 61.91, "step": 11370, "token_acc": 0.8304498269896193, "train_speed(iter/s)": 1.501666 }, { "epoch": 0.4873398740413864, "grad_norm": 0.3814917504787445, "learning_rate": 9.767422526298798e-05, "loss": 0.3852062463760376, "memory(GiB)": 61.91, "step": 11375, "token_acc": 0.9281609195402298, "train_speed(iter/s)": 1.501633 }, { "epoch": 0.4875540893706354, "grad_norm": 0.4015127420425415, "learning_rate": 9.767219619470516e-05, "loss": 0.18698071241378783, "memory(GiB)": 61.91, "step": 11380, "token_acc": 0.9584905660377359, "train_speed(iter/s)": 1.50161 }, { "epoch": 0.4877683046998843, "grad_norm": 3.362959623336792, "learning_rate": 9.767016626279577e-05, "loss": 0.5646921157836914, "memory(GiB)": 61.91, "step": 11385, "token_acc": 0.8943396226415095, "train_speed(iter/s)": 1.501605 }, { "epoch": 0.4879825200291333, "grad_norm": 11.707127571105957, "learning_rate": 9.766813546729663e-05, "loss": 0.3187761068344116, "memory(GiB)": 61.91, "step": 11390, "token_acc": 0.9293286219081273, "train_speed(iter/s)": 1.501536 }, { "epoch": 0.48819673535838226, "grad_norm": 3.4010496139526367, "learning_rate": 9.766610380824448e-05, "loss": 0.6657217502593994, "memory(GiB)": 61.91, "step": 11395, "token_acc": 0.8870431893687708, "train_speed(iter/s)": 1.501525 }, { "epoch": 0.4884109506876312, "grad_norm": 3.1382431983947754, "learning_rate": 9.766407128567617e-05, "loss": 0.22278993129730223, "memory(GiB)": 61.91, "step": 11400, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.501485 }, { "epoch": 0.48862516601688016, "grad_norm": 3.1030924320220947, "learning_rate": 9.766203789962846e-05, "loss": 0.44884448051452636, "memory(GiB)": 61.91, "step": 11405, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.50145 }, { "epoch": 0.48883938134612914, "grad_norm": 1.2568011283874512, "learning_rate": 9.766000365013824e-05, "loss": 0.4228797912597656, "memory(GiB)": 61.91, "step": 11410, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.501434 }, { "epoch": 0.48905359667537807, "grad_norm": 7.209405899047852, "learning_rate": 9.765796853724235e-05, "loss": 0.32353711128234863, "memory(GiB)": 61.91, "step": 11415, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.501445 }, { "epoch": 0.48926781200462705, "grad_norm": 3.2190451622009277, "learning_rate": 9.765593256097764e-05, "loss": 0.4315779685974121, "memory(GiB)": 61.91, "step": 11420, "token_acc": 0.8854489164086687, "train_speed(iter/s)": 1.501429 }, { "epoch": 0.489482027333876, "grad_norm": 0.8573207855224609, "learning_rate": 9.765389572138103e-05, "loss": 0.2522472858428955, "memory(GiB)": 61.91, "step": 11425, "token_acc": 0.9315068493150684, "train_speed(iter/s)": 1.501472 }, { "epoch": 0.48969624266312495, "grad_norm": 3.9562830924987793, "learning_rate": 9.765185801848938e-05, "loss": 0.2380443572998047, "memory(GiB)": 61.91, "step": 11430, "token_acc": 0.9527272727272728, "train_speed(iter/s)": 1.501492 }, { "epoch": 0.4899104579923739, "grad_norm": 11.647363662719727, "learning_rate": 9.764981945233962e-05, "loss": 0.42764482498168943, "memory(GiB)": 61.91, "step": 11435, "token_acc": 0.9126506024096386, "train_speed(iter/s)": 1.501497 }, { "epoch": 0.4901246733216229, "grad_norm": 0.929187536239624, "learning_rate": 9.764778002296868e-05, "loss": 0.35303735733032227, "memory(GiB)": 61.91, "step": 11440, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.50154 }, { "epoch": 0.49033888865087183, "grad_norm": 2.3841586112976074, "learning_rate": 9.764573973041352e-05, "loss": 0.5817358493804932, "memory(GiB)": 61.91, "step": 11445, "token_acc": 0.8960244648318043, "train_speed(iter/s)": 1.501547 }, { "epoch": 0.4905531039801208, "grad_norm": 0.39544421434402466, "learning_rate": 9.764369857471109e-05, "loss": 0.3716617822647095, "memory(GiB)": 61.91, "step": 11450, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.501528 }, { "epoch": 0.4907673193093698, "grad_norm": 1.7395681142807007, "learning_rate": 9.764165655589835e-05, "loss": 0.5188132286071777, "memory(GiB)": 61.91, "step": 11455, "token_acc": 0.8783783783783784, "train_speed(iter/s)": 1.501608 }, { "epoch": 0.49098153463861877, "grad_norm": 6.2855224609375, "learning_rate": 9.763961367401231e-05, "loss": 0.3518547534942627, "memory(GiB)": 61.91, "step": 11460, "token_acc": 0.9132231404958677, "train_speed(iter/s)": 1.501549 }, { "epoch": 0.4911957499678677, "grad_norm": 2.784038782119751, "learning_rate": 9.763756992909e-05, "loss": 0.502281379699707, "memory(GiB)": 61.91, "step": 11465, "token_acc": 0.9093959731543624, "train_speed(iter/s)": 1.501597 }, { "epoch": 0.49140996529711667, "grad_norm": 2.0573761463165283, "learning_rate": 9.763552532116841e-05, "loss": 0.48516192436218264, "memory(GiB)": 61.91, "step": 11470, "token_acc": 0.8954248366013072, "train_speed(iter/s)": 1.501578 }, { "epoch": 0.49162418062636565, "grad_norm": 3.6975748538970947, "learning_rate": 9.76334798502846e-05, "loss": 0.8070209503173829, "memory(GiB)": 61.91, "step": 11475, "token_acc": 0.8418079096045198, "train_speed(iter/s)": 1.50163 }, { "epoch": 0.49183839595561457, "grad_norm": 4.245465278625488, "learning_rate": 9.763143351647561e-05, "loss": 0.38462653160095217, "memory(GiB)": 61.91, "step": 11480, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.501593 }, { "epoch": 0.49205261128486355, "grad_norm": 4.021365165710449, "learning_rate": 9.762938631977852e-05, "loss": 0.5284987926483155, "memory(GiB)": 61.91, "step": 11485, "token_acc": 0.896969696969697, "train_speed(iter/s)": 1.501668 }, { "epoch": 0.49226682661411253, "grad_norm": 5.62999963760376, "learning_rate": 9.762733826023042e-05, "loss": 0.34734354019165037, "memory(GiB)": 61.91, "step": 11490, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.501734 }, { "epoch": 0.49248104194336145, "grad_norm": 2.4938879013061523, "learning_rate": 9.76252893378684e-05, "loss": 0.4212313175201416, "memory(GiB)": 61.91, "step": 11495, "token_acc": 0.8877551020408163, "train_speed(iter/s)": 1.501776 }, { "epoch": 0.49269525727261043, "grad_norm": 4.279657363891602, "learning_rate": 9.76232395527296e-05, "loss": 0.40076308250427245, "memory(GiB)": 61.91, "step": 11500, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.501887 }, { "epoch": 0.49269525727261043, "eval_loss": 2.562253713607788, "eval_runtime": 13.2976, "eval_samples_per_second": 7.52, "eval_steps_per_second": 7.52, "eval_token_acc": 0.45235069885641677, "step": 11500 }, { "epoch": 0.4929094726018594, "grad_norm": 2.085803985595703, "learning_rate": 9.762118890485115e-05, "loss": 0.331924295425415, "memory(GiB)": 61.91, "step": 11505, "token_acc": 0.5907441016333939, "train_speed(iter/s)": 1.499098 }, { "epoch": 0.49312368793110833, "grad_norm": 2.121908664703369, "learning_rate": 9.761913739427017e-05, "loss": 0.3426368713378906, "memory(GiB)": 61.91, "step": 11510, "token_acc": 0.9339622641509434, "train_speed(iter/s)": 1.499161 }, { "epoch": 0.4933379032603573, "grad_norm": 0.07577716559171677, "learning_rate": 9.761708502102384e-05, "loss": 0.4422903537750244, "memory(GiB)": 61.91, "step": 11515, "token_acc": 0.9140625, "train_speed(iter/s)": 1.499144 }, { "epoch": 0.4935521185896063, "grad_norm": 1.9089854955673218, "learning_rate": 9.761503178514938e-05, "loss": 0.27995753288269043, "memory(GiB)": 61.91, "step": 11520, "token_acc": 0.9343629343629344, "train_speed(iter/s)": 1.499045 }, { "epoch": 0.4937663339188552, "grad_norm": 3.9500057697296143, "learning_rate": 9.761297768668393e-05, "loss": 0.743000602722168, "memory(GiB)": 61.91, "step": 11525, "token_acc": 0.8801369863013698, "train_speed(iter/s)": 1.498967 }, { "epoch": 0.4939805492481042, "grad_norm": 2.4674270153045654, "learning_rate": 9.761092272566472e-05, "loss": 0.334085750579834, "memory(GiB)": 61.91, "step": 11530, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.498951 }, { "epoch": 0.4941947645773532, "grad_norm": 2.625473976135254, "learning_rate": 9.7608866902129e-05, "loss": 0.37681214809417723, "memory(GiB)": 61.91, "step": 11535, "token_acc": 0.8838174273858921, "train_speed(iter/s)": 1.499009 }, { "epoch": 0.4944089799066021, "grad_norm": 3.0092153549194336, "learning_rate": 9.760681021611398e-05, "loss": 0.38585724830627444, "memory(GiB)": 61.91, "step": 11540, "token_acc": 0.9147540983606557, "train_speed(iter/s)": 1.499089 }, { "epoch": 0.4946231952358511, "grad_norm": 3.478627920150757, "learning_rate": 9.760475266765694e-05, "loss": 0.3850773811340332, "memory(GiB)": 61.91, "step": 11545, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.499117 }, { "epoch": 0.49483741056510006, "grad_norm": 3.978088140487671, "learning_rate": 9.760269425679514e-05, "loss": 0.4026025295257568, "memory(GiB)": 61.91, "step": 11550, "token_acc": 0.9385964912280702, "train_speed(iter/s)": 1.499149 }, { "epoch": 0.495051625894349, "grad_norm": 3.5149834156036377, "learning_rate": 9.760063498356589e-05, "loss": 0.3726953983306885, "memory(GiB)": 61.91, "step": 11555, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.499127 }, { "epoch": 0.49526584122359796, "grad_norm": 1.8539488315582275, "learning_rate": 9.759857484800647e-05, "loss": 0.2643078327178955, "memory(GiB)": 61.91, "step": 11560, "token_acc": 0.9348659003831418, "train_speed(iter/s)": 1.49909 }, { "epoch": 0.49548005655284694, "grad_norm": 2.2930963039398193, "learning_rate": 9.759651385015423e-05, "loss": 0.3690244913101196, "memory(GiB)": 61.91, "step": 11565, "token_acc": 0.9131832797427653, "train_speed(iter/s)": 1.499094 }, { "epoch": 0.49569427188209586, "grad_norm": 2.4579033851623535, "learning_rate": 9.759445199004649e-05, "loss": 0.4261507034301758, "memory(GiB)": 61.91, "step": 11570, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.499087 }, { "epoch": 0.49590848721134484, "grad_norm": 3.7795209884643555, "learning_rate": 9.75923892677206e-05, "loss": 0.46725921630859374, "memory(GiB)": 61.91, "step": 11575, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.499031 }, { "epoch": 0.4961227025405938, "grad_norm": 2.91925311088562, "learning_rate": 9.759032568321395e-05, "loss": 0.4893782138824463, "memory(GiB)": 61.91, "step": 11580, "token_acc": 0.9083333333333333, "train_speed(iter/s)": 1.499015 }, { "epoch": 0.49633691786984274, "grad_norm": 3.5743446350097656, "learning_rate": 9.758826123656388e-05, "loss": 0.4412864685058594, "memory(GiB)": 61.91, "step": 11585, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.499026 }, { "epoch": 0.4965511331990917, "grad_norm": 1.1577966213226318, "learning_rate": 9.758619592780784e-05, "loss": 0.356041693687439, "memory(GiB)": 61.91, "step": 11590, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.499017 }, { "epoch": 0.4967653485283407, "grad_norm": 2.1124770641326904, "learning_rate": 9.758412975698321e-05, "loss": 0.20717840194702147, "memory(GiB)": 61.91, "step": 11595, "token_acc": 0.9573643410852714, "train_speed(iter/s)": 1.498985 }, { "epoch": 0.4969795638575896, "grad_norm": 2.6313931941986084, "learning_rate": 9.758206272412743e-05, "loss": 0.5898406505584717, "memory(GiB)": 61.91, "step": 11600, "token_acc": 0.886435331230284, "train_speed(iter/s)": 1.498968 }, { "epoch": 0.4971937791868386, "grad_norm": 3.1397464275360107, "learning_rate": 9.757999482927795e-05, "loss": 0.332635760307312, "memory(GiB)": 61.91, "step": 11605, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.498945 }, { "epoch": 0.4974079945160876, "grad_norm": 2.822706699371338, "learning_rate": 9.757792607247224e-05, "loss": 0.6359538078308106, "memory(GiB)": 61.91, "step": 11610, "token_acc": 0.8607594936708861, "train_speed(iter/s)": 1.498975 }, { "epoch": 0.4976222098453365, "grad_norm": 2.7931511402130127, "learning_rate": 9.757585645374777e-05, "loss": 0.5658703327178956, "memory(GiB)": 61.91, "step": 11615, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.498971 }, { "epoch": 0.4978364251745855, "grad_norm": 4.5933685302734375, "learning_rate": 9.757378597314201e-05, "loss": 0.4616353988647461, "memory(GiB)": 61.91, "step": 11620, "token_acc": 0.8968253968253969, "train_speed(iter/s)": 1.499057 }, { "epoch": 0.49805064050383446, "grad_norm": 2.829345464706421, "learning_rate": 9.75717146306925e-05, "loss": 0.33214545249938965, "memory(GiB)": 61.91, "step": 11625, "token_acc": 0.9421221864951769, "train_speed(iter/s)": 1.499042 }, { "epoch": 0.49826485583308344, "grad_norm": 2.046233892440796, "learning_rate": 9.756964242643674e-05, "loss": 0.3090994358062744, "memory(GiB)": 61.91, "step": 11630, "token_acc": 0.9461279461279462, "train_speed(iter/s)": 1.499029 }, { "epoch": 0.49847907116233237, "grad_norm": 1.0769659280776978, "learning_rate": 9.75675693604123e-05, "loss": 0.33114018440246584, "memory(GiB)": 61.91, "step": 11635, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.499096 }, { "epoch": 0.49869328649158134, "grad_norm": 1.7287824153900146, "learning_rate": 9.756549543265671e-05, "loss": 0.2884695291519165, "memory(GiB)": 61.91, "step": 11640, "token_acc": 0.9315068493150684, "train_speed(iter/s)": 1.499087 }, { "epoch": 0.4989075018208303, "grad_norm": 4.361929416656494, "learning_rate": 9.756342064320756e-05, "loss": 0.4384263038635254, "memory(GiB)": 61.91, "step": 11645, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.499066 }, { "epoch": 0.49912171715007925, "grad_norm": 2.2501699924468994, "learning_rate": 9.756134499210243e-05, "loss": 0.4242057800292969, "memory(GiB)": 61.91, "step": 11650, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.499035 }, { "epoch": 0.4993359324793282, "grad_norm": 2.3019497394561768, "learning_rate": 9.755926847937891e-05, "loss": 0.45203371047973634, "memory(GiB)": 61.91, "step": 11655, "token_acc": 0.8972809667673716, "train_speed(iter/s)": 1.498979 }, { "epoch": 0.4995501478085772, "grad_norm": 2.5359861850738525, "learning_rate": 9.755719110507462e-05, "loss": 0.24514362812042237, "memory(GiB)": 61.91, "step": 11660, "token_acc": 0.9391891891891891, "train_speed(iter/s)": 1.498881 }, { "epoch": 0.49976436313782613, "grad_norm": 5.153092384338379, "learning_rate": 9.75551128692272e-05, "loss": 0.3216991901397705, "memory(GiB)": 61.91, "step": 11665, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.498972 }, { "epoch": 0.4999785784670751, "grad_norm": 2.5615012645721436, "learning_rate": 9.755303377187433e-05, "loss": 0.5082923412322998, "memory(GiB)": 61.91, "step": 11670, "token_acc": 0.8951612903225806, "train_speed(iter/s)": 1.499037 }, { "epoch": 0.5001927937963241, "grad_norm": 2.8399922847747803, "learning_rate": 9.755095381305362e-05, "loss": 0.35175347328186035, "memory(GiB)": 61.91, "step": 11675, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.499077 }, { "epoch": 0.500407009125573, "grad_norm": 2.9368789196014404, "learning_rate": 9.754887299280277e-05, "loss": 0.3901956796646118, "memory(GiB)": 61.91, "step": 11680, "token_acc": 0.916083916083916, "train_speed(iter/s)": 1.499074 }, { "epoch": 0.5006212244548219, "grad_norm": 2.6722841262817383, "learning_rate": 9.754679131115949e-05, "loss": 0.634061050415039, "memory(GiB)": 61.91, "step": 11685, "token_acc": 0.8736462093862816, "train_speed(iter/s)": 1.499021 }, { "epoch": 0.500835439784071, "grad_norm": 1.4680615663528442, "learning_rate": 9.754470876816148e-05, "loss": 0.20721962451934814, "memory(GiB)": 61.91, "step": 11690, "token_acc": 0.9537815126050421, "train_speed(iter/s)": 1.499017 }, { "epoch": 0.5010496551133199, "grad_norm": 3.000307559967041, "learning_rate": 9.754262536384649e-05, "loss": 0.3040961980819702, "memory(GiB)": 61.91, "step": 11695, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.498979 }, { "epoch": 0.5012638704425688, "grad_norm": 2.494349241256714, "learning_rate": 9.754054109825223e-05, "loss": 0.3126260995864868, "memory(GiB)": 61.91, "step": 11700, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.498987 }, { "epoch": 0.5014780857718178, "grad_norm": 6.695204734802246, "learning_rate": 9.753845597141647e-05, "loss": 0.6201929092407227, "memory(GiB)": 61.91, "step": 11705, "token_acc": 0.8808777429467085, "train_speed(iter/s)": 1.498966 }, { "epoch": 0.5016923011010668, "grad_norm": 0.4185749292373657, "learning_rate": 9.753636998337698e-05, "loss": 0.6026837825775146, "memory(GiB)": 61.91, "step": 11710, "token_acc": 0.8813559322033898, "train_speed(iter/s)": 1.499016 }, { "epoch": 0.5019065164303158, "grad_norm": 3.727421522140503, "learning_rate": 9.753428313417155e-05, "loss": 0.3028208494186401, "memory(GiB)": 61.91, "step": 11715, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.499003 }, { "epoch": 0.5021207317595647, "grad_norm": 10.740583419799805, "learning_rate": 9.7532195423838e-05, "loss": 0.5771458148956299, "memory(GiB)": 61.91, "step": 11720, "token_acc": 0.8884120171673819, "train_speed(iter/s)": 1.498976 }, { "epoch": 0.5023349470888137, "grad_norm": 4.886120319366455, "learning_rate": 9.753010685241415e-05, "loss": 0.35011651515960696, "memory(GiB)": 61.91, "step": 11725, "token_acc": 0.932806324110672, "train_speed(iter/s)": 1.498937 }, { "epoch": 0.5025491624180627, "grad_norm": 1.2414103746414185, "learning_rate": 9.752801741993781e-05, "loss": 0.2849865913391113, "memory(GiB)": 61.91, "step": 11730, "token_acc": 0.9316239316239316, "train_speed(iter/s)": 1.498926 }, { "epoch": 0.5027633777473116, "grad_norm": 2.718902826309204, "learning_rate": 9.752592712644686e-05, "loss": 0.6359911441802979, "memory(GiB)": 61.91, "step": 11735, "token_acc": 0.8580246913580247, "train_speed(iter/s)": 1.498921 }, { "epoch": 0.5029775930765605, "grad_norm": 2.390573501586914, "learning_rate": 9.752383597197916e-05, "loss": 0.20740032196044922, "memory(GiB)": 61.91, "step": 11740, "token_acc": 0.9519230769230769, "train_speed(iter/s)": 1.49891 }, { "epoch": 0.5031918084058096, "grad_norm": 3.059730291366577, "learning_rate": 9.752174395657259e-05, "loss": 0.24524540901184083, "memory(GiB)": 61.91, "step": 11745, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.498905 }, { "epoch": 0.5034060237350585, "grad_norm": 3.401766061782837, "learning_rate": 9.751965108026505e-05, "loss": 0.2665886402130127, "memory(GiB)": 61.91, "step": 11750, "token_acc": 0.9568106312292359, "train_speed(iter/s)": 1.49889 }, { "epoch": 0.5036202390643074, "grad_norm": 2.4849658012390137, "learning_rate": 9.751755734309443e-05, "loss": 0.41339592933654784, "memory(GiB)": 61.91, "step": 11755, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.498857 }, { "epoch": 0.5038344543935565, "grad_norm": 1.2633259296417236, "learning_rate": 9.751546274509871e-05, "loss": 0.4778024196624756, "memory(GiB)": 61.91, "step": 11760, "token_acc": 0.9098837209302325, "train_speed(iter/s)": 1.498837 }, { "epoch": 0.5040486697228054, "grad_norm": 3.3610551357269287, "learning_rate": 9.75133672863158e-05, "loss": 0.3482205390930176, "memory(GiB)": 61.91, "step": 11765, "token_acc": 0.9288135593220339, "train_speed(iter/s)": 1.498808 }, { "epoch": 0.5042628850520543, "grad_norm": 0.4551701247692108, "learning_rate": 9.751127096678366e-05, "loss": 0.2983554840087891, "memory(GiB)": 61.91, "step": 11770, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.498853 }, { "epoch": 0.5044771003813033, "grad_norm": 0.49810534715652466, "learning_rate": 9.75091737865403e-05, "loss": 0.13609750270843507, "memory(GiB)": 61.91, "step": 11775, "token_acc": 0.9692307692307692, "train_speed(iter/s)": 1.49883 }, { "epoch": 0.5046913157105523, "grad_norm": 4.073742389678955, "learning_rate": 9.750707574562368e-05, "loss": 0.20907375812530518, "memory(GiB)": 61.91, "step": 11780, "token_acc": 0.9645669291338582, "train_speed(iter/s)": 1.498858 }, { "epoch": 0.5049055310398012, "grad_norm": 10.750560760498047, "learning_rate": 9.75049768440718e-05, "loss": 0.6448709011077881, "memory(GiB)": 61.91, "step": 11785, "token_acc": 0.8850931677018633, "train_speed(iter/s)": 1.498826 }, { "epoch": 0.5051197463690502, "grad_norm": 2.8313822746276855, "learning_rate": 9.750287708192272e-05, "loss": 0.3769326448440552, "memory(GiB)": 61.91, "step": 11790, "token_acc": 0.9298245614035088, "train_speed(iter/s)": 1.498811 }, { "epoch": 0.5053339616982991, "grad_norm": 4.731559753417969, "learning_rate": 9.750077645921445e-05, "loss": 0.27373127937316893, "memory(GiB)": 61.91, "step": 11795, "token_acc": 0.9430604982206405, "train_speed(iter/s)": 1.498774 }, { "epoch": 0.5055481770275481, "grad_norm": 2.781751871109009, "learning_rate": 9.749867497598508e-05, "loss": 0.29472975730895995, "memory(GiB)": 61.91, "step": 11800, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.498822 }, { "epoch": 0.5057623923567971, "grad_norm": 3.3122825622558594, "learning_rate": 9.749657263227263e-05, "loss": 0.41118197441101073, "memory(GiB)": 61.91, "step": 11805, "token_acc": 0.9044368600682594, "train_speed(iter/s)": 1.498797 }, { "epoch": 0.505976607686046, "grad_norm": 1.995957612991333, "learning_rate": 9.749446942811522e-05, "loss": 0.41576647758483887, "memory(GiB)": 61.91, "step": 11810, "token_acc": 0.9107142857142857, "train_speed(iter/s)": 1.498782 }, { "epoch": 0.5061908230152949, "grad_norm": 2.804713249206543, "learning_rate": 9.749236536355094e-05, "loss": 0.4786635398864746, "memory(GiB)": 61.91, "step": 11815, "token_acc": 0.9053254437869822, "train_speed(iter/s)": 1.498776 }, { "epoch": 0.506405038344544, "grad_norm": 2.4473214149475098, "learning_rate": 9.749026043861791e-05, "loss": 0.25251524448394774, "memory(GiB)": 61.91, "step": 11820, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.498822 }, { "epoch": 0.5066192536737929, "grad_norm": 3.9618875980377197, "learning_rate": 9.748815465335426e-05, "loss": 0.5148707389831543, "memory(GiB)": 61.91, "step": 11825, "token_acc": 0.8827838827838828, "train_speed(iter/s)": 1.498786 }, { "epoch": 0.5068334690030418, "grad_norm": 4.451826095581055, "learning_rate": 9.748604800779814e-05, "loss": 0.423082160949707, "memory(GiB)": 61.91, "step": 11830, "token_acc": 0.9057971014492754, "train_speed(iter/s)": 1.498772 }, { "epoch": 0.5070476843322909, "grad_norm": 3.118638277053833, "learning_rate": 9.748394050198773e-05, "loss": 0.4295830249786377, "memory(GiB)": 61.91, "step": 11835, "token_acc": 0.9305135951661632, "train_speed(iter/s)": 1.498762 }, { "epoch": 0.5072618996615398, "grad_norm": 2.529816150665283, "learning_rate": 9.748183213596118e-05, "loss": 0.4864049434661865, "memory(GiB)": 61.91, "step": 11840, "token_acc": 0.8876404494382022, "train_speed(iter/s)": 1.498744 }, { "epoch": 0.5074761149907887, "grad_norm": 4.092941761016846, "learning_rate": 9.747972290975671e-05, "loss": 0.47177948951721194, "memory(GiB)": 61.91, "step": 11845, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.498782 }, { "epoch": 0.5076903303200377, "grad_norm": 4.305933952331543, "learning_rate": 9.747761282341253e-05, "loss": 0.6760017395019531, "memory(GiB)": 61.91, "step": 11850, "token_acc": 0.8844884488448845, "train_speed(iter/s)": 1.498755 }, { "epoch": 0.5079045456492867, "grad_norm": 3.8334522247314453, "learning_rate": 9.747550187696684e-05, "loss": 0.42466087341308595, "memory(GiB)": 61.91, "step": 11855, "token_acc": 0.9169675090252708, "train_speed(iter/s)": 1.498828 }, { "epoch": 0.5081187609785356, "grad_norm": 4.99958610534668, "learning_rate": 9.747339007045789e-05, "loss": 0.6982944488525391, "memory(GiB)": 61.91, "step": 11860, "token_acc": 0.8770764119601329, "train_speed(iter/s)": 1.498823 }, { "epoch": 0.5083329763077846, "grad_norm": 3.4633126258850098, "learning_rate": 9.747127740392394e-05, "loss": 0.2717592716217041, "memory(GiB)": 61.91, "step": 11865, "token_acc": 0.9380165289256198, "train_speed(iter/s)": 1.49884 }, { "epoch": 0.5085471916370335, "grad_norm": 3.1933844089508057, "learning_rate": 9.746916387740329e-05, "loss": 0.2939890384674072, "memory(GiB)": 61.91, "step": 11870, "token_acc": 0.9272030651340997, "train_speed(iter/s)": 1.49884 }, { "epoch": 0.5087614069662825, "grad_norm": 0.9778139591217041, "learning_rate": 9.74670494909342e-05, "loss": 0.037975388765335086, "memory(GiB)": 61.91, "step": 11875, "token_acc": 0.9927007299270073, "train_speed(iter/s)": 1.498816 }, { "epoch": 0.5089756222955315, "grad_norm": 5.115503787994385, "learning_rate": 9.746493424455498e-05, "loss": 0.8407869338989258, "memory(GiB)": 61.91, "step": 11880, "token_acc": 0.8445945945945946, "train_speed(iter/s)": 1.498846 }, { "epoch": 0.5091898376247804, "grad_norm": 1.2167021036148071, "learning_rate": 9.746281813830394e-05, "loss": 0.6066072463989258, "memory(GiB)": 61.91, "step": 11885, "token_acc": 0.8660130718954249, "train_speed(iter/s)": 1.498887 }, { "epoch": 0.5094040529540294, "grad_norm": 3.746429681777954, "learning_rate": 9.746070117221944e-05, "loss": 0.35821003913879396, "memory(GiB)": 61.91, "step": 11890, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.498859 }, { "epoch": 0.5096182682832784, "grad_norm": 1.8639062643051147, "learning_rate": 9.745858334633982e-05, "loss": 0.46844019889831545, "memory(GiB)": 61.91, "step": 11895, "token_acc": 0.8931750741839762, "train_speed(iter/s)": 1.498789 }, { "epoch": 0.5098324836125273, "grad_norm": 3.029054880142212, "learning_rate": 9.745646466070342e-05, "loss": 0.5110203266143799, "memory(GiB)": 61.91, "step": 11900, "token_acc": 0.8945578231292517, "train_speed(iter/s)": 1.498798 }, { "epoch": 0.5100466989417762, "grad_norm": 9.540836334228516, "learning_rate": 9.745434511534867e-05, "loss": 0.5523706436157226, "memory(GiB)": 61.91, "step": 11905, "token_acc": 0.8875502008032129, "train_speed(iter/s)": 1.498792 }, { "epoch": 0.5102609142710253, "grad_norm": 5.300279140472412, "learning_rate": 9.745222471031392e-05, "loss": 0.5052375793457031, "memory(GiB)": 61.91, "step": 11910, "token_acc": 0.9178885630498533, "train_speed(iter/s)": 1.498806 }, { "epoch": 0.5104751296002742, "grad_norm": 4.351354598999023, "learning_rate": 9.745010344563763e-05, "loss": 0.4628037452697754, "memory(GiB)": 61.91, "step": 11915, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.498883 }, { "epoch": 0.5106893449295231, "grad_norm": 2.0403239727020264, "learning_rate": 9.744798132135819e-05, "loss": 0.27713894844055176, "memory(GiB)": 61.91, "step": 11920, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.498851 }, { "epoch": 0.5109035602587721, "grad_norm": 2.803433418273926, "learning_rate": 9.744585833751405e-05, "loss": 0.30646069049835206, "memory(GiB)": 61.91, "step": 11925, "token_acc": 0.9371069182389937, "train_speed(iter/s)": 1.498859 }, { "epoch": 0.5111177755880211, "grad_norm": 4.330046653747559, "learning_rate": 9.74437344941437e-05, "loss": 0.4329432487487793, "memory(GiB)": 61.91, "step": 11930, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.498829 }, { "epoch": 0.51133199091727, "grad_norm": 1.8169386386871338, "learning_rate": 9.744160979128559e-05, "loss": 0.5005404472351074, "memory(GiB)": 61.91, "step": 11935, "token_acc": 0.8905109489051095, "train_speed(iter/s)": 1.498846 }, { "epoch": 0.511546206246519, "grad_norm": 1.9133203029632568, "learning_rate": 9.743948422897819e-05, "loss": 0.2761404514312744, "memory(GiB)": 61.91, "step": 11940, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.498817 }, { "epoch": 0.511760421575768, "grad_norm": 2.189591407775879, "learning_rate": 9.743735780726005e-05, "loss": 0.3386735200881958, "memory(GiB)": 61.91, "step": 11945, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.498792 }, { "epoch": 0.5119746369050169, "grad_norm": 1.662418007850647, "learning_rate": 9.743523052616968e-05, "loss": 0.3594545841217041, "memory(GiB)": 61.91, "step": 11950, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.498796 }, { "epoch": 0.5121888522342659, "grad_norm": 3.829617977142334, "learning_rate": 9.743310238574561e-05, "loss": 0.5298149585723877, "memory(GiB)": 61.91, "step": 11955, "token_acc": 0.8937728937728938, "train_speed(iter/s)": 1.498776 }, { "epoch": 0.5124030675635148, "grad_norm": 6.214954376220703, "learning_rate": 9.743097338602638e-05, "loss": 0.5200906753540039, "memory(GiB)": 61.91, "step": 11960, "token_acc": 0.8881578947368421, "train_speed(iter/s)": 1.498708 }, { "epoch": 0.5126172828927638, "grad_norm": 1.762589454650879, "learning_rate": 9.742884352705057e-05, "loss": 0.3737818241119385, "memory(GiB)": 61.91, "step": 11965, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.498672 }, { "epoch": 0.5128314982220128, "grad_norm": 0.8091220855712891, "learning_rate": 9.742671280885678e-05, "loss": 0.43362841606140134, "memory(GiB)": 61.91, "step": 11970, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.498712 }, { "epoch": 0.5130457135512617, "grad_norm": 2.2912375926971436, "learning_rate": 9.74245812314836e-05, "loss": 0.4212669372558594, "memory(GiB)": 61.91, "step": 11975, "token_acc": 0.9216867469879518, "train_speed(iter/s)": 1.498781 }, { "epoch": 0.5132599288805106, "grad_norm": 1.0757673978805542, "learning_rate": 9.742244879496964e-05, "loss": 0.1435302734375, "memory(GiB)": 61.91, "step": 11980, "token_acc": 0.9659090909090909, "train_speed(iter/s)": 1.498776 }, { "epoch": 0.5134741442097597, "grad_norm": 9.667943954467773, "learning_rate": 9.742031549935353e-05, "loss": 0.4683395862579346, "memory(GiB)": 61.91, "step": 11985, "token_acc": 0.9106628242074928, "train_speed(iter/s)": 1.498747 }, { "epoch": 0.5136883595390086, "grad_norm": 3.732602596282959, "learning_rate": 9.741818134467393e-05, "loss": 0.5413120269775391, "memory(GiB)": 61.91, "step": 11990, "token_acc": 0.8902077151335311, "train_speed(iter/s)": 1.49872 }, { "epoch": 0.5139025748682575, "grad_norm": 2.0410916805267334, "learning_rate": 9.741604633096947e-05, "loss": 0.6873550891876221, "memory(GiB)": 61.91, "step": 11995, "token_acc": 0.8812949640287769, "train_speed(iter/s)": 1.498753 }, { "epoch": 0.5141167901975066, "grad_norm": 2.344332218170166, "learning_rate": 9.741391045827888e-05, "loss": 0.36939547061920164, "memory(GiB)": 61.91, "step": 12000, "token_acc": 0.9276018099547512, "train_speed(iter/s)": 1.498751 }, { "epoch": 0.5141167901975066, "eval_loss": 2.4543375968933105, "eval_runtime": 13.9499, "eval_samples_per_second": 7.169, "eval_steps_per_second": 7.169, "eval_token_acc": 0.42052565707133915, "step": 12000 }, { "epoch": 0.5143310055267555, "grad_norm": 1.9370639324188232, "learning_rate": 9.741177372664081e-05, "loss": 0.3360904216766357, "memory(GiB)": 61.91, "step": 12005, "token_acc": 0.5457979225684608, "train_speed(iter/s)": 1.495924 }, { "epoch": 0.5145452208560044, "grad_norm": 2.9757943153381348, "learning_rate": 9.7409636136094e-05, "loss": 0.28811743259429934, "memory(GiB)": 61.91, "step": 12010, "token_acc": 0.9603174603174603, "train_speed(iter/s)": 1.495909 }, { "epoch": 0.5147594361852534, "grad_norm": 1.9000308513641357, "learning_rate": 9.740749768667715e-05, "loss": 0.5960486888885498, "memory(GiB)": 61.91, "step": 12015, "token_acc": 0.8709677419354839, "train_speed(iter/s)": 1.495988 }, { "epoch": 0.5149736515145024, "grad_norm": 2.5099427700042725, "learning_rate": 9.740535837842899e-05, "loss": 0.2592325687408447, "memory(GiB)": 61.91, "step": 12020, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.495997 }, { "epoch": 0.5151878668437513, "grad_norm": 3.237865686416626, "learning_rate": 9.74032182113883e-05, "loss": 0.6058212280273437, "memory(GiB)": 61.91, "step": 12025, "token_acc": 0.8850174216027874, "train_speed(iter/s)": 1.496128 }, { "epoch": 0.5154020821730003, "grad_norm": 8.451794624328613, "learning_rate": 9.740107718559385e-05, "loss": 0.43600993156433104, "memory(GiB)": 61.91, "step": 12030, "token_acc": 0.9115853658536586, "train_speed(iter/s)": 1.496095 }, { "epoch": 0.5156162975022492, "grad_norm": 2.6820943355560303, "learning_rate": 9.739893530108442e-05, "loss": 0.288727855682373, "memory(GiB)": 61.91, "step": 12035, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.496045 }, { "epoch": 0.5158305128314982, "grad_norm": 5.928441047668457, "learning_rate": 9.739679255789881e-05, "loss": 0.4487174987792969, "memory(GiB)": 61.91, "step": 12040, "token_acc": 0.9157088122605364, "train_speed(iter/s)": 1.496046 }, { "epoch": 0.5160447281607472, "grad_norm": 2.544163703918457, "learning_rate": 9.739464895607584e-05, "loss": 0.39589767456054686, "memory(GiB)": 61.91, "step": 12045, "token_acc": 0.9217391304347826, "train_speed(iter/s)": 1.496013 }, { "epoch": 0.5162589434899961, "grad_norm": 2.323228597640991, "learning_rate": 9.739250449565436e-05, "loss": 0.39283027648925783, "memory(GiB)": 61.91, "step": 12050, "token_acc": 0.9131736526946108, "train_speed(iter/s)": 1.495984 }, { "epoch": 0.5164731588192452, "grad_norm": 3.9587008953094482, "learning_rate": 9.739035917667319e-05, "loss": 0.6828359603881836, "memory(GiB)": 61.91, "step": 12055, "token_acc": 0.8597785977859779, "train_speed(iter/s)": 1.496 }, { "epoch": 0.5166873741484941, "grad_norm": 3.4365808963775635, "learning_rate": 9.738821299917121e-05, "loss": 0.22077653408050538, "memory(GiB)": 61.91, "step": 12060, "token_acc": 0.9611307420494699, "train_speed(iter/s)": 1.495975 }, { "epoch": 0.516901589477743, "grad_norm": 4.879485607147217, "learning_rate": 9.738606596318728e-05, "loss": 0.4388290882110596, "memory(GiB)": 61.91, "step": 12065, "token_acc": 0.9026548672566371, "train_speed(iter/s)": 1.49596 }, { "epoch": 0.517115804806992, "grad_norm": 1.3596863746643066, "learning_rate": 9.738391806876034e-05, "loss": 0.2892922401428223, "memory(GiB)": 61.91, "step": 12070, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.495927 }, { "epoch": 0.517330020136241, "grad_norm": 5.179567337036133, "learning_rate": 9.738176931592926e-05, "loss": 0.49228343963623045, "memory(GiB)": 61.91, "step": 12075, "token_acc": 0.9022801302931596, "train_speed(iter/s)": 1.495881 }, { "epoch": 0.5175442354654899, "grad_norm": 2.2047479152679443, "learning_rate": 9.7379619704733e-05, "loss": 0.31847071647644043, "memory(GiB)": 61.91, "step": 12080, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.49586 }, { "epoch": 0.5177584507947389, "grad_norm": 2.0995304584503174, "learning_rate": 9.737746923521046e-05, "loss": 0.4737997055053711, "memory(GiB)": 61.91, "step": 12085, "token_acc": 0.9020979020979021, "train_speed(iter/s)": 1.495846 }, { "epoch": 0.5179726661239878, "grad_norm": 0.3790076971054077, "learning_rate": 9.737531790740062e-05, "loss": 0.3286629915237427, "memory(GiB)": 61.91, "step": 12090, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.495857 }, { "epoch": 0.5181868814532368, "grad_norm": 3.472923517227173, "learning_rate": 9.737316572134247e-05, "loss": 0.39309158325195315, "memory(GiB)": 61.91, "step": 12095, "token_acc": 0.9150326797385621, "train_speed(iter/s)": 1.495916 }, { "epoch": 0.5184010967824858, "grad_norm": 4.3440632820129395, "learning_rate": 9.737101267707498e-05, "loss": 0.8626234054565429, "memory(GiB)": 61.91, "step": 12100, "token_acc": 0.8207885304659498, "train_speed(iter/s)": 1.495996 }, { "epoch": 0.5186153121117347, "grad_norm": 4.244625091552734, "learning_rate": 9.736885877463716e-05, "loss": 0.5035964965820312, "memory(GiB)": 61.91, "step": 12105, "token_acc": 0.8949416342412452, "train_speed(iter/s)": 1.496061 }, { "epoch": 0.5188295274409837, "grad_norm": 1.8647189140319824, "learning_rate": 9.736670401406802e-05, "loss": 0.6073509216308594, "memory(GiB)": 61.91, "step": 12110, "token_acc": 0.8701298701298701, "train_speed(iter/s)": 1.496032 }, { "epoch": 0.5190437427702327, "grad_norm": 3.7943227291107178, "learning_rate": 9.736454839540661e-05, "loss": 0.2654410362243652, "memory(GiB)": 61.91, "step": 12115, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.496025 }, { "epoch": 0.5192579580994816, "grad_norm": 1.8608444929122925, "learning_rate": 9.736239191869197e-05, "loss": 0.32226593494415284, "memory(GiB)": 61.91, "step": 12120, "token_acc": 0.9475806451612904, "train_speed(iter/s)": 1.496034 }, { "epoch": 0.5194721734287305, "grad_norm": 5.080722332000732, "learning_rate": 9.736023458396318e-05, "loss": 0.32826478481292726, "memory(GiB)": 61.91, "step": 12125, "token_acc": 0.9152542372881356, "train_speed(iter/s)": 1.495986 }, { "epoch": 0.5196863887579796, "grad_norm": 3.2507479190826416, "learning_rate": 9.735807639125928e-05, "loss": 0.5881553649902344, "memory(GiB)": 61.91, "step": 12130, "token_acc": 0.8613569321533924, "train_speed(iter/s)": 1.496024 }, { "epoch": 0.5199006040872285, "grad_norm": 3.570373058319092, "learning_rate": 9.735591734061945e-05, "loss": 0.46005725860595703, "memory(GiB)": 61.91, "step": 12135, "token_acc": 0.8853754940711462, "train_speed(iter/s)": 1.496036 }, { "epoch": 0.5201148194164774, "grad_norm": 0.4238989055156708, "learning_rate": 9.735375743208273e-05, "loss": 0.4264807224273682, "memory(GiB)": 61.91, "step": 12140, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.495991 }, { "epoch": 0.5203290347457264, "grad_norm": 2.519395351409912, "learning_rate": 9.735159666568826e-05, "loss": 0.6516101360321045, "memory(GiB)": 61.91, "step": 12145, "token_acc": 0.8357771260997068, "train_speed(iter/s)": 1.495971 }, { "epoch": 0.5205432500749754, "grad_norm": 2.3026540279388428, "learning_rate": 9.734943504147521e-05, "loss": 0.35778758525848386, "memory(GiB)": 61.91, "step": 12150, "token_acc": 0.92, "train_speed(iter/s)": 1.495935 }, { "epoch": 0.5207574654042243, "grad_norm": 2.2468626499176025, "learning_rate": 9.734727255948273e-05, "loss": 0.6154538154602051, "memory(GiB)": 61.91, "step": 12155, "token_acc": 0.8801261829652997, "train_speed(iter/s)": 1.495909 }, { "epoch": 0.5209716807334733, "grad_norm": 0.7530818581581116, "learning_rate": 9.734510921974998e-05, "loss": 0.2256394624710083, "memory(GiB)": 61.91, "step": 12160, "token_acc": 0.9470404984423676, "train_speed(iter/s)": 1.495933 }, { "epoch": 0.5211858960627223, "grad_norm": 1.4180082082748413, "learning_rate": 9.734294502231618e-05, "loss": 0.5919866561889648, "memory(GiB)": 61.91, "step": 12165, "token_acc": 0.8586956521739131, "train_speed(iter/s)": 1.495898 }, { "epoch": 0.5214001113919712, "grad_norm": 3.487553834915161, "learning_rate": 9.73407799672205e-05, "loss": 0.6225837707519531, "memory(GiB)": 61.91, "step": 12170, "token_acc": 0.875, "train_speed(iter/s)": 1.495895 }, { "epoch": 0.5216143267212202, "grad_norm": 0.8870673775672913, "learning_rate": 9.73386140545022e-05, "loss": 0.3407327890396118, "memory(GiB)": 61.91, "step": 12175, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.495825 }, { "epoch": 0.5218285420504691, "grad_norm": 4.447450160980225, "learning_rate": 9.73364472842005e-05, "loss": 0.3827086925506592, "memory(GiB)": 61.91, "step": 12180, "token_acc": 0.9299610894941635, "train_speed(iter/s)": 1.495776 }, { "epoch": 0.5220427573797181, "grad_norm": 4.267645835876465, "learning_rate": 9.733427965635465e-05, "loss": 0.4170802593231201, "memory(GiB)": 61.91, "step": 12185, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.495773 }, { "epoch": 0.5222569727089671, "grad_norm": 4.827301025390625, "learning_rate": 9.73321111710039e-05, "loss": 0.34223456382751466, "memory(GiB)": 61.91, "step": 12190, "token_acc": 0.9213483146067416, "train_speed(iter/s)": 1.49575 }, { "epoch": 0.522471188038216, "grad_norm": 1.1807361841201782, "learning_rate": 9.732994182818757e-05, "loss": 0.2292494535446167, "memory(GiB)": 61.91, "step": 12195, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.495711 }, { "epoch": 0.5226854033674649, "grad_norm": 2.6098792552948, "learning_rate": 9.732777162794496e-05, "loss": 0.2562653303146362, "memory(GiB)": 61.91, "step": 12200, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.495709 }, { "epoch": 0.522899618696714, "grad_norm": 3.451625347137451, "learning_rate": 9.732560057031534e-05, "loss": 0.4077170372009277, "memory(GiB)": 61.91, "step": 12205, "token_acc": 0.9116719242902208, "train_speed(iter/s)": 1.495699 }, { "epoch": 0.5231138340259629, "grad_norm": 2.348679542541504, "learning_rate": 9.732342865533809e-05, "loss": 0.45265998840332033, "memory(GiB)": 61.91, "step": 12210, "token_acc": 0.8951310861423221, "train_speed(iter/s)": 1.495664 }, { "epoch": 0.5233280493552118, "grad_norm": 3.244274139404297, "learning_rate": 9.732125588305252e-05, "loss": 0.31479339599609374, "memory(GiB)": 61.91, "step": 12215, "token_acc": 0.939297124600639, "train_speed(iter/s)": 1.495652 }, { "epoch": 0.5235422646844609, "grad_norm": 1.6108759641647339, "learning_rate": 9.731908225349803e-05, "loss": 0.3947009801864624, "memory(GiB)": 61.91, "step": 12220, "token_acc": 0.9196428571428571, "train_speed(iter/s)": 1.495639 }, { "epoch": 0.5237564800137098, "grad_norm": 3.6354432106018066, "learning_rate": 9.731690776671396e-05, "loss": 0.32954959869384765, "memory(GiB)": 61.91, "step": 12225, "token_acc": 0.9504950495049505, "train_speed(iter/s)": 1.495631 }, { "epoch": 0.5239706953429587, "grad_norm": 2.776472568511963, "learning_rate": 9.731473242273974e-05, "loss": 0.33017525672912595, "memory(GiB)": 61.91, "step": 12230, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.495638 }, { "epoch": 0.5241849106722077, "grad_norm": 2.9493374824523926, "learning_rate": 9.731255622161474e-05, "loss": 0.42521133422851565, "memory(GiB)": 61.91, "step": 12235, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.495686 }, { "epoch": 0.5243991260014567, "grad_norm": 4.754340648651123, "learning_rate": 9.731037916337839e-05, "loss": 0.46363325119018556, "memory(GiB)": 61.91, "step": 12240, "token_acc": 0.8809523809523809, "train_speed(iter/s)": 1.495724 }, { "epoch": 0.5246133413307056, "grad_norm": 2.474245548248291, "learning_rate": 9.730820124807017e-05, "loss": 0.47938175201416017, "memory(GiB)": 61.91, "step": 12245, "token_acc": 0.8982035928143712, "train_speed(iter/s)": 1.495766 }, { "epoch": 0.5248275566599546, "grad_norm": 3.9538309574127197, "learning_rate": 9.730602247572948e-05, "loss": 0.35663666725158694, "memory(GiB)": 61.91, "step": 12250, "token_acc": 0.924, "train_speed(iter/s)": 1.495747 }, { "epoch": 0.5250417719892035, "grad_norm": 3.9435153007507324, "learning_rate": 9.730384284639584e-05, "loss": 0.3821770668029785, "memory(GiB)": 61.91, "step": 12255, "token_acc": 0.9161073825503355, "train_speed(iter/s)": 1.495759 }, { "epoch": 0.5252559873184525, "grad_norm": 2.4234161376953125, "learning_rate": 9.730166236010869e-05, "loss": 0.5741121768951416, "memory(GiB)": 61.91, "step": 12260, "token_acc": 0.875, "train_speed(iter/s)": 1.495743 }, { "epoch": 0.5254702026477015, "grad_norm": 4.430825233459473, "learning_rate": 9.729948101690757e-05, "loss": 0.2634239435195923, "memory(GiB)": 61.91, "step": 12265, "token_acc": 0.9234234234234234, "train_speed(iter/s)": 1.495716 }, { "epoch": 0.5256844179769504, "grad_norm": 3.680152654647827, "learning_rate": 9.729729881683197e-05, "loss": 0.3991286277770996, "memory(GiB)": 61.91, "step": 12270, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.495707 }, { "epoch": 0.5258986333061993, "grad_norm": 2.181058406829834, "learning_rate": 9.729511575992145e-05, "loss": 0.38998115062713623, "memory(GiB)": 61.91, "step": 12275, "token_acc": 0.9127906976744186, "train_speed(iter/s)": 1.495735 }, { "epoch": 0.5261128486354484, "grad_norm": 2.465670347213745, "learning_rate": 9.729293184621552e-05, "loss": 0.2688431262969971, "memory(GiB)": 61.91, "step": 12280, "token_acc": 0.9318885448916409, "train_speed(iter/s)": 1.495745 }, { "epoch": 0.5263270639646973, "grad_norm": 3.1682870388031006, "learning_rate": 9.729074707575377e-05, "loss": 0.5518004894256592, "memory(GiB)": 61.91, "step": 12285, "token_acc": 0.8896551724137931, "train_speed(iter/s)": 1.495721 }, { "epoch": 0.5265412792939462, "grad_norm": 3.041635513305664, "learning_rate": 9.728856144857579e-05, "loss": 0.26360354423522947, "memory(GiB)": 61.91, "step": 12290, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.495708 }, { "epoch": 0.5267554946231953, "grad_norm": 3.7464704513549805, "learning_rate": 9.728637496472115e-05, "loss": 0.27412374019622804, "memory(GiB)": 61.91, "step": 12295, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.495727 }, { "epoch": 0.5269697099524442, "grad_norm": 3.225273370742798, "learning_rate": 9.728418762422947e-05, "loss": 0.4324338912963867, "memory(GiB)": 61.91, "step": 12300, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.495755 }, { "epoch": 0.5271839252816931, "grad_norm": 0.9132187962532043, "learning_rate": 9.728199942714036e-05, "loss": 0.4057960033416748, "memory(GiB)": 61.91, "step": 12305, "token_acc": 0.9140625, "train_speed(iter/s)": 1.495727 }, { "epoch": 0.5273981406109421, "grad_norm": 1.8627687692642212, "learning_rate": 9.727981037349349e-05, "loss": 0.21397392749786376, "memory(GiB)": 61.91, "step": 12310, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.495713 }, { "epoch": 0.5276123559401911, "grad_norm": 0.3842521607875824, "learning_rate": 9.727762046332851e-05, "loss": 0.38910727500915526, "memory(GiB)": 61.91, "step": 12315, "token_acc": 0.9098039215686274, "train_speed(iter/s)": 1.495724 }, { "epoch": 0.52782657126944, "grad_norm": 1.9252909421920776, "learning_rate": 9.727542969668507e-05, "loss": 0.6561356067657471, "memory(GiB)": 61.91, "step": 12320, "token_acc": 0.8634686346863468, "train_speed(iter/s)": 1.495761 }, { "epoch": 0.528040786598689, "grad_norm": 8.956718444824219, "learning_rate": 9.72732380736029e-05, "loss": 0.5487292289733887, "memory(GiB)": 61.91, "step": 12325, "token_acc": 0.8854166666666666, "train_speed(iter/s)": 1.495717 }, { "epoch": 0.528255001927938, "grad_norm": 3.1522910594940186, "learning_rate": 9.727104559412163e-05, "loss": 0.29309003353118895, "memory(GiB)": 61.91, "step": 12330, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.495704 }, { "epoch": 0.5284692172571869, "grad_norm": 4.158008575439453, "learning_rate": 9.726885225828106e-05, "loss": 0.37363224029541015, "memory(GiB)": 61.91, "step": 12335, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.495696 }, { "epoch": 0.5286834325864359, "grad_norm": 3.4331467151641846, "learning_rate": 9.726665806612087e-05, "loss": 0.4514486789703369, "memory(GiB)": 61.91, "step": 12340, "token_acc": 0.908256880733945, "train_speed(iter/s)": 1.495707 }, { "epoch": 0.5288976479156848, "grad_norm": 3.966679096221924, "learning_rate": 9.726446301768085e-05, "loss": 0.6376059532165528, "memory(GiB)": 61.91, "step": 12345, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.495664 }, { "epoch": 0.5291118632449338, "grad_norm": 5.115257263183594, "learning_rate": 9.726226711300074e-05, "loss": 0.7503841400146485, "memory(GiB)": 61.91, "step": 12350, "token_acc": 0.8333333333333334, "train_speed(iter/s)": 1.495618 }, { "epoch": 0.5293260785741828, "grad_norm": 6.8202691078186035, "learning_rate": 9.726007035212033e-05, "loss": 0.39874188899993895, "memory(GiB)": 61.91, "step": 12355, "token_acc": 0.8722222222222222, "train_speed(iter/s)": 1.49558 }, { "epoch": 0.5295402939034317, "grad_norm": 3.2582814693450928, "learning_rate": 9.725787273507939e-05, "loss": 0.6947667598724365, "memory(GiB)": 61.91, "step": 12360, "token_acc": 0.8783783783783784, "train_speed(iter/s)": 1.495561 }, { "epoch": 0.5297545092326806, "grad_norm": 6.8030476570129395, "learning_rate": 9.725567426191777e-05, "loss": 0.6498387813568115, "memory(GiB)": 61.91, "step": 12365, "token_acc": 0.8745519713261649, "train_speed(iter/s)": 1.495666 }, { "epoch": 0.5299687245619297, "grad_norm": 3.68721866607666, "learning_rate": 9.725347493267529e-05, "loss": 0.6059948444366455, "memory(GiB)": 61.91, "step": 12370, "token_acc": 0.875, "train_speed(iter/s)": 1.495671 }, { "epoch": 0.5301829398911786, "grad_norm": 1.7922813892364502, "learning_rate": 9.725127474739176e-05, "loss": 0.49415082931518556, "memory(GiB)": 61.91, "step": 12375, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.49569 }, { "epoch": 0.5303971552204275, "grad_norm": 4.136538982391357, "learning_rate": 9.724907370610707e-05, "loss": 0.23403167724609375, "memory(GiB)": 61.91, "step": 12380, "token_acc": 0.9446640316205533, "train_speed(iter/s)": 1.49568 }, { "epoch": 0.5306113705496766, "grad_norm": 2.1796278953552246, "learning_rate": 9.724687180886109e-05, "loss": 0.2896800756454468, "memory(GiB)": 61.91, "step": 12385, "token_acc": 0.9404388714733543, "train_speed(iter/s)": 1.495676 }, { "epoch": 0.5308255858789255, "grad_norm": 4.327676296234131, "learning_rate": 9.724466905569372e-05, "loss": 0.4006159782409668, "memory(GiB)": 61.91, "step": 12390, "token_acc": 0.9128787878787878, "train_speed(iter/s)": 1.495654 }, { "epoch": 0.5310398012081745, "grad_norm": 0.6979964971542358, "learning_rate": 9.724246544664484e-05, "loss": 0.34466359615325926, "memory(GiB)": 61.91, "step": 12395, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.495618 }, { "epoch": 0.5312540165374234, "grad_norm": 1.3455475568771362, "learning_rate": 9.724026098175438e-05, "loss": 0.36588573455810547, "memory(GiB)": 61.91, "step": 12400, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.495638 }, { "epoch": 0.5314682318666724, "grad_norm": 2.444693088531494, "learning_rate": 9.723805566106227e-05, "loss": 0.3243107795715332, "memory(GiB)": 61.91, "step": 12405, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.495602 }, { "epoch": 0.5316824471959214, "grad_norm": 0.9538853764533997, "learning_rate": 9.723584948460848e-05, "loss": 0.5258162021636963, "memory(GiB)": 61.91, "step": 12410, "token_acc": 0.8720538720538721, "train_speed(iter/s)": 1.495549 }, { "epoch": 0.5318966625251703, "grad_norm": 5.543931007385254, "learning_rate": 9.723364245243296e-05, "loss": 0.31894803047180176, "memory(GiB)": 61.91, "step": 12415, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.495541 }, { "epoch": 0.5321108778544192, "grad_norm": 4.133533477783203, "learning_rate": 9.723143456457571e-05, "loss": 0.48996787071228026, "memory(GiB)": 61.91, "step": 12420, "token_acc": 0.8871595330739299, "train_speed(iter/s)": 1.49562 }, { "epoch": 0.5323250931836683, "grad_norm": 2.500091314315796, "learning_rate": 9.72292258210767e-05, "loss": 0.39491751194000246, "memory(GiB)": 61.91, "step": 12425, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.495651 }, { "epoch": 0.5325393085129172, "grad_norm": 5.879429817199707, "learning_rate": 9.722701622197596e-05, "loss": 0.4640162467956543, "memory(GiB)": 61.91, "step": 12430, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.495611 }, { "epoch": 0.5327535238421661, "grad_norm": 1.2026617527008057, "learning_rate": 9.722480576731354e-05, "loss": 0.27720816135406495, "memory(GiB)": 61.91, "step": 12435, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.495592 }, { "epoch": 0.5329677391714152, "grad_norm": 2.7020487785339355, "learning_rate": 9.722259445712945e-05, "loss": 0.5515942096710205, "memory(GiB)": 61.91, "step": 12440, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.495574 }, { "epoch": 0.5331819545006641, "grad_norm": 1.2416417598724365, "learning_rate": 9.722038229146375e-05, "loss": 0.3095396518707275, "memory(GiB)": 61.91, "step": 12445, "token_acc": 0.9171779141104295, "train_speed(iter/s)": 1.495539 }, { "epoch": 0.533396169829913, "grad_norm": 7.995690822601318, "learning_rate": 9.721816927035654e-05, "loss": 0.2334822177886963, "memory(GiB)": 61.91, "step": 12450, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.495495 }, { "epoch": 0.533610385159162, "grad_norm": 6.231065273284912, "learning_rate": 9.721595539384791e-05, "loss": 0.389294695854187, "memory(GiB)": 61.91, "step": 12455, "token_acc": 0.9264069264069265, "train_speed(iter/s)": 1.495506 }, { "epoch": 0.533824600488411, "grad_norm": 4.365808963775635, "learning_rate": 9.721374066197795e-05, "loss": 0.41240901947021485, "memory(GiB)": 61.91, "step": 12460, "token_acc": 0.9116465863453815, "train_speed(iter/s)": 1.495474 }, { "epoch": 0.5340388158176599, "grad_norm": 3.7024595737457275, "learning_rate": 9.721152507478677e-05, "loss": 0.3579502820968628, "memory(GiB)": 61.91, "step": 12465, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.495521 }, { "epoch": 0.5342530311469089, "grad_norm": 3.141005039215088, "learning_rate": 9.720930863231454e-05, "loss": 0.2831853151321411, "memory(GiB)": 61.91, "step": 12470, "token_acc": 0.9323843416370107, "train_speed(iter/s)": 1.495489 }, { "epoch": 0.5344672464761578, "grad_norm": 1.8041349649429321, "learning_rate": 9.72070913346014e-05, "loss": 0.22698280811309815, "memory(GiB)": 61.91, "step": 12475, "token_acc": 0.9498207885304659, "train_speed(iter/s)": 1.495459 }, { "epoch": 0.5346814618054068, "grad_norm": 2.028538942337036, "learning_rate": 9.720487318168752e-05, "loss": 0.33606393337249757, "memory(GiB)": 61.91, "step": 12480, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.495473 }, { "epoch": 0.5348956771346558, "grad_norm": 2.1764578819274902, "learning_rate": 9.720265417361306e-05, "loss": 0.34403166770935056, "memory(GiB)": 61.91, "step": 12485, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.495454 }, { "epoch": 0.5351098924639047, "grad_norm": 3.033322811126709, "learning_rate": 9.720043431041825e-05, "loss": 0.23395085334777832, "memory(GiB)": 61.91, "step": 12490, "token_acc": 0.9481481481481482, "train_speed(iter/s)": 1.495443 }, { "epoch": 0.5353241077931536, "grad_norm": 6.454158782958984, "learning_rate": 9.719821359214331e-05, "loss": 0.4052289962768555, "memory(GiB)": 61.91, "step": 12495, "token_acc": 0.9042904290429042, "train_speed(iter/s)": 1.495516 }, { "epoch": 0.5355383231224027, "grad_norm": 4.980947494506836, "learning_rate": 9.719599201882845e-05, "loss": 0.37141802310943606, "memory(GiB)": 61.91, "step": 12500, "token_acc": 0.9373040752351097, "train_speed(iter/s)": 1.495561 }, { "epoch": 0.5355383231224027, "eval_loss": 2.5545318126678467, "eval_runtime": 13.376, "eval_samples_per_second": 7.476, "eval_steps_per_second": 7.476, "eval_token_acc": 0.4326923076923077, "step": 12500 }, { "epoch": 0.5357525384516516, "grad_norm": 5.014606475830078, "learning_rate": 9.71937695905139e-05, "loss": 0.3360579490661621, "memory(GiB)": 61.91, "step": 12505, "token_acc": 0.574826560951437, "train_speed(iter/s)": 1.492992 }, { "epoch": 0.5359667537809005, "grad_norm": 0.9198101758956909, "learning_rate": 9.719154630723996e-05, "loss": 0.3297051668167114, "memory(GiB)": 61.91, "step": 12510, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.492989 }, { "epoch": 0.5361809691101496, "grad_norm": 3.3914058208465576, "learning_rate": 9.718932216904689e-05, "loss": 0.45501484870910647, "memory(GiB)": 61.91, "step": 12515, "token_acc": 0.8980392156862745, "train_speed(iter/s)": 1.492966 }, { "epoch": 0.5363951844393985, "grad_norm": 2.0653979778289795, "learning_rate": 9.718709717597498e-05, "loss": 0.24959852695465087, "memory(GiB)": 61.91, "step": 12520, "token_acc": 0.9484848484848485, "train_speed(iter/s)": 1.492923 }, { "epoch": 0.5366093997686474, "grad_norm": 2.810060739517212, "learning_rate": 9.718487132806453e-05, "loss": 0.40211195945739747, "memory(GiB)": 61.91, "step": 12525, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.492898 }, { "epoch": 0.5368236150978964, "grad_norm": 1.7296003103256226, "learning_rate": 9.71826446253559e-05, "loss": 0.23976035118103028, "memory(GiB)": 61.91, "step": 12530, "token_acc": 0.948220064724919, "train_speed(iter/s)": 1.492908 }, { "epoch": 0.5370378304271454, "grad_norm": 2.3204824924468994, "learning_rate": 9.718041706788937e-05, "loss": 0.24142942428588868, "memory(GiB)": 61.91, "step": 12535, "token_acc": 0.9484848484848485, "train_speed(iter/s)": 1.492908 }, { "epoch": 0.5372520457563943, "grad_norm": 3.7545416355133057, "learning_rate": 9.717818865570534e-05, "loss": 0.5297311782836914, "memory(GiB)": 61.91, "step": 12540, "token_acc": 0.8838526912181303, "train_speed(iter/s)": 1.492905 }, { "epoch": 0.5374662610856433, "grad_norm": 4.391420841217041, "learning_rate": 9.717595938884416e-05, "loss": 0.3285609483718872, "memory(GiB)": 61.91, "step": 12545, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.492872 }, { "epoch": 0.5376804764148922, "grad_norm": 3.716578960418701, "learning_rate": 9.717372926734623e-05, "loss": 0.5211369037628174, "memory(GiB)": 61.91, "step": 12550, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.492868 }, { "epoch": 0.5378946917441412, "grad_norm": 4.392141819000244, "learning_rate": 9.717149829125194e-05, "loss": 0.36246490478515625, "memory(GiB)": 61.91, "step": 12555, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.492906 }, { "epoch": 0.5381089070733902, "grad_norm": 4.260949611663818, "learning_rate": 9.71692664606017e-05, "loss": 0.5526125431060791, "memory(GiB)": 61.91, "step": 12560, "token_acc": 0.8706293706293706, "train_speed(iter/s)": 1.492874 }, { "epoch": 0.5383231224026391, "grad_norm": 4.2641401290893555, "learning_rate": 9.716703377543595e-05, "loss": 0.4137476921081543, "memory(GiB)": 61.91, "step": 12565, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.492859 }, { "epoch": 0.538537337731888, "grad_norm": 2.118736982345581, "learning_rate": 9.716480023579516e-05, "loss": 0.2704385995864868, "memory(GiB)": 61.91, "step": 12570, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.492833 }, { "epoch": 0.5387515530611371, "grad_norm": 1.0443006753921509, "learning_rate": 9.716256584171974e-05, "loss": 0.4249587535858154, "memory(GiB)": 61.91, "step": 12575, "token_acc": 0.9161490683229814, "train_speed(iter/s)": 1.492765 }, { "epoch": 0.538965768390386, "grad_norm": 2.501037836074829, "learning_rate": 9.71603305932502e-05, "loss": 0.36573574542999265, "memory(GiB)": 61.91, "step": 12580, "token_acc": 0.9147540983606557, "train_speed(iter/s)": 1.492763 }, { "epoch": 0.5391799837196349, "grad_norm": 1.176613450050354, "learning_rate": 9.715809449042706e-05, "loss": 0.3413656949996948, "memory(GiB)": 61.91, "step": 12585, "token_acc": 0.940677966101695, "train_speed(iter/s)": 1.492729 }, { "epoch": 0.539394199048884, "grad_norm": 7.239649295806885, "learning_rate": 9.715585753329077e-05, "loss": 0.23984005451202392, "memory(GiB)": 61.91, "step": 12590, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.492784 }, { "epoch": 0.5396084143781329, "grad_norm": 4.600289821624756, "learning_rate": 9.715361972188191e-05, "loss": 0.48151121139526365, "memory(GiB)": 61.91, "step": 12595, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.492754 }, { "epoch": 0.5398226297073818, "grad_norm": 2.9765398502349854, "learning_rate": 9.715138105624097e-05, "loss": 0.6090617179870605, "memory(GiB)": 61.91, "step": 12600, "token_acc": 0.8484848484848485, "train_speed(iter/s)": 1.492788 }, { "epoch": 0.5400368450366309, "grad_norm": 3.532555103302002, "learning_rate": 9.714914153640854e-05, "loss": 0.34031665325164795, "memory(GiB)": 61.91, "step": 12605, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.492791 }, { "epoch": 0.5402510603658798, "grad_norm": 3.727386713027954, "learning_rate": 9.714690116242518e-05, "loss": 0.2430474042892456, "memory(GiB)": 61.91, "step": 12610, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.49277 }, { "epoch": 0.5404652756951287, "grad_norm": 6.978740692138672, "learning_rate": 9.714465993433148e-05, "loss": 0.22541074752807616, "memory(GiB)": 61.91, "step": 12615, "token_acc": 0.9520295202952029, "train_speed(iter/s)": 1.492801 }, { "epoch": 0.5406794910243777, "grad_norm": 3.062215805053711, "learning_rate": 9.714241785216804e-05, "loss": 0.393666410446167, "memory(GiB)": 61.91, "step": 12620, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.492795 }, { "epoch": 0.5408937063536267, "grad_norm": 4.969911575317383, "learning_rate": 9.714017491597548e-05, "loss": 0.7478813648223877, "memory(GiB)": 61.91, "step": 12625, "token_acc": 0.8246268656716418, "train_speed(iter/s)": 1.49275 }, { "epoch": 0.5411079216828756, "grad_norm": 3.6489908695220947, "learning_rate": 9.713793112579443e-05, "loss": 0.6580411911010742, "memory(GiB)": 61.91, "step": 12630, "token_acc": 0.8640483383685801, "train_speed(iter/s)": 1.49279 }, { "epoch": 0.5413221370121246, "grad_norm": 2.248872756958008, "learning_rate": 9.713568648166555e-05, "loss": 0.3579464673995972, "memory(GiB)": 61.91, "step": 12635, "token_acc": 0.9316770186335404, "train_speed(iter/s)": 1.492753 }, { "epoch": 0.5415363523413735, "grad_norm": 3.737145185470581, "learning_rate": 9.713344098362948e-05, "loss": 0.36696710586547854, "memory(GiB)": 61.91, "step": 12640, "token_acc": 0.8993055555555556, "train_speed(iter/s)": 1.492721 }, { "epoch": 0.5417505676706225, "grad_norm": 3.96877121925354, "learning_rate": 9.71311946317269e-05, "loss": 0.3942409038543701, "memory(GiB)": 61.91, "step": 12645, "token_acc": 0.908284023668639, "train_speed(iter/s)": 1.492676 }, { "epoch": 0.5419647829998715, "grad_norm": 1.2148613929748535, "learning_rate": 9.712894742599854e-05, "loss": 0.5852044105529786, "memory(GiB)": 61.91, "step": 12650, "token_acc": 0.8528428093645485, "train_speed(iter/s)": 1.492667 }, { "epoch": 0.5421789983291204, "grad_norm": 2.895491600036621, "learning_rate": 9.712669936648507e-05, "loss": 0.2311777353286743, "memory(GiB)": 61.91, "step": 12655, "token_acc": 0.9449152542372882, "train_speed(iter/s)": 1.492634 }, { "epoch": 0.5423932136583693, "grad_norm": 1.3666521310806274, "learning_rate": 9.712445045322725e-05, "loss": 0.4424308776855469, "memory(GiB)": 61.91, "step": 12660, "token_acc": 0.8952095808383234, "train_speed(iter/s)": 1.492617 }, { "epoch": 0.5426074289876184, "grad_norm": 3.0233395099639893, "learning_rate": 9.712220068626578e-05, "loss": 0.35888473987579345, "memory(GiB)": 61.91, "step": 12665, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.492611 }, { "epoch": 0.5428216443168673, "grad_norm": 3.0853729248046875, "learning_rate": 9.711995006564145e-05, "loss": 0.1995012044906616, "memory(GiB)": 61.91, "step": 12670, "token_acc": 0.9501915708812261, "train_speed(iter/s)": 1.492634 }, { "epoch": 0.5430358596461162, "grad_norm": 3.6685714721679688, "learning_rate": 9.711769859139503e-05, "loss": 0.5455767631530761, "memory(GiB)": 61.91, "step": 12675, "token_acc": 0.8806451612903226, "train_speed(iter/s)": 1.492617 }, { "epoch": 0.5432500749753653, "grad_norm": 9.612741470336914, "learning_rate": 9.71154462635673e-05, "loss": 0.6148692607879639, "memory(GiB)": 61.91, "step": 12680, "token_acc": 0.8540540540540541, "train_speed(iter/s)": 1.492634 }, { "epoch": 0.5434642903046142, "grad_norm": 5.480513572692871, "learning_rate": 9.711319308219906e-05, "loss": 0.32258057594299316, "memory(GiB)": 61.91, "step": 12685, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.492643 }, { "epoch": 0.5436785056338631, "grad_norm": 4.095878601074219, "learning_rate": 9.711093904733113e-05, "loss": 0.6041972160339355, "memory(GiB)": 61.91, "step": 12690, "token_acc": 0.865546218487395, "train_speed(iter/s)": 1.492614 }, { "epoch": 0.5438927209631121, "grad_norm": 2.476207733154297, "learning_rate": 9.710868415900433e-05, "loss": 0.6227049827575684, "memory(GiB)": 61.91, "step": 12695, "token_acc": 0.8812260536398467, "train_speed(iter/s)": 1.492618 }, { "epoch": 0.5441069362923611, "grad_norm": 3.75608491897583, "learning_rate": 9.710642841725954e-05, "loss": 0.7794917106628418, "memory(GiB)": 61.91, "step": 12700, "token_acc": 0.8370607028753994, "train_speed(iter/s)": 1.492669 }, { "epoch": 0.54432115162161, "grad_norm": 3.654162645339966, "learning_rate": 9.710417182213762e-05, "loss": 0.374002742767334, "memory(GiB)": 61.91, "step": 12705, "token_acc": 0.9035714285714286, "train_speed(iter/s)": 1.49262 }, { "epoch": 0.544535366950859, "grad_norm": 10.794281959533691, "learning_rate": 9.710191437367942e-05, "loss": 0.6310531616210937, "memory(GiB)": 61.91, "step": 12710, "token_acc": 0.8560311284046692, "train_speed(iter/s)": 1.49266 }, { "epoch": 0.544749582280108, "grad_norm": 3.381904125213623, "learning_rate": 9.709965607192589e-05, "loss": 0.3287661552429199, "memory(GiB)": 61.91, "step": 12715, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.492624 }, { "epoch": 0.5449637976093569, "grad_norm": 0.10352974385023117, "learning_rate": 9.709739691691787e-05, "loss": 0.17052966356277466, "memory(GiB)": 61.91, "step": 12720, "token_acc": 0.9685534591194969, "train_speed(iter/s)": 1.492572 }, { "epoch": 0.5451780129386059, "grad_norm": 2.4556593894958496, "learning_rate": 9.709513690869634e-05, "loss": 0.40941171646118163, "memory(GiB)": 61.91, "step": 12725, "token_acc": 0.8740157480314961, "train_speed(iter/s)": 1.492557 }, { "epoch": 0.5453922282678548, "grad_norm": 4.790131568908691, "learning_rate": 9.709287604730222e-05, "loss": 0.33583390712738037, "memory(GiB)": 61.91, "step": 12730, "token_acc": 0.9404388714733543, "train_speed(iter/s)": 1.492548 }, { "epoch": 0.5456064435971039, "grad_norm": 6.209418296813965, "learning_rate": 9.709061433277647e-05, "loss": 0.343867564201355, "memory(GiB)": 61.91, "step": 12735, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.492542 }, { "epoch": 0.5458206589263528, "grad_norm": 2.5852513313293457, "learning_rate": 9.708835176516008e-05, "loss": 0.30210299491882325, "memory(GiB)": 61.91, "step": 12740, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.492584 }, { "epoch": 0.5460348742556017, "grad_norm": 2.640119791030884, "learning_rate": 9.708608834449402e-05, "loss": 0.6011237144470215, "memory(GiB)": 61.91, "step": 12745, "token_acc": 0.8694158075601375, "train_speed(iter/s)": 1.492647 }, { "epoch": 0.5462490895848507, "grad_norm": 2.866649866104126, "learning_rate": 9.708382407081929e-05, "loss": 0.20138068199157716, "memory(GiB)": 61.91, "step": 12750, "token_acc": 0.946360153256705, "train_speed(iter/s)": 1.492616 }, { "epoch": 0.5464633049140997, "grad_norm": 0.03273199871182442, "learning_rate": 9.708155894417693e-05, "loss": 0.28084986209869384, "memory(GiB)": 61.91, "step": 12755, "token_acc": 0.9294605809128631, "train_speed(iter/s)": 1.492628 }, { "epoch": 0.5466775202433486, "grad_norm": 3.17718243598938, "learning_rate": 9.707929296460796e-05, "loss": 0.21805624961853026, "memory(GiB)": 61.91, "step": 12760, "token_acc": 0.9472049689440993, "train_speed(iter/s)": 1.492619 }, { "epoch": 0.5468917355725976, "grad_norm": 3.171391248703003, "learning_rate": 9.707702613215344e-05, "loss": 0.654906940460205, "memory(GiB)": 61.91, "step": 12765, "token_acc": 0.8745762711864407, "train_speed(iter/s)": 1.492623 }, { "epoch": 0.5471059509018465, "grad_norm": 5.684665679931641, "learning_rate": 9.707475844685443e-05, "loss": 0.7477387905120849, "memory(GiB)": 61.91, "step": 12770, "token_acc": 0.8587896253602305, "train_speed(iter/s)": 1.492625 }, { "epoch": 0.5473201662310955, "grad_norm": 2.6138458251953125, "learning_rate": 9.7072489908752e-05, "loss": 0.4770840644836426, "memory(GiB)": 61.91, "step": 12775, "token_acc": 0.9057971014492754, "train_speed(iter/s)": 1.492594 }, { "epoch": 0.5475343815603445, "grad_norm": 1.2035634517669678, "learning_rate": 9.707022051788725e-05, "loss": 0.2779963254928589, "memory(GiB)": 61.91, "step": 12780, "token_acc": 0.936026936026936, "train_speed(iter/s)": 1.492563 }, { "epoch": 0.5477485968895934, "grad_norm": 12.346540451049805, "learning_rate": 9.706795027430133e-05, "loss": 0.7571014404296875, "memory(GiB)": 61.91, "step": 12785, "token_acc": 0.8469750889679716, "train_speed(iter/s)": 1.492616 }, { "epoch": 0.5479628122188424, "grad_norm": 3.6869888305664062, "learning_rate": 9.706567917803532e-05, "loss": 0.5568064689636231, "memory(GiB)": 61.91, "step": 12790, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.492609 }, { "epoch": 0.5481770275480914, "grad_norm": 0.44182372093200684, "learning_rate": 9.706340722913037e-05, "loss": 0.42725200653076173, "memory(GiB)": 61.91, "step": 12795, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.492605 }, { "epoch": 0.5483912428773403, "grad_norm": 4.720300197601318, "learning_rate": 9.706113442762768e-05, "loss": 0.5793606281280518, "memory(GiB)": 61.91, "step": 12800, "token_acc": 0.8729096989966555, "train_speed(iter/s)": 1.492688 }, { "epoch": 0.5486054582065892, "grad_norm": 4.7276763916015625, "learning_rate": 9.705886077356837e-05, "loss": 0.24990835189819335, "memory(GiB)": 61.91, "step": 12805, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.492721 }, { "epoch": 0.5488196735358383, "grad_norm": 0.07675153017044067, "learning_rate": 9.705658626699365e-05, "loss": 0.324847412109375, "memory(GiB)": 61.91, "step": 12810, "token_acc": 0.948339483394834, "train_speed(iter/s)": 1.492726 }, { "epoch": 0.5490338888650872, "grad_norm": 2.919053316116333, "learning_rate": 9.705431090794472e-05, "loss": 0.3659971714019775, "memory(GiB)": 61.91, "step": 12815, "token_acc": 0.9181818181818182, "train_speed(iter/s)": 1.492733 }, { "epoch": 0.5492481041943361, "grad_norm": 5.662714004516602, "learning_rate": 9.705203469646282e-05, "loss": 0.32527804374694824, "memory(GiB)": 61.91, "step": 12820, "token_acc": 0.9325396825396826, "train_speed(iter/s)": 1.492741 }, { "epoch": 0.5494623195235852, "grad_norm": 0.3888707160949707, "learning_rate": 9.704975763258919e-05, "loss": 0.4678382396697998, "memory(GiB)": 61.91, "step": 12825, "token_acc": 0.8771929824561403, "train_speed(iter/s)": 1.492749 }, { "epoch": 0.5496765348528341, "grad_norm": 3.3857319355010986, "learning_rate": 9.704747971636504e-05, "loss": 0.2564095973968506, "memory(GiB)": 61.91, "step": 12830, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.492748 }, { "epoch": 0.549890750182083, "grad_norm": 3.3380730152130127, "learning_rate": 9.704520094783167e-05, "loss": 0.4442139625549316, "memory(GiB)": 61.91, "step": 12835, "token_acc": 0.9042553191489362, "train_speed(iter/s)": 1.492721 }, { "epoch": 0.550104965511332, "grad_norm": 1.563835859298706, "learning_rate": 9.704292132703034e-05, "loss": 0.24052104949951172, "memory(GiB)": 61.91, "step": 12840, "token_acc": 0.9536423841059603, "train_speed(iter/s)": 1.492692 }, { "epoch": 0.550319180840581, "grad_norm": 3.4636781215667725, "learning_rate": 9.704064085400237e-05, "loss": 0.4224545001983643, "memory(GiB)": 61.91, "step": 12845, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.492716 }, { "epoch": 0.5505333961698299, "grad_norm": 0.023982644081115723, "learning_rate": 9.703835952878908e-05, "loss": 0.2800628185272217, "memory(GiB)": 61.91, "step": 12850, "token_acc": 0.9294871794871795, "train_speed(iter/s)": 1.492777 }, { "epoch": 0.5507476114990789, "grad_norm": 3.5176045894622803, "learning_rate": 9.703607735143176e-05, "loss": 0.4371194839477539, "memory(GiB)": 61.91, "step": 12855, "token_acc": 0.9283582089552239, "train_speed(iter/s)": 1.492761 }, { "epoch": 0.5509618268283278, "grad_norm": 2.217677354812622, "learning_rate": 9.703379432197178e-05, "loss": 0.2891937017440796, "memory(GiB)": 61.91, "step": 12860, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.492776 }, { "epoch": 0.5511760421575768, "grad_norm": 2.3405256271362305, "learning_rate": 9.70315104404505e-05, "loss": 0.5220946788787841, "memory(GiB)": 61.91, "step": 12865, "token_acc": 0.8955696202531646, "train_speed(iter/s)": 1.49271 }, { "epoch": 0.5513902574868258, "grad_norm": 3.2533698081970215, "learning_rate": 9.70292257069093e-05, "loss": 0.2991813659667969, "memory(GiB)": 61.91, "step": 12870, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.492698 }, { "epoch": 0.5516044728160747, "grad_norm": 6.588254451751709, "learning_rate": 9.702694012138953e-05, "loss": 0.3256206035614014, "memory(GiB)": 61.91, "step": 12875, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.492679 }, { "epoch": 0.5518186881453236, "grad_norm": 4.7658586502075195, "learning_rate": 9.702465368393264e-05, "loss": 0.37789602279663087, "memory(GiB)": 61.91, "step": 12880, "token_acc": 0.9139072847682119, "train_speed(iter/s)": 1.492641 }, { "epoch": 0.5520329034745727, "grad_norm": 1.9054632186889648, "learning_rate": 9.702236639458003e-05, "loss": 0.4326024055480957, "memory(GiB)": 61.91, "step": 12885, "token_acc": 0.8997134670487106, "train_speed(iter/s)": 1.492597 }, { "epoch": 0.5522471188038216, "grad_norm": 3.326127767562866, "learning_rate": 9.702007825337316e-05, "loss": 0.4937246322631836, "memory(GiB)": 61.91, "step": 12890, "token_acc": 0.9015748031496063, "train_speed(iter/s)": 1.492574 }, { "epoch": 0.5524613341330705, "grad_norm": 3.0150022506713867, "learning_rate": 9.701778926035344e-05, "loss": 0.3144610643386841, "memory(GiB)": 61.91, "step": 12895, "token_acc": 0.936, "train_speed(iter/s)": 1.492581 }, { "epoch": 0.5526755494623196, "grad_norm": 4.2696990966796875, "learning_rate": 9.701549941556238e-05, "loss": 0.18773559331893921, "memory(GiB)": 61.91, "step": 12900, "token_acc": 0.9533333333333334, "train_speed(iter/s)": 1.49256 }, { "epoch": 0.5528897647915685, "grad_norm": 2.6443424224853516, "learning_rate": 9.701320871904143e-05, "loss": 0.5935259819030761, "memory(GiB)": 61.91, "step": 12905, "token_acc": 0.8741935483870967, "train_speed(iter/s)": 1.492556 }, { "epoch": 0.5531039801208174, "grad_norm": 3.420987844467163, "learning_rate": 9.70109171708321e-05, "loss": 0.3367485046386719, "memory(GiB)": 61.91, "step": 12910, "token_acc": 0.912751677852349, "train_speed(iter/s)": 1.492555 }, { "epoch": 0.5533181954500664, "grad_norm": 1.883061170578003, "learning_rate": 9.700862477097592e-05, "loss": 0.3612550735473633, "memory(GiB)": 61.91, "step": 12915, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.492561 }, { "epoch": 0.5535324107793154, "grad_norm": 3.909189224243164, "learning_rate": 9.700633151951441e-05, "loss": 0.44535222053527834, "memory(GiB)": 61.91, "step": 12920, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.492551 }, { "epoch": 0.5537466261085643, "grad_norm": 5.765448570251465, "learning_rate": 9.700403741648909e-05, "loss": 0.5226988315582275, "memory(GiB)": 61.91, "step": 12925, "token_acc": 0.88671875, "train_speed(iter/s)": 1.492544 }, { "epoch": 0.5539608414378133, "grad_norm": 2.9847395420074463, "learning_rate": 9.700174246194155e-05, "loss": 0.2232513666152954, "memory(GiB)": 61.91, "step": 12930, "token_acc": 0.9440559440559441, "train_speed(iter/s)": 1.492532 }, { "epoch": 0.5541750567670622, "grad_norm": 3.396556854248047, "learning_rate": 9.699944665591336e-05, "loss": 0.4742887496948242, "memory(GiB)": 61.91, "step": 12935, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.49253 }, { "epoch": 0.5543892720963112, "grad_norm": 0.6087324023246765, "learning_rate": 9.699714999844608e-05, "loss": 0.3564779758453369, "memory(GiB)": 61.91, "step": 12940, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.492588 }, { "epoch": 0.5546034874255602, "grad_norm": 5.657891273498535, "learning_rate": 9.699485248958137e-05, "loss": 0.4252163887023926, "memory(GiB)": 61.91, "step": 12945, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.492738 }, { "epoch": 0.5548177027548091, "grad_norm": 5.706553936004639, "learning_rate": 9.69925541293608e-05, "loss": 0.6801621437072753, "memory(GiB)": 61.91, "step": 12950, "token_acc": 0.8591549295774648, "train_speed(iter/s)": 1.492751 }, { "epoch": 0.555031918084058, "grad_norm": 5.870884418487549, "learning_rate": 9.699025491782606e-05, "loss": 0.5833921432495117, "memory(GiB)": 61.91, "step": 12955, "token_acc": 0.8740157480314961, "train_speed(iter/s)": 1.4928 }, { "epoch": 0.5552461334133071, "grad_norm": 1.3007453680038452, "learning_rate": 9.698795485501873e-05, "loss": 0.6328574180603027, "memory(GiB)": 61.91, "step": 12960, "token_acc": 0.8735177865612648, "train_speed(iter/s)": 1.492925 }, { "epoch": 0.555460348742556, "grad_norm": 3.717494010925293, "learning_rate": 9.698565394098054e-05, "loss": 0.21348028182983397, "memory(GiB)": 61.91, "step": 12965, "token_acc": 0.96, "train_speed(iter/s)": 1.493021 }, { "epoch": 0.5556745640718049, "grad_norm": 3.496403694152832, "learning_rate": 9.698335217575316e-05, "loss": 0.38600845336914064, "memory(GiB)": 61.91, "step": 12970, "token_acc": 0.9335443037974683, "train_speed(iter/s)": 1.492953 }, { "epoch": 0.555888779401054, "grad_norm": 1.6071022748947144, "learning_rate": 9.698104955937827e-05, "loss": 0.3222589254379272, "memory(GiB)": 61.91, "step": 12975, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.492967 }, { "epoch": 0.5561029947303029, "grad_norm": 1.717017650604248, "learning_rate": 9.697874609189759e-05, "loss": 0.509594202041626, "memory(GiB)": 61.91, "step": 12980, "token_acc": 0.8975265017667845, "train_speed(iter/s)": 1.493024 }, { "epoch": 0.5563172100595518, "grad_norm": 3.724766254425049, "learning_rate": 9.697644177335287e-05, "loss": 0.2775221586227417, "memory(GiB)": 61.91, "step": 12985, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.493075 }, { "epoch": 0.5565314253888008, "grad_norm": 0.3296225965023041, "learning_rate": 9.697413660378584e-05, "loss": 0.23901448249816895, "memory(GiB)": 61.91, "step": 12990, "token_acc": 0.9503311258278145, "train_speed(iter/s)": 1.493091 }, { "epoch": 0.5567456407180498, "grad_norm": 2.913480281829834, "learning_rate": 9.697183058323826e-05, "loss": 0.2073270559310913, "memory(GiB)": 61.91, "step": 12995, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.493101 }, { "epoch": 0.5569598560472987, "grad_norm": 0.422588974237442, "learning_rate": 9.69695237117519e-05, "loss": 0.38780086040496825, "memory(GiB)": 61.91, "step": 13000, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.493093 }, { "epoch": 0.5569598560472987, "eval_loss": 2.645993709564209, "eval_runtime": 13.6838, "eval_samples_per_second": 7.308, "eval_steps_per_second": 7.308, "eval_token_acc": 0.43041237113402064, "step": 13000 }, { "epoch": 0.5571740713765477, "grad_norm": 2.112321615219116, "learning_rate": 9.696721598936856e-05, "loss": 0.36470842361450195, "memory(GiB)": 61.91, "step": 13005, "token_acc": 0.5755064456721916, "train_speed(iter/s)": 1.490601 }, { "epoch": 0.5573882867057967, "grad_norm": 2.0305707454681396, "learning_rate": 9.696490741613002e-05, "loss": 0.5130202770233154, "memory(GiB)": 61.91, "step": 13010, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.490583 }, { "epoch": 0.5576025020350456, "grad_norm": 3.497251272201538, "learning_rate": 9.696259799207816e-05, "loss": 0.18693649768829346, "memory(GiB)": 61.91, "step": 13015, "token_acc": 0.9575163398692811, "train_speed(iter/s)": 1.49058 }, { "epoch": 0.5578167173642946, "grad_norm": 0.5946825742721558, "learning_rate": 9.696028771725476e-05, "loss": 0.3408238649368286, "memory(GiB)": 61.91, "step": 13020, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.490544 }, { "epoch": 0.5580309326935435, "grad_norm": 4.600757122039795, "learning_rate": 9.695797659170171e-05, "loss": 0.4094806671142578, "memory(GiB)": 61.91, "step": 13025, "token_acc": 0.9159420289855073, "train_speed(iter/s)": 1.490529 }, { "epoch": 0.5582451480227925, "grad_norm": 7.566234588623047, "learning_rate": 9.695566461546086e-05, "loss": 0.6307057857513427, "memory(GiB)": 61.91, "step": 13030, "token_acc": 0.8659420289855072, "train_speed(iter/s)": 1.490547 }, { "epoch": 0.5584593633520415, "grad_norm": 6.146847248077393, "learning_rate": 9.695335178857409e-05, "loss": 0.48227667808532715, "memory(GiB)": 61.91, "step": 13035, "token_acc": 0.9044368600682594, "train_speed(iter/s)": 1.490503 }, { "epoch": 0.5586735786812904, "grad_norm": 4.2266669273376465, "learning_rate": 9.695103811108333e-05, "loss": 0.33376007080078124, "memory(GiB)": 61.91, "step": 13040, "token_acc": 0.9372384937238494, "train_speed(iter/s)": 1.490477 }, { "epoch": 0.5588877940105393, "grad_norm": 1.8757895231246948, "learning_rate": 9.694872358303044e-05, "loss": 0.31785335540771487, "memory(GiB)": 61.91, "step": 13045, "token_acc": 0.9510204081632653, "train_speed(iter/s)": 1.490488 }, { "epoch": 0.5591020093397884, "grad_norm": 2.5348691940307617, "learning_rate": 9.69464082044574e-05, "loss": 0.5548069477081299, "memory(GiB)": 61.91, "step": 13050, "token_acc": 0.8990228013029316, "train_speed(iter/s)": 1.490465 }, { "epoch": 0.5593162246690373, "grad_norm": 2.513394832611084, "learning_rate": 9.694409197540613e-05, "loss": 0.44098672866821287, "memory(GiB)": 61.91, "step": 13055, "token_acc": 0.9022801302931596, "train_speed(iter/s)": 1.490457 }, { "epoch": 0.5595304399982862, "grad_norm": 3.208719253540039, "learning_rate": 9.69417748959186e-05, "loss": 0.4759847640991211, "memory(GiB)": 61.91, "step": 13060, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.490452 }, { "epoch": 0.5597446553275353, "grad_norm": 2.622288942337036, "learning_rate": 9.693945696603678e-05, "loss": 0.569830846786499, "memory(GiB)": 61.91, "step": 13065, "token_acc": 0.8819188191881919, "train_speed(iter/s)": 1.490462 }, { "epoch": 0.5599588706567842, "grad_norm": 5.744207859039307, "learning_rate": 9.693713818580265e-05, "loss": 0.5023719310760498, "memory(GiB)": 61.91, "step": 13070, "token_acc": 0.8924050632911392, "train_speed(iter/s)": 1.490446 }, { "epoch": 0.5601730859860332, "grad_norm": 1.7545088529586792, "learning_rate": 9.693481855525826e-05, "loss": 0.3205952405929565, "memory(GiB)": 61.91, "step": 13075, "token_acc": 0.9246031746031746, "train_speed(iter/s)": 1.490411 }, { "epoch": 0.5603873013152821, "grad_norm": 7.620838165283203, "learning_rate": 9.693249807444559e-05, "loss": 0.6212154865264893, "memory(GiB)": 61.91, "step": 13080, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.49037 }, { "epoch": 0.5606015166445311, "grad_norm": 9.34473705291748, "learning_rate": 9.693017674340669e-05, "loss": 0.3625880241394043, "memory(GiB)": 61.91, "step": 13085, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.490352 }, { "epoch": 0.5608157319737801, "grad_norm": 0.3704056143760681, "learning_rate": 9.69278545621836e-05, "loss": 0.28856773376464845, "memory(GiB)": 61.91, "step": 13090, "token_acc": 0.9509433962264151, "train_speed(iter/s)": 1.490514 }, { "epoch": 0.561029947303029, "grad_norm": 3.4762837886810303, "learning_rate": 9.692553153081842e-05, "loss": 0.28941960334777833, "memory(GiB)": 61.91, "step": 13095, "token_acc": 0.9363957597173145, "train_speed(iter/s)": 1.490494 }, { "epoch": 0.5612441626322779, "grad_norm": 2.495220422744751, "learning_rate": 9.692320764935322e-05, "loss": 0.5314304828643799, "memory(GiB)": 61.91, "step": 13100, "token_acc": 0.8868613138686131, "train_speed(iter/s)": 1.490471 }, { "epoch": 0.561458377961527, "grad_norm": 5.02890157699585, "learning_rate": 9.692088291783009e-05, "loss": 0.49011731147766113, "memory(GiB)": 61.91, "step": 13105, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.490523 }, { "epoch": 0.5616725932907759, "grad_norm": 3.415149211883545, "learning_rate": 9.691855733629115e-05, "loss": 0.3201854944229126, "memory(GiB)": 61.91, "step": 13110, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.490481 }, { "epoch": 0.5618868086200248, "grad_norm": 0.9847811460494995, "learning_rate": 9.691623090477852e-05, "loss": 0.21547043323516846, "memory(GiB)": 61.91, "step": 13115, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.490513 }, { "epoch": 0.5621010239492739, "grad_norm": 5.6877851486206055, "learning_rate": 9.691390362333437e-05, "loss": 0.758168363571167, "memory(GiB)": 61.91, "step": 13120, "token_acc": 0.8380681818181818, "train_speed(iter/s)": 1.490537 }, { "epoch": 0.5623152392785228, "grad_norm": 4.985800743103027, "learning_rate": 9.691157549200085e-05, "loss": 0.3220491886138916, "memory(GiB)": 61.91, "step": 13125, "token_acc": 0.9358974358974359, "train_speed(iter/s)": 1.490552 }, { "epoch": 0.5625294546077717, "grad_norm": 5.033344745635986, "learning_rate": 9.690924651082014e-05, "loss": 0.5973250389099121, "memory(GiB)": 61.91, "step": 13130, "token_acc": 0.8607142857142858, "train_speed(iter/s)": 1.490495 }, { "epoch": 0.5627436699370207, "grad_norm": 4.5838093757629395, "learning_rate": 9.69069166798344e-05, "loss": 0.4568964958190918, "memory(GiB)": 61.91, "step": 13135, "token_acc": 0.8962264150943396, "train_speed(iter/s)": 1.490498 }, { "epoch": 0.5629578852662697, "grad_norm": 2.2657856941223145, "learning_rate": 9.690458599908588e-05, "loss": 0.6534167289733886, "memory(GiB)": 61.91, "step": 13140, "token_acc": 0.8657142857142858, "train_speed(iter/s)": 1.49047 }, { "epoch": 0.5631721005955186, "grad_norm": 4.3585381507873535, "learning_rate": 9.690225446861678e-05, "loss": 0.3150540113449097, "memory(GiB)": 61.91, "step": 13145, "token_acc": 0.9407407407407408, "train_speed(iter/s)": 1.490463 }, { "epoch": 0.5633863159247676, "grad_norm": 3.0564651489257812, "learning_rate": 9.689992208846934e-05, "loss": 0.42705330848693845, "memory(GiB)": 61.91, "step": 13150, "token_acc": 0.910958904109589, "train_speed(iter/s)": 1.490447 }, { "epoch": 0.5636005312540165, "grad_norm": 0.7883842587471008, "learning_rate": 9.689758885868582e-05, "loss": 0.44359073638916013, "memory(GiB)": 61.91, "step": 13155, "token_acc": 0.8807017543859649, "train_speed(iter/s)": 1.490457 }, { "epoch": 0.5638147465832655, "grad_norm": 4.302028179168701, "learning_rate": 9.689525477930848e-05, "loss": 0.24403347969055175, "memory(GiB)": 61.91, "step": 13160, "token_acc": 0.9506578947368421, "train_speed(iter/s)": 1.490454 }, { "epoch": 0.5640289619125145, "grad_norm": 2.2359607219696045, "learning_rate": 9.68929198503796e-05, "loss": 0.3351123332977295, "memory(GiB)": 61.91, "step": 13165, "token_acc": 0.9291338582677166, "train_speed(iter/s)": 1.490452 }, { "epoch": 0.5642431772417634, "grad_norm": 2.803471565246582, "learning_rate": 9.689058407194152e-05, "loss": 0.5444179058074952, "memory(GiB)": 61.91, "step": 13170, "token_acc": 0.8915254237288136, "train_speed(iter/s)": 1.490434 }, { "epoch": 0.5644573925710124, "grad_norm": 5.04897403717041, "learning_rate": 9.688824744403648e-05, "loss": 0.42846102714538575, "memory(GiB)": 61.91, "step": 13175, "token_acc": 0.9134615384615384, "train_speed(iter/s)": 1.49047 }, { "epoch": 0.5646716079002614, "grad_norm": 1.9409801959991455, "learning_rate": 9.688590996670688e-05, "loss": 0.40284295082092286, "memory(GiB)": 61.91, "step": 13180, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.490467 }, { "epoch": 0.5648858232295103, "grad_norm": 2.3601760864257812, "learning_rate": 9.688357163999504e-05, "loss": 0.34731032848358157, "memory(GiB)": 61.91, "step": 13185, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.490546 }, { "epoch": 0.5651000385587592, "grad_norm": 0.8404945135116577, "learning_rate": 9.68812324639433e-05, "loss": 0.17675092220306396, "memory(GiB)": 61.91, "step": 13190, "token_acc": 0.9693486590038314, "train_speed(iter/s)": 1.490555 }, { "epoch": 0.5653142538880083, "grad_norm": 3.971151113510132, "learning_rate": 9.687889243859407e-05, "loss": 0.42297706604003904, "memory(GiB)": 61.91, "step": 13195, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.490528 }, { "epoch": 0.5655284692172572, "grad_norm": 2.9290502071380615, "learning_rate": 9.687655156398972e-05, "loss": 0.38873751163482667, "memory(GiB)": 61.91, "step": 13200, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.490505 }, { "epoch": 0.5657426845465061, "grad_norm": 2.739459276199341, "learning_rate": 9.687420984017268e-05, "loss": 0.4084742069244385, "memory(GiB)": 61.91, "step": 13205, "token_acc": 0.9076433121019108, "train_speed(iter/s)": 1.490491 }, { "epoch": 0.5659568998757551, "grad_norm": 3.737039089202881, "learning_rate": 9.687186726718533e-05, "loss": 0.38274483680725097, "memory(GiB)": 61.91, "step": 13210, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.490518 }, { "epoch": 0.5661711152050041, "grad_norm": 2.3709254264831543, "learning_rate": 9.686952384507017e-05, "loss": 0.5517672538757324, "memory(GiB)": 61.91, "step": 13215, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.490499 }, { "epoch": 0.566385330534253, "grad_norm": 3.0234968662261963, "learning_rate": 9.686717957386959e-05, "loss": 0.43480143547058103, "memory(GiB)": 61.91, "step": 13220, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.490523 }, { "epoch": 0.566599545863502, "grad_norm": 4.56168794631958, "learning_rate": 9.686483445362611e-05, "loss": 0.34618487358093264, "memory(GiB)": 61.91, "step": 13225, "token_acc": 0.9120879120879121, "train_speed(iter/s)": 1.490489 }, { "epoch": 0.566813761192751, "grad_norm": 1.5643035173416138, "learning_rate": 9.686248848438217e-05, "loss": 0.24843056201934816, "memory(GiB)": 61.91, "step": 13230, "token_acc": 0.9361022364217252, "train_speed(iter/s)": 1.490494 }, { "epoch": 0.5670279765219999, "grad_norm": 5.172840118408203, "learning_rate": 9.68601416661803e-05, "loss": 0.6226789474487304, "memory(GiB)": 61.91, "step": 13235, "token_acc": 0.8734693877551021, "train_speed(iter/s)": 1.490466 }, { "epoch": 0.5672421918512489, "grad_norm": 1.5658818483352661, "learning_rate": 9.6857793999063e-05, "loss": 0.4624669075012207, "memory(GiB)": 61.91, "step": 13240, "token_acc": 0.8993506493506493, "train_speed(iter/s)": 1.490443 }, { "epoch": 0.5674564071804978, "grad_norm": 2.891721248626709, "learning_rate": 9.685544548307281e-05, "loss": 0.594964075088501, "memory(GiB)": 61.91, "step": 13245, "token_acc": 0.8926174496644296, "train_speed(iter/s)": 1.490411 }, { "epoch": 0.5676706225097468, "grad_norm": 1.5209288597106934, "learning_rate": 9.685309611825226e-05, "loss": 0.40756726264953613, "memory(GiB)": 61.91, "step": 13250, "token_acc": 0.9118329466357309, "train_speed(iter/s)": 1.490417 }, { "epoch": 0.5678848378389958, "grad_norm": 7.023831367492676, "learning_rate": 9.685074590464394e-05, "loss": 0.11184229850769042, "memory(GiB)": 61.91, "step": 13255, "token_acc": 0.9787234042553191, "train_speed(iter/s)": 1.490381 }, { "epoch": 0.5680990531682447, "grad_norm": 7.077420234680176, "learning_rate": 9.68483948422904e-05, "loss": 0.14034852981567383, "memory(GiB)": 61.91, "step": 13260, "token_acc": 0.961864406779661, "train_speed(iter/s)": 1.490368 }, { "epoch": 0.5683132684974936, "grad_norm": 3.3042736053466797, "learning_rate": 9.684604293123425e-05, "loss": 0.4560530185699463, "memory(GiB)": 61.91, "step": 13265, "token_acc": 0.9012738853503185, "train_speed(iter/s)": 1.490337 }, { "epoch": 0.5685274838267427, "grad_norm": 3.953301191329956, "learning_rate": 9.684369017151806e-05, "loss": 0.38683319091796875, "memory(GiB)": 61.91, "step": 13270, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.490381 }, { "epoch": 0.5687416991559916, "grad_norm": 2.0727710723876953, "learning_rate": 9.68413365631845e-05, "loss": 0.45212907791137696, "memory(GiB)": 61.91, "step": 13275, "token_acc": 0.8957654723127035, "train_speed(iter/s)": 1.490361 }, { "epoch": 0.5689559144852405, "grad_norm": 2.6606338024139404, "learning_rate": 9.68389821062762e-05, "loss": 0.39273929595947266, "memory(GiB)": 61.91, "step": 13280, "token_acc": 0.9224137931034483, "train_speed(iter/s)": 1.490352 }, { "epoch": 0.5691701298144896, "grad_norm": 2.7290172576904297, "learning_rate": 9.683662680083578e-05, "loss": 0.38120744228363035, "memory(GiB)": 61.91, "step": 13285, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.490364 }, { "epoch": 0.5693843451437385, "grad_norm": 3.3262312412261963, "learning_rate": 9.683427064690593e-05, "loss": 0.3335121631622314, "memory(GiB)": 61.91, "step": 13290, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.490364 }, { "epoch": 0.5695985604729874, "grad_norm": 4.114030838012695, "learning_rate": 9.683191364452934e-05, "loss": 0.6350560188293457, "memory(GiB)": 61.91, "step": 13295, "token_acc": 0.8798701298701299, "train_speed(iter/s)": 1.490379 }, { "epoch": 0.5698127758022364, "grad_norm": 1.5411256551742554, "learning_rate": 9.68295557937487e-05, "loss": 0.6142405986785888, "memory(GiB)": 61.91, "step": 13300, "token_acc": 0.8785046728971962, "train_speed(iter/s)": 1.490315 }, { "epoch": 0.5700269911314854, "grad_norm": 6.786955833435059, "learning_rate": 9.682719709460672e-05, "loss": 0.5116767883300781, "memory(GiB)": 61.91, "step": 13305, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.490439 }, { "epoch": 0.5702412064607343, "grad_norm": 2.348670482635498, "learning_rate": 9.682483754714615e-05, "loss": 0.6818351745605469, "memory(GiB)": 61.91, "step": 13310, "token_acc": 0.8688524590163934, "train_speed(iter/s)": 1.490489 }, { "epoch": 0.5704554217899833, "grad_norm": 1.6320552825927734, "learning_rate": 9.682247715140974e-05, "loss": 0.39583237171173097, "memory(GiB)": 61.91, "step": 13315, "token_acc": 0.9388379204892966, "train_speed(iter/s)": 1.490498 }, { "epoch": 0.5706696371192322, "grad_norm": 3.3893632888793945, "learning_rate": 9.68201159074402e-05, "loss": 0.6101117134094238, "memory(GiB)": 61.91, "step": 13320, "token_acc": 0.884375, "train_speed(iter/s)": 1.490461 }, { "epoch": 0.5708838524484812, "grad_norm": 3.37837553024292, "learning_rate": 9.681775381528034e-05, "loss": 0.4471477508544922, "memory(GiB)": 61.91, "step": 13325, "token_acc": 0.8970588235294118, "train_speed(iter/s)": 1.490502 }, { "epoch": 0.5710980677777302, "grad_norm": 1.4890791177749634, "learning_rate": 9.681539087497296e-05, "loss": 0.5617738723754883, "memory(GiB)": 61.91, "step": 13330, "token_acc": 0.8712871287128713, "train_speed(iter/s)": 1.490552 }, { "epoch": 0.5713122831069791, "grad_norm": 1.7019623517990112, "learning_rate": 9.681302708656086e-05, "loss": 0.4303577423095703, "memory(GiB)": 61.91, "step": 13335, "token_acc": 0.8990228013029316, "train_speed(iter/s)": 1.490566 }, { "epoch": 0.571526498436228, "grad_norm": 2.995971441268921, "learning_rate": 9.681066245008687e-05, "loss": 0.43776531219482423, "memory(GiB)": 61.91, "step": 13340, "token_acc": 0.903010033444816, "train_speed(iter/s)": 1.490553 }, { "epoch": 0.5717407137654771, "grad_norm": 2.650177001953125, "learning_rate": 9.68082969655938e-05, "loss": 0.3034083366394043, "memory(GiB)": 61.91, "step": 13345, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.490562 }, { "epoch": 0.571954929094726, "grad_norm": 5.425454139709473, "learning_rate": 9.680593063312454e-05, "loss": 0.5938477039337158, "memory(GiB)": 61.91, "step": 13350, "token_acc": 0.8868613138686131, "train_speed(iter/s)": 1.490533 }, { "epoch": 0.5721691444239749, "grad_norm": 3.6616811752319336, "learning_rate": 9.680356345272193e-05, "loss": 0.23226375579833985, "memory(GiB)": 61.91, "step": 13355, "token_acc": 0.9456869009584664, "train_speed(iter/s)": 1.490593 }, { "epoch": 0.572383359753224, "grad_norm": 1.9663934707641602, "learning_rate": 9.680119542442885e-05, "loss": 0.3447110652923584, "memory(GiB)": 61.91, "step": 13360, "token_acc": 0.9205479452054794, "train_speed(iter/s)": 1.490623 }, { "epoch": 0.5725975750824729, "grad_norm": 2.0027260780334473, "learning_rate": 9.679882654828822e-05, "loss": 0.3962819576263428, "memory(GiB)": 61.91, "step": 13365, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.490642 }, { "epoch": 0.5728117904117218, "grad_norm": 0.6651832461357117, "learning_rate": 9.679645682434295e-05, "loss": 0.32296640872955323, "memory(GiB)": 61.91, "step": 13370, "token_acc": 0.9246031746031746, "train_speed(iter/s)": 1.490639 }, { "epoch": 0.5730260057409708, "grad_norm": 7.477115631103516, "learning_rate": 9.679408625263596e-05, "loss": 0.1644531011581421, "memory(GiB)": 61.91, "step": 13375, "token_acc": 0.9591836734693877, "train_speed(iter/s)": 1.490664 }, { "epoch": 0.5732402210702198, "grad_norm": 2.821225881576538, "learning_rate": 9.679171483321022e-05, "loss": 0.6215180873870849, "memory(GiB)": 61.91, "step": 13380, "token_acc": 0.8739495798319328, "train_speed(iter/s)": 1.490723 }, { "epoch": 0.5734544363994687, "grad_norm": 0.33988627791404724, "learning_rate": 9.678934256610864e-05, "loss": 0.2508926630020142, "memory(GiB)": 61.91, "step": 13385, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.490744 }, { "epoch": 0.5736686517287177, "grad_norm": 0.10258810222148895, "learning_rate": 9.678696945137424e-05, "loss": 0.29040524959564207, "memory(GiB)": 61.91, "step": 13390, "token_acc": 0.9303135888501742, "train_speed(iter/s)": 1.490754 }, { "epoch": 0.5738828670579667, "grad_norm": 4.647797584533691, "learning_rate": 9.678459548905001e-05, "loss": 0.38529324531555176, "memory(GiB)": 61.91, "step": 13395, "token_acc": 0.9309309309309309, "train_speed(iter/s)": 1.490731 }, { "epoch": 0.5740970823872156, "grad_norm": 4.82197380065918, "learning_rate": 9.678222067917893e-05, "loss": 0.3822281837463379, "memory(GiB)": 61.91, "step": 13400, "token_acc": 0.9071428571428571, "train_speed(iter/s)": 1.49071 }, { "epoch": 0.5743112977164646, "grad_norm": 2.553710699081421, "learning_rate": 9.677984502180405e-05, "loss": 0.41651020050048826, "memory(GiB)": 61.91, "step": 13405, "token_acc": 0.8896103896103896, "train_speed(iter/s)": 1.49068 }, { "epoch": 0.5745255130457135, "grad_norm": 10.510475158691406, "learning_rate": 9.677746851696838e-05, "loss": 0.1603337049484253, "memory(GiB)": 61.91, "step": 13410, "token_acc": 0.9703947368421053, "train_speed(iter/s)": 1.490658 }, { "epoch": 0.5747397283749626, "grad_norm": 1.790899634361267, "learning_rate": 9.6775091164715e-05, "loss": 0.37792010307312013, "memory(GiB)": 61.91, "step": 13415, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.490657 }, { "epoch": 0.5749539437042115, "grad_norm": 4.038863658905029, "learning_rate": 9.677271296508697e-05, "loss": 0.548935842514038, "memory(GiB)": 61.91, "step": 13420, "token_acc": 0.8785942492012779, "train_speed(iter/s)": 1.490652 }, { "epoch": 0.5751681590334604, "grad_norm": 1.4322264194488525, "learning_rate": 9.677033391812736e-05, "loss": 0.21003298759460448, "memory(GiB)": 61.91, "step": 13425, "token_acc": 0.9482071713147411, "train_speed(iter/s)": 1.490657 }, { "epoch": 0.5753823743627094, "grad_norm": 7.586784839630127, "learning_rate": 9.676795402387927e-05, "loss": 0.38659253120422366, "memory(GiB)": 61.91, "step": 13430, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.490645 }, { "epoch": 0.5755965896919584, "grad_norm": 1.9697051048278809, "learning_rate": 9.676557328238582e-05, "loss": 0.2939443826675415, "memory(GiB)": 61.91, "step": 13435, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.490708 }, { "epoch": 0.5758108050212073, "grad_norm": 2.1904191970825195, "learning_rate": 9.676319169369016e-05, "loss": 0.3012415409088135, "memory(GiB)": 61.91, "step": 13440, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.490659 }, { "epoch": 0.5760250203504563, "grad_norm": 2.9046692848205566, "learning_rate": 9.67608092578354e-05, "loss": 0.4096056938171387, "memory(GiB)": 61.91, "step": 13445, "token_acc": 0.9225806451612903, "train_speed(iter/s)": 1.490638 }, { "epoch": 0.5762392356797053, "grad_norm": 4.578183174133301, "learning_rate": 9.675842597486471e-05, "loss": 0.45118465423583987, "memory(GiB)": 61.91, "step": 13450, "token_acc": 0.8934169278996865, "train_speed(iter/s)": 1.490643 }, { "epoch": 0.5764534510089542, "grad_norm": 3.1019625663757324, "learning_rate": 9.675604184482129e-05, "loss": 0.46416077613830564, "memory(GiB)": 61.91, "step": 13455, "token_acc": 0.9061224489795918, "train_speed(iter/s)": 1.490714 }, { "epoch": 0.5766676663382032, "grad_norm": 2.354992628097534, "learning_rate": 9.67536568677483e-05, "loss": 0.33783793449401855, "memory(GiB)": 61.91, "step": 13460, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.490717 }, { "epoch": 0.5768818816674521, "grad_norm": 0.8575239777565002, "learning_rate": 9.675127104368895e-05, "loss": 0.34040853977203367, "memory(GiB)": 61.91, "step": 13465, "token_acc": 0.9221183800623053, "train_speed(iter/s)": 1.490734 }, { "epoch": 0.5770960969967011, "grad_norm": 3.928168773651123, "learning_rate": 9.674888437268648e-05, "loss": 0.60721435546875, "memory(GiB)": 61.91, "step": 13470, "token_acc": 0.86328125, "train_speed(iter/s)": 1.490727 }, { "epoch": 0.5773103123259501, "grad_norm": 1.5958478450775146, "learning_rate": 9.674649685478412e-05, "loss": 0.16885098218917846, "memory(GiB)": 61.91, "step": 13475, "token_acc": 0.9639344262295082, "train_speed(iter/s)": 1.49071 }, { "epoch": 0.577524527655199, "grad_norm": 6.546055316925049, "learning_rate": 9.67441084900251e-05, "loss": 0.45613937377929686, "memory(GiB)": 61.91, "step": 13480, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.490693 }, { "epoch": 0.5777387429844479, "grad_norm": 1.8657541275024414, "learning_rate": 9.674171927845272e-05, "loss": 0.19820718765258788, "memory(GiB)": 61.91, "step": 13485, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.490686 }, { "epoch": 0.577952958313697, "grad_norm": 4.075368881225586, "learning_rate": 9.673932922011024e-05, "loss": 0.43367853164672854, "memory(GiB)": 61.91, "step": 13490, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.490626 }, { "epoch": 0.5781671736429459, "grad_norm": 3.092252254486084, "learning_rate": 9.673693831504096e-05, "loss": 0.31474554538726807, "memory(GiB)": 61.91, "step": 13495, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.490664 }, { "epoch": 0.5783813889721948, "grad_norm": 4.00586462020874, "learning_rate": 9.673454656328823e-05, "loss": 0.6068418502807618, "memory(GiB)": 61.91, "step": 13500, "token_acc": 0.8876811594202898, "train_speed(iter/s)": 1.490689 }, { "epoch": 0.5783813889721948, "eval_loss": 2.6489007472991943, "eval_runtime": 12.8789, "eval_samples_per_second": 7.765, "eval_steps_per_second": 7.765, "eval_token_acc": 0.4077025232403719, "step": 13500 }, { "epoch": 0.5785956043014439, "grad_norm": 2.8116018772125244, "learning_rate": 9.673215396489531e-05, "loss": 0.3607691764831543, "memory(GiB)": 61.91, "step": 13505, "token_acc": 0.5429122468659595, "train_speed(iter/s)": 1.488369 }, { "epoch": 0.5788098196306928, "grad_norm": 2.408745288848877, "learning_rate": 9.67297605199056e-05, "loss": 0.2182774543762207, "memory(GiB)": 61.91, "step": 13510, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.488356 }, { "epoch": 0.5790240349599417, "grad_norm": 12.628022193908691, "learning_rate": 9.672736622836245e-05, "loss": 0.6455174922943115, "memory(GiB)": 61.91, "step": 13515, "token_acc": 0.8611111111111112, "train_speed(iter/s)": 1.488354 }, { "epoch": 0.5792382502891907, "grad_norm": 2.722378969192505, "learning_rate": 9.672497109030922e-05, "loss": 0.41347346305847166, "memory(GiB)": 61.91, "step": 13520, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.488377 }, { "epoch": 0.5794524656184397, "grad_norm": 5.904735088348389, "learning_rate": 9.67225751057893e-05, "loss": 0.7364228248596192, "memory(GiB)": 61.91, "step": 13525, "token_acc": 0.8362068965517241, "train_speed(iter/s)": 1.488537 }, { "epoch": 0.5796666809476886, "grad_norm": 3.233441114425659, "learning_rate": 9.672017827484611e-05, "loss": 0.26290345191955566, "memory(GiB)": 61.91, "step": 13530, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.488563 }, { "epoch": 0.5798808962769376, "grad_norm": 3.0749428272247314, "learning_rate": 9.671778059752305e-05, "loss": 0.5654623985290528, "memory(GiB)": 61.91, "step": 13535, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.488556 }, { "epoch": 0.5800951116061865, "grad_norm": 6.65649938583374, "learning_rate": 9.67153820738636e-05, "loss": 0.3867976188659668, "memory(GiB)": 61.91, "step": 13540, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.488502 }, { "epoch": 0.5803093269354355, "grad_norm": 2.6117939949035645, "learning_rate": 9.671298270391114e-05, "loss": 0.2474213123321533, "memory(GiB)": 61.91, "step": 13545, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.488475 }, { "epoch": 0.5805235422646845, "grad_norm": 2.9311411380767822, "learning_rate": 9.671058248770922e-05, "loss": 0.40988454818725584, "memory(GiB)": 61.91, "step": 13550, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.488438 }, { "epoch": 0.5807377575939334, "grad_norm": 1.5507254600524902, "learning_rate": 9.670818142530125e-05, "loss": 0.4317680835723877, "memory(GiB)": 61.91, "step": 13555, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.488409 }, { "epoch": 0.5809519729231823, "grad_norm": 2.513195514678955, "learning_rate": 9.670577951673076e-05, "loss": 0.3436610460281372, "memory(GiB)": 61.91, "step": 13560, "token_acc": 0.9221556886227545, "train_speed(iter/s)": 1.488392 }, { "epoch": 0.5811661882524314, "grad_norm": 0.5031458139419556, "learning_rate": 9.670337676204127e-05, "loss": 0.377254581451416, "memory(GiB)": 61.91, "step": 13565, "token_acc": 0.8969072164948454, "train_speed(iter/s)": 1.488371 }, { "epoch": 0.5813804035816803, "grad_norm": 4.493702411651611, "learning_rate": 9.670097316127631e-05, "loss": 0.44896087646484373, "memory(GiB)": 61.91, "step": 13570, "token_acc": 0.8990228013029316, "train_speed(iter/s)": 1.4884 }, { "epoch": 0.5815946189109292, "grad_norm": 3.885554075241089, "learning_rate": 9.66985687144794e-05, "loss": 0.6597471237182617, "memory(GiB)": 61.91, "step": 13575, "token_acc": 0.8736059479553904, "train_speed(iter/s)": 1.488394 }, { "epoch": 0.5818088342401783, "grad_norm": 4.559040546417236, "learning_rate": 9.66961634216941e-05, "loss": 0.2627657413482666, "memory(GiB)": 61.91, "step": 13580, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.488431 }, { "epoch": 0.5820230495694272, "grad_norm": 4.202556133270264, "learning_rate": 9.669375728296402e-05, "loss": 0.22635350227355958, "memory(GiB)": 61.91, "step": 13585, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.488372 }, { "epoch": 0.5822372648986761, "grad_norm": 7.757063865661621, "learning_rate": 9.66913502983327e-05, "loss": 0.4031222343444824, "memory(GiB)": 61.91, "step": 13590, "token_acc": 0.9018987341772152, "train_speed(iter/s)": 1.488372 }, { "epoch": 0.5824514802279251, "grad_norm": 3.6609575748443604, "learning_rate": 9.668894246784378e-05, "loss": 0.48310256004333496, "memory(GiB)": 61.91, "step": 13595, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.488424 }, { "epoch": 0.5826656955571741, "grad_norm": 2.3712892532348633, "learning_rate": 9.668653379154086e-05, "loss": 0.45130243301391604, "memory(GiB)": 61.91, "step": 13600, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.488461 }, { "epoch": 0.582879910886423, "grad_norm": 3.0951056480407715, "learning_rate": 9.66841242694676e-05, "loss": 0.3148458242416382, "memory(GiB)": 61.91, "step": 13605, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.488453 }, { "epoch": 0.583094126215672, "grad_norm": 1.5078330039978027, "learning_rate": 9.668171390166763e-05, "loss": 0.345990514755249, "memory(GiB)": 61.91, "step": 13610, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.488456 }, { "epoch": 0.583308341544921, "grad_norm": 5.079171657562256, "learning_rate": 9.667930268818462e-05, "loss": 0.49607076644897463, "memory(GiB)": 61.91, "step": 13615, "token_acc": 0.8976897689768977, "train_speed(iter/s)": 1.488472 }, { "epoch": 0.5835225568741699, "grad_norm": 1.0359894037246704, "learning_rate": 9.667689062906226e-05, "loss": 0.26001021862030027, "memory(GiB)": 61.91, "step": 13620, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.488482 }, { "epoch": 0.5837367722034189, "grad_norm": 2.1008126735687256, "learning_rate": 9.667447772434423e-05, "loss": 0.32034189701080323, "memory(GiB)": 61.91, "step": 13625, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.488544 }, { "epoch": 0.5839509875326678, "grad_norm": 4.930720329284668, "learning_rate": 9.667206397407426e-05, "loss": 0.7969061851501464, "memory(GiB)": 61.91, "step": 13630, "token_acc": 0.881578947368421, "train_speed(iter/s)": 1.488572 }, { "epoch": 0.5841652028619168, "grad_norm": 7.479313850402832, "learning_rate": 9.666964937829606e-05, "loss": 0.6103077411651612, "memory(GiB)": 61.91, "step": 13635, "token_acc": 0.871875, "train_speed(iter/s)": 1.488543 }, { "epoch": 0.5843794181911658, "grad_norm": 2.0037899017333984, "learning_rate": 9.666723393705339e-05, "loss": 0.4689946174621582, "memory(GiB)": 61.91, "step": 13640, "token_acc": 0.8952380952380953, "train_speed(iter/s)": 1.488543 }, { "epoch": 0.5845936335204147, "grad_norm": 3.2685415744781494, "learning_rate": 9.666481765038999e-05, "loss": 0.2537943124771118, "memory(GiB)": 61.91, "step": 13645, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.488518 }, { "epoch": 0.5848078488496636, "grad_norm": 4.9026198387146, "learning_rate": 9.666240051834965e-05, "loss": 0.44504776000976565, "memory(GiB)": 61.91, "step": 13650, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.488479 }, { "epoch": 0.5850220641789127, "grad_norm": 0.6264171004295349, "learning_rate": 9.665998254097616e-05, "loss": 0.14431467056274414, "memory(GiB)": 61.91, "step": 13655, "token_acc": 0.9683794466403162, "train_speed(iter/s)": 1.488427 }, { "epoch": 0.5852362795081616, "grad_norm": 1.251159429550171, "learning_rate": 9.66575637183133e-05, "loss": 0.5726042747497558, "memory(GiB)": 61.91, "step": 13660, "token_acc": 0.8604651162790697, "train_speed(iter/s)": 1.488482 }, { "epoch": 0.5854504948374105, "grad_norm": 4.787815570831299, "learning_rate": 9.665514405040491e-05, "loss": 0.979379940032959, "memory(GiB)": 61.91, "step": 13665, "token_acc": 0.8230088495575221, "train_speed(iter/s)": 1.488456 }, { "epoch": 0.5856647101666596, "grad_norm": 4.100827693939209, "learning_rate": 9.665272353729482e-05, "loss": 0.3607770919799805, "memory(GiB)": 61.91, "step": 13670, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.488445 }, { "epoch": 0.5858789254959085, "grad_norm": 2.614910840988159, "learning_rate": 9.665030217902688e-05, "loss": 0.27694082260131836, "memory(GiB)": 61.91, "step": 13675, "token_acc": 0.9475409836065574, "train_speed(iter/s)": 1.488455 }, { "epoch": 0.5860931408251574, "grad_norm": 3.2868263721466064, "learning_rate": 9.664787997564496e-05, "loss": 0.896211814880371, "memory(GiB)": 61.91, "step": 13680, "token_acc": 0.8256227758007118, "train_speed(iter/s)": 1.488507 }, { "epoch": 0.5863073561544064, "grad_norm": 2.4300742149353027, "learning_rate": 9.664545692719293e-05, "loss": 0.44839677810668943, "memory(GiB)": 61.91, "step": 13685, "token_acc": 0.893687707641196, "train_speed(iter/s)": 1.488472 }, { "epoch": 0.5865215714836554, "grad_norm": 4.986358165740967, "learning_rate": 9.66430330337147e-05, "loss": 0.5378251075744629, "memory(GiB)": 61.91, "step": 13690, "token_acc": 0.8866279069767442, "train_speed(iter/s)": 1.488472 }, { "epoch": 0.5867357868129043, "grad_norm": 2.3901846408843994, "learning_rate": 9.664060829525416e-05, "loss": 0.3968550682067871, "memory(GiB)": 61.91, "step": 13695, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.488592 }, { "epoch": 0.5869500021421533, "grad_norm": 0.040224913507699966, "learning_rate": 9.663818271185525e-05, "loss": 0.2545372724533081, "memory(GiB)": 61.91, "step": 13700, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.488565 }, { "epoch": 0.5871642174714022, "grad_norm": 3.493652105331421, "learning_rate": 9.663575628356191e-05, "loss": 0.4006034851074219, "memory(GiB)": 61.91, "step": 13705, "token_acc": 0.9129129129129129, "train_speed(iter/s)": 1.48855 }, { "epoch": 0.5873784328006512, "grad_norm": 2.3909530639648438, "learning_rate": 9.663332901041809e-05, "loss": 0.6866885662078858, "memory(GiB)": 61.91, "step": 13710, "token_acc": 0.8754716981132076, "train_speed(iter/s)": 1.488566 }, { "epoch": 0.5875926481299002, "grad_norm": 2.0919344425201416, "learning_rate": 9.663090089246778e-05, "loss": 0.45888748168945315, "memory(GiB)": 61.91, "step": 13715, "token_acc": 0.8990536277602523, "train_speed(iter/s)": 1.488542 }, { "epoch": 0.5878068634591491, "grad_norm": 3.4094502925872803, "learning_rate": 9.662847192975496e-05, "loss": 0.4969906806945801, "memory(GiB)": 61.91, "step": 13720, "token_acc": 0.8973684210526316, "train_speed(iter/s)": 1.488543 }, { "epoch": 0.588021078788398, "grad_norm": 6.400973320007324, "learning_rate": 9.662604212232362e-05, "loss": 0.45326967239379884, "memory(GiB)": 61.91, "step": 13725, "token_acc": 0.8994082840236687, "train_speed(iter/s)": 1.488572 }, { "epoch": 0.5882352941176471, "grad_norm": 4.781771183013916, "learning_rate": 9.662361147021779e-05, "loss": 0.75138840675354, "memory(GiB)": 61.91, "step": 13730, "token_acc": 0.847457627118644, "train_speed(iter/s)": 1.488546 }, { "epoch": 0.588449509446896, "grad_norm": 14.103879928588867, "learning_rate": 9.662117997348151e-05, "loss": 0.7839416027069092, "memory(GiB)": 61.91, "step": 13735, "token_acc": 0.8212290502793296, "train_speed(iter/s)": 1.48856 }, { "epoch": 0.588663724776145, "grad_norm": 3.6092867851257324, "learning_rate": 9.661874763215881e-05, "loss": 0.5428380012512207, "memory(GiB)": 61.91, "step": 13740, "token_acc": 0.8870967741935484, "train_speed(iter/s)": 1.488622 }, { "epoch": 0.588877940105394, "grad_norm": 2.630082607269287, "learning_rate": 9.661631444629378e-05, "loss": 0.3032931566238403, "memory(GiB)": 61.91, "step": 13745, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.4886 }, { "epoch": 0.5890921554346429, "grad_norm": 2.1401615142822266, "learning_rate": 9.661388041593047e-05, "loss": 0.3926527500152588, "memory(GiB)": 61.91, "step": 13750, "token_acc": 0.9209726443768997, "train_speed(iter/s)": 1.488586 }, { "epoch": 0.5893063707638919, "grad_norm": 3.194918155670166, "learning_rate": 9.661144554111298e-05, "loss": 0.16474244594573975, "memory(GiB)": 61.91, "step": 13755, "token_acc": 0.9715447154471545, "train_speed(iter/s)": 1.488613 }, { "epoch": 0.5895205860931408, "grad_norm": 3.371760845184326, "learning_rate": 9.660900982188545e-05, "loss": 0.41986966133117676, "memory(GiB)": 61.91, "step": 13760, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.488625 }, { "epoch": 0.5897348014223898, "grad_norm": 3.066784620285034, "learning_rate": 9.660657325829198e-05, "loss": 0.32955756187438967, "memory(GiB)": 61.91, "step": 13765, "token_acc": 0.9266862170087976, "train_speed(iter/s)": 1.488626 }, { "epoch": 0.5899490167516388, "grad_norm": 2.9243412017822266, "learning_rate": 9.660413585037671e-05, "loss": 0.3419323205947876, "memory(GiB)": 61.91, "step": 13770, "token_acc": 0.9295302013422819, "train_speed(iter/s)": 1.488601 }, { "epoch": 0.5901632320808877, "grad_norm": 6.241645812988281, "learning_rate": 9.66016975981838e-05, "loss": 0.5576666355133056, "memory(GiB)": 61.91, "step": 13775, "token_acc": 0.8862275449101796, "train_speed(iter/s)": 1.488587 }, { "epoch": 0.5903774474101366, "grad_norm": 1.6669409275054932, "learning_rate": 9.659925850175742e-05, "loss": 0.6932161808013916, "memory(GiB)": 61.91, "step": 13780, "token_acc": 0.8566978193146417, "train_speed(iter/s)": 1.488556 }, { "epoch": 0.5905916627393857, "grad_norm": 3.320474147796631, "learning_rate": 9.659681856114177e-05, "loss": 0.42017450332641604, "memory(GiB)": 61.91, "step": 13785, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.488571 }, { "epoch": 0.5908058780686346, "grad_norm": 5.830926895141602, "learning_rate": 9.659437777638102e-05, "loss": 0.5472935676574707, "memory(GiB)": 61.91, "step": 13790, "token_acc": 0.8712121212121212, "train_speed(iter/s)": 1.488574 }, { "epoch": 0.5910200933978835, "grad_norm": 4.40803337097168, "learning_rate": 9.659193614751942e-05, "loss": 0.2612058877944946, "memory(GiB)": 61.91, "step": 13795, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.488609 }, { "epoch": 0.5912343087271326, "grad_norm": 2.810000419616699, "learning_rate": 9.658949367460119e-05, "loss": 0.3767163038253784, "memory(GiB)": 61.91, "step": 13800, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.488609 }, { "epoch": 0.5914485240563815, "grad_norm": 2.667909860610962, "learning_rate": 9.658705035767059e-05, "loss": 0.28503003120422366, "memory(GiB)": 61.91, "step": 13805, "token_acc": 0.9347181008902077, "train_speed(iter/s)": 1.4886 }, { "epoch": 0.5916627393856304, "grad_norm": 0.7143001556396484, "learning_rate": 9.658460619677185e-05, "loss": 0.6494495868682861, "memory(GiB)": 61.91, "step": 13810, "token_acc": 0.8774703557312253, "train_speed(iter/s)": 1.488629 }, { "epoch": 0.5918769547148794, "grad_norm": 2.17639422416687, "learning_rate": 9.658216119194929e-05, "loss": 0.431962251663208, "memory(GiB)": 61.91, "step": 13815, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.488611 }, { "epoch": 0.5920911700441284, "grad_norm": 0.15772181749343872, "learning_rate": 9.657971534324717e-05, "loss": 0.16030507087707518, "memory(GiB)": 61.91, "step": 13820, "token_acc": 0.9536679536679536, "train_speed(iter/s)": 1.48862 }, { "epoch": 0.5923053853733773, "grad_norm": 4.220511436462402, "learning_rate": 9.657726865070981e-05, "loss": 0.3489703893661499, "memory(GiB)": 61.91, "step": 13825, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.488625 }, { "epoch": 0.5925196007026263, "grad_norm": 1.6556353569030762, "learning_rate": 9.657482111438154e-05, "loss": 0.6313056468963623, "memory(GiB)": 61.91, "step": 13830, "token_acc": 0.8656716417910447, "train_speed(iter/s)": 1.48871 }, { "epoch": 0.5927338160318752, "grad_norm": 4.931127548217773, "learning_rate": 9.657237273430669e-05, "loss": 0.5550161838531494, "memory(GiB)": 61.91, "step": 13835, "token_acc": 0.89272030651341, "train_speed(iter/s)": 1.4887 }, { "epoch": 0.5929480313611242, "grad_norm": 7.529865741729736, "learning_rate": 9.656992351052961e-05, "loss": 0.2418933629989624, "memory(GiB)": 61.91, "step": 13840, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.488695 }, { "epoch": 0.5931622466903732, "grad_norm": 4.04878044128418, "learning_rate": 9.656747344309469e-05, "loss": 0.4411433696746826, "memory(GiB)": 61.91, "step": 13845, "token_acc": 0.9014598540145985, "train_speed(iter/s)": 1.488732 }, { "epoch": 0.5933764620196221, "grad_norm": 3.462559461593628, "learning_rate": 9.656502253204632e-05, "loss": 0.3393659830093384, "memory(GiB)": 61.91, "step": 13850, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.488719 }, { "epoch": 0.593590677348871, "grad_norm": 5.656381130218506, "learning_rate": 9.656257077742886e-05, "loss": 0.2009063720703125, "memory(GiB)": 61.91, "step": 13855, "token_acc": 0.96, "train_speed(iter/s)": 1.488718 }, { "epoch": 0.5938048926781201, "grad_norm": 3.214001178741455, "learning_rate": 9.656011817928676e-05, "loss": 0.5668754577636719, "memory(GiB)": 61.91, "step": 13860, "token_acc": 0.8745387453874539, "train_speed(iter/s)": 1.488701 }, { "epoch": 0.594019108007369, "grad_norm": 1.4025168418884277, "learning_rate": 9.655766473766444e-05, "loss": 0.4490914821624756, "memory(GiB)": 61.91, "step": 13865, "token_acc": 0.9239543726235742, "train_speed(iter/s)": 1.488668 }, { "epoch": 0.5942333233366179, "grad_norm": 0.8365256786346436, "learning_rate": 9.655521045260636e-05, "loss": 0.15911716222763062, "memory(GiB)": 61.91, "step": 13870, "token_acc": 0.9606299212598425, "train_speed(iter/s)": 1.48868 }, { "epoch": 0.594447538665867, "grad_norm": 2.1679015159606934, "learning_rate": 9.655275532415696e-05, "loss": 0.36083278656005857, "memory(GiB)": 61.91, "step": 13875, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.488626 }, { "epoch": 0.5946617539951159, "grad_norm": 2.6328043937683105, "learning_rate": 9.655029935236073e-05, "loss": 0.45126943588256835, "memory(GiB)": 61.91, "step": 13880, "token_acc": 0.887719298245614, "train_speed(iter/s)": 1.488686 }, { "epoch": 0.5948759693243648, "grad_norm": 2.491976022720337, "learning_rate": 9.654784253726216e-05, "loss": 0.4230278491973877, "memory(GiB)": 61.91, "step": 13885, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.488664 }, { "epoch": 0.5950901846536139, "grad_norm": 2.9671273231506348, "learning_rate": 9.654538487890577e-05, "loss": 0.21676340103149414, "memory(GiB)": 61.91, "step": 13890, "token_acc": 0.9430604982206405, "train_speed(iter/s)": 1.488729 }, { "epoch": 0.5953043999828628, "grad_norm": 1.7437989711761475, "learning_rate": 9.654292637733605e-05, "loss": 0.42176222801208496, "memory(GiB)": 61.91, "step": 13895, "token_acc": 0.9252336448598131, "train_speed(iter/s)": 1.488729 }, { "epoch": 0.5955186153121117, "grad_norm": 3.9670486450195312, "learning_rate": 9.654046703259755e-05, "loss": 0.38765735626220704, "memory(GiB)": 61.91, "step": 13900, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.488714 }, { "epoch": 0.5957328306413607, "grad_norm": 2.512817859649658, "learning_rate": 9.653800684473485e-05, "loss": 0.21742568016052247, "memory(GiB)": 61.91, "step": 13905, "token_acc": 0.9415384615384615, "train_speed(iter/s)": 1.488694 }, { "epoch": 0.5959470459706097, "grad_norm": 1.3592387437820435, "learning_rate": 9.653554581379249e-05, "loss": 0.2425626277923584, "memory(GiB)": 61.91, "step": 13910, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.488721 }, { "epoch": 0.5961612612998586, "grad_norm": 1.081842064857483, "learning_rate": 9.653308393981508e-05, "loss": 0.24586520195007325, "memory(GiB)": 61.91, "step": 13915, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.488724 }, { "epoch": 0.5963754766291076, "grad_norm": 3.7341103553771973, "learning_rate": 9.65306212228472e-05, "loss": 0.4090461254119873, "memory(GiB)": 61.91, "step": 13920, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.488743 }, { "epoch": 0.5965896919583565, "grad_norm": 3.1156599521636963, "learning_rate": 9.652815766293344e-05, "loss": 0.37679443359375, "memory(GiB)": 61.91, "step": 13925, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.488745 }, { "epoch": 0.5968039072876055, "grad_norm": 3.778721570968628, "learning_rate": 9.652569326011849e-05, "loss": 0.10195324420928956, "memory(GiB)": 61.91, "step": 13930, "token_acc": 0.9765886287625418, "train_speed(iter/s)": 1.488727 }, { "epoch": 0.5970181226168545, "grad_norm": 4.843070983886719, "learning_rate": 9.652322801444695e-05, "loss": 0.4238940715789795, "memory(GiB)": 61.91, "step": 13935, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.488703 }, { "epoch": 0.5972323379461034, "grad_norm": 2.630650758743286, "learning_rate": 9.652076192596349e-05, "loss": 0.2551581382751465, "memory(GiB)": 61.91, "step": 13940, "token_acc": 0.9450980392156862, "train_speed(iter/s)": 1.488762 }, { "epoch": 0.5974465532753523, "grad_norm": 2.2302188873291016, "learning_rate": 9.65182949947128e-05, "loss": 0.2442610502243042, "memory(GiB)": 61.91, "step": 13945, "token_acc": 0.9478827361563518, "train_speed(iter/s)": 1.488783 }, { "epoch": 0.5976607686046014, "grad_norm": 1.7835274934768677, "learning_rate": 9.651582722073953e-05, "loss": 0.5668976306915283, "memory(GiB)": 61.91, "step": 13950, "token_acc": 0.8870431893687708, "train_speed(iter/s)": 1.488818 }, { "epoch": 0.5978749839338503, "grad_norm": 3.6029205322265625, "learning_rate": 9.651335860408843e-05, "loss": 0.7078385829925538, "memory(GiB)": 61.91, "step": 13955, "token_acc": 0.8634361233480177, "train_speed(iter/s)": 1.488798 }, { "epoch": 0.5980891992630992, "grad_norm": 0.19108808040618896, "learning_rate": 9.651088914480421e-05, "loss": 0.3037719249725342, "memory(GiB)": 61.91, "step": 13960, "token_acc": 0.9190283400809717, "train_speed(iter/s)": 1.488847 }, { "epoch": 0.5983034145923483, "grad_norm": 3.7818682193756104, "learning_rate": 9.650841884293159e-05, "loss": 0.8432294845581054, "memory(GiB)": 61.91, "step": 13965, "token_acc": 0.8464730290456431, "train_speed(iter/s)": 1.488844 }, { "epoch": 0.5985176299215972, "grad_norm": 3.6480066776275635, "learning_rate": 9.650594769851535e-05, "loss": 0.4612950325012207, "memory(GiB)": 61.91, "step": 13970, "token_acc": 0.899390243902439, "train_speed(iter/s)": 1.48883 }, { "epoch": 0.5987318452508461, "grad_norm": 3.233715295791626, "learning_rate": 9.650347571160023e-05, "loss": 0.32309958934783933, "memory(GiB)": 61.91, "step": 13975, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.488848 }, { "epoch": 0.5989460605800951, "grad_norm": 4.039986610412598, "learning_rate": 9.650100288223102e-05, "loss": 0.35417191982269286, "memory(GiB)": 61.91, "step": 13980, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.488952 }, { "epoch": 0.5991602759093441, "grad_norm": 3.758554458618164, "learning_rate": 9.649852921045254e-05, "loss": 0.5582855701446533, "memory(GiB)": 61.91, "step": 13985, "token_acc": 0.8544776119402985, "train_speed(iter/s)": 1.488923 }, { "epoch": 0.599374491238593, "grad_norm": 3.331873655319214, "learning_rate": 9.649605469630958e-05, "loss": 0.30801575183868407, "memory(GiB)": 61.91, "step": 13990, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.488905 }, { "epoch": 0.599588706567842, "grad_norm": 3.3926875591278076, "learning_rate": 9.649357933984697e-05, "loss": 0.46751880645751953, "memory(GiB)": 61.91, "step": 13995, "token_acc": 0.8991596638655462, "train_speed(iter/s)": 1.488872 }, { "epoch": 0.599802921897091, "grad_norm": 0.8655768036842346, "learning_rate": 9.649110314110954e-05, "loss": 0.5064396858215332, "memory(GiB)": 61.91, "step": 14000, "token_acc": 0.8895705521472392, "train_speed(iter/s)": 1.488844 }, { "epoch": 0.599802921897091, "eval_loss": 2.4434611797332764, "eval_runtime": 13.2916, "eval_samples_per_second": 7.524, "eval_steps_per_second": 7.524, "eval_token_acc": 0.43828016643550627, "step": 14000 }, { "epoch": 0.6000171372263399, "grad_norm": 0.8183765411376953, "learning_rate": 9.648862610014218e-05, "loss": 0.16789748668670654, "memory(GiB)": 61.91, "step": 14005, "token_acc": 0.5773930753564155, "train_speed(iter/s)": 1.486612 }, { "epoch": 0.6002313525555889, "grad_norm": 3.8673360347747803, "learning_rate": 9.648614821698975e-05, "loss": 0.3370398998260498, "memory(GiB)": 61.91, "step": 14010, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.486603 }, { "epoch": 0.6004455678848378, "grad_norm": 3.3346216678619385, "learning_rate": 9.648366949169712e-05, "loss": 0.1755645751953125, "memory(GiB)": 61.91, "step": 14015, "token_acc": 0.967948717948718, "train_speed(iter/s)": 1.486612 }, { "epoch": 0.6006597832140868, "grad_norm": 2.916618585586548, "learning_rate": 9.648118992430923e-05, "loss": 0.5253336906433106, "memory(GiB)": 61.91, "step": 14020, "token_acc": 0.8871951219512195, "train_speed(iter/s)": 1.486686 }, { "epoch": 0.6008739985433358, "grad_norm": 4.609787940979004, "learning_rate": 9.647870951487098e-05, "loss": 0.4266552448272705, "memory(GiB)": 61.91, "step": 14025, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.486623 }, { "epoch": 0.6010882138725847, "grad_norm": 3.3727235794067383, "learning_rate": 9.64762282634273e-05, "loss": 0.5904091358184814, "memory(GiB)": 61.91, "step": 14030, "token_acc": 0.8480565371024735, "train_speed(iter/s)": 1.48667 }, { "epoch": 0.6013024292018336, "grad_norm": 0.057956937700510025, "learning_rate": 9.647374617002314e-05, "loss": 0.3255255699157715, "memory(GiB)": 61.91, "step": 14035, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.486682 }, { "epoch": 0.6015166445310827, "grad_norm": 2.872129201889038, "learning_rate": 9.647126323470349e-05, "loss": 0.295322847366333, "memory(GiB)": 61.91, "step": 14040, "token_acc": 0.9467084639498433, "train_speed(iter/s)": 1.486679 }, { "epoch": 0.6017308598603316, "grad_norm": 4.119054317474365, "learning_rate": 9.646877945751332e-05, "loss": 0.45143980979919435, "memory(GiB)": 61.91, "step": 14045, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.486689 }, { "epoch": 0.6019450751895805, "grad_norm": 0.5347427725791931, "learning_rate": 9.646629483849758e-05, "loss": 0.26142168045043945, "memory(GiB)": 61.91, "step": 14050, "token_acc": 0.9627118644067797, "train_speed(iter/s)": 1.486677 }, { "epoch": 0.6021592905188295, "grad_norm": 4.1233110427856445, "learning_rate": 9.646380937770135e-05, "loss": 0.5276237487792969, "memory(GiB)": 61.91, "step": 14055, "token_acc": 0.8878205128205128, "train_speed(iter/s)": 1.48669 }, { "epoch": 0.6023735058480785, "grad_norm": 5.289035320281982, "learning_rate": 9.646132307516961e-05, "loss": 0.7230296611785889, "memory(GiB)": 61.91, "step": 14060, "token_acc": 0.8405797101449275, "train_speed(iter/s)": 1.486677 }, { "epoch": 0.6025877211773274, "grad_norm": 2.596764087677002, "learning_rate": 9.645883593094745e-05, "loss": 0.5383239269256592, "memory(GiB)": 61.91, "step": 14065, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.486723 }, { "epoch": 0.6028019365065764, "grad_norm": 3.7913663387298584, "learning_rate": 9.645634794507987e-05, "loss": 0.6229513168334961, "memory(GiB)": 61.91, "step": 14070, "token_acc": 0.8754448398576512, "train_speed(iter/s)": 1.486744 }, { "epoch": 0.6030161518358254, "grad_norm": 3.7340970039367676, "learning_rate": 9.645385911761196e-05, "loss": 0.46764936447143557, "memory(GiB)": 61.91, "step": 14075, "token_acc": 0.8840579710144928, "train_speed(iter/s)": 1.486701 }, { "epoch": 0.6032303671650744, "grad_norm": 0.1802360713481903, "learning_rate": 9.645136944858883e-05, "loss": 0.40853233337402345, "memory(GiB)": 61.91, "step": 14080, "token_acc": 0.9075907590759076, "train_speed(iter/s)": 1.486682 }, { "epoch": 0.6034445824943233, "grad_norm": 4.657423973083496, "learning_rate": 9.644887893805557e-05, "loss": 0.4033369541168213, "memory(GiB)": 61.91, "step": 14085, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.486677 }, { "epoch": 0.6036587978235722, "grad_norm": 2.252159357070923, "learning_rate": 9.644638758605729e-05, "loss": 0.4970703125, "memory(GiB)": 61.91, "step": 14090, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.486684 }, { "epoch": 0.6038730131528213, "grad_norm": 4.696386337280273, "learning_rate": 9.644389539263911e-05, "loss": 0.6287853240966796, "memory(GiB)": 61.91, "step": 14095, "token_acc": 0.8723404255319149, "train_speed(iter/s)": 1.486729 }, { "epoch": 0.6040872284820702, "grad_norm": 2.03652286529541, "learning_rate": 9.644140235784623e-05, "loss": 0.22837362289428711, "memory(GiB)": 61.91, "step": 14100, "token_acc": 0.9455040871934605, "train_speed(iter/s)": 1.486752 }, { "epoch": 0.6043014438113191, "grad_norm": 6.765116214752197, "learning_rate": 9.643890848172377e-05, "loss": 0.3953569412231445, "memory(GiB)": 61.91, "step": 14105, "token_acc": 0.9295774647887324, "train_speed(iter/s)": 1.486748 }, { "epoch": 0.6045156591405682, "grad_norm": 1.937926173210144, "learning_rate": 9.643641376431691e-05, "loss": 0.3807487964630127, "memory(GiB)": 61.91, "step": 14110, "token_acc": 0.9306122448979591, "train_speed(iter/s)": 1.486756 }, { "epoch": 0.6047298744698171, "grad_norm": 1.630527138710022, "learning_rate": 9.643391820567084e-05, "loss": 0.4083390712738037, "memory(GiB)": 61.91, "step": 14115, "token_acc": 0.9158249158249159, "train_speed(iter/s)": 1.486755 }, { "epoch": 0.604944089799066, "grad_norm": 2.2819087505340576, "learning_rate": 9.643142180583081e-05, "loss": 0.34829862117767335, "memory(GiB)": 61.91, "step": 14120, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.486754 }, { "epoch": 0.605158305128315, "grad_norm": 3.1325066089630127, "learning_rate": 9.642892456484201e-05, "loss": 0.3364971876144409, "memory(GiB)": 61.91, "step": 14125, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.486735 }, { "epoch": 0.605372520457564, "grad_norm": 4.1054253578186035, "learning_rate": 9.642642648274968e-05, "loss": 0.3065436363220215, "memory(GiB)": 61.91, "step": 14130, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.486692 }, { "epoch": 0.6055867357868129, "grad_norm": 2.3581690788269043, "learning_rate": 9.64239275595991e-05, "loss": 0.6818489551544189, "memory(GiB)": 61.91, "step": 14135, "token_acc": 0.8579710144927536, "train_speed(iter/s)": 1.486674 }, { "epoch": 0.6058009511160619, "grad_norm": 11.97305679321289, "learning_rate": 9.642142779543551e-05, "loss": 0.39144129753112794, "memory(GiB)": 61.91, "step": 14140, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.4867 }, { "epoch": 0.6060151664453108, "grad_norm": 0.8987568020820618, "learning_rate": 9.641892719030421e-05, "loss": 0.32344956398010255, "memory(GiB)": 61.91, "step": 14145, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.486692 }, { "epoch": 0.6062293817745598, "grad_norm": 1.0360469818115234, "learning_rate": 9.64164257442505e-05, "loss": 0.28715991973876953, "memory(GiB)": 61.91, "step": 14150, "token_acc": 0.932806324110672, "train_speed(iter/s)": 1.486663 }, { "epoch": 0.6064435971038088, "grad_norm": 5.0235395431518555, "learning_rate": 9.64139234573197e-05, "loss": 0.6477124214172363, "memory(GiB)": 61.91, "step": 14155, "token_acc": 0.8859060402684564, "train_speed(iter/s)": 1.486621 }, { "epoch": 0.6066578124330577, "grad_norm": 4.468339920043945, "learning_rate": 9.641142032955714e-05, "loss": 0.305224084854126, "memory(GiB)": 61.91, "step": 14160, "token_acc": 0.9139344262295082, "train_speed(iter/s)": 1.486605 }, { "epoch": 0.6068720277623066, "grad_norm": 2.692639112472534, "learning_rate": 9.640891636100815e-05, "loss": 0.4382962703704834, "memory(GiB)": 61.91, "step": 14165, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.486567 }, { "epoch": 0.6070862430915557, "grad_norm": 3.5822701454162598, "learning_rate": 9.640641155171812e-05, "loss": 0.43465137481689453, "memory(GiB)": 61.91, "step": 14170, "token_acc": 0.9079365079365079, "train_speed(iter/s)": 1.486594 }, { "epoch": 0.6073004584208046, "grad_norm": 3.488126754760742, "learning_rate": 9.64039059017324e-05, "loss": 0.6095567226409913, "memory(GiB)": 61.91, "step": 14175, "token_acc": 0.8944281524926686, "train_speed(iter/s)": 1.486585 }, { "epoch": 0.6075146737500535, "grad_norm": 2.7439005374908447, "learning_rate": 9.64013994110964e-05, "loss": 0.5629470348358154, "memory(GiB)": 61.91, "step": 14180, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.486633 }, { "epoch": 0.6077288890793026, "grad_norm": 1.6845779418945312, "learning_rate": 9.639889207985552e-05, "loss": 0.34294734001159666, "memory(GiB)": 61.91, "step": 14185, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.486606 }, { "epoch": 0.6079431044085515, "grad_norm": 1.2361080646514893, "learning_rate": 9.639638390805518e-05, "loss": 0.2475717306137085, "memory(GiB)": 61.91, "step": 14190, "token_acc": 0.9496124031007752, "train_speed(iter/s)": 1.486612 }, { "epoch": 0.6081573197378004, "grad_norm": 4.1260666847229, "learning_rate": 9.639387489574083e-05, "loss": 0.15836730003356933, "memory(GiB)": 61.91, "step": 14195, "token_acc": 0.9695945945945946, "train_speed(iter/s)": 1.486568 }, { "epoch": 0.6083715350670494, "grad_norm": 2.052952289581299, "learning_rate": 9.639136504295792e-05, "loss": 0.2942540407180786, "memory(GiB)": 61.91, "step": 14200, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.486562 }, { "epoch": 0.6085857503962984, "grad_norm": 2.4480578899383545, "learning_rate": 9.638885434975189e-05, "loss": 0.38722302913665774, "memory(GiB)": 61.91, "step": 14205, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.486537 }, { "epoch": 0.6087999657255473, "grad_norm": 1.4277976751327515, "learning_rate": 9.638634281616827e-05, "loss": 0.21782875061035156, "memory(GiB)": 61.91, "step": 14210, "token_acc": 0.955719557195572, "train_speed(iter/s)": 1.486521 }, { "epoch": 0.6090141810547963, "grad_norm": 5.976009845733643, "learning_rate": 9.638383044225252e-05, "loss": 0.3191081523895264, "memory(GiB)": 61.91, "step": 14215, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.486528 }, { "epoch": 0.6092283963840452, "grad_norm": 4.49836540222168, "learning_rate": 9.638131722805018e-05, "loss": 0.38584327697753906, "memory(GiB)": 61.91, "step": 14220, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.486586 }, { "epoch": 0.6094426117132942, "grad_norm": 2.2129650115966797, "learning_rate": 9.637880317360677e-05, "loss": 0.3592863082885742, "memory(GiB)": 61.91, "step": 14225, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.486594 }, { "epoch": 0.6096568270425432, "grad_norm": 2.3062875270843506, "learning_rate": 9.637628827896784e-05, "loss": 0.20665173530578612, "memory(GiB)": 61.91, "step": 14230, "token_acc": 0.9651567944250871, "train_speed(iter/s)": 1.486581 }, { "epoch": 0.6098710423717921, "grad_norm": 2.9346299171447754, "learning_rate": 9.637377254417892e-05, "loss": 0.3606394290924072, "memory(GiB)": 61.91, "step": 14235, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.486582 }, { "epoch": 0.610085257701041, "grad_norm": 1.4274080991744995, "learning_rate": 9.637125596928562e-05, "loss": 0.25708489418029784, "memory(GiB)": 61.91, "step": 14240, "token_acc": 0.9431438127090301, "train_speed(iter/s)": 1.486575 }, { "epoch": 0.6102994730302901, "grad_norm": 3.2180304527282715, "learning_rate": 9.636873855433353e-05, "loss": 0.4530495643615723, "memory(GiB)": 61.91, "step": 14245, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.486594 }, { "epoch": 0.610513688359539, "grad_norm": 1.0057686567306519, "learning_rate": 9.636622029936823e-05, "loss": 0.11905807256698608, "memory(GiB)": 61.91, "step": 14250, "token_acc": 0.975, "train_speed(iter/s)": 1.486589 }, { "epoch": 0.6107279036887879, "grad_norm": 3.292430877685547, "learning_rate": 9.636370120443536e-05, "loss": 0.23512375354766846, "memory(GiB)": 61.91, "step": 14255, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.486577 }, { "epoch": 0.610942119018037, "grad_norm": 4.931755542755127, "learning_rate": 9.636118126958056e-05, "loss": 0.43858795166015624, "memory(GiB)": 61.91, "step": 14260, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.486556 }, { "epoch": 0.6111563343472859, "grad_norm": 4.004387855529785, "learning_rate": 9.635866049484945e-05, "loss": 0.30942411422729493, "memory(GiB)": 61.91, "step": 14265, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.486596 }, { "epoch": 0.6113705496765348, "grad_norm": 2.4785938262939453, "learning_rate": 9.635613888028773e-05, "loss": 0.23449358940124512, "memory(GiB)": 61.91, "step": 14270, "token_acc": 0.9438943894389439, "train_speed(iter/s)": 1.486588 }, { "epoch": 0.6115847650057838, "grad_norm": 2.479883909225464, "learning_rate": 9.635361642594107e-05, "loss": 0.593048095703125, "memory(GiB)": 61.91, "step": 14275, "token_acc": 0.8847352024922118, "train_speed(iter/s)": 1.486619 }, { "epoch": 0.6117989803350328, "grad_norm": 3.8209149837493896, "learning_rate": 9.635109313185515e-05, "loss": 0.38460378646850585, "memory(GiB)": 61.91, "step": 14280, "token_acc": 0.8931034482758621, "train_speed(iter/s)": 1.486644 }, { "epoch": 0.6120131956642817, "grad_norm": 3.7049787044525146, "learning_rate": 9.634856899807571e-05, "loss": 0.28540711402893065, "memory(GiB)": 61.91, "step": 14285, "token_acc": 0.9368770764119602, "train_speed(iter/s)": 1.486637 }, { "epoch": 0.6122274109935307, "grad_norm": 3.6475064754486084, "learning_rate": 9.634604402464846e-05, "loss": 0.31600198745727537, "memory(GiB)": 61.91, "step": 14290, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.486612 }, { "epoch": 0.6124416263227797, "grad_norm": 2.1444215774536133, "learning_rate": 9.634351821161916e-05, "loss": 0.28791892528533936, "memory(GiB)": 61.91, "step": 14295, "token_acc": 0.95703125, "train_speed(iter/s)": 1.486614 }, { "epoch": 0.6126558416520286, "grad_norm": 1.262500286102295, "learning_rate": 9.634099155903353e-05, "loss": 0.3598459243774414, "memory(GiB)": 61.91, "step": 14300, "token_acc": 0.9325153374233128, "train_speed(iter/s)": 1.486624 }, { "epoch": 0.6128700569812776, "grad_norm": 2.7544631958007812, "learning_rate": 9.633846406693738e-05, "loss": 0.5233990669250488, "memory(GiB)": 61.91, "step": 14305, "token_acc": 0.8878205128205128, "train_speed(iter/s)": 1.486609 }, { "epoch": 0.6130842723105265, "grad_norm": 5.591941833496094, "learning_rate": 9.633593573537649e-05, "loss": 0.44601001739501955, "memory(GiB)": 61.91, "step": 14310, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.486609 }, { "epoch": 0.6132984876397755, "grad_norm": 5.0450944900512695, "learning_rate": 9.633340656439664e-05, "loss": 0.5231477737426757, "memory(GiB)": 61.91, "step": 14315, "token_acc": 0.8877551020408163, "train_speed(iter/s)": 1.486611 }, { "epoch": 0.6135127029690245, "grad_norm": 1.7608586549758911, "learning_rate": 9.633087655404369e-05, "loss": 0.3960092067718506, "memory(GiB)": 61.91, "step": 14320, "token_acc": 0.904, "train_speed(iter/s)": 1.486662 }, { "epoch": 0.6137269182982734, "grad_norm": 0.6324070692062378, "learning_rate": 9.632834570436343e-05, "loss": 0.4520329475402832, "memory(GiB)": 61.91, "step": 14325, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.486677 }, { "epoch": 0.6139411336275223, "grad_norm": 0.7124397158622742, "learning_rate": 9.632581401540174e-05, "loss": 0.3519623279571533, "memory(GiB)": 61.91, "step": 14330, "token_acc": 0.9214501510574018, "train_speed(iter/s)": 1.486651 }, { "epoch": 0.6141553489567714, "grad_norm": 2.2128617763519287, "learning_rate": 9.632328148720447e-05, "loss": 0.17168896198272704, "memory(GiB)": 61.91, "step": 14335, "token_acc": 0.9711191335740073, "train_speed(iter/s)": 1.486632 }, { "epoch": 0.6143695642860203, "grad_norm": 4.875009059906006, "learning_rate": 9.632074811981749e-05, "loss": 0.48229174613952636, "memory(GiB)": 61.91, "step": 14340, "token_acc": 0.8990228013029316, "train_speed(iter/s)": 1.486679 }, { "epoch": 0.6145837796152692, "grad_norm": 3.771174192428589, "learning_rate": 9.631821391328671e-05, "loss": 0.7388247966766357, "memory(GiB)": 61.91, "step": 14345, "token_acc": 0.8643533123028391, "train_speed(iter/s)": 1.486727 }, { "epoch": 0.6147979949445183, "grad_norm": 1.122465968132019, "learning_rate": 9.631567886765804e-05, "loss": 0.23930554389953612, "memory(GiB)": 61.91, "step": 14350, "token_acc": 0.9616724738675958, "train_speed(iter/s)": 1.48672 }, { "epoch": 0.6150122102737672, "grad_norm": 1.132616400718689, "learning_rate": 9.63131429829774e-05, "loss": 0.07855649590492249, "memory(GiB)": 61.91, "step": 14355, "token_acc": 0.982078853046595, "train_speed(iter/s)": 1.486735 }, { "epoch": 0.6152264256030161, "grad_norm": 5.2229413986206055, "learning_rate": 9.631060625929073e-05, "loss": 0.36440362930297854, "memory(GiB)": 61.91, "step": 14360, "token_acc": 0.9276729559748428, "train_speed(iter/s)": 1.486731 }, { "epoch": 0.6154406409322651, "grad_norm": 2.2336020469665527, "learning_rate": 9.630806869664397e-05, "loss": 0.24940683841705322, "memory(GiB)": 61.91, "step": 14365, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.486718 }, { "epoch": 0.6156548562615141, "grad_norm": 3.5841469764709473, "learning_rate": 9.630553029508312e-05, "loss": 0.41121740341186525, "memory(GiB)": 61.91, "step": 14370, "token_acc": 0.9025270758122743, "train_speed(iter/s)": 1.486678 }, { "epoch": 0.615869071590763, "grad_norm": 8.739190101623535, "learning_rate": 9.630299105465414e-05, "loss": 0.3900779724121094, "memory(GiB)": 61.91, "step": 14375, "token_acc": 0.9148148148148149, "train_speed(iter/s)": 1.486691 }, { "epoch": 0.616083286920012, "grad_norm": 3.017531633377075, "learning_rate": 9.630045097540305e-05, "loss": 0.44404001235961915, "memory(GiB)": 61.91, "step": 14380, "token_acc": 0.8982558139534884, "train_speed(iter/s)": 1.486689 }, { "epoch": 0.6162975022492609, "grad_norm": 3.1917130947113037, "learning_rate": 9.629791005737587e-05, "loss": 0.6213650703430176, "memory(GiB)": 61.91, "step": 14385, "token_acc": 0.8403614457831325, "train_speed(iter/s)": 1.48672 }, { "epoch": 0.6165117175785099, "grad_norm": 2.2005603313446045, "learning_rate": 9.62953683006186e-05, "loss": 0.3143636465072632, "memory(GiB)": 61.91, "step": 14390, "token_acc": 0.9382352941176471, "train_speed(iter/s)": 1.486724 }, { "epoch": 0.6167259329077589, "grad_norm": 2.340540647506714, "learning_rate": 9.62928257051773e-05, "loss": 0.3297745227813721, "memory(GiB)": 61.91, "step": 14395, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.486685 }, { "epoch": 0.6169401482370078, "grad_norm": 3.1718060970306396, "learning_rate": 9.629028227109806e-05, "loss": 0.3801619052886963, "memory(GiB)": 61.91, "step": 14400, "token_acc": 0.916, "train_speed(iter/s)": 1.48668 }, { "epoch": 0.6171543635662567, "grad_norm": 2.3237273693084717, "learning_rate": 9.628773799842692e-05, "loss": 0.49047136306762695, "memory(GiB)": 61.91, "step": 14405, "token_acc": 0.8790035587188612, "train_speed(iter/s)": 1.486671 }, { "epoch": 0.6173685788955058, "grad_norm": 0.4901587665081024, "learning_rate": 9.628519288720998e-05, "loss": 0.4465914249420166, "memory(GiB)": 61.91, "step": 14410, "token_acc": 0.8985074626865671, "train_speed(iter/s)": 1.486674 }, { "epoch": 0.6175827942247547, "grad_norm": 3.085916757583618, "learning_rate": 9.628264693749336e-05, "loss": 0.5802166938781739, "memory(GiB)": 61.91, "step": 14415, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.486726 }, { "epoch": 0.6177970095540037, "grad_norm": 2.8544583320617676, "learning_rate": 9.628010014932315e-05, "loss": 0.39330766201019285, "memory(GiB)": 61.91, "step": 14420, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.486693 }, { "epoch": 0.6180112248832527, "grad_norm": 4.287430763244629, "learning_rate": 9.627755252274555e-05, "loss": 0.4472951412200928, "memory(GiB)": 61.91, "step": 14425, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.486717 }, { "epoch": 0.6182254402125016, "grad_norm": 1.947024941444397, "learning_rate": 9.627500405780664e-05, "loss": 0.2328768253326416, "memory(GiB)": 61.91, "step": 14430, "token_acc": 0.9540636042402827, "train_speed(iter/s)": 1.486686 }, { "epoch": 0.6184396555417506, "grad_norm": 0.5923617482185364, "learning_rate": 9.627245475455265e-05, "loss": 0.3618467807769775, "memory(GiB)": 61.91, "step": 14435, "token_acc": 0.937888198757764, "train_speed(iter/s)": 1.486647 }, { "epoch": 0.6186538708709995, "grad_norm": 3.12911057472229, "learning_rate": 9.626990461302972e-05, "loss": 0.3333358526229858, "memory(GiB)": 61.91, "step": 14440, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.486688 }, { "epoch": 0.6188680862002485, "grad_norm": 3.0926883220672607, "learning_rate": 9.626735363328405e-05, "loss": 0.3877788782119751, "memory(GiB)": 61.91, "step": 14445, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.486669 }, { "epoch": 0.6190823015294975, "grad_norm": 3.95113205909729, "learning_rate": 9.626480181536188e-05, "loss": 0.7216397285461426, "memory(GiB)": 61.91, "step": 14450, "token_acc": 0.8459016393442623, "train_speed(iter/s)": 1.486633 }, { "epoch": 0.6192965168587464, "grad_norm": 3.7041571140289307, "learning_rate": 9.626224915930943e-05, "loss": 0.40782723426818845, "memory(GiB)": 61.91, "step": 14455, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.486633 }, { "epoch": 0.6195107321879953, "grad_norm": 4.347240447998047, "learning_rate": 9.625969566517294e-05, "loss": 0.6652137756347656, "memory(GiB)": 61.91, "step": 14460, "token_acc": 0.8470948012232415, "train_speed(iter/s)": 1.486632 }, { "epoch": 0.6197249475172444, "grad_norm": 0.31364306807518005, "learning_rate": 9.625714133299866e-05, "loss": 0.2773986101150513, "memory(GiB)": 61.91, "step": 14465, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.486612 }, { "epoch": 0.6199391628464933, "grad_norm": 3.6125991344451904, "learning_rate": 9.625458616283289e-05, "loss": 0.4420825481414795, "memory(GiB)": 61.91, "step": 14470, "token_acc": 0.9021406727828746, "train_speed(iter/s)": 1.486602 }, { "epoch": 0.6201533781757422, "grad_norm": 2.0265719890594482, "learning_rate": 9.625203015472189e-05, "loss": 0.2913269758224487, "memory(GiB)": 61.91, "step": 14475, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.486615 }, { "epoch": 0.6203675935049913, "grad_norm": 3.5978167057037354, "learning_rate": 9.624947330871197e-05, "loss": 0.30154991149902344, "memory(GiB)": 61.91, "step": 14480, "token_acc": 0.933993399339934, "train_speed(iter/s)": 1.486638 }, { "epoch": 0.6205818088342402, "grad_norm": 1.7711743116378784, "learning_rate": 9.624691562484945e-05, "loss": 0.3784898519515991, "memory(GiB)": 61.91, "step": 14485, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.48663 }, { "epoch": 0.6207960241634891, "grad_norm": 2.496685743331909, "learning_rate": 9.624435710318067e-05, "loss": 0.5250838279724122, "memory(GiB)": 61.91, "step": 14490, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.486556 }, { "epoch": 0.6210102394927381, "grad_norm": 3.0175371170043945, "learning_rate": 9.6241797743752e-05, "loss": 0.2517117738723755, "memory(GiB)": 61.91, "step": 14495, "token_acc": 0.941747572815534, "train_speed(iter/s)": 1.486562 }, { "epoch": 0.6212244548219871, "grad_norm": 6.15025520324707, "learning_rate": 9.623923754660977e-05, "loss": 0.4684727191925049, "memory(GiB)": 61.91, "step": 14500, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.486589 }, { "epoch": 0.6212244548219871, "eval_loss": 2.817950963973999, "eval_runtime": 13.2356, "eval_samples_per_second": 7.555, "eval_steps_per_second": 7.555, "eval_token_acc": 0.43636363636363634, "step": 14500 }, { "epoch": 0.621438670151236, "grad_norm": 1.7428393363952637, "learning_rate": 9.623667651180039e-05, "loss": 0.2028125286102295, "memory(GiB)": 61.91, "step": 14505, "token_acc": 0.56071076011846, "train_speed(iter/s)": 1.484445 }, { "epoch": 0.621652885480485, "grad_norm": 2.4227089881896973, "learning_rate": 9.623411463937022e-05, "loss": 0.49710450172424314, "memory(GiB)": 61.91, "step": 14510, "token_acc": 0.8702290076335878, "train_speed(iter/s)": 1.484427 }, { "epoch": 0.621867100809734, "grad_norm": 4.13482666015625, "learning_rate": 9.62315519293657e-05, "loss": 0.4304450511932373, "memory(GiB)": 61.91, "step": 14515, "token_acc": 0.9044117647058824, "train_speed(iter/s)": 1.48441 }, { "epoch": 0.6220813161389829, "grad_norm": 4.8451690673828125, "learning_rate": 9.622898838183325e-05, "loss": 0.6394207000732421, "memory(GiB)": 61.91, "step": 14520, "token_acc": 0.8637992831541219, "train_speed(iter/s)": 1.484391 }, { "epoch": 0.6222955314682319, "grad_norm": 3.1908092498779297, "learning_rate": 9.622642399681931e-05, "loss": 0.5209192752838134, "memory(GiB)": 61.91, "step": 14525, "token_acc": 0.9125, "train_speed(iter/s)": 1.484406 }, { "epoch": 0.6225097467974808, "grad_norm": 3.3763744831085205, "learning_rate": 9.622385877437034e-05, "loss": 0.6407839298248291, "memory(GiB)": 61.91, "step": 14530, "token_acc": 0.8440366972477065, "train_speed(iter/s)": 1.484462 }, { "epoch": 0.6227239621267298, "grad_norm": 2.7645857334136963, "learning_rate": 9.622129271453281e-05, "loss": 0.2966806173324585, "memory(GiB)": 61.91, "step": 14535, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.484465 }, { "epoch": 0.6229381774559788, "grad_norm": 2.8770060539245605, "learning_rate": 9.62187258173532e-05, "loss": 0.3563460111618042, "memory(GiB)": 61.91, "step": 14540, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.484475 }, { "epoch": 0.6231523927852277, "grad_norm": 3.1781489849090576, "learning_rate": 9.621615808287801e-05, "loss": 0.3735190868377686, "memory(GiB)": 61.91, "step": 14545, "token_acc": 0.9373134328358209, "train_speed(iter/s)": 1.484441 }, { "epoch": 0.6233666081144766, "grad_norm": 4.3358941078186035, "learning_rate": 9.621358951115377e-05, "loss": 0.44463281631469725, "memory(GiB)": 61.91, "step": 14550, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.484447 }, { "epoch": 0.6235808234437257, "grad_norm": 3.709871768951416, "learning_rate": 9.6211020102227e-05, "loss": 0.39681448936462405, "memory(GiB)": 61.91, "step": 14555, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.484445 }, { "epoch": 0.6237950387729746, "grad_norm": 5.341330051422119, "learning_rate": 9.620844985614425e-05, "loss": 0.21898908615112306, "memory(GiB)": 61.91, "step": 14560, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.484434 }, { "epoch": 0.6240092541022235, "grad_norm": 1.4658693075180054, "learning_rate": 9.620587877295209e-05, "loss": 0.3221198797225952, "memory(GiB)": 61.91, "step": 14565, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.484449 }, { "epoch": 0.6242234694314726, "grad_norm": 3.5065741539001465, "learning_rate": 9.620330685269707e-05, "loss": 0.3495811939239502, "memory(GiB)": 61.91, "step": 14570, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.484444 }, { "epoch": 0.6244376847607215, "grad_norm": 1.926450252532959, "learning_rate": 9.620073409542583e-05, "loss": 0.5926002025604248, "memory(GiB)": 61.91, "step": 14575, "token_acc": 0.8776223776223776, "train_speed(iter/s)": 1.48449 }, { "epoch": 0.6246519000899704, "grad_norm": 5.145088195800781, "learning_rate": 9.619816050118494e-05, "loss": 0.3203765392303467, "memory(GiB)": 61.91, "step": 14580, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.484482 }, { "epoch": 0.6248661154192194, "grad_norm": 4.8615546226501465, "learning_rate": 9.619558607002103e-05, "loss": 0.44176111221313474, "memory(GiB)": 61.91, "step": 14585, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.484508 }, { "epoch": 0.6250803307484684, "grad_norm": 1.7054017782211304, "learning_rate": 9.619301080198075e-05, "loss": 0.4668544292449951, "memory(GiB)": 61.91, "step": 14590, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.484512 }, { "epoch": 0.6252945460777173, "grad_norm": 4.048427104949951, "learning_rate": 9.619043469711074e-05, "loss": 0.46843528747558594, "memory(GiB)": 61.91, "step": 14595, "token_acc": 0.8854748603351955, "train_speed(iter/s)": 1.484495 }, { "epoch": 0.6255087614069663, "grad_norm": 5.763453006744385, "learning_rate": 9.618785775545769e-05, "loss": 0.2939110517501831, "memory(GiB)": 61.91, "step": 14600, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.484479 }, { "epoch": 0.6257229767362152, "grad_norm": 1.7419207096099854, "learning_rate": 9.618527997706825e-05, "loss": 0.6177469730377197, "memory(GiB)": 61.91, "step": 14605, "token_acc": 0.8725490196078431, "train_speed(iter/s)": 1.484472 }, { "epoch": 0.6259371920654642, "grad_norm": 2.4443984031677246, "learning_rate": 9.618270136198915e-05, "loss": 0.497251558303833, "memory(GiB)": 61.91, "step": 14610, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.484526 }, { "epoch": 0.6261514073947132, "grad_norm": 3.1425395011901855, "learning_rate": 9.61801219102671e-05, "loss": 0.6767574310302734, "memory(GiB)": 61.91, "step": 14615, "token_acc": 0.8618181818181818, "train_speed(iter/s)": 1.484577 }, { "epoch": 0.6263656227239621, "grad_norm": 2.114856719970703, "learning_rate": 9.617754162194881e-05, "loss": 0.3845386028289795, "memory(GiB)": 61.91, "step": 14620, "token_acc": 0.909967845659164, "train_speed(iter/s)": 1.484583 }, { "epoch": 0.626579838053211, "grad_norm": 3.8917486667633057, "learning_rate": 9.617496049708103e-05, "loss": 0.42268962860107423, "memory(GiB)": 61.91, "step": 14625, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.484616 }, { "epoch": 0.6267940533824601, "grad_norm": 1.5894486904144287, "learning_rate": 9.617237853571054e-05, "loss": 0.2914065599441528, "memory(GiB)": 61.91, "step": 14630, "token_acc": 0.9250720461095101, "train_speed(iter/s)": 1.484606 }, { "epoch": 0.627008268711709, "grad_norm": 3.577294111251831, "learning_rate": 9.61697957378841e-05, "loss": 0.4904576301574707, "memory(GiB)": 61.91, "step": 14635, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.484579 }, { "epoch": 0.6272224840409579, "grad_norm": 7.394708633422852, "learning_rate": 9.616721210364849e-05, "loss": 0.30845818519592283, "memory(GiB)": 61.91, "step": 14640, "token_acc": 0.9313432835820895, "train_speed(iter/s)": 1.48459 }, { "epoch": 0.627436699370207, "grad_norm": 4.052779674530029, "learning_rate": 9.616462763305052e-05, "loss": 0.21387763023376466, "memory(GiB)": 61.91, "step": 14645, "token_acc": 0.9568627450980393, "train_speed(iter/s)": 1.484597 }, { "epoch": 0.6276509146994559, "grad_norm": 3.3255093097686768, "learning_rate": 9.616204232613701e-05, "loss": 0.39247024059295654, "memory(GiB)": 61.91, "step": 14650, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.48463 }, { "epoch": 0.6278651300287048, "grad_norm": 4.059135913848877, "learning_rate": 9.615945618295483e-05, "loss": 0.3055412292480469, "memory(GiB)": 61.91, "step": 14655, "token_acc": 0.924901185770751, "train_speed(iter/s)": 1.484631 }, { "epoch": 0.6280793453579538, "grad_norm": 1.5347957611083984, "learning_rate": 9.615686920355077e-05, "loss": 0.44037256240844724, "memory(GiB)": 61.91, "step": 14660, "token_acc": 0.9022556390977443, "train_speed(iter/s)": 1.484637 }, { "epoch": 0.6282935606872028, "grad_norm": 3.0786752700805664, "learning_rate": 9.615428138797174e-05, "loss": 0.45560755729675295, "memory(GiB)": 61.91, "step": 14665, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.484633 }, { "epoch": 0.6285077760164517, "grad_norm": 3.1524155139923096, "learning_rate": 9.61516927362646e-05, "loss": 0.5191938877105713, "memory(GiB)": 61.91, "step": 14670, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.484598 }, { "epoch": 0.6287219913457007, "grad_norm": 3.246070623397827, "learning_rate": 9.614910324847626e-05, "loss": 0.3735840320587158, "memory(GiB)": 61.91, "step": 14675, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.484598 }, { "epoch": 0.6289362066749496, "grad_norm": 2.6072261333465576, "learning_rate": 9.614651292465361e-05, "loss": 0.4132176399230957, "memory(GiB)": 61.91, "step": 14680, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.484627 }, { "epoch": 0.6291504220041986, "grad_norm": 2.251004219055176, "learning_rate": 9.614392176484359e-05, "loss": 0.24217648506164552, "memory(GiB)": 61.91, "step": 14685, "token_acc": 0.959409594095941, "train_speed(iter/s)": 1.484617 }, { "epoch": 0.6293646373334476, "grad_norm": 2.780733823776245, "learning_rate": 9.614132976909316e-05, "loss": 0.4391185283660889, "memory(GiB)": 61.91, "step": 14690, "token_acc": 0.9012345679012346, "train_speed(iter/s)": 1.484605 }, { "epoch": 0.6295788526626965, "grad_norm": 1.6242775917053223, "learning_rate": 9.613873693744924e-05, "loss": 0.2058866500854492, "memory(GiB)": 61.91, "step": 14695, "token_acc": 0.946360153256705, "train_speed(iter/s)": 1.484666 }, { "epoch": 0.6297930679919455, "grad_norm": 3.2150087356567383, "learning_rate": 9.613614326995882e-05, "loss": 0.4322181224822998, "memory(GiB)": 61.91, "step": 14700, "token_acc": 0.9153846153846154, "train_speed(iter/s)": 1.484675 }, { "epoch": 0.6300072833211945, "grad_norm": 4.356142044067383, "learning_rate": 9.613354876666889e-05, "loss": 0.46286654472351074, "memory(GiB)": 61.91, "step": 14705, "token_acc": 0.8817891373801917, "train_speed(iter/s)": 1.484677 }, { "epoch": 0.6302214986504434, "grad_norm": 2.106816053390503, "learning_rate": 9.613095342762646e-05, "loss": 0.5676737308502198, "memory(GiB)": 61.91, "step": 14710, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.484753 }, { "epoch": 0.6304357139796923, "grad_norm": 2.594012498855591, "learning_rate": 9.612835725287852e-05, "loss": 0.29802157878875735, "memory(GiB)": 61.91, "step": 14715, "token_acc": 0.9331103678929766, "train_speed(iter/s)": 1.484813 }, { "epoch": 0.6306499293089414, "grad_norm": 3.064887762069702, "learning_rate": 9.61257602424721e-05, "loss": 0.3664767503738403, "memory(GiB)": 61.91, "step": 14720, "token_acc": 0.9363957597173145, "train_speed(iter/s)": 1.484791 }, { "epoch": 0.6308641446381903, "grad_norm": 1.2812520265579224, "learning_rate": 9.612316239645429e-05, "loss": 0.37951207160949707, "memory(GiB)": 61.91, "step": 14725, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.484832 }, { "epoch": 0.6310783599674392, "grad_norm": 4.980966091156006, "learning_rate": 9.612056371487213e-05, "loss": 0.4036360740661621, "memory(GiB)": 61.91, "step": 14730, "token_acc": 0.9225806451612903, "train_speed(iter/s)": 1.484811 }, { "epoch": 0.6312925752966883, "grad_norm": 1.5435926914215088, "learning_rate": 9.611796419777266e-05, "loss": 0.21831693649291992, "memory(GiB)": 61.91, "step": 14735, "token_acc": 0.959866220735786, "train_speed(iter/s)": 1.484858 }, { "epoch": 0.6315067906259372, "grad_norm": 2.822359323501587, "learning_rate": 9.611536384520303e-05, "loss": 0.27899553775787356, "memory(GiB)": 61.91, "step": 14740, "token_acc": 0.9371428571428572, "train_speed(iter/s)": 1.484841 }, { "epoch": 0.6317210059551861, "grad_norm": 2.7643606662750244, "learning_rate": 9.611276265721032e-05, "loss": 0.3614768981933594, "memory(GiB)": 61.91, "step": 14745, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.484814 }, { "epoch": 0.6319352212844351, "grad_norm": 2.053511142730713, "learning_rate": 9.611016063384168e-05, "loss": 0.26288599967956544, "memory(GiB)": 61.91, "step": 14750, "token_acc": 0.9607843137254902, "train_speed(iter/s)": 1.484787 }, { "epoch": 0.6321494366136841, "grad_norm": 3.419545888900757, "learning_rate": 9.61075577751442e-05, "loss": 0.47572989463806153, "memory(GiB)": 61.91, "step": 14755, "token_acc": 0.9072847682119205, "train_speed(iter/s)": 1.484816 }, { "epoch": 0.6323636519429331, "grad_norm": 4.687938213348389, "learning_rate": 9.610495408116507e-05, "loss": 0.7010627746582031, "memory(GiB)": 61.91, "step": 14760, "token_acc": 0.8464163822525598, "train_speed(iter/s)": 1.484818 }, { "epoch": 0.632577867272182, "grad_norm": 3.0458271503448486, "learning_rate": 9.610234955195143e-05, "loss": 0.4775047302246094, "memory(GiB)": 61.91, "step": 14765, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.484805 }, { "epoch": 0.6327920826014309, "grad_norm": 7.711501598358154, "learning_rate": 9.60997441875505e-05, "loss": 0.45972452163696287, "memory(GiB)": 61.91, "step": 14770, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.48484 }, { "epoch": 0.63300629793068, "grad_norm": 3.723446846008301, "learning_rate": 9.609713798800943e-05, "loss": 0.5391488552093506, "memory(GiB)": 61.91, "step": 14775, "token_acc": 0.8893280632411067, "train_speed(iter/s)": 1.484843 }, { "epoch": 0.6332205132599289, "grad_norm": 3.70721435546875, "learning_rate": 9.609453095337548e-05, "loss": 0.432661247253418, "memory(GiB)": 61.91, "step": 14780, "token_acc": 0.8915254237288136, "train_speed(iter/s)": 1.484936 }, { "epoch": 0.6334347285891778, "grad_norm": 4.185970306396484, "learning_rate": 9.609192308369588e-05, "loss": 0.3017240047454834, "memory(GiB)": 61.91, "step": 14785, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.48497 }, { "epoch": 0.6336489439184269, "grad_norm": 0.8727532625198364, "learning_rate": 9.608931437901782e-05, "loss": 0.14391974210739136, "memory(GiB)": 61.91, "step": 14790, "token_acc": 0.9595959595959596, "train_speed(iter/s)": 1.484965 }, { "epoch": 0.6338631592476758, "grad_norm": 3.9483530521392822, "learning_rate": 9.608670483938862e-05, "loss": 0.3390128374099731, "memory(GiB)": 61.91, "step": 14795, "token_acc": 0.9233576642335767, "train_speed(iter/s)": 1.484956 }, { "epoch": 0.6340773745769247, "grad_norm": 3.3456575870513916, "learning_rate": 9.608409446485553e-05, "loss": 0.28879656791687014, "memory(GiB)": 61.91, "step": 14800, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.484965 }, { "epoch": 0.6342915899061737, "grad_norm": 0.9742447733879089, "learning_rate": 9.608148325546583e-05, "loss": 0.3231043338775635, "memory(GiB)": 61.91, "step": 14805, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.484984 }, { "epoch": 0.6345058052354227, "grad_norm": 3.489210844039917, "learning_rate": 9.607887121126684e-05, "loss": 0.42917451858520506, "memory(GiB)": 61.91, "step": 14810, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.484966 }, { "epoch": 0.6347200205646716, "grad_norm": 3.688579559326172, "learning_rate": 9.607625833230586e-05, "loss": 0.6383449077606201, "memory(GiB)": 61.91, "step": 14815, "token_acc": 0.8740157480314961, "train_speed(iter/s)": 1.484956 }, { "epoch": 0.6349342358939206, "grad_norm": 2.410067081451416, "learning_rate": 9.607364461863024e-05, "loss": 0.40448474884033203, "memory(GiB)": 61.91, "step": 14820, "token_acc": 0.9100719424460432, "train_speed(iter/s)": 1.484968 }, { "epoch": 0.6351484512231695, "grad_norm": 3.793860673904419, "learning_rate": 9.607103007028733e-05, "loss": 0.44420762062072755, "memory(GiB)": 61.91, "step": 14825, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.484969 }, { "epoch": 0.6353626665524185, "grad_norm": 3.4886090755462646, "learning_rate": 9.606841468732451e-05, "loss": 0.35037384033203123, "memory(GiB)": 61.91, "step": 14830, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.484993 }, { "epoch": 0.6355768818816675, "grad_norm": 2.152660608291626, "learning_rate": 9.606579846978912e-05, "loss": 0.7115180969238282, "memory(GiB)": 61.91, "step": 14835, "token_acc": 0.8560311284046692, "train_speed(iter/s)": 1.48502 }, { "epoch": 0.6357910972109164, "grad_norm": 2.227879524230957, "learning_rate": 9.606318141772858e-05, "loss": 0.27489328384399414, "memory(GiB)": 61.91, "step": 14840, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.485045 }, { "epoch": 0.6360053125401653, "grad_norm": 3.693131923675537, "learning_rate": 9.606056353119031e-05, "loss": 0.4243523120880127, "memory(GiB)": 61.91, "step": 14845, "token_acc": 0.8869257950530035, "train_speed(iter/s)": 1.485022 }, { "epoch": 0.6362195278694144, "grad_norm": 0.7422344088554382, "learning_rate": 9.60579448102217e-05, "loss": 0.27050397396087644, "memory(GiB)": 61.91, "step": 14850, "token_acc": 0.95, "train_speed(iter/s)": 1.485018 }, { "epoch": 0.6364337431986633, "grad_norm": 5.085533618927002, "learning_rate": 9.605532525487024e-05, "loss": 0.583841609954834, "memory(GiB)": 61.91, "step": 14855, "token_acc": 0.8931297709923665, "train_speed(iter/s)": 1.484989 }, { "epoch": 0.6366479585279122, "grad_norm": 3.134425401687622, "learning_rate": 9.605270486518335e-05, "loss": 0.34767398834228513, "memory(GiB)": 61.91, "step": 14860, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.48497 }, { "epoch": 0.6368621738571613, "grad_norm": 3.1537108421325684, "learning_rate": 9.60500836412085e-05, "loss": 0.613176965713501, "memory(GiB)": 61.91, "step": 14865, "token_acc": 0.8738170347003155, "train_speed(iter/s)": 1.484987 }, { "epoch": 0.6370763891864102, "grad_norm": 1.8521831035614014, "learning_rate": 9.604746158299319e-05, "loss": 0.41345062255859377, "memory(GiB)": 61.91, "step": 14870, "token_acc": 0.9105058365758755, "train_speed(iter/s)": 1.484993 }, { "epoch": 0.6372906045156591, "grad_norm": 2.67592191696167, "learning_rate": 9.604483869058492e-05, "loss": 0.7560207366943359, "memory(GiB)": 61.91, "step": 14875, "token_acc": 0.8396946564885496, "train_speed(iter/s)": 1.484971 }, { "epoch": 0.6375048198449081, "grad_norm": 5.289083957672119, "learning_rate": 9.604221496403121e-05, "loss": 0.5849059104919434, "memory(GiB)": 61.91, "step": 14880, "token_acc": 0.8776223776223776, "train_speed(iter/s)": 1.48496 }, { "epoch": 0.6377190351741571, "grad_norm": 15.396535873413086, "learning_rate": 9.603959040337958e-05, "loss": 0.3866638422012329, "memory(GiB)": 61.91, "step": 14885, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.484964 }, { "epoch": 0.637933250503406, "grad_norm": 2.2538225650787354, "learning_rate": 9.603696500867758e-05, "loss": 0.40776629447937013, "memory(GiB)": 61.91, "step": 14890, "token_acc": 0.9118541033434651, "train_speed(iter/s)": 1.485054 }, { "epoch": 0.638147465832655, "grad_norm": 6.382772445678711, "learning_rate": 9.603433877997278e-05, "loss": 0.8178947448730469, "memory(GiB)": 61.91, "step": 14895, "token_acc": 0.8308823529411765, "train_speed(iter/s)": 1.48516 }, { "epoch": 0.638361681161904, "grad_norm": 4.55464506149292, "learning_rate": 9.603171171731273e-05, "loss": 0.3245693206787109, "memory(GiB)": 61.91, "step": 14900, "token_acc": 0.9377593360995851, "train_speed(iter/s)": 1.485191 }, { "epoch": 0.6385758964911529, "grad_norm": 1.177608609199524, "learning_rate": 9.602908382074508e-05, "loss": 0.2794328212738037, "memory(GiB)": 61.91, "step": 14905, "token_acc": 0.9415384615384615, "train_speed(iter/s)": 1.485192 }, { "epoch": 0.6387901118204019, "grad_norm": 1.1164549589157104, "learning_rate": 9.602645509031735e-05, "loss": 0.340386962890625, "memory(GiB)": 61.91, "step": 14910, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.485159 }, { "epoch": 0.6390043271496508, "grad_norm": 2.795476198196411, "learning_rate": 9.602382552607724e-05, "loss": 0.6486668109893798, "memory(GiB)": 61.91, "step": 14915, "token_acc": 0.8861538461538462, "train_speed(iter/s)": 1.485124 }, { "epoch": 0.6392185424788998, "grad_norm": 2.877755641937256, "learning_rate": 9.602119512807234e-05, "loss": 0.14693330526351928, "memory(GiB)": 61.91, "step": 14920, "token_acc": 0.9644128113879004, "train_speed(iter/s)": 1.485114 }, { "epoch": 0.6394327578081488, "grad_norm": 6.204319953918457, "learning_rate": 9.601856389635034e-05, "loss": 0.4558413982391357, "memory(GiB)": 61.91, "step": 14925, "token_acc": 0.8874172185430463, "train_speed(iter/s)": 1.485099 }, { "epoch": 0.6396469731373977, "grad_norm": 1.7430015802383423, "learning_rate": 9.601593183095887e-05, "loss": 0.2009504795074463, "memory(GiB)": 61.91, "step": 14930, "token_acc": 0.9605263157894737, "train_speed(iter/s)": 1.485077 }, { "epoch": 0.6398611884666466, "grad_norm": 3.99906849861145, "learning_rate": 9.601329893194563e-05, "loss": 0.550875473022461, "memory(GiB)": 61.91, "step": 14935, "token_acc": 0.8865030674846626, "train_speed(iter/s)": 1.485051 }, { "epoch": 0.6400754037958957, "grad_norm": 4.578465938568115, "learning_rate": 9.601066519935832e-05, "loss": 0.5483016014099121, "memory(GiB)": 61.91, "step": 14940, "token_acc": 0.8839285714285714, "train_speed(iter/s)": 1.485071 }, { "epoch": 0.6402896191251446, "grad_norm": 1.4683079719543457, "learning_rate": 9.600803063324465e-05, "loss": 0.48709959983825685, "memory(GiB)": 61.91, "step": 14945, "token_acc": 0.9007633587786259, "train_speed(iter/s)": 1.485054 }, { "epoch": 0.6405038344543935, "grad_norm": 4.422609329223633, "learning_rate": 9.600539523365234e-05, "loss": 0.16610996723175048, "memory(GiB)": 61.91, "step": 14950, "token_acc": 0.9619377162629758, "train_speed(iter/s)": 1.48502 }, { "epoch": 0.6407180497836426, "grad_norm": 7.860347270965576, "learning_rate": 9.600275900062915e-05, "loss": 0.6171211242675781, "memory(GiB)": 61.91, "step": 14955, "token_acc": 0.875, "train_speed(iter/s)": 1.485039 }, { "epoch": 0.6409322651128915, "grad_norm": 6.96934175491333, "learning_rate": 9.600012193422281e-05, "loss": 0.3641418695449829, "memory(GiB)": 61.91, "step": 14960, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.484992 }, { "epoch": 0.6411464804421404, "grad_norm": 3.244460105895996, "learning_rate": 9.599748403448113e-05, "loss": 0.20010690689086913, "memory(GiB)": 61.91, "step": 14965, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.484966 }, { "epoch": 0.6413606957713894, "grad_norm": 2.1829843521118164, "learning_rate": 9.599484530145186e-05, "loss": 0.2176499843597412, "memory(GiB)": 61.91, "step": 14970, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.484953 }, { "epoch": 0.6415749111006384, "grad_norm": 3.357576608657837, "learning_rate": 9.599220573518282e-05, "loss": 0.5286779880523682, "memory(GiB)": 61.91, "step": 14975, "token_acc": 0.8876811594202898, "train_speed(iter/s)": 1.484953 }, { "epoch": 0.6417891264298873, "grad_norm": 3.9801621437072754, "learning_rate": 9.598956533572183e-05, "loss": 0.4387080192565918, "memory(GiB)": 61.91, "step": 14980, "token_acc": 0.8990536277602523, "train_speed(iter/s)": 1.484996 }, { "epoch": 0.6420033417591363, "grad_norm": 0.1717825084924698, "learning_rate": 9.598692410311674e-05, "loss": 0.2945881366729736, "memory(GiB)": 61.91, "step": 14985, "token_acc": 0.9409448818897638, "train_speed(iter/s)": 1.485007 }, { "epoch": 0.6422175570883852, "grad_norm": 3.919212818145752, "learning_rate": 9.598428203741536e-05, "loss": 0.40082559585571287, "memory(GiB)": 61.91, "step": 14990, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.484984 }, { "epoch": 0.6424317724176342, "grad_norm": 3.669501304626465, "learning_rate": 9.598163913866558e-05, "loss": 0.3335289716720581, "memory(GiB)": 61.91, "step": 14995, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.484991 }, { "epoch": 0.6426459877468832, "grad_norm": 6.054754257202148, "learning_rate": 9.597899540691527e-05, "loss": 0.4841428279876709, "memory(GiB)": 61.91, "step": 15000, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.484968 }, { "epoch": 0.6426459877468832, "eval_loss": 2.749873399734497, "eval_runtime": 13.3422, "eval_samples_per_second": 7.495, "eval_steps_per_second": 7.495, "eval_token_acc": 0.41746411483253587, "step": 15000 }, { "epoch": 0.6428602030761321, "grad_norm": 2.8434858322143555, "learning_rate": 9.597635084221234e-05, "loss": 0.2857626676559448, "memory(GiB)": 61.91, "step": 15005, "token_acc": 0.5464285714285714, "train_speed(iter/s)": 1.482867 }, { "epoch": 0.643074418405381, "grad_norm": 2.8151803016662598, "learning_rate": 9.597370544460467e-05, "loss": 0.34791200160980223, "memory(GiB)": 61.91, "step": 15010, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.482838 }, { "epoch": 0.6432886337346301, "grad_norm": 0.15423570573329926, "learning_rate": 9.59710592141402e-05, "loss": 0.31936054229736327, "memory(GiB)": 61.91, "step": 15015, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.482917 }, { "epoch": 0.643502849063879, "grad_norm": 0.060921043157577515, "learning_rate": 9.596841215086689e-05, "loss": 0.20055902004241943, "memory(GiB)": 61.91, "step": 15020, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.483014 }, { "epoch": 0.6437170643931279, "grad_norm": 3.9596028327941895, "learning_rate": 9.596576425483264e-05, "loss": 0.16176793575286866, "memory(GiB)": 61.91, "step": 15025, "token_acc": 0.956989247311828, "train_speed(iter/s)": 1.482998 }, { "epoch": 0.643931279722377, "grad_norm": 2.315037727355957, "learning_rate": 9.596311552608547e-05, "loss": 0.23392224311828613, "memory(GiB)": 61.91, "step": 15030, "token_acc": 0.9463276836158192, "train_speed(iter/s)": 1.482967 }, { "epoch": 0.6441454950516259, "grad_norm": 3.0920605659484863, "learning_rate": 9.596046596467334e-05, "loss": 0.5604113101959228, "memory(GiB)": 61.91, "step": 15035, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.483007 }, { "epoch": 0.6443597103808748, "grad_norm": 6.976906776428223, "learning_rate": 9.595781557064427e-05, "loss": 0.7252992630004883, "memory(GiB)": 61.91, "step": 15040, "token_acc": 0.8481012658227848, "train_speed(iter/s)": 1.483053 }, { "epoch": 0.6445739257101238, "grad_norm": 1.8767651319503784, "learning_rate": 9.595516434404624e-05, "loss": 0.5165930271148682, "memory(GiB)": 61.91, "step": 15045, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.483051 }, { "epoch": 0.6447881410393728, "grad_norm": 2.6300089359283447, "learning_rate": 9.59525122849273e-05, "loss": 0.27689833641052247, "memory(GiB)": 61.91, "step": 15050, "token_acc": 0.9372937293729373, "train_speed(iter/s)": 1.483025 }, { "epoch": 0.6450023563686217, "grad_norm": 4.597753047943115, "learning_rate": 9.594985939333549e-05, "loss": 0.43529610633850097, "memory(GiB)": 61.91, "step": 15055, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.483041 }, { "epoch": 0.6452165716978707, "grad_norm": 3.9214320182800293, "learning_rate": 9.594720566931889e-05, "loss": 0.26026239395141604, "memory(GiB)": 61.91, "step": 15060, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.483059 }, { "epoch": 0.6454307870271196, "grad_norm": 2.1826767921447754, "learning_rate": 9.594455111292555e-05, "loss": 0.6001876354217529, "memory(GiB)": 61.91, "step": 15065, "token_acc": 0.8807692307692307, "train_speed(iter/s)": 1.483057 }, { "epoch": 0.6456450023563686, "grad_norm": 3.546027421951294, "learning_rate": 9.594189572420356e-05, "loss": 0.3328744888305664, "memory(GiB)": 61.91, "step": 15070, "token_acc": 0.9187675070028011, "train_speed(iter/s)": 1.483029 }, { "epoch": 0.6458592176856176, "grad_norm": 1.4402191638946533, "learning_rate": 9.593923950320104e-05, "loss": 0.29089984893798826, "memory(GiB)": 61.91, "step": 15075, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.482994 }, { "epoch": 0.6460734330148665, "grad_norm": 3.9519665241241455, "learning_rate": 9.593658244996609e-05, "loss": 0.41818943023681643, "memory(GiB)": 61.91, "step": 15080, "token_acc": 0.91, "train_speed(iter/s)": 1.482994 }, { "epoch": 0.6462876483441155, "grad_norm": 3.293963670730591, "learning_rate": 9.593392456454686e-05, "loss": 0.40402679443359374, "memory(GiB)": 61.91, "step": 15085, "token_acc": 0.9010238907849829, "train_speed(iter/s)": 1.483031 }, { "epoch": 0.6465018636733645, "grad_norm": 3.9125983715057373, "learning_rate": 9.59312658469915e-05, "loss": 0.3362214803695679, "memory(GiB)": 61.91, "step": 15090, "token_acc": 0.9320987654320988, "train_speed(iter/s)": 1.483021 }, { "epoch": 0.6467160790026134, "grad_norm": 3.4699180126190186, "learning_rate": 9.592860629734819e-05, "loss": 0.4789536952972412, "memory(GiB)": 61.91, "step": 15095, "token_acc": 0.8932038834951457, "train_speed(iter/s)": 1.483017 }, { "epoch": 0.6469302943318624, "grad_norm": 2.525895357131958, "learning_rate": 9.592594591566508e-05, "loss": 0.25381109714508054, "memory(GiB)": 61.91, "step": 15100, "token_acc": 0.9448051948051948, "train_speed(iter/s)": 1.483022 }, { "epoch": 0.6471445096611114, "grad_norm": 3.056114435195923, "learning_rate": 9.592328470199037e-05, "loss": 0.37546095848083494, "memory(GiB)": 61.91, "step": 15105, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.483012 }, { "epoch": 0.6473587249903603, "grad_norm": 5.900458335876465, "learning_rate": 9.592062265637227e-05, "loss": 0.2662632465362549, "memory(GiB)": 61.91, "step": 15110, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.483026 }, { "epoch": 0.6475729403196093, "grad_norm": 4.118362903594971, "learning_rate": 9.591795977885903e-05, "loss": 0.5514111042022705, "memory(GiB)": 61.91, "step": 15115, "token_acc": 0.9030303030303031, "train_speed(iter/s)": 1.483039 }, { "epoch": 0.6477871556488582, "grad_norm": 16.66381072998047, "learning_rate": 9.591529606949887e-05, "loss": 0.466886568069458, "memory(GiB)": 61.91, "step": 15120, "token_acc": 0.9007352941176471, "train_speed(iter/s)": 1.483077 }, { "epoch": 0.6480013709781072, "grad_norm": 2.8863210678100586, "learning_rate": 9.591263152834005e-05, "loss": 0.17088807821273805, "memory(GiB)": 61.91, "step": 15125, "token_acc": 0.9691780821917808, "train_speed(iter/s)": 1.483105 }, { "epoch": 0.6482155863073562, "grad_norm": 3.754732370376587, "learning_rate": 9.590996615543084e-05, "loss": 0.22885172367095946, "memory(GiB)": 61.91, "step": 15130, "token_acc": 0.948936170212766, "train_speed(iter/s)": 1.483116 }, { "epoch": 0.6484298016366051, "grad_norm": 2.608783006668091, "learning_rate": 9.590729995081953e-05, "loss": 0.4702596664428711, "memory(GiB)": 61.91, "step": 15135, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.483104 }, { "epoch": 0.648644016965854, "grad_norm": 0.23440741002559662, "learning_rate": 9.590463291455442e-05, "loss": 0.17348248958587648, "memory(GiB)": 61.91, "step": 15140, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.483083 }, { "epoch": 0.6488582322951031, "grad_norm": 3.4966940879821777, "learning_rate": 9.59019650466838e-05, "loss": 0.5444223403930664, "memory(GiB)": 61.91, "step": 15145, "token_acc": 0.8938775510204081, "train_speed(iter/s)": 1.483141 }, { "epoch": 0.649072447624352, "grad_norm": 2.6982882022857666, "learning_rate": 9.589929634725605e-05, "loss": 0.14031593799591063, "memory(GiB)": 61.91, "step": 15150, "token_acc": 0.9767441860465116, "train_speed(iter/s)": 1.48314 }, { "epoch": 0.6492866629536009, "grad_norm": 3.0679030418395996, "learning_rate": 9.589662681631948e-05, "loss": 0.4532435417175293, "memory(GiB)": 61.91, "step": 15155, "token_acc": 0.8924731182795699, "train_speed(iter/s)": 1.48309 }, { "epoch": 0.64950087828285, "grad_norm": 2.7940385341644287, "learning_rate": 9.589395645392245e-05, "loss": 0.43416681289672854, "memory(GiB)": 61.91, "step": 15160, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.483092 }, { "epoch": 0.6497150936120989, "grad_norm": 3.3787941932678223, "learning_rate": 9.589128526011336e-05, "loss": 0.45481395721435547, "memory(GiB)": 61.91, "step": 15165, "token_acc": 0.9016949152542373, "train_speed(iter/s)": 1.483065 }, { "epoch": 0.6499293089413478, "grad_norm": 0.8866674304008484, "learning_rate": 9.588861323494058e-05, "loss": 0.490221643447876, "memory(GiB)": 61.91, "step": 15170, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.483115 }, { "epoch": 0.6501435242705969, "grad_norm": 0.25974056124687195, "learning_rate": 9.588594037845254e-05, "loss": 0.3499161958694458, "memory(GiB)": 61.91, "step": 15175, "token_acc": 0.9150943396226415, "train_speed(iter/s)": 1.483158 }, { "epoch": 0.6503577395998458, "grad_norm": 2.4710803031921387, "learning_rate": 9.588326669069763e-05, "loss": 0.45882339477539064, "memory(GiB)": 61.91, "step": 15180, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.483164 }, { "epoch": 0.6505719549290947, "grad_norm": 3.4145281314849854, "learning_rate": 9.588059217172432e-05, "loss": 0.5231476783752441, "memory(GiB)": 61.91, "step": 15185, "token_acc": 0.8838028169014085, "train_speed(iter/s)": 1.483138 }, { "epoch": 0.6507861702583437, "grad_norm": 4.653240203857422, "learning_rate": 9.587791682158102e-05, "loss": 0.4043719291687012, "memory(GiB)": 61.91, "step": 15190, "token_acc": 0.9182389937106918, "train_speed(iter/s)": 1.483145 }, { "epoch": 0.6510003855875927, "grad_norm": 2.2239253520965576, "learning_rate": 9.587524064031624e-05, "loss": 0.34597456455230713, "memory(GiB)": 61.91, "step": 15195, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.48314 }, { "epoch": 0.6512146009168416, "grad_norm": 3.719764471054077, "learning_rate": 9.587256362797842e-05, "loss": 0.36766824722290037, "memory(GiB)": 61.91, "step": 15200, "token_acc": 0.9188311688311688, "train_speed(iter/s)": 1.483129 }, { "epoch": 0.6514288162460906, "grad_norm": 4.495085716247559, "learning_rate": 9.586988578461609e-05, "loss": 0.20655531883239747, "memory(GiB)": 61.91, "step": 15205, "token_acc": 0.9507042253521126, "train_speed(iter/s)": 1.483158 }, { "epoch": 0.6516430315753395, "grad_norm": 3.1698853969573975, "learning_rate": 9.586720711027775e-05, "loss": 0.3032225131988525, "memory(GiB)": 61.91, "step": 15210, "token_acc": 0.9425287356321839, "train_speed(iter/s)": 1.48319 }, { "epoch": 0.6518572469045885, "grad_norm": 3.6391234397888184, "learning_rate": 9.586452760501193e-05, "loss": 0.38789565563201905, "memory(GiB)": 61.91, "step": 15215, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.483198 }, { "epoch": 0.6520714622338375, "grad_norm": 2.5931687355041504, "learning_rate": 9.586184726886715e-05, "loss": 0.4439659118652344, "memory(GiB)": 61.91, "step": 15220, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.483183 }, { "epoch": 0.6522856775630864, "grad_norm": 1.1138383150100708, "learning_rate": 9.5859166101892e-05, "loss": 0.3578639030456543, "memory(GiB)": 61.91, "step": 15225, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.48323 }, { "epoch": 0.6524998928923353, "grad_norm": 2.637204170227051, "learning_rate": 9.585648410413503e-05, "loss": 0.46647982597351073, "memory(GiB)": 61.91, "step": 15230, "token_acc": 0.9118773946360154, "train_speed(iter/s)": 1.48322 }, { "epoch": 0.6527141082215844, "grad_norm": 2.3995208740234375, "learning_rate": 9.585380127564484e-05, "loss": 0.33827064037322996, "memory(GiB)": 61.91, "step": 15235, "token_acc": 0.9313725490196079, "train_speed(iter/s)": 1.483256 }, { "epoch": 0.6529283235508333, "grad_norm": 2.1860058307647705, "learning_rate": 9.585111761647002e-05, "loss": 0.48485026359558103, "memory(GiB)": 61.91, "step": 15240, "token_acc": 0.8929889298892989, "train_speed(iter/s)": 1.483326 }, { "epoch": 0.6531425388800822, "grad_norm": 3.6835358142852783, "learning_rate": 9.58484331266592e-05, "loss": 0.3816518783569336, "memory(GiB)": 61.91, "step": 15245, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.483351 }, { "epoch": 0.6533567542093313, "grad_norm": 2.0366384983062744, "learning_rate": 9.5845747806261e-05, "loss": 0.2655291795730591, "memory(GiB)": 61.91, "step": 15250, "token_acc": 0.9375, "train_speed(iter/s)": 1.483368 }, { "epoch": 0.6535709695385802, "grad_norm": 1.876519799232483, "learning_rate": 9.584306165532406e-05, "loss": 0.2224421262741089, "memory(GiB)": 61.91, "step": 15255, "token_acc": 0.967741935483871, "train_speed(iter/s)": 1.483368 }, { "epoch": 0.6537851848678291, "grad_norm": 1.018646240234375, "learning_rate": 9.584037467389708e-05, "loss": 0.280442214012146, "memory(GiB)": 61.91, "step": 15260, "token_acc": 0.9375, "train_speed(iter/s)": 1.483359 }, { "epoch": 0.6539994001970781, "grad_norm": 6.785383701324463, "learning_rate": 9.583768686202869e-05, "loss": 0.36781132221221924, "memory(GiB)": 61.91, "step": 15265, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.483324 }, { "epoch": 0.6542136155263271, "grad_norm": 3.6130948066711426, "learning_rate": 9.583499821976762e-05, "loss": 0.3427476644515991, "memory(GiB)": 61.91, "step": 15270, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.483297 }, { "epoch": 0.654427830855576, "grad_norm": 2.830362319946289, "learning_rate": 9.583230874716253e-05, "loss": 0.3999899625778198, "memory(GiB)": 61.91, "step": 15275, "token_acc": 0.9155672823218998, "train_speed(iter/s)": 1.48324 }, { "epoch": 0.654642046184825, "grad_norm": 2.13804292678833, "learning_rate": 9.582961844426221e-05, "loss": 0.2818603992462158, "memory(GiB)": 61.91, "step": 15280, "token_acc": 0.939873417721519, "train_speed(iter/s)": 1.483217 }, { "epoch": 0.654856261514074, "grad_norm": 2.1057684421539307, "learning_rate": 9.582692731111535e-05, "loss": 0.3193145036697388, "memory(GiB)": 61.91, "step": 15285, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.483244 }, { "epoch": 0.6550704768433229, "grad_norm": 0.7677075862884521, "learning_rate": 9.58242353477707e-05, "loss": 0.1641088008880615, "memory(GiB)": 61.91, "step": 15290, "token_acc": 0.970873786407767, "train_speed(iter/s)": 1.483247 }, { "epoch": 0.6552846921725719, "grad_norm": 4.6232805252075195, "learning_rate": 9.582154255427705e-05, "loss": 0.4167636871337891, "memory(GiB)": 61.91, "step": 15295, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.483293 }, { "epoch": 0.6554989075018208, "grad_norm": 5.0430216789245605, "learning_rate": 9.581884893068315e-05, "loss": 0.4082638263702393, "memory(GiB)": 61.91, "step": 15300, "token_acc": 0.9196787148594378, "train_speed(iter/s)": 1.483346 }, { "epoch": 0.6557131228310698, "grad_norm": 3.114863872528076, "learning_rate": 9.581615447703784e-05, "loss": 0.65341157913208, "memory(GiB)": 61.91, "step": 15305, "token_acc": 0.8547717842323651, "train_speed(iter/s)": 1.483361 }, { "epoch": 0.6559273381603188, "grad_norm": 4.7054901123046875, "learning_rate": 9.581345919338994e-05, "loss": 0.37988481521606443, "memory(GiB)": 61.91, "step": 15310, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.483405 }, { "epoch": 0.6561415534895677, "grad_norm": 3.034242630004883, "learning_rate": 9.58107630797882e-05, "loss": 0.40912837982177735, "memory(GiB)": 61.91, "step": 15315, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.483372 }, { "epoch": 0.6563557688188166, "grad_norm": 3.6366467475891113, "learning_rate": 9.580806613628155e-05, "loss": 0.3159144163131714, "memory(GiB)": 61.91, "step": 15320, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.483337 }, { "epoch": 0.6565699841480657, "grad_norm": 3.738665819168091, "learning_rate": 9.580536836291878e-05, "loss": 0.5010700702667237, "memory(GiB)": 61.91, "step": 15325, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.483356 }, { "epoch": 0.6567841994773146, "grad_norm": 2.003697156906128, "learning_rate": 9.58026697597488e-05, "loss": 0.22606127262115477, "memory(GiB)": 61.91, "step": 15330, "token_acc": 0.947565543071161, "train_speed(iter/s)": 1.483356 }, { "epoch": 0.6569984148065635, "grad_norm": 0.3879861831665039, "learning_rate": 9.579997032682052e-05, "loss": 0.16379634141921998, "memory(GiB)": 61.91, "step": 15335, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.483404 }, { "epoch": 0.6572126301358125, "grad_norm": 0.619060218334198, "learning_rate": 9.579727006418279e-05, "loss": 0.3288090467453003, "memory(GiB)": 61.91, "step": 15340, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.483384 }, { "epoch": 0.6574268454650615, "grad_norm": 3.7311556339263916, "learning_rate": 9.579456897188455e-05, "loss": 0.17155165672302247, "memory(GiB)": 61.91, "step": 15345, "token_acc": 0.953405017921147, "train_speed(iter/s)": 1.483363 }, { "epoch": 0.6576410607943104, "grad_norm": 4.817153453826904, "learning_rate": 9.579186704997474e-05, "loss": 0.594321870803833, "memory(GiB)": 61.91, "step": 15350, "token_acc": 0.8669064748201439, "train_speed(iter/s)": 1.483403 }, { "epoch": 0.6578552761235594, "grad_norm": 7.772558689117432, "learning_rate": 9.57891642985023e-05, "loss": 0.8487940788269043, "memory(GiB)": 61.91, "step": 15355, "token_acc": 0.8181818181818182, "train_speed(iter/s)": 1.483391 }, { "epoch": 0.6580694914528084, "grad_norm": 1.4625623226165771, "learning_rate": 9.57864607175162e-05, "loss": 0.16375558376312255, "memory(GiB)": 61.91, "step": 15360, "token_acc": 0.9414225941422594, "train_speed(iter/s)": 1.483372 }, { "epoch": 0.6582837067820573, "grad_norm": 4.342894554138184, "learning_rate": 9.57837563070654e-05, "loss": 0.40346150398254393, "memory(GiB)": 61.91, "step": 15365, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.483353 }, { "epoch": 0.6584979221113063, "grad_norm": 2.8687925338745117, "learning_rate": 9.578105106719893e-05, "loss": 0.3236011743545532, "memory(GiB)": 61.91, "step": 15370, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.483372 }, { "epoch": 0.6587121374405552, "grad_norm": 0.18687939643859863, "learning_rate": 9.577834499796575e-05, "loss": 0.4040717124938965, "memory(GiB)": 61.91, "step": 15375, "token_acc": 0.900990099009901, "train_speed(iter/s)": 1.483436 }, { "epoch": 0.6589263527698042, "grad_norm": 2.195784091949463, "learning_rate": 9.577563809941492e-05, "loss": 0.270742392539978, "memory(GiB)": 61.91, "step": 15380, "token_acc": 0.9407894736842105, "train_speed(iter/s)": 1.483457 }, { "epoch": 0.6591405680990532, "grad_norm": 3.193660259246826, "learning_rate": 9.577293037159544e-05, "loss": 0.2888221740722656, "memory(GiB)": 61.91, "step": 15385, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.483466 }, { "epoch": 0.6593547834283021, "grad_norm": 5.543472766876221, "learning_rate": 9.577022181455641e-05, "loss": 0.4467325687408447, "memory(GiB)": 61.91, "step": 15390, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.483446 }, { "epoch": 0.659568998757551, "grad_norm": 8.139946937561035, "learning_rate": 9.57675124283469e-05, "loss": 0.7194681644439698, "memory(GiB)": 61.91, "step": 15395, "token_acc": 0.8896797153024911, "train_speed(iter/s)": 1.483464 }, { "epoch": 0.6597832140868001, "grad_norm": 0.5711617469787598, "learning_rate": 9.576480221301593e-05, "loss": 0.4167047023773193, "memory(GiB)": 61.91, "step": 15400, "token_acc": 0.9, "train_speed(iter/s)": 1.483442 }, { "epoch": 0.659997429416049, "grad_norm": 2.100922107696533, "learning_rate": 9.576209116861265e-05, "loss": 0.3912088871002197, "memory(GiB)": 61.91, "step": 15405, "token_acc": 0.9134860050890585, "train_speed(iter/s)": 1.48343 }, { "epoch": 0.6602116447452979, "grad_norm": 4.575545310974121, "learning_rate": 9.575937929518616e-05, "loss": 0.5527102470397949, "memory(GiB)": 61.91, "step": 15410, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.483484 }, { "epoch": 0.660425860074547, "grad_norm": 1.8660027980804443, "learning_rate": 9.575666659278559e-05, "loss": 0.26024441719055175, "memory(GiB)": 61.91, "step": 15415, "token_acc": 0.9355828220858896, "train_speed(iter/s)": 1.483501 }, { "epoch": 0.6606400754037959, "grad_norm": 1.2600493431091309, "learning_rate": 9.575395306146008e-05, "loss": 0.4911402702331543, "memory(GiB)": 61.91, "step": 15420, "token_acc": 0.9021739130434783, "train_speed(iter/s)": 1.483536 }, { "epoch": 0.6608542907330448, "grad_norm": 1.9247502088546753, "learning_rate": 9.575123870125879e-05, "loss": 0.43680343627929685, "memory(GiB)": 61.91, "step": 15425, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.48351 }, { "epoch": 0.6610685060622938, "grad_norm": 5.929807186126709, "learning_rate": 9.574852351223089e-05, "loss": 0.37536671161651614, "memory(GiB)": 61.91, "step": 15430, "token_acc": 0.9217391304347826, "train_speed(iter/s)": 1.483495 }, { "epoch": 0.6612827213915428, "grad_norm": 4.932224750518799, "learning_rate": 9.574580749442557e-05, "loss": 0.5385919094085694, "memory(GiB)": 61.91, "step": 15435, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.483493 }, { "epoch": 0.6614969367207918, "grad_norm": 1.289957880973816, "learning_rate": 9.574309064789205e-05, "loss": 0.2823874711990356, "memory(GiB)": 61.91, "step": 15440, "token_acc": 0.933993399339934, "train_speed(iter/s)": 1.483465 }, { "epoch": 0.6617111520500407, "grad_norm": 4.166236400604248, "learning_rate": 9.57403729726795e-05, "loss": 0.7183233737945557, "memory(GiB)": 61.91, "step": 15445, "token_acc": 0.8628048780487805, "train_speed(iter/s)": 1.483461 }, { "epoch": 0.6619253673792896, "grad_norm": 8.245028495788574, "learning_rate": 9.573765446883722e-05, "loss": 0.3341393232345581, "memory(GiB)": 61.91, "step": 15450, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.483448 }, { "epoch": 0.6621395827085387, "grad_norm": 7.2736592292785645, "learning_rate": 9.57349351364144e-05, "loss": 0.4201539993286133, "memory(GiB)": 61.91, "step": 15455, "token_acc": 0.9105058365758755, "train_speed(iter/s)": 1.483423 }, { "epoch": 0.6623537980377876, "grad_norm": 8.347678184509277, "learning_rate": 9.573221497546035e-05, "loss": 0.942021369934082, "memory(GiB)": 61.91, "step": 15460, "token_acc": 0.8120567375886525, "train_speed(iter/s)": 1.483439 }, { "epoch": 0.6625680133670365, "grad_norm": 3.0991435050964355, "learning_rate": 9.57294939860243e-05, "loss": 0.6733010768890381, "memory(GiB)": 61.91, "step": 15465, "token_acc": 0.8754578754578755, "train_speed(iter/s)": 1.483439 }, { "epoch": 0.6627822286962856, "grad_norm": 0.7671016454696655, "learning_rate": 9.572677216815559e-05, "loss": 0.09834669828414917, "memory(GiB)": 61.91, "step": 15470, "token_acc": 0.9760956175298805, "train_speed(iter/s)": 1.48341 }, { "epoch": 0.6629964440255345, "grad_norm": 5.753828525543213, "learning_rate": 9.572404952190349e-05, "loss": 0.3595176935195923, "memory(GiB)": 61.91, "step": 15475, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.483419 }, { "epoch": 0.6632106593547834, "grad_norm": 2.6901352405548096, "learning_rate": 9.572132604731735e-05, "loss": 0.3294682025909424, "memory(GiB)": 61.91, "step": 15480, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.483405 }, { "epoch": 0.6634248746840324, "grad_norm": 5.839317798614502, "learning_rate": 9.571860174444649e-05, "loss": 0.3485441207885742, "memory(GiB)": 61.91, "step": 15485, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.483384 }, { "epoch": 0.6636390900132814, "grad_norm": 2.522357940673828, "learning_rate": 9.571587661334028e-05, "loss": 0.5577456951141357, "memory(GiB)": 61.91, "step": 15490, "token_acc": 0.8931034482758621, "train_speed(iter/s)": 1.483432 }, { "epoch": 0.6638533053425303, "grad_norm": 3.733240842819214, "learning_rate": 9.571315065404808e-05, "loss": 0.6636384963989258, "memory(GiB)": 61.91, "step": 15495, "token_acc": 0.8461538461538461, "train_speed(iter/s)": 1.483413 }, { "epoch": 0.6640675206717793, "grad_norm": 3.231855630874634, "learning_rate": 9.571042386661928e-05, "loss": 0.20257167816162108, "memory(GiB)": 61.91, "step": 15500, "token_acc": 0.9604863221884499, "train_speed(iter/s)": 1.483391 }, { "epoch": 0.6640675206717793, "eval_loss": 2.5579657554626465, "eval_runtime": 14.015, "eval_samples_per_second": 7.135, "eval_steps_per_second": 7.135, "eval_token_acc": 0.44881889763779526, "step": 15500 }, { "epoch": 0.6642817360010282, "grad_norm": 8.471296310424805, "learning_rate": 9.570769625110325e-05, "loss": 0.4405327796936035, "memory(GiB)": 61.91, "step": 15505, "token_acc": 0.5754189944134078, "train_speed(iter/s)": 1.481298 }, { "epoch": 0.6644959513302772, "grad_norm": 13.33039379119873, "learning_rate": 9.570496780754945e-05, "loss": 0.3610804557800293, "memory(GiB)": 61.91, "step": 15510, "token_acc": 0.94, "train_speed(iter/s)": 1.481362 }, { "epoch": 0.6647101666595262, "grad_norm": 5.2048258781433105, "learning_rate": 9.570223853600727e-05, "loss": 0.5659201622009278, "memory(GiB)": 61.91, "step": 15515, "token_acc": 0.8847352024922118, "train_speed(iter/s)": 1.481424 }, { "epoch": 0.6649243819887751, "grad_norm": 3.679561138153076, "learning_rate": 9.569950843652618e-05, "loss": 0.26436195373535154, "memory(GiB)": 61.91, "step": 15520, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.481426 }, { "epoch": 0.665138597318024, "grad_norm": 3.6464450359344482, "learning_rate": 9.56967775091556e-05, "loss": 0.40215415954589845, "memory(GiB)": 61.91, "step": 15525, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.481502 }, { "epoch": 0.6653528126472731, "grad_norm": 3.2875547409057617, "learning_rate": 9.569404575394505e-05, "loss": 0.5548294067382813, "memory(GiB)": 61.91, "step": 15530, "token_acc": 0.8855218855218855, "train_speed(iter/s)": 1.481491 }, { "epoch": 0.665567027976522, "grad_norm": 3.2569162845611572, "learning_rate": 9.569131317094399e-05, "loss": 0.39796648025512693, "memory(GiB)": 61.91, "step": 15535, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.481493 }, { "epoch": 0.6657812433057709, "grad_norm": 3.5562973022460938, "learning_rate": 9.568857976020193e-05, "loss": 0.3296400547027588, "memory(GiB)": 61.91, "step": 15540, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.481481 }, { "epoch": 0.66599545863502, "grad_norm": 3.6275205612182617, "learning_rate": 9.568584552176838e-05, "loss": 0.6006229400634766, "memory(GiB)": 61.91, "step": 15545, "token_acc": 0.8844984802431611, "train_speed(iter/s)": 1.481479 }, { "epoch": 0.6662096739642689, "grad_norm": 6.46284294128418, "learning_rate": 9.568311045569289e-05, "loss": 0.6438868045806885, "memory(GiB)": 61.91, "step": 15550, "token_acc": 0.851063829787234, "train_speed(iter/s)": 1.481532 }, { "epoch": 0.6664238892935178, "grad_norm": 3.8647687435150146, "learning_rate": 9.568037456202501e-05, "loss": 0.37517514228820803, "memory(GiB)": 61.91, "step": 15555, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.481555 }, { "epoch": 0.6666381046227668, "grad_norm": 2.901510238647461, "learning_rate": 9.567763784081428e-05, "loss": 0.640408706665039, "memory(GiB)": 61.91, "step": 15560, "token_acc": 0.8725490196078431, "train_speed(iter/s)": 1.481572 }, { "epoch": 0.6668523199520158, "grad_norm": 6.680017948150635, "learning_rate": 9.567490029211029e-05, "loss": 0.6082592010498047, "memory(GiB)": 61.91, "step": 15565, "token_acc": 0.8674242424242424, "train_speed(iter/s)": 1.481583 }, { "epoch": 0.6670665352812647, "grad_norm": 2.9801955223083496, "learning_rate": 9.567216191596264e-05, "loss": 0.466887092590332, "memory(GiB)": 61.91, "step": 15570, "token_acc": 0.9, "train_speed(iter/s)": 1.481579 }, { "epoch": 0.6672807506105137, "grad_norm": 2.7941744327545166, "learning_rate": 9.566942271242093e-05, "loss": 0.3779301166534424, "memory(GiB)": 61.91, "step": 15575, "token_acc": 0.9298780487804879, "train_speed(iter/s)": 1.481561 }, { "epoch": 0.6674949659397627, "grad_norm": 1.513960838317871, "learning_rate": 9.566668268153479e-05, "loss": 0.5241959095001221, "memory(GiB)": 61.91, "step": 15580, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.481552 }, { "epoch": 0.6677091812690116, "grad_norm": 2.5155041217803955, "learning_rate": 9.566394182335385e-05, "loss": 0.34528796672821044, "memory(GiB)": 61.91, "step": 15585, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.481538 }, { "epoch": 0.6679233965982606, "grad_norm": 3.066272258758545, "learning_rate": 9.566120013792776e-05, "loss": 0.5831729888916015, "memory(GiB)": 61.91, "step": 15590, "token_acc": 0.8703703703703703, "train_speed(iter/s)": 1.481537 }, { "epoch": 0.6681376119275095, "grad_norm": 3.409740447998047, "learning_rate": 9.56584576253062e-05, "loss": 0.4680851936340332, "memory(GiB)": 61.91, "step": 15595, "token_acc": 0.8912280701754386, "train_speed(iter/s)": 1.481529 }, { "epoch": 0.6683518272567585, "grad_norm": 2.5377256870269775, "learning_rate": 9.565571428553887e-05, "loss": 0.15661001205444336, "memory(GiB)": 61.91, "step": 15600, "token_acc": 0.9527272727272728, "train_speed(iter/s)": 1.481546 }, { "epoch": 0.6685660425860075, "grad_norm": 4.939282417297363, "learning_rate": 9.565297011867543e-05, "loss": 0.4762606620788574, "memory(GiB)": 61.91, "step": 15605, "token_acc": 0.8786764705882353, "train_speed(iter/s)": 1.481511 }, { "epoch": 0.6687802579152564, "grad_norm": 2.9337871074676514, "learning_rate": 9.565022512476561e-05, "loss": 0.6552966117858887, "memory(GiB)": 61.91, "step": 15610, "token_acc": 0.8328445747800587, "train_speed(iter/s)": 1.481473 }, { "epoch": 0.6689944732445053, "grad_norm": 3.017373561859131, "learning_rate": 9.564747930385915e-05, "loss": 0.22021045684814453, "memory(GiB)": 61.91, "step": 15615, "token_acc": 0.9531772575250836, "train_speed(iter/s)": 1.481437 }, { "epoch": 0.6692086885737544, "grad_norm": 3.758607864379883, "learning_rate": 9.564473265600576e-05, "loss": 0.3468992471694946, "memory(GiB)": 61.91, "step": 15620, "token_acc": 0.9153225806451613, "train_speed(iter/s)": 1.481435 }, { "epoch": 0.6694229039030033, "grad_norm": 2.091606378555298, "learning_rate": 9.564198518125523e-05, "loss": 0.6225194454193115, "memory(GiB)": 61.91, "step": 15625, "token_acc": 0.8648648648648649, "train_speed(iter/s)": 1.481457 }, { "epoch": 0.6696371192322522, "grad_norm": 4.352039337158203, "learning_rate": 9.563923687965733e-05, "loss": 0.4844411849975586, "memory(GiB)": 61.91, "step": 15630, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.481438 }, { "epoch": 0.6698513345615013, "grad_norm": 2.9185569286346436, "learning_rate": 9.563648775126184e-05, "loss": 0.4459090232849121, "memory(GiB)": 61.91, "step": 15635, "token_acc": 0.9022082018927445, "train_speed(iter/s)": 1.48143 }, { "epoch": 0.6700655498907502, "grad_norm": 4.6144561767578125, "learning_rate": 9.563373779611855e-05, "loss": 0.5153146743774414, "memory(GiB)": 61.91, "step": 15640, "token_acc": 0.864951768488746, "train_speed(iter/s)": 1.481441 }, { "epoch": 0.6702797652199991, "grad_norm": 4.122740268707275, "learning_rate": 9.563098701427731e-05, "loss": 0.5546058654785156, "memory(GiB)": 61.91, "step": 15645, "token_acc": 0.8880866425992779, "train_speed(iter/s)": 1.481436 }, { "epoch": 0.6704939805492481, "grad_norm": 5.20991849899292, "learning_rate": 9.562823540578791e-05, "loss": 0.27165191173553466, "memory(GiB)": 61.91, "step": 15650, "token_acc": 0.9455252918287937, "train_speed(iter/s)": 1.481448 }, { "epoch": 0.6707081958784971, "grad_norm": 1.9670287370681763, "learning_rate": 9.562548297070025e-05, "loss": 0.15592151880264282, "memory(GiB)": 61.91, "step": 15655, "token_acc": 0.9550173010380623, "train_speed(iter/s)": 1.481457 }, { "epoch": 0.670922411207746, "grad_norm": 2.7613086700439453, "learning_rate": 9.562272970906416e-05, "loss": 0.3607675075531006, "memory(GiB)": 61.91, "step": 15660, "token_acc": 0.8951048951048951, "train_speed(iter/s)": 1.48145 }, { "epoch": 0.671136626536995, "grad_norm": 0.09351155906915665, "learning_rate": 9.561997562092951e-05, "loss": 0.4224907398223877, "memory(GiB)": 61.91, "step": 15665, "token_acc": 0.8880866425992779, "train_speed(iter/s)": 1.481428 }, { "epoch": 0.6713508418662439, "grad_norm": 2.3093526363372803, "learning_rate": 9.561722070634623e-05, "loss": 0.5870115756988525, "memory(GiB)": 61.91, "step": 15670, "token_acc": 0.8881789137380192, "train_speed(iter/s)": 1.481411 }, { "epoch": 0.6715650571954929, "grad_norm": 5.0500946044921875, "learning_rate": 9.561446496536418e-05, "loss": 0.4020216464996338, "memory(GiB)": 61.91, "step": 15675, "token_acc": 0.9328358208955224, "train_speed(iter/s)": 1.481419 }, { "epoch": 0.6717792725247419, "grad_norm": 2.606489658355713, "learning_rate": 9.561170839803331e-05, "loss": 0.5015843868255615, "memory(GiB)": 61.91, "step": 15680, "token_acc": 0.9128787878787878, "train_speed(iter/s)": 1.481395 }, { "epoch": 0.6719934878539908, "grad_norm": 2.5600788593292236, "learning_rate": 9.560895100440357e-05, "loss": 0.3483614206314087, "memory(GiB)": 61.91, "step": 15685, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.481362 }, { "epoch": 0.6722077031832397, "grad_norm": 3.682037830352783, "learning_rate": 9.56061927845249e-05, "loss": 0.4748417854309082, "memory(GiB)": 61.91, "step": 15690, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.481374 }, { "epoch": 0.6724219185124888, "grad_norm": 2.9429361820220947, "learning_rate": 9.560343373844724e-05, "loss": 0.3984123468399048, "memory(GiB)": 61.91, "step": 15695, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.481405 }, { "epoch": 0.6726361338417377, "grad_norm": 3.6299030780792236, "learning_rate": 9.560067386622063e-05, "loss": 0.5585944652557373, "memory(GiB)": 61.91, "step": 15700, "token_acc": 0.8677966101694915, "train_speed(iter/s)": 1.481389 }, { "epoch": 0.6728503491709866, "grad_norm": 3.9927449226379395, "learning_rate": 9.559791316789502e-05, "loss": 0.43685212135314944, "memory(GiB)": 61.91, "step": 15705, "token_acc": 0.8709677419354839, "train_speed(iter/s)": 1.481376 }, { "epoch": 0.6730645645002357, "grad_norm": 4.929154396057129, "learning_rate": 9.559515164352044e-05, "loss": 0.32324469089508057, "memory(GiB)": 61.91, "step": 15710, "token_acc": 0.9153846153846154, "train_speed(iter/s)": 1.481388 }, { "epoch": 0.6732787798294846, "grad_norm": 3.427712917327881, "learning_rate": 9.559238929314692e-05, "loss": 0.4585599899291992, "memory(GiB)": 61.91, "step": 15715, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.481432 }, { "epoch": 0.6734929951587335, "grad_norm": 1.3544883728027344, "learning_rate": 9.558962611682447e-05, "loss": 0.4540881156921387, "memory(GiB)": 61.91, "step": 15720, "token_acc": 0.9093851132686084, "train_speed(iter/s)": 1.481445 }, { "epoch": 0.6737072104879825, "grad_norm": 3.855447292327881, "learning_rate": 9.558686211460322e-05, "loss": 0.32382969856262206, "memory(GiB)": 61.91, "step": 15725, "token_acc": 0.9223880597014925, "train_speed(iter/s)": 1.481454 }, { "epoch": 0.6739214258172315, "grad_norm": 4.459170341491699, "learning_rate": 9.558409728653317e-05, "loss": 0.37713112831115725, "memory(GiB)": 61.91, "step": 15730, "token_acc": 0.9287671232876712, "train_speed(iter/s)": 1.481454 }, { "epoch": 0.6741356411464804, "grad_norm": 0.29431939125061035, "learning_rate": 9.558133163266444e-05, "loss": 0.30568838119506836, "memory(GiB)": 61.91, "step": 15735, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.481455 }, { "epoch": 0.6743498564757294, "grad_norm": 4.275304794311523, "learning_rate": 9.557856515304713e-05, "loss": 0.2838555335998535, "memory(GiB)": 61.91, "step": 15740, "token_acc": 0.9435483870967742, "train_speed(iter/s)": 1.481451 }, { "epoch": 0.6745640718049783, "grad_norm": 2.8395397663116455, "learning_rate": 9.557579784773137e-05, "loss": 0.41010432243347167, "memory(GiB)": 61.91, "step": 15745, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.481437 }, { "epoch": 0.6747782871342273, "grad_norm": 1.2613489627838135, "learning_rate": 9.557302971676727e-05, "loss": 0.35261919498443606, "memory(GiB)": 61.91, "step": 15750, "token_acc": 0.940251572327044, "train_speed(iter/s)": 1.481464 }, { "epoch": 0.6749925024634763, "grad_norm": 1.8311147689819336, "learning_rate": 9.557026076020498e-05, "loss": 0.33149721622467043, "memory(GiB)": 61.91, "step": 15755, "token_acc": 0.9498327759197325, "train_speed(iter/s)": 1.481462 }, { "epoch": 0.6752067177927252, "grad_norm": 2.444195032119751, "learning_rate": 9.556749097809468e-05, "loss": 0.41254091262817383, "memory(GiB)": 61.91, "step": 15760, "token_acc": 0.9124579124579124, "train_speed(iter/s)": 1.481457 }, { "epoch": 0.6754209331219742, "grad_norm": 3.186567783355713, "learning_rate": 9.556472037048651e-05, "loss": 0.45941996574401855, "memory(GiB)": 61.91, "step": 15765, "token_acc": 0.9087136929460581, "train_speed(iter/s)": 1.481485 }, { "epoch": 0.6756351484512232, "grad_norm": 2.2602972984313965, "learning_rate": 9.556194893743071e-05, "loss": 0.6207956790924072, "memory(GiB)": 61.91, "step": 15770, "token_acc": 0.8769716088328076, "train_speed(iter/s)": 1.481515 }, { "epoch": 0.6758493637804721, "grad_norm": 2.0642268657684326, "learning_rate": 9.555917667897746e-05, "loss": 0.33627517223358155, "memory(GiB)": 61.91, "step": 15775, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.481507 }, { "epoch": 0.6760635791097211, "grad_norm": 2.4361753463745117, "learning_rate": 9.555640359517699e-05, "loss": 0.5159164905548096, "memory(GiB)": 61.91, "step": 15780, "token_acc": 0.9093959731543624, "train_speed(iter/s)": 1.48149 }, { "epoch": 0.6762777944389701, "grad_norm": 5.294432163238525, "learning_rate": 9.555362968607952e-05, "loss": 0.5102428436279297, "memory(GiB)": 61.91, "step": 15785, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.481493 }, { "epoch": 0.676492009768219, "grad_norm": 3.033535957336426, "learning_rate": 9.555085495173532e-05, "loss": 0.29211575984954835, "memory(GiB)": 61.91, "step": 15790, "token_acc": 0.9434523809523809, "train_speed(iter/s)": 1.48147 }, { "epoch": 0.676706225097468, "grad_norm": 3.078522205352783, "learning_rate": 9.554807939219467e-05, "loss": 0.3547328472137451, "memory(GiB)": 61.91, "step": 15795, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.481464 }, { "epoch": 0.676920440426717, "grad_norm": 2.648268938064575, "learning_rate": 9.554530300750782e-05, "loss": 0.11839771270751953, "memory(GiB)": 61.91, "step": 15800, "token_acc": 0.967032967032967, "train_speed(iter/s)": 1.481452 }, { "epoch": 0.6771346557559659, "grad_norm": 6.663754463195801, "learning_rate": 9.554252579772509e-05, "loss": 0.45918951034545896, "memory(GiB)": 61.91, "step": 15805, "token_acc": 0.8836363636363637, "train_speed(iter/s)": 1.481452 }, { "epoch": 0.6773488710852149, "grad_norm": 3.0217902660369873, "learning_rate": 9.553974776289678e-05, "loss": 0.3430671691894531, "memory(GiB)": 61.91, "step": 15810, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.481446 }, { "epoch": 0.6775630864144638, "grad_norm": 5.535459518432617, "learning_rate": 9.55369689030732e-05, "loss": 0.7204854965209961, "memory(GiB)": 61.91, "step": 15815, "token_acc": 0.863013698630137, "train_speed(iter/s)": 1.48147 }, { "epoch": 0.6777773017437128, "grad_norm": 3.619563579559326, "learning_rate": 9.553418921830473e-05, "loss": 0.45639710426330565, "memory(GiB)": 61.91, "step": 15820, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.481506 }, { "epoch": 0.6779915170729618, "grad_norm": 6.696113109588623, "learning_rate": 9.55314087086417e-05, "loss": 0.2822275161743164, "memory(GiB)": 61.91, "step": 15825, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.481457 }, { "epoch": 0.6782057324022107, "grad_norm": 1.9170283079147339, "learning_rate": 9.552862737413449e-05, "loss": 0.5009267807006836, "memory(GiB)": 61.91, "step": 15830, "token_acc": 0.8884462151394422, "train_speed(iter/s)": 1.481445 }, { "epoch": 0.6784199477314596, "grad_norm": 2.6807470321655273, "learning_rate": 9.55258452148335e-05, "loss": 0.40252199172973635, "memory(GiB)": 61.91, "step": 15835, "token_acc": 0.907051282051282, "train_speed(iter/s)": 1.481543 }, { "epoch": 0.6786341630607087, "grad_norm": 4.873498916625977, "learning_rate": 9.552306223078911e-05, "loss": 0.6006161212921143, "memory(GiB)": 61.91, "step": 15840, "token_acc": 0.8623188405797102, "train_speed(iter/s)": 1.481628 }, { "epoch": 0.6788483783899576, "grad_norm": 4.266808032989502, "learning_rate": 9.552027842205174e-05, "loss": 0.30486416816711426, "memory(GiB)": 61.91, "step": 15845, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.481624 }, { "epoch": 0.6790625937192065, "grad_norm": 4.427332401275635, "learning_rate": 9.55174937886718e-05, "loss": 0.49683256149291993, "memory(GiB)": 61.91, "step": 15850, "token_acc": 0.8916083916083916, "train_speed(iter/s)": 1.481636 }, { "epoch": 0.6792768090484556, "grad_norm": 3.953134536743164, "learning_rate": 9.551470833069979e-05, "loss": 0.7852727890014648, "memory(GiB)": 61.91, "step": 15855, "token_acc": 0.8470254957507082, "train_speed(iter/s)": 1.48166 }, { "epoch": 0.6794910243777045, "grad_norm": 1.0975964069366455, "learning_rate": 9.551192204818615e-05, "loss": 0.4172943592071533, "memory(GiB)": 61.91, "step": 15860, "token_acc": 0.903010033444816, "train_speed(iter/s)": 1.481651 }, { "epoch": 0.6797052397069534, "grad_norm": 6.03659200668335, "learning_rate": 9.550913494118133e-05, "loss": 0.5827147483825683, "memory(GiB)": 61.91, "step": 15865, "token_acc": 0.8765822784810127, "train_speed(iter/s)": 1.481686 }, { "epoch": 0.6799194550362024, "grad_norm": 4.107854843139648, "learning_rate": 9.550634700973585e-05, "loss": 0.31268310546875, "memory(GiB)": 61.91, "step": 15870, "token_acc": 0.9380165289256198, "train_speed(iter/s)": 1.481669 }, { "epoch": 0.6801336703654514, "grad_norm": 1.4781098365783691, "learning_rate": 9.55035582539002e-05, "loss": 0.5103428840637207, "memory(GiB)": 61.91, "step": 15875, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.481693 }, { "epoch": 0.6803478856947003, "grad_norm": 2.5276315212249756, "learning_rate": 9.550076867372491e-05, "loss": 0.36758670806884763, "memory(GiB)": 61.91, "step": 15880, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.481675 }, { "epoch": 0.6805621010239493, "grad_norm": 3.6798593997955322, "learning_rate": 9.549797826926052e-05, "loss": 0.4272181034088135, "memory(GiB)": 61.91, "step": 15885, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.481664 }, { "epoch": 0.6807763163531982, "grad_norm": 4.274956703186035, "learning_rate": 9.549518704055755e-05, "loss": 0.3734795331954956, "memory(GiB)": 61.91, "step": 15890, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.48164 }, { "epoch": 0.6809905316824472, "grad_norm": 2.4862425327301025, "learning_rate": 9.549239498766661e-05, "loss": 0.6130151748657227, "memory(GiB)": 61.91, "step": 15895, "token_acc": 0.8767605633802817, "train_speed(iter/s)": 1.481655 }, { "epoch": 0.6812047470116962, "grad_norm": 7.011573314666748, "learning_rate": 9.548960211063824e-05, "loss": 0.628237533569336, "memory(GiB)": 61.91, "step": 15900, "token_acc": 0.8673469387755102, "train_speed(iter/s)": 1.481734 }, { "epoch": 0.6814189623409451, "grad_norm": 3.3327603340148926, "learning_rate": 9.548680840952308e-05, "loss": 0.42882094383239744, "memory(GiB)": 61.91, "step": 15905, "token_acc": 0.9144542772861357, "train_speed(iter/s)": 1.48178 }, { "epoch": 0.681633177670194, "grad_norm": 2.8352174758911133, "learning_rate": 9.548401388437169e-05, "loss": 0.3357755422592163, "memory(GiB)": 61.91, "step": 15910, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.481738 }, { "epoch": 0.6818473929994431, "grad_norm": 3.148167848587036, "learning_rate": 9.548121853523475e-05, "loss": 0.6671308040618896, "memory(GiB)": 61.91, "step": 15915, "token_acc": 0.861198738170347, "train_speed(iter/s)": 1.481733 }, { "epoch": 0.682061608328692, "grad_norm": 2.371392250061035, "learning_rate": 9.547842236216285e-05, "loss": 0.314603853225708, "memory(GiB)": 61.91, "step": 15920, "token_acc": 0.9382352941176471, "train_speed(iter/s)": 1.481753 }, { "epoch": 0.6822758236579409, "grad_norm": 3.5529072284698486, "learning_rate": 9.547562536520667e-05, "loss": 0.33336875438690183, "memory(GiB)": 61.91, "step": 15925, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.481854 }, { "epoch": 0.68249003898719, "grad_norm": 4.414220809936523, "learning_rate": 9.547282754441687e-05, "loss": 0.31989617347717286, "memory(GiB)": 61.91, "step": 15930, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.481833 }, { "epoch": 0.6827042543164389, "grad_norm": 4.114873886108398, "learning_rate": 9.547002889984415e-05, "loss": 0.45402207374572756, "memory(GiB)": 61.91, "step": 15935, "token_acc": 0.9138461538461539, "train_speed(iter/s)": 1.481856 }, { "epoch": 0.6829184696456878, "grad_norm": 3.723588228225708, "learning_rate": 9.546722943153921e-05, "loss": 0.4394641876220703, "memory(GiB)": 61.91, "step": 15940, "token_acc": 0.9017341040462428, "train_speed(iter/s)": 1.481856 }, { "epoch": 0.6831326849749368, "grad_norm": 6.376893997192383, "learning_rate": 9.546442913955276e-05, "loss": 0.4547878742218018, "memory(GiB)": 61.91, "step": 15945, "token_acc": 0.8857938718662952, "train_speed(iter/s)": 1.481843 }, { "epoch": 0.6833469003041858, "grad_norm": 0.23020677268505096, "learning_rate": 9.54616280239355e-05, "loss": 0.4190036773681641, "memory(GiB)": 61.91, "step": 15950, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.481846 }, { "epoch": 0.6835611156334347, "grad_norm": 8.47883129119873, "learning_rate": 9.545882608473823e-05, "loss": 0.7333100318908692, "memory(GiB)": 61.91, "step": 15955, "token_acc": 0.8354838709677419, "train_speed(iter/s)": 1.481872 }, { "epoch": 0.6837753309626837, "grad_norm": 3.2420997619628906, "learning_rate": 9.545602332201167e-05, "loss": 0.20959248542785644, "memory(GiB)": 61.91, "step": 15960, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.481873 }, { "epoch": 0.6839895462919326, "grad_norm": 0.3728238642215729, "learning_rate": 9.545321973580661e-05, "loss": 0.2719521999359131, "memory(GiB)": 61.91, "step": 15965, "token_acc": 0.9330543933054394, "train_speed(iter/s)": 1.481864 }, { "epoch": 0.6842037616211816, "grad_norm": 3.0417933464050293, "learning_rate": 9.545041532617382e-05, "loss": 0.3236682415008545, "memory(GiB)": 61.91, "step": 15970, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.481886 }, { "epoch": 0.6844179769504306, "grad_norm": 5.005701541900635, "learning_rate": 9.544761009316414e-05, "loss": 0.36336545944213866, "memory(GiB)": 61.91, "step": 15975, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.481896 }, { "epoch": 0.6846321922796795, "grad_norm": 4.143919944763184, "learning_rate": 9.544480403682836e-05, "loss": 0.45569324493408203, "memory(GiB)": 61.91, "step": 15980, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.481912 }, { "epoch": 0.6848464076089285, "grad_norm": 3.5759494304656982, "learning_rate": 9.544199715721734e-05, "loss": 0.5146461009979248, "memory(GiB)": 61.91, "step": 15985, "token_acc": 0.8896797153024911, "train_speed(iter/s)": 1.481919 }, { "epoch": 0.6850606229381775, "grad_norm": 5.151166915893555, "learning_rate": 9.543918945438189e-05, "loss": 0.3692609786987305, "memory(GiB)": 61.91, "step": 15990, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.481915 }, { "epoch": 0.6852748382674264, "grad_norm": 5.309976100921631, "learning_rate": 9.543638092837291e-05, "loss": 0.4565682888031006, "memory(GiB)": 61.91, "step": 15995, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.481911 }, { "epoch": 0.6854890535966753, "grad_norm": 3.289983034133911, "learning_rate": 9.543357157924126e-05, "loss": 0.4279374122619629, "memory(GiB)": 61.91, "step": 16000, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.481954 }, { "epoch": 0.6854890535966753, "eval_loss": 2.579702854156494, "eval_runtime": 13.1879, "eval_samples_per_second": 7.583, "eval_steps_per_second": 7.583, "eval_token_acc": 0.42398884239888424, "step": 16000 }, { "epoch": 0.6857032689259244, "grad_norm": 3.1717193126678467, "learning_rate": 9.543076140703786e-05, "loss": 0.825126838684082, "memory(GiB)": 61.91, "step": 16005, "token_acc": 0.5506268081002893, "train_speed(iter/s)": 1.480038 }, { "epoch": 0.6859174842551733, "grad_norm": 4.573133945465088, "learning_rate": 9.542795041181359e-05, "loss": 0.7674770355224609, "memory(GiB)": 61.91, "step": 16010, "token_acc": 0.84375, "train_speed(iter/s)": 1.480045 }, { "epoch": 0.6861316995844222, "grad_norm": 1.206395149230957, "learning_rate": 9.542513859361937e-05, "loss": 0.4191195011138916, "memory(GiB)": 61.91, "step": 16015, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.480073 }, { "epoch": 0.6863459149136713, "grad_norm": 2.8553404808044434, "learning_rate": 9.542232595250615e-05, "loss": 0.7442595481872558, "memory(GiB)": 61.91, "step": 16020, "token_acc": 0.8589743589743589, "train_speed(iter/s)": 1.480106 }, { "epoch": 0.6865601302429202, "grad_norm": 3.0150253772735596, "learning_rate": 9.54195124885249e-05, "loss": 0.43084964752197263, "memory(GiB)": 61.91, "step": 16025, "token_acc": 0.8934169278996865, "train_speed(iter/s)": 1.480106 }, { "epoch": 0.6867743455721691, "grad_norm": 3.5962295532226562, "learning_rate": 9.541669820172656e-05, "loss": 0.6882177352905273, "memory(GiB)": 61.91, "step": 16030, "token_acc": 0.8525073746312685, "train_speed(iter/s)": 1.480149 }, { "epoch": 0.6869885609014181, "grad_norm": 2.33845591545105, "learning_rate": 9.541388309216214e-05, "loss": 0.21027984619140624, "memory(GiB)": 61.91, "step": 16035, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.480148 }, { "epoch": 0.6872027762306671, "grad_norm": 4.793637752532959, "learning_rate": 9.541106715988263e-05, "loss": 0.306810998916626, "memory(GiB)": 61.91, "step": 16040, "token_acc": 0.9375, "train_speed(iter/s)": 1.480146 }, { "epoch": 0.687416991559916, "grad_norm": 3.801490068435669, "learning_rate": 9.540825040493903e-05, "loss": 0.40580925941467283, "memory(GiB)": 61.91, "step": 16045, "token_acc": 0.9375, "train_speed(iter/s)": 1.480139 }, { "epoch": 0.687631206889165, "grad_norm": 2.4022035598754883, "learning_rate": 9.540543282738239e-05, "loss": 0.5293187141418457, "memory(GiB)": 61.91, "step": 16050, "token_acc": 0.8689138576779026, "train_speed(iter/s)": 1.48013 }, { "epoch": 0.6878454222184139, "grad_norm": 3.042287588119507, "learning_rate": 9.540261442726373e-05, "loss": 0.3027838706970215, "memory(GiB)": 61.91, "step": 16055, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.480148 }, { "epoch": 0.6880596375476629, "grad_norm": 3.454556941986084, "learning_rate": 9.539979520463413e-05, "loss": 0.3649451732635498, "memory(GiB)": 61.91, "step": 16060, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.480167 }, { "epoch": 0.6882738528769119, "grad_norm": 2.7650389671325684, "learning_rate": 9.539697515954465e-05, "loss": 0.5789755821228028, "memory(GiB)": 61.91, "step": 16065, "token_acc": 0.8840579710144928, "train_speed(iter/s)": 1.480172 }, { "epoch": 0.6884880682061608, "grad_norm": 2.523130178451538, "learning_rate": 9.539415429204636e-05, "loss": 0.3914970397949219, "memory(GiB)": 61.91, "step": 16070, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.480171 }, { "epoch": 0.6887022835354097, "grad_norm": 3.65509295463562, "learning_rate": 9.53913326021904e-05, "loss": 0.4312613010406494, "memory(GiB)": 61.91, "step": 16075, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.48016 }, { "epoch": 0.6889164988646588, "grad_norm": 1.8948485851287842, "learning_rate": 9.538851009002785e-05, "loss": 0.18220894336700438, "memory(GiB)": 61.91, "step": 16080, "token_acc": 0.959866220735786, "train_speed(iter/s)": 1.480159 }, { "epoch": 0.6891307141939077, "grad_norm": 4.382693290710449, "learning_rate": 9.538568675560988e-05, "loss": 0.4242574691772461, "memory(GiB)": 61.91, "step": 16085, "token_acc": 0.9003215434083601, "train_speed(iter/s)": 1.480146 }, { "epoch": 0.6893449295231566, "grad_norm": 3.77968430519104, "learning_rate": 9.538286259898762e-05, "loss": 0.4293569564819336, "memory(GiB)": 61.91, "step": 16090, "token_acc": 0.9050632911392406, "train_speed(iter/s)": 1.480162 }, { "epoch": 0.6895591448524057, "grad_norm": 1.1404070854187012, "learning_rate": 9.538003762021221e-05, "loss": 0.4667935848236084, "memory(GiB)": 61.91, "step": 16095, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.480204 }, { "epoch": 0.6897733601816546, "grad_norm": 6.5467705726623535, "learning_rate": 9.537721181933488e-05, "loss": 0.6287301540374756, "memory(GiB)": 61.91, "step": 16100, "token_acc": 0.8615819209039548, "train_speed(iter/s)": 1.480184 }, { "epoch": 0.6899875755109035, "grad_norm": 4.944956302642822, "learning_rate": 9.537438519640675e-05, "loss": 0.570768165588379, "memory(GiB)": 61.91, "step": 16105, "token_acc": 0.89171974522293, "train_speed(iter/s)": 1.480183 }, { "epoch": 0.6902017908401525, "grad_norm": 4.639111042022705, "learning_rate": 9.53715577514791e-05, "loss": 0.4648996353149414, "memory(GiB)": 61.91, "step": 16110, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.480183 }, { "epoch": 0.6904160061694015, "grad_norm": 2.360802412033081, "learning_rate": 9.536872948460312e-05, "loss": 0.19060324430465697, "memory(GiB)": 61.91, "step": 16115, "token_acc": 0.9592592592592593, "train_speed(iter/s)": 1.480178 }, { "epoch": 0.6906302214986505, "grad_norm": 2.613246440887451, "learning_rate": 9.536590039583004e-05, "loss": 0.21077706813812255, "memory(GiB)": 61.91, "step": 16120, "token_acc": 0.9320987654320988, "train_speed(iter/s)": 1.480155 }, { "epoch": 0.6908444368278994, "grad_norm": 0.06263306736946106, "learning_rate": 9.536307048521112e-05, "loss": 0.374418306350708, "memory(GiB)": 61.91, "step": 16125, "token_acc": 0.9077490774907749, "train_speed(iter/s)": 1.480126 }, { "epoch": 0.6910586521571483, "grad_norm": 3.0156185626983643, "learning_rate": 9.53602397527976e-05, "loss": 0.3358055830001831, "memory(GiB)": 61.91, "step": 16130, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.480119 }, { "epoch": 0.6912728674863974, "grad_norm": 3.4878363609313965, "learning_rate": 9.535740819864081e-05, "loss": 0.48032584190368655, "memory(GiB)": 61.91, "step": 16135, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.480113 }, { "epoch": 0.6914870828156463, "grad_norm": 2.8022868633270264, "learning_rate": 9.535457582279203e-05, "loss": 0.34444143772125246, "memory(GiB)": 61.91, "step": 16140, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.480103 }, { "epoch": 0.6917012981448952, "grad_norm": 4.650376796722412, "learning_rate": 9.535174262530254e-05, "loss": 0.12239797115325927, "memory(GiB)": 61.91, "step": 16145, "token_acc": 0.9793103448275862, "train_speed(iter/s)": 1.480104 }, { "epoch": 0.6919155134741443, "grad_norm": 2.8736023902893066, "learning_rate": 9.53489086062237e-05, "loss": 0.5388351917266846, "memory(GiB)": 61.91, "step": 16150, "token_acc": 0.8877551020408163, "train_speed(iter/s)": 1.480079 }, { "epoch": 0.6921297288033932, "grad_norm": 5.612738609313965, "learning_rate": 9.534607376560684e-05, "loss": 0.6932330131530762, "memory(GiB)": 61.91, "step": 16155, "token_acc": 0.8870431893687708, "train_speed(iter/s)": 1.480064 }, { "epoch": 0.6923439441326421, "grad_norm": 0.0807308629155159, "learning_rate": 9.534323810350332e-05, "loss": 0.5938024997711182, "memory(GiB)": 61.91, "step": 16160, "token_acc": 0.8717948717948718, "train_speed(iter/s)": 1.480016 }, { "epoch": 0.6925581594618911, "grad_norm": 2.4327831268310547, "learning_rate": 9.534040161996449e-05, "loss": 0.7377054214477539, "memory(GiB)": 61.91, "step": 16165, "token_acc": 0.8617021276595744, "train_speed(iter/s)": 1.480058 }, { "epoch": 0.6927723747911401, "grad_norm": 3.0079426765441895, "learning_rate": 9.533756431504177e-05, "loss": 0.3451094150543213, "memory(GiB)": 61.91, "step": 16170, "token_acc": 0.9174917491749175, "train_speed(iter/s)": 1.480055 }, { "epoch": 0.692986590120389, "grad_norm": 2.066725730895996, "learning_rate": 9.533472618878653e-05, "loss": 0.33328244686126707, "memory(GiB)": 61.91, "step": 16175, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.480065 }, { "epoch": 0.693200805449638, "grad_norm": 4.514894962310791, "learning_rate": 9.53318872412502e-05, "loss": 0.7159274101257325, "memory(GiB)": 61.91, "step": 16180, "token_acc": 0.8679867986798679, "train_speed(iter/s)": 1.480069 }, { "epoch": 0.693415020778887, "grad_norm": 2.2582883834838867, "learning_rate": 9.532904747248422e-05, "loss": 0.49430098533630373, "memory(GiB)": 61.91, "step": 16185, "token_acc": 0.9066265060240963, "train_speed(iter/s)": 1.480123 }, { "epoch": 0.6936292361081359, "grad_norm": 2.403585195541382, "learning_rate": 9.532620688254e-05, "loss": 0.5540934085845948, "memory(GiB)": 61.91, "step": 16190, "token_acc": 0.8716981132075472, "train_speed(iter/s)": 1.48012 }, { "epoch": 0.6938434514373849, "grad_norm": 3.2883353233337402, "learning_rate": 9.532336547146904e-05, "loss": 0.41477031707763673, "memory(GiB)": 61.91, "step": 16195, "token_acc": 0.9018987341772152, "train_speed(iter/s)": 1.480115 }, { "epoch": 0.6940576667666338, "grad_norm": 1.7837417125701904, "learning_rate": 9.532052323932279e-05, "loss": 0.4742863655090332, "memory(GiB)": 61.91, "step": 16200, "token_acc": 0.8966666666666666, "train_speed(iter/s)": 1.480122 }, { "epoch": 0.6942718820958828, "grad_norm": 4.919589042663574, "learning_rate": 9.531768018615276e-05, "loss": 0.7809865951538086, "memory(GiB)": 61.91, "step": 16205, "token_acc": 0.8470588235294118, "train_speed(iter/s)": 1.480165 }, { "epoch": 0.6944860974251318, "grad_norm": 5.131477355957031, "learning_rate": 9.531483631201044e-05, "loss": 0.5495792388916015, "memory(GiB)": 61.91, "step": 16210, "token_acc": 0.8850574712643678, "train_speed(iter/s)": 1.480169 }, { "epoch": 0.6947003127543807, "grad_norm": 2.3777599334716797, "learning_rate": 9.531199161694732e-05, "loss": 0.5106991767883301, "memory(GiB)": 61.91, "step": 16215, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.480166 }, { "epoch": 0.6949145280836296, "grad_norm": 2.0256621837615967, "learning_rate": 9.5309146101015e-05, "loss": 0.35914623737335205, "memory(GiB)": 61.91, "step": 16220, "token_acc": 0.9294117647058824, "train_speed(iter/s)": 1.480163 }, { "epoch": 0.6951287434128787, "grad_norm": 4.8575119972229, "learning_rate": 9.530629976426499e-05, "loss": 0.3849184989929199, "memory(GiB)": 61.91, "step": 16225, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.480143 }, { "epoch": 0.6953429587421276, "grad_norm": 3.6568636894226074, "learning_rate": 9.530345260674885e-05, "loss": 0.2740474700927734, "memory(GiB)": 61.91, "step": 16230, "token_acc": 0.9277566539923955, "train_speed(iter/s)": 1.480174 }, { "epoch": 0.6955571740713765, "grad_norm": 3.297774076461792, "learning_rate": 9.530060462851818e-05, "loss": 0.42949376106262205, "memory(GiB)": 61.91, "step": 16235, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.480218 }, { "epoch": 0.6957713894006256, "grad_norm": 3.8572871685028076, "learning_rate": 9.529775582962455e-05, "loss": 0.49840555191040037, "memory(GiB)": 61.91, "step": 16240, "token_acc": 0.8924731182795699, "train_speed(iter/s)": 1.480222 }, { "epoch": 0.6959856047298745, "grad_norm": 5.665950298309326, "learning_rate": 9.529490621011958e-05, "loss": 0.3551926136016846, "memory(GiB)": 61.91, "step": 16245, "token_acc": 0.9247648902821317, "train_speed(iter/s)": 1.480278 }, { "epoch": 0.6961998200591234, "grad_norm": 3.3979365825653076, "learning_rate": 9.529205577005491e-05, "loss": 0.38051939010620117, "memory(GiB)": 61.91, "step": 16250, "token_acc": 0.9300291545189504, "train_speed(iter/s)": 1.480286 }, { "epoch": 0.6964140353883724, "grad_norm": 5.01914644241333, "learning_rate": 9.528920450948215e-05, "loss": 0.3453675270080566, "memory(GiB)": 61.91, "step": 16255, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.480292 }, { "epoch": 0.6966282507176214, "grad_norm": 4.103448390960693, "learning_rate": 9.528635242845297e-05, "loss": 0.5251649856567383, "memory(GiB)": 61.91, "step": 16260, "token_acc": 0.9224137931034483, "train_speed(iter/s)": 1.480308 }, { "epoch": 0.6968424660468703, "grad_norm": 4.562344074249268, "learning_rate": 9.528349952701902e-05, "loss": 0.36369097232818604, "memory(GiB)": 61.91, "step": 16265, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.480313 }, { "epoch": 0.6970566813761193, "grad_norm": 1.0909693241119385, "learning_rate": 9.528064580523201e-05, "loss": 0.2936516761779785, "memory(GiB)": 61.91, "step": 16270, "token_acc": 0.9159891598915989, "train_speed(iter/s)": 1.480324 }, { "epoch": 0.6972708967053682, "grad_norm": 2.369767904281616, "learning_rate": 9.527779126314362e-05, "loss": 0.4679872035980225, "memory(GiB)": 61.91, "step": 16275, "token_acc": 0.8916083916083916, "train_speed(iter/s)": 1.480373 }, { "epoch": 0.6974851120346172, "grad_norm": 0.500034749507904, "learning_rate": 9.527493590080557e-05, "loss": 0.40371031761169435, "memory(GiB)": 61.91, "step": 16280, "token_acc": 0.9184397163120568, "train_speed(iter/s)": 1.480396 }, { "epoch": 0.6976993273638662, "grad_norm": 3.466543436050415, "learning_rate": 9.527207971826959e-05, "loss": 0.5843483448028565, "memory(GiB)": 61.91, "step": 16285, "token_acc": 0.8758169934640523, "train_speed(iter/s)": 1.480388 }, { "epoch": 0.6979135426931151, "grad_norm": 2.788017988204956, "learning_rate": 9.52692227155874e-05, "loss": 0.24493088722229003, "memory(GiB)": 61.91, "step": 16290, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.480373 }, { "epoch": 0.698127758022364, "grad_norm": 2.5382473468780518, "learning_rate": 9.526636489281078e-05, "loss": 0.2528695106506348, "memory(GiB)": 61.91, "step": 16295, "token_acc": 0.9525316455696202, "train_speed(iter/s)": 1.480398 }, { "epoch": 0.6983419733516131, "grad_norm": 3.3628792762756348, "learning_rate": 9.526350624999153e-05, "loss": 0.16815507411956787, "memory(GiB)": 61.91, "step": 16300, "token_acc": 0.9601593625498008, "train_speed(iter/s)": 1.48039 }, { "epoch": 0.698556188680862, "grad_norm": 3.9932966232299805, "learning_rate": 9.526064678718137e-05, "loss": 0.4950117111206055, "memory(GiB)": 61.91, "step": 16305, "token_acc": 0.9011857707509882, "train_speed(iter/s)": 1.480451 }, { "epoch": 0.6987704040101109, "grad_norm": 2.40407133102417, "learning_rate": 9.525778650443214e-05, "loss": 0.2374722719192505, "memory(GiB)": 61.91, "step": 16310, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.480454 }, { "epoch": 0.69898461933936, "grad_norm": 3.667633533477783, "learning_rate": 9.525492540179563e-05, "loss": 0.3298379182815552, "memory(GiB)": 61.91, "step": 16315, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.480446 }, { "epoch": 0.6991988346686089, "grad_norm": 2.8265085220336914, "learning_rate": 9.525206347932373e-05, "loss": 0.3808382272720337, "memory(GiB)": 61.91, "step": 16320, "token_acc": 0.9299610894941635, "train_speed(iter/s)": 1.480486 }, { "epoch": 0.6994130499978578, "grad_norm": 2.5881175994873047, "learning_rate": 9.524920073706824e-05, "loss": 0.30810182094573973, "memory(GiB)": 61.91, "step": 16325, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.480478 }, { "epoch": 0.6996272653271068, "grad_norm": 2.9233310222625732, "learning_rate": 9.524633717508103e-05, "loss": 0.34280738830566404, "memory(GiB)": 61.91, "step": 16330, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.480452 }, { "epoch": 0.6998414806563558, "grad_norm": 3.6014957427978516, "learning_rate": 9.524347279341397e-05, "loss": 0.341717004776001, "memory(GiB)": 61.91, "step": 16335, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.480441 }, { "epoch": 0.7000556959856047, "grad_norm": 2.3197576999664307, "learning_rate": 9.524060759211897e-05, "loss": 0.5460198879241943, "memory(GiB)": 61.91, "step": 16340, "token_acc": 0.8948863636363636, "train_speed(iter/s)": 1.480435 }, { "epoch": 0.7002699113148537, "grad_norm": 0.16807520389556885, "learning_rate": 9.523774157124791e-05, "loss": 0.14502352476119995, "memory(GiB)": 61.91, "step": 16345, "token_acc": 0.9723320158102767, "train_speed(iter/s)": 1.480433 }, { "epoch": 0.7004841266441026, "grad_norm": 4.0745344161987305, "learning_rate": 9.523487473085274e-05, "loss": 0.4727954387664795, "memory(GiB)": 61.91, "step": 16350, "token_acc": 0.8877887788778878, "train_speed(iter/s)": 1.480428 }, { "epoch": 0.7006983419733516, "grad_norm": 1.7719545364379883, "learning_rate": 9.523200707098537e-05, "loss": 0.5752057075500489, "memory(GiB)": 61.91, "step": 16355, "token_acc": 0.9146919431279621, "train_speed(iter/s)": 1.480431 }, { "epoch": 0.7009125573026006, "grad_norm": 3.190648078918457, "learning_rate": 9.522913859169778e-05, "loss": 0.3660828351974487, "memory(GiB)": 61.91, "step": 16360, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.480421 }, { "epoch": 0.7011267726318495, "grad_norm": 5.664255142211914, "learning_rate": 9.52262692930419e-05, "loss": 0.8895369529724121, "memory(GiB)": 61.91, "step": 16365, "token_acc": 0.8461538461538461, "train_speed(iter/s)": 1.48046 }, { "epoch": 0.7013409879610984, "grad_norm": 7.215083599090576, "learning_rate": 9.522339917506973e-05, "loss": 0.5255794048309326, "memory(GiB)": 61.91, "step": 16370, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.480474 }, { "epoch": 0.7015552032903475, "grad_norm": 0.08328226208686829, "learning_rate": 9.522052823783325e-05, "loss": 0.15123497247695922, "memory(GiB)": 61.91, "step": 16375, "token_acc": 0.967391304347826, "train_speed(iter/s)": 1.480493 }, { "epoch": 0.7017694186195964, "grad_norm": 9.059599876403809, "learning_rate": 9.521765648138449e-05, "loss": 0.20889406204223632, "memory(GiB)": 61.91, "step": 16380, "token_acc": 0.9548872180451128, "train_speed(iter/s)": 1.480499 }, { "epoch": 0.7019836339488453, "grad_norm": 7.689831733703613, "learning_rate": 9.521478390577546e-05, "loss": 0.35178546905517577, "memory(GiB)": 61.91, "step": 16385, "token_acc": 0.9390681003584229, "train_speed(iter/s)": 1.480502 }, { "epoch": 0.7021978492780944, "grad_norm": 1.194166660308838, "learning_rate": 9.521191051105823e-05, "loss": 0.5222110748291016, "memory(GiB)": 61.91, "step": 16390, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.480555 }, { "epoch": 0.7024120646073433, "grad_norm": 1.6429449319839478, "learning_rate": 9.520903629728479e-05, "loss": 0.3597019910812378, "memory(GiB)": 61.91, "step": 16395, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.480572 }, { "epoch": 0.7026262799365922, "grad_norm": 5.310493469238281, "learning_rate": 9.520616126450726e-05, "loss": 0.4792961120605469, "memory(GiB)": 61.91, "step": 16400, "token_acc": 0.8855218855218855, "train_speed(iter/s)": 1.480537 }, { "epoch": 0.7028404952658412, "grad_norm": 3.854611396789551, "learning_rate": 9.520328541277772e-05, "loss": 0.5979731559753418, "memory(GiB)": 61.91, "step": 16405, "token_acc": 0.8671328671328671, "train_speed(iter/s)": 1.480512 }, { "epoch": 0.7030547105950902, "grad_norm": 3.0502285957336426, "learning_rate": 9.520040874214827e-05, "loss": 0.36093270778656006, "memory(GiB)": 61.91, "step": 16410, "token_acc": 0.9177215189873418, "train_speed(iter/s)": 1.480489 }, { "epoch": 0.7032689259243391, "grad_norm": 0.6400442719459534, "learning_rate": 9.519753125267101e-05, "loss": 0.4716456413269043, "memory(GiB)": 61.91, "step": 16415, "token_acc": 0.9037037037037037, "train_speed(iter/s)": 1.480469 }, { "epoch": 0.7034831412535881, "grad_norm": 5.470909595489502, "learning_rate": 9.519465294439805e-05, "loss": 0.40007600784301756, "memory(GiB)": 61.91, "step": 16420, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.480515 }, { "epoch": 0.703697356582837, "grad_norm": 0.6272264122962952, "learning_rate": 9.519177381738158e-05, "loss": 0.2540931224822998, "memory(GiB)": 61.91, "step": 16425, "token_acc": 0.9498069498069498, "train_speed(iter/s)": 1.480565 }, { "epoch": 0.703911571912086, "grad_norm": 4.417055606842041, "learning_rate": 9.518889387167373e-05, "loss": 0.5250715732574462, "memory(GiB)": 61.91, "step": 16430, "token_acc": 0.8885630498533724, "train_speed(iter/s)": 1.480601 }, { "epoch": 0.704125787241335, "grad_norm": 3.5807437896728516, "learning_rate": 9.518601310732667e-05, "loss": 0.5242449283599854, "memory(GiB)": 61.91, "step": 16435, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.4806 }, { "epoch": 0.7043400025705839, "grad_norm": 3.099679946899414, "learning_rate": 9.518313152439259e-05, "loss": 0.5968850612640381, "memory(GiB)": 61.91, "step": 16440, "token_acc": 0.8975155279503105, "train_speed(iter/s)": 1.480601 }, { "epoch": 0.7045542178998329, "grad_norm": 3.103273391723633, "learning_rate": 9.51802491229237e-05, "loss": 0.5999731540679931, "memory(GiB)": 61.91, "step": 16445, "token_acc": 0.88, "train_speed(iter/s)": 1.480608 }, { "epoch": 0.7047684332290819, "grad_norm": 5.614880561828613, "learning_rate": 9.517736590297223e-05, "loss": 0.9482495307922363, "memory(GiB)": 61.91, "step": 16450, "token_acc": 0.8229508196721311, "train_speed(iter/s)": 1.480592 }, { "epoch": 0.7049826485583308, "grad_norm": 3.2638866901397705, "learning_rate": 9.517448186459038e-05, "loss": 0.527281665802002, "memory(GiB)": 61.91, "step": 16455, "token_acc": 0.8931750741839762, "train_speed(iter/s)": 1.480598 }, { "epoch": 0.7051968638875799, "grad_norm": 2.292140483856201, "learning_rate": 9.517159700783042e-05, "loss": 0.16148282289505006, "memory(GiB)": 61.91, "step": 16460, "token_acc": 0.95, "train_speed(iter/s)": 1.480604 }, { "epoch": 0.7054110792168288, "grad_norm": 3.986051321029663, "learning_rate": 9.516871133274461e-05, "loss": 0.40984277725219725, "memory(GiB)": 61.91, "step": 16465, "token_acc": 0.9126984126984127, "train_speed(iter/s)": 1.480605 }, { "epoch": 0.7056252945460777, "grad_norm": 2.7484424114227295, "learning_rate": 9.51658248393852e-05, "loss": 0.31502413749694824, "memory(GiB)": 61.91, "step": 16470, "token_acc": 0.9390581717451524, "train_speed(iter/s)": 1.48056 }, { "epoch": 0.7058395098753267, "grad_norm": 2.631208658218384, "learning_rate": 9.516293752780454e-05, "loss": 0.43389172554016114, "memory(GiB)": 61.91, "step": 16475, "token_acc": 0.9127725856697819, "train_speed(iter/s)": 1.480567 }, { "epoch": 0.7060537252045757, "grad_norm": 2.1906211376190186, "learning_rate": 9.516004939805488e-05, "loss": 0.2535202980041504, "memory(GiB)": 61.91, "step": 16480, "token_acc": 0.9490909090909091, "train_speed(iter/s)": 1.48056 }, { "epoch": 0.7062679405338246, "grad_norm": 4.081280708312988, "learning_rate": 9.515716045018856e-05, "loss": 0.3797840356826782, "memory(GiB)": 61.91, "step": 16485, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.480565 }, { "epoch": 0.7064821558630736, "grad_norm": 3.1640737056732178, "learning_rate": 9.515427068425793e-05, "loss": 0.42271575927734373, "memory(GiB)": 61.91, "step": 16490, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.480539 }, { "epoch": 0.7066963711923225, "grad_norm": 0.6654239892959595, "learning_rate": 9.515138010031532e-05, "loss": 0.45027952194213866, "memory(GiB)": 61.91, "step": 16495, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.480552 }, { "epoch": 0.7069105865215715, "grad_norm": 9.167838096618652, "learning_rate": 9.51484886984131e-05, "loss": 0.653744125366211, "memory(GiB)": 61.91, "step": 16500, "token_acc": 0.8724137931034482, "train_speed(iter/s)": 1.480549 }, { "epoch": 0.7069105865215715, "eval_loss": 2.425776720046997, "eval_runtime": 12.9344, "eval_samples_per_second": 7.731, "eval_steps_per_second": 7.731, "eval_token_acc": 0.4424778761061947, "step": 16500 }, { "epoch": 0.7071248018508205, "grad_norm": 3.6010444164276123, "learning_rate": 9.514559647860366e-05, "loss": 0.446878719329834, "memory(GiB)": 61.91, "step": 16505, "token_acc": 0.5674217907227616, "train_speed(iter/s)": 1.478733 }, { "epoch": 0.7073390171800694, "grad_norm": 5.683206558227539, "learning_rate": 9.514270344093939e-05, "loss": 0.6978508472442627, "memory(GiB)": 61.91, "step": 16510, "token_acc": 0.8552188552188552, "train_speed(iter/s)": 1.478718 }, { "epoch": 0.7075532325093183, "grad_norm": 4.054269313812256, "learning_rate": 9.513980958547269e-05, "loss": 0.21841344833374024, "memory(GiB)": 61.91, "step": 16515, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.478686 }, { "epoch": 0.7077674478385674, "grad_norm": 4.232163906097412, "learning_rate": 9.5136914912256e-05, "loss": 0.3219402313232422, "memory(GiB)": 61.91, "step": 16520, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.478651 }, { "epoch": 0.7079816631678163, "grad_norm": 3.3038744926452637, "learning_rate": 9.513401942134177e-05, "loss": 0.3281768798828125, "memory(GiB)": 61.91, "step": 16525, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.478626 }, { "epoch": 0.7081958784970652, "grad_norm": 2.9385945796966553, "learning_rate": 9.513112311278243e-05, "loss": 0.2951833248138428, "memory(GiB)": 61.91, "step": 16530, "token_acc": 0.9338842975206612, "train_speed(iter/s)": 1.478628 }, { "epoch": 0.7084100938263143, "grad_norm": 3.38702654838562, "learning_rate": 9.512822598663045e-05, "loss": 0.3069169521331787, "memory(GiB)": 61.91, "step": 16535, "token_acc": 0.954225352112676, "train_speed(iter/s)": 1.478644 }, { "epoch": 0.7086243091555632, "grad_norm": 2.326622724533081, "learning_rate": 9.512532804293832e-05, "loss": 0.25887150764465333, "memory(GiB)": 61.91, "step": 16540, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.478623 }, { "epoch": 0.7088385244848121, "grad_norm": 1.7776824235916138, "learning_rate": 9.512242928175857e-05, "loss": 0.16993857622146608, "memory(GiB)": 61.91, "step": 16545, "token_acc": 0.96, "train_speed(iter/s)": 1.478612 }, { "epoch": 0.7090527398140611, "grad_norm": 3.5754122734069824, "learning_rate": 9.511952970314365e-05, "loss": 0.42432847023010256, "memory(GiB)": 61.91, "step": 16550, "token_acc": 0.908256880733945, "train_speed(iter/s)": 1.478659 }, { "epoch": 0.7092669551433101, "grad_norm": 3.1265692710876465, "learning_rate": 9.511662930714614e-05, "loss": 0.27005255222320557, "memory(GiB)": 61.91, "step": 16555, "token_acc": 0.9363957597173145, "train_speed(iter/s)": 1.478639 }, { "epoch": 0.709481170472559, "grad_norm": 3.778080463409424, "learning_rate": 9.511372809381856e-05, "loss": 0.1452672600746155, "memory(GiB)": 61.91, "step": 16560, "token_acc": 0.9702602230483272, "train_speed(iter/s)": 1.47862 }, { "epoch": 0.709695385801808, "grad_norm": 1.500020146369934, "learning_rate": 9.511082606321348e-05, "loss": 0.30020599365234374, "memory(GiB)": 61.91, "step": 16565, "token_acc": 0.9287833827893175, "train_speed(iter/s)": 1.478592 }, { "epoch": 0.7099096011310569, "grad_norm": 0.5511215329170227, "learning_rate": 9.510792321538348e-05, "loss": 0.36534240245819094, "memory(GiB)": 61.91, "step": 16570, "token_acc": 0.9161676646706587, "train_speed(iter/s)": 1.478603 }, { "epoch": 0.7101238164603059, "grad_norm": 4.993802547454834, "learning_rate": 9.510501955038112e-05, "loss": 0.23202104568481446, "memory(GiB)": 61.91, "step": 16575, "token_acc": 0.9425675675675675, "train_speed(iter/s)": 1.47861 }, { "epoch": 0.7103380317895549, "grad_norm": 3.4167864322662354, "learning_rate": 9.510211506825904e-05, "loss": 0.4467190742492676, "memory(GiB)": 61.91, "step": 16580, "token_acc": 0.9022556390977443, "train_speed(iter/s)": 1.478612 }, { "epoch": 0.7105522471188038, "grad_norm": 1.6731311082839966, "learning_rate": 9.509920976906981e-05, "loss": 0.3860288143157959, "memory(GiB)": 61.91, "step": 16585, "token_acc": 0.9329268292682927, "train_speed(iter/s)": 1.478615 }, { "epoch": 0.7107664624480527, "grad_norm": 2.592042922973633, "learning_rate": 9.50963036528661e-05, "loss": 0.3284921169281006, "memory(GiB)": 61.91, "step": 16590, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.478615 }, { "epoch": 0.7109806777773018, "grad_norm": 1.1994259357452393, "learning_rate": 9.509339671970054e-05, "loss": 0.38009142875671387, "memory(GiB)": 61.91, "step": 16595, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.478617 }, { "epoch": 0.7111948931065507, "grad_norm": 5.69264554977417, "learning_rate": 9.50904889696258e-05, "loss": 0.7819517135620118, "memory(GiB)": 61.91, "step": 16600, "token_acc": 0.832089552238806, "train_speed(iter/s)": 1.478687 }, { "epoch": 0.7114091084357996, "grad_norm": 8.653632164001465, "learning_rate": 9.508758040269457e-05, "loss": 0.4432826995849609, "memory(GiB)": 61.91, "step": 16605, "token_acc": 0.9053497942386831, "train_speed(iter/s)": 1.478672 }, { "epoch": 0.7116233237650487, "grad_norm": 1.7594597339630127, "learning_rate": 9.50846710189595e-05, "loss": 0.6085412979125977, "memory(GiB)": 61.91, "step": 16610, "token_acc": 0.8801498127340824, "train_speed(iter/s)": 1.478704 }, { "epoch": 0.7118375390942976, "grad_norm": 3.428231716156006, "learning_rate": 9.508176081847333e-05, "loss": 0.749599552154541, "memory(GiB)": 61.91, "step": 16615, "token_acc": 0.8657718120805369, "train_speed(iter/s)": 1.478714 }, { "epoch": 0.7120517544235465, "grad_norm": 1.9439531564712524, "learning_rate": 9.507884980128879e-05, "loss": 0.6030141353607178, "memory(GiB)": 61.91, "step": 16620, "token_acc": 0.8740740740740741, "train_speed(iter/s)": 1.4787 }, { "epoch": 0.7122659697527955, "grad_norm": 5.608951568603516, "learning_rate": 9.507593796745858e-05, "loss": 0.3460185766220093, "memory(GiB)": 61.91, "step": 16625, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.478767 }, { "epoch": 0.7124801850820445, "grad_norm": 5.1683669090271, "learning_rate": 9.507302531703549e-05, "loss": 0.2166452646255493, "memory(GiB)": 61.91, "step": 16630, "token_acc": 0.948, "train_speed(iter/s)": 1.478769 }, { "epoch": 0.7126944004112934, "grad_norm": 3.6581408977508545, "learning_rate": 9.507011185007224e-05, "loss": 0.3365183353424072, "memory(GiB)": 61.91, "step": 16635, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.478777 }, { "epoch": 0.7129086157405424, "grad_norm": 3.9710147380828857, "learning_rate": 9.506719756662163e-05, "loss": 0.21444578170776368, "memory(GiB)": 61.91, "step": 16640, "token_acc": 0.9552845528455285, "train_speed(iter/s)": 1.478771 }, { "epoch": 0.7131228310697914, "grad_norm": 2.7805449962615967, "learning_rate": 9.50642824667365e-05, "loss": 0.414628791809082, "memory(GiB)": 61.91, "step": 16645, "token_acc": 0.8993055555555556, "train_speed(iter/s)": 1.478781 }, { "epoch": 0.7133370463990403, "grad_norm": 2.2715916633605957, "learning_rate": 9.506136655046957e-05, "loss": 0.2948322772979736, "memory(GiB)": 61.91, "step": 16650, "token_acc": 0.9335443037974683, "train_speed(iter/s)": 1.478764 }, { "epoch": 0.7135512617282893, "grad_norm": 2.723388433456421, "learning_rate": 9.505844981787374e-05, "loss": 0.4392858028411865, "memory(GiB)": 61.91, "step": 16655, "token_acc": 0.9002849002849003, "train_speed(iter/s)": 1.47874 }, { "epoch": 0.7137654770575382, "grad_norm": 4.174272060394287, "learning_rate": 9.505553226900181e-05, "loss": 0.17654931545257568, "memory(GiB)": 61.91, "step": 16660, "token_acc": 0.9671532846715328, "train_speed(iter/s)": 1.47873 }, { "epoch": 0.7139796923867872, "grad_norm": 3.990247964859009, "learning_rate": 9.505261390390667e-05, "loss": 0.29314160346984863, "memory(GiB)": 61.91, "step": 16665, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.478731 }, { "epoch": 0.7141939077160362, "grad_norm": 2.4353983402252197, "learning_rate": 9.504969472264114e-05, "loss": 0.7174554824829101, "memory(GiB)": 61.91, "step": 16670, "token_acc": 0.8859060402684564, "train_speed(iter/s)": 1.478708 }, { "epoch": 0.7144081230452851, "grad_norm": 4.9088945388793945, "learning_rate": 9.504677472525816e-05, "loss": 0.317915153503418, "memory(GiB)": 61.91, "step": 16675, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.47867 }, { "epoch": 0.714622338374534, "grad_norm": 2.1576263904571533, "learning_rate": 9.504385391181058e-05, "loss": 0.39096879959106445, "memory(GiB)": 61.91, "step": 16680, "token_acc": 0.924901185770751, "train_speed(iter/s)": 1.478648 }, { "epoch": 0.7148365537037831, "grad_norm": 2.4548680782318115, "learning_rate": 9.504093228235132e-05, "loss": 0.28214068412780763, "memory(GiB)": 61.91, "step": 16685, "token_acc": 0.9365558912386707, "train_speed(iter/s)": 1.478654 }, { "epoch": 0.715050769033032, "grad_norm": 2.538191556930542, "learning_rate": 9.503800983693334e-05, "loss": 0.30012688636779783, "memory(GiB)": 61.91, "step": 16690, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.478646 }, { "epoch": 0.7152649843622809, "grad_norm": 2.425138473510742, "learning_rate": 9.503508657560956e-05, "loss": 0.37040019035339355, "memory(GiB)": 61.91, "step": 16695, "token_acc": 0.9272151898734177, "train_speed(iter/s)": 1.478683 }, { "epoch": 0.71547919969153, "grad_norm": 3.737959623336792, "learning_rate": 9.503216249843294e-05, "loss": 0.4499513626098633, "memory(GiB)": 61.91, "step": 16700, "token_acc": 0.9208211143695014, "train_speed(iter/s)": 1.478671 }, { "epoch": 0.7156934150207789, "grad_norm": 4.051141262054443, "learning_rate": 9.502923760545644e-05, "loss": 0.4181673049926758, "memory(GiB)": 61.91, "step": 16705, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.478664 }, { "epoch": 0.7159076303500278, "grad_norm": 4.779979228973389, "learning_rate": 9.502631189673307e-05, "loss": 0.5061701774597168, "memory(GiB)": 61.91, "step": 16710, "token_acc": 0.8884758364312267, "train_speed(iter/s)": 1.478694 }, { "epoch": 0.7161218456792768, "grad_norm": 2.3193085193634033, "learning_rate": 9.502338537231583e-05, "loss": 0.4542228698730469, "memory(GiB)": 61.91, "step": 16715, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.47872 }, { "epoch": 0.7163360610085258, "grad_norm": 3.8117752075195312, "learning_rate": 9.502045803225772e-05, "loss": 0.4845607280731201, "memory(GiB)": 61.91, "step": 16720, "token_acc": 0.8855218855218855, "train_speed(iter/s)": 1.478763 }, { "epoch": 0.7165502763377747, "grad_norm": 0.16169549524784088, "learning_rate": 9.501752987661177e-05, "loss": 0.16406153440475463, "memory(GiB)": 61.91, "step": 16725, "token_acc": 0.9571865443425076, "train_speed(iter/s)": 1.478768 }, { "epoch": 0.7167644916670237, "grad_norm": 2.302211284637451, "learning_rate": 9.501460090543105e-05, "loss": 0.25293536186218263, "memory(GiB)": 61.91, "step": 16730, "token_acc": 0.9543973941368078, "train_speed(iter/s)": 1.478779 }, { "epoch": 0.7169787069962726, "grad_norm": 4.780746936798096, "learning_rate": 9.50116711187686e-05, "loss": 0.5391230583190918, "memory(GiB)": 61.91, "step": 16735, "token_acc": 0.9, "train_speed(iter/s)": 1.478808 }, { "epoch": 0.7171929223255216, "grad_norm": 5.386836528778076, "learning_rate": 9.500874051667751e-05, "loss": 0.3859040021896362, "memory(GiB)": 61.91, "step": 16740, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.478851 }, { "epoch": 0.7174071376547706, "grad_norm": 2.6586215496063232, "learning_rate": 9.500580909921086e-05, "loss": 0.36942481994628906, "memory(GiB)": 61.91, "step": 16745, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.478904 }, { "epoch": 0.7176213529840195, "grad_norm": 5.946472644805908, "learning_rate": 9.500287686642177e-05, "loss": 0.6627528190612793, "memory(GiB)": 61.91, "step": 16750, "token_acc": 0.8607594936708861, "train_speed(iter/s)": 1.478867 }, { "epoch": 0.7178355683132684, "grad_norm": 3.149681568145752, "learning_rate": 9.499994381836334e-05, "loss": 0.46861581802368163, "memory(GiB)": 61.91, "step": 16755, "token_acc": 0.8996865203761756, "train_speed(iter/s)": 1.478878 }, { "epoch": 0.7180497836425175, "grad_norm": 2.037706136703491, "learning_rate": 9.499700995508871e-05, "loss": 0.2680126428604126, "memory(GiB)": 61.91, "step": 16760, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.478882 }, { "epoch": 0.7182639989717664, "grad_norm": 0.904888391494751, "learning_rate": 9.499407527665103e-05, "loss": 0.3646254301071167, "memory(GiB)": 61.91, "step": 16765, "token_acc": 0.9292035398230089, "train_speed(iter/s)": 1.478893 }, { "epoch": 0.7184782143010153, "grad_norm": 3.8573648929595947, "learning_rate": 9.499113978310348e-05, "loss": 0.3028018712997437, "memory(GiB)": 61.91, "step": 16770, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.478898 }, { "epoch": 0.7186924296302644, "grad_norm": 3.1975979804992676, "learning_rate": 9.498820347449923e-05, "loss": 0.42732553482055663, "memory(GiB)": 61.91, "step": 16775, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.478893 }, { "epoch": 0.7189066449595133, "grad_norm": 3.8299107551574707, "learning_rate": 9.498526635089147e-05, "loss": 0.36214327812194824, "memory(GiB)": 61.91, "step": 16780, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.47887 }, { "epoch": 0.7191208602887622, "grad_norm": 3.5709424018859863, "learning_rate": 9.498232841233341e-05, "loss": 0.5615796089172364, "memory(GiB)": 61.91, "step": 16785, "token_acc": 0.8941176470588236, "train_speed(iter/s)": 1.478874 }, { "epoch": 0.7193350756180112, "grad_norm": 3.4626235961914062, "learning_rate": 9.497938965887827e-05, "loss": 0.34398274421691893, "memory(GiB)": 61.91, "step": 16790, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.478861 }, { "epoch": 0.7195492909472602, "grad_norm": 2.0272302627563477, "learning_rate": 9.497645009057929e-05, "loss": 0.35986919403076173, "memory(GiB)": 61.91, "step": 16795, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.478874 }, { "epoch": 0.7197635062765092, "grad_norm": 7.792052268981934, "learning_rate": 9.497350970748973e-05, "loss": 0.30637288093566895, "memory(GiB)": 61.91, "step": 16800, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.478892 }, { "epoch": 0.7199777216057581, "grad_norm": 3.2225024700164795, "learning_rate": 9.497056850966286e-05, "loss": 0.2817051887512207, "memory(GiB)": 61.91, "step": 16805, "token_acc": 0.9392712550607287, "train_speed(iter/s)": 1.47891 }, { "epoch": 0.720191936935007, "grad_norm": 3.4371817111968994, "learning_rate": 9.496762649715194e-05, "loss": 0.8064935684204102, "memory(GiB)": 61.91, "step": 16810, "token_acc": 0.8372093023255814, "train_speed(iter/s)": 1.478961 }, { "epoch": 0.7204061522642561, "grad_norm": 3.0468761920928955, "learning_rate": 9.496468367001027e-05, "loss": 0.6901190280914307, "memory(GiB)": 61.91, "step": 16815, "token_acc": 0.8844765342960289, "train_speed(iter/s)": 1.478975 }, { "epoch": 0.720620367593505, "grad_norm": 0.46146515011787415, "learning_rate": 9.496174002829121e-05, "loss": 0.47258877754211426, "memory(GiB)": 61.91, "step": 16820, "token_acc": 0.9049295774647887, "train_speed(iter/s)": 1.478937 }, { "epoch": 0.7208345829227539, "grad_norm": 3.2101874351501465, "learning_rate": 9.495879557204803e-05, "loss": 0.42347307205200196, "memory(GiB)": 61.91, "step": 16825, "token_acc": 0.901060070671378, "train_speed(iter/s)": 1.478926 }, { "epoch": 0.721048798252003, "grad_norm": 1.7745095491409302, "learning_rate": 9.49558503013341e-05, "loss": 0.3780593156814575, "memory(GiB)": 61.91, "step": 16830, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.478943 }, { "epoch": 0.7212630135812519, "grad_norm": 2.729671001434326, "learning_rate": 9.495290421620278e-05, "loss": 0.47467756271362305, "memory(GiB)": 61.91, "step": 16835, "token_acc": 0.8837209302325582, "train_speed(iter/s)": 1.478995 }, { "epoch": 0.7214772289105008, "grad_norm": 2.454490900039673, "learning_rate": 9.494995731670742e-05, "loss": 0.39114794731140134, "memory(GiB)": 61.91, "step": 16840, "token_acc": 0.9107142857142857, "train_speed(iter/s)": 1.478976 }, { "epoch": 0.7216914442397498, "grad_norm": 2.73335862159729, "learning_rate": 9.494700960290141e-05, "loss": 0.26662445068359375, "memory(GiB)": 61.91, "step": 16845, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.478957 }, { "epoch": 0.7219056595689988, "grad_norm": 2.4433934688568115, "learning_rate": 9.494406107483817e-05, "loss": 0.2215137481689453, "memory(GiB)": 61.91, "step": 16850, "token_acc": 0.9515418502202643, "train_speed(iter/s)": 1.478965 }, { "epoch": 0.7221198748982477, "grad_norm": 1.8655914068222046, "learning_rate": 9.49411117325711e-05, "loss": 0.21569766998291015, "memory(GiB)": 61.91, "step": 16855, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.478973 }, { "epoch": 0.7223340902274967, "grad_norm": 2.6329591274261475, "learning_rate": 9.493816157615363e-05, "loss": 0.4536569118499756, "memory(GiB)": 61.91, "step": 16860, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.47897 }, { "epoch": 0.7225483055567457, "grad_norm": 5.293385982513428, "learning_rate": 9.493521060563921e-05, "loss": 0.42713289260864257, "memory(GiB)": 61.91, "step": 16865, "token_acc": 0.9153846153846154, "train_speed(iter/s)": 1.478972 }, { "epoch": 0.7227625208859946, "grad_norm": 0.4555695056915283, "learning_rate": 9.49322588210813e-05, "loss": 0.2052109956741333, "memory(GiB)": 61.91, "step": 16870, "token_acc": 0.9609120521172638, "train_speed(iter/s)": 1.478968 }, { "epoch": 0.7229767362152436, "grad_norm": 0.8822590112686157, "learning_rate": 9.492930622253336e-05, "loss": 0.25387885570526125, "memory(GiB)": 61.91, "step": 16875, "token_acc": 0.937984496124031, "train_speed(iter/s)": 1.478981 }, { "epoch": 0.7231909515444925, "grad_norm": 9.263008117675781, "learning_rate": 9.49263528100489e-05, "loss": 0.4524045467376709, "memory(GiB)": 61.91, "step": 16880, "token_acc": 0.925531914893617, "train_speed(iter/s)": 1.478965 }, { "epoch": 0.7234051668737415, "grad_norm": 11.093587875366211, "learning_rate": 9.492339858368141e-05, "loss": 0.42981886863708496, "memory(GiB)": 61.91, "step": 16885, "token_acc": 0.9015748031496063, "train_speed(iter/s)": 1.479 }, { "epoch": 0.7236193822029905, "grad_norm": 2.848594903945923, "learning_rate": 9.492044354348442e-05, "loss": 0.44877915382385253, "memory(GiB)": 61.91, "step": 16890, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.479015 }, { "epoch": 0.7238335975322394, "grad_norm": 1.795229434967041, "learning_rate": 9.491748768951145e-05, "loss": 0.30874109268188477, "memory(GiB)": 61.91, "step": 16895, "token_acc": 0.9213483146067416, "train_speed(iter/s)": 1.479037 }, { "epoch": 0.7240478128614883, "grad_norm": 7.745279788970947, "learning_rate": 9.491453102181606e-05, "loss": 0.2635792255401611, "memory(GiB)": 61.91, "step": 16900, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.479049 }, { "epoch": 0.7242620281907374, "grad_norm": 0.2829608619213104, "learning_rate": 9.49115735404518e-05, "loss": 0.33588447570800783, "memory(GiB)": 61.91, "step": 16905, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.47904 }, { "epoch": 0.7244762435199863, "grad_norm": 1.8242017030715942, "learning_rate": 9.490861524547225e-05, "loss": 0.32953524589538574, "memory(GiB)": 61.91, "step": 16910, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.479025 }, { "epoch": 0.7246904588492352, "grad_norm": 3.5310580730438232, "learning_rate": 9.490565613693102e-05, "loss": 0.6338255882263184, "memory(GiB)": 61.91, "step": 16915, "token_acc": 0.8633093525179856, "train_speed(iter/s)": 1.479043 }, { "epoch": 0.7249046741784843, "grad_norm": 2.4312288761138916, "learning_rate": 9.49026962148817e-05, "loss": 0.35823655128479004, "memory(GiB)": 61.91, "step": 16920, "token_acc": 0.9297124600638977, "train_speed(iter/s)": 1.479036 }, { "epoch": 0.7251188895077332, "grad_norm": 2.9038209915161133, "learning_rate": 9.489973547937792e-05, "loss": 0.2355365753173828, "memory(GiB)": 61.91, "step": 16925, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.479038 }, { "epoch": 0.7253331048369821, "grad_norm": 0.14195992052555084, "learning_rate": 9.48967739304733e-05, "loss": 0.21048190593719482, "memory(GiB)": 61.91, "step": 16930, "token_acc": 0.9701492537313433, "train_speed(iter/s)": 1.479053 }, { "epoch": 0.7255473201662311, "grad_norm": 1.6530735492706299, "learning_rate": 9.489381156822152e-05, "loss": 0.347200345993042, "memory(GiB)": 61.91, "step": 16935, "token_acc": 0.9404761904761905, "train_speed(iter/s)": 1.479076 }, { "epoch": 0.7257615354954801, "grad_norm": 2.095832109451294, "learning_rate": 9.489084839267621e-05, "loss": 0.3959364414215088, "memory(GiB)": 61.91, "step": 16940, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.479154 }, { "epoch": 0.725975750824729, "grad_norm": 5.884769916534424, "learning_rate": 9.488788440389109e-05, "loss": 0.38692560195922854, "memory(GiB)": 61.91, "step": 16945, "token_acc": 0.9260450160771704, "train_speed(iter/s)": 1.479135 }, { "epoch": 0.726189966153978, "grad_norm": 3.9021294116973877, "learning_rate": 9.488491960191984e-05, "loss": 0.629886245727539, "memory(GiB)": 61.91, "step": 16950, "token_acc": 0.8539682539682539, "train_speed(iter/s)": 1.479129 }, { "epoch": 0.7264041814832269, "grad_norm": 4.22376823425293, "learning_rate": 9.488195398681614e-05, "loss": 0.17271900177001953, "memory(GiB)": 61.91, "step": 16955, "token_acc": 0.9619377162629758, "train_speed(iter/s)": 1.479144 }, { "epoch": 0.7266183968124759, "grad_norm": 1.718185305595398, "learning_rate": 9.487898755863376e-05, "loss": 0.2986595153808594, "memory(GiB)": 61.91, "step": 16960, "token_acc": 0.9, "train_speed(iter/s)": 1.479119 }, { "epoch": 0.7268326121417249, "grad_norm": 3.350248098373413, "learning_rate": 9.487602031742642e-05, "loss": 0.4539472579956055, "memory(GiB)": 61.91, "step": 16965, "token_acc": 0.8896321070234113, "train_speed(iter/s)": 1.479133 }, { "epoch": 0.7270468274709738, "grad_norm": 3.041759490966797, "learning_rate": 9.487305226324786e-05, "loss": 0.3453991889953613, "memory(GiB)": 61.91, "step": 16970, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.479122 }, { "epoch": 0.7272610428002227, "grad_norm": 5.081351280212402, "learning_rate": 9.487008339615187e-05, "loss": 0.3193485736846924, "memory(GiB)": 61.91, "step": 16975, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.479114 }, { "epoch": 0.7274752581294718, "grad_norm": 1.6957095861434937, "learning_rate": 9.486711371619224e-05, "loss": 0.3843179702758789, "memory(GiB)": 61.91, "step": 16980, "token_acc": 0.9261538461538461, "train_speed(iter/s)": 1.479143 }, { "epoch": 0.7276894734587207, "grad_norm": 4.409252166748047, "learning_rate": 9.486414322342274e-05, "loss": 0.6688547134399414, "memory(GiB)": 61.91, "step": 16985, "token_acc": 0.8718861209964412, "train_speed(iter/s)": 1.479172 }, { "epoch": 0.7279036887879696, "grad_norm": 3.892345428466797, "learning_rate": 9.486117191789721e-05, "loss": 0.27278711795806887, "memory(GiB)": 61.91, "step": 16990, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.47923 }, { "epoch": 0.7281179041172187, "grad_norm": 0.8310908079147339, "learning_rate": 9.485819979966946e-05, "loss": 0.3929093599319458, "memory(GiB)": 61.91, "step": 16995, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.479224 }, { "epoch": 0.7283321194464676, "grad_norm": 11.09154987335205, "learning_rate": 9.485522686879336e-05, "loss": 0.47129507064819337, "memory(GiB)": 61.91, "step": 17000, "token_acc": 0.8885350318471338, "train_speed(iter/s)": 1.479269 }, { "epoch": 0.7283321194464676, "eval_loss": 2.3239338397979736, "eval_runtime": 13.1954, "eval_samples_per_second": 7.578, "eval_steps_per_second": 7.578, "eval_token_acc": 0.44224924012158057, "step": 17000 }, { "epoch": 0.7285463347757165, "grad_norm": 2.3646836280822754, "learning_rate": 9.485225312532274e-05, "loss": 0.5024572372436523, "memory(GiB)": 61.91, "step": 17005, "token_acc": 0.5814931650893796, "train_speed(iter/s)": 1.477465 }, { "epoch": 0.7287605501049655, "grad_norm": 1.2927544116973877, "learning_rate": 9.484927856931146e-05, "loss": 0.45227856636047364, "memory(GiB)": 61.91, "step": 17010, "token_acc": 0.9, "train_speed(iter/s)": 1.477473 }, { "epoch": 0.7289747654342145, "grad_norm": 1.175212025642395, "learning_rate": 9.484630320081343e-05, "loss": 0.33919315338134765, "memory(GiB)": 61.91, "step": 17015, "token_acc": 0.9388489208633094, "train_speed(iter/s)": 1.477481 }, { "epoch": 0.7291889807634634, "grad_norm": 3.0363781452178955, "learning_rate": 9.484332701988257e-05, "loss": 0.1452466607093811, "memory(GiB)": 61.91, "step": 17020, "token_acc": 0.9737827715355806, "train_speed(iter/s)": 1.477534 }, { "epoch": 0.7294031960927124, "grad_norm": 4.164963722229004, "learning_rate": 9.484035002657276e-05, "loss": 0.377247953414917, "memory(GiB)": 61.91, "step": 17025, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.477526 }, { "epoch": 0.7296174114219613, "grad_norm": 4.539981842041016, "learning_rate": 9.483737222093794e-05, "loss": 0.5046974658966065, "memory(GiB)": 61.91, "step": 17030, "token_acc": 0.9022556390977443, "train_speed(iter/s)": 1.477546 }, { "epoch": 0.7298316267512103, "grad_norm": 3.188782215118408, "learning_rate": 9.483439360303208e-05, "loss": 0.36030192375183107, "memory(GiB)": 61.91, "step": 17035, "token_acc": 0.9279279279279279, "train_speed(iter/s)": 1.477558 }, { "epoch": 0.7300458420804593, "grad_norm": 7.767899513244629, "learning_rate": 9.48314141729091e-05, "loss": 0.41483268737792967, "memory(GiB)": 61.91, "step": 17040, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.477547 }, { "epoch": 0.7302600574097082, "grad_norm": 8.000993728637695, "learning_rate": 9.482843393062302e-05, "loss": 0.4156181335449219, "memory(GiB)": 61.91, "step": 17045, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.47754 }, { "epoch": 0.7304742727389572, "grad_norm": 1.8839514255523682, "learning_rate": 9.482545287622779e-05, "loss": 0.3313511610031128, "memory(GiB)": 61.91, "step": 17050, "token_acc": 0.9552238805970149, "train_speed(iter/s)": 1.477578 }, { "epoch": 0.7306884880682062, "grad_norm": 0.5159255862236023, "learning_rate": 9.482247100977744e-05, "loss": 0.47060284614562986, "memory(GiB)": 61.91, "step": 17055, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.477569 }, { "epoch": 0.7309027033974551, "grad_norm": 5.750687122344971, "learning_rate": 9.481948833132597e-05, "loss": 0.6924715995788574, "memory(GiB)": 61.91, "step": 17060, "token_acc": 0.8651685393258427, "train_speed(iter/s)": 1.477565 }, { "epoch": 0.731116918726704, "grad_norm": 6.304527759552002, "learning_rate": 9.481650484092744e-05, "loss": 0.45029525756835936, "memory(GiB)": 61.91, "step": 17065, "token_acc": 0.900709219858156, "train_speed(iter/s)": 1.477574 }, { "epoch": 0.7313311340559531, "grad_norm": 1.9238650798797607, "learning_rate": 9.481352053863585e-05, "loss": 0.5572210788726807, "memory(GiB)": 61.91, "step": 17070, "token_acc": 0.9109311740890689, "train_speed(iter/s)": 1.477601 }, { "epoch": 0.731545349385202, "grad_norm": 3.1551709175109863, "learning_rate": 9.481053542450533e-05, "loss": 0.30603413581848143, "memory(GiB)": 61.91, "step": 17075, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.477648 }, { "epoch": 0.7317595647144509, "grad_norm": 3.1677050590515137, "learning_rate": 9.480754949858992e-05, "loss": 0.3045662879943848, "memory(GiB)": 61.91, "step": 17080, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.477653 }, { "epoch": 0.7319737800437, "grad_norm": 0.8074802756309509, "learning_rate": 9.480456276094372e-05, "loss": 0.40825982093811036, "memory(GiB)": 61.91, "step": 17085, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.477654 }, { "epoch": 0.7321879953729489, "grad_norm": 2.906280517578125, "learning_rate": 9.480157521162084e-05, "loss": 0.4046309471130371, "memory(GiB)": 61.91, "step": 17090, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.477643 }, { "epoch": 0.7324022107021978, "grad_norm": 0.04284346103668213, "learning_rate": 9.479858685067538e-05, "loss": 0.3298440456390381, "memory(GiB)": 61.91, "step": 17095, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.477654 }, { "epoch": 0.7326164260314468, "grad_norm": 4.901258945465088, "learning_rate": 9.479559767816151e-05, "loss": 0.49830193519592286, "memory(GiB)": 61.91, "step": 17100, "token_acc": 0.8892857142857142, "train_speed(iter/s)": 1.477638 }, { "epoch": 0.7328306413606958, "grad_norm": 1.7963131666183472, "learning_rate": 9.479260769413335e-05, "loss": 0.13559969663619995, "memory(GiB)": 61.91, "step": 17105, "token_acc": 0.9712230215827338, "train_speed(iter/s)": 1.477681 }, { "epoch": 0.7330448566899447, "grad_norm": 2.9472110271453857, "learning_rate": 9.47896168986451e-05, "loss": 0.24849390983581543, "memory(GiB)": 61.91, "step": 17110, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.477656 }, { "epoch": 0.7332590720191937, "grad_norm": 4.483307361602783, "learning_rate": 9.478662529175091e-05, "loss": 0.29392037391662595, "memory(GiB)": 61.91, "step": 17115, "token_acc": 0.9297124600638977, "train_speed(iter/s)": 1.477657 }, { "epoch": 0.7334732873484426, "grad_norm": 0.30746111273765564, "learning_rate": 9.4783632873505e-05, "loss": 0.4034391403198242, "memory(GiB)": 61.91, "step": 17120, "token_acc": 0.9172932330827067, "train_speed(iter/s)": 1.477677 }, { "epoch": 0.7336875026776916, "grad_norm": 4.617684841156006, "learning_rate": 9.478063964396156e-05, "loss": 0.5957664489746094, "memory(GiB)": 61.91, "step": 17125, "token_acc": 0.8836477987421384, "train_speed(iter/s)": 1.4777 }, { "epoch": 0.7339017180069406, "grad_norm": 2.0648787021636963, "learning_rate": 9.477764560317483e-05, "loss": 0.31573073863983153, "memory(GiB)": 61.91, "step": 17130, "token_acc": 0.9378698224852071, "train_speed(iter/s)": 1.4777 }, { "epoch": 0.7341159333361895, "grad_norm": 6.537891387939453, "learning_rate": 9.477465075119905e-05, "loss": 0.21034841537475585, "memory(GiB)": 61.91, "step": 17135, "token_acc": 0.9613899613899614, "train_speed(iter/s)": 1.477713 }, { "epoch": 0.7343301486654386, "grad_norm": 1.7473180294036865, "learning_rate": 9.477165508808847e-05, "loss": 0.34116837978363035, "memory(GiB)": 61.91, "step": 17140, "token_acc": 0.93125, "train_speed(iter/s)": 1.477703 }, { "epoch": 0.7345443639946875, "grad_norm": 2.908355951309204, "learning_rate": 9.476865861389735e-05, "loss": 0.5733506202697753, "memory(GiB)": 61.91, "step": 17145, "token_acc": 0.8905109489051095, "train_speed(iter/s)": 1.477678 }, { "epoch": 0.7347585793239364, "grad_norm": 4.829767227172852, "learning_rate": 9.476566132867999e-05, "loss": 0.5183637619018555, "memory(GiB)": 61.91, "step": 17150, "token_acc": 0.8870292887029289, "train_speed(iter/s)": 1.477682 }, { "epoch": 0.7349727946531854, "grad_norm": 2.7687861919403076, "learning_rate": 9.476266323249068e-05, "loss": 0.2762128829956055, "memory(GiB)": 61.91, "step": 17155, "token_acc": 0.9433333333333334, "train_speed(iter/s)": 1.477667 }, { "epoch": 0.7351870099824344, "grad_norm": 2.258364200592041, "learning_rate": 9.475966432538372e-05, "loss": 0.3654314517974854, "memory(GiB)": 61.91, "step": 17160, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.477643 }, { "epoch": 0.7354012253116833, "grad_norm": 2.8369171619415283, "learning_rate": 9.475666460741348e-05, "loss": 0.30525870323181153, "memory(GiB)": 61.91, "step": 17165, "token_acc": 0.9392712550607287, "train_speed(iter/s)": 1.477625 }, { "epoch": 0.7356154406409323, "grad_norm": 2.965449333190918, "learning_rate": 9.475366407863427e-05, "loss": 0.2810215473175049, "memory(GiB)": 61.91, "step": 17170, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.477645 }, { "epoch": 0.7358296559701812, "grad_norm": 3.6696553230285645, "learning_rate": 9.475066273910043e-05, "loss": 0.29972167015075685, "memory(GiB)": 61.91, "step": 17175, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.47764 }, { "epoch": 0.7360438712994302, "grad_norm": 5.074547290802002, "learning_rate": 9.474766058886637e-05, "loss": 0.42393884658813474, "memory(GiB)": 61.91, "step": 17180, "token_acc": 0.9149659863945578, "train_speed(iter/s)": 1.477629 }, { "epoch": 0.7362580866286792, "grad_norm": 2.2468879222869873, "learning_rate": 9.474465762798646e-05, "loss": 0.2680697441101074, "memory(GiB)": 61.91, "step": 17185, "token_acc": 0.9457364341085271, "train_speed(iter/s)": 1.47764 }, { "epoch": 0.7364723019579281, "grad_norm": 2.8894429206848145, "learning_rate": 9.474165385651511e-05, "loss": 0.5299309730529785, "memory(GiB)": 61.91, "step": 17190, "token_acc": 0.892128279883382, "train_speed(iter/s)": 1.47764 }, { "epoch": 0.736686517287177, "grad_norm": 4.193004131317139, "learning_rate": 9.473864927450673e-05, "loss": 0.5200443267822266, "memory(GiB)": 61.91, "step": 17195, "token_acc": 0.8835341365461847, "train_speed(iter/s)": 1.477641 }, { "epoch": 0.7369007326164261, "grad_norm": 2.5429577827453613, "learning_rate": 9.473564388201575e-05, "loss": 0.4157592296600342, "memory(GiB)": 61.91, "step": 17200, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.477696 }, { "epoch": 0.737114947945675, "grad_norm": 3.172999858856201, "learning_rate": 9.473263767909659e-05, "loss": 0.2731677770614624, "memory(GiB)": 61.91, "step": 17205, "token_acc": 0.9369085173501577, "train_speed(iter/s)": 1.477738 }, { "epoch": 0.7373291632749239, "grad_norm": 6.126654148101807, "learning_rate": 9.472963066580375e-05, "loss": 0.3269988536834717, "memory(GiB)": 61.91, "step": 17210, "token_acc": 0.9283018867924528, "train_speed(iter/s)": 1.477741 }, { "epoch": 0.737543378604173, "grad_norm": 2.791569471359253, "learning_rate": 9.472662284219171e-05, "loss": 0.2903048276901245, "memory(GiB)": 61.91, "step": 17215, "token_acc": 0.9409722222222222, "train_speed(iter/s)": 1.477745 }, { "epoch": 0.7377575939334219, "grad_norm": 1.3037389516830444, "learning_rate": 9.472361420831493e-05, "loss": 0.2257984161376953, "memory(GiB)": 61.91, "step": 17220, "token_acc": 0.9563758389261745, "train_speed(iter/s)": 1.477745 }, { "epoch": 0.7379718092626708, "grad_norm": 0.4125943183898926, "learning_rate": 9.472060476422791e-05, "loss": 0.19991921186447142, "memory(GiB)": 61.91, "step": 17225, "token_acc": 0.9644012944983819, "train_speed(iter/s)": 1.477748 }, { "epoch": 0.7381860245919198, "grad_norm": 0.6402518153190613, "learning_rate": 9.471759450998521e-05, "loss": 0.33398435115814207, "memory(GiB)": 61.91, "step": 17230, "token_acc": 0.9380530973451328, "train_speed(iter/s)": 1.47776 }, { "epoch": 0.7384002399211688, "grad_norm": 9.231948852539062, "learning_rate": 9.471458344564132e-05, "loss": 0.33410935401916503, "memory(GiB)": 61.91, "step": 17235, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.477759 }, { "epoch": 0.7386144552504177, "grad_norm": 3.824552059173584, "learning_rate": 9.47115715712508e-05, "loss": 0.4316185474395752, "memory(GiB)": 61.91, "step": 17240, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.477731 }, { "epoch": 0.7388286705796667, "grad_norm": 3.7314553260803223, "learning_rate": 9.470855888686822e-05, "loss": 0.6067300796508789, "memory(GiB)": 61.91, "step": 17245, "token_acc": 0.8674698795180723, "train_speed(iter/s)": 1.477725 }, { "epoch": 0.7390428859089156, "grad_norm": 2.727473020553589, "learning_rate": 9.470554539254816e-05, "loss": 0.46352238655090333, "memory(GiB)": 61.91, "step": 17250, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.477719 }, { "epoch": 0.7392571012381646, "grad_norm": 4.518321990966797, "learning_rate": 9.470253108834522e-05, "loss": 0.4388858795166016, "memory(GiB)": 61.91, "step": 17255, "token_acc": 0.9116809116809117, "train_speed(iter/s)": 1.477723 }, { "epoch": 0.7394713165674136, "grad_norm": 3.36692214012146, "learning_rate": 9.469951597431398e-05, "loss": 0.49880733489990237, "memory(GiB)": 61.91, "step": 17260, "token_acc": 0.8968481375358166, "train_speed(iter/s)": 1.477772 }, { "epoch": 0.7396855318966625, "grad_norm": 2.9298083782196045, "learning_rate": 9.469650005050907e-05, "loss": 0.424791145324707, "memory(GiB)": 61.91, "step": 17265, "token_acc": 0.9006622516556292, "train_speed(iter/s)": 1.477743 }, { "epoch": 0.7398997472259115, "grad_norm": 3.5731608867645264, "learning_rate": 9.469348331698516e-05, "loss": 0.34513149261474607, "memory(GiB)": 61.91, "step": 17270, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.477726 }, { "epoch": 0.7401139625551605, "grad_norm": 2.967603921890259, "learning_rate": 9.469046577379685e-05, "loss": 0.5188796520233154, "memory(GiB)": 61.91, "step": 17275, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.477722 }, { "epoch": 0.7403281778844094, "grad_norm": 1.688162922859192, "learning_rate": 9.468744742099885e-05, "loss": 0.2880302906036377, "memory(GiB)": 61.91, "step": 17280, "token_acc": 0.9357429718875502, "train_speed(iter/s)": 1.477723 }, { "epoch": 0.7405423932136583, "grad_norm": 3.26340913772583, "learning_rate": 9.468442825864578e-05, "loss": 0.18981293439865113, "memory(GiB)": 61.91, "step": 17285, "token_acc": 0.9606557377049181, "train_speed(iter/s)": 1.477727 }, { "epoch": 0.7407566085429074, "grad_norm": 2.622612237930298, "learning_rate": 9.468140828679242e-05, "loss": 0.3792809724807739, "memory(GiB)": 61.91, "step": 17290, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.477732 }, { "epoch": 0.7409708238721563, "grad_norm": 1.1876327991485596, "learning_rate": 9.467838750549341e-05, "loss": 0.2549848794937134, "memory(GiB)": 61.91, "step": 17295, "token_acc": 0.9288135593220339, "train_speed(iter/s)": 1.477717 }, { "epoch": 0.7411850392014052, "grad_norm": 5.4677629470825195, "learning_rate": 9.46753659148035e-05, "loss": 0.4644014358520508, "memory(GiB)": 61.91, "step": 17300, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.477764 }, { "epoch": 0.7413992545306543, "grad_norm": 2.130826473236084, "learning_rate": 9.467234351477743e-05, "loss": 0.08875709772109985, "memory(GiB)": 61.91, "step": 17305, "token_acc": 0.9795221843003413, "train_speed(iter/s)": 1.477781 }, { "epoch": 0.7416134698599032, "grad_norm": 2.095438241958618, "learning_rate": 9.466932030546996e-05, "loss": 0.308350133895874, "memory(GiB)": 61.91, "step": 17310, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.477783 }, { "epoch": 0.7418276851891521, "grad_norm": 2.5861356258392334, "learning_rate": 9.466629628693584e-05, "loss": 0.2735862731933594, "memory(GiB)": 61.91, "step": 17315, "token_acc": 0.9385964912280702, "train_speed(iter/s)": 1.477791 }, { "epoch": 0.7420419005184011, "grad_norm": 2.66814923286438, "learning_rate": 9.466327145922986e-05, "loss": 0.2877746820449829, "memory(GiB)": 61.91, "step": 17320, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.477789 }, { "epoch": 0.74225611584765, "grad_norm": 4.614184379577637, "learning_rate": 9.466024582240683e-05, "loss": 0.48297271728515623, "memory(GiB)": 61.91, "step": 17325, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.477844 }, { "epoch": 0.742470331176899, "grad_norm": 4.332878112792969, "learning_rate": 9.465721937652155e-05, "loss": 0.32562201023101806, "memory(GiB)": 61.91, "step": 17330, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.47783 }, { "epoch": 0.742684546506148, "grad_norm": 2.2279374599456787, "learning_rate": 9.465419212162883e-05, "loss": 0.5097407817840576, "memory(GiB)": 61.91, "step": 17335, "token_acc": 0.903010033444816, "train_speed(iter/s)": 1.477817 }, { "epoch": 0.7428987618353969, "grad_norm": 2.414546489715576, "learning_rate": 9.465116405778357e-05, "loss": 0.3425808668136597, "memory(GiB)": 61.91, "step": 17340, "token_acc": 0.9293478260869565, "train_speed(iter/s)": 1.4778 }, { "epoch": 0.7431129771646459, "grad_norm": 1.8277654647827148, "learning_rate": 9.464813518504055e-05, "loss": 0.6318798065185547, "memory(GiB)": 61.91, "step": 17345, "token_acc": 0.8637873754152824, "train_speed(iter/s)": 1.477839 }, { "epoch": 0.7433271924938949, "grad_norm": 6.9997382164001465, "learning_rate": 9.464510550345469e-05, "loss": 0.321658992767334, "memory(GiB)": 61.91, "step": 17350, "token_acc": 0.9461538461538461, "train_speed(iter/s)": 1.477807 }, { "epoch": 0.7435414078231438, "grad_norm": 3.0880422592163086, "learning_rate": 9.464207501308086e-05, "loss": 0.4595590114593506, "memory(GiB)": 61.91, "step": 17355, "token_acc": 0.9072847682119205, "train_speed(iter/s)": 1.47782 }, { "epoch": 0.7437556231523927, "grad_norm": 2.7987210750579834, "learning_rate": 9.463904371397397e-05, "loss": 0.4362345218658447, "memory(GiB)": 61.91, "step": 17360, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.477821 }, { "epoch": 0.7439698384816418, "grad_norm": 0.3616732358932495, "learning_rate": 9.463601160618892e-05, "loss": 0.5091505527496338, "memory(GiB)": 61.91, "step": 17365, "token_acc": 0.9071428571428571, "train_speed(iter/s)": 1.477829 }, { "epoch": 0.7441840538108907, "grad_norm": 5.983371257781982, "learning_rate": 9.463297868978066e-05, "loss": 0.32327704429626464, "memory(GiB)": 61.91, "step": 17370, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.47789 }, { "epoch": 0.7443982691401396, "grad_norm": 4.627436637878418, "learning_rate": 9.462994496480411e-05, "loss": 0.45624513626098634, "memory(GiB)": 61.91, "step": 17375, "token_acc": 0.8945454545454545, "train_speed(iter/s)": 1.477887 }, { "epoch": 0.7446124844693887, "grad_norm": 5.4143290519714355, "learning_rate": 9.462691043131424e-05, "loss": 0.3950221300125122, "memory(GiB)": 61.91, "step": 17380, "token_acc": 0.9054545454545454, "train_speed(iter/s)": 1.477893 }, { "epoch": 0.7448266997986376, "grad_norm": 5.77822208404541, "learning_rate": 9.462387508936605e-05, "loss": 0.46863589286804197, "memory(GiB)": 61.91, "step": 17385, "token_acc": 0.9083665338645418, "train_speed(iter/s)": 1.477892 }, { "epoch": 0.7450409151278865, "grad_norm": 2.396498441696167, "learning_rate": 9.462083893901448e-05, "loss": 0.3713083028793335, "memory(GiB)": 61.91, "step": 17390, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.477904 }, { "epoch": 0.7452551304571355, "grad_norm": 2.9574763774871826, "learning_rate": 9.461780198031456e-05, "loss": 0.2941826105117798, "memory(GiB)": 61.91, "step": 17395, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.47791 }, { "epoch": 0.7454693457863845, "grad_norm": 2.1571266651153564, "learning_rate": 9.46147642133213e-05, "loss": 0.5120114326477051, "memory(GiB)": 61.91, "step": 17400, "token_acc": 0.9042145593869731, "train_speed(iter/s)": 1.477908 }, { "epoch": 0.7456835611156334, "grad_norm": 2.74545955657959, "learning_rate": 9.461172563808973e-05, "loss": 0.2208326578140259, "memory(GiB)": 61.91, "step": 17405, "token_acc": 0.9469026548672567, "train_speed(iter/s)": 1.4779 }, { "epoch": 0.7458977764448824, "grad_norm": 11.954442024230957, "learning_rate": 9.46086862546749e-05, "loss": 0.41226611137390134, "memory(GiB)": 61.91, "step": 17410, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.477952 }, { "epoch": 0.7461119917741313, "grad_norm": 5.097442626953125, "learning_rate": 9.460564606313189e-05, "loss": 0.49616084098815916, "memory(GiB)": 61.91, "step": 17415, "token_acc": 0.9124579124579124, "train_speed(iter/s)": 1.478018 }, { "epoch": 0.7463262071033803, "grad_norm": 2.71186900138855, "learning_rate": 9.460260506351573e-05, "loss": 0.35230896472930906, "memory(GiB)": 61.91, "step": 17420, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.477986 }, { "epoch": 0.7465404224326293, "grad_norm": 2.9569613933563232, "learning_rate": 9.459956325588155e-05, "loss": 0.5140594005584717, "memory(GiB)": 61.91, "step": 17425, "token_acc": 0.8745519713261649, "train_speed(iter/s)": 1.478005 }, { "epoch": 0.7467546377618782, "grad_norm": 2.5948476791381836, "learning_rate": 9.459652064028445e-05, "loss": 0.3534525156021118, "memory(GiB)": 61.91, "step": 17430, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.478023 }, { "epoch": 0.7469688530911271, "grad_norm": 8.797500610351562, "learning_rate": 9.459347721677954e-05, "loss": 0.28751847743988035, "memory(GiB)": 61.91, "step": 17435, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.478007 }, { "epoch": 0.7471830684203762, "grad_norm": 1.7213914394378662, "learning_rate": 9.459043298542196e-05, "loss": 0.301419734954834, "memory(GiB)": 61.91, "step": 17440, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.478015 }, { "epoch": 0.7473972837496251, "grad_norm": 3.540888547897339, "learning_rate": 9.458738794626687e-05, "loss": 0.4858253479003906, "memory(GiB)": 61.91, "step": 17445, "token_acc": 0.9163346613545816, "train_speed(iter/s)": 1.478072 }, { "epoch": 0.747611499078874, "grad_norm": 4.56614875793457, "learning_rate": 9.45843420993694e-05, "loss": 0.4814272880554199, "memory(GiB)": 61.91, "step": 17450, "token_acc": 0.9021739130434783, "train_speed(iter/s)": 1.478059 }, { "epoch": 0.7478257144081231, "grad_norm": 4.591188907623291, "learning_rate": 9.458129544478476e-05, "loss": 0.31876330375671386, "memory(GiB)": 61.91, "step": 17455, "token_acc": 0.9050632911392406, "train_speed(iter/s)": 1.478065 }, { "epoch": 0.748039929737372, "grad_norm": 0.9824604392051697, "learning_rate": 9.457824798256813e-05, "loss": 0.44881691932678225, "memory(GiB)": 61.91, "step": 17460, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.478015 }, { "epoch": 0.7482541450666209, "grad_norm": 5.891805648803711, "learning_rate": 9.457519971277471e-05, "loss": 0.5973117828369141, "memory(GiB)": 61.91, "step": 17465, "token_acc": 0.8685015290519877, "train_speed(iter/s)": 1.478084 }, { "epoch": 0.74846836039587, "grad_norm": 1.2742502689361572, "learning_rate": 9.457215063545974e-05, "loss": 0.3084941148757935, "memory(GiB)": 61.91, "step": 17470, "token_acc": 0.9429530201342282, "train_speed(iter/s)": 1.478122 }, { "epoch": 0.7486825757251189, "grad_norm": 0.14228518307209015, "learning_rate": 9.456910075067847e-05, "loss": 0.5034667015075683, "memory(GiB)": 61.91, "step": 17475, "token_acc": 0.8797250859106529, "train_speed(iter/s)": 1.478172 }, { "epoch": 0.7488967910543679, "grad_norm": 3.5319247245788574, "learning_rate": 9.45660500584861e-05, "loss": 0.3700082540512085, "memory(GiB)": 61.91, "step": 17480, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.478166 }, { "epoch": 0.7491110063836168, "grad_norm": 4.446175575256348, "learning_rate": 9.456299855893794e-05, "loss": 0.5079458236694336, "memory(GiB)": 61.91, "step": 17485, "token_acc": 0.8812260536398467, "train_speed(iter/s)": 1.47813 }, { "epoch": 0.7493252217128658, "grad_norm": 8.405726432800293, "learning_rate": 9.455994625208926e-05, "loss": 0.3055213451385498, "memory(GiB)": 61.91, "step": 17490, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.4782 }, { "epoch": 0.7495394370421148, "grad_norm": 0.6327483654022217, "learning_rate": 9.455689313799535e-05, "loss": 0.4532972812652588, "memory(GiB)": 61.91, "step": 17495, "token_acc": 0.8877887788778878, "train_speed(iter/s)": 1.47822 }, { "epoch": 0.7497536523713637, "grad_norm": 1.654052734375, "learning_rate": 9.455383921671153e-05, "loss": 0.34325644969940183, "memory(GiB)": 61.91, "step": 17500, "token_acc": 0.928125, "train_speed(iter/s)": 1.478212 }, { "epoch": 0.7497536523713637, "eval_loss": 2.3542885780334473, "eval_runtime": 13.4713, "eval_samples_per_second": 7.423, "eval_steps_per_second": 7.423, "eval_token_acc": 0.48419540229885055, "step": 17500 }, { "epoch": 0.7499678677006126, "grad_norm": 2.0985755920410156, "learning_rate": 9.45507844882931e-05, "loss": 0.47764906883239744, "memory(GiB)": 61.91, "step": 17505, "token_acc": 0.5901981230448383, "train_speed(iter/s)": 1.47643 }, { "epoch": 0.7501820830298617, "grad_norm": 1.7728626728057861, "learning_rate": 9.454772895279543e-05, "loss": 0.32654538154602053, "memory(GiB)": 61.91, "step": 17510, "token_acc": 0.9256198347107438, "train_speed(iter/s)": 1.47641 }, { "epoch": 0.7503962983591106, "grad_norm": 3.0053038597106934, "learning_rate": 9.454467261027386e-05, "loss": 0.3941663265228271, "memory(GiB)": 61.91, "step": 17515, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.476423 }, { "epoch": 0.7506105136883595, "grad_norm": 1.0535264015197754, "learning_rate": 9.454161546078375e-05, "loss": 0.4197384834289551, "memory(GiB)": 61.91, "step": 17520, "token_acc": 0.8901515151515151, "train_speed(iter/s)": 1.476423 }, { "epoch": 0.7508247290176085, "grad_norm": 0.21003127098083496, "learning_rate": 9.453855750438049e-05, "loss": 0.3816428899765015, "memory(GiB)": 61.91, "step": 17525, "token_acc": 0.9088145896656535, "train_speed(iter/s)": 1.47643 }, { "epoch": 0.7510389443468575, "grad_norm": 2.819819450378418, "learning_rate": 9.45354987411195e-05, "loss": 0.34066023826599123, "memory(GiB)": 61.91, "step": 17530, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.476438 }, { "epoch": 0.7512531596761064, "grad_norm": 0.44067782163619995, "learning_rate": 9.453243917105617e-05, "loss": 0.29637837409973145, "memory(GiB)": 61.91, "step": 17535, "token_acc": 0.9138888888888889, "train_speed(iter/s)": 1.476429 }, { "epoch": 0.7514673750053554, "grad_norm": 3.7418322563171387, "learning_rate": 9.452937879424592e-05, "loss": 0.37384185791015623, "memory(GiB)": 61.91, "step": 17540, "token_acc": 0.9179331306990881, "train_speed(iter/s)": 1.476452 }, { "epoch": 0.7516815903346044, "grad_norm": 4.822911262512207, "learning_rate": 9.452631761074419e-05, "loss": 0.6348400592803956, "memory(GiB)": 61.91, "step": 17545, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.476439 }, { "epoch": 0.7518958056638533, "grad_norm": 4.522646427154541, "learning_rate": 9.452325562060647e-05, "loss": 0.34326183795928955, "memory(GiB)": 61.91, "step": 17550, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.476422 }, { "epoch": 0.7521100209931023, "grad_norm": 3.580265998840332, "learning_rate": 9.45201928238882e-05, "loss": 0.24520506858825683, "memory(GiB)": 61.91, "step": 17555, "token_acc": 0.9490445859872612, "train_speed(iter/s)": 1.476406 }, { "epoch": 0.7523242363223512, "grad_norm": 2.7262051105499268, "learning_rate": 9.451712922064488e-05, "loss": 0.4785344123840332, "memory(GiB)": 61.91, "step": 17560, "token_acc": 0.8896321070234113, "train_speed(iter/s)": 1.476406 }, { "epoch": 0.7525384516516002, "grad_norm": 2.6038026809692383, "learning_rate": 9.4514064810932e-05, "loss": 0.34695513248443605, "memory(GiB)": 61.91, "step": 17565, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.476376 }, { "epoch": 0.7527526669808492, "grad_norm": 1.3370826244354248, "learning_rate": 9.451099959480507e-05, "loss": 0.34910407066345217, "memory(GiB)": 61.91, "step": 17570, "token_acc": 0.9289940828402367, "train_speed(iter/s)": 1.476369 }, { "epoch": 0.7529668823100981, "grad_norm": 0.3969125747680664, "learning_rate": 9.450793357231962e-05, "loss": 0.4009547233581543, "memory(GiB)": 61.91, "step": 17575, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.476336 }, { "epoch": 0.753181097639347, "grad_norm": 3.830777645111084, "learning_rate": 9.450486674353122e-05, "loss": 0.47237286567687986, "memory(GiB)": 61.91, "step": 17580, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.47633 }, { "epoch": 0.7533953129685961, "grad_norm": 4.370207786560059, "learning_rate": 9.450179910849541e-05, "loss": 0.6460176467895508, "memory(GiB)": 61.91, "step": 17585, "token_acc": 0.8319559228650137, "train_speed(iter/s)": 1.476311 }, { "epoch": 0.753609528297845, "grad_norm": 0.672476053237915, "learning_rate": 9.449873066726775e-05, "loss": 0.38372995853424074, "memory(GiB)": 61.91, "step": 17590, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.476321 }, { "epoch": 0.7538237436270939, "grad_norm": 1.7714412212371826, "learning_rate": 9.449566141990384e-05, "loss": 0.34878993034362793, "memory(GiB)": 61.91, "step": 17595, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.476329 }, { "epoch": 0.754037958956343, "grad_norm": 0.06497224420309067, "learning_rate": 9.449259136645929e-05, "loss": 0.48102560043334963, "memory(GiB)": 61.91, "step": 17600, "token_acc": 0.9095238095238095, "train_speed(iter/s)": 1.476334 }, { "epoch": 0.7542521742855919, "grad_norm": 1.3018341064453125, "learning_rate": 9.448952050698972e-05, "loss": 0.4425107479095459, "memory(GiB)": 61.91, "step": 17605, "token_acc": 0.8921933085501859, "train_speed(iter/s)": 1.476388 }, { "epoch": 0.7544663896148408, "grad_norm": 2.3196256160736084, "learning_rate": 9.448644884155075e-05, "loss": 0.3804525375366211, "memory(GiB)": 61.91, "step": 17610, "token_acc": 0.9025157232704403, "train_speed(iter/s)": 1.476406 }, { "epoch": 0.7546806049440898, "grad_norm": 2.625708818435669, "learning_rate": 9.448337637019801e-05, "loss": 0.46379523277282714, "memory(GiB)": 61.91, "step": 17615, "token_acc": 0.9105431309904153, "train_speed(iter/s)": 1.476407 }, { "epoch": 0.7548948202733388, "grad_norm": 3.342862129211426, "learning_rate": 9.44803030929872e-05, "loss": 0.4454464912414551, "memory(GiB)": 61.91, "step": 17620, "token_acc": 0.9171974522292994, "train_speed(iter/s)": 1.476391 }, { "epoch": 0.7551090356025877, "grad_norm": 3.2727174758911133, "learning_rate": 9.447722900997396e-05, "loss": 0.6218725204467773, "memory(GiB)": 61.91, "step": 17625, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.476412 }, { "epoch": 0.7553232509318367, "grad_norm": 0.04385213553905487, "learning_rate": 9.447415412121399e-05, "loss": 0.410170841217041, "memory(GiB)": 61.91, "step": 17630, "token_acc": 0.9121338912133892, "train_speed(iter/s)": 1.476416 }, { "epoch": 0.7555374662610856, "grad_norm": 2.546023368835449, "learning_rate": 9.4471078426763e-05, "loss": 0.5976030826568604, "memory(GiB)": 61.91, "step": 17635, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.476431 }, { "epoch": 0.7557516815903346, "grad_norm": 4.837190628051758, "learning_rate": 9.44680019266767e-05, "loss": 0.5206684112548828, "memory(GiB)": 61.91, "step": 17640, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.476434 }, { "epoch": 0.7559658969195836, "grad_norm": 1.5311986207962036, "learning_rate": 9.446492462101087e-05, "loss": 0.11898102760314941, "memory(GiB)": 61.91, "step": 17645, "token_acc": 0.9700996677740864, "train_speed(iter/s)": 1.47645 }, { "epoch": 0.7561801122488325, "grad_norm": 3.4006803035736084, "learning_rate": 9.446184650982117e-05, "loss": 0.4393166542053223, "memory(GiB)": 61.91, "step": 17650, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.476472 }, { "epoch": 0.7563943275780814, "grad_norm": 4.03226375579834, "learning_rate": 9.445876759316344e-05, "loss": 0.3960193872451782, "memory(GiB)": 61.91, "step": 17655, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.476468 }, { "epoch": 0.7566085429073305, "grad_norm": 0.12888787686824799, "learning_rate": 9.445568787109342e-05, "loss": 0.307785177230835, "memory(GiB)": 61.91, "step": 17660, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.47646 }, { "epoch": 0.7568227582365794, "grad_norm": 3.0396568775177, "learning_rate": 9.445260734366692e-05, "loss": 0.2591418743133545, "memory(GiB)": 61.91, "step": 17665, "token_acc": 0.9407894736842105, "train_speed(iter/s)": 1.476502 }, { "epoch": 0.7570369735658283, "grad_norm": 1.6026294231414795, "learning_rate": 9.444952601093972e-05, "loss": 0.5990015029907226, "memory(GiB)": 61.91, "step": 17670, "token_acc": 0.8826979472140762, "train_speed(iter/s)": 1.476479 }, { "epoch": 0.7572511888950774, "grad_norm": 2.518064022064209, "learning_rate": 9.444644387296767e-05, "loss": 0.3972053050994873, "memory(GiB)": 61.91, "step": 17675, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.476459 }, { "epoch": 0.7574654042243263, "grad_norm": 4.108645915985107, "learning_rate": 9.444336092980661e-05, "loss": 0.6597655773162842, "memory(GiB)": 61.91, "step": 17680, "token_acc": 0.8661087866108786, "train_speed(iter/s)": 1.47645 }, { "epoch": 0.7576796195535752, "grad_norm": 3.1483347415924072, "learning_rate": 9.444027718151235e-05, "loss": 0.2701606035232544, "memory(GiB)": 61.91, "step": 17685, "token_acc": 0.940625, "train_speed(iter/s)": 1.476443 }, { "epoch": 0.7578938348828242, "grad_norm": 0.27617496252059937, "learning_rate": 9.443719262814079e-05, "loss": 0.1592450737953186, "memory(GiB)": 61.91, "step": 17690, "token_acc": 0.9585062240663901, "train_speed(iter/s)": 1.476457 }, { "epoch": 0.7581080502120732, "grad_norm": 0.34637513756752014, "learning_rate": 9.44341072697478e-05, "loss": 0.272267746925354, "memory(GiB)": 61.91, "step": 17695, "token_acc": 0.9442231075697212, "train_speed(iter/s)": 1.476463 }, { "epoch": 0.7583222655413221, "grad_norm": 2.8971304893493652, "learning_rate": 9.443102110638928e-05, "loss": 0.36293227672576905, "memory(GiB)": 61.91, "step": 17700, "token_acc": 0.9301675977653632, "train_speed(iter/s)": 1.476482 }, { "epoch": 0.7585364808705711, "grad_norm": 2.9195828437805176, "learning_rate": 9.442793413812113e-05, "loss": 0.4400651931762695, "memory(GiB)": 61.91, "step": 17705, "token_acc": 0.9008620689655172, "train_speed(iter/s)": 1.476486 }, { "epoch": 0.75875069619982, "grad_norm": 4.574628829956055, "learning_rate": 9.442484636499927e-05, "loss": 0.3307775020599365, "memory(GiB)": 61.91, "step": 17710, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.476475 }, { "epoch": 0.758964911529069, "grad_norm": 2.2554097175598145, "learning_rate": 9.442175778707965e-05, "loss": 0.5268548011779786, "memory(GiB)": 61.91, "step": 17715, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.476467 }, { "epoch": 0.759179126858318, "grad_norm": 3.4524965286254883, "learning_rate": 9.441866840441822e-05, "loss": 0.474748420715332, "memory(GiB)": 61.91, "step": 17720, "token_acc": 0.8953068592057761, "train_speed(iter/s)": 1.476488 }, { "epoch": 0.7593933421875669, "grad_norm": 1.743773102760315, "learning_rate": 9.441557821707094e-05, "loss": 0.43621106147766114, "memory(GiB)": 61.91, "step": 17725, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.476485 }, { "epoch": 0.7596075575168159, "grad_norm": 3.4695370197296143, "learning_rate": 9.44124872250938e-05, "loss": 0.3078224420547485, "memory(GiB)": 61.91, "step": 17730, "token_acc": 0.9478260869565217, "train_speed(iter/s)": 1.476498 }, { "epoch": 0.7598217728460649, "grad_norm": 4.111319065093994, "learning_rate": 9.440939542854278e-05, "loss": 0.3372620105743408, "memory(GiB)": 61.91, "step": 17735, "token_acc": 0.9355828220858896, "train_speed(iter/s)": 1.476504 }, { "epoch": 0.7600359881753138, "grad_norm": 4.328671932220459, "learning_rate": 9.440630282747392e-05, "loss": 1.0008514404296875, "memory(GiB)": 61.91, "step": 17740, "token_acc": 0.8181818181818182, "train_speed(iter/s)": 1.476523 }, { "epoch": 0.7602502035045627, "grad_norm": 1.3275572061538696, "learning_rate": 9.44032094219432e-05, "loss": 0.0550759494304657, "memory(GiB)": 61.91, "step": 17745, "token_acc": 0.991701244813278, "train_speed(iter/s)": 1.476524 }, { "epoch": 0.7604644188338118, "grad_norm": 2.9855692386627197, "learning_rate": 9.440011521200671e-05, "loss": 0.4399075508117676, "memory(GiB)": 61.91, "step": 17750, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.476524 }, { "epoch": 0.7606786341630607, "grad_norm": 5.056507587432861, "learning_rate": 9.439702019772047e-05, "loss": 0.4407635688781738, "memory(GiB)": 61.91, "step": 17755, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.476514 }, { "epoch": 0.7608928494923096, "grad_norm": 2.100281000137329, "learning_rate": 9.439392437914058e-05, "loss": 0.35784847736358644, "memory(GiB)": 61.91, "step": 17760, "token_acc": 0.9120234604105572, "train_speed(iter/s)": 1.476484 }, { "epoch": 0.7611070648215587, "grad_norm": 1.3742643594741821, "learning_rate": 9.439082775632309e-05, "loss": 0.3371132850646973, "memory(GiB)": 61.91, "step": 17765, "token_acc": 0.9312714776632303, "train_speed(iter/s)": 1.476488 }, { "epoch": 0.7613212801508076, "grad_norm": 2.6546413898468018, "learning_rate": 9.43877303293241e-05, "loss": 0.2862443685531616, "memory(GiB)": 61.91, "step": 17770, "token_acc": 0.9390681003584229, "train_speed(iter/s)": 1.476483 }, { "epoch": 0.7615354954800565, "grad_norm": 3.882566213607788, "learning_rate": 9.438463209819976e-05, "loss": 0.36263742446899416, "memory(GiB)": 61.91, "step": 17775, "token_acc": 0.926605504587156, "train_speed(iter/s)": 1.476467 }, { "epoch": 0.7617497108093055, "grad_norm": 4.397487163543701, "learning_rate": 9.438153306300616e-05, "loss": 0.41292428970336914, "memory(GiB)": 61.91, "step": 17780, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.476467 }, { "epoch": 0.7619639261385545, "grad_norm": 3.875135660171509, "learning_rate": 9.437843322379947e-05, "loss": 0.799160099029541, "memory(GiB)": 61.91, "step": 17785, "token_acc": 0.8216783216783217, "train_speed(iter/s)": 1.476496 }, { "epoch": 0.7621781414678034, "grad_norm": 0.5637069940567017, "learning_rate": 9.437533258063582e-05, "loss": 0.2674410820007324, "memory(GiB)": 61.91, "step": 17790, "token_acc": 0.953405017921147, "train_speed(iter/s)": 1.476539 }, { "epoch": 0.7623923567970524, "grad_norm": 0.590112030506134, "learning_rate": 9.437223113357141e-05, "loss": 0.3939110279083252, "memory(GiB)": 61.91, "step": 17795, "token_acc": 0.9375, "train_speed(iter/s)": 1.476521 }, { "epoch": 0.7626065721263013, "grad_norm": 3.66007399559021, "learning_rate": 9.436912888266239e-05, "loss": 0.5299183368682862, "memory(GiB)": 61.91, "step": 17800, "token_acc": 0.884375, "train_speed(iter/s)": 1.476517 }, { "epoch": 0.7628207874555503, "grad_norm": 2.091958522796631, "learning_rate": 9.436602582796498e-05, "loss": 0.4302510738372803, "memory(GiB)": 61.91, "step": 17805, "token_acc": 0.9022801302931596, "train_speed(iter/s)": 1.47653 }, { "epoch": 0.7630350027847993, "grad_norm": 1.0386624336242676, "learning_rate": 9.43629219695354e-05, "loss": 0.3885274171829224, "memory(GiB)": 61.91, "step": 17810, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.476527 }, { "epoch": 0.7632492181140482, "grad_norm": 2.8409454822540283, "learning_rate": 9.435981730742986e-05, "loss": 0.1685233235359192, "memory(GiB)": 61.91, "step": 17815, "token_acc": 0.9634146341463414, "train_speed(iter/s)": 1.47654 }, { "epoch": 0.7634634334432973, "grad_norm": 2.203361988067627, "learning_rate": 9.435671184170463e-05, "loss": 0.2962444543838501, "memory(GiB)": 61.91, "step": 17820, "token_acc": 0.9409722222222222, "train_speed(iter/s)": 1.476543 }, { "epoch": 0.7636776487725462, "grad_norm": 4.074455261230469, "learning_rate": 9.435360557241595e-05, "loss": 0.17392344474792482, "memory(GiB)": 61.91, "step": 17825, "token_acc": 0.95703125, "train_speed(iter/s)": 1.47654 }, { "epoch": 0.7638918641017951, "grad_norm": 3.75576114654541, "learning_rate": 9.43504984996201e-05, "loss": 0.5283247947692871, "memory(GiB)": 61.91, "step": 17830, "token_acc": 0.895397489539749, "train_speed(iter/s)": 1.476561 }, { "epoch": 0.7641060794310441, "grad_norm": 3.7666265964508057, "learning_rate": 9.434739062337337e-05, "loss": 0.31411659717559814, "memory(GiB)": 61.91, "step": 17835, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.476551 }, { "epoch": 0.7643202947602931, "grad_norm": 0.5788799524307251, "learning_rate": 9.434428194373205e-05, "loss": 0.44597601890563965, "memory(GiB)": 61.91, "step": 17840, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.476555 }, { "epoch": 0.764534510089542, "grad_norm": 1.794717788696289, "learning_rate": 9.434117246075246e-05, "loss": 0.35425355434417727, "memory(GiB)": 61.91, "step": 17845, "token_acc": 0.9296875, "train_speed(iter/s)": 1.476581 }, { "epoch": 0.764748725418791, "grad_norm": 4.7251152992248535, "learning_rate": 9.433806217449094e-05, "loss": 0.620120906829834, "memory(GiB)": 61.91, "step": 17850, "token_acc": 0.8736842105263158, "train_speed(iter/s)": 1.476618 }, { "epoch": 0.7649629407480399, "grad_norm": 1.712943434715271, "learning_rate": 9.433495108500384e-05, "loss": 0.45328407287597655, "memory(GiB)": 61.91, "step": 17855, "token_acc": 0.9295774647887324, "train_speed(iter/s)": 1.476625 }, { "epoch": 0.7651771560772889, "grad_norm": 3.2541568279266357, "learning_rate": 9.43318391923475e-05, "loss": 0.515521240234375, "memory(GiB)": 61.91, "step": 17860, "token_acc": 0.8906752411575563, "train_speed(iter/s)": 1.476607 }, { "epoch": 0.7653913714065379, "grad_norm": 3.237934112548828, "learning_rate": 9.432872649657832e-05, "loss": 0.3444105625152588, "memory(GiB)": 61.91, "step": 17865, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.476603 }, { "epoch": 0.7656055867357868, "grad_norm": 3.921712636947632, "learning_rate": 9.432561299775267e-05, "loss": 0.42067670822143555, "memory(GiB)": 61.91, "step": 17870, "token_acc": 0.9161073825503355, "train_speed(iter/s)": 1.476664 }, { "epoch": 0.7658198020650357, "grad_norm": 5.253240585327148, "learning_rate": 9.432249869592696e-05, "loss": 0.355410099029541, "memory(GiB)": 61.91, "step": 17875, "token_acc": 0.9051724137931034, "train_speed(iter/s)": 1.476667 }, { "epoch": 0.7660340173942848, "grad_norm": 8.360040664672852, "learning_rate": 9.431938359115759e-05, "loss": 0.259171199798584, "memory(GiB)": 61.91, "step": 17880, "token_acc": 0.9481481481481482, "train_speed(iter/s)": 1.476668 }, { "epoch": 0.7662482327235337, "grad_norm": 1.3716843128204346, "learning_rate": 9.431626768350104e-05, "loss": 0.2558110237121582, "memory(GiB)": 61.91, "step": 17885, "token_acc": 0.9401993355481728, "train_speed(iter/s)": 1.476674 }, { "epoch": 0.7664624480527826, "grad_norm": 0.07192708551883698, "learning_rate": 9.43131509730137e-05, "loss": 0.1768442153930664, "memory(GiB)": 61.91, "step": 17890, "token_acc": 0.9563318777292577, "train_speed(iter/s)": 1.476683 }, { "epoch": 0.7666766633820317, "grad_norm": 0.4258044362068176, "learning_rate": 9.43100334597521e-05, "loss": 0.3609853982925415, "memory(GiB)": 61.91, "step": 17895, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.476682 }, { "epoch": 0.7668908787112806, "grad_norm": 2.807299852371216, "learning_rate": 9.430691514377264e-05, "loss": 0.3459017753601074, "memory(GiB)": 61.91, "step": 17900, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.476675 }, { "epoch": 0.7671050940405295, "grad_norm": 4.105527400970459, "learning_rate": 9.430379602513186e-05, "loss": 0.6321208000183105, "memory(GiB)": 61.91, "step": 17905, "token_acc": 0.8769716088328076, "train_speed(iter/s)": 1.476664 }, { "epoch": 0.7673193093697785, "grad_norm": 1.592163324356079, "learning_rate": 9.430067610388626e-05, "loss": 0.4225130558013916, "memory(GiB)": 61.91, "step": 17910, "token_acc": 0.9188311688311688, "train_speed(iter/s)": 1.476668 }, { "epoch": 0.7675335246990275, "grad_norm": 3.1511833667755127, "learning_rate": 9.429755538009235e-05, "loss": 0.3337214469909668, "memory(GiB)": 61.91, "step": 17915, "token_acc": 0.9137380191693291, "train_speed(iter/s)": 1.476655 }, { "epoch": 0.7677477400282764, "grad_norm": 0.574252188205719, "learning_rate": 9.429443385380667e-05, "loss": 0.45998053550720214, "memory(GiB)": 61.91, "step": 17920, "token_acc": 0.9, "train_speed(iter/s)": 1.476634 }, { "epoch": 0.7679619553575254, "grad_norm": 2.0766303539276123, "learning_rate": 9.429131152508577e-05, "loss": 0.37004790306091306, "memory(GiB)": 61.91, "step": 17925, "token_acc": 0.9295302013422819, "train_speed(iter/s)": 1.476644 }, { "epoch": 0.7681761706867744, "grad_norm": 1.2767369747161865, "learning_rate": 9.42881883939862e-05, "loss": 0.22735412120819093, "memory(GiB)": 61.91, "step": 17930, "token_acc": 0.9529411764705882, "train_speed(iter/s)": 1.476746 }, { "epoch": 0.7683903860160233, "grad_norm": 8.654136657714844, "learning_rate": 9.428506446056458e-05, "loss": 0.5335653305053711, "memory(GiB)": 61.91, "step": 17935, "token_acc": 0.89937106918239, "train_speed(iter/s)": 1.476725 }, { "epoch": 0.7686046013452723, "grad_norm": 1.8025439977645874, "learning_rate": 9.428193972487746e-05, "loss": 0.4268326759338379, "memory(GiB)": 61.91, "step": 17940, "token_acc": 0.9110512129380054, "train_speed(iter/s)": 1.476782 }, { "epoch": 0.7688188166745212, "grad_norm": 2.801381826400757, "learning_rate": 9.427881418698147e-05, "loss": 0.40581378936767576, "memory(GiB)": 61.91, "step": 17945, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.476839 }, { "epoch": 0.7690330320037702, "grad_norm": 5.900826930999756, "learning_rate": 9.427568784693321e-05, "loss": 0.5754095077514648, "memory(GiB)": 61.91, "step": 17950, "token_acc": 0.8736462093862816, "train_speed(iter/s)": 1.476827 }, { "epoch": 0.7692472473330192, "grad_norm": 3.050471544265747, "learning_rate": 9.427256070478934e-05, "loss": 0.34565713405609133, "memory(GiB)": 61.91, "step": 17955, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.476818 }, { "epoch": 0.7694614626622681, "grad_norm": 3.461519479751587, "learning_rate": 9.426943276060649e-05, "loss": 0.5484074592590332, "memory(GiB)": 61.91, "step": 17960, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.476835 }, { "epoch": 0.769675677991517, "grad_norm": 3.97141170501709, "learning_rate": 9.426630401444136e-05, "loss": 0.6407596588134765, "memory(GiB)": 61.91, "step": 17965, "token_acc": 0.8689024390243902, "train_speed(iter/s)": 1.47687 }, { "epoch": 0.7698898933207661, "grad_norm": 2.143667221069336, "learning_rate": 9.426317446635059e-05, "loss": 0.44272284507751464, "memory(GiB)": 61.91, "step": 17970, "token_acc": 0.9161490683229814, "train_speed(iter/s)": 1.476869 }, { "epoch": 0.770104108650015, "grad_norm": 2.1256141662597656, "learning_rate": 9.426004411639089e-05, "loss": 0.4420326709747314, "memory(GiB)": 61.91, "step": 17975, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.476912 }, { "epoch": 0.7703183239792639, "grad_norm": 3.1965503692626953, "learning_rate": 9.425691296461898e-05, "loss": 0.25966382026672363, "memory(GiB)": 61.91, "step": 17980, "token_acc": 0.9510703363914373, "train_speed(iter/s)": 1.476894 }, { "epoch": 0.770532539308513, "grad_norm": 2.943720579147339, "learning_rate": 9.425378101109158e-05, "loss": 0.2289384126663208, "memory(GiB)": 61.91, "step": 17985, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.476894 }, { "epoch": 0.7707467546377619, "grad_norm": 8.741105079650879, "learning_rate": 9.425064825586541e-05, "loss": 0.6280754566192627, "memory(GiB)": 61.91, "step": 17990, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.47693 }, { "epoch": 0.7709609699670108, "grad_norm": 2.2523515224456787, "learning_rate": 9.424751469899724e-05, "loss": 0.454746150970459, "memory(GiB)": 61.91, "step": 17995, "token_acc": 0.8970099667774086, "train_speed(iter/s)": 1.476924 }, { "epoch": 0.7711751852962598, "grad_norm": 1.521958351135254, "learning_rate": 9.424438034054385e-05, "loss": 0.1399255633354187, "memory(GiB)": 61.91, "step": 18000, "token_acc": 0.952076677316294, "train_speed(iter/s)": 1.476926 }, { "epoch": 0.7711751852962598, "eval_loss": 2.5564470291137695, "eval_runtime": 13.0166, "eval_samples_per_second": 7.683, "eval_steps_per_second": 7.683, "eval_token_acc": 0.4300653594771242, "step": 18000 }, { "epoch": 0.7713894006255088, "grad_norm": 4.690343379974365, "learning_rate": 9.424124518056199e-05, "loss": 0.29893133640289304, "memory(GiB)": 61.91, "step": 18005, "token_acc": 0.5512437810945273, "train_speed(iter/s)": 1.475231 }, { "epoch": 0.7716036159547577, "grad_norm": 0.6310479044914246, "learning_rate": 9.423810921910848e-05, "loss": 0.39267845153808595, "memory(GiB)": 61.91, "step": 18010, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.475237 }, { "epoch": 0.7718178312840067, "grad_norm": 2.586376428604126, "learning_rate": 9.42349724562401e-05, "loss": 0.19487950801849366, "memory(GiB)": 61.91, "step": 18015, "token_acc": 0.9678456591639871, "train_speed(iter/s)": 1.475221 }, { "epoch": 0.7720320466132556, "grad_norm": 3.1441826820373535, "learning_rate": 9.423183489201373e-05, "loss": 0.5759323120117188, "memory(GiB)": 61.91, "step": 18020, "token_acc": 0.8798701298701299, "train_speed(iter/s)": 1.475229 }, { "epoch": 0.7722462619425046, "grad_norm": 4.201291084289551, "learning_rate": 9.422869652648617e-05, "loss": 0.3275307178497314, "memory(GiB)": 61.91, "step": 18025, "token_acc": 0.9098712446351931, "train_speed(iter/s)": 1.475211 }, { "epoch": 0.7724604772717536, "grad_norm": 3.617377519607544, "learning_rate": 9.422555735971426e-05, "loss": 0.3146941661834717, "memory(GiB)": 61.91, "step": 18030, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.475226 }, { "epoch": 0.7726746926010025, "grad_norm": 5.3858723640441895, "learning_rate": 9.422241739175491e-05, "loss": 0.6596278190612793, "memory(GiB)": 61.91, "step": 18035, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.475217 }, { "epoch": 0.7728889079302514, "grad_norm": 2.7037596702575684, "learning_rate": 9.421927662266499e-05, "loss": 0.3119417667388916, "memory(GiB)": 61.91, "step": 18040, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.475225 }, { "epoch": 0.7731031232595005, "grad_norm": 4.783984184265137, "learning_rate": 9.421613505250138e-05, "loss": 0.46303248405456543, "memory(GiB)": 61.91, "step": 18045, "token_acc": 0.9219219219219219, "train_speed(iter/s)": 1.475224 }, { "epoch": 0.7733173385887494, "grad_norm": 2.9250478744506836, "learning_rate": 9.421299268132103e-05, "loss": 0.3121185302734375, "memory(GiB)": 61.91, "step": 18050, "token_acc": 0.9304635761589404, "train_speed(iter/s)": 1.475228 }, { "epoch": 0.7735315539179983, "grad_norm": 3.2610762119293213, "learning_rate": 9.420984950918082e-05, "loss": 0.3206181049346924, "memory(GiB)": 61.91, "step": 18055, "token_acc": 0.9299610894941635, "train_speed(iter/s)": 1.475233 }, { "epoch": 0.7737457692472474, "grad_norm": 2.605062246322632, "learning_rate": 9.42067055361377e-05, "loss": 0.3707383155822754, "memory(GiB)": 61.91, "step": 18060, "token_acc": 0.9082278481012658, "train_speed(iter/s)": 1.475223 }, { "epoch": 0.7739599845764963, "grad_norm": 2.9702630043029785, "learning_rate": 9.420356076224865e-05, "loss": 0.1993948459625244, "memory(GiB)": 61.91, "step": 18065, "token_acc": 0.953125, "train_speed(iter/s)": 1.475221 }, { "epoch": 0.7741741999057452, "grad_norm": 2.4142966270446777, "learning_rate": 9.420041518757066e-05, "loss": 0.31743559837341306, "memory(GiB)": 61.91, "step": 18070, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.47521 }, { "epoch": 0.7743884152349942, "grad_norm": 1.5340408086776733, "learning_rate": 9.419726881216066e-05, "loss": 0.519460916519165, "memory(GiB)": 61.91, "step": 18075, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.475284 }, { "epoch": 0.7746026305642432, "grad_norm": 2.030808687210083, "learning_rate": 9.419412163607567e-05, "loss": 0.3509959697723389, "memory(GiB)": 61.91, "step": 18080, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.475331 }, { "epoch": 0.7748168458934921, "grad_norm": 2.6986589431762695, "learning_rate": 9.419097365937272e-05, "loss": 0.3808340072631836, "memory(GiB)": 61.91, "step": 18085, "token_acc": 0.9155844155844156, "train_speed(iter/s)": 1.475328 }, { "epoch": 0.7750310612227411, "grad_norm": 1.1533554792404175, "learning_rate": 9.418782488210882e-05, "loss": 0.46178278923034666, "memory(GiB)": 61.91, "step": 18090, "token_acc": 0.8977635782747604, "train_speed(iter/s)": 1.475323 }, { "epoch": 0.77524527655199, "grad_norm": 2.110898017883301, "learning_rate": 9.418467530434103e-05, "loss": 0.5668289184570312, "memory(GiB)": 61.91, "step": 18095, "token_acc": 0.884, "train_speed(iter/s)": 1.475312 }, { "epoch": 0.775459491881239, "grad_norm": 2.4103446006774902, "learning_rate": 9.41815249261264e-05, "loss": 0.23775846958160402, "memory(GiB)": 61.91, "step": 18100, "token_acc": 0.937984496124031, "train_speed(iter/s)": 1.475323 }, { "epoch": 0.775673707210488, "grad_norm": 5.103067398071289, "learning_rate": 9.417837374752199e-05, "loss": 0.33349971771240233, "memory(GiB)": 61.91, "step": 18105, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.475368 }, { "epoch": 0.7758879225397369, "grad_norm": 2.5990865230560303, "learning_rate": 9.417522176858491e-05, "loss": 0.33001110553741453, "memory(GiB)": 61.91, "step": 18110, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.475371 }, { "epoch": 0.7761021378689859, "grad_norm": 2.1576690673828125, "learning_rate": 9.417206898937223e-05, "loss": 0.32697248458862305, "memory(GiB)": 61.91, "step": 18115, "token_acc": 0.9452554744525548, "train_speed(iter/s)": 1.475361 }, { "epoch": 0.7763163531982349, "grad_norm": 4.651826858520508, "learning_rate": 9.41689154099411e-05, "loss": 0.4662612438201904, "memory(GiB)": 61.91, "step": 18120, "token_acc": 0.9039039039039038, "train_speed(iter/s)": 1.475362 }, { "epoch": 0.7765305685274838, "grad_norm": 2.8461434841156006, "learning_rate": 9.416576103034862e-05, "loss": 0.27122697830200193, "memory(GiB)": 61.91, "step": 18125, "token_acc": 0.9347181008902077, "train_speed(iter/s)": 1.475349 }, { "epoch": 0.7767447838567327, "grad_norm": 4.126341342926025, "learning_rate": 9.416260585065195e-05, "loss": 0.47978878021240234, "memory(GiB)": 61.91, "step": 18130, "token_acc": 0.9093851132686084, "train_speed(iter/s)": 1.475332 }, { "epoch": 0.7769589991859818, "grad_norm": 1.502657175064087, "learning_rate": 9.415944987090826e-05, "loss": 0.21162753105163573, "memory(GiB)": 61.91, "step": 18135, "token_acc": 0.9571428571428572, "train_speed(iter/s)": 1.47532 }, { "epoch": 0.7771732145152307, "grad_norm": 1.6383600234985352, "learning_rate": 9.415629309117471e-05, "loss": 0.2996227502822876, "memory(GiB)": 61.91, "step": 18140, "token_acc": 0.9404761904761905, "train_speed(iter/s)": 1.475308 }, { "epoch": 0.7773874298444796, "grad_norm": 0.4740476608276367, "learning_rate": 9.415313551150847e-05, "loss": 0.50497407913208, "memory(GiB)": 61.91, "step": 18145, "token_acc": 0.8855218855218855, "train_speed(iter/s)": 1.475293 }, { "epoch": 0.7776016451737287, "grad_norm": 3.0299289226531982, "learning_rate": 9.414997713196678e-05, "loss": 0.2419410228729248, "memory(GiB)": 61.91, "step": 18150, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.475322 }, { "epoch": 0.7778158605029776, "grad_norm": 3.042367935180664, "learning_rate": 9.414681795260683e-05, "loss": 0.6752778053283691, "memory(GiB)": 61.91, "step": 18155, "token_acc": 0.8156028368794326, "train_speed(iter/s)": 1.475381 }, { "epoch": 0.7780300758322266, "grad_norm": 4.10979700088501, "learning_rate": 9.414365797348586e-05, "loss": 0.4149197578430176, "memory(GiB)": 61.91, "step": 18160, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.475389 }, { "epoch": 0.7782442911614755, "grad_norm": 0.4377210736274719, "learning_rate": 9.414049719466111e-05, "loss": 0.10137203931808472, "memory(GiB)": 61.91, "step": 18165, "token_acc": 0.9664179104477612, "train_speed(iter/s)": 1.475378 }, { "epoch": 0.7784585064907245, "grad_norm": 0.824683666229248, "learning_rate": 9.413733561618985e-05, "loss": 0.4958193778991699, "memory(GiB)": 61.91, "step": 18170, "token_acc": 0.8949416342412452, "train_speed(iter/s)": 1.475405 }, { "epoch": 0.7786727218199735, "grad_norm": 4.913130760192871, "learning_rate": 9.413417323812936e-05, "loss": 0.42053661346435545, "memory(GiB)": 61.91, "step": 18175, "token_acc": 0.8982456140350877, "train_speed(iter/s)": 1.475388 }, { "epoch": 0.7788869371492224, "grad_norm": 3.716764211654663, "learning_rate": 9.41310100605369e-05, "loss": 0.7530553817749024, "memory(GiB)": 61.91, "step": 18180, "token_acc": 0.8539682539682539, "train_speed(iter/s)": 1.475356 }, { "epoch": 0.7791011524784713, "grad_norm": 2.9686596393585205, "learning_rate": 9.412784608346983e-05, "loss": 0.4850471973419189, "memory(GiB)": 61.91, "step": 18185, "token_acc": 0.8956521739130435, "train_speed(iter/s)": 1.475351 }, { "epoch": 0.7793153678077204, "grad_norm": 1.6737639904022217, "learning_rate": 9.41246813069854e-05, "loss": 0.34420387744903563, "memory(GiB)": 61.91, "step": 18190, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.475358 }, { "epoch": 0.7795295831369693, "grad_norm": 5.964578151702881, "learning_rate": 9.412151573114098e-05, "loss": 0.5602612972259522, "memory(GiB)": 61.91, "step": 18195, "token_acc": 0.8795620437956204, "train_speed(iter/s)": 1.475395 }, { "epoch": 0.7797437984662182, "grad_norm": 2.513779640197754, "learning_rate": 9.411834935599393e-05, "loss": 0.4416038513183594, "memory(GiB)": 61.91, "step": 18200, "token_acc": 0.8984962406015038, "train_speed(iter/s)": 1.475367 }, { "epoch": 0.7799580137954673, "grad_norm": 3.5727601051330566, "learning_rate": 9.411518218160158e-05, "loss": 0.3451981544494629, "memory(GiB)": 61.91, "step": 18205, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.475384 }, { "epoch": 0.7801722291247162, "grad_norm": 1.8249051570892334, "learning_rate": 9.411201420802134e-05, "loss": 0.2401799201965332, "memory(GiB)": 61.91, "step": 18210, "token_acc": 0.9431818181818182, "train_speed(iter/s)": 1.475361 }, { "epoch": 0.7803864444539651, "grad_norm": 5.3668928146362305, "learning_rate": 9.410884543531056e-05, "loss": 0.2728896617889404, "memory(GiB)": 61.91, "step": 18215, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.475325 }, { "epoch": 0.7806006597832141, "grad_norm": 7.242773532867432, "learning_rate": 9.410567586352668e-05, "loss": 0.31612999439239503, "memory(GiB)": 61.91, "step": 18220, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.475304 }, { "epoch": 0.7808148751124631, "grad_norm": 2.543660879135132, "learning_rate": 9.41025054927271e-05, "loss": 0.3086737871170044, "memory(GiB)": 61.91, "step": 18225, "token_acc": 0.9383116883116883, "train_speed(iter/s)": 1.475294 }, { "epoch": 0.781029090441712, "grad_norm": 4.4405412673950195, "learning_rate": 9.409933432296927e-05, "loss": 0.43038363456726075, "memory(GiB)": 61.91, "step": 18230, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.475282 }, { "epoch": 0.781243305770961, "grad_norm": 3.361912488937378, "learning_rate": 9.409616235431062e-05, "loss": 0.4404590606689453, "memory(GiB)": 61.91, "step": 18235, "token_acc": 0.9014925373134328, "train_speed(iter/s)": 1.475258 }, { "epoch": 0.7814575211002099, "grad_norm": 4.304111957550049, "learning_rate": 9.409298958680864e-05, "loss": 0.3563318967819214, "memory(GiB)": 61.91, "step": 18240, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.475291 }, { "epoch": 0.7816717364294589, "grad_norm": 4.392543792724609, "learning_rate": 9.408981602052078e-05, "loss": 0.3089850425720215, "memory(GiB)": 61.91, "step": 18245, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.475292 }, { "epoch": 0.7818859517587079, "grad_norm": 4.504976749420166, "learning_rate": 9.408664165550453e-05, "loss": 0.8227139472961426, "memory(GiB)": 61.91, "step": 18250, "token_acc": 0.8242811501597445, "train_speed(iter/s)": 1.47527 }, { "epoch": 0.7821001670879568, "grad_norm": 1.4801713228225708, "learning_rate": 9.408346649181742e-05, "loss": 0.22419450283050538, "memory(GiB)": 61.91, "step": 18255, "token_acc": 0.9505813953488372, "train_speed(iter/s)": 1.475291 }, { "epoch": 0.7823143824172057, "grad_norm": 0.36407697200775146, "learning_rate": 9.408029052951698e-05, "loss": 0.3018055438995361, "memory(GiB)": 61.91, "step": 18260, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.475283 }, { "epoch": 0.7825285977464548, "grad_norm": 3.7339563369750977, "learning_rate": 9.407711376866071e-05, "loss": 0.43625102043151853, "memory(GiB)": 61.91, "step": 18265, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.475316 }, { "epoch": 0.7827428130757037, "grad_norm": 4.981937885284424, "learning_rate": 9.407393620930618e-05, "loss": 0.3895270586013794, "memory(GiB)": 61.91, "step": 18270, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.475324 }, { "epoch": 0.7829570284049526, "grad_norm": 3.34212064743042, "learning_rate": 9.407075785151094e-05, "loss": 0.40848913192749026, "memory(GiB)": 61.91, "step": 18275, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.475311 }, { "epoch": 0.7831712437342017, "grad_norm": 2.970597267150879, "learning_rate": 9.406757869533259e-05, "loss": 0.1914145827293396, "memory(GiB)": 61.91, "step": 18280, "token_acc": 0.9577922077922078, "train_speed(iter/s)": 1.475295 }, { "epoch": 0.7833854590634506, "grad_norm": 3.3834140300750732, "learning_rate": 9.406439874082871e-05, "loss": 0.397034740447998, "memory(GiB)": 61.91, "step": 18285, "token_acc": 0.916083916083916, "train_speed(iter/s)": 1.475302 }, { "epoch": 0.7835996743926995, "grad_norm": 6.06129789352417, "learning_rate": 9.406121798805692e-05, "loss": 0.5157209396362304, "memory(GiB)": 61.91, "step": 18290, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.475324 }, { "epoch": 0.7838138897219485, "grad_norm": 2.664497137069702, "learning_rate": 9.405803643707482e-05, "loss": 0.6046819686889648, "memory(GiB)": 61.91, "step": 18295, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.475297 }, { "epoch": 0.7840281050511975, "grad_norm": 3.0954744815826416, "learning_rate": 9.405485408794005e-05, "loss": 0.5451520919799805, "memory(GiB)": 61.91, "step": 18300, "token_acc": 0.8868613138686131, "train_speed(iter/s)": 1.475294 }, { "epoch": 0.7842423203804464, "grad_norm": 5.036472320556641, "learning_rate": 9.40516709407103e-05, "loss": 0.34125123023986814, "memory(GiB)": 61.91, "step": 18305, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.475283 }, { "epoch": 0.7844565357096954, "grad_norm": 2.4226138591766357, "learning_rate": 9.40484869954432e-05, "loss": 0.4034876823425293, "memory(GiB)": 61.91, "step": 18310, "token_acc": 0.9034749034749034, "train_speed(iter/s)": 1.47527 }, { "epoch": 0.7846707510389443, "grad_norm": 2.0947017669677734, "learning_rate": 9.404530225219643e-05, "loss": 0.4193308353424072, "memory(GiB)": 61.91, "step": 18315, "token_acc": 0.9247648902821317, "train_speed(iter/s)": 1.475249 }, { "epoch": 0.7848849663681933, "grad_norm": 1.9820621013641357, "learning_rate": 9.404211671102769e-05, "loss": 0.5283358097076416, "memory(GiB)": 61.91, "step": 18320, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.475254 }, { "epoch": 0.7850991816974423, "grad_norm": 0.6188111305236816, "learning_rate": 9.403893037199469e-05, "loss": 0.3248807668685913, "memory(GiB)": 61.91, "step": 18325, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.475253 }, { "epoch": 0.7853133970266912, "grad_norm": 4.409011363983154, "learning_rate": 9.403574323515516e-05, "loss": 0.5854901790618896, "memory(GiB)": 61.91, "step": 18330, "token_acc": 0.8731343283582089, "train_speed(iter/s)": 1.475236 }, { "epoch": 0.7855276123559402, "grad_norm": 3.93644380569458, "learning_rate": 9.403255530056682e-05, "loss": 0.3459939956665039, "memory(GiB)": 61.91, "step": 18335, "token_acc": 0.9174041297935103, "train_speed(iter/s)": 1.475268 }, { "epoch": 0.7857418276851892, "grad_norm": 3.0632126331329346, "learning_rate": 9.402936656828745e-05, "loss": 0.3423941612243652, "memory(GiB)": 61.91, "step": 18340, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.475297 }, { "epoch": 0.7859560430144381, "grad_norm": 3.6476686000823975, "learning_rate": 9.40261770383748e-05, "loss": 0.9022815704345704, "memory(GiB)": 61.91, "step": 18345, "token_acc": 0.8338278931750742, "train_speed(iter/s)": 1.475282 }, { "epoch": 0.786170258343687, "grad_norm": 3.154538631439209, "learning_rate": 9.402298671088665e-05, "loss": 0.33768658638000487, "memory(GiB)": 61.91, "step": 18350, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.475291 }, { "epoch": 0.7863844736729361, "grad_norm": 2.4833741188049316, "learning_rate": 9.401979558588081e-05, "loss": 0.38443498611450194, "memory(GiB)": 61.91, "step": 18355, "token_acc": 0.9148148148148149, "train_speed(iter/s)": 1.47538 }, { "epoch": 0.786598689002185, "grad_norm": 4.310610771179199, "learning_rate": 9.401660366341506e-05, "loss": 0.5771257400512695, "memory(GiB)": 61.91, "step": 18360, "token_acc": 0.8734693877551021, "train_speed(iter/s)": 1.475424 }, { "epoch": 0.7868129043314339, "grad_norm": 1.4788991212844849, "learning_rate": 9.401341094354725e-05, "loss": 0.6203030586242676, "memory(GiB)": 61.91, "step": 18365, "token_acc": 0.8907849829351536, "train_speed(iter/s)": 1.475396 }, { "epoch": 0.787027119660683, "grad_norm": 2.8922224044799805, "learning_rate": 9.401021742633523e-05, "loss": 0.2692288398742676, "memory(GiB)": 61.91, "step": 18370, "token_acc": 0.9323076923076923, "train_speed(iter/s)": 1.475371 }, { "epoch": 0.7872413349899319, "grad_norm": 4.602039813995361, "learning_rate": 9.400702311183681e-05, "loss": 0.6428001880645752, "memory(GiB)": 61.91, "step": 18375, "token_acc": 0.8546099290780141, "train_speed(iter/s)": 1.475401 }, { "epoch": 0.7874555503191808, "grad_norm": 4.733827590942383, "learning_rate": 9.40038280001099e-05, "loss": 0.5406495094299316, "memory(GiB)": 61.91, "step": 18380, "token_acc": 0.8670886075949367, "train_speed(iter/s)": 1.475398 }, { "epoch": 0.7876697656484298, "grad_norm": 4.539280414581299, "learning_rate": 9.400063209121235e-05, "loss": 0.45218782424926757, "memory(GiB)": 61.91, "step": 18385, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.47538 }, { "epoch": 0.7878839809776788, "grad_norm": 2.3652846813201904, "learning_rate": 9.39974353852021e-05, "loss": 0.4602104663848877, "memory(GiB)": 61.91, "step": 18390, "token_acc": 0.8935361216730038, "train_speed(iter/s)": 1.475342 }, { "epoch": 0.7880981963069277, "grad_norm": 6.795609951019287, "learning_rate": 9.399423788213701e-05, "loss": 0.5180245399475097, "memory(GiB)": 61.91, "step": 18395, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.475422 }, { "epoch": 0.7883124116361767, "grad_norm": 5.803899765014648, "learning_rate": 9.399103958207505e-05, "loss": 0.46486449241638184, "memory(GiB)": 61.91, "step": 18400, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.47542 }, { "epoch": 0.7885266269654256, "grad_norm": 4.889878273010254, "learning_rate": 9.398784048507414e-05, "loss": 0.37122206687927245, "memory(GiB)": 61.91, "step": 18405, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.475422 }, { "epoch": 0.7887408422946746, "grad_norm": 1.5110496282577515, "learning_rate": 9.398464059119222e-05, "loss": 0.2590285062789917, "memory(GiB)": 61.91, "step": 18410, "token_acc": 0.9514563106796117, "train_speed(iter/s)": 1.475409 }, { "epoch": 0.7889550576239236, "grad_norm": 4.9905686378479, "learning_rate": 9.398143990048727e-05, "loss": 0.5189781188964844, "memory(GiB)": 61.91, "step": 18415, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.475402 }, { "epoch": 0.7891692729531725, "grad_norm": 4.14054536819458, "learning_rate": 9.397823841301729e-05, "loss": 0.39918835163116456, "memory(GiB)": 61.91, "step": 18420, "token_acc": 0.907051282051282, "train_speed(iter/s)": 1.475394 }, { "epoch": 0.7893834882824214, "grad_norm": 6.6264729499816895, "learning_rate": 9.397503612884028e-05, "loss": 0.3877793550491333, "memory(GiB)": 61.91, "step": 18425, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.475395 }, { "epoch": 0.7895977036116705, "grad_norm": 2.2090871334075928, "learning_rate": 9.397183304801422e-05, "loss": 0.4285701274871826, "memory(GiB)": 61.91, "step": 18430, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.4754 }, { "epoch": 0.7898119189409194, "grad_norm": 2.3123679161071777, "learning_rate": 9.396862917059716e-05, "loss": 0.3271916627883911, "memory(GiB)": 61.91, "step": 18435, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.475406 }, { "epoch": 0.7900261342701683, "grad_norm": 7.65413761138916, "learning_rate": 9.396542449664714e-05, "loss": 0.6744177341461182, "memory(GiB)": 61.91, "step": 18440, "token_acc": 0.8557046979865772, "train_speed(iter/s)": 1.475398 }, { "epoch": 0.7902403495994174, "grad_norm": 1.2584947347640991, "learning_rate": 9.396221902622221e-05, "loss": 0.2628985643386841, "memory(GiB)": 61.91, "step": 18445, "token_acc": 0.9465408805031447, "train_speed(iter/s)": 1.475409 }, { "epoch": 0.7904545649286663, "grad_norm": 5.150904178619385, "learning_rate": 9.395901275938047e-05, "loss": 0.5920158863067627, "memory(GiB)": 61.91, "step": 18450, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.475444 }, { "epoch": 0.7906687802579152, "grad_norm": 3.6417105197906494, "learning_rate": 9.395580569617993e-05, "loss": 0.39792284965515134, "memory(GiB)": 61.91, "step": 18455, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.475458 }, { "epoch": 0.7908829955871642, "grad_norm": 1.864035725593567, "learning_rate": 9.395259783667876e-05, "loss": 0.14321712255477906, "memory(GiB)": 61.91, "step": 18460, "token_acc": 0.9614243323442137, "train_speed(iter/s)": 1.475463 }, { "epoch": 0.7910972109164132, "grad_norm": 4.1704182624816895, "learning_rate": 9.394938918093506e-05, "loss": 0.29643776416778567, "memory(GiB)": 61.91, "step": 18465, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.475467 }, { "epoch": 0.7913114262456621, "grad_norm": 2.4637370109558105, "learning_rate": 9.394617972900692e-05, "loss": 0.6764262676239013, "memory(GiB)": 61.91, "step": 18470, "token_acc": 0.8434504792332268, "train_speed(iter/s)": 1.47548 }, { "epoch": 0.7915256415749111, "grad_norm": 1.5648157596588135, "learning_rate": 9.394296948095251e-05, "loss": 0.4197638988494873, "memory(GiB)": 61.91, "step": 18475, "token_acc": 0.8707482993197279, "train_speed(iter/s)": 1.475551 }, { "epoch": 0.79173985690416, "grad_norm": 9.901230812072754, "learning_rate": 9.393975843683e-05, "loss": 0.37912323474884035, "memory(GiB)": 61.91, "step": 18480, "token_acc": 0.9190751445086706, "train_speed(iter/s)": 1.475591 }, { "epoch": 0.791954072233409, "grad_norm": 2.578077793121338, "learning_rate": 9.393654659669755e-05, "loss": 0.37273871898651123, "memory(GiB)": 61.91, "step": 18485, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.475597 }, { "epoch": 0.792168287562658, "grad_norm": 7.595118999481201, "learning_rate": 9.393333396061335e-05, "loss": 0.5484381675720215, "memory(GiB)": 61.91, "step": 18490, "token_acc": 0.9003215434083601, "train_speed(iter/s)": 1.475622 }, { "epoch": 0.7923825028919069, "grad_norm": 3.5842859745025635, "learning_rate": 9.393012052863556e-05, "loss": 0.297391939163208, "memory(GiB)": 61.91, "step": 18495, "token_acc": 0.9501661129568106, "train_speed(iter/s)": 1.475642 }, { "epoch": 0.792596718221156, "grad_norm": 5.659254550933838, "learning_rate": 9.392690630082246e-05, "loss": 0.5155001640319824, "memory(GiB)": 61.91, "step": 18500, "token_acc": 0.9183006535947712, "train_speed(iter/s)": 1.475648 }, { "epoch": 0.792596718221156, "eval_loss": 2.574023723602295, "eval_runtime": 13.5331, "eval_samples_per_second": 7.389, "eval_steps_per_second": 7.389, "eval_token_acc": 0.422475106685633, "step": 18500 }, { "epoch": 0.7928109335504049, "grad_norm": 3.272918224334717, "learning_rate": 9.392369127723224e-05, "loss": 0.5598132610321045, "memory(GiB)": 61.91, "step": 18505, "token_acc": 0.5686080947680158, "train_speed(iter/s)": 1.473928 }, { "epoch": 0.7930251488796538, "grad_norm": 4.966953277587891, "learning_rate": 9.392047545792312e-05, "loss": 0.2361004114151001, "memory(GiB)": 61.91, "step": 18510, "token_acc": 0.9370860927152318, "train_speed(iter/s)": 1.473906 }, { "epoch": 0.7932393642089028, "grad_norm": 5.0572052001953125, "learning_rate": 9.391725884295342e-05, "loss": 0.40151634216308596, "memory(GiB)": 61.91, "step": 18515, "token_acc": 0.8850931677018633, "train_speed(iter/s)": 1.473904 }, { "epoch": 0.7934535795381518, "grad_norm": 1.6856962442398071, "learning_rate": 9.391404143238137e-05, "loss": 0.24655508995056152, "memory(GiB)": 61.91, "step": 18520, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.473907 }, { "epoch": 0.7936677948674007, "grad_norm": 2.0310492515563965, "learning_rate": 9.391082322626526e-05, "loss": 0.2624941349029541, "memory(GiB)": 61.91, "step": 18525, "token_acc": 0.9496644295302014, "train_speed(iter/s)": 1.473907 }, { "epoch": 0.7938820101966497, "grad_norm": 2.181705951690674, "learning_rate": 9.39076042246634e-05, "loss": 0.16793471574783325, "memory(GiB)": 61.91, "step": 18530, "token_acc": 0.9633699633699634, "train_speed(iter/s)": 1.473916 }, { "epoch": 0.7940962255258986, "grad_norm": 2.3664073944091797, "learning_rate": 9.39043844276341e-05, "loss": 0.46283588409423826, "memory(GiB)": 61.91, "step": 18535, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.47393 }, { "epoch": 0.7943104408551476, "grad_norm": 1.4814714193344116, "learning_rate": 9.390116383523568e-05, "loss": 0.39252486228942873, "memory(GiB)": 61.91, "step": 18540, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.473983 }, { "epoch": 0.7945246561843966, "grad_norm": 4.718359470367432, "learning_rate": 9.38979424475265e-05, "loss": 0.3776960849761963, "memory(GiB)": 61.91, "step": 18545, "token_acc": 0.927797833935018, "train_speed(iter/s)": 1.473983 }, { "epoch": 0.7947388715136455, "grad_norm": 2.585422992706299, "learning_rate": 9.389472026456492e-05, "loss": 0.30879027843475343, "memory(GiB)": 61.91, "step": 18550, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.473958 }, { "epoch": 0.7949530868428945, "grad_norm": 1.2046576738357544, "learning_rate": 9.38914972864093e-05, "loss": 0.13352725505828858, "memory(GiB)": 61.91, "step": 18555, "token_acc": 0.9659090909090909, "train_speed(iter/s)": 1.473957 }, { "epoch": 0.7951673021721435, "grad_norm": 0.11178230494260788, "learning_rate": 9.388827351311804e-05, "loss": 0.43127026557922366, "memory(GiB)": 61.91, "step": 18560, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.473965 }, { "epoch": 0.7953815175013924, "grad_norm": 2.999084234237671, "learning_rate": 9.388504894474953e-05, "loss": 0.6221551418304443, "memory(GiB)": 61.91, "step": 18565, "token_acc": 0.8856088560885609, "train_speed(iter/s)": 1.474028 }, { "epoch": 0.7955957328306413, "grad_norm": 3.237025022506714, "learning_rate": 9.38818235813622e-05, "loss": 0.33475098609924314, "memory(GiB)": 61.91, "step": 18570, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.474101 }, { "epoch": 0.7958099481598904, "grad_norm": 5.419332027435303, "learning_rate": 9.387859742301445e-05, "loss": 0.22651875019073486, "memory(GiB)": 61.91, "step": 18575, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.474114 }, { "epoch": 0.7960241634891393, "grad_norm": 4.656375885009766, "learning_rate": 9.387537046976476e-05, "loss": 0.41849446296691895, "memory(GiB)": 61.91, "step": 18580, "token_acc": 0.9049295774647887, "train_speed(iter/s)": 1.474121 }, { "epoch": 0.7962383788183882, "grad_norm": 3.3053672313690186, "learning_rate": 9.387214272167157e-05, "loss": 0.4264240741729736, "memory(GiB)": 61.91, "step": 18585, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.47413 }, { "epoch": 0.7964525941476372, "grad_norm": 1.3435789346694946, "learning_rate": 9.386891417879335e-05, "loss": 0.40169439315795896, "memory(GiB)": 61.91, "step": 18590, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.474117 }, { "epoch": 0.7966668094768862, "grad_norm": 6.600622177124023, "learning_rate": 9.386568484118862e-05, "loss": 0.26682462692260744, "memory(GiB)": 61.91, "step": 18595, "token_acc": 0.9537815126050421, "train_speed(iter/s)": 1.47411 }, { "epoch": 0.7968810248061351, "grad_norm": 1.584809422492981, "learning_rate": 9.386245470891584e-05, "loss": 0.43613252639770506, "memory(GiB)": 61.91, "step": 18600, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.474131 }, { "epoch": 0.7970952401353841, "grad_norm": 3.7048964500427246, "learning_rate": 9.385922378203356e-05, "loss": 0.18313544988632202, "memory(GiB)": 61.91, "step": 18605, "token_acc": 0.975609756097561, "train_speed(iter/s)": 1.474136 }, { "epoch": 0.797309455464633, "grad_norm": 0.7785220742225647, "learning_rate": 9.38559920606003e-05, "loss": 0.3190826654434204, "memory(GiB)": 61.91, "step": 18610, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.474172 }, { "epoch": 0.797523670793882, "grad_norm": 3.739743709564209, "learning_rate": 9.38527595446746e-05, "loss": 0.4396231651306152, "memory(GiB)": 61.91, "step": 18615, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.474165 }, { "epoch": 0.797737886123131, "grad_norm": 4.534667491912842, "learning_rate": 9.384952623431502e-05, "loss": 0.8241065979003906, "memory(GiB)": 61.91, "step": 18620, "token_acc": 0.8192419825072886, "train_speed(iter/s)": 1.474125 }, { "epoch": 0.7979521014523799, "grad_norm": 1.1816273927688599, "learning_rate": 9.384629212958014e-05, "loss": 0.1785948395729065, "memory(GiB)": 61.91, "step": 18625, "token_acc": 0.9695817490494296, "train_speed(iter/s)": 1.474125 }, { "epoch": 0.7981663167816289, "grad_norm": 1.8985686302185059, "learning_rate": 9.384305723052855e-05, "loss": 0.31483049392700196, "memory(GiB)": 61.91, "step": 18630, "token_acc": 0.9430379746835443, "train_speed(iter/s)": 1.474124 }, { "epoch": 0.7983805321108779, "grad_norm": 2.2994425296783447, "learning_rate": 9.383982153721884e-05, "loss": 0.39060699939727783, "memory(GiB)": 61.91, "step": 18635, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.474118 }, { "epoch": 0.7985947474401268, "grad_norm": 4.656024932861328, "learning_rate": 9.383658504970965e-05, "loss": 0.4783134460449219, "memory(GiB)": 61.91, "step": 18640, "token_acc": 0.8991935483870968, "train_speed(iter/s)": 1.474094 }, { "epoch": 0.7988089627693757, "grad_norm": 2.3416354656219482, "learning_rate": 9.383334776805958e-05, "loss": 0.6098880290985107, "memory(GiB)": 61.91, "step": 18645, "token_acc": 0.8666666666666667, "train_speed(iter/s)": 1.474059 }, { "epoch": 0.7990231780986248, "grad_norm": 5.026739597320557, "learning_rate": 9.383010969232731e-05, "loss": 0.4009675979614258, "memory(GiB)": 61.91, "step": 18650, "token_acc": 0.9180887372013652, "train_speed(iter/s)": 1.474037 }, { "epoch": 0.7992373934278737, "grad_norm": 3.422908306121826, "learning_rate": 9.382687082257149e-05, "loss": 0.6899025917053223, "memory(GiB)": 61.91, "step": 18655, "token_acc": 0.8584070796460177, "train_speed(iter/s)": 1.474065 }, { "epoch": 0.7994516087571226, "grad_norm": 3.3283779621124268, "learning_rate": 9.38236311588508e-05, "loss": 0.21002719402313233, "memory(GiB)": 61.91, "step": 18660, "token_acc": 0.9609929078014184, "train_speed(iter/s)": 1.474047 }, { "epoch": 0.7996658240863717, "grad_norm": 2.304401159286499, "learning_rate": 9.38203907012239e-05, "loss": 0.41063852310180665, "memory(GiB)": 61.91, "step": 18665, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.474063 }, { "epoch": 0.7998800394156206, "grad_norm": 3.561155080795288, "learning_rate": 9.381714944974953e-05, "loss": 0.4834311008453369, "memory(GiB)": 61.91, "step": 18670, "token_acc": 0.8849557522123894, "train_speed(iter/s)": 1.474087 }, { "epoch": 0.8000942547448695, "grad_norm": 2.300720691680908, "learning_rate": 9.38139074044864e-05, "loss": 0.32204110622406007, "memory(GiB)": 61.91, "step": 18675, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.474054 }, { "epoch": 0.8003084700741185, "grad_norm": 3.604433298110962, "learning_rate": 9.381066456549321e-05, "loss": 0.31633501052856444, "memory(GiB)": 61.91, "step": 18680, "token_acc": 0.940625, "train_speed(iter/s)": 1.474066 }, { "epoch": 0.8005226854033675, "grad_norm": 0.2422514110803604, "learning_rate": 9.380742093282876e-05, "loss": 0.2056957006454468, "memory(GiB)": 61.91, "step": 18685, "token_acc": 0.9602446483180428, "train_speed(iter/s)": 1.47407 }, { "epoch": 0.8007369007326164, "grad_norm": 1.373374104499817, "learning_rate": 9.380417650655178e-05, "loss": 0.11875449419021607, "memory(GiB)": 61.91, "step": 18690, "token_acc": 0.9797101449275363, "train_speed(iter/s)": 1.474069 }, { "epoch": 0.8009511160618654, "grad_norm": 2.9152965545654297, "learning_rate": 9.380093128672104e-05, "loss": 0.6401498317718506, "memory(GiB)": 61.91, "step": 18695, "token_acc": 0.8754863813229572, "train_speed(iter/s)": 1.474065 }, { "epoch": 0.8011653313911143, "grad_norm": 4.159124851226807, "learning_rate": 9.379768527339534e-05, "loss": 0.36798532009124757, "memory(GiB)": 61.91, "step": 18700, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.47407 }, { "epoch": 0.8013795467203633, "grad_norm": 3.8517603874206543, "learning_rate": 9.379443846663349e-05, "loss": 0.6658937931060791, "memory(GiB)": 61.91, "step": 18705, "token_acc": 0.8620689655172413, "train_speed(iter/s)": 1.474051 }, { "epoch": 0.8015937620496123, "grad_norm": 3.101155996322632, "learning_rate": 9.379119086649432e-05, "loss": 0.3228920936584473, "memory(GiB)": 61.91, "step": 18710, "token_acc": 0.9295774647887324, "train_speed(iter/s)": 1.474068 }, { "epoch": 0.8018079773788612, "grad_norm": 3.422792911529541, "learning_rate": 9.378794247303664e-05, "loss": 0.41062288284301757, "memory(GiB)": 61.91, "step": 18715, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.474086 }, { "epoch": 0.8020221927081101, "grad_norm": 4.622722625732422, "learning_rate": 9.378469328631929e-05, "loss": 0.5445349216461182, "memory(GiB)": 61.91, "step": 18720, "token_acc": 0.8713235294117647, "train_speed(iter/s)": 1.474092 }, { "epoch": 0.8022364080373592, "grad_norm": 3.0639758110046387, "learning_rate": 9.378144330640118e-05, "loss": 0.26751368045806884, "memory(GiB)": 61.91, "step": 18725, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.474142 }, { "epoch": 0.8024506233666081, "grad_norm": 3.4552083015441895, "learning_rate": 9.377819253334113e-05, "loss": 0.21582088470458985, "memory(GiB)": 61.91, "step": 18730, "token_acc": 0.952, "train_speed(iter/s)": 1.474151 }, { "epoch": 0.802664838695857, "grad_norm": 3.4900455474853516, "learning_rate": 9.377494096719805e-05, "loss": 0.7300946235656738, "memory(GiB)": 61.91, "step": 18735, "token_acc": 0.8607142857142858, "train_speed(iter/s)": 1.474155 }, { "epoch": 0.8028790540251061, "grad_norm": 2.1262834072113037, "learning_rate": 9.377168860803085e-05, "loss": 0.19719691276550294, "memory(GiB)": 61.91, "step": 18740, "token_acc": 0.9605263157894737, "train_speed(iter/s)": 1.474143 }, { "epoch": 0.803093269354355, "grad_norm": 2.8604021072387695, "learning_rate": 9.376843545589846e-05, "loss": 0.40342154502868655, "memory(GiB)": 61.91, "step": 18745, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.474157 }, { "epoch": 0.8033074846836039, "grad_norm": 2.741079807281494, "learning_rate": 9.37651815108598e-05, "loss": 0.5112120628356933, "memory(GiB)": 61.91, "step": 18750, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.474146 }, { "epoch": 0.803521700012853, "grad_norm": 5.040750503540039, "learning_rate": 9.376192677297383e-05, "loss": 0.4081984043121338, "memory(GiB)": 61.91, "step": 18755, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.474145 }, { "epoch": 0.8037359153421019, "grad_norm": 0.6238551735877991, "learning_rate": 9.37586712422995e-05, "loss": 0.6019916534423828, "memory(GiB)": 61.91, "step": 18760, "token_acc": 0.8850574712643678, "train_speed(iter/s)": 1.474112 }, { "epoch": 0.8039501306713508, "grad_norm": 3.3715832233428955, "learning_rate": 9.375541491889579e-05, "loss": 0.46813116073608396, "memory(GiB)": 61.91, "step": 18765, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.47412 }, { "epoch": 0.8041643460005998, "grad_norm": 2.386761426925659, "learning_rate": 9.375215780282169e-05, "loss": 0.3985734939575195, "memory(GiB)": 61.91, "step": 18770, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.474107 }, { "epoch": 0.8043785613298488, "grad_norm": 4.078367710113525, "learning_rate": 9.374889989413622e-05, "loss": 0.5971104145050049, "memory(GiB)": 61.91, "step": 18775, "token_acc": 0.8945147679324894, "train_speed(iter/s)": 1.47411 }, { "epoch": 0.8045927766590977, "grad_norm": 5.44532585144043, "learning_rate": 9.374564119289837e-05, "loss": 0.7557895660400391, "memory(GiB)": 61.91, "step": 18780, "token_acc": 0.8349514563106796, "train_speed(iter/s)": 1.474161 }, { "epoch": 0.8048069919883467, "grad_norm": 7.29603385925293, "learning_rate": 9.37423816991672e-05, "loss": 0.4730524063110352, "memory(GiB)": 61.91, "step": 18785, "token_acc": 0.8933333333333333, "train_speed(iter/s)": 1.474192 }, { "epoch": 0.8050212073175956, "grad_norm": 2.860114097595215, "learning_rate": 9.373912141300177e-05, "loss": 0.2779209136962891, "memory(GiB)": 61.91, "step": 18790, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.474195 }, { "epoch": 0.8052354226468446, "grad_norm": 3.792863130569458, "learning_rate": 9.37358603344611e-05, "loss": 0.3281688928604126, "memory(GiB)": 61.91, "step": 18795, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.474209 }, { "epoch": 0.8054496379760936, "grad_norm": 4.797098636627197, "learning_rate": 9.37325984636043e-05, "loss": 0.5995500087738037, "memory(GiB)": 61.91, "step": 18800, "token_acc": 0.8639455782312925, "train_speed(iter/s)": 1.474182 }, { "epoch": 0.8056638533053425, "grad_norm": 3.3216679096221924, "learning_rate": 9.372933580049047e-05, "loss": 0.3559230327606201, "memory(GiB)": 61.91, "step": 18805, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.474177 }, { "epoch": 0.8058780686345914, "grad_norm": 2.3485395908355713, "learning_rate": 9.372607234517868e-05, "loss": 0.6152901172637939, "memory(GiB)": 61.91, "step": 18810, "token_acc": 0.8925925925925926, "train_speed(iter/s)": 1.474189 }, { "epoch": 0.8060922839638405, "grad_norm": 3.7759811878204346, "learning_rate": 9.372280809772808e-05, "loss": 0.28699588775634766, "memory(GiB)": 61.91, "step": 18815, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.4742 }, { "epoch": 0.8063064992930894, "grad_norm": 1.3878213167190552, "learning_rate": 9.371954305819779e-05, "loss": 0.4122321128845215, "memory(GiB)": 61.91, "step": 18820, "token_acc": 0.9225806451612903, "train_speed(iter/s)": 1.474176 }, { "epoch": 0.8065207146223383, "grad_norm": 2.0881776809692383, "learning_rate": 9.371627722664698e-05, "loss": 0.3179079532623291, "memory(GiB)": 61.91, "step": 18825, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.474157 }, { "epoch": 0.8067349299515874, "grad_norm": 4.9934983253479, "learning_rate": 9.371301060313477e-05, "loss": 0.41310868263244627, "memory(GiB)": 61.91, "step": 18830, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.474153 }, { "epoch": 0.8069491452808363, "grad_norm": 4.637001037597656, "learning_rate": 9.370974318772038e-05, "loss": 0.2987504243850708, "memory(GiB)": 61.91, "step": 18835, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.474152 }, { "epoch": 0.8071633606100853, "grad_norm": 2.6170268058776855, "learning_rate": 9.370647498046302e-05, "loss": 0.3300474166870117, "memory(GiB)": 61.91, "step": 18840, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.474156 }, { "epoch": 0.8073775759393342, "grad_norm": 6.155663013458252, "learning_rate": 9.370320598142183e-05, "loss": 0.3968080520629883, "memory(GiB)": 61.91, "step": 18845, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.474171 }, { "epoch": 0.8075917912685832, "grad_norm": 4.704369068145752, "learning_rate": 9.369993619065608e-05, "loss": 0.35061140060424806, "memory(GiB)": 61.91, "step": 18850, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.474152 }, { "epoch": 0.8078060065978322, "grad_norm": 9.355813026428223, "learning_rate": 9.369666560822498e-05, "loss": 0.3230825662612915, "memory(GiB)": 61.91, "step": 18855, "token_acc": 0.9255813953488372, "train_speed(iter/s)": 1.474149 }, { "epoch": 0.8080202219270811, "grad_norm": 3.7218616008758545, "learning_rate": 9.36933942341878e-05, "loss": 0.4278896331787109, "memory(GiB)": 61.91, "step": 18860, "token_acc": 0.8953488372093024, "train_speed(iter/s)": 1.474133 }, { "epoch": 0.80823443725633, "grad_norm": 0.6616474390029907, "learning_rate": 9.369012206860381e-05, "loss": 0.44423184394836424, "memory(GiB)": 61.91, "step": 18865, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.474161 }, { "epoch": 0.8084486525855791, "grad_norm": 6.573335647583008, "learning_rate": 9.368684911153225e-05, "loss": 0.5402679920196534, "memory(GiB)": 61.91, "step": 18870, "token_acc": 0.9063670411985019, "train_speed(iter/s)": 1.474167 }, { "epoch": 0.808662867914828, "grad_norm": 2.6111624240875244, "learning_rate": 9.368357536303245e-05, "loss": 0.34785354137420654, "memory(GiB)": 61.91, "step": 18875, "token_acc": 0.9345794392523364, "train_speed(iter/s)": 1.474187 }, { "epoch": 0.8088770832440769, "grad_norm": 6.986985683441162, "learning_rate": 9.36803008231637e-05, "loss": 0.4593794822692871, "memory(GiB)": 61.91, "step": 18880, "token_acc": 0.8876404494382022, "train_speed(iter/s)": 1.4742 }, { "epoch": 0.809091298573326, "grad_norm": 7.402266502380371, "learning_rate": 9.367702549198533e-05, "loss": 0.4630213737487793, "memory(GiB)": 61.91, "step": 18885, "token_acc": 0.9076305220883534, "train_speed(iter/s)": 1.474228 }, { "epoch": 0.8093055139025749, "grad_norm": 2.8385252952575684, "learning_rate": 9.367374936955666e-05, "loss": 0.45731048583984374, "memory(GiB)": 61.91, "step": 18890, "token_acc": 0.8898305084745762, "train_speed(iter/s)": 1.474263 }, { "epoch": 0.8095197292318238, "grad_norm": 2.6554296016693115, "learning_rate": 9.367047245593705e-05, "loss": 0.48888463973999025, "memory(GiB)": 61.91, "step": 18895, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.474282 }, { "epoch": 0.8097339445610728, "grad_norm": 1.9531031847000122, "learning_rate": 9.366719475118588e-05, "loss": 0.24012346267700196, "memory(GiB)": 61.91, "step": 18900, "token_acc": 0.9517684887459807, "train_speed(iter/s)": 1.474291 }, { "epoch": 0.8099481598903218, "grad_norm": 2.2696683406829834, "learning_rate": 9.36639162553625e-05, "loss": 0.25869848728179934, "memory(GiB)": 61.91, "step": 18905, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.47431 }, { "epoch": 0.8101623752195707, "grad_norm": 5.082864761352539, "learning_rate": 9.366063696852634e-05, "loss": 0.4358375549316406, "memory(GiB)": 61.91, "step": 18910, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.474363 }, { "epoch": 0.8103765905488197, "grad_norm": 3.258709669113159, "learning_rate": 9.365735689073676e-05, "loss": 0.32673428058624265, "memory(GiB)": 61.91, "step": 18915, "token_acc": 0.9119496855345912, "train_speed(iter/s)": 1.474364 }, { "epoch": 0.8105908058780686, "grad_norm": 0.0666128545999527, "learning_rate": 9.365407602205322e-05, "loss": 0.28112177848815917, "memory(GiB)": 61.91, "step": 18920, "token_acc": 0.9540229885057471, "train_speed(iter/s)": 1.474372 }, { "epoch": 0.8108050212073176, "grad_norm": 4.607535362243652, "learning_rate": 9.365079436253513e-05, "loss": 0.5724672317504883, "memory(GiB)": 61.91, "step": 18925, "token_acc": 0.8805460750853242, "train_speed(iter/s)": 1.474422 }, { "epoch": 0.8110192365365666, "grad_norm": 2.0325076580047607, "learning_rate": 9.364751191224197e-05, "loss": 0.32991025447845457, "memory(GiB)": 61.91, "step": 18930, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.474439 }, { "epoch": 0.8112334518658155, "grad_norm": 2.2790427207946777, "learning_rate": 9.364422867123317e-05, "loss": 0.619900894165039, "memory(GiB)": 61.91, "step": 18935, "token_acc": 0.8678571428571429, "train_speed(iter/s)": 1.474434 }, { "epoch": 0.8114476671950644, "grad_norm": 4.718181133270264, "learning_rate": 9.364094463956823e-05, "loss": 0.596822452545166, "memory(GiB)": 61.91, "step": 18940, "token_acc": 0.8787878787878788, "train_speed(iter/s)": 1.474423 }, { "epoch": 0.8116618825243135, "grad_norm": 2.8911564350128174, "learning_rate": 9.363765981730664e-05, "loss": 0.2434302806854248, "memory(GiB)": 61.91, "step": 18945, "token_acc": 0.9388489208633094, "train_speed(iter/s)": 1.474417 }, { "epoch": 0.8118760978535624, "grad_norm": 1.4176558256149292, "learning_rate": 9.363437420450791e-05, "loss": 0.49009385108947756, "memory(GiB)": 61.91, "step": 18950, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.474416 }, { "epoch": 0.8120903131828113, "grad_norm": 2.526258707046509, "learning_rate": 9.363108780123155e-05, "loss": 0.32928004264831545, "memory(GiB)": 61.91, "step": 18955, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.4744 }, { "epoch": 0.8123045285120604, "grad_norm": 2.693472385406494, "learning_rate": 9.362780060753712e-05, "loss": 0.5690230846405029, "memory(GiB)": 61.91, "step": 18960, "token_acc": 0.8689024390243902, "train_speed(iter/s)": 1.474391 }, { "epoch": 0.8125187438413093, "grad_norm": 3.4599695205688477, "learning_rate": 9.362451262348414e-05, "loss": 0.4181980133056641, "memory(GiB)": 61.91, "step": 18965, "token_acc": 0.926605504587156, "train_speed(iter/s)": 1.474441 }, { "epoch": 0.8127329591705582, "grad_norm": 3.047734022140503, "learning_rate": 9.362122384913221e-05, "loss": 0.24616765975952148, "memory(GiB)": 61.91, "step": 18970, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.474412 }, { "epoch": 0.8129471744998072, "grad_norm": 0.4647904932498932, "learning_rate": 9.361793428454087e-05, "loss": 0.5162190437316895, "memory(GiB)": 61.91, "step": 18975, "token_acc": 0.897196261682243, "train_speed(iter/s)": 1.474433 }, { "epoch": 0.8131613898290562, "grad_norm": 3.0282905101776123, "learning_rate": 9.361464392976975e-05, "loss": 0.3116239070892334, "memory(GiB)": 61.91, "step": 18980, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.474429 }, { "epoch": 0.8133756051583051, "grad_norm": 3.4018967151641846, "learning_rate": 9.361135278487843e-05, "loss": 0.42847495079040526, "memory(GiB)": 61.91, "step": 18985, "token_acc": 0.8995983935742972, "train_speed(iter/s)": 1.474427 }, { "epoch": 0.8135898204875541, "grad_norm": 3.2361583709716797, "learning_rate": 9.360806084992655e-05, "loss": 0.4822692394256592, "memory(GiB)": 61.91, "step": 18990, "token_acc": 0.9008264462809917, "train_speed(iter/s)": 1.474418 }, { "epoch": 0.813804035816803, "grad_norm": 1.6058682203292847, "learning_rate": 9.360476812497374e-05, "loss": 0.2878437519073486, "memory(GiB)": 61.91, "step": 18995, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.474393 }, { "epoch": 0.814018251146052, "grad_norm": 3.3386144638061523, "learning_rate": 9.360147461007964e-05, "loss": 0.8450477600097657, "memory(GiB)": 61.91, "step": 19000, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.474394 }, { "epoch": 0.814018251146052, "eval_loss": 2.439305067062378, "eval_runtime": 13.8739, "eval_samples_per_second": 7.208, "eval_steps_per_second": 7.208, "eval_token_acc": 0.44487179487179485, "step": 19000 }, { "epoch": 0.814232466475301, "grad_norm": 2.59073543548584, "learning_rate": 9.359818030530394e-05, "loss": 0.30266761779785156, "memory(GiB)": 61.91, "step": 19005, "token_acc": 0.5733082706766918, "train_speed(iter/s)": 1.472712 }, { "epoch": 0.8144466818045499, "grad_norm": 1.5947685241699219, "learning_rate": 9.359488521070629e-05, "loss": 0.228110933303833, "memory(GiB)": 61.91, "step": 19010, "token_acc": 0.9251968503937008, "train_speed(iter/s)": 1.472685 }, { "epoch": 0.8146608971337989, "grad_norm": 3.439403533935547, "learning_rate": 9.359158932634642e-05, "loss": 0.5216547012329101, "memory(GiB)": 61.91, "step": 19015, "token_acc": 0.8835616438356164, "train_speed(iter/s)": 1.472686 }, { "epoch": 0.8148751124630479, "grad_norm": 3.48807954788208, "learning_rate": 9.3588292652284e-05, "loss": 0.4797548294067383, "memory(GiB)": 61.91, "step": 19020, "token_acc": 0.9040697674418605, "train_speed(iter/s)": 1.472681 }, { "epoch": 0.8150893277922968, "grad_norm": 3.1717610359191895, "learning_rate": 9.358499518857877e-05, "loss": 0.6077849388122558, "memory(GiB)": 61.91, "step": 19025, "token_acc": 0.8694029850746269, "train_speed(iter/s)": 1.472674 }, { "epoch": 0.8153035431215457, "grad_norm": 4.88322114944458, "learning_rate": 9.358169693529048e-05, "loss": 0.24203591346740722, "memory(GiB)": 61.91, "step": 19030, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.472678 }, { "epoch": 0.8155177584507948, "grad_norm": 2.191582679748535, "learning_rate": 9.357839789247886e-05, "loss": 0.1745816469192505, "memory(GiB)": 61.91, "step": 19035, "token_acc": 0.9630996309963099, "train_speed(iter/s)": 1.472665 }, { "epoch": 0.8157319737800437, "grad_norm": 2.9046220779418945, "learning_rate": 9.357509806020369e-05, "loss": 0.3799008369445801, "memory(GiB)": 61.91, "step": 19040, "token_acc": 0.9392857142857143, "train_speed(iter/s)": 1.472707 }, { "epoch": 0.8159461891092926, "grad_norm": 6.391145706176758, "learning_rate": 9.357179743852471e-05, "loss": 0.7935803413391114, "memory(GiB)": 61.91, "step": 19045, "token_acc": 0.8317152103559871, "train_speed(iter/s)": 1.472717 }, { "epoch": 0.8161604044385417, "grad_norm": 4.065824031829834, "learning_rate": 9.356849602750177e-05, "loss": 0.30049724578857423, "memory(GiB)": 61.91, "step": 19050, "token_acc": 0.9250814332247557, "train_speed(iter/s)": 1.472706 }, { "epoch": 0.8163746197677906, "grad_norm": 3.638380527496338, "learning_rate": 9.356519382719467e-05, "loss": 0.5196584701538086, "memory(GiB)": 61.91, "step": 19055, "token_acc": 0.891156462585034, "train_speed(iter/s)": 1.472689 }, { "epoch": 0.8165888350970395, "grad_norm": 3.2578444480895996, "learning_rate": 9.356189083766318e-05, "loss": 0.3574237823486328, "memory(GiB)": 61.91, "step": 19060, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.472684 }, { "epoch": 0.8168030504262885, "grad_norm": 3.715829610824585, "learning_rate": 9.35585870589672e-05, "loss": 0.40453262329101564, "memory(GiB)": 61.91, "step": 19065, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.472686 }, { "epoch": 0.8170172657555375, "grad_norm": 3.0946171283721924, "learning_rate": 9.355528249116653e-05, "loss": 0.31367878913879393, "memory(GiB)": 61.91, "step": 19070, "token_acc": 0.9446064139941691, "train_speed(iter/s)": 1.472675 }, { "epoch": 0.8172314810847864, "grad_norm": 2.619549512863159, "learning_rate": 9.355197713432109e-05, "loss": 0.5052403450012207, "memory(GiB)": 61.91, "step": 19075, "token_acc": 0.8996763754045307, "train_speed(iter/s)": 1.472651 }, { "epoch": 0.8174456964140354, "grad_norm": 3.74698805809021, "learning_rate": 9.354867098849071e-05, "loss": 0.8028478622436523, "memory(GiB)": 61.91, "step": 19080, "token_acc": 0.8210116731517509, "train_speed(iter/s)": 1.472647 }, { "epoch": 0.8176599117432843, "grad_norm": 3.5552828311920166, "learning_rate": 9.35453640537353e-05, "loss": 0.4255534172058105, "memory(GiB)": 61.91, "step": 19085, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.47264 }, { "epoch": 0.8178741270725333, "grad_norm": 3.9059505462646484, "learning_rate": 9.354205633011479e-05, "loss": 0.3600472927093506, "memory(GiB)": 61.91, "step": 19090, "token_acc": 0.9203539823008849, "train_speed(iter/s)": 1.47262 }, { "epoch": 0.8180883424017823, "grad_norm": 1.1627929210662842, "learning_rate": 9.353874781768908e-05, "loss": 0.29270546436309813, "memory(GiB)": 61.91, "step": 19095, "token_acc": 0.9201520912547528, "train_speed(iter/s)": 1.47264 }, { "epoch": 0.8183025577310312, "grad_norm": 2.7965238094329834, "learning_rate": 9.353543851651809e-05, "loss": 0.3217471122741699, "memory(GiB)": 61.91, "step": 19100, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.472652 }, { "epoch": 0.8185167730602801, "grad_norm": 2.8606653213500977, "learning_rate": 9.353212842666181e-05, "loss": 0.38212993144989016, "memory(GiB)": 61.91, "step": 19105, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.472687 }, { "epoch": 0.8187309883895292, "grad_norm": 1.8077986240386963, "learning_rate": 9.352881754818019e-05, "loss": 0.37478744983673096, "memory(GiB)": 61.91, "step": 19110, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.472686 }, { "epoch": 0.8189452037187781, "grad_norm": 2.803656578063965, "learning_rate": 9.352550588113319e-05, "loss": 0.37788023948669436, "memory(GiB)": 61.91, "step": 19115, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.472717 }, { "epoch": 0.819159419048027, "grad_norm": 2.832307815551758, "learning_rate": 9.352219342558083e-05, "loss": 0.2159646987915039, "memory(GiB)": 61.91, "step": 19120, "token_acc": 0.958041958041958, "train_speed(iter/s)": 1.472726 }, { "epoch": 0.8193736343772761, "grad_norm": 2.089731216430664, "learning_rate": 9.351888018158312e-05, "loss": 0.42191877365112307, "memory(GiB)": 61.91, "step": 19125, "token_acc": 0.9150326797385621, "train_speed(iter/s)": 1.472721 }, { "epoch": 0.819587849706525, "grad_norm": 5.491288661956787, "learning_rate": 9.351556614920005e-05, "loss": 0.47777667045593264, "memory(GiB)": 61.91, "step": 19130, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.472756 }, { "epoch": 0.8198020650357739, "grad_norm": 8.668190002441406, "learning_rate": 9.351225132849169e-05, "loss": 0.6943512916564941, "memory(GiB)": 61.91, "step": 19135, "token_acc": 0.8413284132841329, "train_speed(iter/s)": 1.47274 }, { "epoch": 0.8200162803650229, "grad_norm": 4.6993727684021, "learning_rate": 9.350893571951808e-05, "loss": 0.3699460506439209, "memory(GiB)": 61.91, "step": 19140, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.472722 }, { "epoch": 0.8202304956942719, "grad_norm": 2.303630828857422, "learning_rate": 9.350561932233928e-05, "loss": 0.4959548473358154, "memory(GiB)": 61.91, "step": 19145, "token_acc": 0.8981132075471698, "train_speed(iter/s)": 1.472699 }, { "epoch": 0.8204447110235208, "grad_norm": 2.99418306350708, "learning_rate": 9.350230213701537e-05, "loss": 0.303556489944458, "memory(GiB)": 61.91, "step": 19150, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.472742 }, { "epoch": 0.8206589263527698, "grad_norm": 2.9417405128479004, "learning_rate": 9.349898416360646e-05, "loss": 0.46001410484313965, "memory(GiB)": 61.91, "step": 19155, "token_acc": 0.8798701298701299, "train_speed(iter/s)": 1.47272 }, { "epoch": 0.8208731416820187, "grad_norm": 3.5693695545196533, "learning_rate": 9.349566540217265e-05, "loss": 0.440250825881958, "memory(GiB)": 61.91, "step": 19160, "token_acc": 0.8905109489051095, "train_speed(iter/s)": 1.472715 }, { "epoch": 0.8210873570112677, "grad_norm": 2.6258506774902344, "learning_rate": 9.349234585277404e-05, "loss": 0.19887547492980956, "memory(GiB)": 61.91, "step": 19165, "token_acc": 0.9433198380566802, "train_speed(iter/s)": 1.472703 }, { "epoch": 0.8213015723405167, "grad_norm": 3.6009066104888916, "learning_rate": 9.348902551547081e-05, "loss": 0.5374238014221191, "memory(GiB)": 61.91, "step": 19170, "token_acc": 0.8809523809523809, "train_speed(iter/s)": 1.472711 }, { "epoch": 0.8215157876697656, "grad_norm": 5.308638095855713, "learning_rate": 9.348570439032306e-05, "loss": 0.5686711311340332, "memory(GiB)": 61.91, "step": 19175, "token_acc": 0.88, "train_speed(iter/s)": 1.472714 }, { "epoch": 0.8217300029990147, "grad_norm": 3.057790517807007, "learning_rate": 9.348238247739101e-05, "loss": 0.30715460777282716, "memory(GiB)": 61.91, "step": 19180, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.472754 }, { "epoch": 0.8219442183282636, "grad_norm": 4.371999263763428, "learning_rate": 9.34790597767348e-05, "loss": 0.35199265480041503, "memory(GiB)": 61.91, "step": 19185, "token_acc": 0.9135135135135135, "train_speed(iter/s)": 1.472758 }, { "epoch": 0.8221584336575125, "grad_norm": 3.6531121730804443, "learning_rate": 9.347573628841462e-05, "loss": 0.1371090292930603, "memory(GiB)": 61.91, "step": 19190, "token_acc": 0.9726443768996961, "train_speed(iter/s)": 1.472753 }, { "epoch": 0.8223726489867615, "grad_norm": 7.031786918640137, "learning_rate": 9.347241201249071e-05, "loss": 0.5809253692626953, "memory(GiB)": 61.91, "step": 19195, "token_acc": 0.8833922261484098, "train_speed(iter/s)": 1.472739 }, { "epoch": 0.8225868643160105, "grad_norm": 2.096583604812622, "learning_rate": 9.346908694902327e-05, "loss": 0.14714035987854004, "memory(GiB)": 61.91, "step": 19200, "token_acc": 0.9649805447470817, "train_speed(iter/s)": 1.472734 }, { "epoch": 0.8228010796452594, "grad_norm": 0.17493672668933868, "learning_rate": 9.346576109807255e-05, "loss": 0.28557643890380857, "memory(GiB)": 61.91, "step": 19205, "token_acc": 0.9528619528619529, "train_speed(iter/s)": 1.472728 }, { "epoch": 0.8230152949745084, "grad_norm": 4.258440017700195, "learning_rate": 9.346243445969877e-05, "loss": 0.23144361972808838, "memory(GiB)": 61.91, "step": 19210, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.472726 }, { "epoch": 0.8232295103037574, "grad_norm": 4.072005271911621, "learning_rate": 9.345910703396225e-05, "loss": 0.5084473133087158, "memory(GiB)": 61.91, "step": 19215, "token_acc": 0.8943396226415095, "train_speed(iter/s)": 1.47271 }, { "epoch": 0.8234437256330063, "grad_norm": 4.0961174964904785, "learning_rate": 9.345577882092322e-05, "loss": 0.6102946281433106, "memory(GiB)": 61.91, "step": 19220, "token_acc": 0.8831168831168831, "train_speed(iter/s)": 1.472685 }, { "epoch": 0.8236579409622553, "grad_norm": 4.212896823883057, "learning_rate": 9.345244982064201e-05, "loss": 0.5795434474945068, "memory(GiB)": 61.91, "step": 19225, "token_acc": 0.889344262295082, "train_speed(iter/s)": 1.472677 }, { "epoch": 0.8238721562915042, "grad_norm": 5.359926700592041, "learning_rate": 9.344912003317888e-05, "loss": 0.5850218772888184, "memory(GiB)": 61.91, "step": 19230, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.47269 }, { "epoch": 0.8240863716207532, "grad_norm": 4.482243537902832, "learning_rate": 9.344578945859421e-05, "loss": 0.43910512924194334, "memory(GiB)": 61.91, "step": 19235, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.472691 }, { "epoch": 0.8243005869500022, "grad_norm": 2.1947062015533447, "learning_rate": 9.344245809694829e-05, "loss": 0.3532389163970947, "memory(GiB)": 61.91, "step": 19240, "token_acc": 0.9348659003831418, "train_speed(iter/s)": 1.472735 }, { "epoch": 0.8245148022792511, "grad_norm": 4.864712715148926, "learning_rate": 9.34391259483015e-05, "loss": 0.3359057903289795, "memory(GiB)": 61.91, "step": 19245, "token_acc": 0.926605504587156, "train_speed(iter/s)": 1.472728 }, { "epoch": 0.8247290176085, "grad_norm": 6.054345607757568, "learning_rate": 9.343579301271419e-05, "loss": 0.49776368141174315, "memory(GiB)": 61.91, "step": 19250, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.472731 }, { "epoch": 0.8249432329377491, "grad_norm": 1.5577641725540161, "learning_rate": 9.343245929024674e-05, "loss": 0.19263956546783448, "memory(GiB)": 61.91, "step": 19255, "token_acc": 0.9566666666666667, "train_speed(iter/s)": 1.472715 }, { "epoch": 0.825157448266998, "grad_norm": 2.759538412094116, "learning_rate": 9.342912478095955e-05, "loss": 0.13361979722976686, "memory(GiB)": 61.91, "step": 19260, "token_acc": 0.9702602230483272, "train_speed(iter/s)": 1.472754 }, { "epoch": 0.8253716635962469, "grad_norm": 2.1765236854553223, "learning_rate": 9.342578948491303e-05, "loss": 0.4039007663726807, "memory(GiB)": 61.91, "step": 19265, "token_acc": 0.9325153374233128, "train_speed(iter/s)": 1.472756 }, { "epoch": 0.825585878925496, "grad_norm": 4.276610851287842, "learning_rate": 9.342245340216759e-05, "loss": 0.31638140678405763, "memory(GiB)": 61.91, "step": 19270, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.472799 }, { "epoch": 0.8258000942547449, "grad_norm": 4.581518173217773, "learning_rate": 9.341911653278367e-05, "loss": 0.3911776304244995, "memory(GiB)": 61.91, "step": 19275, "token_acc": 0.9079497907949791, "train_speed(iter/s)": 1.472825 }, { "epoch": 0.8260143095839938, "grad_norm": 3.986985206604004, "learning_rate": 9.341577887682173e-05, "loss": 0.5428249835968018, "memory(GiB)": 61.91, "step": 19280, "token_acc": 0.8988095238095238, "train_speed(iter/s)": 1.472826 }, { "epoch": 0.8262285249132428, "grad_norm": 2.997633934020996, "learning_rate": 9.341244043434224e-05, "loss": 0.45124268531799316, "memory(GiB)": 61.91, "step": 19285, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.472818 }, { "epoch": 0.8264427402424918, "grad_norm": 0.1601894050836563, "learning_rate": 9.340910120540564e-05, "loss": 0.37769279479980467, "memory(GiB)": 61.91, "step": 19290, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.472794 }, { "epoch": 0.8266569555717407, "grad_norm": 5.740532398223877, "learning_rate": 9.340576119007249e-05, "loss": 0.3863893985748291, "memory(GiB)": 61.91, "step": 19295, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.472788 }, { "epoch": 0.8268711709009897, "grad_norm": 4.799330711364746, "learning_rate": 9.340242038840322e-05, "loss": 0.3972153186798096, "memory(GiB)": 61.91, "step": 19300, "token_acc": 0.9137380191693291, "train_speed(iter/s)": 1.472789 }, { "epoch": 0.8270853862302386, "grad_norm": 1.4455618858337402, "learning_rate": 9.33990788004584e-05, "loss": 0.2682557344436646, "memory(GiB)": 61.91, "step": 19305, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.472782 }, { "epoch": 0.8272996015594876, "grad_norm": 2.0476417541503906, "learning_rate": 9.339573642629857e-05, "loss": 0.3413562536239624, "memory(GiB)": 61.91, "step": 19310, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.472786 }, { "epoch": 0.8275138168887366, "grad_norm": 2.2752909660339355, "learning_rate": 9.339239326598426e-05, "loss": 0.30293362140655516, "memory(GiB)": 61.91, "step": 19315, "token_acc": 0.9228571428571428, "train_speed(iter/s)": 1.472801 }, { "epoch": 0.8277280322179855, "grad_norm": 2.9143218994140625, "learning_rate": 9.338904931957603e-05, "loss": 0.680009412765503, "memory(GiB)": 61.91, "step": 19320, "token_acc": 0.8653846153846154, "train_speed(iter/s)": 1.472779 }, { "epoch": 0.8279422475472344, "grad_norm": 5.008427143096924, "learning_rate": 9.338570458713447e-05, "loss": 0.411362361907959, "memory(GiB)": 61.91, "step": 19325, "token_acc": 0.9027237354085603, "train_speed(iter/s)": 1.472865 }, { "epoch": 0.8281564628764835, "grad_norm": 3.0995407104492188, "learning_rate": 9.338235906872019e-05, "loss": 0.5149702072143555, "memory(GiB)": 61.91, "step": 19330, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.472868 }, { "epoch": 0.8283706782057324, "grad_norm": 2.428689479827881, "learning_rate": 9.337901276439376e-05, "loss": 0.6144418239593505, "memory(GiB)": 61.91, "step": 19335, "token_acc": 0.8650137741046832, "train_speed(iter/s)": 1.472861 }, { "epoch": 0.8285848935349813, "grad_norm": 2.458258628845215, "learning_rate": 9.337566567421583e-05, "loss": 0.19618405103683473, "memory(GiB)": 61.91, "step": 19340, "token_acc": 0.9653846153846154, "train_speed(iter/s)": 1.472866 }, { "epoch": 0.8287991088642304, "grad_norm": 2.3355448246002197, "learning_rate": 9.337231779824703e-05, "loss": 0.29977974891662595, "memory(GiB)": 61.91, "step": 19345, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.47286 }, { "epoch": 0.8290133241934793, "grad_norm": 0.9014405012130737, "learning_rate": 9.3368969136548e-05, "loss": 0.0906871497631073, "memory(GiB)": 61.91, "step": 19350, "token_acc": 0.976271186440678, "train_speed(iter/s)": 1.472865 }, { "epoch": 0.8292275395227282, "grad_norm": 2.082282066345215, "learning_rate": 9.33656196891794e-05, "loss": 0.2911111831665039, "memory(GiB)": 61.91, "step": 19355, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.472862 }, { "epoch": 0.8294417548519772, "grad_norm": 4.015933513641357, "learning_rate": 9.336226945620194e-05, "loss": 0.3188253164291382, "memory(GiB)": 61.91, "step": 19360, "token_acc": 0.9441176470588235, "train_speed(iter/s)": 1.472877 }, { "epoch": 0.8296559701812262, "grad_norm": 3.6134047508239746, "learning_rate": 9.335891843767629e-05, "loss": 0.23585994243621827, "memory(GiB)": 61.91, "step": 19365, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.472877 }, { "epoch": 0.8298701855104751, "grad_norm": 5.540928840637207, "learning_rate": 9.335556663366314e-05, "loss": 0.5639671325683594, "memory(GiB)": 61.91, "step": 19370, "token_acc": 0.8988326848249028, "train_speed(iter/s)": 1.472879 }, { "epoch": 0.8300844008397241, "grad_norm": 3.872849225997925, "learning_rate": 9.335221404422325e-05, "loss": 0.3930069923400879, "memory(GiB)": 61.91, "step": 19375, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.472899 }, { "epoch": 0.830298616168973, "grad_norm": 4.165943622589111, "learning_rate": 9.334886066941733e-05, "loss": 0.4462653636932373, "memory(GiB)": 61.91, "step": 19380, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.472925 }, { "epoch": 0.830512831498222, "grad_norm": 4.067141532897949, "learning_rate": 9.334550650930613e-05, "loss": 0.33069891929626466, "memory(GiB)": 61.91, "step": 19385, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.472906 }, { "epoch": 0.830727046827471, "grad_norm": 2.40267276763916, "learning_rate": 9.334215156395042e-05, "loss": 0.3941451072692871, "memory(GiB)": 61.91, "step": 19390, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.472921 }, { "epoch": 0.8309412621567199, "grad_norm": 5.130852699279785, "learning_rate": 9.333879583341098e-05, "loss": 0.4472458839416504, "memory(GiB)": 61.91, "step": 19395, "token_acc": 0.8939393939393939, "train_speed(iter/s)": 1.472921 }, { "epoch": 0.8311554774859689, "grad_norm": 3.2712209224700928, "learning_rate": 9.33354393177486e-05, "loss": 0.17288806438446044, "memory(GiB)": 61.91, "step": 19400, "token_acc": 0.96484375, "train_speed(iter/s)": 1.472953 }, { "epoch": 0.8313696928152179, "grad_norm": 2.483206033706665, "learning_rate": 9.333208201702407e-05, "loss": 0.28351900577545164, "memory(GiB)": 61.91, "step": 19405, "token_acc": 0.9361022364217252, "train_speed(iter/s)": 1.472925 }, { "epoch": 0.8315839081444668, "grad_norm": 3.9488375186920166, "learning_rate": 9.332872393129823e-05, "loss": 0.44515290260314944, "memory(GiB)": 61.91, "step": 19410, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.472903 }, { "epoch": 0.8317981234737157, "grad_norm": 1.6634522676467896, "learning_rate": 9.332536506063192e-05, "loss": 0.2669835090637207, "memory(GiB)": 61.91, "step": 19415, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.4729 }, { "epoch": 0.8320123388029648, "grad_norm": 1.3530875444412231, "learning_rate": 9.332200540508598e-05, "loss": 0.3492603063583374, "memory(GiB)": 61.91, "step": 19420, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.472884 }, { "epoch": 0.8322265541322137, "grad_norm": 3.6625425815582275, "learning_rate": 9.331864496472125e-05, "loss": 0.5048963069915772, "memory(GiB)": 61.91, "step": 19425, "token_acc": 0.9015384615384615, "train_speed(iter/s)": 1.472923 }, { "epoch": 0.8324407694614626, "grad_norm": 1.1776357889175415, "learning_rate": 9.331528373959865e-05, "loss": 0.5741371631622314, "memory(GiB)": 61.91, "step": 19430, "token_acc": 0.8916083916083916, "train_speed(iter/s)": 1.472911 }, { "epoch": 0.8326549847907116, "grad_norm": 9.338237762451172, "learning_rate": 9.331192172977905e-05, "loss": 0.46270976066589353, "memory(GiB)": 61.91, "step": 19435, "token_acc": 0.9018987341772152, "train_speed(iter/s)": 1.472906 }, { "epoch": 0.8328692001199606, "grad_norm": 9.46874713897705, "learning_rate": 9.330855893532335e-05, "loss": 0.3568193197250366, "memory(GiB)": 61.91, "step": 19440, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.472907 }, { "epoch": 0.8330834154492095, "grad_norm": 5.1518754959106445, "learning_rate": 9.330519535629249e-05, "loss": 0.6339482307434082, "memory(GiB)": 61.91, "step": 19445, "token_acc": 0.8775510204081632, "train_speed(iter/s)": 1.472983 }, { "epoch": 0.8332976307784585, "grad_norm": 0.6205894947052002, "learning_rate": 9.330183099274739e-05, "loss": 0.457975435256958, "memory(GiB)": 61.91, "step": 19450, "token_acc": 0.8893280632411067, "train_speed(iter/s)": 1.472969 }, { "epoch": 0.8335118461077075, "grad_norm": 5.514767646789551, "learning_rate": 9.329846584474899e-05, "loss": 0.338388729095459, "memory(GiB)": 61.91, "step": 19455, "token_acc": 0.9343283582089552, "train_speed(iter/s)": 1.472984 }, { "epoch": 0.8337260614369564, "grad_norm": 3.650951623916626, "learning_rate": 9.329509991235829e-05, "loss": 0.5828299522399902, "memory(GiB)": 61.91, "step": 19460, "token_acc": 0.896875, "train_speed(iter/s)": 1.472977 }, { "epoch": 0.8339402767662054, "grad_norm": 1.7858247756958008, "learning_rate": 9.329173319563622e-05, "loss": 0.4154338359832764, "memory(GiB)": 61.91, "step": 19465, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.472967 }, { "epoch": 0.8341544920954543, "grad_norm": 4.056972980499268, "learning_rate": 9.328836569464379e-05, "loss": 0.3688324451446533, "memory(GiB)": 61.91, "step": 19470, "token_acc": 0.905511811023622, "train_speed(iter/s)": 1.472972 }, { "epoch": 0.8343687074247033, "grad_norm": 2.705793619155884, "learning_rate": 9.328499740944201e-05, "loss": 0.2579374313354492, "memory(GiB)": 61.91, "step": 19475, "token_acc": 0.9343065693430657, "train_speed(iter/s)": 1.473001 }, { "epoch": 0.8345829227539523, "grad_norm": 3.4004242420196533, "learning_rate": 9.328162834009192e-05, "loss": 0.3307967185974121, "memory(GiB)": 61.91, "step": 19480, "token_acc": 0.9174311926605505, "train_speed(iter/s)": 1.473028 }, { "epoch": 0.8347971380832012, "grad_norm": 4.374651908874512, "learning_rate": 9.32782584866545e-05, "loss": 0.37250757217407227, "memory(GiB)": 61.91, "step": 19485, "token_acc": 0.91015625, "train_speed(iter/s)": 1.473106 }, { "epoch": 0.8350113534124501, "grad_norm": 3.9501664638519287, "learning_rate": 9.327488784919084e-05, "loss": 0.4289849281311035, "memory(GiB)": 61.91, "step": 19490, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.473096 }, { "epoch": 0.8352255687416992, "grad_norm": 1.8994230031967163, "learning_rate": 9.3271516427762e-05, "loss": 0.4600456714630127, "memory(GiB)": 61.91, "step": 19495, "token_acc": 0.891640866873065, "train_speed(iter/s)": 1.473129 }, { "epoch": 0.8354397840709481, "grad_norm": 2.688825845718384, "learning_rate": 9.326814422242905e-05, "loss": 0.4552611351013184, "memory(GiB)": 61.91, "step": 19500, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.473118 }, { "epoch": 0.8354397840709481, "eval_loss": 2.5957086086273193, "eval_runtime": 13.5208, "eval_samples_per_second": 7.396, "eval_steps_per_second": 7.396, "eval_token_acc": 0.4501347708894879, "step": 19500 }, { "epoch": 0.835653999400197, "grad_norm": 1.680031180381775, "learning_rate": 9.326477123325306e-05, "loss": 0.41425323486328125, "memory(GiB)": 61.91, "step": 19505, "token_acc": 0.5855072463768116, "train_speed(iter/s)": 1.471535 }, { "epoch": 0.8358682147294461, "grad_norm": 3.567096710205078, "learning_rate": 9.326139746029516e-05, "loss": 0.3464086055755615, "memory(GiB)": 61.91, "step": 19510, "token_acc": 0.9296875, "train_speed(iter/s)": 1.47155 }, { "epoch": 0.836082430058695, "grad_norm": 2.3329124450683594, "learning_rate": 9.325802290361647e-05, "loss": 0.5270912647247314, "memory(GiB)": 61.91, "step": 19515, "token_acc": 0.9105058365758755, "train_speed(iter/s)": 1.471558 }, { "epoch": 0.836296645387944, "grad_norm": 6.764080047607422, "learning_rate": 9.32546475632781e-05, "loss": 0.3815983772277832, "memory(GiB)": 61.91, "step": 19520, "token_acc": 0.9272030651340997, "train_speed(iter/s)": 1.471561 }, { "epoch": 0.8365108607171929, "grad_norm": 5.95261287689209, "learning_rate": 9.325127143934122e-05, "loss": 0.39394464492797854, "memory(GiB)": 61.91, "step": 19525, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.471566 }, { "epoch": 0.8367250760464419, "grad_norm": 2.433192729949951, "learning_rate": 9.324789453186699e-05, "loss": 0.4622025012969971, "memory(GiB)": 61.91, "step": 19530, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.471546 }, { "epoch": 0.8369392913756909, "grad_norm": 2.9949519634246826, "learning_rate": 9.32445168409166e-05, "loss": 0.28785841464996337, "memory(GiB)": 61.91, "step": 19535, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.471558 }, { "epoch": 0.8371535067049398, "grad_norm": 5.3140645027160645, "learning_rate": 9.324113836655119e-05, "loss": 0.5126853942871094, "memory(GiB)": 61.91, "step": 19540, "token_acc": 0.8786764705882353, "train_speed(iter/s)": 1.47154 }, { "epoch": 0.8373677220341887, "grad_norm": 11.70101547241211, "learning_rate": 9.323775910883202e-05, "loss": 0.2675257921218872, "memory(GiB)": 61.91, "step": 19545, "token_acc": 0.9453376205787781, "train_speed(iter/s)": 1.471541 }, { "epoch": 0.8375819373634378, "grad_norm": 3.498056411743164, "learning_rate": 9.323437906782026e-05, "loss": 0.4317812919616699, "memory(GiB)": 61.91, "step": 19550, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.471532 }, { "epoch": 0.8377961526926867, "grad_norm": 3.1353650093078613, "learning_rate": 9.323099824357717e-05, "loss": 0.3121234893798828, "memory(GiB)": 61.91, "step": 19555, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.471519 }, { "epoch": 0.8380103680219356, "grad_norm": 3.3862123489379883, "learning_rate": 9.3227616636164e-05, "loss": 0.4029379844665527, "memory(GiB)": 61.91, "step": 19560, "token_acc": 0.9057971014492754, "train_speed(iter/s)": 1.47151 }, { "epoch": 0.8382245833511847, "grad_norm": 2.7181103229522705, "learning_rate": 9.322423424564201e-05, "loss": 0.48924813270568845, "memory(GiB)": 61.91, "step": 19565, "token_acc": 0.8885245901639345, "train_speed(iter/s)": 1.471503 }, { "epoch": 0.8384387986804336, "grad_norm": 0.13268689811229706, "learning_rate": 9.322085107207245e-05, "loss": 0.2576186418533325, "memory(GiB)": 61.91, "step": 19570, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.471496 }, { "epoch": 0.8386530140096825, "grad_norm": 5.9556660652160645, "learning_rate": 9.321746711551665e-05, "loss": 0.6194622993469239, "memory(GiB)": 61.91, "step": 19575, "token_acc": 0.8754448398576512, "train_speed(iter/s)": 1.471489 }, { "epoch": 0.8388672293389315, "grad_norm": 8.855518341064453, "learning_rate": 9.321408237603589e-05, "loss": 0.4577075481414795, "memory(GiB)": 61.91, "step": 19580, "token_acc": 0.911042944785276, "train_speed(iter/s)": 1.471481 }, { "epoch": 0.8390814446681805, "grad_norm": 2.042194128036499, "learning_rate": 9.321069685369147e-05, "loss": 0.18877631425857544, "memory(GiB)": 61.91, "step": 19585, "token_acc": 0.9694915254237289, "train_speed(iter/s)": 1.471477 }, { "epoch": 0.8392956599974294, "grad_norm": 6.207080841064453, "learning_rate": 9.320731054854474e-05, "loss": 0.7610092163085938, "memory(GiB)": 61.91, "step": 19590, "token_acc": 0.853035143769968, "train_speed(iter/s)": 1.471472 }, { "epoch": 0.8395098753266784, "grad_norm": 0.06736212968826294, "learning_rate": 9.320392346065707e-05, "loss": 0.2847642421722412, "memory(GiB)": 61.91, "step": 19595, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.471528 }, { "epoch": 0.8397240906559273, "grad_norm": 5.991504192352295, "learning_rate": 9.320053559008979e-05, "loss": 0.45770835876464844, "memory(GiB)": 61.91, "step": 19600, "token_acc": 0.9314516129032258, "train_speed(iter/s)": 1.471512 }, { "epoch": 0.8399383059851763, "grad_norm": 3.5598912239074707, "learning_rate": 9.319714693690429e-05, "loss": 0.3171184301376343, "memory(GiB)": 61.91, "step": 19605, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.471512 }, { "epoch": 0.8401525213144253, "grad_norm": 3.8613409996032715, "learning_rate": 9.319375750116194e-05, "loss": 0.4055172920227051, "memory(GiB)": 61.91, "step": 19610, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.471512 }, { "epoch": 0.8403667366436742, "grad_norm": 4.739055156707764, "learning_rate": 9.319036728292415e-05, "loss": 0.4142427444458008, "memory(GiB)": 61.91, "step": 19615, "token_acc": 0.9003831417624522, "train_speed(iter/s)": 1.471515 }, { "epoch": 0.8405809519729232, "grad_norm": 1.6333402395248413, "learning_rate": 9.318697628225235e-05, "loss": 0.27523250579833985, "memory(GiB)": 61.91, "step": 19620, "token_acc": 0.9375, "train_speed(iter/s)": 1.471554 }, { "epoch": 0.8407951673021722, "grad_norm": 2.2876830101013184, "learning_rate": 9.318358449920795e-05, "loss": 0.18862769603729249, "memory(GiB)": 61.91, "step": 19625, "token_acc": 0.9711191335740073, "train_speed(iter/s)": 1.471566 }, { "epoch": 0.8410093826314211, "grad_norm": 2.8671374320983887, "learning_rate": 9.318019193385242e-05, "loss": 0.3607171535491943, "memory(GiB)": 61.91, "step": 19630, "token_acc": 0.9383116883116883, "train_speed(iter/s)": 1.471588 }, { "epoch": 0.84122359796067, "grad_norm": 2.1109578609466553, "learning_rate": 9.317679858624721e-05, "loss": 0.5994267463684082, "memory(GiB)": 61.91, "step": 19635, "token_acc": 0.888135593220339, "train_speed(iter/s)": 1.471592 }, { "epoch": 0.8414378132899191, "grad_norm": 4.771324634552002, "learning_rate": 9.317340445645377e-05, "loss": 0.5655547142028808, "memory(GiB)": 61.91, "step": 19640, "token_acc": 0.8641975308641975, "train_speed(iter/s)": 1.471648 }, { "epoch": 0.841652028619168, "grad_norm": 2.5655357837677, "learning_rate": 9.317000954453364e-05, "loss": 0.6138123035430908, "memory(GiB)": 61.91, "step": 19645, "token_acc": 0.8900709219858156, "train_speed(iter/s)": 1.471676 }, { "epoch": 0.8418662439484169, "grad_norm": 3.39235258102417, "learning_rate": 9.316661385054825e-05, "loss": 0.4506237983703613, "memory(GiB)": 61.91, "step": 19650, "token_acc": 0.9052287581699346, "train_speed(iter/s)": 1.471705 }, { "epoch": 0.842080459277666, "grad_norm": 2.2666776180267334, "learning_rate": 9.316321737455919e-05, "loss": 0.3023213863372803, "memory(GiB)": 61.91, "step": 19655, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.471743 }, { "epoch": 0.8422946746069149, "grad_norm": 3.395669460296631, "learning_rate": 9.315982011662794e-05, "loss": 0.3260129451751709, "memory(GiB)": 61.91, "step": 19660, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.471759 }, { "epoch": 0.8425088899361638, "grad_norm": 5.267627239227295, "learning_rate": 9.315642207681607e-05, "loss": 0.6995327472686768, "memory(GiB)": 61.91, "step": 19665, "token_acc": 0.8512658227848101, "train_speed(iter/s)": 1.47177 }, { "epoch": 0.8427231052654128, "grad_norm": 3.3722968101501465, "learning_rate": 9.315302325518514e-05, "loss": 0.36089329719543456, "memory(GiB)": 61.91, "step": 19670, "token_acc": 0.9202898550724637, "train_speed(iter/s)": 1.471749 }, { "epoch": 0.8429373205946618, "grad_norm": 8.115644454956055, "learning_rate": 9.31496236517967e-05, "loss": 0.6061041831970215, "memory(GiB)": 61.91, "step": 19675, "token_acc": 0.9019073569482289, "train_speed(iter/s)": 1.471807 }, { "epoch": 0.8431515359239107, "grad_norm": 3.678255796432495, "learning_rate": 9.314622326671233e-05, "loss": 0.5306111812591553, "memory(GiB)": 61.91, "step": 19680, "token_acc": 0.8757961783439491, "train_speed(iter/s)": 1.471835 }, { "epoch": 0.8433657512531597, "grad_norm": 2.04264235496521, "learning_rate": 9.314282209999368e-05, "loss": 0.5059330463409424, "memory(GiB)": 61.91, "step": 19685, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.47186 }, { "epoch": 0.8435799665824086, "grad_norm": 4.152689456939697, "learning_rate": 9.313942015170233e-05, "loss": 0.770423412322998, "memory(GiB)": 61.91, "step": 19690, "token_acc": 0.849112426035503, "train_speed(iter/s)": 1.471851 }, { "epoch": 0.8437941819116576, "grad_norm": 4.532154560089111, "learning_rate": 9.31360174218999e-05, "loss": 0.4731895446777344, "memory(GiB)": 61.91, "step": 19695, "token_acc": 0.8939393939393939, "train_speed(iter/s)": 1.471835 }, { "epoch": 0.8440083972409066, "grad_norm": 3.135373830795288, "learning_rate": 9.313261391064807e-05, "loss": 0.42493476867675783, "memory(GiB)": 61.91, "step": 19700, "token_acc": 0.9023569023569024, "train_speed(iter/s)": 1.471861 }, { "epoch": 0.8442226125701555, "grad_norm": 4.659185886383057, "learning_rate": 9.312920961800847e-05, "loss": 0.39133548736572266, "memory(GiB)": 61.91, "step": 19705, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.47186 }, { "epoch": 0.8444368278994044, "grad_norm": 1.7986674308776855, "learning_rate": 9.312580454404278e-05, "loss": 0.2670176029205322, "memory(GiB)": 61.91, "step": 19710, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.47186 }, { "epoch": 0.8446510432286535, "grad_norm": 4.140810489654541, "learning_rate": 9.312239868881268e-05, "loss": 0.35165934562683104, "memory(GiB)": 61.91, "step": 19715, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.47185 }, { "epoch": 0.8448652585579024, "grad_norm": 6.392854690551758, "learning_rate": 9.311899205237989e-05, "loss": 0.5227189540863038, "memory(GiB)": 61.91, "step": 19720, "token_acc": 0.8986928104575164, "train_speed(iter/s)": 1.47183 }, { "epoch": 0.8450794738871513, "grad_norm": 7.27832555770874, "learning_rate": 9.31155846348061e-05, "loss": 0.5432696342468262, "memory(GiB)": 61.91, "step": 19725, "token_acc": 0.8848484848484849, "train_speed(iter/s)": 1.471823 }, { "epoch": 0.8452936892164004, "grad_norm": 2.5845816135406494, "learning_rate": 9.311217643615304e-05, "loss": 0.8471970558166504, "memory(GiB)": 61.91, "step": 19730, "token_acc": 0.8271276595744681, "train_speed(iter/s)": 1.471844 }, { "epoch": 0.8455079045456493, "grad_norm": 0.7281131148338318, "learning_rate": 9.310876745648247e-05, "loss": 0.16599799394607545, "memory(GiB)": 61.91, "step": 19735, "token_acc": 0.968421052631579, "train_speed(iter/s)": 1.471889 }, { "epoch": 0.8457221198748982, "grad_norm": 6.5422773361206055, "learning_rate": 9.310535769585615e-05, "loss": 0.41595001220703126, "memory(GiB)": 61.91, "step": 19740, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.471875 }, { "epoch": 0.8459363352041472, "grad_norm": 3.186211347579956, "learning_rate": 9.310194715433583e-05, "loss": 0.7690975666046143, "memory(GiB)": 61.91, "step": 19745, "token_acc": 0.8598726114649682, "train_speed(iter/s)": 1.471896 }, { "epoch": 0.8461505505333962, "grad_norm": 3.912923574447632, "learning_rate": 9.309853583198328e-05, "loss": 0.40468683242797854, "memory(GiB)": 61.91, "step": 19750, "token_acc": 0.9120234604105572, "train_speed(iter/s)": 1.471898 }, { "epoch": 0.8463647658626451, "grad_norm": 5.038135528564453, "learning_rate": 9.309512372886036e-05, "loss": 0.5050382137298584, "memory(GiB)": 61.91, "step": 19755, "token_acc": 0.9126506024096386, "train_speed(iter/s)": 1.471899 }, { "epoch": 0.8465789811918941, "grad_norm": 1.8981236219406128, "learning_rate": 9.309171084502883e-05, "loss": 0.6356444835662842, "memory(GiB)": 61.91, "step": 19760, "token_acc": 0.8605341246290801, "train_speed(iter/s)": 1.471893 }, { "epoch": 0.846793196521143, "grad_norm": 1.8713494539260864, "learning_rate": 9.308829718055054e-05, "loss": 0.24184052944183348, "memory(GiB)": 61.91, "step": 19765, "token_acc": 0.9393063583815029, "train_speed(iter/s)": 1.471895 }, { "epoch": 0.847007411850392, "grad_norm": 1.6851919889450073, "learning_rate": 9.308488273548732e-05, "loss": 0.17262299060821534, "memory(GiB)": 61.91, "step": 19770, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.471911 }, { "epoch": 0.847221627179641, "grad_norm": 4.5855865478515625, "learning_rate": 9.308146750990106e-05, "loss": 0.30072736740112305, "memory(GiB)": 61.91, "step": 19775, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.471939 }, { "epoch": 0.8474358425088899, "grad_norm": 0.6932580471038818, "learning_rate": 9.307805150385356e-05, "loss": 0.39714038372039795, "memory(GiB)": 61.91, "step": 19780, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.471944 }, { "epoch": 0.8476500578381388, "grad_norm": 2.5599021911621094, "learning_rate": 9.307463471740678e-05, "loss": 0.35082030296325684, "memory(GiB)": 61.91, "step": 19785, "token_acc": 0.9112627986348123, "train_speed(iter/s)": 1.47192 }, { "epoch": 0.8478642731673879, "grad_norm": 1.313427209854126, "learning_rate": 9.307121715062257e-05, "loss": 0.3513232946395874, "memory(GiB)": 61.91, "step": 19790, "token_acc": 0.9251497005988024, "train_speed(iter/s)": 1.471914 }, { "epoch": 0.8480784884966368, "grad_norm": 2.6140010356903076, "learning_rate": 9.306779880356284e-05, "loss": 0.3670816898345947, "memory(GiB)": 61.91, "step": 19795, "token_acc": 0.9106382978723404, "train_speed(iter/s)": 1.471913 }, { "epoch": 0.8482927038258857, "grad_norm": 2.5445475578308105, "learning_rate": 9.306437967628956e-05, "loss": 0.5435813426971435, "memory(GiB)": 61.91, "step": 19800, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.471903 }, { "epoch": 0.8485069191551348, "grad_norm": 2.0050761699676514, "learning_rate": 9.306095976886464e-05, "loss": 0.2378622770309448, "memory(GiB)": 61.91, "step": 19805, "token_acc": 0.9558011049723757, "train_speed(iter/s)": 1.471899 }, { "epoch": 0.8487211344843837, "grad_norm": 4.822188377380371, "learning_rate": 9.305753908135003e-05, "loss": 0.520949125289917, "memory(GiB)": 61.91, "step": 19810, "token_acc": 0.885558583106267, "train_speed(iter/s)": 1.471877 }, { "epoch": 0.8489353498136326, "grad_norm": 1.3905186653137207, "learning_rate": 9.305411761380771e-05, "loss": 0.5374310493469239, "memory(GiB)": 61.91, "step": 19815, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.471935 }, { "epoch": 0.8491495651428816, "grad_norm": 4.218071460723877, "learning_rate": 9.305069536629967e-05, "loss": 0.45853352546691895, "memory(GiB)": 61.91, "step": 19820, "token_acc": 0.8943089430894309, "train_speed(iter/s)": 1.471915 }, { "epoch": 0.8493637804721306, "grad_norm": 0.12447553128004074, "learning_rate": 9.30472723388879e-05, "loss": 0.371260404586792, "memory(GiB)": 61.91, "step": 19825, "token_acc": 0.9224376731301939, "train_speed(iter/s)": 1.471893 }, { "epoch": 0.8495779958013795, "grad_norm": 2.613992691040039, "learning_rate": 9.30438485316344e-05, "loss": 0.4059854507446289, "memory(GiB)": 61.91, "step": 19830, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.471902 }, { "epoch": 0.8497922111306285, "grad_norm": 1.0069931745529175, "learning_rate": 9.30404239446012e-05, "loss": 0.16037768125534058, "memory(GiB)": 61.91, "step": 19835, "token_acc": 0.9608540925266904, "train_speed(iter/s)": 1.471935 }, { "epoch": 0.8500064264598775, "grad_norm": 3.861947536468506, "learning_rate": 9.303699857785035e-05, "loss": 0.5155746936798096, "memory(GiB)": 61.91, "step": 19840, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.471928 }, { "epoch": 0.8502206417891264, "grad_norm": 1.1237943172454834, "learning_rate": 9.303357243144392e-05, "loss": 0.2743096828460693, "memory(GiB)": 61.91, "step": 19845, "token_acc": 0.9531772575250836, "train_speed(iter/s)": 1.471925 }, { "epoch": 0.8504348571183754, "grad_norm": 3.7272391319274902, "learning_rate": 9.303014550544394e-05, "loss": 0.1996646285057068, "memory(GiB)": 61.91, "step": 19850, "token_acc": 0.9566666666666667, "train_speed(iter/s)": 1.471911 }, { "epoch": 0.8506490724476243, "grad_norm": 3.2184457778930664, "learning_rate": 9.30267177999125e-05, "loss": 0.3648238658905029, "memory(GiB)": 61.91, "step": 19855, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.471908 }, { "epoch": 0.8508632877768734, "grad_norm": 5.93240213394165, "learning_rate": 9.302328931491172e-05, "loss": 0.6529603004455566, "memory(GiB)": 61.91, "step": 19860, "token_acc": 0.8694158075601375, "train_speed(iter/s)": 1.47189 }, { "epoch": 0.8510775031061223, "grad_norm": 2.273712396621704, "learning_rate": 9.301986005050369e-05, "loss": 0.553393030166626, "memory(GiB)": 61.91, "step": 19865, "token_acc": 0.8790035587188612, "train_speed(iter/s)": 1.471883 }, { "epoch": 0.8512917184353712, "grad_norm": 3.3166182041168213, "learning_rate": 9.301643000675054e-05, "loss": 0.5587393283843994, "memory(GiB)": 61.91, "step": 19870, "token_acc": 0.8795986622073578, "train_speed(iter/s)": 1.471869 }, { "epoch": 0.8515059337646202, "grad_norm": 1.9719276428222656, "learning_rate": 9.30129991837144e-05, "loss": 0.3009920358657837, "memory(GiB)": 61.91, "step": 19875, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.471879 }, { "epoch": 0.8517201490938692, "grad_norm": 2.2975635528564453, "learning_rate": 9.300956758145742e-05, "loss": 0.354633092880249, "memory(GiB)": 61.91, "step": 19880, "token_acc": 0.9084967320261438, "train_speed(iter/s)": 1.471891 }, { "epoch": 0.8519343644231181, "grad_norm": 3.512188196182251, "learning_rate": 9.30061352000418e-05, "loss": 0.49337472915649416, "memory(GiB)": 61.91, "step": 19885, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.47189 }, { "epoch": 0.8521485797523671, "grad_norm": 1.4285236597061157, "learning_rate": 9.30027020395297e-05, "loss": 0.2660136461257935, "memory(GiB)": 61.91, "step": 19890, "token_acc": 0.945054945054945, "train_speed(iter/s)": 1.471874 }, { "epoch": 0.852362795081616, "grad_norm": 2.2853806018829346, "learning_rate": 9.299926809998329e-05, "loss": 0.11218425035476684, "memory(GiB)": 61.91, "step": 19895, "token_acc": 0.9713114754098361, "train_speed(iter/s)": 1.47188 }, { "epoch": 0.852577010410865, "grad_norm": 1.8349584341049194, "learning_rate": 9.299583338146483e-05, "loss": 0.5171759128570557, "memory(GiB)": 61.91, "step": 19900, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.471892 }, { "epoch": 0.852791225740114, "grad_norm": 1.8868131637573242, "learning_rate": 9.299239788403649e-05, "loss": 0.3040379762649536, "memory(GiB)": 61.91, "step": 19905, "token_acc": 0.9448818897637795, "train_speed(iter/s)": 1.471891 }, { "epoch": 0.8530054410693629, "grad_norm": 3.666639566421509, "learning_rate": 9.298896160776054e-05, "loss": 0.7052950859069824, "memory(GiB)": 61.91, "step": 19910, "token_acc": 0.850187265917603, "train_speed(iter/s)": 1.471863 }, { "epoch": 0.8532196563986119, "grad_norm": 4.224941730499268, "learning_rate": 9.298552455269923e-05, "loss": 0.35053489208221433, "memory(GiB)": 61.91, "step": 19915, "token_acc": 0.9224137931034483, "train_speed(iter/s)": 1.471862 }, { "epoch": 0.8534338717278609, "grad_norm": 3.5377261638641357, "learning_rate": 9.298208671891482e-05, "loss": 0.4311653137207031, "memory(GiB)": 61.91, "step": 19920, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.471875 }, { "epoch": 0.8536480870571098, "grad_norm": 4.543691635131836, "learning_rate": 9.297864810646958e-05, "loss": 0.45357646942138674, "memory(GiB)": 61.91, "step": 19925, "token_acc": 0.883495145631068, "train_speed(iter/s)": 1.471911 }, { "epoch": 0.8538623023863587, "grad_norm": 2.096405029296875, "learning_rate": 9.297520871542583e-05, "loss": 0.12067979574203491, "memory(GiB)": 61.91, "step": 19930, "token_acc": 0.9754385964912281, "train_speed(iter/s)": 1.471903 }, { "epoch": 0.8540765177156078, "grad_norm": 2.7295875549316406, "learning_rate": 9.297176854584582e-05, "loss": 0.13116652965545655, "memory(GiB)": 61.91, "step": 19935, "token_acc": 0.9706959706959707, "train_speed(iter/s)": 1.471894 }, { "epoch": 0.8542907330448567, "grad_norm": 3.653522253036499, "learning_rate": 9.296832759779195e-05, "loss": 0.3280023097991943, "memory(GiB)": 61.91, "step": 19940, "token_acc": 0.9308176100628931, "train_speed(iter/s)": 1.471895 }, { "epoch": 0.8545049483741056, "grad_norm": 1.3786207437515259, "learning_rate": 9.29648858713265e-05, "loss": 0.5261239528656005, "memory(GiB)": 61.91, "step": 19945, "token_acc": 0.8834586466165414, "train_speed(iter/s)": 1.471863 }, { "epoch": 0.8547191637033547, "grad_norm": 0.8776336908340454, "learning_rate": 9.296144336651185e-05, "loss": 0.5163407325744629, "memory(GiB)": 61.91, "step": 19950, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.47187 }, { "epoch": 0.8549333790326036, "grad_norm": 3.3514463901519775, "learning_rate": 9.295800008341033e-05, "loss": 0.7294382095336914, "memory(GiB)": 61.91, "step": 19955, "token_acc": 0.855457227138643, "train_speed(iter/s)": 1.471906 }, { "epoch": 0.8551475943618525, "grad_norm": 6.463566303253174, "learning_rate": 9.295455602208438e-05, "loss": 0.5850791931152344, "memory(GiB)": 61.91, "step": 19960, "token_acc": 0.8929765886287625, "train_speed(iter/s)": 1.471898 }, { "epoch": 0.8553618096911015, "grad_norm": 1.3091460466384888, "learning_rate": 9.295111118259632e-05, "loss": 0.30006554126739504, "memory(GiB)": 61.91, "step": 19965, "token_acc": 0.9392857142857143, "train_speed(iter/s)": 1.471973 }, { "epoch": 0.8555760250203505, "grad_norm": 1.5328657627105713, "learning_rate": 9.29476655650086e-05, "loss": 0.38243653774261477, "memory(GiB)": 61.91, "step": 19970, "token_acc": 0.9192546583850931, "train_speed(iter/s)": 1.472023 }, { "epoch": 0.8557902403495994, "grad_norm": 2.081610679626465, "learning_rate": 9.294421916938363e-05, "loss": 0.5400920391082764, "memory(GiB)": 61.91, "step": 19975, "token_acc": 0.8754578754578755, "train_speed(iter/s)": 1.472029 }, { "epoch": 0.8560044556788484, "grad_norm": 2.4541804790496826, "learning_rate": 9.294077199578384e-05, "loss": 0.28597092628479004, "memory(GiB)": 61.91, "step": 19980, "token_acc": 0.9405204460966543, "train_speed(iter/s)": 1.472035 }, { "epoch": 0.8562186710080973, "grad_norm": 2.325462579727173, "learning_rate": 9.293732404427169e-05, "loss": 0.6353636741638183, "memory(GiB)": 61.91, "step": 19985, "token_acc": 0.8770491803278688, "train_speed(iter/s)": 1.47203 }, { "epoch": 0.8564328863373463, "grad_norm": 3.061760902404785, "learning_rate": 9.293387531490964e-05, "loss": 0.23824920654296874, "memory(GiB)": 61.91, "step": 19990, "token_acc": 0.946875, "train_speed(iter/s)": 1.472041 }, { "epoch": 0.8566471016665953, "grad_norm": 2.4279537200927734, "learning_rate": 9.293042580776015e-05, "loss": 0.2872252702713013, "memory(GiB)": 61.91, "step": 19995, "token_acc": 0.9263565891472868, "train_speed(iter/s)": 1.472075 }, { "epoch": 0.8568613169958442, "grad_norm": 2.666952610015869, "learning_rate": 9.292697552288574e-05, "loss": 0.3319696426391602, "memory(GiB)": 61.91, "step": 20000, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.472069 }, { "epoch": 0.8568613169958442, "eval_loss": 2.5874948501586914, "eval_runtime": 13.3386, "eval_samples_per_second": 7.497, "eval_steps_per_second": 7.497, "eval_token_acc": 0.42136498516320475, "step": 20000 }, { "epoch": 0.8570755323250931, "grad_norm": 5.899051666259766, "learning_rate": 9.292352446034889e-05, "loss": 0.5617931842803955, "memory(GiB)": 61.91, "step": 20005, "token_acc": 0.555439330543933, "train_speed(iter/s)": 1.470515 }, { "epoch": 0.8572897476543422, "grad_norm": 1.9955731630325317, "learning_rate": 9.292007262021213e-05, "loss": 0.3105919361114502, "memory(GiB)": 61.91, "step": 20010, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.470536 }, { "epoch": 0.8575039629835911, "grad_norm": 4.718433380126953, "learning_rate": 9.291662000253799e-05, "loss": 0.40196914672851564, "memory(GiB)": 61.91, "step": 20015, "token_acc": 0.8864468864468864, "train_speed(iter/s)": 1.470538 }, { "epoch": 0.85771817831284, "grad_norm": 1.6956686973571777, "learning_rate": 9.291316660738903e-05, "loss": 0.25438950061798093, "memory(GiB)": 61.91, "step": 20020, "token_acc": 0.939209726443769, "train_speed(iter/s)": 1.47057 }, { "epoch": 0.8579323936420891, "grad_norm": 1.0845787525177002, "learning_rate": 9.290971243482778e-05, "loss": 0.37956478595733645, "memory(GiB)": 61.91, "step": 20025, "token_acc": 0.9190031152647975, "train_speed(iter/s)": 1.470568 }, { "epoch": 0.858146608971338, "grad_norm": 4.024974822998047, "learning_rate": 9.290625748491686e-05, "loss": 0.7599750518798828, "memory(GiB)": 61.91, "step": 20030, "token_acc": 0.8475609756097561, "train_speed(iter/s)": 1.470598 }, { "epoch": 0.8583608243005869, "grad_norm": 3.2862374782562256, "learning_rate": 9.290280175771882e-05, "loss": 0.2909094333648682, "memory(GiB)": 61.91, "step": 20035, "token_acc": 0.9458333333333333, "train_speed(iter/s)": 1.470602 }, { "epoch": 0.858575039629836, "grad_norm": 1.980050802230835, "learning_rate": 9.289934525329628e-05, "loss": 0.4299412727355957, "memory(GiB)": 61.91, "step": 20040, "token_acc": 0.90234375, "train_speed(iter/s)": 1.470622 }, { "epoch": 0.8587892549590849, "grad_norm": 8.636500358581543, "learning_rate": 9.289588797171186e-05, "loss": 0.4340050220489502, "memory(GiB)": 61.91, "step": 20045, "token_acc": 0.9149659863945578, "train_speed(iter/s)": 1.470756 }, { "epoch": 0.8590034702883338, "grad_norm": 2.5225956439971924, "learning_rate": 9.289242991302821e-05, "loss": 0.3914666652679443, "memory(GiB)": 61.91, "step": 20050, "token_acc": 0.8873239436619719, "train_speed(iter/s)": 1.470744 }, { "epoch": 0.8592176856175828, "grad_norm": 3.282421588897705, "learning_rate": 9.288897107730794e-05, "loss": 0.5906006813049316, "memory(GiB)": 61.91, "step": 20055, "token_acc": 0.875, "train_speed(iter/s)": 1.470778 }, { "epoch": 0.8594319009468318, "grad_norm": 3.961298942565918, "learning_rate": 9.288551146461373e-05, "loss": 0.38909914493560793, "memory(GiB)": 61.91, "step": 20060, "token_acc": 0.9041666666666667, "train_speed(iter/s)": 1.470764 }, { "epoch": 0.8596461162760807, "grad_norm": 2.4347612857818604, "learning_rate": 9.288205107500825e-05, "loss": 0.6195327281951905, "memory(GiB)": 61.91, "step": 20065, "token_acc": 0.8648648648648649, "train_speed(iter/s)": 1.470808 }, { "epoch": 0.8598603316053297, "grad_norm": 3.616971254348755, "learning_rate": 9.287858990855418e-05, "loss": 0.39984588623046874, "memory(GiB)": 61.91, "step": 20070, "token_acc": 0.9201277955271565, "train_speed(iter/s)": 1.470788 }, { "epoch": 0.8600745469345786, "grad_norm": 4.9394402503967285, "learning_rate": 9.287512796531423e-05, "loss": 0.3668442487716675, "memory(GiB)": 61.91, "step": 20075, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.470813 }, { "epoch": 0.8602887622638276, "grad_norm": 3.415799856185913, "learning_rate": 9.287166524535111e-05, "loss": 0.4930895805358887, "memory(GiB)": 61.91, "step": 20080, "token_acc": 0.8642857142857143, "train_speed(iter/s)": 1.470785 }, { "epoch": 0.8605029775930766, "grad_norm": 2.4158530235290527, "learning_rate": 9.286820174872758e-05, "loss": 0.5660175800323486, "memory(GiB)": 61.91, "step": 20085, "token_acc": 0.8803680981595092, "train_speed(iter/s)": 1.470801 }, { "epoch": 0.8607171929223255, "grad_norm": 3.700089693069458, "learning_rate": 9.286473747550635e-05, "loss": 0.2886655330657959, "memory(GiB)": 61.91, "step": 20090, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.470783 }, { "epoch": 0.8609314082515744, "grad_norm": 0.8853837251663208, "learning_rate": 9.286127242575019e-05, "loss": 0.187231183052063, "memory(GiB)": 61.91, "step": 20095, "token_acc": 0.9628482972136223, "train_speed(iter/s)": 1.470791 }, { "epoch": 0.8611456235808235, "grad_norm": 3.345749855041504, "learning_rate": 9.285780659952188e-05, "loss": 0.4053945541381836, "memory(GiB)": 61.91, "step": 20100, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.470808 }, { "epoch": 0.8613598389100724, "grad_norm": 3.787198543548584, "learning_rate": 9.285433999688419e-05, "loss": 0.32616052627563474, "memory(GiB)": 61.91, "step": 20105, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.470808 }, { "epoch": 0.8615740542393213, "grad_norm": 5.233165740966797, "learning_rate": 9.285087261789993e-05, "loss": 0.5410067558288574, "memory(GiB)": 61.91, "step": 20110, "token_acc": 0.8791208791208791, "train_speed(iter/s)": 1.470826 }, { "epoch": 0.8617882695685704, "grad_norm": 0.8143125772476196, "learning_rate": 9.284740446263191e-05, "loss": 0.08311324715614318, "memory(GiB)": 61.91, "step": 20115, "token_acc": 0.9810126582278481, "train_speed(iter/s)": 1.470825 }, { "epoch": 0.8620024848978193, "grad_norm": 3.2865731716156006, "learning_rate": 9.284393553114298e-05, "loss": 0.5665054321289062, "memory(GiB)": 61.91, "step": 20120, "token_acc": 0.8864468864468864, "train_speed(iter/s)": 1.470816 }, { "epoch": 0.8622167002270682, "grad_norm": 3.846881151199341, "learning_rate": 9.284046582349596e-05, "loss": 0.3428520202636719, "memory(GiB)": 61.91, "step": 20125, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.470839 }, { "epoch": 0.8624309155563172, "grad_norm": 5.846823215484619, "learning_rate": 9.283699533975372e-05, "loss": 0.48529872894287107, "memory(GiB)": 61.91, "step": 20130, "token_acc": 0.9097744360902256, "train_speed(iter/s)": 1.470824 }, { "epoch": 0.8626451308855662, "grad_norm": 3.064837694168091, "learning_rate": 9.283352407997912e-05, "loss": 0.4784424304962158, "memory(GiB)": 61.91, "step": 20135, "token_acc": 0.8854489164086687, "train_speed(iter/s)": 1.470793 }, { "epoch": 0.8628593462148151, "grad_norm": 3.0719735622406006, "learning_rate": 9.283005204423504e-05, "loss": 0.31009361743927, "memory(GiB)": 61.91, "step": 20140, "token_acc": 0.9169960474308301, "train_speed(iter/s)": 1.470812 }, { "epoch": 0.8630735615440641, "grad_norm": 0.6391885280609131, "learning_rate": 9.28265792325844e-05, "loss": 0.2515124320983887, "memory(GiB)": 61.91, "step": 20145, "token_acc": 0.9510869565217391, "train_speed(iter/s)": 1.470803 }, { "epoch": 0.863287776873313, "grad_norm": 0.9864287376403809, "learning_rate": 9.282310564509009e-05, "loss": 0.3010514736175537, "memory(GiB)": 61.91, "step": 20150, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.470804 }, { "epoch": 0.863501992202562, "grad_norm": 3.1748859882354736, "learning_rate": 9.281963128181508e-05, "loss": 0.3412697076797485, "memory(GiB)": 61.91, "step": 20155, "token_acc": 0.9260450160771704, "train_speed(iter/s)": 1.470804 }, { "epoch": 0.863716207531811, "grad_norm": 3.3222877979278564, "learning_rate": 9.281615614282225e-05, "loss": 0.4447681427001953, "memory(GiB)": 61.91, "step": 20160, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.47079 }, { "epoch": 0.8639304228610599, "grad_norm": 3.368241786956787, "learning_rate": 9.28126802281746e-05, "loss": 0.31921777725219724, "memory(GiB)": 61.91, "step": 20165, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.470791 }, { "epoch": 0.8641446381903088, "grad_norm": 4.449585914611816, "learning_rate": 9.280920353793508e-05, "loss": 0.5786193370819092, "memory(GiB)": 61.91, "step": 20170, "token_acc": 0.8757961783439491, "train_speed(iter/s)": 1.470788 }, { "epoch": 0.8643588535195579, "grad_norm": 5.571672439575195, "learning_rate": 9.280572607216668e-05, "loss": 0.5033541679382324, "memory(GiB)": 61.91, "step": 20175, "token_acc": 0.8758620689655172, "train_speed(iter/s)": 1.470804 }, { "epoch": 0.8645730688488068, "grad_norm": 3.843304395675659, "learning_rate": 9.280224783093241e-05, "loss": 0.4961112022399902, "memory(GiB)": 61.91, "step": 20180, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.4708 }, { "epoch": 0.8647872841780557, "grad_norm": 3.6941306591033936, "learning_rate": 9.279876881429526e-05, "loss": 0.4249885082244873, "memory(GiB)": 61.91, "step": 20185, "token_acc": 0.915129151291513, "train_speed(iter/s)": 1.470805 }, { "epoch": 0.8650014995073048, "grad_norm": 2.8257648944854736, "learning_rate": 9.279528902231827e-05, "loss": 0.3045284032821655, "memory(GiB)": 61.91, "step": 20190, "token_acc": 0.9283489096573209, "train_speed(iter/s)": 1.470834 }, { "epoch": 0.8652157148365537, "grad_norm": 2.070969820022583, "learning_rate": 9.279180845506446e-05, "loss": 0.15169551372528076, "memory(GiB)": 61.91, "step": 20195, "token_acc": 0.9750889679715302, "train_speed(iter/s)": 1.470857 }, { "epoch": 0.8654299301658027, "grad_norm": 3.058532953262329, "learning_rate": 9.27883271125969e-05, "loss": 0.4051656246185303, "memory(GiB)": 61.91, "step": 20200, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.470906 }, { "epoch": 0.8656441454950516, "grad_norm": 3.8719303607940674, "learning_rate": 9.278484499497868e-05, "loss": 0.40245370864868163, "memory(GiB)": 61.91, "step": 20205, "token_acc": 0.9060402684563759, "train_speed(iter/s)": 1.470888 }, { "epoch": 0.8658583608243006, "grad_norm": 4.998509407043457, "learning_rate": 9.278136210227284e-05, "loss": 0.2352292060852051, "memory(GiB)": 61.91, "step": 20210, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.470887 }, { "epoch": 0.8660725761535496, "grad_norm": 0.4039033055305481, "learning_rate": 9.277787843454248e-05, "loss": 0.3789071083068848, "memory(GiB)": 61.91, "step": 20215, "token_acc": 0.9152542372881356, "train_speed(iter/s)": 1.470879 }, { "epoch": 0.8662867914827985, "grad_norm": 4.707996368408203, "learning_rate": 9.277439399185074e-05, "loss": 0.5556529045104981, "memory(GiB)": 61.91, "step": 20220, "token_acc": 0.8859315589353612, "train_speed(iter/s)": 1.470906 }, { "epoch": 0.8665010068120474, "grad_norm": 2.0808002948760986, "learning_rate": 9.277090877426074e-05, "loss": 0.33698134422302245, "memory(GiB)": 61.91, "step": 20225, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.470899 }, { "epoch": 0.8667152221412965, "grad_norm": 2.5169596672058105, "learning_rate": 9.276742278183558e-05, "loss": 0.32741289138793944, "memory(GiB)": 61.91, "step": 20230, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.470889 }, { "epoch": 0.8669294374705454, "grad_norm": 5.539857864379883, "learning_rate": 9.276393601463844e-05, "loss": 0.421093225479126, "memory(GiB)": 61.91, "step": 20235, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.470893 }, { "epoch": 0.8671436527997943, "grad_norm": 3.5388846397399902, "learning_rate": 9.276044847273249e-05, "loss": 0.4861351490020752, "memory(GiB)": 61.91, "step": 20240, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.470901 }, { "epoch": 0.8673578681290434, "grad_norm": 1.8168584108352661, "learning_rate": 9.27569601561809e-05, "loss": 0.29187147617340087, "memory(GiB)": 61.91, "step": 20245, "token_acc": 0.9437229437229437, "train_speed(iter/s)": 1.470903 }, { "epoch": 0.8675720834582923, "grad_norm": 3.5806045532226562, "learning_rate": 9.275347106504689e-05, "loss": 0.36395626068115233, "memory(GiB)": 61.91, "step": 20250, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.47092 }, { "epoch": 0.8677862987875412, "grad_norm": 4.734711170196533, "learning_rate": 9.274998119939362e-05, "loss": 0.3267072200775146, "memory(GiB)": 61.91, "step": 20255, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.470916 }, { "epoch": 0.8680005141167902, "grad_norm": 1.8125243186950684, "learning_rate": 9.274649055928434e-05, "loss": 0.17592849731445312, "memory(GiB)": 61.91, "step": 20260, "token_acc": 0.9685314685314685, "train_speed(iter/s)": 1.470923 }, { "epoch": 0.8682147294460392, "grad_norm": 2.4889023303985596, "learning_rate": 9.274299914478228e-05, "loss": 0.47710418701171875, "memory(GiB)": 61.91, "step": 20265, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.4709 }, { "epoch": 0.8684289447752881, "grad_norm": 5.094547271728516, "learning_rate": 9.273950695595071e-05, "loss": 0.4860203266143799, "memory(GiB)": 61.91, "step": 20270, "token_acc": 0.8679867986798679, "train_speed(iter/s)": 1.470887 }, { "epoch": 0.8686431601045371, "grad_norm": 2.9492194652557373, "learning_rate": 9.273601399285287e-05, "loss": 0.33748438358306887, "memory(GiB)": 61.91, "step": 20275, "token_acc": 0.9423868312757202, "train_speed(iter/s)": 1.470868 }, { "epoch": 0.868857375433786, "grad_norm": 2.8262240886688232, "learning_rate": 9.273252025555205e-05, "loss": 0.5211102962493896, "memory(GiB)": 61.91, "step": 20280, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.470862 }, { "epoch": 0.869071590763035, "grad_norm": 2.1415047645568848, "learning_rate": 9.272902574411153e-05, "loss": 0.4004986763000488, "memory(GiB)": 61.91, "step": 20285, "token_acc": 0.910394265232975, "train_speed(iter/s)": 1.470871 }, { "epoch": 0.869285806092284, "grad_norm": 3.3779759407043457, "learning_rate": 9.272553045859464e-05, "loss": 0.46398348808288575, "memory(GiB)": 61.91, "step": 20290, "token_acc": 0.8839590443686007, "train_speed(iter/s)": 1.470906 }, { "epoch": 0.8695000214215329, "grad_norm": 3.2674472332000732, "learning_rate": 9.272203439906469e-05, "loss": 0.27725958824157715, "memory(GiB)": 61.91, "step": 20295, "token_acc": 0.9402390438247012, "train_speed(iter/s)": 1.470902 }, { "epoch": 0.8697142367507819, "grad_norm": 0.27419528365135193, "learning_rate": 9.271853756558497e-05, "loss": 0.1272291898727417, "memory(GiB)": 61.91, "step": 20300, "token_acc": 0.9581749049429658, "train_speed(iter/s)": 1.470914 }, { "epoch": 0.8699284520800309, "grad_norm": 1.919814944267273, "learning_rate": 9.271503995821891e-05, "loss": 0.28002777099609377, "memory(GiB)": 61.91, "step": 20305, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.470909 }, { "epoch": 0.8701426674092798, "grad_norm": 2.8603670597076416, "learning_rate": 9.271154157702982e-05, "loss": 0.3877793073654175, "memory(GiB)": 61.91, "step": 20310, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.470926 }, { "epoch": 0.8703568827385287, "grad_norm": 4.851675987243652, "learning_rate": 9.270804242208109e-05, "loss": 0.4519011974334717, "memory(GiB)": 61.91, "step": 20315, "token_acc": 0.901685393258427, "train_speed(iter/s)": 1.471004 }, { "epoch": 0.8705710980677778, "grad_norm": 2.747835636138916, "learning_rate": 9.27045424934361e-05, "loss": 0.6600792407989502, "memory(GiB)": 61.91, "step": 20320, "token_acc": 0.8458904109589042, "train_speed(iter/s)": 1.470991 }, { "epoch": 0.8707853133970267, "grad_norm": 0.33447086811065674, "learning_rate": 9.270104179115825e-05, "loss": 0.2837267637252808, "memory(GiB)": 61.91, "step": 20325, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.47099 }, { "epoch": 0.8709995287262756, "grad_norm": 3.061060905456543, "learning_rate": 9.2697540315311e-05, "loss": 0.3605178356170654, "memory(GiB)": 61.91, "step": 20330, "token_acc": 0.9003115264797508, "train_speed(iter/s)": 1.470962 }, { "epoch": 0.8712137440555247, "grad_norm": 2.6664257049560547, "learning_rate": 9.269403806595775e-05, "loss": 0.42342686653137207, "memory(GiB)": 61.91, "step": 20335, "token_acc": 0.9067524115755627, "train_speed(iter/s)": 1.471021 }, { "epoch": 0.8714279593847736, "grad_norm": 3.8074936866760254, "learning_rate": 9.269053504316194e-05, "loss": 0.5013735294342041, "memory(GiB)": 61.91, "step": 20340, "token_acc": 0.8980392156862745, "train_speed(iter/s)": 1.471048 }, { "epoch": 0.8716421747140225, "grad_norm": 0.629012405872345, "learning_rate": 9.268703124698704e-05, "loss": 0.39875454902648927, "memory(GiB)": 61.91, "step": 20345, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.471029 }, { "epoch": 0.8718563900432715, "grad_norm": 2.0433337688446045, "learning_rate": 9.268352667749653e-05, "loss": 0.3396777153015137, "memory(GiB)": 61.91, "step": 20350, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.471016 }, { "epoch": 0.8720706053725205, "grad_norm": 4.542453765869141, "learning_rate": 9.268002133475388e-05, "loss": 0.37538139820098876, "memory(GiB)": 61.91, "step": 20355, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.47102 }, { "epoch": 0.8722848207017694, "grad_norm": 5.2698235511779785, "learning_rate": 9.267651521882263e-05, "loss": 0.5297595500946045, "memory(GiB)": 61.91, "step": 20360, "token_acc": 0.8566433566433567, "train_speed(iter/s)": 1.471047 }, { "epoch": 0.8724990360310184, "grad_norm": 1.0837794542312622, "learning_rate": 9.267300832976626e-05, "loss": 0.2728140830993652, "memory(GiB)": 61.91, "step": 20365, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.471029 }, { "epoch": 0.8727132513602673, "grad_norm": 3.025268793106079, "learning_rate": 9.266950066764832e-05, "loss": 0.5232598781585693, "memory(GiB)": 61.91, "step": 20370, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.471016 }, { "epoch": 0.8729274666895163, "grad_norm": 3.443796157836914, "learning_rate": 9.266599223253234e-05, "loss": 0.38428282737731934, "memory(GiB)": 61.91, "step": 20375, "token_acc": 0.9271255060728745, "train_speed(iter/s)": 1.471029 }, { "epoch": 0.8731416820187653, "grad_norm": 6.164388656616211, "learning_rate": 9.266248302448188e-05, "loss": 0.5353055000305176, "memory(GiB)": 61.91, "step": 20380, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.471098 }, { "epoch": 0.8733558973480142, "grad_norm": 2.455932378768921, "learning_rate": 9.265897304356055e-05, "loss": 0.5279929161071777, "memory(GiB)": 61.91, "step": 20385, "token_acc": 0.8932384341637011, "train_speed(iter/s)": 1.471086 }, { "epoch": 0.8735701126772631, "grad_norm": 2.3730509281158447, "learning_rate": 9.265546228983189e-05, "loss": 0.3904967546463013, "memory(GiB)": 61.91, "step": 20390, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.47107 }, { "epoch": 0.8737843280065122, "grad_norm": 2.647158145904541, "learning_rate": 9.26519507633595e-05, "loss": 0.2832775354385376, "memory(GiB)": 61.91, "step": 20395, "token_acc": 0.9397590361445783, "train_speed(iter/s)": 1.471059 }, { "epoch": 0.8739985433357611, "grad_norm": 4.059156894683838, "learning_rate": 9.264843846420702e-05, "loss": 0.3772773027420044, "memory(GiB)": 61.91, "step": 20400, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.471045 }, { "epoch": 0.87421275866501, "grad_norm": 5.386110305786133, "learning_rate": 9.264492539243808e-05, "loss": 0.3716228723526001, "memory(GiB)": 61.91, "step": 20405, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.471043 }, { "epoch": 0.8744269739942591, "grad_norm": 0.08316881209611893, "learning_rate": 9.26414115481163e-05, "loss": 0.20720417499542237, "memory(GiB)": 61.91, "step": 20410, "token_acc": 0.9402390438247012, "train_speed(iter/s)": 1.471077 }, { "epoch": 0.874641189323508, "grad_norm": 2.0852210521698, "learning_rate": 9.263789693130535e-05, "loss": 0.23512818813323974, "memory(GiB)": 61.91, "step": 20415, "token_acc": 0.9498432601880877, "train_speed(iter/s)": 1.471067 }, { "epoch": 0.8748554046527569, "grad_norm": 5.843173027038574, "learning_rate": 9.263438154206892e-05, "loss": 0.30675501823425294, "memory(GiB)": 61.91, "step": 20420, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.471063 }, { "epoch": 0.8750696199820059, "grad_norm": 4.7368316650390625, "learning_rate": 9.263086538047064e-05, "loss": 0.5433887958526611, "memory(GiB)": 61.91, "step": 20425, "token_acc": 0.8905660377358491, "train_speed(iter/s)": 1.471044 }, { "epoch": 0.8752838353112549, "grad_norm": 1.196048617362976, "learning_rate": 9.262734844657425e-05, "loss": 0.36014442443847655, "memory(GiB)": 61.91, "step": 20430, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.471028 }, { "epoch": 0.8754980506405038, "grad_norm": 2.621941328048706, "learning_rate": 9.262383074044347e-05, "loss": 0.3916118383407593, "memory(GiB)": 61.91, "step": 20435, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.471029 }, { "epoch": 0.8757122659697528, "grad_norm": 6.432663917541504, "learning_rate": 9.262031226214201e-05, "loss": 0.7440526008605957, "memory(GiB)": 61.91, "step": 20440, "token_acc": 0.867109634551495, "train_speed(iter/s)": 1.471022 }, { "epoch": 0.8759264812990017, "grad_norm": 4.009224891662598, "learning_rate": 9.261679301173361e-05, "loss": 0.6484699249267578, "memory(GiB)": 61.91, "step": 20445, "token_acc": 0.8848684210526315, "train_speed(iter/s)": 1.47103 }, { "epoch": 0.8761406966282507, "grad_norm": 5.817619323730469, "learning_rate": 9.261327298928203e-05, "loss": 0.5252953052520752, "memory(GiB)": 61.91, "step": 20450, "token_acc": 0.8835341365461847, "train_speed(iter/s)": 1.471031 }, { "epoch": 0.8763549119574997, "grad_norm": 3.3331449031829834, "learning_rate": 9.260975219485104e-05, "loss": 0.985097599029541, "memory(GiB)": 61.91, "step": 20455, "token_acc": 0.811377245508982, "train_speed(iter/s)": 1.471129 }, { "epoch": 0.8765691272867486, "grad_norm": 1.1041663885116577, "learning_rate": 9.26062306285044e-05, "loss": 0.23234739303588867, "memory(GiB)": 61.91, "step": 20460, "token_acc": 0.9471698113207547, "train_speed(iter/s)": 1.471143 }, { "epoch": 0.8767833426159976, "grad_norm": 1.2077059745788574, "learning_rate": 9.260270829030594e-05, "loss": 0.45517644882202146, "memory(GiB)": 61.91, "step": 20465, "token_acc": 0.9125, "train_speed(iter/s)": 1.471127 }, { "epoch": 0.8769975579452466, "grad_norm": 5.726842403411865, "learning_rate": 9.259918518031944e-05, "loss": 0.4931472301483154, "memory(GiB)": 61.91, "step": 20470, "token_acc": 0.8910505836575876, "train_speed(iter/s)": 1.471116 }, { "epoch": 0.8772117732744955, "grad_norm": 4.1705732345581055, "learning_rate": 9.259566129860874e-05, "loss": 0.30225400924682616, "memory(GiB)": 61.91, "step": 20475, "token_acc": 0.9383116883116883, "train_speed(iter/s)": 1.471107 }, { "epoch": 0.8774259886037444, "grad_norm": 10.478804588317871, "learning_rate": 9.25921366452377e-05, "loss": 0.44098901748657227, "memory(GiB)": 61.91, "step": 20480, "token_acc": 0.9045801526717557, "train_speed(iter/s)": 1.471143 }, { "epoch": 0.8776402039329935, "grad_norm": 1.982035517692566, "learning_rate": 9.258861122027013e-05, "loss": 0.3743525743484497, "memory(GiB)": 61.91, "step": 20485, "token_acc": 0.9186046511627907, "train_speed(iter/s)": 1.471183 }, { "epoch": 0.8778544192622424, "grad_norm": 3.042750835418701, "learning_rate": 9.258508502376992e-05, "loss": 0.36660161018371584, "memory(GiB)": 61.91, "step": 20490, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.471189 }, { "epoch": 0.8780686345914913, "grad_norm": 2.342512607574463, "learning_rate": 9.258155805580095e-05, "loss": 0.3879528522491455, "memory(GiB)": 61.91, "step": 20495, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.471227 }, { "epoch": 0.8782828499207403, "grad_norm": 4.236050128936768, "learning_rate": 9.257803031642711e-05, "loss": 0.3503885269165039, "memory(GiB)": 61.91, "step": 20500, "token_acc": 0.9216300940438872, "train_speed(iter/s)": 1.47121 }, { "epoch": 0.8782828499207403, "eval_loss": 2.654681921005249, "eval_runtime": 13.0954, "eval_samples_per_second": 7.636, "eval_steps_per_second": 7.636, "eval_token_acc": 0.4090909090909091, "step": 20500 }, { "epoch": 0.8784970652499893, "grad_norm": 5.393502235412598, "learning_rate": 9.257450180571232e-05, "loss": 0.5845614910125733, "memory(GiB)": 61.91, "step": 20505, "token_acc": 0.5372993389990557, "train_speed(iter/s)": 1.469732 }, { "epoch": 0.8787112805792382, "grad_norm": 2.5605697631835938, "learning_rate": 9.257097252372047e-05, "loss": 0.33208436965942384, "memory(GiB)": 61.91, "step": 20510, "token_acc": 0.9304635761589404, "train_speed(iter/s)": 1.469746 }, { "epoch": 0.8789254959084872, "grad_norm": 4.5631422996521, "learning_rate": 9.256744247051553e-05, "loss": 0.4456012725830078, "memory(GiB)": 61.91, "step": 20515, "token_acc": 0.8870056497175142, "train_speed(iter/s)": 1.469791 }, { "epoch": 0.8791397112377362, "grad_norm": 2.600782871246338, "learning_rate": 9.256391164616144e-05, "loss": 0.4852243423461914, "memory(GiB)": 61.91, "step": 20520, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.469791 }, { "epoch": 0.8793539265669851, "grad_norm": 2.8911192417144775, "learning_rate": 9.256038005072216e-05, "loss": 0.6145119190216064, "memory(GiB)": 61.91, "step": 20525, "token_acc": 0.8925925925925926, "train_speed(iter/s)": 1.469777 }, { "epoch": 0.8795681418962341, "grad_norm": 2.902102470397949, "learning_rate": 9.255684768426168e-05, "loss": 0.31998457908630373, "memory(GiB)": 61.91, "step": 20530, "token_acc": 0.9183006535947712, "train_speed(iter/s)": 1.469776 }, { "epoch": 0.879782357225483, "grad_norm": 6.201425552368164, "learning_rate": 9.255331454684395e-05, "loss": 0.48004980087280275, "memory(GiB)": 61.91, "step": 20535, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.469783 }, { "epoch": 0.8799965725547321, "grad_norm": 3.5748789310455322, "learning_rate": 9.254978063853303e-05, "loss": 0.31202819347381594, "memory(GiB)": 61.91, "step": 20540, "token_acc": 0.9368029739776952, "train_speed(iter/s)": 1.469774 }, { "epoch": 0.880210787883981, "grad_norm": 3.452876091003418, "learning_rate": 9.254624595939293e-05, "loss": 0.2782243251800537, "memory(GiB)": 61.91, "step": 20545, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.469762 }, { "epoch": 0.8804250032132299, "grad_norm": 4.855597019195557, "learning_rate": 9.254271050948767e-05, "loss": 0.3939187049865723, "memory(GiB)": 61.91, "step": 20550, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.469741 }, { "epoch": 0.880639218542479, "grad_norm": 2.6236846446990967, "learning_rate": 9.253917428888129e-05, "loss": 0.5342060089111328, "memory(GiB)": 61.91, "step": 20555, "token_acc": 0.8905325443786982, "train_speed(iter/s)": 1.469752 }, { "epoch": 0.8808534338717279, "grad_norm": 3.9176530838012695, "learning_rate": 9.253563729763786e-05, "loss": 0.7897815704345703, "memory(GiB)": 61.91, "step": 20560, "token_acc": 0.8191489361702128, "train_speed(iter/s)": 1.469847 }, { "epoch": 0.8810676492009768, "grad_norm": 3.4497034549713135, "learning_rate": 9.253209953582147e-05, "loss": 0.3669548511505127, "memory(GiB)": 61.91, "step": 20565, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.469861 }, { "epoch": 0.8812818645302258, "grad_norm": 2.8337817192077637, "learning_rate": 9.25285610034962e-05, "loss": 0.5095669746398925, "memory(GiB)": 61.91, "step": 20570, "token_acc": 0.8961937716262975, "train_speed(iter/s)": 1.469847 }, { "epoch": 0.8814960798594748, "grad_norm": 0.6971103549003601, "learning_rate": 9.252502170072615e-05, "loss": 0.4159690380096436, "memory(GiB)": 61.91, "step": 20575, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.469867 }, { "epoch": 0.8817102951887237, "grad_norm": 0.8404387831687927, "learning_rate": 9.252148162757542e-05, "loss": 0.35258615016937256, "memory(GiB)": 61.91, "step": 20580, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.469887 }, { "epoch": 0.8819245105179727, "grad_norm": 1.8977062702178955, "learning_rate": 9.251794078410818e-05, "loss": 0.26389617919921876, "memory(GiB)": 61.91, "step": 20585, "token_acc": 0.9281914893617021, "train_speed(iter/s)": 1.46988 }, { "epoch": 0.8821387258472216, "grad_norm": 2.8386340141296387, "learning_rate": 9.251439917038856e-05, "loss": 0.4016404151916504, "memory(GiB)": 61.91, "step": 20590, "token_acc": 0.9300411522633745, "train_speed(iter/s)": 1.469884 }, { "epoch": 0.8823529411764706, "grad_norm": 4.149711608886719, "learning_rate": 9.251085678648072e-05, "loss": 0.44855446815490724, "memory(GiB)": 61.91, "step": 20595, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.469877 }, { "epoch": 0.8825671565057196, "grad_norm": 4.27059268951416, "learning_rate": 9.250731363244882e-05, "loss": 0.57105393409729, "memory(GiB)": 61.91, "step": 20600, "token_acc": 0.8825503355704698, "train_speed(iter/s)": 1.469863 }, { "epoch": 0.8827813718349685, "grad_norm": 0.7378469109535217, "learning_rate": 9.250376970835706e-05, "loss": 0.4087355136871338, "memory(GiB)": 61.91, "step": 20605, "token_acc": 0.913946587537092, "train_speed(iter/s)": 1.469854 }, { "epoch": 0.8829955871642174, "grad_norm": 2.172455310821533, "learning_rate": 9.250022501426965e-05, "loss": 0.30805253982543945, "memory(GiB)": 61.91, "step": 20610, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.469892 }, { "epoch": 0.8832098024934665, "grad_norm": 3.545656681060791, "learning_rate": 9.249667955025079e-05, "loss": 0.45917425155639646, "memory(GiB)": 61.91, "step": 20615, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.469889 }, { "epoch": 0.8834240178227154, "grad_norm": 2.050723075866699, "learning_rate": 9.24931333163647e-05, "loss": 0.26000337600708007, "memory(GiB)": 61.91, "step": 20620, "token_acc": 0.9519774011299436, "train_speed(iter/s)": 1.46989 }, { "epoch": 0.8836382331519643, "grad_norm": 3.761932611465454, "learning_rate": 9.248958631267566e-05, "loss": 0.29191017150878906, "memory(GiB)": 61.91, "step": 20625, "token_acc": 0.9446254071661238, "train_speed(iter/s)": 1.469893 }, { "epoch": 0.8838524484812134, "grad_norm": 3.858949899673462, "learning_rate": 9.248603853924789e-05, "loss": 0.3816622257232666, "memory(GiB)": 61.91, "step": 20630, "token_acc": 0.9225806451612903, "train_speed(iter/s)": 1.469892 }, { "epoch": 0.8840666638104623, "grad_norm": 2.3824822902679443, "learning_rate": 9.248248999614568e-05, "loss": 0.4334630012512207, "memory(GiB)": 61.91, "step": 20635, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.46993 }, { "epoch": 0.8842808791397112, "grad_norm": 3.9659693241119385, "learning_rate": 9.247894068343332e-05, "loss": 0.3457240104675293, "memory(GiB)": 61.91, "step": 20640, "token_acc": 0.9163179916317992, "train_speed(iter/s)": 1.469917 }, { "epoch": 0.8844950944689602, "grad_norm": 3.462836742401123, "learning_rate": 9.24753906011751e-05, "loss": 0.3277305603027344, "memory(GiB)": 61.91, "step": 20645, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.469965 }, { "epoch": 0.8847093097982092, "grad_norm": 3.7420461177825928, "learning_rate": 9.247183974943532e-05, "loss": 0.3756612777709961, "memory(GiB)": 61.91, "step": 20650, "token_acc": 0.9208860759493671, "train_speed(iter/s)": 1.469968 }, { "epoch": 0.8849235251274581, "grad_norm": 1.8029277324676514, "learning_rate": 9.246828812827834e-05, "loss": 0.3801274299621582, "memory(GiB)": 61.91, "step": 20655, "token_acc": 0.9120879120879121, "train_speed(iter/s)": 1.469958 }, { "epoch": 0.8851377404567071, "grad_norm": 2.0436856746673584, "learning_rate": 9.246473573776845e-05, "loss": 0.21556942462921141, "memory(GiB)": 61.91, "step": 20660, "token_acc": 0.950381679389313, "train_speed(iter/s)": 1.469934 }, { "epoch": 0.885351955785956, "grad_norm": 3.6523873805999756, "learning_rate": 9.246118257797007e-05, "loss": 0.4493208885192871, "memory(GiB)": 61.91, "step": 20665, "token_acc": 0.8794788273615635, "train_speed(iter/s)": 1.46993 }, { "epoch": 0.885566171115205, "grad_norm": 6.215827465057373, "learning_rate": 9.245762864894754e-05, "loss": 0.23746700286865235, "memory(GiB)": 61.91, "step": 20670, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.46995 }, { "epoch": 0.885780386444454, "grad_norm": 4.692248344421387, "learning_rate": 9.245407395076522e-05, "loss": 0.709587287902832, "memory(GiB)": 61.91, "step": 20675, "token_acc": 0.8364312267657993, "train_speed(iter/s)": 1.469964 }, { "epoch": 0.8859946017737029, "grad_norm": 0.1074208989739418, "learning_rate": 9.245051848348753e-05, "loss": 0.3345240592956543, "memory(GiB)": 61.91, "step": 20680, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.46996 }, { "epoch": 0.8862088171029519, "grad_norm": 4.640208721160889, "learning_rate": 9.244696224717889e-05, "loss": 0.2177915096282959, "memory(GiB)": 61.91, "step": 20685, "token_acc": 0.9560439560439561, "train_speed(iter/s)": 1.470003 }, { "epoch": 0.8864230324322009, "grad_norm": 1.579383373260498, "learning_rate": 9.244340524190368e-05, "loss": 0.6324481010437012, "memory(GiB)": 61.91, "step": 20690, "token_acc": 0.8830769230769231, "train_speed(iter/s)": 1.470021 }, { "epoch": 0.8866372477614498, "grad_norm": 1.55363929271698, "learning_rate": 9.243984746772638e-05, "loss": 0.5042998313903808, "memory(GiB)": 61.91, "step": 20695, "token_acc": 0.9019073569482289, "train_speed(iter/s)": 1.470008 }, { "epoch": 0.8868514630906987, "grad_norm": 0.22636151313781738, "learning_rate": 9.243628892471145e-05, "loss": 0.3779905796051025, "memory(GiB)": 61.91, "step": 20700, "token_acc": 0.9202898550724637, "train_speed(iter/s)": 1.469991 }, { "epoch": 0.8870656784199478, "grad_norm": 1.284071445465088, "learning_rate": 9.243272961292331e-05, "loss": 0.12181166410446168, "memory(GiB)": 61.91, "step": 20705, "token_acc": 0.9726027397260274, "train_speed(iter/s)": 1.469978 }, { "epoch": 0.8872798937491967, "grad_norm": 4.952624797821045, "learning_rate": 9.242916953242649e-05, "loss": 0.5259307861328125, "memory(GiB)": 61.91, "step": 20710, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.469961 }, { "epoch": 0.8874941090784456, "grad_norm": 2.1396164894104004, "learning_rate": 9.242560868328545e-05, "loss": 0.45645833015441895, "memory(GiB)": 61.91, "step": 20715, "token_acc": 0.8938906752411575, "train_speed(iter/s)": 1.469959 }, { "epoch": 0.8877083244076946, "grad_norm": 2.696761131286621, "learning_rate": 9.242204706556472e-05, "loss": 0.37336955070495603, "memory(GiB)": 61.91, "step": 20720, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.469926 }, { "epoch": 0.8879225397369436, "grad_norm": 2.4662747383117676, "learning_rate": 9.241848467932881e-05, "loss": 0.6372010707855225, "memory(GiB)": 61.91, "step": 20725, "token_acc": 0.886435331230284, "train_speed(iter/s)": 1.469896 }, { "epoch": 0.8881367550661925, "grad_norm": 2.2339935302734375, "learning_rate": 9.241492152464225e-05, "loss": 0.5099318504333497, "memory(GiB)": 61.91, "step": 20730, "token_acc": 0.8901515151515151, "train_speed(iter/s)": 1.469876 }, { "epoch": 0.8883509703954415, "grad_norm": 1.2749433517456055, "learning_rate": 9.24113576015696e-05, "loss": 0.35081846714019777, "memory(GiB)": 61.91, "step": 20735, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.469866 }, { "epoch": 0.8885651857246905, "grad_norm": 1.0432847738265991, "learning_rate": 9.240779291017542e-05, "loss": 0.21951467990875245, "memory(GiB)": 61.91, "step": 20740, "token_acc": 0.966542750929368, "train_speed(iter/s)": 1.469889 }, { "epoch": 0.8887794010539394, "grad_norm": 2.4450995922088623, "learning_rate": 9.240422745052431e-05, "loss": 0.449696683883667, "memory(GiB)": 61.91, "step": 20745, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.469909 }, { "epoch": 0.8889936163831884, "grad_norm": 6.32750940322876, "learning_rate": 9.240066122268081e-05, "loss": 0.3290694713592529, "memory(GiB)": 61.91, "step": 20750, "token_acc": 0.9063545150501672, "train_speed(iter/s)": 1.469912 }, { "epoch": 0.8892078317124373, "grad_norm": 3.356121063232422, "learning_rate": 9.239709422670958e-05, "loss": 0.27981808185577395, "memory(GiB)": 61.91, "step": 20755, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.469904 }, { "epoch": 0.8894220470416863, "grad_norm": 1.514607548713684, "learning_rate": 9.23935264626752e-05, "loss": 0.16967442035675048, "memory(GiB)": 61.91, "step": 20760, "token_acc": 0.9574468085106383, "train_speed(iter/s)": 1.469927 }, { "epoch": 0.8896362623709353, "grad_norm": 3.0608346462249756, "learning_rate": 9.238995793064233e-05, "loss": 0.6112927913665771, "memory(GiB)": 61.91, "step": 20765, "token_acc": 0.8910256410256411, "train_speed(iter/s)": 1.469987 }, { "epoch": 0.8898504777001842, "grad_norm": 2.5715832710266113, "learning_rate": 9.238638863067561e-05, "loss": 0.7595057010650634, "memory(GiB)": 61.91, "step": 20770, "token_acc": 0.840782122905028, "train_speed(iter/s)": 1.470015 }, { "epoch": 0.8900646930294331, "grad_norm": 1.7982739210128784, "learning_rate": 9.23828185628397e-05, "loss": 0.24884500503540039, "memory(GiB)": 61.91, "step": 20775, "token_acc": 0.959866220735786, "train_speed(iter/s)": 1.470021 }, { "epoch": 0.8902789083586822, "grad_norm": 1.380895972251892, "learning_rate": 9.237924772719927e-05, "loss": 0.5679807186126709, "memory(GiB)": 61.91, "step": 20780, "token_acc": 0.8849315068493151, "train_speed(iter/s)": 1.470006 }, { "epoch": 0.8904931236879311, "grad_norm": 3.490095853805542, "learning_rate": 9.237567612381901e-05, "loss": 0.5837003707885742, "memory(GiB)": 61.91, "step": 20785, "token_acc": 0.8766233766233766, "train_speed(iter/s)": 1.470049 }, { "epoch": 0.89070733901718, "grad_norm": 3.419269323348999, "learning_rate": 9.237210375276361e-05, "loss": 0.21716742515563964, "memory(GiB)": 61.91, "step": 20790, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.470031 }, { "epoch": 0.8909215543464291, "grad_norm": 6.25709867477417, "learning_rate": 9.236853061409783e-05, "loss": 0.20150697231292725, "memory(GiB)": 61.91, "step": 20795, "token_acc": 0.9504132231404959, "train_speed(iter/s)": 1.470064 }, { "epoch": 0.891135769675678, "grad_norm": 3.0650343894958496, "learning_rate": 9.236495670788636e-05, "loss": 0.5902454376220703, "memory(GiB)": 61.91, "step": 20800, "token_acc": 0.875, "train_speed(iter/s)": 1.470093 }, { "epoch": 0.8913499850049269, "grad_norm": 1.9880532026290894, "learning_rate": 9.236138203419396e-05, "loss": 0.43593473434448243, "memory(GiB)": 61.91, "step": 20805, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.470089 }, { "epoch": 0.8915642003341759, "grad_norm": 2.4334452152252197, "learning_rate": 9.23578065930854e-05, "loss": 0.34554603099823, "memory(GiB)": 61.91, "step": 20810, "token_acc": 0.9045936395759717, "train_speed(iter/s)": 1.470088 }, { "epoch": 0.8917784156634249, "grad_norm": 3.139000415802002, "learning_rate": 9.235423038462542e-05, "loss": 0.33843035697937013, "memory(GiB)": 61.91, "step": 20815, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.470104 }, { "epoch": 0.8919926309926738, "grad_norm": 4.979912281036377, "learning_rate": 9.235065340887882e-05, "loss": 0.4225703239440918, "memory(GiB)": 61.91, "step": 20820, "token_acc": 0.9139344262295082, "train_speed(iter/s)": 1.470122 }, { "epoch": 0.8922068463219228, "grad_norm": 5.9896626472473145, "learning_rate": 9.234707566591042e-05, "loss": 0.563661003112793, "memory(GiB)": 61.91, "step": 20825, "token_acc": 0.8759398496240601, "train_speed(iter/s)": 1.470116 }, { "epoch": 0.8924210616511717, "grad_norm": 3.5042896270751953, "learning_rate": 9.234349715578501e-05, "loss": 0.4656111717224121, "memory(GiB)": 61.91, "step": 20830, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.470105 }, { "epoch": 0.8926352769804207, "grad_norm": 5.914404392242432, "learning_rate": 9.233991787856743e-05, "loss": 0.5232911109924316, "memory(GiB)": 61.91, "step": 20835, "token_acc": 0.9202453987730062, "train_speed(iter/s)": 1.470129 }, { "epoch": 0.8928494923096697, "grad_norm": 2.1118712425231934, "learning_rate": 9.233633783432252e-05, "loss": 0.21433866024017334, "memory(GiB)": 61.91, "step": 20840, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.470118 }, { "epoch": 0.8930637076389186, "grad_norm": 3.029348850250244, "learning_rate": 9.233275702311514e-05, "loss": 0.48177351951599123, "memory(GiB)": 61.91, "step": 20845, "token_acc": 0.8953068592057761, "train_speed(iter/s)": 1.470132 }, { "epoch": 0.8932779229681675, "grad_norm": 3.4902336597442627, "learning_rate": 9.232917544501015e-05, "loss": 0.6631650447845459, "memory(GiB)": 61.91, "step": 20850, "token_acc": 0.8542274052478134, "train_speed(iter/s)": 1.470121 }, { "epoch": 0.8934921382974166, "grad_norm": 2.977479934692383, "learning_rate": 9.232559310007245e-05, "loss": 0.19341800212860108, "memory(GiB)": 61.91, "step": 20855, "token_acc": 0.9579288025889967, "train_speed(iter/s)": 1.470124 }, { "epoch": 0.8937063536266655, "grad_norm": 0.5782030820846558, "learning_rate": 9.23220099883669e-05, "loss": 0.3001437187194824, "memory(GiB)": 61.91, "step": 20860, "token_acc": 0.9457831325301205, "train_speed(iter/s)": 1.470121 }, { "epoch": 0.8939205689559145, "grad_norm": 3.369854211807251, "learning_rate": 9.231842610995847e-05, "loss": 0.4585575103759766, "memory(GiB)": 61.91, "step": 20865, "token_acc": 0.9115646258503401, "train_speed(iter/s)": 1.470118 }, { "epoch": 0.8941347842851635, "grad_norm": 5.6356658935546875, "learning_rate": 9.231484146491203e-05, "loss": 0.49897942543029783, "memory(GiB)": 61.91, "step": 20870, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.470112 }, { "epoch": 0.8943489996144124, "grad_norm": 3.2243142127990723, "learning_rate": 9.231125605329256e-05, "loss": 0.37024102210998533, "memory(GiB)": 61.91, "step": 20875, "token_acc": 0.9211356466876972, "train_speed(iter/s)": 1.470095 }, { "epoch": 0.8945632149436614, "grad_norm": 3.677781820297241, "learning_rate": 9.230766987516498e-05, "loss": 0.22300753593444825, "memory(GiB)": 61.91, "step": 20880, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.470129 }, { "epoch": 0.8947774302729103, "grad_norm": 13.043365478515625, "learning_rate": 9.230408293059427e-05, "loss": 0.49459133148193357, "memory(GiB)": 61.91, "step": 20885, "token_acc": 0.8954703832752613, "train_speed(iter/s)": 1.470136 }, { "epoch": 0.8949916456021593, "grad_norm": 1.5378832817077637, "learning_rate": 9.230049521964544e-05, "loss": 0.38436174392700195, "memory(GiB)": 61.91, "step": 20890, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.470128 }, { "epoch": 0.8952058609314083, "grad_norm": 5.490140914916992, "learning_rate": 9.229690674238343e-05, "loss": 0.4281723976135254, "memory(GiB)": 61.91, "step": 20895, "token_acc": 0.8997050147492626, "train_speed(iter/s)": 1.470145 }, { "epoch": 0.8954200762606572, "grad_norm": 3.94907283782959, "learning_rate": 9.22933174988733e-05, "loss": 0.2334757089614868, "memory(GiB)": 61.91, "step": 20900, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.470126 }, { "epoch": 0.8956342915899062, "grad_norm": 3.798694610595703, "learning_rate": 9.228972748918002e-05, "loss": 0.3139692544937134, "memory(GiB)": 61.91, "step": 20905, "token_acc": 0.9425287356321839, "train_speed(iter/s)": 1.470161 }, { "epoch": 0.8958485069191552, "grad_norm": 2.0100677013397217, "learning_rate": 9.228613671336867e-05, "loss": 0.19742195606231688, "memory(GiB)": 61.91, "step": 20910, "token_acc": 0.95578231292517, "train_speed(iter/s)": 1.470168 }, { "epoch": 0.8960627222484041, "grad_norm": 1.268965482711792, "learning_rate": 9.228254517150429e-05, "loss": 0.37749547958374025, "memory(GiB)": 61.91, "step": 20915, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.470172 }, { "epoch": 0.896276937577653, "grad_norm": 5.493825912475586, "learning_rate": 9.227895286365194e-05, "loss": 0.2918544292449951, "memory(GiB)": 61.91, "step": 20920, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.470178 }, { "epoch": 0.8964911529069021, "grad_norm": 3.4826717376708984, "learning_rate": 9.227535978987669e-05, "loss": 0.2407521963119507, "memory(GiB)": 61.91, "step": 20925, "token_acc": 0.9380804953560371, "train_speed(iter/s)": 1.470217 }, { "epoch": 0.896705368236151, "grad_norm": 2.1452136039733887, "learning_rate": 9.227176595024364e-05, "loss": 0.24792542457580566, "memory(GiB)": 61.91, "step": 20930, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.470209 }, { "epoch": 0.8969195835653999, "grad_norm": 9.311711311340332, "learning_rate": 9.22681713448179e-05, "loss": 0.4112142562866211, "memory(GiB)": 61.91, "step": 20935, "token_acc": 0.902127659574468, "train_speed(iter/s)": 1.470189 }, { "epoch": 0.897133798894649, "grad_norm": 0.9551988244056702, "learning_rate": 9.226457597366459e-05, "loss": 0.5080307006835938, "memory(GiB)": 61.91, "step": 20940, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.470192 }, { "epoch": 0.8973480142238979, "grad_norm": 4.892697334289551, "learning_rate": 9.226097983684883e-05, "loss": 0.6797909736633301, "memory(GiB)": 61.91, "step": 20945, "token_acc": 0.8235294117647058, "train_speed(iter/s)": 1.470168 }, { "epoch": 0.8975622295531468, "grad_norm": 2.0195531845092773, "learning_rate": 9.225738293443578e-05, "loss": 0.3499652147293091, "memory(GiB)": 61.91, "step": 20950, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.470168 }, { "epoch": 0.8977764448823958, "grad_norm": 0.4601273536682129, "learning_rate": 9.22537852664906e-05, "loss": 0.4303931713104248, "memory(GiB)": 61.91, "step": 20955, "token_acc": 0.8875968992248062, "train_speed(iter/s)": 1.470172 }, { "epoch": 0.8979906602116448, "grad_norm": 8.655726432800293, "learning_rate": 9.225018683307846e-05, "loss": 0.7715602874755859, "memory(GiB)": 61.91, "step": 20960, "token_acc": 0.8339622641509434, "train_speed(iter/s)": 1.470152 }, { "epoch": 0.8982048755408937, "grad_norm": 2.676051616668701, "learning_rate": 9.224658763426454e-05, "loss": 0.39538040161132815, "memory(GiB)": 61.91, "step": 20965, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.470146 }, { "epoch": 0.8984190908701427, "grad_norm": 4.208296298980713, "learning_rate": 9.224298767011407e-05, "loss": 0.34768152236938477, "memory(GiB)": 61.91, "step": 20970, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.47016 }, { "epoch": 0.8986333061993916, "grad_norm": 1.6938073635101318, "learning_rate": 9.223938694069227e-05, "loss": 0.3236730098724365, "memory(GiB)": 61.91, "step": 20975, "token_acc": 0.9225806451612903, "train_speed(iter/s)": 1.47015 }, { "epoch": 0.8988475215286406, "grad_norm": 3.403122663497925, "learning_rate": 9.223578544606432e-05, "loss": 0.3810574054718018, "memory(GiB)": 61.91, "step": 20980, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.470143 }, { "epoch": 0.8990617368578896, "grad_norm": 9.691448211669922, "learning_rate": 9.223218318629551e-05, "loss": 0.3260735034942627, "memory(GiB)": 61.91, "step": 20985, "token_acc": 0.9409722222222222, "train_speed(iter/s)": 1.470123 }, { "epoch": 0.8992759521871385, "grad_norm": 2.5972323417663574, "learning_rate": 9.222858016145108e-05, "loss": 0.582308006286621, "memory(GiB)": 61.91, "step": 20990, "token_acc": 0.8905109489051095, "train_speed(iter/s)": 1.470121 }, { "epoch": 0.8994901675163874, "grad_norm": 2.213144302368164, "learning_rate": 9.22249763715963e-05, "loss": 0.45563273429870604, "memory(GiB)": 61.91, "step": 20995, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.470136 }, { "epoch": 0.8997043828456365, "grad_norm": 1.6468422412872314, "learning_rate": 9.222137181679648e-05, "loss": 0.4763401985168457, "memory(GiB)": 61.91, "step": 21000, "token_acc": 0.8872727272727273, "train_speed(iter/s)": 1.470136 }, { "epoch": 0.8997043828456365, "eval_loss": 2.4301607608795166, "eval_runtime": 13.3228, "eval_samples_per_second": 7.506, "eval_steps_per_second": 7.506, "eval_token_acc": 0.43898809523809523, "step": 21000 }, { "epoch": 0.8999185981748854, "grad_norm": 5.576371669769287, "learning_rate": 9.221776649711689e-05, "loss": 0.711960744857788, "memory(GiB)": 61.91, "step": 21005, "token_acc": 0.5627615062761506, "train_speed(iter/s)": 1.468677 }, { "epoch": 0.9001328135041343, "grad_norm": 1.782555103302002, "learning_rate": 9.221416041262285e-05, "loss": 0.3265284299850464, "memory(GiB)": 61.91, "step": 21010, "token_acc": 0.9418604651162791, "train_speed(iter/s)": 1.46871 }, { "epoch": 0.9003470288333834, "grad_norm": 3.746281623840332, "learning_rate": 9.22105535633797e-05, "loss": 0.4999196529388428, "memory(GiB)": 61.91, "step": 21015, "token_acc": 0.887719298245614, "train_speed(iter/s)": 1.468795 }, { "epoch": 0.9005612441626323, "grad_norm": 3.613847017288208, "learning_rate": 9.220694594945278e-05, "loss": 0.5048034191131592, "memory(GiB)": 61.91, "step": 21020, "token_acc": 0.8841463414634146, "train_speed(iter/s)": 1.468817 }, { "epoch": 0.9007754594918812, "grad_norm": 3.0073487758636475, "learning_rate": 9.220333757090745e-05, "loss": 0.2156372547149658, "memory(GiB)": 61.91, "step": 21025, "token_acc": 0.9621993127147767, "train_speed(iter/s)": 1.468829 }, { "epoch": 0.9009896748211302, "grad_norm": 3.043901205062866, "learning_rate": 9.219972842780907e-05, "loss": 0.24177703857421876, "memory(GiB)": 61.91, "step": 21030, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.468802 }, { "epoch": 0.9012038901503792, "grad_norm": 4.064138889312744, "learning_rate": 9.219611852022301e-05, "loss": 0.6333346366882324, "memory(GiB)": 61.91, "step": 21035, "token_acc": 0.8622754491017964, "train_speed(iter/s)": 1.468842 }, { "epoch": 0.9014181054796281, "grad_norm": 3.9614241123199463, "learning_rate": 9.219250784821467e-05, "loss": 0.5782087802886963, "memory(GiB)": 61.91, "step": 21040, "token_acc": 0.8538681948424068, "train_speed(iter/s)": 1.468914 }, { "epoch": 0.9016323208088771, "grad_norm": 2.234959363937378, "learning_rate": 9.218889641184949e-05, "loss": 0.3065903663635254, "memory(GiB)": 61.91, "step": 21045, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.468905 }, { "epoch": 0.901846536138126, "grad_norm": 5.204192161560059, "learning_rate": 9.218528421119287e-05, "loss": 0.398516321182251, "memory(GiB)": 61.91, "step": 21050, "token_acc": 0.9057971014492754, "train_speed(iter/s)": 1.468902 }, { "epoch": 0.902060751467375, "grad_norm": 1.9443687200546265, "learning_rate": 9.218167124631025e-05, "loss": 0.3728534698486328, "memory(GiB)": 61.91, "step": 21055, "token_acc": 0.9338842975206612, "train_speed(iter/s)": 1.468917 }, { "epoch": 0.902274966796624, "grad_norm": 3.7165210247039795, "learning_rate": 9.21780575172671e-05, "loss": 0.39874801635742185, "memory(GiB)": 61.91, "step": 21060, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.46889 }, { "epoch": 0.9024891821258729, "grad_norm": 3.9779069423675537, "learning_rate": 9.217444302412886e-05, "loss": 0.5322491645812988, "memory(GiB)": 61.91, "step": 21065, "token_acc": 0.8546099290780141, "train_speed(iter/s)": 1.468898 }, { "epoch": 0.9027033974551218, "grad_norm": 2.97896671295166, "learning_rate": 9.217082776696101e-05, "loss": 0.23320324420928956, "memory(GiB)": 61.91, "step": 21070, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.4689 }, { "epoch": 0.9029176127843709, "grad_norm": 4.310017108917236, "learning_rate": 9.216721174582907e-05, "loss": 0.8566571235656738, "memory(GiB)": 61.91, "step": 21075, "token_acc": 0.8323170731707317, "train_speed(iter/s)": 1.468894 }, { "epoch": 0.9031318281136198, "grad_norm": 1.8396245241165161, "learning_rate": 9.216359496079851e-05, "loss": 0.40108566284179686, "memory(GiB)": 61.91, "step": 21080, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.468877 }, { "epoch": 0.9033460434428687, "grad_norm": 4.817530155181885, "learning_rate": 9.215997741193491e-05, "loss": 0.5295067310333252, "memory(GiB)": 61.91, "step": 21085, "token_acc": 0.8754325259515571, "train_speed(iter/s)": 1.468874 }, { "epoch": 0.9035602587721178, "grad_norm": 2.4668655395507812, "learning_rate": 9.215635909930376e-05, "loss": 0.5407426357269287, "memory(GiB)": 61.91, "step": 21090, "token_acc": 0.8801369863013698, "train_speed(iter/s)": 1.468901 }, { "epoch": 0.9037744741013667, "grad_norm": 2.441227912902832, "learning_rate": 9.21527400229706e-05, "loss": 0.5110808372497558, "memory(GiB)": 61.91, "step": 21095, "token_acc": 0.9005681818181818, "train_speed(iter/s)": 1.468901 }, { "epoch": 0.9039886894306156, "grad_norm": 3.674844264984131, "learning_rate": 9.214912018300103e-05, "loss": 0.5080351829528809, "memory(GiB)": 61.91, "step": 21100, "token_acc": 0.8915254237288136, "train_speed(iter/s)": 1.468902 }, { "epoch": 0.9042029047598646, "grad_norm": 5.887238025665283, "learning_rate": 9.214549957946061e-05, "loss": 0.4701183319091797, "memory(GiB)": 61.91, "step": 21105, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.468904 }, { "epoch": 0.9044171200891136, "grad_norm": 5.291126251220703, "learning_rate": 9.214187821241492e-05, "loss": 0.46682538986206057, "memory(GiB)": 61.91, "step": 21110, "token_acc": 0.9243027888446215, "train_speed(iter/s)": 1.468908 }, { "epoch": 0.9046313354183625, "grad_norm": 2.9230358600616455, "learning_rate": 9.213825608192959e-05, "loss": 0.44749794006347654, "memory(GiB)": 61.91, "step": 21115, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.468902 }, { "epoch": 0.9048455507476115, "grad_norm": 3.0748167037963867, "learning_rate": 9.213463318807021e-05, "loss": 0.31810860633850097, "memory(GiB)": 61.91, "step": 21120, "token_acc": 0.9372937293729373, "train_speed(iter/s)": 1.468898 }, { "epoch": 0.9050597660768605, "grad_norm": 1.4669369459152222, "learning_rate": 9.213100953090241e-05, "loss": 0.26803786754608155, "memory(GiB)": 61.91, "step": 21125, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.468889 }, { "epoch": 0.9052739814061094, "grad_norm": 7.287233829498291, "learning_rate": 9.212738511049187e-05, "loss": 0.3265950679779053, "memory(GiB)": 61.91, "step": 21130, "token_acc": 0.9128787878787878, "train_speed(iter/s)": 1.468913 }, { "epoch": 0.9054881967353584, "grad_norm": 1.516561508178711, "learning_rate": 9.212375992690423e-05, "loss": 0.2428133487701416, "memory(GiB)": 61.91, "step": 21135, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.468912 }, { "epoch": 0.9057024120646073, "grad_norm": 0.18434983491897583, "learning_rate": 9.212013398020516e-05, "loss": 0.2659222841262817, "memory(GiB)": 61.91, "step": 21140, "token_acc": 0.947565543071161, "train_speed(iter/s)": 1.4689 }, { "epoch": 0.9059166273938563, "grad_norm": 4.327808856964111, "learning_rate": 9.211650727046033e-05, "loss": 0.7205610275268555, "memory(GiB)": 61.91, "step": 21145, "token_acc": 0.8446215139442231, "train_speed(iter/s)": 1.468881 }, { "epoch": 0.9061308427231053, "grad_norm": 1.7535452842712402, "learning_rate": 9.211287979773548e-05, "loss": 0.4063247203826904, "memory(GiB)": 61.91, "step": 21150, "token_acc": 0.90234375, "train_speed(iter/s)": 1.468888 }, { "epoch": 0.9063450580523542, "grad_norm": 2.796550989151001, "learning_rate": 9.210925156209632e-05, "loss": 0.4236761569976807, "memory(GiB)": 61.91, "step": 21155, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.468906 }, { "epoch": 0.9065592733816031, "grad_norm": 4.28469705581665, "learning_rate": 9.210562256360855e-05, "loss": 0.5242139816284179, "memory(GiB)": 61.91, "step": 21160, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.46891 }, { "epoch": 0.9067734887108522, "grad_norm": 3.1400394439697266, "learning_rate": 9.210199280233794e-05, "loss": 0.5179316520690918, "memory(GiB)": 61.91, "step": 21165, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.468969 }, { "epoch": 0.9069877040401011, "grad_norm": 3.3325257301330566, "learning_rate": 9.209836227835022e-05, "loss": 0.3620790481567383, "memory(GiB)": 61.91, "step": 21170, "token_acc": 0.9308943089430894, "train_speed(iter/s)": 1.468978 }, { "epoch": 0.90720191936935, "grad_norm": 1.3673030138015747, "learning_rate": 9.20947309917112e-05, "loss": 0.36810133457183836, "memory(GiB)": 61.91, "step": 21175, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.46898 }, { "epoch": 0.907416134698599, "grad_norm": 2.4307329654693604, "learning_rate": 9.209109894248662e-05, "loss": 0.24785809516906737, "memory(GiB)": 61.91, "step": 21180, "token_acc": 0.9504504504504504, "train_speed(iter/s)": 1.468972 }, { "epoch": 0.907630350027848, "grad_norm": 3.12860369682312, "learning_rate": 9.208746613074231e-05, "loss": 0.4231432914733887, "memory(GiB)": 61.91, "step": 21185, "token_acc": 0.9054545454545454, "train_speed(iter/s)": 1.468971 }, { "epoch": 0.9078445653570969, "grad_norm": 3.9119021892547607, "learning_rate": 9.208383255654406e-05, "loss": 0.4599141597747803, "memory(GiB)": 61.91, "step": 21190, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.468954 }, { "epoch": 0.9080587806863459, "grad_norm": 1.9521663188934326, "learning_rate": 9.20801982199577e-05, "loss": 0.4041721820831299, "memory(GiB)": 61.91, "step": 21195, "token_acc": 0.9201277955271565, "train_speed(iter/s)": 1.468945 }, { "epoch": 0.9082729960155949, "grad_norm": 2.9989755153656006, "learning_rate": 9.20765631210491e-05, "loss": 0.5964997291564942, "memory(GiB)": 61.91, "step": 21200, "token_acc": 0.8742138364779874, "train_speed(iter/s)": 1.468942 }, { "epoch": 0.9084872113448439, "grad_norm": 4.3423237800598145, "learning_rate": 9.207292725988405e-05, "loss": 0.43865213394165037, "memory(GiB)": 61.91, "step": 21205, "token_acc": 0.89568345323741, "train_speed(iter/s)": 1.468954 }, { "epoch": 0.9087014266740928, "grad_norm": 1.333600640296936, "learning_rate": 9.206929063652849e-05, "loss": 0.2001824140548706, "memory(GiB)": 61.91, "step": 21210, "token_acc": 0.934640522875817, "train_speed(iter/s)": 1.468953 }, { "epoch": 0.9089156420033417, "grad_norm": 4.251082420349121, "learning_rate": 9.206565325104826e-05, "loss": 0.48429279327392577, "memory(GiB)": 61.91, "step": 21215, "token_acc": 0.8884120171673819, "train_speed(iter/s)": 1.468951 }, { "epoch": 0.9091298573325908, "grad_norm": 6.138099670410156, "learning_rate": 9.206201510350925e-05, "loss": 0.42626237869262695, "memory(GiB)": 61.91, "step": 21220, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.468961 }, { "epoch": 0.9093440726618397, "grad_norm": 3.1755292415618896, "learning_rate": 9.205837619397738e-05, "loss": 0.5806573867797852, "memory(GiB)": 61.91, "step": 21225, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.468948 }, { "epoch": 0.9095582879910886, "grad_norm": 5.708640098571777, "learning_rate": 9.205473652251858e-05, "loss": 0.39664862155914304, "memory(GiB)": 61.91, "step": 21230, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.468959 }, { "epoch": 0.9097725033203377, "grad_norm": 2.797381639480591, "learning_rate": 9.205109608919878e-05, "loss": 0.5492545604705811, "memory(GiB)": 61.91, "step": 21235, "token_acc": 0.8917910447761194, "train_speed(iter/s)": 1.468937 }, { "epoch": 0.9099867186495866, "grad_norm": 3.780402660369873, "learning_rate": 9.20474548940839e-05, "loss": 0.32891438007354734, "memory(GiB)": 61.91, "step": 21240, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.468922 }, { "epoch": 0.9102009339788355, "grad_norm": 2.2886500358581543, "learning_rate": 9.204381293723996e-05, "loss": 0.5610685348510742, "memory(GiB)": 61.91, "step": 21245, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.468914 }, { "epoch": 0.9104151493080845, "grad_norm": 2.907201051712036, "learning_rate": 9.204017021873289e-05, "loss": 0.38972561359405516, "memory(GiB)": 61.91, "step": 21250, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.468912 }, { "epoch": 0.9106293646373335, "grad_norm": 2.669767379760742, "learning_rate": 9.20365267386287e-05, "loss": 0.22498085498809814, "memory(GiB)": 61.91, "step": 21255, "token_acc": 0.952054794520548, "train_speed(iter/s)": 1.468965 }, { "epoch": 0.9108435799665824, "grad_norm": 2.1855711936950684, "learning_rate": 9.203288249699341e-05, "loss": 0.30702483654022217, "memory(GiB)": 61.91, "step": 21260, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.468981 }, { "epoch": 0.9110577952958314, "grad_norm": 1.6837031841278076, "learning_rate": 9.202923749389302e-05, "loss": 0.5667271137237548, "memory(GiB)": 61.91, "step": 21265, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.468976 }, { "epoch": 0.9112720106250803, "grad_norm": 3.6862032413482666, "learning_rate": 9.202559172939355e-05, "loss": 0.5549705505371094, "memory(GiB)": 61.91, "step": 21270, "token_acc": 0.9045454545454545, "train_speed(iter/s)": 1.468984 }, { "epoch": 0.9114862259543293, "grad_norm": 3.7247517108917236, "learning_rate": 9.202194520356108e-05, "loss": 0.25127615928649905, "memory(GiB)": 61.91, "step": 21275, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.468998 }, { "epoch": 0.9117004412835783, "grad_norm": 2.1001250743865967, "learning_rate": 9.201829791646165e-05, "loss": 0.44761133193969727, "memory(GiB)": 61.91, "step": 21280, "token_acc": 0.912751677852349, "train_speed(iter/s)": 1.469042 }, { "epoch": 0.9119146566128272, "grad_norm": 7.292211532592773, "learning_rate": 9.201464986816132e-05, "loss": 0.40395331382751465, "memory(GiB)": 61.91, "step": 21285, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.469055 }, { "epoch": 0.9121288719420761, "grad_norm": 2.17932391166687, "learning_rate": 9.201100105872622e-05, "loss": 0.36630499362945557, "memory(GiB)": 61.91, "step": 21290, "token_acc": 0.9112627986348123, "train_speed(iter/s)": 1.469047 }, { "epoch": 0.9123430872713252, "grad_norm": 1.934751033782959, "learning_rate": 9.200735148822241e-05, "loss": 0.5192175388336182, "memory(GiB)": 61.91, "step": 21295, "token_acc": 0.8922413793103449, "train_speed(iter/s)": 1.469067 }, { "epoch": 0.9125573026005741, "grad_norm": 1.818652868270874, "learning_rate": 9.200370115671604e-05, "loss": 0.5283898830413818, "memory(GiB)": 61.91, "step": 21300, "token_acc": 0.8953068592057761, "train_speed(iter/s)": 1.469069 }, { "epoch": 0.912771517929823, "grad_norm": 7.61631965637207, "learning_rate": 9.20000500642732e-05, "loss": 0.4101680278778076, "memory(GiB)": 61.91, "step": 21305, "token_acc": 0.9455252918287937, "train_speed(iter/s)": 1.469047 }, { "epoch": 0.9129857332590721, "grad_norm": 5.630564212799072, "learning_rate": 9.199639821096006e-05, "loss": 0.4321906089782715, "memory(GiB)": 61.91, "step": 21310, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.469071 }, { "epoch": 0.913199948588321, "grad_norm": 5.04205846786499, "learning_rate": 9.199274559684277e-05, "loss": 0.37345871925354, "memory(GiB)": 61.91, "step": 21315, "token_acc": 0.9349593495934959, "train_speed(iter/s)": 1.469125 }, { "epoch": 0.9134141639175699, "grad_norm": 3.7404470443725586, "learning_rate": 9.198909222198751e-05, "loss": 0.28591485023498536, "memory(GiB)": 61.91, "step": 21320, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.469133 }, { "epoch": 0.913628379246819, "grad_norm": 4.263899326324463, "learning_rate": 9.198543808646045e-05, "loss": 0.4653331756591797, "memory(GiB)": 61.91, "step": 21325, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.46913 }, { "epoch": 0.9138425945760679, "grad_norm": 2.5885558128356934, "learning_rate": 9.19817831903278e-05, "loss": 0.22964158058166503, "memory(GiB)": 61.91, "step": 21330, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.46913 }, { "epoch": 0.9140568099053168, "grad_norm": 1.9484418630599976, "learning_rate": 9.197812753365575e-05, "loss": 0.2303131103515625, "memory(GiB)": 61.91, "step": 21335, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.469173 }, { "epoch": 0.9142710252345658, "grad_norm": 4.612428188323975, "learning_rate": 9.197447111651055e-05, "loss": 0.579350757598877, "memory(GiB)": 61.91, "step": 21340, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.469204 }, { "epoch": 0.9144852405638147, "grad_norm": 3.776289224624634, "learning_rate": 9.197081393895843e-05, "loss": 0.5877717018127442, "memory(GiB)": 61.91, "step": 21345, "token_acc": 0.8827361563517915, "train_speed(iter/s)": 1.469191 }, { "epoch": 0.9146994558930637, "grad_norm": 7.371421813964844, "learning_rate": 9.196715600106564e-05, "loss": 0.4563504695892334, "memory(GiB)": 61.91, "step": 21350, "token_acc": 0.8981723237597912, "train_speed(iter/s)": 1.469209 }, { "epoch": 0.9149136712223127, "grad_norm": 4.879343509674072, "learning_rate": 9.196349730289845e-05, "loss": 0.5559205055236817, "memory(GiB)": 61.91, "step": 21355, "token_acc": 0.8576388888888888, "train_speed(iter/s)": 1.469211 }, { "epoch": 0.9151278865515616, "grad_norm": 2.8758363723754883, "learning_rate": 9.195983784452315e-05, "loss": 0.4233862400054932, "memory(GiB)": 61.91, "step": 21360, "token_acc": 0.9105263157894737, "train_speed(iter/s)": 1.469184 }, { "epoch": 0.9153421018808106, "grad_norm": 3.2779462337493896, "learning_rate": 9.195617762600601e-05, "loss": 0.13059602975845336, "memory(GiB)": 61.91, "step": 21365, "token_acc": 0.9736070381231672, "train_speed(iter/s)": 1.469164 }, { "epoch": 0.9155563172100596, "grad_norm": 3.2037782669067383, "learning_rate": 9.195251664741337e-05, "loss": 0.1968465805053711, "memory(GiB)": 61.91, "step": 21370, "token_acc": 0.9572368421052632, "train_speed(iter/s)": 1.469195 }, { "epoch": 0.9157705325393085, "grad_norm": 0.19991984963417053, "learning_rate": 9.194885490881153e-05, "loss": 0.4513296604156494, "memory(GiB)": 61.91, "step": 21375, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.469172 }, { "epoch": 0.9159847478685574, "grad_norm": 3.6746139526367188, "learning_rate": 9.194519241026684e-05, "loss": 0.31049184799194335, "memory(GiB)": 61.91, "step": 21380, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.469162 }, { "epoch": 0.9161989631978065, "grad_norm": 3.1778290271759033, "learning_rate": 9.194152915184564e-05, "loss": 0.4972710609436035, "memory(GiB)": 61.91, "step": 21385, "token_acc": 0.8729641693811075, "train_speed(iter/s)": 1.469185 }, { "epoch": 0.9164131785270554, "grad_norm": 3.5028743743896484, "learning_rate": 9.193786513361428e-05, "loss": 0.4686120986938477, "memory(GiB)": 61.91, "step": 21390, "token_acc": 0.8892215568862275, "train_speed(iter/s)": 1.469212 }, { "epoch": 0.9166273938563043, "grad_norm": 4.076226234436035, "learning_rate": 9.193420035563916e-05, "loss": 0.31940126419067383, "memory(GiB)": 61.91, "step": 21395, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.469211 }, { "epoch": 0.9168416091855534, "grad_norm": 3.7301125526428223, "learning_rate": 9.193053481798667e-05, "loss": 0.5384344577789306, "memory(GiB)": 61.91, "step": 21400, "token_acc": 0.8939393939393939, "train_speed(iter/s)": 1.469209 }, { "epoch": 0.9170558245148023, "grad_norm": 3.6700236797332764, "learning_rate": 9.192686852072321e-05, "loss": 0.5279745101928711, "memory(GiB)": 61.91, "step": 21405, "token_acc": 0.8759398496240601, "train_speed(iter/s)": 1.469215 }, { "epoch": 0.9172700398440512, "grad_norm": 5.704214096069336, "learning_rate": 9.192320146391518e-05, "loss": 0.31001012325286864, "memory(GiB)": 61.91, "step": 21410, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.469227 }, { "epoch": 0.9174842551733002, "grad_norm": 5.04512357711792, "learning_rate": 9.191953364762904e-05, "loss": 0.5324447631835938, "memory(GiB)": 61.91, "step": 21415, "token_acc": 0.8790035587188612, "train_speed(iter/s)": 1.469238 }, { "epoch": 0.9176984705025492, "grad_norm": 3.31048846244812, "learning_rate": 9.191586507193122e-05, "loss": 0.3500903844833374, "memory(GiB)": 61.91, "step": 21420, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.469219 }, { "epoch": 0.9179126858317981, "grad_norm": 3.0110814571380615, "learning_rate": 9.191219573688819e-05, "loss": 0.282306170463562, "memory(GiB)": 61.91, "step": 21425, "token_acc": 0.9586776859504132, "train_speed(iter/s)": 1.469214 }, { "epoch": 0.9181269011610471, "grad_norm": 1.8160474300384521, "learning_rate": 9.190852564256641e-05, "loss": 0.3427959680557251, "memory(GiB)": 61.91, "step": 21430, "token_acc": 0.9157894736842105, "train_speed(iter/s)": 1.469246 }, { "epoch": 0.918341116490296, "grad_norm": 0.09367097914218903, "learning_rate": 9.190485478903238e-05, "loss": 0.60965576171875, "memory(GiB)": 61.91, "step": 21435, "token_acc": 0.875, "train_speed(iter/s)": 1.469285 }, { "epoch": 0.918555331819545, "grad_norm": 1.5077035427093506, "learning_rate": 9.190118317635259e-05, "loss": 0.5646876335144043, "memory(GiB)": 61.91, "step": 21440, "token_acc": 0.903010033444816, "train_speed(iter/s)": 1.469327 }, { "epoch": 0.918769547148794, "grad_norm": 4.373021602630615, "learning_rate": 9.189751080459357e-05, "loss": 0.461979341506958, "memory(GiB)": 61.91, "step": 21445, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.469314 }, { "epoch": 0.9189837624780429, "grad_norm": 2.6770918369293213, "learning_rate": 9.189383767382182e-05, "loss": 0.39444169998168943, "memory(GiB)": 61.91, "step": 21450, "token_acc": 0.8969072164948454, "train_speed(iter/s)": 1.469331 }, { "epoch": 0.9191979778072918, "grad_norm": 2.621260643005371, "learning_rate": 9.189016378410393e-05, "loss": 0.7233175277709961, "memory(GiB)": 61.91, "step": 21455, "token_acc": 0.8721804511278195, "train_speed(iter/s)": 1.469375 }, { "epoch": 0.9194121931365409, "grad_norm": 4.042331695556641, "learning_rate": 9.188648913550641e-05, "loss": 0.34184041023254397, "memory(GiB)": 61.91, "step": 21460, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.469371 }, { "epoch": 0.9196264084657898, "grad_norm": 1.2194174528121948, "learning_rate": 9.188281372809584e-05, "loss": 0.35287740230560305, "memory(GiB)": 61.91, "step": 21465, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.469355 }, { "epoch": 0.9198406237950387, "grad_norm": 4.231706142425537, "learning_rate": 9.187913756193882e-05, "loss": 0.3508771896362305, "memory(GiB)": 61.91, "step": 21470, "token_acc": 0.9331210191082803, "train_speed(iter/s)": 1.469356 }, { "epoch": 0.9200548391242878, "grad_norm": 7.7826080322265625, "learning_rate": 9.187546063710193e-05, "loss": 0.3142568826675415, "memory(GiB)": 61.91, "step": 21475, "token_acc": 0.941717791411043, "train_speed(iter/s)": 1.469381 }, { "epoch": 0.9202690544535367, "grad_norm": 2.4413230419158936, "learning_rate": 9.18717829536518e-05, "loss": 0.2336583614349365, "memory(GiB)": 61.91, "step": 21480, "token_acc": 0.9461077844311377, "train_speed(iter/s)": 1.469376 }, { "epoch": 0.9204832697827856, "grad_norm": 13.328428268432617, "learning_rate": 9.186810451165502e-05, "loss": 0.4836394786834717, "memory(GiB)": 61.91, "step": 21485, "token_acc": 0.889273356401384, "train_speed(iter/s)": 1.469454 }, { "epoch": 0.9206974851120346, "grad_norm": 1.6326860189437866, "learning_rate": 9.186442531117828e-05, "loss": 0.14376662969589232, "memory(GiB)": 61.91, "step": 21490, "token_acc": 0.9608540925266904, "train_speed(iter/s)": 1.469446 }, { "epoch": 0.9209117004412836, "grad_norm": 3.1716983318328857, "learning_rate": 9.18607453522882e-05, "loss": 0.4406903743743896, "memory(GiB)": 61.91, "step": 21495, "token_acc": 0.9152542372881356, "train_speed(iter/s)": 1.469444 }, { "epoch": 0.9211259157705325, "grad_norm": 4.195250034332275, "learning_rate": 9.185706463505143e-05, "loss": 0.41508235931396487, "memory(GiB)": 61.91, "step": 21500, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.469437 }, { "epoch": 0.9211259157705325, "eval_loss": 2.4809060096740723, "eval_runtime": 13.5809, "eval_samples_per_second": 7.363, "eval_steps_per_second": 7.363, "eval_token_acc": 0.47137150466045274, "step": 21500 }, { "epoch": 0.9213401310997815, "grad_norm": 5.0348076820373535, "learning_rate": 9.185338315953468e-05, "loss": 0.5375232219696044, "memory(GiB)": 61.91, "step": 21505, "token_acc": 0.6016869728209935, "train_speed(iter/s)": 1.467996 }, { "epoch": 0.9215543464290304, "grad_norm": 3.5491511821746826, "learning_rate": 9.184970092580463e-05, "loss": 0.35980677604675293, "memory(GiB)": 61.91, "step": 21510, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.467997 }, { "epoch": 0.9217685617582794, "grad_norm": 3.6351449489593506, "learning_rate": 9.1846017933928e-05, "loss": 0.6836559772491455, "memory(GiB)": 61.91, "step": 21515, "token_acc": 0.8622950819672132, "train_speed(iter/s)": 1.468031 }, { "epoch": 0.9219827770875284, "grad_norm": 5.040965557098389, "learning_rate": 9.184233418397148e-05, "loss": 0.5772891044616699, "memory(GiB)": 61.91, "step": 21520, "token_acc": 0.8552631578947368, "train_speed(iter/s)": 1.46805 }, { "epoch": 0.9221969924167773, "grad_norm": 2.6130921840667725, "learning_rate": 9.183864967600184e-05, "loss": 0.33608782291412354, "memory(GiB)": 61.91, "step": 21525, "token_acc": 0.9131832797427653, "train_speed(iter/s)": 1.468032 }, { "epoch": 0.9224112077460263, "grad_norm": 3.7164864540100098, "learning_rate": 9.183496441008581e-05, "loss": 0.6200734138488769, "memory(GiB)": 61.91, "step": 21530, "token_acc": 0.901840490797546, "train_speed(iter/s)": 1.468053 }, { "epoch": 0.9226254230752753, "grad_norm": 3.2450459003448486, "learning_rate": 9.183127838629016e-05, "loss": 0.34793431758880616, "memory(GiB)": 61.91, "step": 21535, "token_acc": 0.932, "train_speed(iter/s)": 1.468063 }, { "epoch": 0.9228396384045242, "grad_norm": 3.1909561157226562, "learning_rate": 9.182759160468164e-05, "loss": 0.502778959274292, "memory(GiB)": 61.91, "step": 21540, "token_acc": 0.8908554572271387, "train_speed(iter/s)": 1.468055 }, { "epoch": 0.9230538537337732, "grad_norm": 2.9813499450683594, "learning_rate": 9.182390406532708e-05, "loss": 0.3325430631637573, "memory(GiB)": 61.91, "step": 21545, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.468059 }, { "epoch": 0.9232680690630222, "grad_norm": 0.42148756980895996, "learning_rate": 9.182021576829326e-05, "loss": 0.2874005317687988, "memory(GiB)": 61.91, "step": 21550, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.468051 }, { "epoch": 0.9234822843922711, "grad_norm": 7.309770107269287, "learning_rate": 9.1816526713647e-05, "loss": 0.41666626930236816, "memory(GiB)": 61.91, "step": 21555, "token_acc": 0.9102167182662538, "train_speed(iter/s)": 1.468058 }, { "epoch": 0.9236964997215201, "grad_norm": 2.943063259124756, "learning_rate": 9.181283690145514e-05, "loss": 0.44800772666931155, "memory(GiB)": 61.91, "step": 21560, "token_acc": 0.9005847953216374, "train_speed(iter/s)": 1.468119 }, { "epoch": 0.923910715050769, "grad_norm": 2.407881021499634, "learning_rate": 9.18091463317845e-05, "loss": 0.5137041568756103, "memory(GiB)": 61.91, "step": 21565, "token_acc": 0.89419795221843, "train_speed(iter/s)": 1.468111 }, { "epoch": 0.924124930380018, "grad_norm": 3.065321207046509, "learning_rate": 9.180545500470197e-05, "loss": 0.1917765974998474, "memory(GiB)": 61.91, "step": 21570, "token_acc": 0.9442724458204335, "train_speed(iter/s)": 1.468135 }, { "epoch": 0.924339145709267, "grad_norm": 5.5132341384887695, "learning_rate": 9.18017629202744e-05, "loss": 0.3915544033050537, "memory(GiB)": 61.91, "step": 21575, "token_acc": 0.9260869565217391, "train_speed(iter/s)": 1.468147 }, { "epoch": 0.9245533610385159, "grad_norm": 4.298677444458008, "learning_rate": 9.179807007856867e-05, "loss": 0.5810704231262207, "memory(GiB)": 61.91, "step": 21580, "token_acc": 0.8877551020408163, "train_speed(iter/s)": 1.468184 }, { "epoch": 0.9247675763677649, "grad_norm": 2.4810447692871094, "learning_rate": 9.179437647965172e-05, "loss": 0.2753650188446045, "memory(GiB)": 61.91, "step": 21585, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.468185 }, { "epoch": 0.9249817916970139, "grad_norm": 0.9998043775558472, "learning_rate": 9.179068212359041e-05, "loss": 0.31003599166870116, "memory(GiB)": 61.91, "step": 21590, "token_acc": 0.9081632653061225, "train_speed(iter/s)": 1.468171 }, { "epoch": 0.9251960070262628, "grad_norm": 1.3406322002410889, "learning_rate": 9.178698701045169e-05, "loss": 0.5939372539520263, "memory(GiB)": 61.91, "step": 21595, "token_acc": 0.8881118881118881, "train_speed(iter/s)": 1.468168 }, { "epoch": 0.9254102223555117, "grad_norm": 4.452731609344482, "learning_rate": 9.178329114030251e-05, "loss": 0.6086047172546387, "memory(GiB)": 61.91, "step": 21600, "token_acc": 0.8770764119601329, "train_speed(iter/s)": 1.468188 }, { "epoch": 0.9256244376847608, "grad_norm": 4.703988075256348, "learning_rate": 9.177959451320981e-05, "loss": 0.43970203399658203, "memory(GiB)": 61.91, "step": 21605, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.468251 }, { "epoch": 0.9258386530140097, "grad_norm": 4.260828495025635, "learning_rate": 9.177589712924055e-05, "loss": 0.2748166799545288, "memory(GiB)": 61.91, "step": 21610, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.468258 }, { "epoch": 0.9260528683432586, "grad_norm": 2.5444605350494385, "learning_rate": 9.177219898846175e-05, "loss": 0.46671123504638673, "memory(GiB)": 61.91, "step": 21615, "token_acc": 0.9028571428571428, "train_speed(iter/s)": 1.468236 }, { "epoch": 0.9262670836725077, "grad_norm": 0.20816941559314728, "learning_rate": 9.176850009094037e-05, "loss": 0.18379536867141724, "memory(GiB)": 61.91, "step": 21620, "token_acc": 0.9477124183006536, "train_speed(iter/s)": 1.468265 }, { "epoch": 0.9264812990017566, "grad_norm": 2.2215750217437744, "learning_rate": 9.176480043674343e-05, "loss": 0.6445996284484863, "memory(GiB)": 61.91, "step": 21625, "token_acc": 0.8631921824104235, "train_speed(iter/s)": 1.468308 }, { "epoch": 0.9266955143310055, "grad_norm": 2.9829514026641846, "learning_rate": 9.176110002593794e-05, "loss": 0.2611949920654297, "memory(GiB)": 61.91, "step": 21630, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.468312 }, { "epoch": 0.9269097296602545, "grad_norm": 0.739586353302002, "learning_rate": 9.175739885859095e-05, "loss": 0.2256244421005249, "memory(GiB)": 61.91, "step": 21635, "token_acc": 0.9447852760736196, "train_speed(iter/s)": 1.46831 }, { "epoch": 0.9271239449895035, "grad_norm": 1.6529383659362793, "learning_rate": 9.175369693476951e-05, "loss": 0.407302713394165, "memory(GiB)": 61.91, "step": 21640, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.468305 }, { "epoch": 0.9273381603187524, "grad_norm": 4.504682540893555, "learning_rate": 9.17499942545407e-05, "loss": 0.552332067489624, "memory(GiB)": 61.91, "step": 21645, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.468325 }, { "epoch": 0.9275523756480014, "grad_norm": 2.626634359359741, "learning_rate": 9.174629081797156e-05, "loss": 0.3850921869277954, "memory(GiB)": 61.91, "step": 21650, "token_acc": 0.9237804878048781, "train_speed(iter/s)": 1.468361 }, { "epoch": 0.9277665909772503, "grad_norm": 4.137327194213867, "learning_rate": 9.174258662512921e-05, "loss": 0.6813245296478272, "memory(GiB)": 61.91, "step": 21655, "token_acc": 0.8544061302681992, "train_speed(iter/s)": 1.468361 }, { "epoch": 0.9279808063064993, "grad_norm": 2.16678786277771, "learning_rate": 9.173888167608074e-05, "loss": 0.31565237045288086, "memory(GiB)": 61.91, "step": 21660, "token_acc": 0.933852140077821, "train_speed(iter/s)": 1.46834 }, { "epoch": 0.9281950216357483, "grad_norm": 4.856224536895752, "learning_rate": 9.173517597089328e-05, "loss": 0.42600555419921876, "memory(GiB)": 61.91, "step": 21665, "token_acc": 0.9125475285171103, "train_speed(iter/s)": 1.468332 }, { "epoch": 0.9284092369649972, "grad_norm": 3.3635854721069336, "learning_rate": 9.173146950963396e-05, "loss": 0.5583606719970703, "memory(GiB)": 61.91, "step": 21670, "token_acc": 0.869281045751634, "train_speed(iter/s)": 1.468336 }, { "epoch": 0.9286234522942461, "grad_norm": 2.0386464595794678, "learning_rate": 9.17277622923699e-05, "loss": 0.36098499298095704, "memory(GiB)": 61.91, "step": 21675, "token_acc": 0.9010238907849829, "train_speed(iter/s)": 1.46836 }, { "epoch": 0.9288376676234952, "grad_norm": 3.757216453552246, "learning_rate": 9.172405431916831e-05, "loss": 0.3979825019836426, "memory(GiB)": 61.91, "step": 21680, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.468339 }, { "epoch": 0.9290518829527441, "grad_norm": 3.399961471557617, "learning_rate": 9.172034559009632e-05, "loss": 0.43909454345703125, "memory(GiB)": 61.91, "step": 21685, "token_acc": 0.9112903225806451, "train_speed(iter/s)": 1.468346 }, { "epoch": 0.929266098281993, "grad_norm": 4.831272125244141, "learning_rate": 9.171663610522114e-05, "loss": 0.5059815406799316, "memory(GiB)": 61.91, "step": 21690, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.468358 }, { "epoch": 0.9294803136112421, "grad_norm": 2.395486354827881, "learning_rate": 9.171292586460996e-05, "loss": 0.21482341289520263, "memory(GiB)": 61.91, "step": 21695, "token_acc": 0.9368029739776952, "train_speed(iter/s)": 1.468353 }, { "epoch": 0.929694528940491, "grad_norm": 2.5342960357666016, "learning_rate": 9.170921486833e-05, "loss": 0.3123177051544189, "memory(GiB)": 61.91, "step": 21700, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.468364 }, { "epoch": 0.9299087442697399, "grad_norm": 6.398664474487305, "learning_rate": 9.170550311644848e-05, "loss": 0.29514780044555666, "memory(GiB)": 61.91, "step": 21705, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.46836 }, { "epoch": 0.9301229595989889, "grad_norm": 10.130684852600098, "learning_rate": 9.170179060903265e-05, "loss": 0.4303934097290039, "memory(GiB)": 61.91, "step": 21710, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.468385 }, { "epoch": 0.9303371749282379, "grad_norm": 1.6480915546417236, "learning_rate": 9.169807734614976e-05, "loss": 0.3645016670227051, "memory(GiB)": 61.91, "step": 21715, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.468385 }, { "epoch": 0.9305513902574868, "grad_norm": 0.2999686896800995, "learning_rate": 9.16943633278671e-05, "loss": 0.1993804931640625, "memory(GiB)": 61.91, "step": 21720, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.468389 }, { "epoch": 0.9307656055867358, "grad_norm": 5.946140289306641, "learning_rate": 9.169064855425191e-05, "loss": 0.5237468242645263, "memory(GiB)": 61.91, "step": 21725, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.468381 }, { "epoch": 0.9309798209159847, "grad_norm": 2.3099989891052246, "learning_rate": 9.168693302537155e-05, "loss": 0.12125378847122192, "memory(GiB)": 61.91, "step": 21730, "token_acc": 0.9662921348314607, "train_speed(iter/s)": 1.468389 }, { "epoch": 0.9311940362452337, "grad_norm": 4.910868167877197, "learning_rate": 9.168321674129326e-05, "loss": 0.7533376693725586, "memory(GiB)": 61.91, "step": 21735, "token_acc": 0.8374558303886925, "train_speed(iter/s)": 1.468406 }, { "epoch": 0.9314082515744827, "grad_norm": 2.5448858737945557, "learning_rate": 9.16794997020844e-05, "loss": 0.5284440994262696, "memory(GiB)": 61.91, "step": 21740, "token_acc": 0.8783783783783784, "train_speed(iter/s)": 1.468443 }, { "epoch": 0.9316224669037316, "grad_norm": 5.13730001449585, "learning_rate": 9.167578190781232e-05, "loss": 0.47591609954833985, "memory(GiB)": 61.91, "step": 21745, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.468473 }, { "epoch": 0.9318366822329806, "grad_norm": 3.113262176513672, "learning_rate": 9.167206335854435e-05, "loss": 0.44594440460205076, "memory(GiB)": 61.91, "step": 21750, "token_acc": 0.903010033444816, "train_speed(iter/s)": 1.468502 }, { "epoch": 0.9320508975622296, "grad_norm": 5.642615795135498, "learning_rate": 9.166834405434785e-05, "loss": 0.3996292591094971, "memory(GiB)": 61.91, "step": 21755, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.468549 }, { "epoch": 0.9322651128914785, "grad_norm": 2.782771110534668, "learning_rate": 9.166462399529021e-05, "loss": 0.2934854984283447, "memory(GiB)": 61.91, "step": 21760, "token_acc": 0.9291338582677166, "train_speed(iter/s)": 1.468528 }, { "epoch": 0.9324793282207274, "grad_norm": 2.485308885574341, "learning_rate": 9.166090318143883e-05, "loss": 0.33316378593444823, "memory(GiB)": 61.91, "step": 21765, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.46853 }, { "epoch": 0.9326935435499765, "grad_norm": 4.183074474334717, "learning_rate": 9.165718161286111e-05, "loss": 0.2629737377166748, "memory(GiB)": 61.91, "step": 21770, "token_acc": 0.9507692307692308, "train_speed(iter/s)": 1.468589 }, { "epoch": 0.9329077588792254, "grad_norm": 9.6895170211792, "learning_rate": 9.165345928962446e-05, "loss": 0.446839714050293, "memory(GiB)": 61.91, "step": 21775, "token_acc": 0.909375, "train_speed(iter/s)": 1.468582 }, { "epoch": 0.9331219742084743, "grad_norm": 1.9767284393310547, "learning_rate": 9.164973621179634e-05, "loss": 0.3707670927047729, "memory(GiB)": 61.91, "step": 21780, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.468574 }, { "epoch": 0.9333361895377233, "grad_norm": 3.75480580329895, "learning_rate": 9.164601237944415e-05, "loss": 0.5814951419830322, "memory(GiB)": 61.91, "step": 21785, "token_acc": 0.8786764705882353, "train_speed(iter/s)": 1.468569 }, { "epoch": 0.9335504048669723, "grad_norm": 4.223403453826904, "learning_rate": 9.16422877926354e-05, "loss": 0.48333230018615725, "memory(GiB)": 61.91, "step": 21790, "token_acc": 0.907258064516129, "train_speed(iter/s)": 1.468565 }, { "epoch": 0.9337646201962212, "grad_norm": 7.418219566345215, "learning_rate": 9.163856245143752e-05, "loss": 0.30231361389160155, "memory(GiB)": 61.91, "step": 21795, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.468544 }, { "epoch": 0.9339788355254702, "grad_norm": 0.9733916521072388, "learning_rate": 9.163483635591804e-05, "loss": 0.5750089168548584, "memory(GiB)": 61.91, "step": 21800, "token_acc": 0.8605442176870748, "train_speed(iter/s)": 1.468541 }, { "epoch": 0.9341930508547192, "grad_norm": 3.0943830013275146, "learning_rate": 9.163110950614445e-05, "loss": 0.5243294715881348, "memory(GiB)": 61.91, "step": 21805, "token_acc": 0.8904109589041096, "train_speed(iter/s)": 1.468537 }, { "epoch": 0.9344072661839681, "grad_norm": 2.1970760822296143, "learning_rate": 9.162738190218424e-05, "loss": 0.5049887657165527, "memory(GiB)": 61.91, "step": 21810, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.468535 }, { "epoch": 0.9346214815132171, "grad_norm": 2.471345901489258, "learning_rate": 9.162365354410496e-05, "loss": 0.3774231433868408, "memory(GiB)": 61.91, "step": 21815, "token_acc": 0.9022556390977443, "train_speed(iter/s)": 1.468522 }, { "epoch": 0.934835696842466, "grad_norm": 2.2825400829315186, "learning_rate": 9.161992443197416e-05, "loss": 0.6198227405548096, "memory(GiB)": 61.91, "step": 21820, "token_acc": 0.8742857142857143, "train_speed(iter/s)": 1.46855 }, { "epoch": 0.935049912171715, "grad_norm": 8.822504043579102, "learning_rate": 9.161619456585937e-05, "loss": 0.5631507396697998, "memory(GiB)": 61.91, "step": 21825, "token_acc": 0.8767605633802817, "train_speed(iter/s)": 1.468598 }, { "epoch": 0.935264127500964, "grad_norm": 0.9258486032485962, "learning_rate": 9.161246394582818e-05, "loss": 0.21416058540344238, "memory(GiB)": 61.91, "step": 21830, "token_acc": 0.9573643410852714, "train_speed(iter/s)": 1.468604 }, { "epoch": 0.9354783428302129, "grad_norm": 0.7375249266624451, "learning_rate": 9.160873257194818e-05, "loss": 0.5469198226928711, "memory(GiB)": 61.91, "step": 21835, "token_acc": 0.889589905362776, "train_speed(iter/s)": 1.468594 }, { "epoch": 0.9356925581594618, "grad_norm": 2.162390947341919, "learning_rate": 9.160500044428696e-05, "loss": 0.15717933177947999, "memory(GiB)": 61.91, "step": 21840, "token_acc": 0.9575289575289575, "train_speed(iter/s)": 1.468609 }, { "epoch": 0.9359067734887109, "grad_norm": 3.077003240585327, "learning_rate": 9.160126756291211e-05, "loss": 0.1973029136657715, "memory(GiB)": 61.91, "step": 21845, "token_acc": 0.9623824451410659, "train_speed(iter/s)": 1.468605 }, { "epoch": 0.9361209888179598, "grad_norm": 4.814973831176758, "learning_rate": 9.15975339278913e-05, "loss": 0.4534627914428711, "memory(GiB)": 61.91, "step": 21850, "token_acc": 0.9015384615384615, "train_speed(iter/s)": 1.468597 }, { "epoch": 0.9363352041472087, "grad_norm": 3.5661003589630127, "learning_rate": 9.159379953929213e-05, "loss": 0.34124555587768557, "memory(GiB)": 61.91, "step": 21855, "token_acc": 0.9375, "train_speed(iter/s)": 1.468606 }, { "epoch": 0.9365494194764578, "grad_norm": 0.24279503524303436, "learning_rate": 9.159006439718226e-05, "loss": 0.39367103576660156, "memory(GiB)": 61.91, "step": 21860, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.468587 }, { "epoch": 0.9367636348057067, "grad_norm": 5.398018836975098, "learning_rate": 9.158632850162935e-05, "loss": 0.9319599151611329, "memory(GiB)": 61.91, "step": 21865, "token_acc": 0.81640625, "train_speed(iter/s)": 1.468583 }, { "epoch": 0.9369778501349556, "grad_norm": 1.7811599969863892, "learning_rate": 9.158259185270108e-05, "loss": 0.1912021040916443, "memory(GiB)": 61.91, "step": 21870, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.468573 }, { "epoch": 0.9371920654642046, "grad_norm": 3.930480718612671, "learning_rate": 9.157885445046519e-05, "loss": 0.3817904472351074, "memory(GiB)": 61.91, "step": 21875, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.468562 }, { "epoch": 0.9374062807934536, "grad_norm": 1.0631310939788818, "learning_rate": 9.157511629498932e-05, "loss": 0.24165065288543702, "memory(GiB)": 61.91, "step": 21880, "token_acc": 0.9465648854961832, "train_speed(iter/s)": 1.468612 }, { "epoch": 0.9376204961227026, "grad_norm": 3.7608025074005127, "learning_rate": 9.157137738634122e-05, "loss": 0.41273512840271, "memory(GiB)": 61.91, "step": 21885, "token_acc": 0.9105058365758755, "train_speed(iter/s)": 1.468643 }, { "epoch": 0.9378347114519515, "grad_norm": 1.4621498584747314, "learning_rate": 9.156763772458862e-05, "loss": 0.5931309223175049, "memory(GiB)": 61.91, "step": 21890, "token_acc": 0.8614130434782609, "train_speed(iter/s)": 1.468755 }, { "epoch": 0.9380489267812004, "grad_norm": 12.01050853729248, "learning_rate": 9.156389730979928e-05, "loss": 0.562861967086792, "memory(GiB)": 61.91, "step": 21895, "token_acc": 0.8742138364779874, "train_speed(iter/s)": 1.468777 }, { "epoch": 0.9382631421104495, "grad_norm": 3.3016483783721924, "learning_rate": 9.156015614204094e-05, "loss": 0.5169822692871093, "memory(GiB)": 61.91, "step": 21900, "token_acc": 0.8850574712643678, "train_speed(iter/s)": 1.468782 }, { "epoch": 0.9384773574396984, "grad_norm": 2.3634750843048096, "learning_rate": 9.155641422138139e-05, "loss": 0.3248711109161377, "memory(GiB)": 61.91, "step": 21905, "token_acc": 0.9389312977099237, "train_speed(iter/s)": 1.468796 }, { "epoch": 0.9386915727689473, "grad_norm": 3.5388264656066895, "learning_rate": 9.15526715478884e-05, "loss": 0.3158034563064575, "memory(GiB)": 61.91, "step": 21910, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.468793 }, { "epoch": 0.9389057880981964, "grad_norm": 2.4606473445892334, "learning_rate": 9.15489281216298e-05, "loss": 0.24230554103851318, "memory(GiB)": 61.91, "step": 21915, "token_acc": 0.940677966101695, "train_speed(iter/s)": 1.468768 }, { "epoch": 0.9391200034274453, "grad_norm": 8.201204299926758, "learning_rate": 9.154518394267338e-05, "loss": 0.43552207946777344, "memory(GiB)": 61.91, "step": 21920, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.468754 }, { "epoch": 0.9393342187566942, "grad_norm": 2.507315158843994, "learning_rate": 9.154143901108699e-05, "loss": 0.27986721992492675, "memory(GiB)": 61.91, "step": 21925, "token_acc": 0.94, "train_speed(iter/s)": 1.468726 }, { "epoch": 0.9395484340859432, "grad_norm": 5.884382724761963, "learning_rate": 9.153769332693847e-05, "loss": 0.4830728530883789, "memory(GiB)": 61.91, "step": 21930, "token_acc": 0.8942307692307693, "train_speed(iter/s)": 1.468723 }, { "epoch": 0.9397626494151922, "grad_norm": 4.490473747253418, "learning_rate": 9.153394689029566e-05, "loss": 0.4647077560424805, "memory(GiB)": 61.91, "step": 21935, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.468717 }, { "epoch": 0.9399768647444411, "grad_norm": 4.24407434463501, "learning_rate": 9.153019970122643e-05, "loss": 0.2609121799468994, "memory(GiB)": 61.91, "step": 21940, "token_acc": 0.943089430894309, "train_speed(iter/s)": 1.46871 }, { "epoch": 0.9401910800736901, "grad_norm": 4.920910358428955, "learning_rate": 9.15264517597987e-05, "loss": 0.45217576026916506, "memory(GiB)": 61.91, "step": 21945, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.468714 }, { "epoch": 0.940405295402939, "grad_norm": 1.928147315979004, "learning_rate": 9.152270306608031e-05, "loss": 0.5785836219787598, "memory(GiB)": 61.91, "step": 21950, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.468723 }, { "epoch": 0.940619510732188, "grad_norm": 6.729263782501221, "learning_rate": 9.151895362013922e-05, "loss": 0.7683110237121582, "memory(GiB)": 61.91, "step": 21955, "token_acc": 0.8605577689243028, "train_speed(iter/s)": 1.468791 }, { "epoch": 0.940833726061437, "grad_norm": 3.0432956218719482, "learning_rate": 9.151520342204334e-05, "loss": 0.5410183429718017, "memory(GiB)": 61.91, "step": 21960, "token_acc": 0.8778877887788779, "train_speed(iter/s)": 1.468841 }, { "epoch": 0.9410479413906859, "grad_norm": 4.007168292999268, "learning_rate": 9.151145247186061e-05, "loss": 0.3860992670059204, "memory(GiB)": 61.91, "step": 21965, "token_acc": 0.9233128834355828, "train_speed(iter/s)": 1.468839 }, { "epoch": 0.9412621567199349, "grad_norm": 7.444249153137207, "learning_rate": 9.150770076965895e-05, "loss": 0.8896987915039063, "memory(GiB)": 61.91, "step": 21970, "token_acc": 0.8233082706766918, "train_speed(iter/s)": 1.468851 }, { "epoch": 0.9414763720491839, "grad_norm": 2.3637804985046387, "learning_rate": 9.15039483155064e-05, "loss": 0.35094847679138186, "memory(GiB)": 61.91, "step": 21975, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.468852 }, { "epoch": 0.9416905873784328, "grad_norm": 1.595467209815979, "learning_rate": 9.150019510947086e-05, "loss": 0.3369180202484131, "memory(GiB)": 61.91, "step": 21980, "token_acc": 0.9400630914826499, "train_speed(iter/s)": 1.468841 }, { "epoch": 0.9419048027076817, "grad_norm": 9.078901290893555, "learning_rate": 9.149644115162035e-05, "loss": 0.3941490650177002, "memory(GiB)": 61.91, "step": 21985, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.468823 }, { "epoch": 0.9421190180369308, "grad_norm": 3.327976942062378, "learning_rate": 9.149268644202289e-05, "loss": 0.6081852436065673, "memory(GiB)": 61.91, "step": 21990, "token_acc": 0.8776978417266187, "train_speed(iter/s)": 1.46884 }, { "epoch": 0.9423332333661797, "grad_norm": 2.6609435081481934, "learning_rate": 9.148893098074649e-05, "loss": 0.2604940414428711, "memory(GiB)": 61.91, "step": 21995, "token_acc": 0.9401993355481728, "train_speed(iter/s)": 1.468842 }, { "epoch": 0.9425474486954286, "grad_norm": 4.201693534851074, "learning_rate": 9.148517476785918e-05, "loss": 0.31285412311553956, "memory(GiB)": 61.91, "step": 22000, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.468842 }, { "epoch": 0.9425474486954286, "eval_loss": 2.6864635944366455, "eval_runtime": 13.7071, "eval_samples_per_second": 7.295, "eval_steps_per_second": 7.295, "eval_token_acc": 0.4018445322793149, "step": 22000 }, { "epoch": 0.9427616640246776, "grad_norm": 0.37354639172554016, "learning_rate": 9.148141780342903e-05, "loss": 0.47384963035583494, "memory(GiB)": 61.91, "step": 22005, "token_acc": 0.5462184873949579, "train_speed(iter/s)": 1.467417 }, { "epoch": 0.9429758793539266, "grad_norm": 4.4734063148498535, "learning_rate": 9.147766008752407e-05, "loss": 0.4271239757537842, "memory(GiB)": 61.91, "step": 22010, "token_acc": 0.8915254237288136, "train_speed(iter/s)": 1.467407 }, { "epoch": 0.9431900946831755, "grad_norm": 3.5342743396759033, "learning_rate": 9.14739016202124e-05, "loss": 0.21099598407745362, "memory(GiB)": 61.91, "step": 22015, "token_acc": 0.9501779359430605, "train_speed(iter/s)": 1.467413 }, { "epoch": 0.9434043100124245, "grad_norm": 5.982921600341797, "learning_rate": 9.14701424015621e-05, "loss": 0.6093832015991211, "memory(GiB)": 61.91, "step": 22020, "token_acc": 0.875, "train_speed(iter/s)": 1.467409 }, { "epoch": 0.9436185253416735, "grad_norm": 0.34842735528945923, "learning_rate": 9.146638243164125e-05, "loss": 0.2463322877883911, "memory(GiB)": 61.91, "step": 22025, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.467407 }, { "epoch": 0.9438327406709224, "grad_norm": 1.6568808555603027, "learning_rate": 9.1462621710518e-05, "loss": 0.16732157468795777, "memory(GiB)": 61.91, "step": 22030, "token_acc": 0.9562043795620438, "train_speed(iter/s)": 1.467395 }, { "epoch": 0.9440469560001714, "grad_norm": 1.5235697031021118, "learning_rate": 9.145886023826044e-05, "loss": 0.6104533195495605, "memory(GiB)": 61.91, "step": 22035, "token_acc": 0.9018181818181819, "train_speed(iter/s)": 1.4674 }, { "epoch": 0.9442611713294203, "grad_norm": 0.07455126941204071, "learning_rate": 9.145509801493677e-05, "loss": 0.5838310241699218, "memory(GiB)": 61.91, "step": 22040, "token_acc": 0.8783269961977186, "train_speed(iter/s)": 1.467419 }, { "epoch": 0.9444753866586693, "grad_norm": 3.9855546951293945, "learning_rate": 9.145133504061509e-05, "loss": 0.46494288444519044, "memory(GiB)": 61.91, "step": 22045, "token_acc": 0.894927536231884, "train_speed(iter/s)": 1.467403 }, { "epoch": 0.9446896019879183, "grad_norm": 4.242485046386719, "learning_rate": 9.14475713153636e-05, "loss": 0.3579071521759033, "memory(GiB)": 61.91, "step": 22050, "token_acc": 0.9131736526946108, "train_speed(iter/s)": 1.467395 }, { "epoch": 0.9449038173171672, "grad_norm": 8.145403861999512, "learning_rate": 9.144380683925044e-05, "loss": 0.539143180847168, "memory(GiB)": 61.91, "step": 22055, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.467394 }, { "epoch": 0.9451180326464161, "grad_norm": 1.9371577501296997, "learning_rate": 9.144004161234388e-05, "loss": 0.3176047086715698, "memory(GiB)": 61.91, "step": 22060, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.467389 }, { "epoch": 0.9453322479756652, "grad_norm": 3.149219274520874, "learning_rate": 9.143627563471209e-05, "loss": 0.34988203048706057, "memory(GiB)": 61.91, "step": 22065, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.467378 }, { "epoch": 0.9455464633049141, "grad_norm": 1.1061185598373413, "learning_rate": 9.143250890642327e-05, "loss": 0.4117379665374756, "memory(GiB)": 61.91, "step": 22070, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.467377 }, { "epoch": 0.945760678634163, "grad_norm": 15.019400596618652, "learning_rate": 9.142874142754572e-05, "loss": 0.6722813129425049, "memory(GiB)": 61.91, "step": 22075, "token_acc": 0.8736059479553904, "train_speed(iter/s)": 1.467435 }, { "epoch": 0.9459748939634121, "grad_norm": 2.8176960945129395, "learning_rate": 9.142497319814764e-05, "loss": 0.3606797456741333, "memory(GiB)": 61.91, "step": 22080, "token_acc": 0.9197707736389685, "train_speed(iter/s)": 1.46745 }, { "epoch": 0.946189109292661, "grad_norm": 4.598263263702393, "learning_rate": 9.142120421829729e-05, "loss": 0.5131877422332763, "memory(GiB)": 61.91, "step": 22085, "token_acc": 0.9061371841155235, "train_speed(iter/s)": 1.467418 }, { "epoch": 0.9464033246219099, "grad_norm": 4.7928619384765625, "learning_rate": 9.141743448806301e-05, "loss": 0.3072411060333252, "memory(GiB)": 61.91, "step": 22090, "token_acc": 0.9378531073446328, "train_speed(iter/s)": 1.467433 }, { "epoch": 0.9466175399511589, "grad_norm": 4.520856857299805, "learning_rate": 9.141366400751301e-05, "loss": 0.23764169216156006, "memory(GiB)": 61.91, "step": 22095, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.467478 }, { "epoch": 0.9468317552804079, "grad_norm": 2.09917950630188, "learning_rate": 9.140989277671567e-05, "loss": 0.3776474714279175, "memory(GiB)": 61.91, "step": 22100, "token_acc": 0.9367088607594937, "train_speed(iter/s)": 1.46753 }, { "epoch": 0.9470459706096568, "grad_norm": 4.244811058044434, "learning_rate": 9.140612079573927e-05, "loss": 0.37911081314086914, "memory(GiB)": 61.91, "step": 22105, "token_acc": 0.9161073825503355, "train_speed(iter/s)": 1.467533 }, { "epoch": 0.9472601859389058, "grad_norm": 7.26774787902832, "learning_rate": 9.140234806465214e-05, "loss": 0.40667333602905276, "memory(GiB)": 61.91, "step": 22110, "token_acc": 0.9017094017094017, "train_speed(iter/s)": 1.46752 }, { "epoch": 0.9474744012681547, "grad_norm": 2.320934534072876, "learning_rate": 9.139857458352263e-05, "loss": 0.40746731758117677, "memory(GiB)": 61.91, "step": 22115, "token_acc": 0.9295302013422819, "train_speed(iter/s)": 1.467493 }, { "epoch": 0.9476886165974037, "grad_norm": 3.4059460163116455, "learning_rate": 9.139480035241912e-05, "loss": 0.2704026222229004, "memory(GiB)": 61.91, "step": 22120, "token_acc": 0.932806324110672, "train_speed(iter/s)": 1.467522 }, { "epoch": 0.9479028319266527, "grad_norm": 3.425511598587036, "learning_rate": 9.139102537140996e-05, "loss": 0.36084067821502686, "memory(GiB)": 61.91, "step": 22125, "token_acc": 0.9293286219081273, "train_speed(iter/s)": 1.467523 }, { "epoch": 0.9481170472559016, "grad_norm": 0.608832061290741, "learning_rate": 9.138724964056355e-05, "loss": 0.1984044075012207, "memory(GiB)": 61.91, "step": 22130, "token_acc": 0.9571984435797666, "train_speed(iter/s)": 1.467537 }, { "epoch": 0.9483312625851505, "grad_norm": 4.046356678009033, "learning_rate": 9.13834731599483e-05, "loss": 0.3591649532318115, "memory(GiB)": 61.91, "step": 22135, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.467579 }, { "epoch": 0.9485454779143996, "grad_norm": 2.5247511863708496, "learning_rate": 9.13796959296326e-05, "loss": 0.46526460647583007, "memory(GiB)": 61.91, "step": 22140, "token_acc": 0.9115646258503401, "train_speed(iter/s)": 1.46758 }, { "epoch": 0.9487596932436485, "grad_norm": 1.9445387125015259, "learning_rate": 9.137591794968489e-05, "loss": 0.3836775302886963, "memory(GiB)": 61.91, "step": 22145, "token_acc": 0.9326599326599326, "train_speed(iter/s)": 1.467565 }, { "epoch": 0.9489739085728974, "grad_norm": 1.2154020071029663, "learning_rate": 9.137213922017363e-05, "loss": 0.1609580397605896, "memory(GiB)": 61.91, "step": 22150, "token_acc": 0.9705882352941176, "train_speed(iter/s)": 1.467564 }, { "epoch": 0.9491881239021465, "grad_norm": 6.556546211242676, "learning_rate": 9.136835974116724e-05, "loss": 0.37168872356414795, "memory(GiB)": 61.91, "step": 22155, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.46754 }, { "epoch": 0.9494023392313954, "grad_norm": 3.9003090858459473, "learning_rate": 9.136457951273423e-05, "loss": 0.3579442262649536, "memory(GiB)": 61.91, "step": 22160, "token_acc": 0.930635838150289, "train_speed(iter/s)": 1.467526 }, { "epoch": 0.9496165545606443, "grad_norm": 5.424684047698975, "learning_rate": 9.136079853494304e-05, "loss": 0.6327717781066895, "memory(GiB)": 61.91, "step": 22165, "token_acc": 0.8622950819672132, "train_speed(iter/s)": 1.467563 }, { "epoch": 0.9498307698898933, "grad_norm": 0.14198103547096252, "learning_rate": 9.135701680786218e-05, "loss": 0.23424289226531983, "memory(GiB)": 61.91, "step": 22170, "token_acc": 0.9451476793248945, "train_speed(iter/s)": 1.467564 }, { "epoch": 0.9500449852191423, "grad_norm": 6.476556777954102, "learning_rate": 9.135323433156018e-05, "loss": 0.27534635066986085, "memory(GiB)": 61.91, "step": 22175, "token_acc": 0.9488372093023256, "train_speed(iter/s)": 1.467566 }, { "epoch": 0.9502592005483912, "grad_norm": 4.816646099090576, "learning_rate": 9.134945110610554e-05, "loss": 0.2359365463256836, "memory(GiB)": 61.91, "step": 22180, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.467576 }, { "epoch": 0.9504734158776402, "grad_norm": 3.0063388347625732, "learning_rate": 9.134566713156679e-05, "loss": 0.3660297155380249, "memory(GiB)": 61.91, "step": 22185, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.467606 }, { "epoch": 0.9506876312068891, "grad_norm": 11.726957321166992, "learning_rate": 9.134188240801251e-05, "loss": 0.7190132141113281, "memory(GiB)": 61.91, "step": 22190, "token_acc": 0.870253164556962, "train_speed(iter/s)": 1.467605 }, { "epoch": 0.9509018465361381, "grad_norm": 3.5979745388031006, "learning_rate": 9.133809693551125e-05, "loss": 0.421830415725708, "memory(GiB)": 61.91, "step": 22195, "token_acc": 0.9054054054054054, "train_speed(iter/s)": 1.467604 }, { "epoch": 0.9511160618653871, "grad_norm": 0.9612870812416077, "learning_rate": 9.133431071413158e-05, "loss": 0.5331704616546631, "memory(GiB)": 61.91, "step": 22200, "token_acc": 0.8907849829351536, "train_speed(iter/s)": 1.467602 }, { "epoch": 0.951330277194636, "grad_norm": 9.472886085510254, "learning_rate": 9.13305237439421e-05, "loss": 0.36295955181121825, "memory(GiB)": 61.91, "step": 22205, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.467586 }, { "epoch": 0.951544492523885, "grad_norm": 0.37650957703590393, "learning_rate": 9.13267360250114e-05, "loss": 0.10992643833160401, "memory(GiB)": 61.91, "step": 22210, "token_acc": 0.9689655172413794, "train_speed(iter/s)": 1.467568 }, { "epoch": 0.951758707853134, "grad_norm": 2.342907190322876, "learning_rate": 9.132294755740814e-05, "loss": 0.6292852878570556, "memory(GiB)": 61.91, "step": 22215, "token_acc": 0.8885245901639345, "train_speed(iter/s)": 1.467601 }, { "epoch": 0.9519729231823829, "grad_norm": 1.9482372999191284, "learning_rate": 9.131915834120088e-05, "loss": 0.336916446685791, "memory(GiB)": 61.91, "step": 22220, "token_acc": 0.9174917491749175, "train_speed(iter/s)": 1.467591 }, { "epoch": 0.952187138511632, "grad_norm": 0.6206032037734985, "learning_rate": 9.131536837645833e-05, "loss": 0.5354894638061524, "memory(GiB)": 61.91, "step": 22225, "token_acc": 0.9053627760252366, "train_speed(iter/s)": 1.467581 }, { "epoch": 0.9524013538408809, "grad_norm": 4.015766143798828, "learning_rate": 9.131157766324912e-05, "loss": 0.3910503625869751, "memory(GiB)": 61.91, "step": 22230, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.467573 }, { "epoch": 0.9526155691701298, "grad_norm": 0.1864416003227234, "learning_rate": 9.130778620164193e-05, "loss": 0.1978856325149536, "memory(GiB)": 61.91, "step": 22235, "token_acc": 0.9573770491803278, "train_speed(iter/s)": 1.467576 }, { "epoch": 0.9528297844993788, "grad_norm": 1.8641152381896973, "learning_rate": 9.130399399170544e-05, "loss": 0.600673770904541, "memory(GiB)": 61.91, "step": 22240, "token_acc": 0.8849315068493151, "train_speed(iter/s)": 1.467564 }, { "epoch": 0.9530439998286278, "grad_norm": 3.894437789916992, "learning_rate": 9.130020103350836e-05, "loss": 0.5011430740356445, "memory(GiB)": 61.91, "step": 22245, "token_acc": 0.9083333333333333, "train_speed(iter/s)": 1.467552 }, { "epoch": 0.9532582151578767, "grad_norm": 0.9876573085784912, "learning_rate": 9.12964073271194e-05, "loss": 0.11070897579193115, "memory(GiB)": 61.91, "step": 22250, "token_acc": 0.970873786407767, "train_speed(iter/s)": 1.467551 }, { "epoch": 0.9534724304871257, "grad_norm": 3.1903834342956543, "learning_rate": 9.129261287260726e-05, "loss": 0.5551234722137451, "memory(GiB)": 61.91, "step": 22255, "token_acc": 0.8856304985337243, "train_speed(iter/s)": 1.46753 }, { "epoch": 0.9536866458163746, "grad_norm": 0.4327496886253357, "learning_rate": 9.128881767004072e-05, "loss": 0.19804818630218507, "memory(GiB)": 61.91, "step": 22260, "token_acc": 0.9498069498069498, "train_speed(iter/s)": 1.467524 }, { "epoch": 0.9539008611456236, "grad_norm": 2.652514696121216, "learning_rate": 9.12850217194885e-05, "loss": 0.6420618057250976, "memory(GiB)": 61.91, "step": 22265, "token_acc": 0.8524096385542169, "train_speed(iter/s)": 1.467552 }, { "epoch": 0.9541150764748726, "grad_norm": 2.181469440460205, "learning_rate": 9.12812250210194e-05, "loss": 0.3005370616912842, "memory(GiB)": 61.91, "step": 22270, "token_acc": 0.9338235294117647, "train_speed(iter/s)": 1.467567 }, { "epoch": 0.9543292918041215, "grad_norm": 3.3608345985412598, "learning_rate": 9.127742757470217e-05, "loss": 0.4011042594909668, "memory(GiB)": 61.91, "step": 22275, "token_acc": 0.9233576642335767, "train_speed(iter/s)": 1.46759 }, { "epoch": 0.9545435071333704, "grad_norm": 2.2149758338928223, "learning_rate": 9.127362938060563e-05, "loss": 0.3083944320678711, "memory(GiB)": 61.91, "step": 22280, "token_acc": 0.9361022364217252, "train_speed(iter/s)": 1.467565 }, { "epoch": 0.9547577224626195, "grad_norm": 3.811053991317749, "learning_rate": 9.126983043879857e-05, "loss": 0.48477878570556643, "memory(GiB)": 61.91, "step": 22285, "token_acc": 0.9033333333333333, "train_speed(iter/s)": 1.46757 }, { "epoch": 0.9549719377918684, "grad_norm": 5.421357154846191, "learning_rate": 9.126603074934982e-05, "loss": 0.4821812152862549, "memory(GiB)": 61.91, "step": 22290, "token_acc": 0.875, "train_speed(iter/s)": 1.467566 }, { "epoch": 0.9551861531211173, "grad_norm": 3.3017706871032715, "learning_rate": 9.126223031232822e-05, "loss": 0.47586774826049805, "memory(GiB)": 61.91, "step": 22295, "token_acc": 0.9112903225806451, "train_speed(iter/s)": 1.467572 }, { "epoch": 0.9554003684503664, "grad_norm": 4.029461860656738, "learning_rate": 9.125842912780259e-05, "loss": 0.3086519479751587, "memory(GiB)": 61.91, "step": 22300, "token_acc": 0.9462025316455697, "train_speed(iter/s)": 1.467576 }, { "epoch": 0.9556145837796153, "grad_norm": 5.007809638977051, "learning_rate": 9.125462719584183e-05, "loss": 0.5117624759674072, "memory(GiB)": 61.91, "step": 22305, "token_acc": 0.8733333333333333, "train_speed(iter/s)": 1.467603 }, { "epoch": 0.9558287991088642, "grad_norm": 6.0248212814331055, "learning_rate": 9.125082451651479e-05, "loss": 0.5615583419799804, "memory(GiB)": 61.91, "step": 22310, "token_acc": 0.8787878787878788, "train_speed(iter/s)": 1.467633 }, { "epoch": 0.9560430144381132, "grad_norm": 3.623988151550293, "learning_rate": 9.124702108989036e-05, "loss": 0.49052066802978517, "memory(GiB)": 61.91, "step": 22315, "token_acc": 0.903010033444816, "train_speed(iter/s)": 1.467623 }, { "epoch": 0.9562572297673622, "grad_norm": 5.618017196655273, "learning_rate": 9.124321691603747e-05, "loss": 0.5810177326202393, "memory(GiB)": 61.91, "step": 22320, "token_acc": 0.889763779527559, "train_speed(iter/s)": 1.467696 }, { "epoch": 0.9564714450966111, "grad_norm": 2.131089687347412, "learning_rate": 9.123941199502501e-05, "loss": 0.33435850143432616, "memory(GiB)": 61.91, "step": 22325, "token_acc": 0.9190751445086706, "train_speed(iter/s)": 1.4677 }, { "epoch": 0.9566856604258601, "grad_norm": 6.61406135559082, "learning_rate": 9.12356063269219e-05, "loss": 0.1339050531387329, "memory(GiB)": 61.91, "step": 22330, "token_acc": 0.9636363636363636, "train_speed(iter/s)": 1.467719 }, { "epoch": 0.956899875755109, "grad_norm": 5.483438014984131, "learning_rate": 9.123179991179711e-05, "loss": 0.5129273891448974, "memory(GiB)": 61.91, "step": 22335, "token_acc": 0.8795620437956204, "train_speed(iter/s)": 1.467755 }, { "epoch": 0.957114091084358, "grad_norm": 3.1934163570404053, "learning_rate": 9.122799274971959e-05, "loss": 0.5013748168945312, "memory(GiB)": 61.91, "step": 22340, "token_acc": 0.8991825613079019, "train_speed(iter/s)": 1.467779 }, { "epoch": 0.957328306413607, "grad_norm": 2.149122476577759, "learning_rate": 9.12241848407583e-05, "loss": 0.35799126625061034, "memory(GiB)": 61.91, "step": 22345, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.467799 }, { "epoch": 0.9575425217428559, "grad_norm": 5.2121500968933105, "learning_rate": 9.122037618498225e-05, "loss": 0.32495272159576416, "memory(GiB)": 61.91, "step": 22350, "token_acc": 0.9336734693877551, "train_speed(iter/s)": 1.467806 }, { "epoch": 0.9577567370721048, "grad_norm": 2.289686679840088, "learning_rate": 9.12165667824604e-05, "loss": 0.436721134185791, "memory(GiB)": 61.91, "step": 22355, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.467808 }, { "epoch": 0.9579709524013539, "grad_norm": 6.328433513641357, "learning_rate": 9.121275663326178e-05, "loss": 0.5549020290374755, "memory(GiB)": 61.91, "step": 22360, "token_acc": 0.891566265060241, "train_speed(iter/s)": 1.467793 }, { "epoch": 0.9581851677306028, "grad_norm": 1.0749785900115967, "learning_rate": 9.120894573745542e-05, "loss": 0.30508639812469485, "memory(GiB)": 61.91, "step": 22365, "token_acc": 0.9369747899159664, "train_speed(iter/s)": 1.467801 }, { "epoch": 0.9583993830598517, "grad_norm": 3.816537380218506, "learning_rate": 9.120513409511033e-05, "loss": 0.7308573722839355, "memory(GiB)": 61.91, "step": 22370, "token_acc": 0.8586206896551725, "train_speed(iter/s)": 1.467852 }, { "epoch": 0.9586135983891008, "grad_norm": 3.2269225120544434, "learning_rate": 9.12013217062956e-05, "loss": 0.22136139869689941, "memory(GiB)": 61.91, "step": 22375, "token_acc": 0.953405017921147, "train_speed(iter/s)": 1.46787 }, { "epoch": 0.9588278137183497, "grad_norm": 2.831312656402588, "learning_rate": 9.119750857108027e-05, "loss": 0.5857760906219482, "memory(GiB)": 61.91, "step": 22380, "token_acc": 0.8945783132530121, "train_speed(iter/s)": 1.467908 }, { "epoch": 0.9590420290475986, "grad_norm": 1.791275143623352, "learning_rate": 9.119369468953344e-05, "loss": 0.16478101015090943, "memory(GiB)": 61.91, "step": 22385, "token_acc": 0.9647058823529412, "train_speed(iter/s)": 1.467907 }, { "epoch": 0.9592562443768476, "grad_norm": 3.5680434703826904, "learning_rate": 9.118988006172418e-05, "loss": 0.44626750946044924, "memory(GiB)": 61.91, "step": 22390, "token_acc": 0.9064327485380117, "train_speed(iter/s)": 1.467886 }, { "epoch": 0.9594704597060966, "grad_norm": 3.1389763355255127, "learning_rate": 9.11860646877216e-05, "loss": 0.27337968349456787, "memory(GiB)": 61.91, "step": 22395, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.467876 }, { "epoch": 0.9596846750353455, "grad_norm": 4.181807994842529, "learning_rate": 9.118224856759482e-05, "loss": 0.19652308225631715, "memory(GiB)": 61.91, "step": 22400, "token_acc": 0.9568106312292359, "train_speed(iter/s)": 1.467862 }, { "epoch": 0.9598988903645945, "grad_norm": 3.2786691188812256, "learning_rate": 9.117843170141297e-05, "loss": 0.3506011962890625, "memory(GiB)": 61.91, "step": 22405, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.46785 }, { "epoch": 0.9601131056938434, "grad_norm": 4.515069961547852, "learning_rate": 9.117461408924521e-05, "loss": 0.4957274436950684, "memory(GiB)": 61.91, "step": 22410, "token_acc": 0.8644688644688645, "train_speed(iter/s)": 1.467848 }, { "epoch": 0.9603273210230924, "grad_norm": 3.6373491287231445, "learning_rate": 9.11707957311607e-05, "loss": 0.358294677734375, "memory(GiB)": 61.91, "step": 22415, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.467846 }, { "epoch": 0.9605415363523414, "grad_norm": 1.7846280336380005, "learning_rate": 9.116697662722859e-05, "loss": 0.32404446601867676, "memory(GiB)": 61.91, "step": 22420, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.467838 }, { "epoch": 0.9607557516815903, "grad_norm": 4.620795726776123, "learning_rate": 9.116315677751807e-05, "loss": 0.3525334119796753, "memory(GiB)": 61.91, "step": 22425, "token_acc": 0.9324894514767933, "train_speed(iter/s)": 1.467812 }, { "epoch": 0.9609699670108393, "grad_norm": 2.1255245208740234, "learning_rate": 9.115933618209838e-05, "loss": 0.5580577850341797, "memory(GiB)": 61.91, "step": 22430, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.46785 }, { "epoch": 0.9611841823400883, "grad_norm": 0.656931459903717, "learning_rate": 9.115551484103869e-05, "loss": 0.34498775005340576, "memory(GiB)": 61.91, "step": 22435, "token_acc": 0.9400749063670412, "train_speed(iter/s)": 1.467908 }, { "epoch": 0.9613983976693372, "grad_norm": 5.368595123291016, "learning_rate": 9.115169275440825e-05, "loss": 0.421937370300293, "memory(GiB)": 61.91, "step": 22440, "token_acc": 0.8934707903780069, "train_speed(iter/s)": 1.467939 }, { "epoch": 0.9616126129985861, "grad_norm": 4.023448944091797, "learning_rate": 9.114786992227629e-05, "loss": 0.461160945892334, "memory(GiB)": 61.91, "step": 22445, "token_acc": 0.8932038834951457, "train_speed(iter/s)": 1.46797 }, { "epoch": 0.9618268283278352, "grad_norm": 1.2711139917373657, "learning_rate": 9.114404634471205e-05, "loss": 0.682069969177246, "memory(GiB)": 61.91, "step": 22450, "token_acc": 0.9021406727828746, "train_speed(iter/s)": 1.467974 }, { "epoch": 0.9620410436570841, "grad_norm": 2.49837327003479, "learning_rate": 9.114022202178483e-05, "loss": 0.5080674648284912, "memory(GiB)": 61.91, "step": 22455, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.468005 }, { "epoch": 0.962255258986333, "grad_norm": 5.150381565093994, "learning_rate": 9.113639695356388e-05, "loss": 0.5699351787567138, "memory(GiB)": 61.91, "step": 22460, "token_acc": 0.8768768768768769, "train_speed(iter/s)": 1.468003 }, { "epoch": 0.962469474315582, "grad_norm": 0.38493117690086365, "learning_rate": 9.113257114011852e-05, "loss": 0.2862043857574463, "memory(GiB)": 61.91, "step": 22465, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.468023 }, { "epoch": 0.962683689644831, "grad_norm": 8.213302612304688, "learning_rate": 9.112874458151805e-05, "loss": 0.5726579666137696, "memory(GiB)": 61.91, "step": 22470, "token_acc": 0.9046052631578947, "train_speed(iter/s)": 1.468044 }, { "epoch": 0.9628979049740799, "grad_norm": 3.255385637283325, "learning_rate": 9.112491727783179e-05, "loss": 0.5179494857788086, "memory(GiB)": 61.91, "step": 22475, "token_acc": 0.8840125391849529, "train_speed(iter/s)": 1.46804 }, { "epoch": 0.9631121203033289, "grad_norm": 2.541604995727539, "learning_rate": 9.112108922912907e-05, "loss": 0.2757928133010864, "memory(GiB)": 61.91, "step": 22480, "token_acc": 0.9463722397476341, "train_speed(iter/s)": 1.468086 }, { "epoch": 0.9633263356325779, "grad_norm": 4.0006103515625, "learning_rate": 9.111726043547926e-05, "loss": 0.5233933448791503, "memory(GiB)": 61.91, "step": 22485, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.468088 }, { "epoch": 0.9635405509618268, "grad_norm": 4.771426677703857, "learning_rate": 9.111343089695168e-05, "loss": 0.18716444969177246, "memory(GiB)": 61.91, "step": 22490, "token_acc": 0.9475806451612904, "train_speed(iter/s)": 1.468145 }, { "epoch": 0.9637547662910758, "grad_norm": 2.943324327468872, "learning_rate": 9.110960061361575e-05, "loss": 0.36272666454315183, "memory(GiB)": 61.91, "step": 22495, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.468142 }, { "epoch": 0.9639689816203247, "grad_norm": 4.363766670227051, "learning_rate": 9.110576958554085e-05, "loss": 0.9681766510009766, "memory(GiB)": 61.91, "step": 22500, "token_acc": 0.7987012987012987, "train_speed(iter/s)": 1.468166 }, { "epoch": 0.9639689816203247, "eval_loss": 2.6133882999420166, "eval_runtime": 13.4996, "eval_samples_per_second": 7.408, "eval_steps_per_second": 7.408, "eval_token_acc": 0.4236024844720497, "step": 22500 }, { "epoch": 0.9641831969495737, "grad_norm": 10.396462440490723, "learning_rate": 9.110193781279635e-05, "loss": 0.3440162181854248, "memory(GiB)": 61.91, "step": 22505, "token_acc": 0.560106856634016, "train_speed(iter/s)": 1.466791 }, { "epoch": 0.9643974122788227, "grad_norm": 0.7376004457473755, "learning_rate": 9.109810529545171e-05, "loss": 0.4431018829345703, "memory(GiB)": 61.91, "step": 22510, "token_acc": 0.909433962264151, "train_speed(iter/s)": 1.466803 }, { "epoch": 0.9646116276080716, "grad_norm": 3.4292876720428467, "learning_rate": 9.109427203357632e-05, "loss": 0.32602872848510744, "memory(GiB)": 61.91, "step": 22515, "token_acc": 0.904320987654321, "train_speed(iter/s)": 1.466802 }, { "epoch": 0.9648258429373205, "grad_norm": 4.269575595855713, "learning_rate": 9.109043802723967e-05, "loss": 0.45380420684814454, "memory(GiB)": 61.91, "step": 22520, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.466815 }, { "epoch": 0.9650400582665696, "grad_norm": 3.65321946144104, "learning_rate": 9.108660327651116e-05, "loss": 0.3738923311233521, "memory(GiB)": 61.91, "step": 22525, "token_acc": 0.926530612244898, "train_speed(iter/s)": 1.466804 }, { "epoch": 0.9652542735958185, "grad_norm": 1.4909251928329468, "learning_rate": 9.10827677814603e-05, "loss": 0.14607330560684204, "memory(GiB)": 61.91, "step": 22530, "token_acc": 0.9619377162629758, "train_speed(iter/s)": 1.466817 }, { "epoch": 0.9654684889250674, "grad_norm": 0.9541783928871155, "learning_rate": 9.107893154215656e-05, "loss": 0.31940200328826907, "memory(GiB)": 61.91, "step": 22535, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.466855 }, { "epoch": 0.9656827042543165, "grad_norm": 5.5046281814575195, "learning_rate": 9.107509455866945e-05, "loss": 0.27488677501678466, "memory(GiB)": 61.91, "step": 22540, "token_acc": 0.9378698224852071, "train_speed(iter/s)": 1.466878 }, { "epoch": 0.9658969195835654, "grad_norm": 4.118765354156494, "learning_rate": 9.107125683106848e-05, "loss": 0.41463871002197267, "memory(GiB)": 61.91, "step": 22545, "token_acc": 0.8873720136518771, "train_speed(iter/s)": 1.4669 }, { "epoch": 0.9661111349128143, "grad_norm": 1.8921499252319336, "learning_rate": 9.106741835942314e-05, "loss": 0.32947168350219724, "memory(GiB)": 61.91, "step": 22550, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.466894 }, { "epoch": 0.9663253502420633, "grad_norm": 0.6701685786247253, "learning_rate": 9.106357914380299e-05, "loss": 0.321948766708374, "memory(GiB)": 61.91, "step": 22555, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.466918 }, { "epoch": 0.9665395655713123, "grad_norm": 0.5912522673606873, "learning_rate": 9.105973918427759e-05, "loss": 0.3507030487060547, "memory(GiB)": 61.91, "step": 22560, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.466945 }, { "epoch": 0.9667537809005613, "grad_norm": 3.008070230484009, "learning_rate": 9.105589848091651e-05, "loss": 0.5276268005371094, "memory(GiB)": 61.91, "step": 22565, "token_acc": 0.8951310861423221, "train_speed(iter/s)": 1.466947 }, { "epoch": 0.9669679962298102, "grad_norm": 5.073894023895264, "learning_rate": 9.105205703378931e-05, "loss": 0.3554256439208984, "memory(GiB)": 61.91, "step": 22570, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.466952 }, { "epoch": 0.9671822115590591, "grad_norm": 5.506412982940674, "learning_rate": 9.104821484296559e-05, "loss": 0.44730396270751954, "memory(GiB)": 61.91, "step": 22575, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.466998 }, { "epoch": 0.9673964268883082, "grad_norm": 6.124819278717041, "learning_rate": 9.104437190851493e-05, "loss": 0.542087459564209, "memory(GiB)": 61.91, "step": 22580, "token_acc": 0.88671875, "train_speed(iter/s)": 1.466991 }, { "epoch": 0.9676106422175571, "grad_norm": 1.4492192268371582, "learning_rate": 9.104052823050699e-05, "loss": 0.45547022819519045, "memory(GiB)": 61.91, "step": 22585, "token_acc": 0.9104938271604939, "train_speed(iter/s)": 1.466984 }, { "epoch": 0.967824857546806, "grad_norm": 4.878481864929199, "learning_rate": 9.103668380901138e-05, "loss": 0.3959322929382324, "memory(GiB)": 61.91, "step": 22590, "token_acc": 0.9118541033434651, "train_speed(iter/s)": 1.466953 }, { "epoch": 0.9680390728760551, "grad_norm": 4.652645111083984, "learning_rate": 9.103283864409775e-05, "loss": 0.37216198444366455, "memory(GiB)": 61.91, "step": 22595, "token_acc": 0.9346153846153846, "train_speed(iter/s)": 1.466962 }, { "epoch": 0.968253288205304, "grad_norm": 5.647254467010498, "learning_rate": 9.102899273583575e-05, "loss": 0.8148353576660157, "memory(GiB)": 61.91, "step": 22600, "token_acc": 0.8439716312056738, "train_speed(iter/s)": 1.466977 }, { "epoch": 0.9684675035345529, "grad_norm": 1.1464730501174927, "learning_rate": 9.102514608429507e-05, "loss": 0.3738710880279541, "memory(GiB)": 61.91, "step": 22605, "token_acc": 0.9020979020979021, "train_speed(iter/s)": 1.466988 }, { "epoch": 0.9686817188638019, "grad_norm": 3.6540586948394775, "learning_rate": 9.102129868954537e-05, "loss": 0.31359329223632815, "memory(GiB)": 61.91, "step": 22610, "token_acc": 0.953405017921147, "train_speed(iter/s)": 1.466989 }, { "epoch": 0.9688959341930509, "grad_norm": 0.45914551615715027, "learning_rate": 9.101745055165635e-05, "loss": 0.1510934591293335, "memory(GiB)": 61.91, "step": 22615, "token_acc": 0.9607843137254902, "train_speed(iter/s)": 1.466983 }, { "epoch": 0.9691101495222998, "grad_norm": 4.30312967300415, "learning_rate": 9.101360167069777e-05, "loss": 0.4194404125213623, "memory(GiB)": 61.91, "step": 22620, "token_acc": 0.9109792284866469, "train_speed(iter/s)": 1.466963 }, { "epoch": 0.9693243648515488, "grad_norm": 0.39689433574676514, "learning_rate": 9.100975204673929e-05, "loss": 0.16106451749801637, "memory(GiB)": 61.91, "step": 22625, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.466963 }, { "epoch": 0.9695385801807977, "grad_norm": 3.3761978149414062, "learning_rate": 9.10059016798507e-05, "loss": 0.42366819381713866, "memory(GiB)": 61.91, "step": 22630, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.46699 }, { "epoch": 0.9697527955100467, "grad_norm": 2.0866873264312744, "learning_rate": 9.100205057010174e-05, "loss": 0.21488714218139648, "memory(GiB)": 61.91, "step": 22635, "token_acc": 0.9558359621451105, "train_speed(iter/s)": 1.466979 }, { "epoch": 0.9699670108392957, "grad_norm": 2.843698740005493, "learning_rate": 9.099819871756215e-05, "loss": 0.8273063659667969, "memory(GiB)": 61.91, "step": 22640, "token_acc": 0.8344827586206897, "train_speed(iter/s)": 1.467028 }, { "epoch": 0.9701812261685446, "grad_norm": 2.5688352584838867, "learning_rate": 9.099434612230175e-05, "loss": 0.4158743381500244, "memory(GiB)": 61.91, "step": 22645, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.467062 }, { "epoch": 0.9703954414977936, "grad_norm": 2.3692450523376465, "learning_rate": 9.099049278439029e-05, "loss": 0.4909754753112793, "memory(GiB)": 61.91, "step": 22650, "token_acc": 0.9187279151943463, "train_speed(iter/s)": 1.467048 }, { "epoch": 0.9706096568270426, "grad_norm": 4.009530544281006, "learning_rate": 9.098663870389763e-05, "loss": 0.53192138671875, "memory(GiB)": 61.91, "step": 22655, "token_acc": 0.8753462603878116, "train_speed(iter/s)": 1.467048 }, { "epoch": 0.9708238721562915, "grad_norm": 0.4658305048942566, "learning_rate": 9.098278388089354e-05, "loss": 0.33517186641693114, "memory(GiB)": 61.91, "step": 22660, "token_acc": 0.9251101321585903, "train_speed(iter/s)": 1.467084 }, { "epoch": 0.9710380874855404, "grad_norm": 2.894105911254883, "learning_rate": 9.097892831544789e-05, "loss": 0.24039785861968993, "memory(GiB)": 61.91, "step": 22665, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.467081 }, { "epoch": 0.9712523028147895, "grad_norm": 2.1339268684387207, "learning_rate": 9.097507200763052e-05, "loss": 0.38661694526672363, "memory(GiB)": 61.91, "step": 22670, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.467099 }, { "epoch": 0.9714665181440384, "grad_norm": 2.4960134029388428, "learning_rate": 9.097121495751126e-05, "loss": 0.6261242389678955, "memory(GiB)": 61.91, "step": 22675, "token_acc": 0.8620689655172413, "train_speed(iter/s)": 1.467094 }, { "epoch": 0.9716807334732873, "grad_norm": 2.5292131900787354, "learning_rate": 9.096735716516001e-05, "loss": 0.37179586887359617, "memory(GiB)": 61.91, "step": 22680, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.467099 }, { "epoch": 0.9718949488025364, "grad_norm": 2.4643478393554688, "learning_rate": 9.096349863064666e-05, "loss": 0.7186147689819335, "memory(GiB)": 61.91, "step": 22685, "token_acc": 0.8371212121212122, "train_speed(iter/s)": 1.467192 }, { "epoch": 0.9721091641317853, "grad_norm": 3.2507870197296143, "learning_rate": 9.09596393540411e-05, "loss": 0.4861918926239014, "memory(GiB)": 61.91, "step": 22690, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.467188 }, { "epoch": 0.9723233794610342, "grad_norm": 4.054476737976074, "learning_rate": 9.095577933541326e-05, "loss": 0.31674144268035886, "memory(GiB)": 61.91, "step": 22695, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.4672 }, { "epoch": 0.9725375947902832, "grad_norm": 3.62605357170105, "learning_rate": 9.095191857483305e-05, "loss": 0.43697395324707033, "memory(GiB)": 61.91, "step": 22700, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.467203 }, { "epoch": 0.9727518101195322, "grad_norm": 4.2331013679504395, "learning_rate": 9.094805707237041e-05, "loss": 0.3778168439865112, "memory(GiB)": 61.91, "step": 22705, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.467221 }, { "epoch": 0.9729660254487811, "grad_norm": 5.261354923248291, "learning_rate": 9.094419482809534e-05, "loss": 0.7905144691467285, "memory(GiB)": 61.91, "step": 22710, "token_acc": 0.8440366972477065, "train_speed(iter/s)": 1.467262 }, { "epoch": 0.9731802407780301, "grad_norm": 3.8394057750701904, "learning_rate": 9.094033184207774e-05, "loss": 0.29835102558135984, "memory(GiB)": 61.91, "step": 22715, "token_acc": 0.9138755980861244, "train_speed(iter/s)": 1.46726 }, { "epoch": 0.973394456107279, "grad_norm": 3.8342833518981934, "learning_rate": 9.093646811438762e-05, "loss": 0.42450265884399413, "memory(GiB)": 61.91, "step": 22720, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.467265 }, { "epoch": 0.973608671436528, "grad_norm": 1.0380003452301025, "learning_rate": 9.0932603645095e-05, "loss": 0.4169100284576416, "memory(GiB)": 61.91, "step": 22725, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.467247 }, { "epoch": 0.973822886765777, "grad_norm": 4.247737407684326, "learning_rate": 9.092873843426986e-05, "loss": 0.5221052169799805, "memory(GiB)": 61.91, "step": 22730, "token_acc": 0.8945686900958466, "train_speed(iter/s)": 1.467249 }, { "epoch": 0.9740371020950259, "grad_norm": 3.9924652576446533, "learning_rate": 9.092487248198222e-05, "loss": 0.2715971231460571, "memory(GiB)": 61.91, "step": 22735, "token_acc": 0.9433198380566802, "train_speed(iter/s)": 1.467245 }, { "epoch": 0.9742513174242748, "grad_norm": 5.439665794372559, "learning_rate": 9.092100578830214e-05, "loss": 0.43852987289428713, "memory(GiB)": 61.91, "step": 22740, "token_acc": 0.9073482428115016, "train_speed(iter/s)": 1.467269 }, { "epoch": 0.9744655327535239, "grad_norm": 3.2469286918640137, "learning_rate": 9.091713835329964e-05, "loss": 0.5105165481567383, "memory(GiB)": 61.91, "step": 22745, "token_acc": 0.8933333333333333, "train_speed(iter/s)": 1.467261 }, { "epoch": 0.9746797480827728, "grad_norm": 3.2909281253814697, "learning_rate": 9.091327017704479e-05, "loss": 0.4127491474151611, "memory(GiB)": 61.91, "step": 22750, "token_acc": 0.9148264984227129, "train_speed(iter/s)": 1.467235 }, { "epoch": 0.9748939634120217, "grad_norm": 3.50614333152771, "learning_rate": 9.090940125960769e-05, "loss": 0.33447291851043703, "memory(GiB)": 61.91, "step": 22755, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.467224 }, { "epoch": 0.9751081787412708, "grad_norm": 3.6761868000030518, "learning_rate": 9.090553160105839e-05, "loss": 0.38726153373718264, "memory(GiB)": 61.91, "step": 22760, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.467228 }, { "epoch": 0.9753223940705197, "grad_norm": 2.542299509048462, "learning_rate": 9.090166120146702e-05, "loss": 0.31950788497924804, "memory(GiB)": 61.91, "step": 22765, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.467242 }, { "epoch": 0.9755366093997686, "grad_norm": 1.9300401210784912, "learning_rate": 9.08977900609037e-05, "loss": 0.5971720695495606, "memory(GiB)": 61.91, "step": 22770, "token_acc": 0.8586206896551725, "train_speed(iter/s)": 1.467296 }, { "epoch": 0.9757508247290176, "grad_norm": 2.1128714084625244, "learning_rate": 9.089391817943853e-05, "loss": 0.3215195894241333, "memory(GiB)": 61.91, "step": 22775, "token_acc": 0.9061371841155235, "train_speed(iter/s)": 1.467292 }, { "epoch": 0.9759650400582666, "grad_norm": 2.8073298931121826, "learning_rate": 9.089004555714168e-05, "loss": 0.3955368518829346, "memory(GiB)": 61.91, "step": 22780, "token_acc": 0.9063670411985019, "train_speed(iter/s)": 1.467313 }, { "epoch": 0.9761792553875155, "grad_norm": 5.319118499755859, "learning_rate": 9.08861721940833e-05, "loss": 0.5572036743164063, "memory(GiB)": 61.91, "step": 22785, "token_acc": 0.86875, "train_speed(iter/s)": 1.467327 }, { "epoch": 0.9763934707167645, "grad_norm": 0.07403390109539032, "learning_rate": 9.088229809033355e-05, "loss": 0.4871349334716797, "memory(GiB)": 61.91, "step": 22790, "token_acc": 0.875, "train_speed(iter/s)": 1.46736 }, { "epoch": 0.9766076860460134, "grad_norm": 4.482929229736328, "learning_rate": 9.087842324596262e-05, "loss": 0.4333138942718506, "memory(GiB)": 61.91, "step": 22795, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.467365 }, { "epoch": 0.9768219013752624, "grad_norm": 2.1911461353302, "learning_rate": 9.087454766104071e-05, "loss": 0.253869366645813, "memory(GiB)": 61.91, "step": 22800, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.467393 }, { "epoch": 0.9770361167045114, "grad_norm": 0.11205966025590897, "learning_rate": 9.087067133563803e-05, "loss": 0.2503319263458252, "memory(GiB)": 61.91, "step": 22805, "token_acc": 0.9317269076305221, "train_speed(iter/s)": 1.467368 }, { "epoch": 0.9772503320337603, "grad_norm": 2.9930312633514404, "learning_rate": 9.086679426982479e-05, "loss": 0.44928412437438964, "memory(GiB)": 61.91, "step": 22810, "token_acc": 0.9085173501577287, "train_speed(iter/s)": 1.467359 }, { "epoch": 0.9774645473630093, "grad_norm": 4.701198101043701, "learning_rate": 9.086291646367123e-05, "loss": 0.3456684350967407, "memory(GiB)": 61.91, "step": 22815, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.467368 }, { "epoch": 0.9776787626922583, "grad_norm": 4.550952434539795, "learning_rate": 9.085903791724761e-05, "loss": 0.6531576156616211, "memory(GiB)": 61.91, "step": 22820, "token_acc": 0.8565737051792829, "train_speed(iter/s)": 1.467435 }, { "epoch": 0.9778929780215072, "grad_norm": 1.775841474533081, "learning_rate": 9.085515863062419e-05, "loss": 0.547609519958496, "memory(GiB)": 61.91, "step": 22825, "token_acc": 0.8595890410958904, "train_speed(iter/s)": 1.467438 }, { "epoch": 0.9781071933507561, "grad_norm": 4.896491050720215, "learning_rate": 9.085127860387126e-05, "loss": 0.5056403160095215, "memory(GiB)": 61.91, "step": 22830, "token_acc": 0.9036144578313253, "train_speed(iter/s)": 1.467417 }, { "epoch": 0.9783214086800052, "grad_norm": 4.149960994720459, "learning_rate": 9.084739783705909e-05, "loss": 0.32451980113983153, "memory(GiB)": 61.91, "step": 22835, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.46741 }, { "epoch": 0.9785356240092541, "grad_norm": 3.365799903869629, "learning_rate": 9.084351633025798e-05, "loss": 0.3388540744781494, "memory(GiB)": 61.91, "step": 22840, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.4674 }, { "epoch": 0.978749839338503, "grad_norm": 1.8322138786315918, "learning_rate": 9.083963408353825e-05, "loss": 0.1649320125579834, "memory(GiB)": 61.91, "step": 22845, "token_acc": 0.9689922480620154, "train_speed(iter/s)": 1.467407 }, { "epoch": 0.978964054667752, "grad_norm": 4.165931701660156, "learning_rate": 9.083575109697027e-05, "loss": 0.331998872756958, "memory(GiB)": 61.91, "step": 22850, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.467384 }, { "epoch": 0.979178269997001, "grad_norm": 3.4234819412231445, "learning_rate": 9.083186737062432e-05, "loss": 0.346051287651062, "memory(GiB)": 61.91, "step": 22855, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.467369 }, { "epoch": 0.9793924853262499, "grad_norm": 2.548560857772827, "learning_rate": 9.082798290457081e-05, "loss": 0.29793710708618165, "memory(GiB)": 61.91, "step": 22860, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.467365 }, { "epoch": 0.9796067006554989, "grad_norm": 2.5122900009155273, "learning_rate": 9.082409769888008e-05, "loss": 0.4655618190765381, "memory(GiB)": 61.91, "step": 22865, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.467374 }, { "epoch": 0.9798209159847479, "grad_norm": 4.179500102996826, "learning_rate": 9.082021175362252e-05, "loss": 0.4870340347290039, "memory(GiB)": 61.91, "step": 22870, "token_acc": 0.8793103448275862, "train_speed(iter/s)": 1.467377 }, { "epoch": 0.9800351313139968, "grad_norm": 1.5908443927764893, "learning_rate": 9.081632506886854e-05, "loss": 0.4044306755065918, "memory(GiB)": 61.91, "step": 22875, "token_acc": 0.9190031152647975, "train_speed(iter/s)": 1.467397 }, { "epoch": 0.9802493466432458, "grad_norm": 2.0280659198760986, "learning_rate": 9.081243764468854e-05, "loss": 0.4685558795928955, "memory(GiB)": 61.91, "step": 22880, "token_acc": 0.9017857142857143, "train_speed(iter/s)": 1.467408 }, { "epoch": 0.9804635619724947, "grad_norm": 2.4970626831054688, "learning_rate": 9.080854948115295e-05, "loss": 0.34035229682922363, "memory(GiB)": 61.91, "step": 22885, "token_acc": 0.9304635761589404, "train_speed(iter/s)": 1.467427 }, { "epoch": 0.9806777773017437, "grad_norm": 6.476522922515869, "learning_rate": 9.080466057833221e-05, "loss": 0.5590397834777832, "memory(GiB)": 61.91, "step": 22890, "token_acc": 0.9020408163265307, "train_speed(iter/s)": 1.467421 }, { "epoch": 0.9808919926309927, "grad_norm": 3.67911434173584, "learning_rate": 9.080077093629675e-05, "loss": 0.1436244010925293, "memory(GiB)": 61.91, "step": 22895, "token_acc": 0.9598540145985401, "train_speed(iter/s)": 1.467422 }, { "epoch": 0.9811062079602416, "grad_norm": 4.9731268882751465, "learning_rate": 9.079688055511707e-05, "loss": 0.3296573877334595, "memory(GiB)": 61.91, "step": 22900, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.467449 }, { "epoch": 0.9813204232894907, "grad_norm": 2.9628140926361084, "learning_rate": 9.079298943486361e-05, "loss": 0.49567241668701173, "memory(GiB)": 61.91, "step": 22905, "token_acc": 0.8932038834951457, "train_speed(iter/s)": 1.467451 }, { "epoch": 0.9815346386187396, "grad_norm": 3.39810848236084, "learning_rate": 9.078909757560687e-05, "loss": 0.2028059720993042, "memory(GiB)": 61.91, "step": 22910, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.467453 }, { "epoch": 0.9817488539479885, "grad_norm": 3.356431722640991, "learning_rate": 9.07852049774174e-05, "loss": 0.46726598739624026, "memory(GiB)": 61.91, "step": 22915, "token_acc": 0.8844884488448845, "train_speed(iter/s)": 1.467524 }, { "epoch": 0.9819630692772375, "grad_norm": 2.2710084915161133, "learning_rate": 9.078131164036565e-05, "loss": 0.3028480291366577, "memory(GiB)": 61.91, "step": 22920, "token_acc": 0.9375, "train_speed(iter/s)": 1.467542 }, { "epoch": 0.9821772846064865, "grad_norm": 2.8824594020843506, "learning_rate": 9.07774175645222e-05, "loss": 0.6599407196044922, "memory(GiB)": 61.91, "step": 22925, "token_acc": 0.8934707903780069, "train_speed(iter/s)": 1.467535 }, { "epoch": 0.9823914999357354, "grad_norm": 5.740154266357422, "learning_rate": 9.077352274995757e-05, "loss": 0.5597051620483399, "memory(GiB)": 61.91, "step": 22930, "token_acc": 0.8806584362139918, "train_speed(iter/s)": 1.467518 }, { "epoch": 0.9826057152649844, "grad_norm": 2.952100992202759, "learning_rate": 9.076962719674233e-05, "loss": 0.23585197925567628, "memory(GiB)": 61.91, "step": 22935, "token_acc": 0.9423868312757202, "train_speed(iter/s)": 1.46753 }, { "epoch": 0.9828199305942333, "grad_norm": 4.040224552154541, "learning_rate": 9.076573090494704e-05, "loss": 0.49897093772888185, "memory(GiB)": 61.91, "step": 22940, "token_acc": 0.900709219858156, "train_speed(iter/s)": 1.467527 }, { "epoch": 0.9830341459234823, "grad_norm": 1.7992702722549438, "learning_rate": 9.076183387464232e-05, "loss": 0.34796526432037356, "memory(GiB)": 61.91, "step": 22945, "token_acc": 0.9423868312757202, "train_speed(iter/s)": 1.467535 }, { "epoch": 0.9832483612527313, "grad_norm": 4.571271896362305, "learning_rate": 9.075793610589871e-05, "loss": 0.35993621349334715, "memory(GiB)": 61.91, "step": 22950, "token_acc": 0.9066147859922179, "train_speed(iter/s)": 1.467526 }, { "epoch": 0.9834625765819802, "grad_norm": 3.348897695541382, "learning_rate": 9.075403759878687e-05, "loss": 0.42169771194458006, "memory(GiB)": 61.91, "step": 22955, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.467504 }, { "epoch": 0.9836767919112291, "grad_norm": 4.686835289001465, "learning_rate": 9.075013835337742e-05, "loss": 0.46544132232666013, "memory(GiB)": 61.91, "step": 22960, "token_acc": 0.9112627986348123, "train_speed(iter/s)": 1.467518 }, { "epoch": 0.9838910072404782, "grad_norm": 9.308765411376953, "learning_rate": 9.074623836974097e-05, "loss": 0.2516660213470459, "memory(GiB)": 61.91, "step": 22965, "token_acc": 0.9507575757575758, "train_speed(iter/s)": 1.467536 }, { "epoch": 0.9841052225697271, "grad_norm": 5.000901699066162, "learning_rate": 9.074233764794818e-05, "loss": 0.5643965721130371, "memory(GiB)": 61.91, "step": 22970, "token_acc": 0.881578947368421, "train_speed(iter/s)": 1.467553 }, { "epoch": 0.984319437898976, "grad_norm": 2.312708616256714, "learning_rate": 9.073843618806974e-05, "loss": 0.3907627582550049, "memory(GiB)": 61.91, "step": 22975, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.467559 }, { "epoch": 0.9845336532282251, "grad_norm": 5.159397125244141, "learning_rate": 9.073453399017631e-05, "loss": 0.3154179334640503, "memory(GiB)": 61.91, "step": 22980, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.467533 }, { "epoch": 0.984747868557474, "grad_norm": 3.7507987022399902, "learning_rate": 9.073063105433859e-05, "loss": 0.46868071556091306, "memory(GiB)": 61.91, "step": 22985, "token_acc": 0.9075342465753424, "train_speed(iter/s)": 1.467525 }, { "epoch": 0.9849620838867229, "grad_norm": 3.410108804702759, "learning_rate": 9.072672738062726e-05, "loss": 0.3694835424423218, "memory(GiB)": 61.91, "step": 22990, "token_acc": 0.9241706161137441, "train_speed(iter/s)": 1.467538 }, { "epoch": 0.9851762992159719, "grad_norm": 5.171638488769531, "learning_rate": 9.072282296911308e-05, "loss": 0.4230008125305176, "memory(GiB)": 61.91, "step": 22995, "token_acc": 0.9131944444444444, "train_speed(iter/s)": 1.467543 }, { "epoch": 0.9853905145452209, "grad_norm": 3.866205930709839, "learning_rate": 9.071891781986675e-05, "loss": 0.3414628267288208, "memory(GiB)": 61.91, "step": 23000, "token_acc": 0.924812030075188, "train_speed(iter/s)": 1.467528 }, { "epoch": 0.9853905145452209, "eval_loss": 2.8018884658813477, "eval_runtime": 13.9523, "eval_samples_per_second": 7.167, "eval_steps_per_second": 7.167, "eval_token_acc": 0.4074074074074074, "step": 23000 }, { "epoch": 0.9856047298744698, "grad_norm": 4.5356316566467285, "learning_rate": 9.071501193295903e-05, "loss": 0.6438232421875, "memory(GiB)": 61.91, "step": 23005, "token_acc": 0.5313688212927756, "train_speed(iter/s)": 1.466173 }, { "epoch": 0.9858189452037188, "grad_norm": 4.7228569984436035, "learning_rate": 9.071110530846067e-05, "loss": 0.71490478515625, "memory(GiB)": 61.91, "step": 23010, "token_acc": 0.8690095846645367, "train_speed(iter/s)": 1.46617 }, { "epoch": 0.9860331605329677, "grad_norm": 1.2189868688583374, "learning_rate": 9.070719794644245e-05, "loss": 0.40151104927062986, "memory(GiB)": 61.91, "step": 23015, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.466177 }, { "epoch": 0.9862473758622167, "grad_norm": 3.7882189750671387, "learning_rate": 9.070328984697516e-05, "loss": 0.47653851509094236, "memory(GiB)": 61.91, "step": 23020, "token_acc": 0.896, "train_speed(iter/s)": 1.466186 }, { "epoch": 0.9864615911914657, "grad_norm": 3.042577028274536, "learning_rate": 9.069938101012958e-05, "loss": 0.28176016807556153, "memory(GiB)": 61.91, "step": 23025, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.466173 }, { "epoch": 0.9866758065207146, "grad_norm": 3.0926737785339355, "learning_rate": 9.069547143597655e-05, "loss": 0.20814967155456543, "memory(GiB)": 61.91, "step": 23030, "token_acc": 0.9616724738675958, "train_speed(iter/s)": 1.466174 }, { "epoch": 0.9868900218499636, "grad_norm": 5.371574401855469, "learning_rate": 9.069156112458685e-05, "loss": 0.354787540435791, "memory(GiB)": 61.91, "step": 23035, "token_acc": 0.9294478527607362, "train_speed(iter/s)": 1.46616 }, { "epoch": 0.9871042371792126, "grad_norm": 2.7587740421295166, "learning_rate": 9.068765007603137e-05, "loss": 0.38357689380645754, "memory(GiB)": 61.91, "step": 23040, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.466146 }, { "epoch": 0.9873184525084615, "grad_norm": 2.922731876373291, "learning_rate": 9.068373829038095e-05, "loss": 0.3913354635238647, "memory(GiB)": 61.91, "step": 23045, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.466146 }, { "epoch": 0.9875326678377104, "grad_norm": 4.297885417938232, "learning_rate": 9.067982576770644e-05, "loss": 0.20912642478942872, "memory(GiB)": 61.91, "step": 23050, "token_acc": 0.9506849315068493, "train_speed(iter/s)": 1.466137 }, { "epoch": 0.9877468831669595, "grad_norm": 4.305956840515137, "learning_rate": 9.067591250807872e-05, "loss": 0.22772567272186278, "memory(GiB)": 61.91, "step": 23055, "token_acc": 0.9421221864951769, "train_speed(iter/s)": 1.466132 }, { "epoch": 0.9879610984962084, "grad_norm": 3.4289517402648926, "learning_rate": 9.067199851156869e-05, "loss": 0.35645499229431155, "memory(GiB)": 61.91, "step": 23060, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.466137 }, { "epoch": 0.9881753138254573, "grad_norm": 3.0420477390289307, "learning_rate": 9.066808377824725e-05, "loss": 0.28496618270874025, "memory(GiB)": 61.91, "step": 23065, "token_acc": 0.943952802359882, "train_speed(iter/s)": 1.466147 }, { "epoch": 0.9883895291547063, "grad_norm": 7.919363021850586, "learning_rate": 9.066416830818531e-05, "loss": 0.3788649082183838, "memory(GiB)": 61.91, "step": 23070, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.46618 }, { "epoch": 0.9886037444839553, "grad_norm": 1.373322606086731, "learning_rate": 9.066025210145384e-05, "loss": 0.43091492652893065, "memory(GiB)": 61.91, "step": 23075, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.466165 }, { "epoch": 0.9888179598132042, "grad_norm": 2.3561184406280518, "learning_rate": 9.065633515812376e-05, "loss": 0.31202499866485595, "memory(GiB)": 61.91, "step": 23080, "token_acc": 0.92, "train_speed(iter/s)": 1.466164 }, { "epoch": 0.9890321751424532, "grad_norm": 3.0452964305877686, "learning_rate": 9.0652417478266e-05, "loss": 0.3210011959075928, "memory(GiB)": 61.91, "step": 23085, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.466183 }, { "epoch": 0.9892463904717022, "grad_norm": 2.93115496635437, "learning_rate": 9.064849906195159e-05, "loss": 0.4121109485626221, "memory(GiB)": 61.91, "step": 23090, "token_acc": 0.9107142857142857, "train_speed(iter/s)": 1.466165 }, { "epoch": 0.9894606058009511, "grad_norm": 5.3507208824157715, "learning_rate": 9.064457990925149e-05, "loss": 0.41944360733032227, "memory(GiB)": 61.91, "step": 23095, "token_acc": 0.890625, "train_speed(iter/s)": 1.466149 }, { "epoch": 0.9896748211302001, "grad_norm": 2.316462278366089, "learning_rate": 9.064066002023668e-05, "loss": 0.2755529165267944, "memory(GiB)": 61.91, "step": 23100, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.466157 }, { "epoch": 0.989889036459449, "grad_norm": 2.3346195220947266, "learning_rate": 9.06367393949782e-05, "loss": 0.3700258731842041, "memory(GiB)": 61.91, "step": 23105, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.466141 }, { "epoch": 0.990103251788698, "grad_norm": 2.1577906608581543, "learning_rate": 9.063281803354707e-05, "loss": 0.2636056661605835, "memory(GiB)": 61.91, "step": 23110, "token_acc": 0.936, "train_speed(iter/s)": 1.466153 }, { "epoch": 0.990317467117947, "grad_norm": 1.8787915706634521, "learning_rate": 9.062889593601432e-05, "loss": 0.27724628448486327, "memory(GiB)": 61.91, "step": 23115, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.466134 }, { "epoch": 0.9905316824471959, "grad_norm": 2.152647018432617, "learning_rate": 9.0624973102451e-05, "loss": 0.4210536003112793, "memory(GiB)": 61.91, "step": 23120, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.466139 }, { "epoch": 0.9907458977764448, "grad_norm": 2.9005849361419678, "learning_rate": 9.062104953292819e-05, "loss": 0.4311054229736328, "memory(GiB)": 61.91, "step": 23125, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.466151 }, { "epoch": 0.9909601131056939, "grad_norm": 3.3991777896881104, "learning_rate": 9.061712522751696e-05, "loss": 0.3234874725341797, "memory(GiB)": 61.91, "step": 23130, "token_acc": 0.9451612903225807, "train_speed(iter/s)": 1.466148 }, { "epoch": 0.9911743284349428, "grad_norm": 4.2745137214660645, "learning_rate": 9.06132001862884e-05, "loss": 0.30898351669311525, "memory(GiB)": 61.91, "step": 23135, "token_acc": 0.934640522875817, "train_speed(iter/s)": 1.466201 }, { "epoch": 0.9913885437641917, "grad_norm": 5.174011707305908, "learning_rate": 9.060927440931362e-05, "loss": 0.5215951919555664, "memory(GiB)": 61.91, "step": 23140, "token_acc": 0.901840490797546, "train_speed(iter/s)": 1.466249 }, { "epoch": 0.9916027590934408, "grad_norm": 4.024851322174072, "learning_rate": 9.060534789666374e-05, "loss": 0.23213725090026854, "memory(GiB)": 61.91, "step": 23145, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.466235 }, { "epoch": 0.9918169744226897, "grad_norm": 5.033786296844482, "learning_rate": 9.06014206484099e-05, "loss": 0.3277372121810913, "memory(GiB)": 61.91, "step": 23150, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.466229 }, { "epoch": 0.9920311897519386, "grad_norm": 4.290957450866699, "learning_rate": 9.059749266462324e-05, "loss": 0.48192148208618163, "memory(GiB)": 61.91, "step": 23155, "token_acc": 0.891566265060241, "train_speed(iter/s)": 1.466252 }, { "epoch": 0.9922454050811876, "grad_norm": 4.1790900230407715, "learning_rate": 9.05935639453749e-05, "loss": 0.20256810188293456, "memory(GiB)": 61.91, "step": 23160, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.466233 }, { "epoch": 0.9924596204104366, "grad_norm": 3.9452850818634033, "learning_rate": 9.058963449073607e-05, "loss": 0.3903980255126953, "memory(GiB)": 61.91, "step": 23165, "token_acc": 0.9295774647887324, "train_speed(iter/s)": 1.466248 }, { "epoch": 0.9926738357396855, "grad_norm": 2.163417339324951, "learning_rate": 9.058570430077795e-05, "loss": 0.5196821689605713, "memory(GiB)": 61.91, "step": 23170, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.466297 }, { "epoch": 0.9928880510689345, "grad_norm": 5.080713272094727, "learning_rate": 9.058177337557172e-05, "loss": 0.6870219230651855, "memory(GiB)": 61.91, "step": 23175, "token_acc": 0.8442367601246106, "train_speed(iter/s)": 1.466316 }, { "epoch": 0.9931022663981834, "grad_norm": 3.430781126022339, "learning_rate": 9.057784171518861e-05, "loss": 0.3689839839935303, "memory(GiB)": 61.91, "step": 23180, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.466285 }, { "epoch": 0.9933164817274324, "grad_norm": 4.090959072113037, "learning_rate": 9.057390931969981e-05, "loss": 0.4228992462158203, "memory(GiB)": 61.91, "step": 23185, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.466275 }, { "epoch": 0.9935306970566814, "grad_norm": 4.610325336456299, "learning_rate": 9.056997618917659e-05, "loss": 0.5709116458892822, "memory(GiB)": 61.91, "step": 23190, "token_acc": 0.8685258964143426, "train_speed(iter/s)": 1.466298 }, { "epoch": 0.9937449123859303, "grad_norm": 5.033995628356934, "learning_rate": 9.056604232369019e-05, "loss": 0.6102637767791748, "memory(GiB)": 61.91, "step": 23195, "token_acc": 0.8714859437751004, "train_speed(iter/s)": 1.466282 }, { "epoch": 0.9939591277151792, "grad_norm": 3.39082407951355, "learning_rate": 9.056210772331188e-05, "loss": 0.7244359493255615, "memory(GiB)": 61.91, "step": 23200, "token_acc": 0.875, "train_speed(iter/s)": 1.466274 }, { "epoch": 0.9941733430444283, "grad_norm": 1.8970284461975098, "learning_rate": 9.055817238811295e-05, "loss": 0.5684709072113037, "memory(GiB)": 61.91, "step": 23205, "token_acc": 0.8770226537216829, "train_speed(iter/s)": 1.466281 }, { "epoch": 0.9943875583736772, "grad_norm": 3.8140368461608887, "learning_rate": 9.055423631816466e-05, "loss": 0.3557960033416748, "memory(GiB)": 61.91, "step": 23210, "token_acc": 0.9201183431952663, "train_speed(iter/s)": 1.466277 }, { "epoch": 0.9946017737029261, "grad_norm": 3.936457633972168, "learning_rate": 9.055029951353835e-05, "loss": 0.4290026664733887, "memory(GiB)": 61.91, "step": 23215, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.466298 }, { "epoch": 0.9948159890321752, "grad_norm": 4.572265625, "learning_rate": 9.054636197430533e-05, "loss": 0.41004114151000975, "memory(GiB)": 61.91, "step": 23220, "token_acc": 0.8862068965517241, "train_speed(iter/s)": 1.4663 }, { "epoch": 0.9950302043614241, "grad_norm": 4.204168796539307, "learning_rate": 9.054242370053691e-05, "loss": 0.43382911682128905, "memory(GiB)": 61.91, "step": 23225, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.466288 }, { "epoch": 0.995244419690673, "grad_norm": 1.5903034210205078, "learning_rate": 9.053848469230446e-05, "loss": 0.3283071041107178, "memory(GiB)": 61.91, "step": 23230, "token_acc": 0.9223880597014925, "train_speed(iter/s)": 1.466292 }, { "epoch": 0.995458635019922, "grad_norm": 4.326785087585449, "learning_rate": 9.053454494967935e-05, "loss": 0.49473347663879397, "memory(GiB)": 61.91, "step": 23235, "token_acc": 0.876984126984127, "train_speed(iter/s)": 1.46628 }, { "epoch": 0.995672850349171, "grad_norm": 5.735963821411133, "learning_rate": 9.053060447273291e-05, "loss": 0.5998233795166016, "memory(GiB)": 61.91, "step": 23240, "token_acc": 0.8525179856115108, "train_speed(iter/s)": 1.466261 }, { "epoch": 0.99588706567842, "grad_norm": 4.444169521331787, "learning_rate": 9.052666326153656e-05, "loss": 0.4996207237243652, "memory(GiB)": 61.91, "step": 23245, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.46626 }, { "epoch": 0.9961012810076689, "grad_norm": 3.0266075134277344, "learning_rate": 9.052272131616168e-05, "loss": 0.38396737575531004, "memory(GiB)": 61.91, "step": 23250, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.466251 }, { "epoch": 0.9963154963369178, "grad_norm": 3.2546558380126953, "learning_rate": 9.051877863667969e-05, "loss": 0.5909715175628663, "memory(GiB)": 61.91, "step": 23255, "token_acc": 0.8835341365461847, "train_speed(iter/s)": 1.466246 }, { "epoch": 0.9965297116661669, "grad_norm": 1.6458323001861572, "learning_rate": 9.051483522316202e-05, "loss": 0.19717875719070435, "memory(GiB)": 61.91, "step": 23260, "token_acc": 0.9488054607508533, "train_speed(iter/s)": 1.466259 }, { "epoch": 0.9967439269954158, "grad_norm": 5.961434841156006, "learning_rate": 9.05108910756801e-05, "loss": 0.4791823387145996, "memory(GiB)": 61.91, "step": 23265, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.466288 }, { "epoch": 0.9969581423246647, "grad_norm": 2.838625431060791, "learning_rate": 9.050694619430539e-05, "loss": 0.46123476028442384, "memory(GiB)": 61.91, "step": 23270, "token_acc": 0.9036544850498339, "train_speed(iter/s)": 1.466283 }, { "epoch": 0.9971723576539138, "grad_norm": 3.1425623893737793, "learning_rate": 9.050300057910936e-05, "loss": 0.3685252904891968, "memory(GiB)": 61.91, "step": 23275, "token_acc": 0.936, "train_speed(iter/s)": 1.466277 }, { "epoch": 0.9973865729831627, "grad_norm": 4.533308506011963, "learning_rate": 9.049905423016347e-05, "loss": 0.5994612693786621, "memory(GiB)": 61.91, "step": 23280, "token_acc": 0.8484848484848485, "train_speed(iter/s)": 1.466302 }, { "epoch": 0.9976007883124116, "grad_norm": 3.043637990951538, "learning_rate": 9.049510714753922e-05, "loss": 0.49548187255859377, "memory(GiB)": 61.91, "step": 23285, "token_acc": 0.9113149847094801, "train_speed(iter/s)": 1.466302 }, { "epoch": 0.9978150036416606, "grad_norm": 2.361222505569458, "learning_rate": 9.049115933130811e-05, "loss": 0.3626401424407959, "memory(GiB)": 61.91, "step": 23290, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.466278 }, { "epoch": 0.9980292189709096, "grad_norm": 3.5664327144622803, "learning_rate": 9.048721078154168e-05, "loss": 0.21756746768951415, "memory(GiB)": 61.91, "step": 23295, "token_acc": 0.9554794520547946, "train_speed(iter/s)": 1.466275 }, { "epoch": 0.9982434343001585, "grad_norm": 1.6589657068252563, "learning_rate": 9.048326149831143e-05, "loss": 0.20720133781433106, "memory(GiB)": 61.91, "step": 23300, "token_acc": 0.9413793103448276, "train_speed(iter/s)": 1.466268 }, { "epoch": 0.9984576496294075, "grad_norm": 6.977700710296631, "learning_rate": 9.047931148168894e-05, "loss": 0.4335311412811279, "memory(GiB)": 61.91, "step": 23305, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.466252 }, { "epoch": 0.9986718649586565, "grad_norm": 2.216754674911499, "learning_rate": 9.047536073174573e-05, "loss": 0.41321487426757814, "memory(GiB)": 61.91, "step": 23310, "token_acc": 0.9161490683229814, "train_speed(iter/s)": 1.466222 }, { "epoch": 0.9988860802879054, "grad_norm": 1.8785909414291382, "learning_rate": 9.047140924855342e-05, "loss": 0.6249926567077637, "memory(GiB)": 61.91, "step": 23315, "token_acc": 0.8746268656716418, "train_speed(iter/s)": 1.466214 }, { "epoch": 0.9991002956171544, "grad_norm": 1.247135043144226, "learning_rate": 9.046745703218356e-05, "loss": 0.17793688774108887, "memory(GiB)": 61.91, "step": 23320, "token_acc": 0.96, "train_speed(iter/s)": 1.466266 }, { "epoch": 0.9993145109464033, "grad_norm": 3.6000959873199463, "learning_rate": 9.046350408270772e-05, "loss": 0.37376389503479, "memory(GiB)": 61.91, "step": 23325, "token_acc": 0.926984126984127, "train_speed(iter/s)": 1.466274 }, { "epoch": 0.9995287262756523, "grad_norm": 1.988287091255188, "learning_rate": 9.045955040019758e-05, "loss": 0.46655926704406736, "memory(GiB)": 61.91, "step": 23330, "token_acc": 0.9087719298245615, "train_speed(iter/s)": 1.466243 }, { "epoch": 0.9997429416049013, "grad_norm": 2.605344295501709, "learning_rate": 9.045559598472472e-05, "loss": 0.3812110900878906, "memory(GiB)": 61.91, "step": 23335, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.466272 }, { "epoch": 0.9999571569341502, "grad_norm": 6.514379978179932, "learning_rate": 9.045164083636079e-05, "loss": 0.5162374973297119, "memory(GiB)": 61.91, "step": 23340, "token_acc": 0.8844984802431611, "train_speed(iter/s)": 1.466289 }, { "epoch": 1.0001713722633991, "grad_norm": 0.22389034926891327, "learning_rate": 9.044768495517744e-05, "loss": 0.43283467292785643, "memory(GiB)": 61.91, "step": 23345, "token_acc": 0.9097744360902256, "train_speed(iter/s)": 1.466301 }, { "epoch": 1.0003855875926482, "grad_norm": 3.21712327003479, "learning_rate": 9.044372834124632e-05, "loss": 0.7594629287719726, "memory(GiB)": 61.91, "step": 23350, "token_acc": 0.8366013071895425, "train_speed(iter/s)": 1.466303 }, { "epoch": 1.000599802921897, "grad_norm": 2.753504991531372, "learning_rate": 9.043977099463914e-05, "loss": 0.4891964912414551, "memory(GiB)": 61.91, "step": 23355, "token_acc": 0.9170854271356784, "train_speed(iter/s)": 1.466294 }, { "epoch": 1.000814018251146, "grad_norm": 3.037426710128784, "learning_rate": 9.043581291542757e-05, "loss": 0.41431436538696287, "memory(GiB)": 61.91, "step": 23360, "token_acc": 0.934375, "train_speed(iter/s)": 1.466277 }, { "epoch": 1.001028233580395, "grad_norm": 2.8651130199432373, "learning_rate": 9.043185410368332e-05, "loss": 0.6977530479431152, "memory(GiB)": 61.91, "step": 23365, "token_acc": 0.8633093525179856, "train_speed(iter/s)": 1.466266 }, { "epoch": 1.0012424489096439, "grad_norm": 2.355426788330078, "learning_rate": 9.042789455947808e-05, "loss": 0.2588301181793213, "memory(GiB)": 61.91, "step": 23370, "token_acc": 0.9509433962264151, "train_speed(iter/s)": 1.466276 }, { "epoch": 1.001456664238893, "grad_norm": 6.826366424560547, "learning_rate": 9.042393428288363e-05, "loss": 0.4513858795166016, "memory(GiB)": 61.91, "step": 23375, "token_acc": 0.8942307692307693, "train_speed(iter/s)": 1.466295 }, { "epoch": 1.001670879568142, "grad_norm": 3.9340813159942627, "learning_rate": 9.041997327397169e-05, "loss": 0.3208312511444092, "memory(GiB)": 61.91, "step": 23380, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.466336 }, { "epoch": 1.0018850948973907, "grad_norm": 1.9665848016738892, "learning_rate": 9.0416011532814e-05, "loss": 0.3248921871185303, "memory(GiB)": 61.91, "step": 23385, "token_acc": 0.9271137026239067, "train_speed(iter/s)": 1.466327 }, { "epoch": 1.0020993102266398, "grad_norm": 4.530271053314209, "learning_rate": 9.041204905948236e-05, "loss": 0.4966284275054932, "memory(GiB)": 61.91, "step": 23390, "token_acc": 0.8920634920634921, "train_speed(iter/s)": 1.466337 }, { "epoch": 1.0023135255558888, "grad_norm": 4.857135772705078, "learning_rate": 9.040808585404854e-05, "loss": 0.6355748653411866, "memory(GiB)": 61.91, "step": 23395, "token_acc": 0.8414239482200647, "train_speed(iter/s)": 1.466356 }, { "epoch": 1.0025277408851376, "grad_norm": 1.4380847215652466, "learning_rate": 9.040412191658434e-05, "loss": 0.24257736206054686, "memory(GiB)": 61.91, "step": 23400, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.466342 }, { "epoch": 1.0027419562143867, "grad_norm": 1.4551470279693604, "learning_rate": 9.040015724716157e-05, "loss": 0.23666529655456542, "memory(GiB)": 61.91, "step": 23405, "token_acc": 0.9718875502008032, "train_speed(iter/s)": 1.466349 }, { "epoch": 1.0029561715436357, "grad_norm": 4.318110942840576, "learning_rate": 9.039619184585204e-05, "loss": 0.46571621894836424, "memory(GiB)": 61.91, "step": 23410, "token_acc": 0.9125, "train_speed(iter/s)": 1.466345 }, { "epoch": 1.0031703868728847, "grad_norm": 3.5381014347076416, "learning_rate": 9.039222571272763e-05, "loss": 0.5022389411926269, "memory(GiB)": 61.91, "step": 23415, "token_acc": 0.8566433566433567, "train_speed(iter/s)": 1.466355 }, { "epoch": 1.0033846022021335, "grad_norm": 6.335610389709473, "learning_rate": 9.038825884786013e-05, "loss": 0.5737714767456055, "memory(GiB)": 61.91, "step": 23420, "token_acc": 0.8526315789473684, "train_speed(iter/s)": 1.466385 }, { "epoch": 1.0035988175313826, "grad_norm": 2.6478395462036133, "learning_rate": 9.038429125132143e-05, "loss": 0.27043585777282714, "memory(GiB)": 61.91, "step": 23425, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.466374 }, { "epoch": 1.0038130328606316, "grad_norm": 3.0995121002197266, "learning_rate": 9.038032292318343e-05, "loss": 0.3798614501953125, "memory(GiB)": 61.91, "step": 23430, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.466426 }, { "epoch": 1.0040272481898804, "grad_norm": 2.3428080081939697, "learning_rate": 9.037635386351801e-05, "loss": 0.29378395080566405, "memory(GiB)": 61.91, "step": 23435, "token_acc": 0.9235474006116208, "train_speed(iter/s)": 1.466436 }, { "epoch": 1.0042414635191295, "grad_norm": 3.297528028488159, "learning_rate": 9.037238407239705e-05, "loss": 0.39637892246246337, "memory(GiB)": 61.91, "step": 23440, "token_acc": 0.8985074626865671, "train_speed(iter/s)": 1.46643 }, { "epoch": 1.0044556788483785, "grad_norm": 0.13324862718582153, "learning_rate": 9.036841354989248e-05, "loss": 0.2587049722671509, "memory(GiB)": 61.91, "step": 23445, "token_acc": 0.9469387755102041, "train_speed(iter/s)": 1.466443 }, { "epoch": 1.0046698941776273, "grad_norm": 2.8279895782470703, "learning_rate": 9.036444229607623e-05, "loss": 0.37244181632995604, "memory(GiB)": 61.91, "step": 23450, "token_acc": 0.9233576642335767, "train_speed(iter/s)": 1.46644 }, { "epoch": 1.0048841095068763, "grad_norm": 2.424670457839966, "learning_rate": 9.036047031102024e-05, "loss": 0.16059852838516236, "memory(GiB)": 61.91, "step": 23455, "token_acc": 0.9607843137254902, "train_speed(iter/s)": 1.466433 }, { "epoch": 1.0050983248361254, "grad_norm": 3.1816744804382324, "learning_rate": 9.035649759479648e-05, "loss": 0.5765062808990479, "memory(GiB)": 61.91, "step": 23460, "token_acc": 0.864406779661017, "train_speed(iter/s)": 1.46643 }, { "epoch": 1.0053125401653742, "grad_norm": 1.5164985656738281, "learning_rate": 9.03525241474769e-05, "loss": 0.39667363166809083, "memory(GiB)": 61.91, "step": 23465, "token_acc": 0.9153225806451613, "train_speed(iter/s)": 1.466439 }, { "epoch": 1.0055267554946232, "grad_norm": 1.8255109786987305, "learning_rate": 9.034854996913349e-05, "loss": 0.2590620517730713, "memory(GiB)": 61.91, "step": 23470, "token_acc": 0.9511278195488722, "train_speed(iter/s)": 1.466455 }, { "epoch": 1.0057409708238723, "grad_norm": 3.093670129776001, "learning_rate": 9.034457505983825e-05, "loss": 0.20003871917724608, "memory(GiB)": 61.91, "step": 23475, "token_acc": 0.9486301369863014, "train_speed(iter/s)": 1.466449 }, { "epoch": 1.005955186153121, "grad_norm": 2.040165901184082, "learning_rate": 9.034059941966318e-05, "loss": 0.41846427917480467, "memory(GiB)": 61.91, "step": 23480, "token_acc": 0.908284023668639, "train_speed(iter/s)": 1.466429 }, { "epoch": 1.00616940148237, "grad_norm": 2.276426315307617, "learning_rate": 9.033662304868031e-05, "loss": 0.3062429666519165, "memory(GiB)": 61.91, "step": 23485, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.466434 }, { "epoch": 1.0063836168116191, "grad_norm": 3.0900518894195557, "learning_rate": 9.033264594696169e-05, "loss": 0.30295767784118655, "memory(GiB)": 61.91, "step": 23490, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.466439 }, { "epoch": 1.006597832140868, "grad_norm": 7.886228084564209, "learning_rate": 9.032866811457935e-05, "loss": 0.3275092124938965, "memory(GiB)": 61.91, "step": 23495, "token_acc": 0.9532374100719424, "train_speed(iter/s)": 1.466473 }, { "epoch": 1.006812047470117, "grad_norm": 2.3839709758758545, "learning_rate": 9.032468955160533e-05, "loss": 0.3025867700576782, "memory(GiB)": 61.91, "step": 23500, "token_acc": 0.9397590361445783, "train_speed(iter/s)": 1.466459 }, { "epoch": 1.006812047470117, "eval_loss": 2.469367265701294, "eval_runtime": 13.9982, "eval_samples_per_second": 7.144, "eval_steps_per_second": 7.144, "eval_token_acc": 0.4483260553129549, "step": 23500 }, { "epoch": 1.007026262799366, "grad_norm": 1.615294098854065, "learning_rate": 9.032071025811175e-05, "loss": 0.5064341068267822, "memory(GiB)": 61.91, "step": 23505, "token_acc": 0.5769633507853403, "train_speed(iter/s)": 1.465154 }, { "epoch": 1.0072404781286148, "grad_norm": 4.048498153686523, "learning_rate": 9.031673023417069e-05, "loss": 0.36421771049499513, "memory(GiB)": 61.91, "step": 23510, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.465138 }, { "epoch": 1.0074546934578639, "grad_norm": 2.4360225200653076, "learning_rate": 9.031274947985422e-05, "loss": 0.24397542476654052, "memory(GiB)": 61.91, "step": 23515, "token_acc": 0.95, "train_speed(iter/s)": 1.465134 }, { "epoch": 1.007668908787113, "grad_norm": 2.519141435623169, "learning_rate": 9.03087679952345e-05, "loss": 0.26824772357940674, "memory(GiB)": 61.91, "step": 23520, "token_acc": 0.9495548961424333, "train_speed(iter/s)": 1.465124 }, { "epoch": 1.0078831241163617, "grad_norm": 3.696718454360962, "learning_rate": 9.030478578038361e-05, "loss": 0.25292823314666746, "memory(GiB)": 61.91, "step": 23525, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.465125 }, { "epoch": 1.0080973394456108, "grad_norm": 4.095617771148682, "learning_rate": 9.030080283537374e-05, "loss": 0.3337285280227661, "memory(GiB)": 61.91, "step": 23530, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.465169 }, { "epoch": 1.0083115547748598, "grad_norm": 5.484187602996826, "learning_rate": 9.029681916027701e-05, "loss": 0.3777590751647949, "memory(GiB)": 61.91, "step": 23535, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.465204 }, { "epoch": 1.0085257701041086, "grad_norm": 8.735016822814941, "learning_rate": 9.029283475516561e-05, "loss": 0.33849124908447265, "memory(GiB)": 61.91, "step": 23540, "token_acc": 0.9304635761589404, "train_speed(iter/s)": 1.465178 }, { "epoch": 1.0087399854333576, "grad_norm": 2.73144268989563, "learning_rate": 9.028884962011169e-05, "loss": 0.3165081024169922, "memory(GiB)": 61.91, "step": 23545, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.465181 }, { "epoch": 1.0089542007626067, "grad_norm": 9.543110847473145, "learning_rate": 9.028486375518748e-05, "loss": 0.575448751449585, "memory(GiB)": 61.91, "step": 23550, "token_acc": 0.8790849673202614, "train_speed(iter/s)": 1.46522 }, { "epoch": 1.0091684160918555, "grad_norm": 7.58178186416626, "learning_rate": 9.028087716046516e-05, "loss": 0.43502011299133303, "memory(GiB)": 61.91, "step": 23555, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.46524 }, { "epoch": 1.0093826314211045, "grad_norm": 0.39661312103271484, "learning_rate": 9.027688983601699e-05, "loss": 0.5027358531951904, "memory(GiB)": 61.91, "step": 23560, "token_acc": 0.900709219858156, "train_speed(iter/s)": 1.465265 }, { "epoch": 1.0095968467503535, "grad_norm": 3.94197154045105, "learning_rate": 9.027290178191515e-05, "loss": 0.4046445369720459, "memory(GiB)": 61.91, "step": 23565, "token_acc": 0.902834008097166, "train_speed(iter/s)": 1.465269 }, { "epoch": 1.0098110620796024, "grad_norm": 3.701342821121216, "learning_rate": 9.026891299823192e-05, "loss": 0.5746723175048828, "memory(GiB)": 61.91, "step": 23570, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.46529 }, { "epoch": 1.0100252774088514, "grad_norm": 9.59444522857666, "learning_rate": 9.026492348503957e-05, "loss": 0.28274238109588623, "memory(GiB)": 61.91, "step": 23575, "token_acc": 0.9461538461538461, "train_speed(iter/s)": 1.465333 }, { "epoch": 1.0102394927381004, "grad_norm": 0.4904763400554657, "learning_rate": 9.026093324241035e-05, "loss": 0.24711041450500487, "memory(GiB)": 61.91, "step": 23580, "token_acc": 0.9504950495049505, "train_speed(iter/s)": 1.465339 }, { "epoch": 1.0104537080673492, "grad_norm": 6.157111167907715, "learning_rate": 9.025694227041656e-05, "loss": 0.36108989715576173, "memory(GiB)": 61.91, "step": 23585, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.465326 }, { "epoch": 1.0106679233965983, "grad_norm": 9.211114883422852, "learning_rate": 9.025295056913049e-05, "loss": 0.12766200304031372, "memory(GiB)": 61.91, "step": 23590, "token_acc": 0.9719626168224299, "train_speed(iter/s)": 1.465329 }, { "epoch": 1.0108821387258473, "grad_norm": 3.0461602210998535, "learning_rate": 9.024895813862446e-05, "loss": 0.34656329154968263, "memory(GiB)": 61.91, "step": 23595, "token_acc": 0.9134328358208955, "train_speed(iter/s)": 1.465363 }, { "epoch": 1.0110963540550961, "grad_norm": 3.0216598510742188, "learning_rate": 9.024496497897082e-05, "loss": 0.38831591606140137, "memory(GiB)": 61.91, "step": 23600, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.465363 }, { "epoch": 1.0113105693843452, "grad_norm": 1.6130406856536865, "learning_rate": 9.024097109024186e-05, "loss": 0.5789651870727539, "memory(GiB)": 61.91, "step": 23605, "token_acc": 0.8830409356725146, "train_speed(iter/s)": 1.465348 }, { "epoch": 1.0115247847135942, "grad_norm": 8.610644340515137, "learning_rate": 9.023697647250995e-05, "loss": 0.635160255432129, "memory(GiB)": 61.91, "step": 23610, "token_acc": 0.8957654723127035, "train_speed(iter/s)": 1.465329 }, { "epoch": 1.011739000042843, "grad_norm": 6.728611946105957, "learning_rate": 9.023298112584749e-05, "loss": 0.2811427593231201, "memory(GiB)": 61.91, "step": 23615, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.465318 }, { "epoch": 1.011953215372092, "grad_norm": 4.121522903442383, "learning_rate": 9.022898505032685e-05, "loss": 0.3907188415527344, "memory(GiB)": 61.91, "step": 23620, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.465323 }, { "epoch": 1.012167430701341, "grad_norm": 3.289065361022949, "learning_rate": 9.022498824602037e-05, "loss": 0.2839103937149048, "memory(GiB)": 61.91, "step": 23625, "token_acc": 0.9435736677115988, "train_speed(iter/s)": 1.465335 }, { "epoch": 1.0123816460305899, "grad_norm": 5.088583469390869, "learning_rate": 9.022099071300052e-05, "loss": 0.5283496856689454, "memory(GiB)": 61.91, "step": 23630, "token_acc": 0.8566433566433567, "train_speed(iter/s)": 1.465349 }, { "epoch": 1.012595861359839, "grad_norm": 1.2132079601287842, "learning_rate": 9.021699245133967e-05, "loss": 0.26539108753204343, "memory(GiB)": 61.91, "step": 23635, "token_acc": 0.9409937888198758, "train_speed(iter/s)": 1.465351 }, { "epoch": 1.012810076689088, "grad_norm": 2.5080668926239014, "learning_rate": 9.02129934611103e-05, "loss": 0.28379638195037843, "memory(GiB)": 61.91, "step": 23640, "token_acc": 0.9503311258278145, "train_speed(iter/s)": 1.465339 }, { "epoch": 1.0130242920183368, "grad_norm": 3.9254150390625, "learning_rate": 9.020899374238481e-05, "loss": 0.6180932998657227, "memory(GiB)": 61.91, "step": 23645, "token_acc": 0.8877887788778878, "train_speed(iter/s)": 1.465347 }, { "epoch": 1.0132385073475858, "grad_norm": 1.524520754814148, "learning_rate": 9.020499329523569e-05, "loss": 0.26360454559326174, "memory(GiB)": 61.91, "step": 23650, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.465333 }, { "epoch": 1.0134527226768348, "grad_norm": 2.9095935821533203, "learning_rate": 9.02009921197354e-05, "loss": 0.3064687728881836, "memory(GiB)": 61.91, "step": 23655, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.465365 }, { "epoch": 1.0136669380060837, "grad_norm": 3.9265029430389404, "learning_rate": 9.019699021595642e-05, "loss": 0.4506204605102539, "memory(GiB)": 61.91, "step": 23660, "token_acc": 0.9025270758122743, "train_speed(iter/s)": 1.465372 }, { "epoch": 1.0138811533353327, "grad_norm": 3.024338483810425, "learning_rate": 9.019298758397127e-05, "loss": 0.3629601240158081, "memory(GiB)": 61.91, "step": 23665, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.465352 }, { "epoch": 1.0140953686645817, "grad_norm": 3.0038468837738037, "learning_rate": 9.018898422385243e-05, "loss": 0.4109188556671143, "memory(GiB)": 61.91, "step": 23670, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.465357 }, { "epoch": 1.0143095839938305, "grad_norm": 8.002450942993164, "learning_rate": 9.018498013567244e-05, "loss": 0.5566529750823974, "memory(GiB)": 61.91, "step": 23675, "token_acc": 0.8658892128279884, "train_speed(iter/s)": 1.465348 }, { "epoch": 1.0145237993230796, "grad_norm": 1.4162464141845703, "learning_rate": 9.018097531950385e-05, "loss": 0.20337629318237305, "memory(GiB)": 61.91, "step": 23680, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.465331 }, { "epoch": 1.0147380146523286, "grad_norm": 3.031485080718994, "learning_rate": 9.01769697754192e-05, "loss": 0.4739412307739258, "memory(GiB)": 61.91, "step": 23685, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.465339 }, { "epoch": 1.0149522299815774, "grad_norm": 1.5300397872924805, "learning_rate": 9.017296350349105e-05, "loss": 0.4788696765899658, "memory(GiB)": 61.91, "step": 23690, "token_acc": 0.8901515151515151, "train_speed(iter/s)": 1.465375 }, { "epoch": 1.0151664453108264, "grad_norm": 3.886418581008911, "learning_rate": 9.016895650379198e-05, "loss": 0.30700781345367434, "memory(GiB)": 61.91, "step": 23695, "token_acc": 0.9351851851851852, "train_speed(iter/s)": 1.465373 }, { "epoch": 1.0153806606400755, "grad_norm": 1.7262898683547974, "learning_rate": 9.016494877639457e-05, "loss": 0.3223842144012451, "memory(GiB)": 61.91, "step": 23700, "token_acc": 0.9297124600638977, "train_speed(iter/s)": 1.465341 }, { "epoch": 1.0155948759693243, "grad_norm": 0.08459437638521194, "learning_rate": 9.016094032137144e-05, "loss": 0.44769678115844724, "memory(GiB)": 61.91, "step": 23705, "token_acc": 0.930327868852459, "train_speed(iter/s)": 1.465341 }, { "epoch": 1.0158090912985733, "grad_norm": 4.825450897216797, "learning_rate": 9.015693113879521e-05, "loss": 0.552297830581665, "memory(GiB)": 61.91, "step": 23710, "token_acc": 0.90625, "train_speed(iter/s)": 1.465369 }, { "epoch": 1.0160233066278224, "grad_norm": 2.290187120437622, "learning_rate": 9.01529212287385e-05, "loss": 0.5609296798706055, "memory(GiB)": 61.91, "step": 23715, "token_acc": 0.8723404255319149, "train_speed(iter/s)": 1.465397 }, { "epoch": 1.0162375219570712, "grad_norm": 4.897535800933838, "learning_rate": 9.014891059127395e-05, "loss": 0.2322864532470703, "memory(GiB)": 61.91, "step": 23720, "token_acc": 0.9592592592592593, "train_speed(iter/s)": 1.465397 }, { "epoch": 1.0164517372863202, "grad_norm": 5.353052139282227, "learning_rate": 9.014489922647423e-05, "loss": 0.7771756172180175, "memory(GiB)": 61.91, "step": 23725, "token_acc": 0.8647686832740213, "train_speed(iter/s)": 1.465394 }, { "epoch": 1.0166659526155692, "grad_norm": 2.4255247116088867, "learning_rate": 9.014088713441199e-05, "loss": 0.27210915088653564, "memory(GiB)": 61.91, "step": 23730, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.465415 }, { "epoch": 1.016880167944818, "grad_norm": 13.28315258026123, "learning_rate": 9.013687431515994e-05, "loss": 0.25912630558013916, "memory(GiB)": 61.91, "step": 23735, "token_acc": 0.95, "train_speed(iter/s)": 1.46544 }, { "epoch": 1.017094383274067, "grad_norm": 0.2015436738729477, "learning_rate": 9.013286076879075e-05, "loss": 0.27670879364013673, "memory(GiB)": 61.91, "step": 23740, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.46546 }, { "epoch": 1.0173085986033161, "grad_norm": 3.1965699195861816, "learning_rate": 9.012884649537715e-05, "loss": 0.4143507957458496, "memory(GiB)": 61.91, "step": 23745, "token_acc": 0.9182389937106918, "train_speed(iter/s)": 1.465452 }, { "epoch": 1.017522813932565, "grad_norm": 3.8803138732910156, "learning_rate": 9.012483149499184e-05, "loss": 0.3869894027709961, "memory(GiB)": 61.91, "step": 23750, "token_acc": 0.9255014326647565, "train_speed(iter/s)": 1.465449 }, { "epoch": 1.017737029261814, "grad_norm": 8.221329689025879, "learning_rate": 9.012081576770757e-05, "loss": 0.3318179130554199, "memory(GiB)": 61.91, "step": 23755, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.465445 }, { "epoch": 1.017951244591063, "grad_norm": 3.0845487117767334, "learning_rate": 9.011679931359708e-05, "loss": 0.3559630632400513, "memory(GiB)": 61.91, "step": 23760, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.465443 }, { "epoch": 1.0181654599203118, "grad_norm": 1.5996146202087402, "learning_rate": 9.011278213273315e-05, "loss": 0.418825101852417, "memory(GiB)": 61.91, "step": 23765, "token_acc": 0.9308943089430894, "train_speed(iter/s)": 1.465455 }, { "epoch": 1.0183796752495609, "grad_norm": 5.632864952087402, "learning_rate": 9.010876422518854e-05, "loss": 0.4143717288970947, "memory(GiB)": 61.91, "step": 23770, "token_acc": 0.9159663865546218, "train_speed(iter/s)": 1.465456 }, { "epoch": 1.01859389057881, "grad_norm": 3.0984368324279785, "learning_rate": 9.010474559103604e-05, "loss": 0.5069024085998535, "memory(GiB)": 61.91, "step": 23775, "token_acc": 0.9093567251461988, "train_speed(iter/s)": 1.465489 }, { "epoch": 1.0188081059080587, "grad_norm": 2.284766435623169, "learning_rate": 9.010072623034845e-05, "loss": 0.3870377540588379, "memory(GiB)": 61.91, "step": 23780, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.465488 }, { "epoch": 1.0190223212373077, "grad_norm": 3.195847511291504, "learning_rate": 9.00967061431986e-05, "loss": 0.5520201683044433, "memory(GiB)": 61.91, "step": 23785, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.465539 }, { "epoch": 1.0192365365665568, "grad_norm": 3.7913758754730225, "learning_rate": 9.009268532965929e-05, "loss": 0.47884197235107423, "memory(GiB)": 61.91, "step": 23790, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.465535 }, { "epoch": 1.0194507518958056, "grad_norm": 1.4986008405685425, "learning_rate": 9.008866378980338e-05, "loss": 0.4291853427886963, "memory(GiB)": 61.91, "step": 23795, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.465551 }, { "epoch": 1.0196649672250546, "grad_norm": 5.391343116760254, "learning_rate": 9.008464152370371e-05, "loss": 0.6348857879638672, "memory(GiB)": 61.91, "step": 23800, "token_acc": 0.875968992248062, "train_speed(iter/s)": 1.465552 }, { "epoch": 1.0198791825543037, "grad_norm": 2.7656822204589844, "learning_rate": 9.008061853143318e-05, "loss": 0.7851232528686524, "memory(GiB)": 61.91, "step": 23805, "token_acc": 0.8095238095238095, "train_speed(iter/s)": 1.465556 }, { "epoch": 1.0200933978835525, "grad_norm": 4.616089344024658, "learning_rate": 9.00765948130646e-05, "loss": 0.4948453903198242, "memory(GiB)": 61.91, "step": 23810, "token_acc": 0.8726591760299626, "train_speed(iter/s)": 1.465558 }, { "epoch": 1.0203076132128015, "grad_norm": 2.38692307472229, "learning_rate": 9.007257036867095e-05, "loss": 0.35584383010864257, "memory(GiB)": 61.91, "step": 23815, "token_acc": 0.9223300970873787, "train_speed(iter/s)": 1.465565 }, { "epoch": 1.0205218285420505, "grad_norm": 3.6535544395446777, "learning_rate": 9.006854519832509e-05, "loss": 0.3924243927001953, "memory(GiB)": 61.91, "step": 23820, "token_acc": 0.9002849002849003, "train_speed(iter/s)": 1.46555 }, { "epoch": 1.0207360438712993, "grad_norm": 1.4373676776885986, "learning_rate": 9.006451930209995e-05, "loss": 0.30323379039764403, "memory(GiB)": 61.91, "step": 23825, "token_acc": 0.936046511627907, "train_speed(iter/s)": 1.465556 }, { "epoch": 1.0209502592005484, "grad_norm": 0.54873126745224, "learning_rate": 9.006049268006844e-05, "loss": 0.4213856220245361, "memory(GiB)": 61.91, "step": 23830, "token_acc": 0.9014084507042254, "train_speed(iter/s)": 1.465565 }, { "epoch": 1.0211644745297974, "grad_norm": 2.612874746322632, "learning_rate": 9.005646533230354e-05, "loss": 0.45401439666748045, "memory(GiB)": 61.91, "step": 23835, "token_acc": 0.9108910891089109, "train_speed(iter/s)": 1.465592 }, { "epoch": 1.0213786898590462, "grad_norm": 3.547144889831543, "learning_rate": 9.005243725887819e-05, "loss": 0.33333210945129393, "memory(GiB)": 61.91, "step": 23840, "token_acc": 0.9432314410480349, "train_speed(iter/s)": 1.465604 }, { "epoch": 1.0215929051882953, "grad_norm": 3.859700918197632, "learning_rate": 9.004840845986538e-05, "loss": 0.28083953857421873, "memory(GiB)": 61.91, "step": 23845, "token_acc": 0.941908713692946, "train_speed(iter/s)": 1.465611 }, { "epoch": 1.0218071205175443, "grad_norm": 1.8149458169937134, "learning_rate": 9.004437893533807e-05, "loss": 0.41157379150390627, "memory(GiB)": 61.91, "step": 23850, "token_acc": 0.9085173501577287, "train_speed(iter/s)": 1.465642 }, { "epoch": 1.022021335846793, "grad_norm": 7.2644124031066895, "learning_rate": 9.004034868536929e-05, "loss": 0.2830932140350342, "memory(GiB)": 61.91, "step": 23855, "token_acc": 0.9455782312925171, "train_speed(iter/s)": 1.465639 }, { "epoch": 1.0222355511760421, "grad_norm": 1.5202754735946655, "learning_rate": 9.0036317710032e-05, "loss": 0.30549819469451905, "memory(GiB)": 61.91, "step": 23860, "token_acc": 0.9462025316455697, "train_speed(iter/s)": 1.465642 }, { "epoch": 1.0224497665052912, "grad_norm": 3.795874834060669, "learning_rate": 9.003228600939926e-05, "loss": 0.3573035001754761, "memory(GiB)": 61.91, "step": 23865, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.465646 }, { "epoch": 1.02266398183454, "grad_norm": 3.455781936645508, "learning_rate": 9.002825358354414e-05, "loss": 0.5387766361236572, "memory(GiB)": 61.91, "step": 23870, "token_acc": 0.8878205128205128, "train_speed(iter/s)": 1.465633 }, { "epoch": 1.022878197163789, "grad_norm": 2.7534501552581787, "learning_rate": 9.002422043253962e-05, "loss": 0.19465169906616211, "memory(GiB)": 61.91, "step": 23875, "token_acc": 0.9567901234567902, "train_speed(iter/s)": 1.465634 }, { "epoch": 1.023092412493038, "grad_norm": 4.161811828613281, "learning_rate": 9.002018655645882e-05, "loss": 0.38434433937072754, "memory(GiB)": 61.91, "step": 23880, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.465669 }, { "epoch": 1.0233066278222869, "grad_norm": 3.3509883880615234, "learning_rate": 9.00161519553748e-05, "loss": 0.2794214725494385, "memory(GiB)": 61.91, "step": 23885, "token_acc": 0.9540636042402827, "train_speed(iter/s)": 1.465643 }, { "epoch": 1.023520843151536, "grad_norm": 6.030032157897949, "learning_rate": 9.001211662936065e-05, "loss": 0.4510654926300049, "memory(GiB)": 61.91, "step": 23890, "token_acc": 0.875, "train_speed(iter/s)": 1.46568 }, { "epoch": 1.023735058480785, "grad_norm": 3.849942922592163, "learning_rate": 9.000808057848946e-05, "loss": 0.2655608654022217, "memory(GiB)": 61.91, "step": 23895, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.465677 }, { "epoch": 1.0239492738100338, "grad_norm": 4.946908950805664, "learning_rate": 9.000404380283435e-05, "loss": 0.6014955997467041, "memory(GiB)": 61.91, "step": 23900, "token_acc": 0.888030888030888, "train_speed(iter/s)": 1.465691 }, { "epoch": 1.0241634891392828, "grad_norm": 5.117305755615234, "learning_rate": 9.000000630246848e-05, "loss": 0.4203506946563721, "memory(GiB)": 61.91, "step": 23905, "token_acc": 0.8727272727272727, "train_speed(iter/s)": 1.465683 }, { "epoch": 1.0243777044685318, "grad_norm": 4.915281295776367, "learning_rate": 8.999596807746497e-05, "loss": 0.604158592224121, "memory(GiB)": 61.91, "step": 23910, "token_acc": 0.8778877887788779, "train_speed(iter/s)": 1.465678 }, { "epoch": 1.0245919197977806, "grad_norm": 3.1406631469726562, "learning_rate": 8.999192912789697e-05, "loss": 0.3629206895828247, "memory(GiB)": 61.91, "step": 23915, "token_acc": 0.9260450160771704, "train_speed(iter/s)": 1.465675 }, { "epoch": 1.0248061351270297, "grad_norm": 1.7814267873764038, "learning_rate": 8.998788945383768e-05, "loss": 0.21764559745788575, "memory(GiB)": 61.91, "step": 23920, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.465703 }, { "epoch": 1.0250203504562787, "grad_norm": 0.5259619355201721, "learning_rate": 8.998384905536025e-05, "loss": 0.13449478149414062, "memory(GiB)": 61.91, "step": 23925, "token_acc": 0.9776951672862454, "train_speed(iter/s)": 1.465703 }, { "epoch": 1.0252345657855275, "grad_norm": 3.3351895809173584, "learning_rate": 8.997980793253789e-05, "loss": 0.40390958786010744, "memory(GiB)": 61.91, "step": 23930, "token_acc": 0.907051282051282, "train_speed(iter/s)": 1.465694 }, { "epoch": 1.0254487811147766, "grad_norm": 2.7225615978240967, "learning_rate": 8.99757660854438e-05, "loss": 0.3746323585510254, "memory(GiB)": 61.91, "step": 23935, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.465677 }, { "epoch": 1.0256629964440256, "grad_norm": 1.1734360456466675, "learning_rate": 8.99717235141512e-05, "loss": 0.285518217086792, "memory(GiB)": 61.91, "step": 23940, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.465684 }, { "epoch": 1.0258772117732744, "grad_norm": 2.768935203552246, "learning_rate": 8.996768021873334e-05, "loss": 0.4000732898712158, "memory(GiB)": 61.91, "step": 23945, "token_acc": 0.9074626865671642, "train_speed(iter/s)": 1.465668 }, { "epoch": 1.0260914271025234, "grad_norm": 1.2003343105316162, "learning_rate": 8.996363619926346e-05, "loss": 0.24114615917205812, "memory(GiB)": 61.91, "step": 23950, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.465696 }, { "epoch": 1.0263056424317725, "grad_norm": 4.34152364730835, "learning_rate": 8.995959145581482e-05, "loss": 0.5346135139465332, "memory(GiB)": 61.91, "step": 23955, "token_acc": 0.8544061302681992, "train_speed(iter/s)": 1.465711 }, { "epoch": 1.0265198577610213, "grad_norm": 7.844885349273682, "learning_rate": 8.995554598846069e-05, "loss": 0.3114687204360962, "memory(GiB)": 61.91, "step": 23960, "token_acc": 0.9376947040498442, "train_speed(iter/s)": 1.465687 }, { "epoch": 1.0267340730902703, "grad_norm": 4.174375057220459, "learning_rate": 8.995149979727437e-05, "loss": 0.2836555004119873, "memory(GiB)": 61.91, "step": 23965, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.465702 }, { "epoch": 1.0269482884195194, "grad_norm": 1.7498981952667236, "learning_rate": 8.994745288232916e-05, "loss": 0.1777945041656494, "memory(GiB)": 61.91, "step": 23970, "token_acc": 0.9479166666666666, "train_speed(iter/s)": 1.465708 }, { "epoch": 1.0271625037487682, "grad_norm": 3.24495530128479, "learning_rate": 8.994340524369836e-05, "loss": 0.3749157190322876, "memory(GiB)": 61.91, "step": 23975, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.465724 }, { "epoch": 1.0273767190780172, "grad_norm": 2.2031073570251465, "learning_rate": 8.993935688145529e-05, "loss": 0.2984520673751831, "memory(GiB)": 61.91, "step": 23980, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.46573 }, { "epoch": 1.0275909344072662, "grad_norm": 1.7824747562408447, "learning_rate": 8.993530779567332e-05, "loss": 0.5164326190948486, "memory(GiB)": 61.91, "step": 23985, "token_acc": 0.899641577060932, "train_speed(iter/s)": 1.46573 }, { "epoch": 1.027805149736515, "grad_norm": 2.182659149169922, "learning_rate": 8.993125798642579e-05, "loss": 0.18115880489349365, "memory(GiB)": 61.91, "step": 23990, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.465721 }, { "epoch": 1.028019365065764, "grad_norm": 4.258477687835693, "learning_rate": 8.992720745378605e-05, "loss": 0.19324620962142944, "memory(GiB)": 61.91, "step": 23995, "token_acc": 0.9528985507246377, "train_speed(iter/s)": 1.465704 }, { "epoch": 1.0282335803950131, "grad_norm": 3.3171730041503906, "learning_rate": 8.99231561978275e-05, "loss": 0.3769431829452515, "memory(GiB)": 61.91, "step": 24000, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.465703 }, { "epoch": 1.0282335803950131, "eval_loss": 2.421964406967163, "eval_runtime": 14.0585, "eval_samples_per_second": 7.113, "eval_steps_per_second": 7.113, "eval_token_acc": 0.4547707558859975, "step": 24000 }, { "epoch": 1.028447795724262, "grad_norm": 0.6051115989685059, "learning_rate": 8.991910421862352e-05, "loss": 0.5557611465454102, "memory(GiB)": 61.91, "step": 24005, "token_acc": 0.5623268698060941, "train_speed(iter/s)": 1.464404 }, { "epoch": 1.028662011053511, "grad_norm": 3.6435952186584473, "learning_rate": 8.991505151624752e-05, "loss": 0.40444722175598147, "memory(GiB)": 61.91, "step": 24010, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.464394 }, { "epoch": 1.02887622638276, "grad_norm": 2.667833089828491, "learning_rate": 8.991099809077292e-05, "loss": 0.37153091430664065, "memory(GiB)": 61.91, "step": 24015, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.464381 }, { "epoch": 1.0290904417120088, "grad_norm": 6.376405239105225, "learning_rate": 8.990694394227317e-05, "loss": 0.6703934669494629, "memory(GiB)": 61.91, "step": 24020, "token_acc": 0.8825396825396825, "train_speed(iter/s)": 1.464383 }, { "epoch": 1.0293046570412578, "grad_norm": 2.192556142807007, "learning_rate": 8.990288907082168e-05, "loss": 0.30700173377990725, "memory(GiB)": 61.91, "step": 24025, "token_acc": 0.9370860927152318, "train_speed(iter/s)": 1.464365 }, { "epoch": 1.0295188723705069, "grad_norm": 1.5167392492294312, "learning_rate": 8.989883347649191e-05, "loss": 0.6016419410705567, "memory(GiB)": 61.91, "step": 24030, "token_acc": 0.8674698795180723, "train_speed(iter/s)": 1.464352 }, { "epoch": 1.0297330876997557, "grad_norm": 3.5724446773529053, "learning_rate": 8.989477715935735e-05, "loss": 0.3023569107055664, "memory(GiB)": 61.91, "step": 24035, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.464348 }, { "epoch": 1.0299473030290047, "grad_norm": 5.434454441070557, "learning_rate": 8.989072011949148e-05, "loss": 0.4067696571350098, "memory(GiB)": 61.91, "step": 24040, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.464351 }, { "epoch": 1.0301615183582538, "grad_norm": 3.908027172088623, "learning_rate": 8.988666235696779e-05, "loss": 0.3161868333816528, "memory(GiB)": 61.91, "step": 24045, "token_acc": 0.9458333333333333, "train_speed(iter/s)": 1.464361 }, { "epoch": 1.0303757336875026, "grad_norm": 5.010776519775391, "learning_rate": 8.98826038718598e-05, "loss": 0.421356725692749, "memory(GiB)": 61.91, "step": 24050, "token_acc": 0.9254658385093167, "train_speed(iter/s)": 1.464398 }, { "epoch": 1.0305899490167516, "grad_norm": 4.870143890380859, "learning_rate": 8.987854466424103e-05, "loss": 0.4914794921875, "memory(GiB)": 61.91, "step": 24055, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.464394 }, { "epoch": 1.0308041643460006, "grad_norm": 1.539344072341919, "learning_rate": 8.987448473418502e-05, "loss": 0.309219217300415, "memory(GiB)": 61.91, "step": 24060, "token_acc": 0.9297124600638977, "train_speed(iter/s)": 1.464394 }, { "epoch": 1.0310183796752495, "grad_norm": 6.891141414642334, "learning_rate": 8.98704240817653e-05, "loss": 0.48237247467041017, "memory(GiB)": 61.91, "step": 24065, "token_acc": 0.8782287822878229, "train_speed(iter/s)": 1.464382 }, { "epoch": 1.0312325950044985, "grad_norm": 9.305587768554688, "learning_rate": 8.986636270705545e-05, "loss": 0.5514328002929687, "memory(GiB)": 61.91, "step": 24070, "token_acc": 0.8618421052631579, "train_speed(iter/s)": 1.464388 }, { "epoch": 1.0314468103337475, "grad_norm": 4.008392810821533, "learning_rate": 8.986230061012905e-05, "loss": 0.4085784912109375, "memory(GiB)": 61.91, "step": 24075, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.464386 }, { "epoch": 1.0316610256629963, "grad_norm": 3.2478981018066406, "learning_rate": 8.985823779105968e-05, "loss": 0.6436156749725341, "memory(GiB)": 61.91, "step": 24080, "token_acc": 0.8733333333333333, "train_speed(iter/s)": 1.464398 }, { "epoch": 1.0318752409922454, "grad_norm": 1.4090534448623657, "learning_rate": 8.985417424992093e-05, "loss": 0.33253023624420164, "memory(GiB)": 61.91, "step": 24085, "token_acc": 0.9481481481481482, "train_speed(iter/s)": 1.464414 }, { "epoch": 1.0320894563214944, "grad_norm": 3.732822895050049, "learning_rate": 8.985010998678642e-05, "loss": 0.35651991367340086, "memory(GiB)": 61.91, "step": 24090, "token_acc": 0.9218009478672986, "train_speed(iter/s)": 1.464413 }, { "epoch": 1.0323036716507432, "grad_norm": 5.69921875, "learning_rate": 8.984604500172982e-05, "loss": 0.5643367767333984, "memory(GiB)": 61.91, "step": 24095, "token_acc": 0.8746081504702194, "train_speed(iter/s)": 1.464427 }, { "epoch": 1.0325178869799922, "grad_norm": 1.5702533721923828, "learning_rate": 8.984197929482471e-05, "loss": 0.23799962997436525, "memory(GiB)": 61.91, "step": 24100, "token_acc": 0.9641255605381166, "train_speed(iter/s)": 1.464436 }, { "epoch": 1.0327321023092413, "grad_norm": 0.3736724257469177, "learning_rate": 8.983791286614476e-05, "loss": 0.3604216337203979, "memory(GiB)": 61.91, "step": 24105, "token_acc": 0.9188405797101449, "train_speed(iter/s)": 1.464437 }, { "epoch": 1.03294631763849, "grad_norm": 7.146617889404297, "learning_rate": 8.983384571576367e-05, "loss": 0.5076184272766113, "memory(GiB)": 61.91, "step": 24110, "token_acc": 0.8858267716535433, "train_speed(iter/s)": 1.464474 }, { "epoch": 1.0331605329677391, "grad_norm": 2.8069584369659424, "learning_rate": 8.98297778437551e-05, "loss": 0.5033679485321045, "memory(GiB)": 61.91, "step": 24115, "token_acc": 0.8884615384615384, "train_speed(iter/s)": 1.464504 }, { "epoch": 1.0333747482969882, "grad_norm": 2.150357961654663, "learning_rate": 8.982570925019273e-05, "loss": 0.48045029640197756, "memory(GiB)": 61.91, "step": 24120, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.464499 }, { "epoch": 1.0335889636262372, "grad_norm": 4.750834941864014, "learning_rate": 8.982163993515027e-05, "loss": 0.41797494888305664, "memory(GiB)": 61.91, "step": 24125, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.464499 }, { "epoch": 1.033803178955486, "grad_norm": 5.105282783508301, "learning_rate": 8.981756989870147e-05, "loss": 0.4597672462463379, "memory(GiB)": 61.91, "step": 24130, "token_acc": 0.886986301369863, "train_speed(iter/s)": 1.464485 }, { "epoch": 1.034017394284735, "grad_norm": 4.628366470336914, "learning_rate": 8.981349914092002e-05, "loss": 0.32115945816040037, "memory(GiB)": 61.91, "step": 24135, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.464512 }, { "epoch": 1.034231609613984, "grad_norm": 3.567861318588257, "learning_rate": 8.98094276618797e-05, "loss": 0.35840439796447754, "memory(GiB)": 61.91, "step": 24140, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.464514 }, { "epoch": 1.034445824943233, "grad_norm": 2.5147645473480225, "learning_rate": 8.980535546165422e-05, "loss": 0.2598418235778809, "memory(GiB)": 61.91, "step": 24145, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.464527 }, { "epoch": 1.034660040272482, "grad_norm": 5.6054277420043945, "learning_rate": 8.980128254031743e-05, "loss": 0.7183452129364014, "memory(GiB)": 61.91, "step": 24150, "token_acc": 0.8598484848484849, "train_speed(iter/s)": 1.464556 }, { "epoch": 1.034874255601731, "grad_norm": 1.148579478263855, "learning_rate": 8.979720889794305e-05, "loss": 0.2743793249130249, "memory(GiB)": 61.91, "step": 24155, "token_acc": 0.9543859649122807, "train_speed(iter/s)": 1.464556 }, { "epoch": 1.0350884709309798, "grad_norm": 2.0797195434570312, "learning_rate": 8.979313453460489e-05, "loss": 0.6344887733459472, "memory(GiB)": 61.91, "step": 24160, "token_acc": 0.8532934131736527, "train_speed(iter/s)": 1.464565 }, { "epoch": 1.0353026862602288, "grad_norm": 3.153362274169922, "learning_rate": 8.978905945037678e-05, "loss": 0.3576789379119873, "memory(GiB)": 61.91, "step": 24165, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.464611 }, { "epoch": 1.0355169015894778, "grad_norm": 6.813891410827637, "learning_rate": 8.978498364533252e-05, "loss": 0.490116548538208, "memory(GiB)": 61.91, "step": 24170, "token_acc": 0.8877887788778878, "train_speed(iter/s)": 1.464588 }, { "epoch": 1.0357311169187267, "grad_norm": 0.4202563762664795, "learning_rate": 8.978090711954598e-05, "loss": 0.5096145629882812, "memory(GiB)": 61.91, "step": 24175, "token_acc": 0.9046052631578947, "train_speed(iter/s)": 1.46458 }, { "epoch": 1.0359453322479757, "grad_norm": 2.6798930168151855, "learning_rate": 8.977682987309097e-05, "loss": 0.5132251739501953, "memory(GiB)": 61.91, "step": 24180, "token_acc": 0.8982456140350877, "train_speed(iter/s)": 1.464591 }, { "epoch": 1.0361595475772247, "grad_norm": 2.239856481552124, "learning_rate": 8.977275190604138e-05, "loss": 0.431731128692627, "memory(GiB)": 61.91, "step": 24185, "token_acc": 0.9197530864197531, "train_speed(iter/s)": 1.464587 }, { "epoch": 1.0363737629064735, "grad_norm": 5.426468849182129, "learning_rate": 8.97686732184711e-05, "loss": 0.2138045310974121, "memory(GiB)": 61.91, "step": 24190, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.464575 }, { "epoch": 1.0365879782357226, "grad_norm": 3.2703423500061035, "learning_rate": 8.976459381045396e-05, "loss": 0.49705729484558103, "memory(GiB)": 61.91, "step": 24195, "token_acc": 0.8950819672131147, "train_speed(iter/s)": 1.464557 }, { "epoch": 1.0368021935649716, "grad_norm": 3.2299559116363525, "learning_rate": 8.976051368206393e-05, "loss": 0.5994937419891357, "memory(GiB)": 61.91, "step": 24200, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.464564 }, { "epoch": 1.0370164088942204, "grad_norm": 3.224350690841675, "learning_rate": 8.97564328333749e-05, "loss": 0.3096506118774414, "memory(GiB)": 61.91, "step": 24205, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.464575 }, { "epoch": 1.0372306242234695, "grad_norm": 3.6686999797821045, "learning_rate": 8.975235126446077e-05, "loss": 0.35379881858825685, "memory(GiB)": 61.91, "step": 24210, "token_acc": 0.9158878504672897, "train_speed(iter/s)": 1.464593 }, { "epoch": 1.0374448395527185, "grad_norm": 2.865050792694092, "learning_rate": 8.974826897539553e-05, "loss": 0.34299044609069823, "memory(GiB)": 61.91, "step": 24215, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.464584 }, { "epoch": 1.0376590548819673, "grad_norm": 4.1757941246032715, "learning_rate": 8.97441859662531e-05, "loss": 0.21633481979370117, "memory(GiB)": 61.91, "step": 24220, "token_acc": 0.9378698224852071, "train_speed(iter/s)": 1.464594 }, { "epoch": 1.0378732702112163, "grad_norm": 1.6024999618530273, "learning_rate": 8.974010223710744e-05, "loss": 0.32485003471374513, "memory(GiB)": 61.91, "step": 24225, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.464573 }, { "epoch": 1.0380874855404654, "grad_norm": 5.2512898445129395, "learning_rate": 8.973601778803257e-05, "loss": 0.31087527275085447, "memory(GiB)": 61.91, "step": 24230, "token_acc": 0.9183006535947712, "train_speed(iter/s)": 1.464572 }, { "epoch": 1.0383017008697142, "grad_norm": 4.432954788208008, "learning_rate": 8.973193261910247e-05, "loss": 0.4209420680999756, "memory(GiB)": 61.91, "step": 24235, "token_acc": 0.9108527131782945, "train_speed(iter/s)": 1.464558 }, { "epoch": 1.0385159161989632, "grad_norm": 1.7838047742843628, "learning_rate": 8.972784673039111e-05, "loss": 0.38697524070739747, "memory(GiB)": 61.91, "step": 24240, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.464567 }, { "epoch": 1.0387301315282123, "grad_norm": 3.2540478706359863, "learning_rate": 8.972376012197256e-05, "loss": 0.3679241418838501, "memory(GiB)": 61.91, "step": 24245, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.464579 }, { "epoch": 1.038944346857461, "grad_norm": 0.34556567668914795, "learning_rate": 8.971967279392083e-05, "loss": 0.24760043621063232, "memory(GiB)": 61.91, "step": 24250, "token_acc": 0.9540983606557377, "train_speed(iter/s)": 1.464577 }, { "epoch": 1.03915856218671, "grad_norm": 2.6779847145080566, "learning_rate": 8.971558474630997e-05, "loss": 0.25387375354766845, "memory(GiB)": 61.91, "step": 24255, "token_acc": 0.9448051948051948, "train_speed(iter/s)": 1.464579 }, { "epoch": 1.0393727775159591, "grad_norm": 10.152575492858887, "learning_rate": 8.971149597921402e-05, "loss": 0.35585870742797854, "memory(GiB)": 61.91, "step": 24260, "token_acc": 0.933993399339934, "train_speed(iter/s)": 1.464603 }, { "epoch": 1.039586992845208, "grad_norm": 5.448152542114258, "learning_rate": 8.970740649270708e-05, "loss": 0.4424184799194336, "memory(GiB)": 61.91, "step": 24265, "token_acc": 0.9235127478753541, "train_speed(iter/s)": 1.464601 }, { "epoch": 1.039801208174457, "grad_norm": 4.517699718475342, "learning_rate": 8.970331628686323e-05, "loss": 0.37652125358581545, "memory(GiB)": 61.91, "step": 24270, "token_acc": 0.92578125, "train_speed(iter/s)": 1.464607 }, { "epoch": 1.040015423503706, "grad_norm": 4.656617641448975, "learning_rate": 8.969922536175654e-05, "loss": 0.20780539512634277, "memory(GiB)": 61.91, "step": 24275, "token_acc": 0.9649122807017544, "train_speed(iter/s)": 1.464608 }, { "epoch": 1.0402296388329548, "grad_norm": 5.385461330413818, "learning_rate": 8.969513371746116e-05, "loss": 0.7299694538116455, "memory(GiB)": 61.91, "step": 24280, "token_acc": 0.8438661710037175, "train_speed(iter/s)": 1.464611 }, { "epoch": 1.0404438541622039, "grad_norm": 2.6314682960510254, "learning_rate": 8.96910413540512e-05, "loss": 0.45586442947387695, "memory(GiB)": 61.91, "step": 24285, "token_acc": 0.9022801302931596, "train_speed(iter/s)": 1.464617 }, { "epoch": 1.040658069491453, "grad_norm": 5.021481037139893, "learning_rate": 8.968694827160078e-05, "loss": 0.44088234901428225, "memory(GiB)": 61.91, "step": 24290, "token_acc": 0.9197530864197531, "train_speed(iter/s)": 1.464599 }, { "epoch": 1.0408722848207017, "grad_norm": 0.8950258493423462, "learning_rate": 8.968285447018406e-05, "loss": 0.23418760299682617, "memory(GiB)": 61.91, "step": 24295, "token_acc": 0.9590443686006825, "train_speed(iter/s)": 1.464588 }, { "epoch": 1.0410865001499507, "grad_norm": 3.1786627769470215, "learning_rate": 8.967875994987521e-05, "loss": 0.2624032735824585, "memory(GiB)": 61.91, "step": 24300, "token_acc": 0.9282868525896414, "train_speed(iter/s)": 1.464597 }, { "epoch": 1.0413007154791998, "grad_norm": 2.7386865615844727, "learning_rate": 8.967466471074841e-05, "loss": 0.6103688240051269, "memory(GiB)": 61.91, "step": 24305, "token_acc": 0.8680351906158358, "train_speed(iter/s)": 1.464632 }, { "epoch": 1.0415149308084486, "grad_norm": 2.946176290512085, "learning_rate": 8.967056875287783e-05, "loss": 0.2909055709838867, "memory(GiB)": 61.91, "step": 24310, "token_acc": 0.9312169312169312, "train_speed(iter/s)": 1.464654 }, { "epoch": 1.0417291461376976, "grad_norm": 3.299198627471924, "learning_rate": 8.966647207633769e-05, "loss": 0.30119829177856444, "memory(GiB)": 61.91, "step": 24315, "token_acc": 0.9244186046511628, "train_speed(iter/s)": 1.464656 }, { "epoch": 1.0419433614669467, "grad_norm": 6.524297714233398, "learning_rate": 8.966237468120218e-05, "loss": 0.33791408538818357, "memory(GiB)": 61.91, "step": 24320, "token_acc": 0.9299363057324841, "train_speed(iter/s)": 1.464661 }, { "epoch": 1.0421575767961955, "grad_norm": 5.097473621368408, "learning_rate": 8.965827656754557e-05, "loss": 0.39647159576416013, "memory(GiB)": 61.91, "step": 24325, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.464656 }, { "epoch": 1.0423717921254445, "grad_norm": 6.352043628692627, "learning_rate": 8.965417773544207e-05, "loss": 0.6294098377227784, "memory(GiB)": 61.91, "step": 24330, "token_acc": 0.8738738738738738, "train_speed(iter/s)": 1.464623 }, { "epoch": 1.0425860074546935, "grad_norm": 0.5737296342849731, "learning_rate": 8.965007818496593e-05, "loss": 0.20080997943878173, "memory(GiB)": 61.91, "step": 24335, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.464614 }, { "epoch": 1.0428002227839424, "grad_norm": 1.7697242498397827, "learning_rate": 8.964597791619145e-05, "loss": 0.38774738311767576, "memory(GiB)": 61.91, "step": 24340, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.464612 }, { "epoch": 1.0430144381131914, "grad_norm": 3.4896583557128906, "learning_rate": 8.964187692919288e-05, "loss": 0.374206018447876, "memory(GiB)": 61.91, "step": 24345, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.46461 }, { "epoch": 1.0432286534424404, "grad_norm": 2.0575408935546875, "learning_rate": 8.963777522404451e-05, "loss": 0.17304366827011108, "memory(GiB)": 61.91, "step": 24350, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.464627 }, { "epoch": 1.0434428687716892, "grad_norm": 2.240955114364624, "learning_rate": 8.963367280082067e-05, "loss": 0.4267293453216553, "memory(GiB)": 61.91, "step": 24355, "token_acc": 0.8959537572254336, "train_speed(iter/s)": 1.464621 }, { "epoch": 1.0436570841009383, "grad_norm": 4.518520832061768, "learning_rate": 8.962956965959568e-05, "loss": 0.43558921813964846, "memory(GiB)": 61.91, "step": 24360, "token_acc": 0.8878787878787879, "train_speed(iter/s)": 1.464657 }, { "epoch": 1.0438712994301873, "grad_norm": 2.9348843097686768, "learning_rate": 8.962546580044384e-05, "loss": 0.32297143936157224, "memory(GiB)": 61.91, "step": 24365, "token_acc": 0.9331476323119777, "train_speed(iter/s)": 1.464656 }, { "epoch": 1.0440855147594361, "grad_norm": 2.9803168773651123, "learning_rate": 8.962136122343952e-05, "loss": 0.4230632781982422, "memory(GiB)": 61.91, "step": 24370, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.464657 }, { "epoch": 1.0442997300886852, "grad_norm": 4.076694011688232, "learning_rate": 8.961725592865708e-05, "loss": 0.2613226890563965, "memory(GiB)": 61.91, "step": 24375, "token_acc": 0.9355828220858896, "train_speed(iter/s)": 1.464642 }, { "epoch": 1.0445139454179342, "grad_norm": 4.294290065765381, "learning_rate": 8.961314991617089e-05, "loss": 0.23920989036560059, "memory(GiB)": 61.91, "step": 24380, "token_acc": 0.939873417721519, "train_speed(iter/s)": 1.464635 }, { "epoch": 1.044728160747183, "grad_norm": 5.321024417877197, "learning_rate": 8.960904318605532e-05, "loss": 0.2719660043716431, "memory(GiB)": 61.91, "step": 24385, "token_acc": 0.9727626459143969, "train_speed(iter/s)": 1.464634 }, { "epoch": 1.044942376076432, "grad_norm": 4.041585445404053, "learning_rate": 8.960493573838477e-05, "loss": 0.21441121101379396, "memory(GiB)": 61.91, "step": 24390, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.464629 }, { "epoch": 1.045156591405681, "grad_norm": 1.3916192054748535, "learning_rate": 8.960082757323367e-05, "loss": 0.3945186614990234, "memory(GiB)": 61.91, "step": 24395, "token_acc": 0.9291338582677166, "train_speed(iter/s)": 1.46463 }, { "epoch": 1.0453708067349299, "grad_norm": 5.567774295806885, "learning_rate": 8.95967186906764e-05, "loss": 0.7327682495117187, "memory(GiB)": 61.91, "step": 24400, "token_acc": 0.8651315789473685, "train_speed(iter/s)": 1.464625 }, { "epoch": 1.045585022064179, "grad_norm": 3.443730592727661, "learning_rate": 8.959260909078746e-05, "loss": 0.45744829177856444, "memory(GiB)": 61.91, "step": 24405, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.46464 }, { "epoch": 1.045799237393428, "grad_norm": 6.3759589195251465, "learning_rate": 8.958849877364125e-05, "loss": 0.26079578399658204, "memory(GiB)": 61.91, "step": 24410, "token_acc": 0.9395973154362416, "train_speed(iter/s)": 1.464627 }, { "epoch": 1.0460134527226768, "grad_norm": 5.873168468475342, "learning_rate": 8.958438773931226e-05, "loss": 0.3855307579040527, "memory(GiB)": 61.91, "step": 24415, "token_acc": 0.9154228855721394, "train_speed(iter/s)": 1.464609 }, { "epoch": 1.0462276680519258, "grad_norm": 4.777133941650391, "learning_rate": 8.958027598787495e-05, "loss": 0.41317110061645507, "memory(GiB)": 61.91, "step": 24420, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.4646 }, { "epoch": 1.0464418833811748, "grad_norm": 3.342287540435791, "learning_rate": 8.957616351940381e-05, "loss": 0.3870342493057251, "memory(GiB)": 61.91, "step": 24425, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.46459 }, { "epoch": 1.0466560987104236, "grad_norm": 4.336524963378906, "learning_rate": 8.957205033397333e-05, "loss": 0.22227773666381836, "memory(GiB)": 61.91, "step": 24430, "token_acc": 0.952191235059761, "train_speed(iter/s)": 1.464584 }, { "epoch": 1.0468703140396727, "grad_norm": 1.838070034980774, "learning_rate": 8.956793643165805e-05, "loss": 0.30626659393310546, "memory(GiB)": 61.91, "step": 24435, "token_acc": 0.9369627507163324, "train_speed(iter/s)": 1.464588 }, { "epoch": 1.0470845293689217, "grad_norm": 3.0602188110351562, "learning_rate": 8.956382181253248e-05, "loss": 0.5171976089477539, "memory(GiB)": 61.91, "step": 24440, "token_acc": 0.889967637540453, "train_speed(iter/s)": 1.464615 }, { "epoch": 1.0472987446981705, "grad_norm": 5.2733588218688965, "learning_rate": 8.955970647667115e-05, "loss": 0.4559472560882568, "memory(GiB)": 61.91, "step": 24445, "token_acc": 0.9107142857142857, "train_speed(iter/s)": 1.464636 }, { "epoch": 1.0475129600274196, "grad_norm": 1.6563656330108643, "learning_rate": 8.955559042414865e-05, "loss": 0.14235712289810182, "memory(GiB)": 61.91, "step": 24450, "token_acc": 0.9703947368421053, "train_speed(iter/s)": 1.464638 }, { "epoch": 1.0477271753566686, "grad_norm": 4.546403408050537, "learning_rate": 8.955147365503952e-05, "loss": 0.3481938362121582, "memory(GiB)": 61.91, "step": 24455, "token_acc": 0.9240924092409241, "train_speed(iter/s)": 1.464659 }, { "epoch": 1.0479413906859174, "grad_norm": 5.60993766784668, "learning_rate": 8.954735616941834e-05, "loss": 0.28456690311431887, "memory(GiB)": 61.91, "step": 24460, "token_acc": 0.9446808510638298, "train_speed(iter/s)": 1.464656 }, { "epoch": 1.0481556060151664, "grad_norm": 3.9068949222564697, "learning_rate": 8.954323796735969e-05, "loss": 0.23663861751556398, "memory(GiB)": 61.91, "step": 24465, "token_acc": 0.9471544715447154, "train_speed(iter/s)": 1.46468 }, { "epoch": 1.0483698213444155, "grad_norm": 2.339041233062744, "learning_rate": 8.953911904893821e-05, "loss": 0.4617782115936279, "memory(GiB)": 61.91, "step": 24470, "token_acc": 0.8938906752411575, "train_speed(iter/s)": 1.464704 }, { "epoch": 1.0485840366736643, "grad_norm": 2.016204833984375, "learning_rate": 8.953499941422849e-05, "loss": 0.5554094791412354, "memory(GiB)": 61.91, "step": 24475, "token_acc": 0.8705501618122977, "train_speed(iter/s)": 1.464706 }, { "epoch": 1.0487982520029133, "grad_norm": 3.9696414470672607, "learning_rate": 8.953087906330514e-05, "loss": 0.4019582271575928, "memory(GiB)": 61.91, "step": 24480, "token_acc": 0.9202898550724637, "train_speed(iter/s)": 1.464708 }, { "epoch": 1.0490124673321624, "grad_norm": 5.826642990112305, "learning_rate": 8.952675799624288e-05, "loss": 0.513193416595459, "memory(GiB)": 61.91, "step": 24485, "token_acc": 0.8874172185430463, "train_speed(iter/s)": 1.464686 }, { "epoch": 1.0492266826614112, "grad_norm": 2.8811776638031006, "learning_rate": 8.952263621311629e-05, "loss": 0.28107781410217286, "memory(GiB)": 61.91, "step": 24490, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.464705 }, { "epoch": 1.0494408979906602, "grad_norm": 3.4632833003997803, "learning_rate": 8.951851371400008e-05, "loss": 0.27753098011016847, "memory(GiB)": 61.91, "step": 24495, "token_acc": 0.9375, "train_speed(iter/s)": 1.464717 }, { "epoch": 1.0496551133199092, "grad_norm": 2.1803221702575684, "learning_rate": 8.951439049896892e-05, "loss": 0.46370563507080076, "memory(GiB)": 61.91, "step": 24500, "token_acc": 0.9003215434083601, "train_speed(iter/s)": 1.464726 }, { "epoch": 1.0496551133199092, "eval_loss": 2.581887722015381, "eval_runtime": 13.8188, "eval_samples_per_second": 7.237, "eval_steps_per_second": 7.237, "eval_token_acc": 0.42140921409214094, "step": 24500 }, { "epoch": 1.049869328649158, "grad_norm": 4.523056507110596, "learning_rate": 8.951026656809753e-05, "loss": 0.3761757850646973, "memory(GiB)": 61.91, "step": 24505, "token_acc": 0.5439672801635992, "train_speed(iter/s)": 1.463442 }, { "epoch": 1.050083543978407, "grad_norm": 3.1562626361846924, "learning_rate": 8.950614192146058e-05, "loss": 0.26401805877685547, "memory(GiB)": 61.91, "step": 24510, "token_acc": 0.9401993355481728, "train_speed(iter/s)": 1.463445 }, { "epoch": 1.0502977593076561, "grad_norm": 3.197272539138794, "learning_rate": 8.950201655913279e-05, "loss": 0.42534689903259276, "memory(GiB)": 61.91, "step": 24515, "token_acc": 0.9128787878787878, "train_speed(iter/s)": 1.46345 }, { "epoch": 1.050511974636905, "grad_norm": 2.8793113231658936, "learning_rate": 8.949789048118894e-05, "loss": 0.5593438148498535, "memory(GiB)": 61.91, "step": 24520, "token_acc": 0.8961937716262975, "train_speed(iter/s)": 1.463472 }, { "epoch": 1.050726189966154, "grad_norm": 3.943530797958374, "learning_rate": 8.949376368770375e-05, "loss": 0.3187270641326904, "memory(GiB)": 61.91, "step": 24525, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.463474 }, { "epoch": 1.050940405295403, "grad_norm": 4.713124752044678, "learning_rate": 8.948963617875197e-05, "loss": 0.5596467018127441, "memory(GiB)": 61.91, "step": 24530, "token_acc": 0.8941605839416058, "train_speed(iter/s)": 1.463447 }, { "epoch": 1.0511546206246518, "grad_norm": 4.850879192352295, "learning_rate": 8.94855079544084e-05, "loss": 0.2769219160079956, "memory(GiB)": 61.91, "step": 24535, "token_acc": 0.9401993355481728, "train_speed(iter/s)": 1.463448 }, { "epoch": 1.0513688359539008, "grad_norm": 0.12854355573654175, "learning_rate": 8.94813790147478e-05, "loss": 0.2537151098251343, "memory(GiB)": 61.91, "step": 24540, "token_acc": 0.9453781512605042, "train_speed(iter/s)": 1.463419 }, { "epoch": 1.0515830512831499, "grad_norm": 2.161783218383789, "learning_rate": 8.947724935984499e-05, "loss": 0.20682380199432374, "memory(GiB)": 61.91, "step": 24545, "token_acc": 0.94375, "train_speed(iter/s)": 1.46341 }, { "epoch": 1.0517972666123987, "grad_norm": 2.990401029586792, "learning_rate": 8.947311898977477e-05, "loss": 0.48949708938598635, "memory(GiB)": 61.91, "step": 24550, "token_acc": 0.8963414634146342, "train_speed(iter/s)": 1.463395 }, { "epoch": 1.0520114819416477, "grad_norm": 4.989822864532471, "learning_rate": 8.946898790461197e-05, "loss": 0.7246774196624756, "memory(GiB)": 61.91, "step": 24555, "token_acc": 0.8481848184818482, "train_speed(iter/s)": 1.463419 }, { "epoch": 1.0522256972708968, "grad_norm": 0.16161809861660004, "learning_rate": 8.946485610443144e-05, "loss": 0.5387742996215821, "memory(GiB)": 61.91, "step": 24560, "token_acc": 0.8896321070234113, "train_speed(iter/s)": 1.463421 }, { "epoch": 1.0524399126001456, "grad_norm": 4.49443244934082, "learning_rate": 8.946072358930802e-05, "loss": 0.5307668209075928, "memory(GiB)": 61.91, "step": 24565, "token_acc": 0.8651026392961877, "train_speed(iter/s)": 1.463423 }, { "epoch": 1.0526541279293946, "grad_norm": 11.624452590942383, "learning_rate": 8.945659035931656e-05, "loss": 0.47968149185180664, "memory(GiB)": 61.91, "step": 24570, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.463432 }, { "epoch": 1.0528683432586436, "grad_norm": 2.9155590534210205, "learning_rate": 8.945245641453197e-05, "loss": 0.5393363475799561, "memory(GiB)": 61.91, "step": 24575, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.46345 }, { "epoch": 1.0530825585878925, "grad_norm": 6.699670791625977, "learning_rate": 8.94483217550291e-05, "loss": 0.5843441963195801, "memory(GiB)": 61.91, "step": 24580, "token_acc": 0.8904109589041096, "train_speed(iter/s)": 1.463466 }, { "epoch": 1.0532967739171415, "grad_norm": 3.1946828365325928, "learning_rate": 8.94441863808829e-05, "loss": 0.457660436630249, "memory(GiB)": 61.91, "step": 24585, "token_acc": 0.8778135048231511, "train_speed(iter/s)": 1.463462 }, { "epoch": 1.0535109892463905, "grad_norm": 4.133587837219238, "learning_rate": 8.944005029216824e-05, "loss": 0.4464540958404541, "memory(GiB)": 61.91, "step": 24590, "token_acc": 0.9171974522292994, "train_speed(iter/s)": 1.463466 }, { "epoch": 1.0537252045756393, "grad_norm": 4.191961765289307, "learning_rate": 8.94359134889601e-05, "loss": 0.5787910461425781, "memory(GiB)": 61.91, "step": 24595, "token_acc": 0.8698630136986302, "train_speed(iter/s)": 1.463458 }, { "epoch": 1.0539394199048884, "grad_norm": 1.294704556465149, "learning_rate": 8.943177597133336e-05, "loss": 0.2509007453918457, "memory(GiB)": 61.91, "step": 24600, "token_acc": 0.9386503067484663, "train_speed(iter/s)": 1.463462 }, { "epoch": 1.0541536352341374, "grad_norm": 5.287006378173828, "learning_rate": 8.942763773936304e-05, "loss": 0.3590357780456543, "memory(GiB)": 61.91, "step": 24605, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.46346 }, { "epoch": 1.0543678505633862, "grad_norm": 6.0106120109558105, "learning_rate": 8.942349879312406e-05, "loss": 0.3875889301300049, "memory(GiB)": 61.91, "step": 24610, "token_acc": 0.9087136929460581, "train_speed(iter/s)": 1.463465 }, { "epoch": 1.0545820658926353, "grad_norm": 3.205077648162842, "learning_rate": 8.941935913269142e-05, "loss": 0.15592634677886963, "memory(GiB)": 61.91, "step": 24615, "token_acc": 0.9535714285714286, "train_speed(iter/s)": 1.463451 }, { "epoch": 1.0547962812218843, "grad_norm": 2.217055082321167, "learning_rate": 8.94152187581401e-05, "loss": 0.3313002109527588, "memory(GiB)": 61.91, "step": 24620, "token_acc": 0.9335260115606936, "train_speed(iter/s)": 1.463441 }, { "epoch": 1.055010496551133, "grad_norm": 2.336862325668335, "learning_rate": 8.941107766954513e-05, "loss": 0.2223813056945801, "memory(GiB)": 61.91, "step": 24625, "token_acc": 0.9491017964071856, "train_speed(iter/s)": 1.463429 }, { "epoch": 1.0552247118803821, "grad_norm": 2.2205827236175537, "learning_rate": 8.940693586698152e-05, "loss": 0.26647419929504396, "memory(GiB)": 61.91, "step": 24630, "token_acc": 0.9530685920577617, "train_speed(iter/s)": 1.463443 }, { "epoch": 1.0554389272096312, "grad_norm": 2.743354558944702, "learning_rate": 8.940279335052428e-05, "loss": 0.6133790016174316, "memory(GiB)": 61.91, "step": 24635, "token_acc": 0.8858024691358025, "train_speed(iter/s)": 1.463469 }, { "epoch": 1.05565314253888, "grad_norm": 1.262387990951538, "learning_rate": 8.939865012024849e-05, "loss": 0.3659468412399292, "memory(GiB)": 61.91, "step": 24640, "token_acc": 0.9316239316239316, "train_speed(iter/s)": 1.463446 }, { "epoch": 1.055867357868129, "grad_norm": 5.1507344245910645, "learning_rate": 8.939450617622919e-05, "loss": 0.13844575881958007, "memory(GiB)": 61.91, "step": 24645, "token_acc": 0.9630872483221476, "train_speed(iter/s)": 1.463467 }, { "epoch": 1.056081573197378, "grad_norm": 1.9977318048477173, "learning_rate": 8.939036151854147e-05, "loss": 0.46593294143676756, "memory(GiB)": 61.91, "step": 24650, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.463456 }, { "epoch": 1.0562957885266269, "grad_norm": 0.044040482491254807, "learning_rate": 8.93862161472604e-05, "loss": 0.540539026260376, "memory(GiB)": 61.91, "step": 24655, "token_acc": 0.8840579710144928, "train_speed(iter/s)": 1.463463 }, { "epoch": 1.056510003855876, "grad_norm": 3.416168689727783, "learning_rate": 8.938207006246106e-05, "loss": 0.3566168785095215, "memory(GiB)": 61.91, "step": 24660, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.463465 }, { "epoch": 1.056724219185125, "grad_norm": 2.509399175643921, "learning_rate": 8.937792326421859e-05, "loss": 0.4582388401031494, "memory(GiB)": 61.91, "step": 24665, "token_acc": 0.8873720136518771, "train_speed(iter/s)": 1.463457 }, { "epoch": 1.0569384345143737, "grad_norm": 1.3250961303710938, "learning_rate": 8.937377575260809e-05, "loss": 0.34017324447631836, "memory(GiB)": 61.91, "step": 24670, "token_acc": 0.9087947882736156, "train_speed(iter/s)": 1.463459 }, { "epoch": 1.0571526498436228, "grad_norm": 5.463022708892822, "learning_rate": 8.93696275277047e-05, "loss": 0.5743795394897461, "memory(GiB)": 61.91, "step": 24675, "token_acc": 0.8858858858858859, "train_speed(iter/s)": 1.463465 }, { "epoch": 1.0573668651728718, "grad_norm": 3.259611129760742, "learning_rate": 8.93654785895836e-05, "loss": 0.3714089870452881, "memory(GiB)": 61.91, "step": 24680, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.463475 }, { "epoch": 1.0575810805021206, "grad_norm": 3.1880557537078857, "learning_rate": 8.936132893831992e-05, "loss": 0.36184766292572024, "memory(GiB)": 61.91, "step": 24685, "token_acc": 0.899390243902439, "train_speed(iter/s)": 1.463467 }, { "epoch": 1.0577952958313697, "grad_norm": 4.762054920196533, "learning_rate": 8.935717857398883e-05, "loss": 0.5450374126434326, "memory(GiB)": 61.91, "step": 24690, "token_acc": 0.9, "train_speed(iter/s)": 1.463451 }, { "epoch": 1.0580095111606187, "grad_norm": 4.3264994621276855, "learning_rate": 8.935302749666554e-05, "loss": 0.4252921581268311, "memory(GiB)": 61.91, "step": 24695, "token_acc": 0.9156976744186046, "train_speed(iter/s)": 1.463448 }, { "epoch": 1.0582237264898675, "grad_norm": 0.6226086616516113, "learning_rate": 8.934887570642523e-05, "loss": 0.305343770980835, "memory(GiB)": 61.91, "step": 24700, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.463464 }, { "epoch": 1.0584379418191165, "grad_norm": 4.047313213348389, "learning_rate": 8.934472320334312e-05, "loss": 0.4131124019622803, "memory(GiB)": 61.91, "step": 24705, "token_acc": 0.9039735099337748, "train_speed(iter/s)": 1.463454 }, { "epoch": 1.0586521571483656, "grad_norm": 2.9302256107330322, "learning_rate": 8.934056998749447e-05, "loss": 0.49488101005554197, "memory(GiB)": 61.91, "step": 24710, "token_acc": 0.8588588588588588, "train_speed(iter/s)": 1.463481 }, { "epoch": 1.0588663724776144, "grad_norm": 2.868764638900757, "learning_rate": 8.933641605895447e-05, "loss": 0.433966064453125, "memory(GiB)": 61.91, "step": 24715, "token_acc": 0.8957654723127035, "train_speed(iter/s)": 1.463515 }, { "epoch": 1.0590805878068634, "grad_norm": 8.793230056762695, "learning_rate": 8.93322614177984e-05, "loss": 0.3274407386779785, "memory(GiB)": 61.91, "step": 24720, "token_acc": 0.9357429718875502, "train_speed(iter/s)": 1.463518 }, { "epoch": 1.0592948031361125, "grad_norm": 2.3424408435821533, "learning_rate": 8.932810606410151e-05, "loss": 0.304095196723938, "memory(GiB)": 61.91, "step": 24725, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.463499 }, { "epoch": 1.0595090184653613, "grad_norm": 1.8642646074295044, "learning_rate": 8.932394999793909e-05, "loss": 0.40365915298461913, "memory(GiB)": 61.91, "step": 24730, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.463495 }, { "epoch": 1.0597232337946103, "grad_norm": 3.6033267974853516, "learning_rate": 8.931979321938643e-05, "loss": 0.500331974029541, "memory(GiB)": 61.91, "step": 24735, "token_acc": 0.8926380368098159, "train_speed(iter/s)": 1.463513 }, { "epoch": 1.0599374491238593, "grad_norm": 2.8488059043884277, "learning_rate": 8.931563572851883e-05, "loss": 0.26050543785095215, "memory(GiB)": 61.91, "step": 24740, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.463504 }, { "epoch": 1.0601516644531084, "grad_norm": 5.4557108879089355, "learning_rate": 8.93114775254116e-05, "loss": 0.7832996368408203, "memory(GiB)": 61.91, "step": 24745, "token_acc": 0.8533333333333334, "train_speed(iter/s)": 1.463521 }, { "epoch": 1.0603658797823572, "grad_norm": 1.754604697227478, "learning_rate": 8.930731861014009e-05, "loss": 0.27332515716552735, "memory(GiB)": 61.91, "step": 24750, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.463509 }, { "epoch": 1.0605800951116062, "grad_norm": 1.303481101989746, "learning_rate": 8.930315898277961e-05, "loss": 0.18530147075653075, "memory(GiB)": 61.91, "step": 24755, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.463532 }, { "epoch": 1.0607943104408553, "grad_norm": 4.179767608642578, "learning_rate": 8.929899864340556e-05, "loss": 0.45236945152282715, "memory(GiB)": 61.91, "step": 24760, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.463544 }, { "epoch": 1.061008525770104, "grad_norm": 3.072788715362549, "learning_rate": 8.929483759209327e-05, "loss": 0.19269709587097167, "memory(GiB)": 61.91, "step": 24765, "token_acc": 0.9718875502008032, "train_speed(iter/s)": 1.463551 }, { "epoch": 1.061222741099353, "grad_norm": 4.7244486808776855, "learning_rate": 8.929067582891813e-05, "loss": 0.5001522541046143, "memory(GiB)": 61.91, "step": 24770, "token_acc": 0.9096385542168675, "train_speed(iter/s)": 1.463553 }, { "epoch": 1.0614369564286021, "grad_norm": 4.894017696380615, "learning_rate": 8.928651335395556e-05, "loss": 0.4955124378204346, "memory(GiB)": 61.91, "step": 24775, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.463574 }, { "epoch": 1.061651171757851, "grad_norm": 2.9096834659576416, "learning_rate": 8.928235016728093e-05, "loss": 0.40262584686279296, "memory(GiB)": 61.91, "step": 24780, "token_acc": 0.9034267912772586, "train_speed(iter/s)": 1.463568 }, { "epoch": 1.0618653870871, "grad_norm": 2.329826593399048, "learning_rate": 8.927818626896968e-05, "loss": 0.46176509857177733, "memory(GiB)": 61.91, "step": 24785, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.463593 }, { "epoch": 1.062079602416349, "grad_norm": 2.8342161178588867, "learning_rate": 8.927402165909724e-05, "loss": 0.45001873970031736, "memory(GiB)": 61.91, "step": 24790, "token_acc": 0.9145569620253164, "train_speed(iter/s)": 1.463582 }, { "epoch": 1.0622938177455978, "grad_norm": 2.7624945640563965, "learning_rate": 8.926985633773906e-05, "loss": 0.6400355815887451, "memory(GiB)": 61.91, "step": 24795, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.463605 }, { "epoch": 1.0625080330748469, "grad_norm": 4.156216144561768, "learning_rate": 8.92656903049706e-05, "loss": 0.22993097305297852, "memory(GiB)": 61.91, "step": 24800, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.463611 }, { "epoch": 1.062722248404096, "grad_norm": 5.344137191772461, "learning_rate": 8.92615235608673e-05, "loss": 0.46294326782226564, "memory(GiB)": 61.91, "step": 24805, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.463669 }, { "epoch": 1.0629364637333447, "grad_norm": 4.553225517272949, "learning_rate": 8.925735610550469e-05, "loss": 0.5529368400573731, "memory(GiB)": 61.91, "step": 24810, "token_acc": 0.861764705882353, "train_speed(iter/s)": 1.463684 }, { "epoch": 1.0631506790625938, "grad_norm": 2.958220958709717, "learning_rate": 8.925318793895825e-05, "loss": 0.36243481636047364, "memory(GiB)": 61.91, "step": 24815, "token_acc": 0.8972602739726028, "train_speed(iter/s)": 1.463697 }, { "epoch": 1.0633648943918428, "grad_norm": 3.3008368015289307, "learning_rate": 8.924901906130348e-05, "loss": 0.47784738540649413, "memory(GiB)": 61.91, "step": 24820, "token_acc": 0.9225806451612903, "train_speed(iter/s)": 1.463685 }, { "epoch": 1.0635791097210916, "grad_norm": 2.8839049339294434, "learning_rate": 8.924484947261592e-05, "loss": 0.3755256414413452, "memory(GiB)": 61.91, "step": 24825, "token_acc": 0.9206896551724137, "train_speed(iter/s)": 1.463681 }, { "epoch": 1.0637933250503406, "grad_norm": 6.298213005065918, "learning_rate": 8.92406791729711e-05, "loss": 0.4611805438995361, "memory(GiB)": 61.91, "step": 24830, "token_acc": 0.8812949640287769, "train_speed(iter/s)": 1.463658 }, { "epoch": 1.0640075403795897, "grad_norm": 6.608723163604736, "learning_rate": 8.923650816244455e-05, "loss": 0.5829193115234375, "memory(GiB)": 61.91, "step": 24835, "token_acc": 0.8745762711864407, "train_speed(iter/s)": 1.463658 }, { "epoch": 1.0642217557088385, "grad_norm": 3.4404544830322266, "learning_rate": 8.923233644111187e-05, "loss": 0.4655473709106445, "memory(GiB)": 61.91, "step": 24840, "token_acc": 0.8817204301075269, "train_speed(iter/s)": 1.463653 }, { "epoch": 1.0644359710380875, "grad_norm": 4.406962871551514, "learning_rate": 8.922816400904859e-05, "loss": 0.6765220642089844, "memory(GiB)": 61.91, "step": 24845, "token_acc": 0.8525641025641025, "train_speed(iter/s)": 1.463643 }, { "epoch": 1.0646501863673365, "grad_norm": 0.7809498310089111, "learning_rate": 8.922399086633032e-05, "loss": 0.2585753440856934, "memory(GiB)": 61.91, "step": 24850, "token_acc": 0.9291338582677166, "train_speed(iter/s)": 1.463636 }, { "epoch": 1.0648644016965854, "grad_norm": 4.0026655197143555, "learning_rate": 8.921981701303267e-05, "loss": 0.20240280628204346, "memory(GiB)": 61.91, "step": 24855, "token_acc": 0.9742647058823529, "train_speed(iter/s)": 1.463652 }, { "epoch": 1.0650786170258344, "grad_norm": 2.499643564224243, "learning_rate": 8.921564244923124e-05, "loss": 0.3598951816558838, "memory(GiB)": 61.91, "step": 24860, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.463646 }, { "epoch": 1.0652928323550834, "grad_norm": 2.647608757019043, "learning_rate": 8.921146717500166e-05, "loss": 0.5103109359741211, "memory(GiB)": 61.91, "step": 24865, "token_acc": 0.8854625550660793, "train_speed(iter/s)": 1.463641 }, { "epoch": 1.0655070476843322, "grad_norm": 3.645301580429077, "learning_rate": 8.920729119041957e-05, "loss": 0.40702085494995116, "memory(GiB)": 61.91, "step": 24870, "token_acc": 0.901060070671378, "train_speed(iter/s)": 1.463654 }, { "epoch": 1.0657212630135813, "grad_norm": 0.49799811840057373, "learning_rate": 8.920311449556062e-05, "loss": 0.2589221239089966, "memory(GiB)": 61.91, "step": 24875, "token_acc": 0.9420731707317073, "train_speed(iter/s)": 1.463635 }, { "epoch": 1.0659354783428303, "grad_norm": 2.3787267208099365, "learning_rate": 8.919893709050049e-05, "loss": 0.5414438247680664, "memory(GiB)": 61.91, "step": 24880, "token_acc": 0.8972809667673716, "train_speed(iter/s)": 1.463638 }, { "epoch": 1.0661496936720791, "grad_norm": 4.094865322113037, "learning_rate": 8.919475897531482e-05, "loss": 0.29633493423461915, "memory(GiB)": 61.91, "step": 24885, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.46363 }, { "epoch": 1.0663639090013282, "grad_norm": 4.348586559295654, "learning_rate": 8.919058015007934e-05, "loss": 0.4576865196228027, "memory(GiB)": 61.91, "step": 24890, "token_acc": 0.9, "train_speed(iter/s)": 1.463633 }, { "epoch": 1.0665781243305772, "grad_norm": 4.508138179779053, "learning_rate": 8.918640061486974e-05, "loss": 0.4288358211517334, "memory(GiB)": 61.91, "step": 24895, "token_acc": 0.9131832797427653, "train_speed(iter/s)": 1.463666 }, { "epoch": 1.066792339659826, "grad_norm": 6.435828685760498, "learning_rate": 8.918222036976172e-05, "loss": 0.5891474723815918, "memory(GiB)": 61.91, "step": 24900, "token_acc": 0.8716216216216216, "train_speed(iter/s)": 1.463683 }, { "epoch": 1.067006554989075, "grad_norm": 3.5111541748046875, "learning_rate": 8.917803941483101e-05, "loss": 0.30353994369506837, "memory(GiB)": 61.91, "step": 24905, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.463683 }, { "epoch": 1.067220770318324, "grad_norm": 1.9401183128356934, "learning_rate": 8.917385775015338e-05, "loss": 0.34528138637542727, "memory(GiB)": 61.91, "step": 24910, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.46367 }, { "epoch": 1.0674349856475729, "grad_norm": 9.288268089294434, "learning_rate": 8.916967537580457e-05, "loss": 0.329301643371582, "memory(GiB)": 61.91, "step": 24915, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.463708 }, { "epoch": 1.067649200976822, "grad_norm": 4.7642717361450195, "learning_rate": 8.916549229186036e-05, "loss": 0.3683889150619507, "memory(GiB)": 61.91, "step": 24920, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.463718 }, { "epoch": 1.067863416306071, "grad_norm": 3.6249523162841797, "learning_rate": 8.916130849839649e-05, "loss": 0.3878005504608154, "memory(GiB)": 61.91, "step": 24925, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.463686 }, { "epoch": 1.0680776316353198, "grad_norm": 3.421154022216797, "learning_rate": 8.915712399548879e-05, "loss": 0.2848489761352539, "memory(GiB)": 61.91, "step": 24930, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.463682 }, { "epoch": 1.0682918469645688, "grad_norm": 8.362064361572266, "learning_rate": 8.915293878321308e-05, "loss": 0.2502378225326538, "memory(GiB)": 61.91, "step": 24935, "token_acc": 0.9446640316205533, "train_speed(iter/s)": 1.463675 }, { "epoch": 1.0685060622938178, "grad_norm": 2.1216957569122314, "learning_rate": 8.914875286164512e-05, "loss": 0.3043825626373291, "memory(GiB)": 61.91, "step": 24940, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.463672 }, { "epoch": 1.0687202776230667, "grad_norm": 4.274103164672852, "learning_rate": 8.914456623086078e-05, "loss": 0.6435746192932129, "memory(GiB)": 61.91, "step": 24945, "token_acc": 0.8384615384615385, "train_speed(iter/s)": 1.463671 }, { "epoch": 1.0689344929523157, "grad_norm": 3.7751007080078125, "learning_rate": 8.914037889093591e-05, "loss": 0.4034872055053711, "memory(GiB)": 61.91, "step": 24950, "token_acc": 0.9158576051779935, "train_speed(iter/s)": 1.463667 }, { "epoch": 1.0691487082815647, "grad_norm": 5.098366737365723, "learning_rate": 8.913619084194638e-05, "loss": 0.48452329635620117, "memory(GiB)": 61.91, "step": 24955, "token_acc": 0.9098591549295775, "train_speed(iter/s)": 1.463713 }, { "epoch": 1.0693629236108135, "grad_norm": 4.837723255157471, "learning_rate": 8.913200208396801e-05, "loss": 0.7379373550415039, "memory(GiB)": 61.91, "step": 24960, "token_acc": 0.8547008547008547, "train_speed(iter/s)": 1.463693 }, { "epoch": 1.0695771389400626, "grad_norm": 4.730515480041504, "learning_rate": 8.912781261707674e-05, "loss": 0.37705981731414795, "memory(GiB)": 61.91, "step": 24965, "token_acc": 0.9175531914893617, "train_speed(iter/s)": 1.463707 }, { "epoch": 1.0697913542693116, "grad_norm": 0.41072091460227966, "learning_rate": 8.912362244134842e-05, "loss": 0.18380309343338014, "memory(GiB)": 61.91, "step": 24970, "token_acc": 0.9589905362776026, "train_speed(iter/s)": 1.463735 }, { "epoch": 1.0700055695985604, "grad_norm": 3.805476665496826, "learning_rate": 8.911943155685897e-05, "loss": 0.17919005155563356, "memory(GiB)": 61.91, "step": 24975, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.463739 }, { "epoch": 1.0702197849278094, "grad_norm": 1.8773678541183472, "learning_rate": 8.911523996368434e-05, "loss": 0.18148293495178222, "memory(GiB)": 61.91, "step": 24980, "token_acc": 0.9683098591549296, "train_speed(iter/s)": 1.463764 }, { "epoch": 1.0704340002570585, "grad_norm": 2.188059091567993, "learning_rate": 8.911104766190045e-05, "loss": 0.23359532356262208, "memory(GiB)": 61.91, "step": 24985, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.463792 }, { "epoch": 1.0706482155863073, "grad_norm": 3.9614980220794678, "learning_rate": 8.910685465158324e-05, "loss": 0.401519775390625, "memory(GiB)": 61.91, "step": 24990, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.463803 }, { "epoch": 1.0708624309155563, "grad_norm": 4.403512001037598, "learning_rate": 8.910266093280866e-05, "loss": 0.3135676860809326, "memory(GiB)": 61.91, "step": 24995, "token_acc": 0.9282868525896414, "train_speed(iter/s)": 1.463794 }, { "epoch": 1.0710766462448054, "grad_norm": 2.090351104736328, "learning_rate": 8.90984665056527e-05, "loss": 0.4066732406616211, "memory(GiB)": 61.91, "step": 25000, "token_acc": 0.9172932330827067, "train_speed(iter/s)": 1.463792 }, { "epoch": 1.0710766462448054, "eval_loss": 2.409970760345459, "eval_runtime": 13.7652, "eval_samples_per_second": 7.265, "eval_steps_per_second": 7.265, "eval_token_acc": 0.4330042313117066, "step": 25000 }, { "epoch": 1.0712908615740542, "grad_norm": 0.228685662150383, "learning_rate": 8.909427137019136e-05, "loss": 0.15805572271347046, "memory(GiB)": 61.91, "step": 25005, "token_acc": 0.5811359026369168, "train_speed(iter/s)": 1.462527 }, { "epoch": 1.0715050769033032, "grad_norm": 4.091732501983643, "learning_rate": 8.90900755265006e-05, "loss": 0.5353171825408936, "memory(GiB)": 61.91, "step": 25010, "token_acc": 0.8817204301075269, "train_speed(iter/s)": 1.462506 }, { "epoch": 1.0717192922325522, "grad_norm": 3.826026439666748, "learning_rate": 8.908587897465644e-05, "loss": 0.3465597152709961, "memory(GiB)": 61.91, "step": 25015, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.462524 }, { "epoch": 1.071933507561801, "grad_norm": 4.4526848793029785, "learning_rate": 8.908168171473494e-05, "loss": 0.6048482894897461, "memory(GiB)": 61.91, "step": 25020, "token_acc": 0.8590163934426229, "train_speed(iter/s)": 1.462535 }, { "epoch": 1.07214772289105, "grad_norm": 2.495469570159912, "learning_rate": 8.907748374681211e-05, "loss": 0.13592972755432128, "memory(GiB)": 61.91, "step": 25025, "token_acc": 0.9668874172185431, "train_speed(iter/s)": 1.462541 }, { "epoch": 1.0723619382202991, "grad_norm": 1.3913660049438477, "learning_rate": 8.907328507096399e-05, "loss": 0.5762334346771241, "memory(GiB)": 61.91, "step": 25030, "token_acc": 0.8781362007168458, "train_speed(iter/s)": 1.462569 }, { "epoch": 1.072576153549548, "grad_norm": 5.104248046875, "learning_rate": 8.906908568726669e-05, "loss": 0.35464026927948, "memory(GiB)": 61.91, "step": 25035, "token_acc": 0.9222520107238605, "train_speed(iter/s)": 1.462579 }, { "epoch": 1.072790368878797, "grad_norm": 2.8235175609588623, "learning_rate": 8.906488559579623e-05, "loss": 0.1746814727783203, "memory(GiB)": 61.91, "step": 25040, "token_acc": 0.9527272727272728, "train_speed(iter/s)": 1.462621 }, { "epoch": 1.073004584208046, "grad_norm": 2.6833760738372803, "learning_rate": 8.906068479662871e-05, "loss": 0.5139699935913086, "memory(GiB)": 61.91, "step": 25045, "token_acc": 0.9184397163120568, "train_speed(iter/s)": 1.462614 }, { "epoch": 1.0732187995372948, "grad_norm": 3.5112621784210205, "learning_rate": 8.905648328984026e-05, "loss": 0.35757970809936523, "memory(GiB)": 61.91, "step": 25050, "token_acc": 0.91015625, "train_speed(iter/s)": 1.46264 }, { "epoch": 1.0734330148665439, "grad_norm": 4.577869415283203, "learning_rate": 8.905228107550697e-05, "loss": 0.5617709159851074, "memory(GiB)": 61.91, "step": 25055, "token_acc": 0.8801261829652997, "train_speed(iter/s)": 1.462627 }, { "epoch": 1.073647230195793, "grad_norm": 0.9779202938079834, "learning_rate": 8.904807815370499e-05, "loss": 0.22143781185150146, "memory(GiB)": 61.91, "step": 25060, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.462627 }, { "epoch": 1.0738614455250417, "grad_norm": 2.0550382137298584, "learning_rate": 8.904387452451042e-05, "loss": 0.2850539445877075, "memory(GiB)": 61.91, "step": 25065, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.462631 }, { "epoch": 1.0740756608542907, "grad_norm": 4.75274658203125, "learning_rate": 8.903967018799946e-05, "loss": 0.3070229530334473, "memory(GiB)": 61.91, "step": 25070, "token_acc": 0.9343065693430657, "train_speed(iter/s)": 1.462621 }, { "epoch": 1.0742898761835398, "grad_norm": 4.14871072769165, "learning_rate": 8.903546514424824e-05, "loss": 0.443074369430542, "memory(GiB)": 61.91, "step": 25075, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.462644 }, { "epoch": 1.0745040915127886, "grad_norm": 0.5455694198608398, "learning_rate": 8.903125939333294e-05, "loss": 0.40418376922607424, "memory(GiB)": 61.91, "step": 25080, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.462643 }, { "epoch": 1.0747183068420376, "grad_norm": 0.7578938603401184, "learning_rate": 8.902705293532978e-05, "loss": 0.3441060781478882, "memory(GiB)": 61.91, "step": 25085, "token_acc": 0.933993399339934, "train_speed(iter/s)": 1.462668 }, { "epoch": 1.0749325221712867, "grad_norm": 3.748610019683838, "learning_rate": 8.902284577031494e-05, "loss": 0.22135050296783448, "memory(GiB)": 61.91, "step": 25090, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.462671 }, { "epoch": 1.0751467375005355, "grad_norm": 2.808198928833008, "learning_rate": 8.901863789836464e-05, "loss": 0.24704573154449463, "memory(GiB)": 61.91, "step": 25095, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.462685 }, { "epoch": 1.0753609528297845, "grad_norm": 2.515803337097168, "learning_rate": 8.901442931955512e-05, "loss": 0.3248444080352783, "memory(GiB)": 61.91, "step": 25100, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.462701 }, { "epoch": 1.0755751681590335, "grad_norm": 3.623593807220459, "learning_rate": 8.90102200339626e-05, "loss": 0.3366007566452026, "memory(GiB)": 61.91, "step": 25105, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.462734 }, { "epoch": 1.0757893834882823, "grad_norm": 5.1452741622924805, "learning_rate": 8.900601004166335e-05, "loss": 0.546142864227295, "memory(GiB)": 61.91, "step": 25110, "token_acc": 0.903954802259887, "train_speed(iter/s)": 1.46271 }, { "epoch": 1.0760035988175314, "grad_norm": 2.1082112789154053, "learning_rate": 8.900179934273366e-05, "loss": 0.3791569709777832, "memory(GiB)": 61.91, "step": 25115, "token_acc": 0.9211356466876972, "train_speed(iter/s)": 1.462718 }, { "epoch": 1.0762178141467804, "grad_norm": 8.615592956542969, "learning_rate": 8.899758793724978e-05, "loss": 0.4793417453765869, "memory(GiB)": 61.91, "step": 25120, "token_acc": 0.8981132075471698, "train_speed(iter/s)": 1.462775 }, { "epoch": 1.0764320294760292, "grad_norm": 10.395878791809082, "learning_rate": 8.8993375825288e-05, "loss": 0.2862694263458252, "memory(GiB)": 61.91, "step": 25125, "token_acc": 0.9323308270676691, "train_speed(iter/s)": 1.46283 }, { "epoch": 1.0766462448052783, "grad_norm": 3.636275053024292, "learning_rate": 8.898916300692463e-05, "loss": 0.24978837966918946, "memory(GiB)": 61.91, "step": 25130, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.462816 }, { "epoch": 1.0768604601345273, "grad_norm": 2.033215284347534, "learning_rate": 8.898494948223604e-05, "loss": 0.47611422538757325, "memory(GiB)": 61.91, "step": 25135, "token_acc": 0.9006622516556292, "train_speed(iter/s)": 1.462817 }, { "epoch": 1.077074675463776, "grad_norm": 4.658651351928711, "learning_rate": 8.898073525129848e-05, "loss": 0.4858135223388672, "memory(GiB)": 61.91, "step": 25140, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.462814 }, { "epoch": 1.0772888907930251, "grad_norm": 7.769630432128906, "learning_rate": 8.897652031418834e-05, "loss": 0.6140228271484375, "memory(GiB)": 61.91, "step": 25145, "token_acc": 0.862453531598513, "train_speed(iter/s)": 1.462879 }, { "epoch": 1.0775031061222742, "grad_norm": 2.520064115524292, "learning_rate": 8.8972304670982e-05, "loss": 0.14724371433258057, "memory(GiB)": 61.91, "step": 25150, "token_acc": 0.9710144927536232, "train_speed(iter/s)": 1.462883 }, { "epoch": 1.077717321451523, "grad_norm": 2.481902837753296, "learning_rate": 8.896808832175576e-05, "loss": 0.3497974634170532, "memory(GiB)": 61.91, "step": 25155, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.462883 }, { "epoch": 1.077931536780772, "grad_norm": 3.923410654067993, "learning_rate": 8.896387126658605e-05, "loss": 0.4088460445404053, "memory(GiB)": 61.91, "step": 25160, "token_acc": 0.9044117647058824, "train_speed(iter/s)": 1.462879 }, { "epoch": 1.078145752110021, "grad_norm": 2.801046133041382, "learning_rate": 8.895965350554929e-05, "loss": 0.26686644554138184, "memory(GiB)": 61.91, "step": 25165, "token_acc": 0.93, "train_speed(iter/s)": 1.462866 }, { "epoch": 1.0783599674392699, "grad_norm": 0.24932678043842316, "learning_rate": 8.895543503872183e-05, "loss": 0.16468576192855836, "memory(GiB)": 61.91, "step": 25170, "token_acc": 0.9575289575289575, "train_speed(iter/s)": 1.462872 }, { "epoch": 1.078574182768519, "grad_norm": 0.6644452810287476, "learning_rate": 8.895121586618014e-05, "loss": 0.6380305290222168, "memory(GiB)": 61.91, "step": 25175, "token_acc": 0.8755020080321285, "train_speed(iter/s)": 1.462851 }, { "epoch": 1.078788398097768, "grad_norm": 3.240630626678467, "learning_rate": 8.894699598800064e-05, "loss": 0.3646794080734253, "memory(GiB)": 61.91, "step": 25180, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.462843 }, { "epoch": 1.0790026134270168, "grad_norm": 3.887277841567993, "learning_rate": 8.894277540425977e-05, "loss": 0.526161003112793, "memory(GiB)": 61.91, "step": 25185, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.462826 }, { "epoch": 1.0792168287562658, "grad_norm": 3.63022780418396, "learning_rate": 8.893855411503398e-05, "loss": 0.5759621620178222, "memory(GiB)": 61.91, "step": 25190, "token_acc": 0.8912280701754386, "train_speed(iter/s)": 1.462835 }, { "epoch": 1.0794310440855148, "grad_norm": 5.664218902587891, "learning_rate": 8.893433212039974e-05, "loss": 0.6657480239868164, "memory(GiB)": 61.91, "step": 25195, "token_acc": 0.8771331058020477, "train_speed(iter/s)": 1.462841 }, { "epoch": 1.0796452594147636, "grad_norm": 3.331249713897705, "learning_rate": 8.893010942043359e-05, "loss": 0.3688171863555908, "memory(GiB)": 61.91, "step": 25200, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.462851 }, { "epoch": 1.0798594747440127, "grad_norm": 1.8218655586242676, "learning_rate": 8.892588601521197e-05, "loss": 0.2277174711227417, "memory(GiB)": 61.91, "step": 25205, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.46284 }, { "epoch": 1.0800736900732617, "grad_norm": 1.4769761562347412, "learning_rate": 8.89216619048114e-05, "loss": 0.3569038391113281, "memory(GiB)": 61.91, "step": 25210, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.462835 }, { "epoch": 1.0802879054025105, "grad_norm": 0.698101818561554, "learning_rate": 8.891743708930842e-05, "loss": 0.14968445301055908, "memory(GiB)": 61.91, "step": 25215, "token_acc": 0.9620253164556962, "train_speed(iter/s)": 1.462841 }, { "epoch": 1.0805021207317596, "grad_norm": 1.0063821077346802, "learning_rate": 8.891321156877957e-05, "loss": 0.20954930782318115, "memory(GiB)": 61.91, "step": 25220, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.462843 }, { "epoch": 1.0807163360610086, "grad_norm": 2.674788236618042, "learning_rate": 8.890898534330136e-05, "loss": 0.44927129745483396, "memory(GiB)": 61.91, "step": 25225, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.462838 }, { "epoch": 1.0809305513902574, "grad_norm": 3.1945652961730957, "learning_rate": 8.89047584129504e-05, "loss": 0.3681762218475342, "memory(GiB)": 61.91, "step": 25230, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.462834 }, { "epoch": 1.0811447667195064, "grad_norm": 2.5937066078186035, "learning_rate": 8.890053077780325e-05, "loss": 0.28696560859680176, "memory(GiB)": 61.91, "step": 25235, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.462837 }, { "epoch": 1.0813589820487555, "grad_norm": 1.969464659690857, "learning_rate": 8.88963024379365e-05, "loss": 0.5010621547698975, "memory(GiB)": 61.91, "step": 25240, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.462837 }, { "epoch": 1.0815731973780043, "grad_norm": 0.615304172039032, "learning_rate": 8.889207339342673e-05, "loss": 0.40453338623046875, "memory(GiB)": 61.91, "step": 25245, "token_acc": 0.9081967213114754, "train_speed(iter/s)": 1.462855 }, { "epoch": 1.0817874127072533, "grad_norm": 5.109377861022949, "learning_rate": 8.888784364435056e-05, "loss": 0.26205923557281496, "memory(GiB)": 61.91, "step": 25250, "token_acc": 0.9449838187702265, "train_speed(iter/s)": 1.462854 }, { "epoch": 1.0820016280365023, "grad_norm": 4.434478759765625, "learning_rate": 8.888361319078464e-05, "loss": 0.5640530109405517, "memory(GiB)": 61.91, "step": 25255, "token_acc": 0.8637992831541219, "train_speed(iter/s)": 1.462855 }, { "epoch": 1.0822158433657512, "grad_norm": 6.821907997131348, "learning_rate": 8.887938203280559e-05, "loss": 0.6305492401123047, "memory(GiB)": 61.91, "step": 25260, "token_acc": 0.874251497005988, "train_speed(iter/s)": 1.462843 }, { "epoch": 1.0824300586950002, "grad_norm": 2.9749338626861572, "learning_rate": 8.887515017049006e-05, "loss": 0.3896113157272339, "memory(GiB)": 61.91, "step": 25265, "token_acc": 0.9058441558441559, "train_speed(iter/s)": 1.462842 }, { "epoch": 1.0826442740242492, "grad_norm": 3.366748094558716, "learning_rate": 8.887091760391471e-05, "loss": 0.33929243087768557, "memory(GiB)": 61.91, "step": 25270, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.462838 }, { "epoch": 1.082858489353498, "grad_norm": 2.0692217350006104, "learning_rate": 8.886668433315622e-05, "loss": 0.397383451461792, "memory(GiB)": 61.91, "step": 25275, "token_acc": 0.8975155279503105, "train_speed(iter/s)": 1.462836 }, { "epoch": 1.083072704682747, "grad_norm": 30.156509399414062, "learning_rate": 8.88624503582913e-05, "loss": 0.4812010765075684, "memory(GiB)": 61.91, "step": 25280, "token_acc": 0.9087301587301587, "train_speed(iter/s)": 1.462868 }, { "epoch": 1.0832869200119961, "grad_norm": 0.6205039024353027, "learning_rate": 8.885821567939663e-05, "loss": 0.33576979637146, "memory(GiB)": 61.91, "step": 25285, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.46286 }, { "epoch": 1.083501135341245, "grad_norm": 2.5365982055664062, "learning_rate": 8.885398029654892e-05, "loss": 0.2362661838531494, "memory(GiB)": 61.91, "step": 25290, "token_acc": 0.9488054607508533, "train_speed(iter/s)": 1.46287 }, { "epoch": 1.083715350670494, "grad_norm": 1.851911187171936, "learning_rate": 8.884974420982493e-05, "loss": 0.22056131362915038, "memory(GiB)": 61.91, "step": 25295, "token_acc": 0.9607843137254902, "train_speed(iter/s)": 1.462859 }, { "epoch": 1.083929565999743, "grad_norm": 7.808811664581299, "learning_rate": 8.884550741930135e-05, "loss": 0.5282489776611328, "memory(GiB)": 61.91, "step": 25300, "token_acc": 0.8875379939209727, "train_speed(iter/s)": 1.462877 }, { "epoch": 1.0841437813289918, "grad_norm": 8.61439323425293, "learning_rate": 8.884126992505498e-05, "loss": 0.42961935997009276, "memory(GiB)": 61.91, "step": 25305, "token_acc": 0.890282131661442, "train_speed(iter/s)": 1.462902 }, { "epoch": 1.0843579966582408, "grad_norm": 3.740543842315674, "learning_rate": 8.883703172716258e-05, "loss": 0.35737314224243166, "memory(GiB)": 61.91, "step": 25310, "token_acc": 0.9357541899441341, "train_speed(iter/s)": 1.462915 }, { "epoch": 1.0845722119874899, "grad_norm": 0.029797568917274475, "learning_rate": 8.883279282570089e-05, "loss": 0.19379490613937378, "memory(GiB)": 61.91, "step": 25315, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.462925 }, { "epoch": 1.0847864273167387, "grad_norm": 3.8831422328948975, "learning_rate": 8.882855322074674e-05, "loss": 0.35411949157714845, "memory(GiB)": 61.91, "step": 25320, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.462926 }, { "epoch": 1.0850006426459877, "grad_norm": 6.7855544090271, "learning_rate": 8.882431291237693e-05, "loss": 0.5354075908660889, "memory(GiB)": 61.91, "step": 25325, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.462951 }, { "epoch": 1.0852148579752368, "grad_norm": 2.5457420349121094, "learning_rate": 8.882007190066827e-05, "loss": 0.5731357574462891, "memory(GiB)": 61.91, "step": 25330, "token_acc": 0.8856088560885609, "train_speed(iter/s)": 1.462956 }, { "epoch": 1.0854290733044856, "grad_norm": 7.45316743850708, "learning_rate": 8.881583018569761e-05, "loss": 0.2582524538040161, "memory(GiB)": 61.91, "step": 25335, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.462937 }, { "epoch": 1.0856432886337346, "grad_norm": 5.669838905334473, "learning_rate": 8.881158776754175e-05, "loss": 0.40042386054992674, "memory(GiB)": 61.91, "step": 25340, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.462927 }, { "epoch": 1.0858575039629836, "grad_norm": 3.774674415588379, "learning_rate": 8.880734464627757e-05, "loss": 0.3499144077301025, "memory(GiB)": 61.91, "step": 25345, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.462958 }, { "epoch": 1.0860717192922325, "grad_norm": 2.943087100982666, "learning_rate": 8.880310082198196e-05, "loss": 0.3465965747833252, "memory(GiB)": 61.91, "step": 25350, "token_acc": 0.9290123456790124, "train_speed(iter/s)": 1.462958 }, { "epoch": 1.0862859346214815, "grad_norm": 2.5265161991119385, "learning_rate": 8.879885629473176e-05, "loss": 0.41710567474365234, "memory(GiB)": 61.91, "step": 25355, "token_acc": 0.8929889298892989, "train_speed(iter/s)": 1.462959 }, { "epoch": 1.0865001499507305, "grad_norm": 4.733047962188721, "learning_rate": 8.87946110646039e-05, "loss": 0.44867305755615233, "memory(GiB)": 61.91, "step": 25360, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.462988 }, { "epoch": 1.0867143652799793, "grad_norm": 4.866017818450928, "learning_rate": 8.879036513167523e-05, "loss": 0.26465625762939454, "memory(GiB)": 61.91, "step": 25365, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.462988 }, { "epoch": 1.0869285806092284, "grad_norm": 3.216681480407715, "learning_rate": 8.878611849602274e-05, "loss": 0.41363911628723143, "memory(GiB)": 61.91, "step": 25370, "token_acc": 0.912621359223301, "train_speed(iter/s)": 1.462982 }, { "epoch": 1.0871427959384774, "grad_norm": 4.08538293838501, "learning_rate": 8.878187115772331e-05, "loss": 0.49055957794189453, "memory(GiB)": 61.91, "step": 25375, "token_acc": 0.8957055214723927, "train_speed(iter/s)": 1.46301 }, { "epoch": 1.0873570112677262, "grad_norm": 4.663366794586182, "learning_rate": 8.877762311685391e-05, "loss": 0.43918437957763673, "memory(GiB)": 61.91, "step": 25380, "token_acc": 0.9054545454545454, "train_speed(iter/s)": 1.463018 }, { "epoch": 1.0875712265969752, "grad_norm": 1.3237236738204956, "learning_rate": 8.87733743734915e-05, "loss": 0.3800185680389404, "memory(GiB)": 61.91, "step": 25385, "token_acc": 0.9203296703296703, "train_speed(iter/s)": 1.463017 }, { "epoch": 1.0877854419262243, "grad_norm": 2.6623470783233643, "learning_rate": 8.876912492771303e-05, "loss": 0.3958677530288696, "memory(GiB)": 61.91, "step": 25390, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.463022 }, { "epoch": 1.087999657255473, "grad_norm": 1.49717378616333, "learning_rate": 8.876487477959551e-05, "loss": 0.45929994583129885, "memory(GiB)": 61.91, "step": 25395, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.463012 }, { "epoch": 1.0882138725847221, "grad_norm": 3.0187296867370605, "learning_rate": 8.876062392921591e-05, "loss": 0.5400813102722168, "memory(GiB)": 61.91, "step": 25400, "token_acc": 0.8625429553264605, "train_speed(iter/s)": 1.463016 }, { "epoch": 1.0884280879139712, "grad_norm": 0.5320674180984497, "learning_rate": 8.875637237665124e-05, "loss": 0.20714907646179198, "memory(GiB)": 61.91, "step": 25405, "token_acc": 0.9689922480620154, "train_speed(iter/s)": 1.463016 }, { "epoch": 1.08864230324322, "grad_norm": 5.187716007232666, "learning_rate": 8.875212012197853e-05, "loss": 0.747398853302002, "memory(GiB)": 61.91, "step": 25410, "token_acc": 0.8581818181818182, "train_speed(iter/s)": 1.463032 }, { "epoch": 1.088856518572469, "grad_norm": 1.979594349861145, "learning_rate": 8.874786716527482e-05, "loss": 0.5119746685028076, "memory(GiB)": 61.91, "step": 25415, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.463034 }, { "epoch": 1.089070733901718, "grad_norm": 3.005169630050659, "learning_rate": 8.874361350661713e-05, "loss": 0.28762526512145997, "memory(GiB)": 61.91, "step": 25420, "token_acc": 0.9305019305019305, "train_speed(iter/s)": 1.463013 }, { "epoch": 1.0892849492309669, "grad_norm": 2.158773183822632, "learning_rate": 8.873935914608256e-05, "loss": 0.37089555263519286, "memory(GiB)": 61.91, "step": 25425, "token_acc": 0.9162011173184358, "train_speed(iter/s)": 1.463023 }, { "epoch": 1.089499164560216, "grad_norm": 1.2758687734603882, "learning_rate": 8.873510408374814e-05, "loss": 0.366333532333374, "memory(GiB)": 61.91, "step": 25430, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.463027 }, { "epoch": 1.089713379889465, "grad_norm": 1.6552425622940063, "learning_rate": 8.873084831969098e-05, "loss": 0.3514272451400757, "memory(GiB)": 61.91, "step": 25435, "token_acc": 0.9221556886227545, "train_speed(iter/s)": 1.463022 }, { "epoch": 1.0899275952187137, "grad_norm": 4.64774751663208, "learning_rate": 8.872659185398817e-05, "loss": 0.6251749038696289, "memory(GiB)": 61.91, "step": 25440, "token_acc": 0.858085808580858, "train_speed(iter/s)": 1.463054 }, { "epoch": 1.0901418105479628, "grad_norm": 2.954789876937866, "learning_rate": 8.872233468671683e-05, "loss": 0.3360616445541382, "memory(GiB)": 61.91, "step": 25445, "token_acc": 0.924, "train_speed(iter/s)": 1.463056 }, { "epoch": 1.0903560258772118, "grad_norm": 5.879907608032227, "learning_rate": 8.871807681795406e-05, "loss": 0.5345702171325684, "memory(GiB)": 61.91, "step": 25450, "token_acc": 0.8848920863309353, "train_speed(iter/s)": 1.463087 }, { "epoch": 1.0905702412064606, "grad_norm": 2.535137891769409, "learning_rate": 8.8713818247777e-05, "loss": 0.42162532806396485, "memory(GiB)": 61.91, "step": 25455, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.463101 }, { "epoch": 1.0907844565357097, "grad_norm": 1.8338370323181152, "learning_rate": 8.870955897626283e-05, "loss": 0.39129176139831545, "memory(GiB)": 61.91, "step": 25460, "token_acc": 0.9129032258064517, "train_speed(iter/s)": 1.463122 }, { "epoch": 1.0909986718649587, "grad_norm": 2.5455667972564697, "learning_rate": 8.870529900348867e-05, "loss": 0.7869563579559327, "memory(GiB)": 61.91, "step": 25465, "token_acc": 0.8435374149659864, "train_speed(iter/s)": 1.463112 }, { "epoch": 1.0912128871942075, "grad_norm": 3.762604236602783, "learning_rate": 8.870103832953172e-05, "loss": 0.5257608413696289, "memory(GiB)": 61.91, "step": 25470, "token_acc": 0.9051094890510949, "train_speed(iter/s)": 1.463113 }, { "epoch": 1.0914271025234565, "grad_norm": 2.8788902759552, "learning_rate": 8.869677695446914e-05, "loss": 0.5383646965026856, "memory(GiB)": 61.91, "step": 25475, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.463128 }, { "epoch": 1.0916413178527056, "grad_norm": 2.302339792251587, "learning_rate": 8.869251487837816e-05, "loss": 0.458376407623291, "memory(GiB)": 61.91, "step": 25480, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.46316 }, { "epoch": 1.0918555331819544, "grad_norm": 4.270304203033447, "learning_rate": 8.868825210133598e-05, "loss": 0.663810920715332, "memory(GiB)": 61.91, "step": 25485, "token_acc": 0.8729641693811075, "train_speed(iter/s)": 1.46319 }, { "epoch": 1.0920697485112034, "grad_norm": 1.3363163471221924, "learning_rate": 8.86839886234198e-05, "loss": 0.5032050609588623, "memory(GiB)": 61.91, "step": 25490, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.463204 }, { "epoch": 1.0922839638404525, "grad_norm": 1.5518182516098022, "learning_rate": 8.86797244447069e-05, "loss": 0.4348761081695557, "memory(GiB)": 61.91, "step": 25495, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.463225 }, { "epoch": 1.0924981791697013, "grad_norm": 3.966857433319092, "learning_rate": 8.86754595652745e-05, "loss": 0.5428637981414794, "memory(GiB)": 61.91, "step": 25500, "token_acc": 0.8778409090909091, "train_speed(iter/s)": 1.463233 }, { "epoch": 1.0924981791697013, "eval_loss": 2.4473960399627686, "eval_runtime": 11.8427, "eval_samples_per_second": 8.444, "eval_steps_per_second": 8.444, "eval_token_acc": 0.46208869814020026, "step": 25500 }, { "epoch": 1.0927123944989503, "grad_norm": 3.0310986042022705, "learning_rate": 8.867119398519986e-05, "loss": 0.2647587299346924, "memory(GiB)": 61.91, "step": 25505, "token_acc": 0.5818181818181818, "train_speed(iter/s)": 1.462166 }, { "epoch": 1.0929266098281993, "grad_norm": 4.473127841949463, "learning_rate": 8.866692770456026e-05, "loss": 0.5191736698150635, "memory(GiB)": 61.91, "step": 25510, "token_acc": 0.9012345679012346, "train_speed(iter/s)": 1.462211 }, { "epoch": 1.0931408251574484, "grad_norm": 2.7050929069519043, "learning_rate": 8.866266072343301e-05, "loss": 0.4386690616607666, "memory(GiB)": 61.91, "step": 25515, "token_acc": 0.9180790960451978, "train_speed(iter/s)": 1.462225 }, { "epoch": 1.0933550404866972, "grad_norm": 3.027987241744995, "learning_rate": 8.865839304189538e-05, "loss": 0.35813226699829104, "memory(GiB)": 61.91, "step": 25520, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.462227 }, { "epoch": 1.0935692558159462, "grad_norm": 2.442445755004883, "learning_rate": 8.865412466002472e-05, "loss": 0.4285855293273926, "memory(GiB)": 61.91, "step": 25525, "token_acc": 0.9113149847094801, "train_speed(iter/s)": 1.462219 }, { "epoch": 1.0937834711451953, "grad_norm": 2.3580482006073, "learning_rate": 8.86498555778983e-05, "loss": 0.4725804805755615, "memory(GiB)": 61.91, "step": 25530, "token_acc": 0.9, "train_speed(iter/s)": 1.462215 }, { "epoch": 1.093997686474444, "grad_norm": 3.6616616249084473, "learning_rate": 8.86455857955935e-05, "loss": 0.1727069616317749, "memory(GiB)": 61.91, "step": 25535, "token_acc": 0.9478527607361963, "train_speed(iter/s)": 1.462223 }, { "epoch": 1.094211901803693, "grad_norm": 7.300609111785889, "learning_rate": 8.864131531318766e-05, "loss": 0.439729642868042, "memory(GiB)": 61.91, "step": 25540, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.46224 }, { "epoch": 1.0944261171329421, "grad_norm": 2.9304041862487793, "learning_rate": 8.863704413075816e-05, "loss": 0.6195690631866455, "memory(GiB)": 61.91, "step": 25545, "token_acc": 0.8961038961038961, "train_speed(iter/s)": 1.462276 }, { "epoch": 1.094640332462191, "grad_norm": 3.695221424102783, "learning_rate": 8.863277224838234e-05, "loss": 0.5634300231933593, "memory(GiB)": 61.91, "step": 25550, "token_acc": 0.8843283582089553, "train_speed(iter/s)": 1.462298 }, { "epoch": 1.09485454779144, "grad_norm": 3.3114211559295654, "learning_rate": 8.862849966613763e-05, "loss": 0.23432071208953859, "memory(GiB)": 61.91, "step": 25555, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.462312 }, { "epoch": 1.095068763120689, "grad_norm": 3.130274534225464, "learning_rate": 8.862422638410139e-05, "loss": 0.3334404706954956, "memory(GiB)": 61.91, "step": 25560, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.462297 }, { "epoch": 1.0952829784499378, "grad_norm": 3.812450885772705, "learning_rate": 8.861995240235106e-05, "loss": 0.31398777961730956, "memory(GiB)": 61.91, "step": 25565, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.462309 }, { "epoch": 1.0954971937791869, "grad_norm": 3.378427743911743, "learning_rate": 8.861567772096408e-05, "loss": 0.2816780090332031, "memory(GiB)": 61.91, "step": 25570, "token_acc": 0.95, "train_speed(iter/s)": 1.46232 }, { "epoch": 1.095711409108436, "grad_norm": 2.5424489974975586, "learning_rate": 8.861140234001785e-05, "loss": 0.38444080352783205, "memory(GiB)": 61.91, "step": 25575, "token_acc": 0.9029850746268657, "train_speed(iter/s)": 1.462337 }, { "epoch": 1.0959256244376847, "grad_norm": 4.114015102386475, "learning_rate": 8.860712625958987e-05, "loss": 0.6762954235076905, "memory(GiB)": 61.91, "step": 25580, "token_acc": 0.8594249201277955, "train_speed(iter/s)": 1.462359 }, { "epoch": 1.0961398397669337, "grad_norm": 5.402930736541748, "learning_rate": 8.860284947975758e-05, "loss": 0.4496199607849121, "memory(GiB)": 61.91, "step": 25585, "token_acc": 0.8776978417266187, "train_speed(iter/s)": 1.462359 }, { "epoch": 1.0963540550961828, "grad_norm": 3.0589101314544678, "learning_rate": 8.859857200059845e-05, "loss": 0.39393432140350343, "memory(GiB)": 61.91, "step": 25590, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.462373 }, { "epoch": 1.0965682704254316, "grad_norm": 1.7756133079528809, "learning_rate": 8.859429382218998e-05, "loss": 0.2643697738647461, "memory(GiB)": 61.91, "step": 25595, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.462414 }, { "epoch": 1.0967824857546806, "grad_norm": 1.9535704851150513, "learning_rate": 8.859001494460968e-05, "loss": 0.21287319660186768, "memory(GiB)": 61.91, "step": 25600, "token_acc": 0.940625, "train_speed(iter/s)": 1.462405 }, { "epoch": 1.0969967010839297, "grad_norm": 2.799534797668457, "learning_rate": 8.858573536793504e-05, "loss": 0.27471189498901366, "memory(GiB)": 61.91, "step": 25605, "token_acc": 0.9442724458204335, "train_speed(iter/s)": 1.462414 }, { "epoch": 1.0972109164131785, "grad_norm": 3.4930942058563232, "learning_rate": 8.858145509224363e-05, "loss": 0.6470661163330078, "memory(GiB)": 61.91, "step": 25610, "token_acc": 0.8538961038961039, "train_speed(iter/s)": 1.462409 }, { "epoch": 1.0974251317424275, "grad_norm": 4.832221984863281, "learning_rate": 8.857717411761296e-05, "loss": 0.43185877799987793, "memory(GiB)": 61.91, "step": 25615, "token_acc": 0.9105058365758755, "train_speed(iter/s)": 1.462418 }, { "epoch": 1.0976393470716765, "grad_norm": 2.0932860374450684, "learning_rate": 8.857289244412059e-05, "loss": 0.4046803951263428, "memory(GiB)": 61.91, "step": 25620, "token_acc": 0.897196261682243, "train_speed(iter/s)": 1.462439 }, { "epoch": 1.0978535624009254, "grad_norm": 3.1558215618133545, "learning_rate": 8.85686100718441e-05, "loss": 0.4110370635986328, "memory(GiB)": 61.91, "step": 25625, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.462454 }, { "epoch": 1.0980677777301744, "grad_norm": 6.352552890777588, "learning_rate": 8.856432700086104e-05, "loss": 0.3852397441864014, "memory(GiB)": 61.91, "step": 25630, "token_acc": 0.9279279279279279, "train_speed(iter/s)": 1.462453 }, { "epoch": 1.0982819930594234, "grad_norm": 2.6966769695281982, "learning_rate": 8.856004323124903e-05, "loss": 0.4180790424346924, "memory(GiB)": 61.91, "step": 25635, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.462463 }, { "epoch": 1.0984962083886722, "grad_norm": 7.430751323699951, "learning_rate": 8.855575876308566e-05, "loss": 0.5509214401245117, "memory(GiB)": 61.91, "step": 25640, "token_acc": 0.8515151515151516, "train_speed(iter/s)": 1.462435 }, { "epoch": 1.0987104237179213, "grad_norm": 2.179262399673462, "learning_rate": 8.855147359644856e-05, "loss": 0.21901309490203857, "memory(GiB)": 61.91, "step": 25645, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.462428 }, { "epoch": 1.0989246390471703, "grad_norm": 0.5742531418800354, "learning_rate": 8.854718773141535e-05, "loss": 0.421583366394043, "memory(GiB)": 61.91, "step": 25650, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.462419 }, { "epoch": 1.0991388543764191, "grad_norm": 3.2867512702941895, "learning_rate": 8.854290116806367e-05, "loss": 0.43399825096130373, "memory(GiB)": 61.91, "step": 25655, "token_acc": 0.9, "train_speed(iter/s)": 1.462418 }, { "epoch": 1.0993530697056682, "grad_norm": 3.334643602371216, "learning_rate": 8.853861390647118e-05, "loss": 0.5151684761047364, "memory(GiB)": 61.91, "step": 25660, "token_acc": 0.897887323943662, "train_speed(iter/s)": 1.462416 }, { "epoch": 1.0995672850349172, "grad_norm": 6.875061511993408, "learning_rate": 8.853432594671554e-05, "loss": 0.40060272216796877, "memory(GiB)": 61.91, "step": 25665, "token_acc": 0.9343065693430657, "train_speed(iter/s)": 1.462424 }, { "epoch": 1.099781500364166, "grad_norm": 4.736055850982666, "learning_rate": 8.853003728887446e-05, "loss": 0.39192359447479247, "memory(GiB)": 61.91, "step": 25670, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.462446 }, { "epoch": 1.099995715693415, "grad_norm": 1.5845576524734497, "learning_rate": 8.85257479330256e-05, "loss": 0.33023924827575685, "memory(GiB)": 61.91, "step": 25675, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.462504 }, { "epoch": 1.100209931022664, "grad_norm": 4.0633087158203125, "learning_rate": 8.852145787924666e-05, "loss": 0.4188316822052002, "memory(GiB)": 61.91, "step": 25680, "token_acc": 0.9198717948717948, "train_speed(iter/s)": 1.462513 }, { "epoch": 1.1004241463519129, "grad_norm": 6.944761753082275, "learning_rate": 8.851716712761538e-05, "loss": 0.560029935836792, "memory(GiB)": 61.91, "step": 25685, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.462555 }, { "epoch": 1.100638361681162, "grad_norm": 5.15505838394165, "learning_rate": 8.851287567820949e-05, "loss": 0.2748559951782227, "memory(GiB)": 61.91, "step": 25690, "token_acc": 0.9553264604810997, "train_speed(iter/s)": 1.462544 }, { "epoch": 1.100852577010411, "grad_norm": 5.872371196746826, "learning_rate": 8.850858353110674e-05, "loss": 0.5358783721923828, "memory(GiB)": 61.91, "step": 25695, "token_acc": 0.8825503355704698, "train_speed(iter/s)": 1.462517 }, { "epoch": 1.1010667923396598, "grad_norm": 5.624542236328125, "learning_rate": 8.850429068638487e-05, "loss": 0.42310667037963867, "memory(GiB)": 61.91, "step": 25700, "token_acc": 0.8944099378881988, "train_speed(iter/s)": 1.46252 }, { "epoch": 1.1012810076689088, "grad_norm": 3.241826295852661, "learning_rate": 8.849999714412165e-05, "loss": 0.1837466835975647, "memory(GiB)": 61.91, "step": 25705, "token_acc": 0.9620253164556962, "train_speed(iter/s)": 1.462512 }, { "epoch": 1.1014952229981578, "grad_norm": 3.443141222000122, "learning_rate": 8.849570290439486e-05, "loss": 0.31424198150634763, "memory(GiB)": 61.91, "step": 25710, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.462513 }, { "epoch": 1.1017094383274066, "grad_norm": 3.0369327068328857, "learning_rate": 8.84914079672823e-05, "loss": 0.3812888145446777, "memory(GiB)": 61.91, "step": 25715, "token_acc": 0.9192546583850931, "train_speed(iter/s)": 1.462506 }, { "epoch": 1.1019236536566557, "grad_norm": 1.579086184501648, "learning_rate": 8.84871123328618e-05, "loss": 0.6522722721099854, "memory(GiB)": 61.91, "step": 25720, "token_acc": 0.8566775244299675, "train_speed(iter/s)": 1.462488 }, { "epoch": 1.1021378689859047, "grad_norm": 9.20272159576416, "learning_rate": 8.848281600121114e-05, "loss": 0.34494576454162595, "memory(GiB)": 61.91, "step": 25725, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.462493 }, { "epoch": 1.1023520843151535, "grad_norm": 1.4920995235443115, "learning_rate": 8.847851897240815e-05, "loss": 0.31069986820220946, "memory(GiB)": 61.91, "step": 25730, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.462494 }, { "epoch": 1.1025662996444026, "grad_norm": 2.090115785598755, "learning_rate": 8.847422124653072e-05, "loss": 0.3923627853393555, "memory(GiB)": 61.91, "step": 25735, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.462505 }, { "epoch": 1.1027805149736516, "grad_norm": 2.7149574756622314, "learning_rate": 8.846992282365667e-05, "loss": 0.30125668048858645, "memory(GiB)": 61.91, "step": 25740, "token_acc": 0.9358974358974359, "train_speed(iter/s)": 1.462542 }, { "epoch": 1.1029947303029004, "grad_norm": 0.14514748752117157, "learning_rate": 8.846562370386389e-05, "loss": 0.26002140045166017, "memory(GiB)": 61.91, "step": 25745, "token_acc": 0.9401993355481728, "train_speed(iter/s)": 1.462553 }, { "epoch": 1.1032089456321494, "grad_norm": 3.914837598800659, "learning_rate": 8.846132388723023e-05, "loss": 0.3982586622238159, "memory(GiB)": 61.91, "step": 25750, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.462558 }, { "epoch": 1.1034231609613985, "grad_norm": 5.314088344573975, "learning_rate": 8.845702337383363e-05, "loss": 0.3666435480117798, "memory(GiB)": 61.91, "step": 25755, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.462559 }, { "epoch": 1.1036373762906473, "grad_norm": 5.537886142730713, "learning_rate": 8.845272216375196e-05, "loss": 0.37426419258117677, "memory(GiB)": 61.91, "step": 25760, "token_acc": 0.91015625, "train_speed(iter/s)": 1.462576 }, { "epoch": 1.1038515916198963, "grad_norm": 6.6959333419799805, "learning_rate": 8.844842025706316e-05, "loss": 0.6918998718261719, "memory(GiB)": 61.91, "step": 25765, "token_acc": 0.8586572438162544, "train_speed(iter/s)": 1.462597 }, { "epoch": 1.1040658069491454, "grad_norm": 0.5444590449333191, "learning_rate": 8.844411765384517e-05, "loss": 0.40291452407836914, "memory(GiB)": 61.91, "step": 25770, "token_acc": 0.8945783132530121, "train_speed(iter/s)": 1.462585 }, { "epoch": 1.1042800222783942, "grad_norm": 3.045780658721924, "learning_rate": 8.843981435417592e-05, "loss": 0.4199867248535156, "memory(GiB)": 61.91, "step": 25775, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.462566 }, { "epoch": 1.1044942376076432, "grad_norm": 3.6586849689483643, "learning_rate": 8.843551035813337e-05, "loss": 0.39733223915100097, "memory(GiB)": 61.91, "step": 25780, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.462562 }, { "epoch": 1.1047084529368922, "grad_norm": 0.856523871421814, "learning_rate": 8.84312056657955e-05, "loss": 0.46988687515258787, "memory(GiB)": 61.91, "step": 25785, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.462566 }, { "epoch": 1.104922668266141, "grad_norm": 2.635715961456299, "learning_rate": 8.842690027724029e-05, "loss": 0.25634231567382815, "memory(GiB)": 61.91, "step": 25790, "token_acc": 0.9375, "train_speed(iter/s)": 1.462569 }, { "epoch": 1.10513688359539, "grad_norm": 2.0435597896575928, "learning_rate": 8.842259419254573e-05, "loss": 0.20999600887298583, "memory(GiB)": 61.91, "step": 25795, "token_acc": 0.9369747899159664, "train_speed(iter/s)": 1.462553 }, { "epoch": 1.1053510989246391, "grad_norm": 0.8108588457107544, "learning_rate": 8.841828741178984e-05, "loss": 0.1909146785736084, "memory(GiB)": 61.91, "step": 25800, "token_acc": 0.9773755656108597, "train_speed(iter/s)": 1.462547 }, { "epoch": 1.105565314253888, "grad_norm": 5.369544506072998, "learning_rate": 8.841397993505062e-05, "loss": 0.4492218017578125, "memory(GiB)": 61.91, "step": 25805, "token_acc": 0.9155844155844156, "train_speed(iter/s)": 1.462554 }, { "epoch": 1.105779529583137, "grad_norm": 3.5922462940216064, "learning_rate": 8.840967176240612e-05, "loss": 0.48829221725463867, "memory(GiB)": 61.91, "step": 25810, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.46256 }, { "epoch": 1.105993744912386, "grad_norm": 4.329902648925781, "learning_rate": 8.840536289393439e-05, "loss": 0.7044106960296631, "memory(GiB)": 61.91, "step": 25815, "token_acc": 0.8355263157894737, "train_speed(iter/s)": 1.462552 }, { "epoch": 1.1062079602416348, "grad_norm": 5.652036190032959, "learning_rate": 8.840105332971348e-05, "loss": 0.5730079650878906, "memory(GiB)": 61.91, "step": 25820, "token_acc": 0.890625, "train_speed(iter/s)": 1.46253 }, { "epoch": 1.1064221755708838, "grad_norm": 4.295072078704834, "learning_rate": 8.839674306982148e-05, "loss": 0.398391580581665, "memory(GiB)": 61.91, "step": 25825, "token_acc": 0.9267399267399268, "train_speed(iter/s)": 1.462544 }, { "epoch": 1.1066363909001329, "grad_norm": 3.323746919631958, "learning_rate": 8.839243211433645e-05, "loss": 0.23300089836120605, "memory(GiB)": 61.91, "step": 25830, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.462541 }, { "epoch": 1.1068506062293817, "grad_norm": 2.5753722190856934, "learning_rate": 8.838812046333648e-05, "loss": 0.4785348892211914, "memory(GiB)": 61.91, "step": 25835, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.462584 }, { "epoch": 1.1070648215586307, "grad_norm": 3.165036201477051, "learning_rate": 8.838380811689973e-05, "loss": 0.38180112838745117, "memory(GiB)": 61.91, "step": 25840, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.46259 }, { "epoch": 1.1072790368878798, "grad_norm": 5.510438442230225, "learning_rate": 8.837949507510427e-05, "loss": 0.2891889810562134, "memory(GiB)": 61.91, "step": 25845, "token_acc": 0.9370860927152318, "train_speed(iter/s)": 1.462589 }, { "epoch": 1.1074932522171286, "grad_norm": 2.0826663970947266, "learning_rate": 8.837518133802826e-05, "loss": 0.1692386507987976, "memory(GiB)": 61.91, "step": 25850, "token_acc": 0.9686520376175548, "train_speed(iter/s)": 1.462604 }, { "epoch": 1.1077074675463776, "grad_norm": 3.4737603664398193, "learning_rate": 8.837086690574983e-05, "loss": 0.3102865219116211, "memory(GiB)": 61.91, "step": 25855, "token_acc": 0.9246861924686193, "train_speed(iter/s)": 1.462604 }, { "epoch": 1.1079216828756266, "grad_norm": 3.8852028846740723, "learning_rate": 8.836655177834716e-05, "loss": 0.5614929676055909, "memory(GiB)": 61.91, "step": 25860, "token_acc": 0.9026217228464419, "train_speed(iter/s)": 1.462585 }, { "epoch": 1.1081358982048755, "grad_norm": 1.5113072395324707, "learning_rate": 8.836223595589842e-05, "loss": 0.3846569061279297, "memory(GiB)": 61.91, "step": 25865, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.462576 }, { "epoch": 1.1083501135341245, "grad_norm": 3.7111425399780273, "learning_rate": 8.835791943848178e-05, "loss": 0.32469854354858396, "memory(GiB)": 61.91, "step": 25870, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.462591 }, { "epoch": 1.1085643288633735, "grad_norm": 5.516271591186523, "learning_rate": 8.835360222617545e-05, "loss": 0.5753870964050293, "memory(GiB)": 61.91, "step": 25875, "token_acc": 0.8772563176895307, "train_speed(iter/s)": 1.462645 }, { "epoch": 1.1087785441926223, "grad_norm": 14.346359252929688, "learning_rate": 8.834928431905764e-05, "loss": 0.3491765022277832, "memory(GiB)": 61.91, "step": 25880, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.462679 }, { "epoch": 1.1089927595218714, "grad_norm": 7.797201633453369, "learning_rate": 8.834496571720657e-05, "loss": 0.15641179084777831, "memory(GiB)": 61.91, "step": 25885, "token_acc": 0.9724770642201835, "train_speed(iter/s)": 1.462689 }, { "epoch": 1.1092069748511204, "grad_norm": 2.635921001434326, "learning_rate": 8.834064642070045e-05, "loss": 0.2182243585586548, "memory(GiB)": 61.91, "step": 25890, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.462697 }, { "epoch": 1.1094211901803692, "grad_norm": 4.201181411743164, "learning_rate": 8.833632642961759e-05, "loss": 0.45209469795227053, "memory(GiB)": 61.91, "step": 25895, "token_acc": 0.9027237354085603, "train_speed(iter/s)": 1.462695 }, { "epoch": 1.1096354055096183, "grad_norm": 3.0628347396850586, "learning_rate": 8.83320057440362e-05, "loss": 0.35988383293151854, "memory(GiB)": 61.91, "step": 25900, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.46272 }, { "epoch": 1.1098496208388673, "grad_norm": 0.08065354079008102, "learning_rate": 8.832768436403455e-05, "loss": 0.3029822587966919, "memory(GiB)": 61.91, "step": 25905, "token_acc": 0.9406392694063926, "train_speed(iter/s)": 1.462709 }, { "epoch": 1.110063836168116, "grad_norm": 4.140479564666748, "learning_rate": 8.832336228969098e-05, "loss": 0.4738334655761719, "memory(GiB)": 61.91, "step": 25910, "token_acc": 0.8991097922848664, "train_speed(iter/s)": 1.462683 }, { "epoch": 1.1102780514973651, "grad_norm": 1.6978429555892944, "learning_rate": 8.831903952108372e-05, "loss": 0.31353259086608887, "memory(GiB)": 61.91, "step": 25915, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.462681 }, { "epoch": 1.1104922668266142, "grad_norm": 3.5904202461242676, "learning_rate": 8.831471605829113e-05, "loss": 0.45062990188598634, "memory(GiB)": 61.91, "step": 25920, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.462663 }, { "epoch": 1.110706482155863, "grad_norm": 4.93032169342041, "learning_rate": 8.831039190139152e-05, "loss": 0.4715071201324463, "memory(GiB)": 61.91, "step": 25925, "token_acc": 0.9, "train_speed(iter/s)": 1.462656 }, { "epoch": 1.110920697485112, "grad_norm": 2.0727381706237793, "learning_rate": 8.830606705046321e-05, "loss": 0.6093162536621094, "memory(GiB)": 61.91, "step": 25930, "token_acc": 0.8754208754208754, "train_speed(iter/s)": 1.462645 }, { "epoch": 1.111134912814361, "grad_norm": 3.761521816253662, "learning_rate": 8.830174150558456e-05, "loss": 0.43672747611999513, "memory(GiB)": 61.91, "step": 25935, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.462636 }, { "epoch": 1.1113491281436099, "grad_norm": 3.003472328186035, "learning_rate": 8.829741526683394e-05, "loss": 0.33553566932678225, "memory(GiB)": 61.91, "step": 25940, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.462646 }, { "epoch": 1.111563343472859, "grad_norm": 2.529301643371582, "learning_rate": 8.829308833428972e-05, "loss": 0.1377146363258362, "memory(GiB)": 61.91, "step": 25945, "token_acc": 0.9562043795620438, "train_speed(iter/s)": 1.46265 }, { "epoch": 1.111777558802108, "grad_norm": 4.39158296585083, "learning_rate": 8.828876070803028e-05, "loss": 0.30743117332458497, "memory(GiB)": 61.91, "step": 25950, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.462657 }, { "epoch": 1.1119917741313567, "grad_norm": 2.907616376876831, "learning_rate": 8.828443238813402e-05, "loss": 0.4554884433746338, "memory(GiB)": 61.91, "step": 25955, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.46266 }, { "epoch": 1.1122059894606058, "grad_norm": 1.3503397703170776, "learning_rate": 8.828010337467936e-05, "loss": 0.2182377576828003, "memory(GiB)": 61.91, "step": 25960, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.462661 }, { "epoch": 1.1124202047898548, "grad_norm": 2.1970860958099365, "learning_rate": 8.82757736677447e-05, "loss": 0.3443499803543091, "memory(GiB)": 61.91, "step": 25965, "token_acc": 0.9116719242902208, "train_speed(iter/s)": 1.462636 }, { "epoch": 1.1126344201191036, "grad_norm": 0.16376492381095886, "learning_rate": 8.82714432674085e-05, "loss": 0.3272589921951294, "memory(GiB)": 61.91, "step": 25970, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.462629 }, { "epoch": 1.1128486354483527, "grad_norm": 0.7045992612838745, "learning_rate": 8.826711217374921e-05, "loss": 0.21618645191192626, "memory(GiB)": 61.91, "step": 25975, "token_acc": 0.9566787003610109, "train_speed(iter/s)": 1.462639 }, { "epoch": 1.1130628507776017, "grad_norm": 3.08687162399292, "learning_rate": 8.826278038684529e-05, "loss": 0.30884385108947754, "memory(GiB)": 61.91, "step": 25980, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.462623 }, { "epoch": 1.1132770661068505, "grad_norm": 4.081678867340088, "learning_rate": 8.82584479067752e-05, "loss": 0.4381399631500244, "memory(GiB)": 61.91, "step": 25985, "token_acc": 0.9064516129032258, "train_speed(iter/s)": 1.462629 }, { "epoch": 1.1134912814360995, "grad_norm": 4.202171325683594, "learning_rate": 8.825411473361745e-05, "loss": 0.46744394302368164, "memory(GiB)": 61.91, "step": 25990, "token_acc": 0.8902439024390244, "train_speed(iter/s)": 1.46263 }, { "epoch": 1.1137054967653486, "grad_norm": 5.836882591247559, "learning_rate": 8.824978086745051e-05, "loss": 0.5948639392852784, "memory(GiB)": 61.91, "step": 25995, "token_acc": 0.8783269961977186, "train_speed(iter/s)": 1.462635 }, { "epoch": 1.1139197120945974, "grad_norm": 0.019692501053214073, "learning_rate": 8.824544630835293e-05, "loss": 0.35597076416015627, "memory(GiB)": 61.91, "step": 26000, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.462642 }, { "epoch": 1.1139197120945974, "eval_loss": 2.8603272438049316, "eval_runtime": 12.8826, "eval_samples_per_second": 7.762, "eval_steps_per_second": 7.762, "eval_token_acc": 0.402088772845953, "step": 26000 }, { "epoch": 1.1141339274238464, "grad_norm": 3.9984090328216553, "learning_rate": 8.82411110564032e-05, "loss": 0.477640438079834, "memory(GiB)": 61.91, "step": 26005, "token_acc": 0.5439924314096499, "train_speed(iter/s)": 1.461508 }, { "epoch": 1.1143481427530955, "grad_norm": 2.167558193206787, "learning_rate": 8.823677511167986e-05, "loss": 0.4058560371398926, "memory(GiB)": 61.91, "step": 26010, "token_acc": 0.9300911854103343, "train_speed(iter/s)": 1.461524 }, { "epoch": 1.1145623580823443, "grad_norm": 4.482082366943359, "learning_rate": 8.823243847426148e-05, "loss": 0.20967702865600585, "memory(GiB)": 61.91, "step": 26015, "token_acc": 0.9602649006622517, "train_speed(iter/s)": 1.461537 }, { "epoch": 1.1147765734115933, "grad_norm": 1.9614057540893555, "learning_rate": 8.822810114422662e-05, "loss": 0.3087871551513672, "memory(GiB)": 61.91, "step": 26020, "token_acc": 0.934984520123839, "train_speed(iter/s)": 1.461528 }, { "epoch": 1.1149907887408423, "grad_norm": 4.167447090148926, "learning_rate": 8.822376312165384e-05, "loss": 0.26484408378601076, "memory(GiB)": 61.91, "step": 26025, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.461532 }, { "epoch": 1.1152050040700912, "grad_norm": 1.0285189151763916, "learning_rate": 8.821942440662172e-05, "loss": 0.48116049766540525, "memory(GiB)": 61.91, "step": 26030, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.461547 }, { "epoch": 1.1154192193993402, "grad_norm": 4.51882266998291, "learning_rate": 8.821508499920889e-05, "loss": 0.5329698085784912, "memory(GiB)": 61.91, "step": 26035, "token_acc": 0.875, "train_speed(iter/s)": 1.461554 }, { "epoch": 1.1156334347285892, "grad_norm": 3.5376172065734863, "learning_rate": 8.821074489949395e-05, "loss": 0.22811331748962402, "memory(GiB)": 61.91, "step": 26040, "token_acc": 0.946360153256705, "train_speed(iter/s)": 1.461563 }, { "epoch": 1.115847650057838, "grad_norm": 2.4727799892425537, "learning_rate": 8.820640410755551e-05, "loss": 0.41037464141845703, "memory(GiB)": 61.91, "step": 26045, "token_acc": 0.9250814332247557, "train_speed(iter/s)": 1.46158 }, { "epoch": 1.116061865387087, "grad_norm": 4.7784600257873535, "learning_rate": 8.820206262347221e-05, "loss": 0.7095525741577149, "memory(GiB)": 61.91, "step": 26050, "token_acc": 0.868421052631579, "train_speed(iter/s)": 1.461582 }, { "epoch": 1.116276080716336, "grad_norm": 1.9152454137802124, "learning_rate": 8.819772044732274e-05, "loss": 0.4355663776397705, "memory(GiB)": 61.91, "step": 26055, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.461607 }, { "epoch": 1.1164902960455851, "grad_norm": 1.8125394582748413, "learning_rate": 8.81933775791857e-05, "loss": 0.391705846786499, "memory(GiB)": 61.91, "step": 26060, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.461618 }, { "epoch": 1.116704511374834, "grad_norm": 2.8288371562957764, "learning_rate": 8.818903401913983e-05, "loss": 0.46057939529418945, "memory(GiB)": 61.91, "step": 26065, "token_acc": 0.9006410256410257, "train_speed(iter/s)": 1.461622 }, { "epoch": 1.116918726704083, "grad_norm": 7.917285442352295, "learning_rate": 8.818468976726377e-05, "loss": 0.3275259256362915, "memory(GiB)": 61.91, "step": 26070, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.46163 }, { "epoch": 1.117132942033332, "grad_norm": 4.810873985290527, "learning_rate": 8.818034482363622e-05, "loss": 0.2278437852859497, "memory(GiB)": 61.91, "step": 26075, "token_acc": 0.950381679389313, "train_speed(iter/s)": 1.461625 }, { "epoch": 1.1173471573625808, "grad_norm": 0.0541328489780426, "learning_rate": 8.817599918833593e-05, "loss": 0.528840160369873, "memory(GiB)": 61.91, "step": 26080, "token_acc": 0.8796992481203008, "train_speed(iter/s)": 1.461629 }, { "epoch": 1.1175613726918299, "grad_norm": 2.85356068611145, "learning_rate": 8.817165286144158e-05, "loss": 0.3979561567306519, "memory(GiB)": 61.91, "step": 26085, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.461635 }, { "epoch": 1.117775588021079, "grad_norm": 4.269800186157227, "learning_rate": 8.816730584303194e-05, "loss": 0.33117525577545165, "memory(GiB)": 61.91, "step": 26090, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.461628 }, { "epoch": 1.1179898033503277, "grad_norm": 1.470747709274292, "learning_rate": 8.816295813318576e-05, "loss": 0.23719658851623535, "memory(GiB)": 61.91, "step": 26095, "token_acc": 0.9349593495934959, "train_speed(iter/s)": 1.461633 }, { "epoch": 1.1182040186795768, "grad_norm": 0.3618052303791046, "learning_rate": 8.815860973198177e-05, "loss": 0.21009986400604247, "memory(GiB)": 61.91, "step": 26100, "token_acc": 0.956140350877193, "train_speed(iter/s)": 1.461649 }, { "epoch": 1.1184182340088258, "grad_norm": 3.4741058349609375, "learning_rate": 8.81542606394988e-05, "loss": 0.2764773368835449, "memory(GiB)": 61.91, "step": 26105, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.461648 }, { "epoch": 1.1186324493380746, "grad_norm": 0.8171009421348572, "learning_rate": 8.814991085581559e-05, "loss": 0.23196308612823485, "memory(GiB)": 61.91, "step": 26110, "token_acc": 0.9606299212598425, "train_speed(iter/s)": 1.461666 }, { "epoch": 1.1188466646673236, "grad_norm": 7.960484027862549, "learning_rate": 8.814556038101097e-05, "loss": 0.35217013359069826, "memory(GiB)": 61.91, "step": 26115, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.461677 }, { "epoch": 1.1190608799965727, "grad_norm": 6.349799633026123, "learning_rate": 8.814120921516372e-05, "loss": 0.4199717044830322, "memory(GiB)": 61.91, "step": 26120, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.461701 }, { "epoch": 1.1192750953258215, "grad_norm": 0.9499421715736389, "learning_rate": 8.81368573583527e-05, "loss": 0.09501857757568359, "memory(GiB)": 61.91, "step": 26125, "token_acc": 0.9735973597359736, "train_speed(iter/s)": 1.461697 }, { "epoch": 1.1194893106550705, "grad_norm": 4.508030414581299, "learning_rate": 8.813250481065673e-05, "loss": 0.3168467044830322, "memory(GiB)": 61.91, "step": 26130, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.46171 }, { "epoch": 1.1197035259843195, "grad_norm": 0.3455658555030823, "learning_rate": 8.812815157215466e-05, "loss": 0.22859432697296142, "memory(GiB)": 61.91, "step": 26135, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.461698 }, { "epoch": 1.1199177413135684, "grad_norm": 0.6234250664710999, "learning_rate": 8.812379764292535e-05, "loss": 0.1639164447784424, "memory(GiB)": 61.91, "step": 26140, "token_acc": 0.9696969696969697, "train_speed(iter/s)": 1.461703 }, { "epoch": 1.1201319566428174, "grad_norm": 7.03093147277832, "learning_rate": 8.81194430230477e-05, "loss": 0.5368332862854004, "memory(GiB)": 61.91, "step": 26145, "token_acc": 0.8848484848484849, "train_speed(iter/s)": 1.461704 }, { "epoch": 1.1203461719720664, "grad_norm": 3.659715414047241, "learning_rate": 8.811508771260058e-05, "loss": 0.35149564743041994, "memory(GiB)": 61.91, "step": 26150, "token_acc": 0.927797833935018, "train_speed(iter/s)": 1.461718 }, { "epoch": 1.1205603873013152, "grad_norm": 6.870719909667969, "learning_rate": 8.811073171166288e-05, "loss": 0.5246157646179199, "memory(GiB)": 61.91, "step": 26155, "token_acc": 0.8832116788321168, "train_speed(iter/s)": 1.461717 }, { "epoch": 1.1207746026305643, "grad_norm": 2.9553987979888916, "learning_rate": 8.810637502031354e-05, "loss": 0.22522883415222167, "memory(GiB)": 61.91, "step": 26160, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.461703 }, { "epoch": 1.1209888179598133, "grad_norm": 1.4802271127700806, "learning_rate": 8.810201763863145e-05, "loss": 0.35056252479553224, "memory(GiB)": 61.91, "step": 26165, "token_acc": 0.91701244813278, "train_speed(iter/s)": 1.46171 }, { "epoch": 1.1212030332890621, "grad_norm": 2.2727434635162354, "learning_rate": 8.809765956669558e-05, "loss": 0.22972548007965088, "memory(GiB)": 61.91, "step": 26170, "token_acc": 0.9540636042402827, "train_speed(iter/s)": 1.461705 }, { "epoch": 1.1214172486183112, "grad_norm": 5.035428047180176, "learning_rate": 8.809330080458487e-05, "loss": 0.609769630432129, "memory(GiB)": 61.91, "step": 26175, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.461699 }, { "epoch": 1.1216314639475602, "grad_norm": 4.777257442474365, "learning_rate": 8.808894135237828e-05, "loss": 0.550982666015625, "memory(GiB)": 61.91, "step": 26180, "token_acc": 0.8766066838046273, "train_speed(iter/s)": 1.461685 }, { "epoch": 1.121845679276809, "grad_norm": 2.739004611968994, "learning_rate": 8.808458121015477e-05, "loss": 0.15362064838409423, "memory(GiB)": 61.91, "step": 26185, "token_acc": 0.9629629629629629, "train_speed(iter/s)": 1.46168 }, { "epoch": 1.122059894606058, "grad_norm": 2.7337646484375, "learning_rate": 8.808022037799337e-05, "loss": 0.28790113925933836, "memory(GiB)": 61.91, "step": 26190, "token_acc": 0.9453376205787781, "train_speed(iter/s)": 1.461672 }, { "epoch": 1.122274109935307, "grad_norm": 1.677972674369812, "learning_rate": 8.807585885597305e-05, "loss": 0.40721921920776366, "memory(GiB)": 61.91, "step": 26195, "token_acc": 0.8912280701754386, "train_speed(iter/s)": 1.461703 }, { "epoch": 1.1224883252645559, "grad_norm": 5.349055767059326, "learning_rate": 8.807149664417283e-05, "loss": 0.46950383186340333, "memory(GiB)": 61.91, "step": 26200, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.461696 }, { "epoch": 1.122702540593805, "grad_norm": 0.5679978132247925, "learning_rate": 8.806713374267172e-05, "loss": 0.6093727111816406, "memory(GiB)": 61.91, "step": 26205, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.461724 }, { "epoch": 1.122916755923054, "grad_norm": 2.052076816558838, "learning_rate": 8.806277015154877e-05, "loss": 0.6248156547546386, "memory(GiB)": 61.91, "step": 26210, "token_acc": 0.881578947368421, "train_speed(iter/s)": 1.461709 }, { "epoch": 1.1231309712523028, "grad_norm": 2.506183385848999, "learning_rate": 8.805840587088305e-05, "loss": 0.40237765312194823, "memory(GiB)": 61.91, "step": 26215, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.461711 }, { "epoch": 1.1233451865815518, "grad_norm": 1.0284641981124878, "learning_rate": 8.80540409007536e-05, "loss": 0.2713034629821777, "memory(GiB)": 61.91, "step": 26220, "token_acc": 0.9455252918287937, "train_speed(iter/s)": 1.461703 }, { "epoch": 1.1235594019108008, "grad_norm": 0.7613735795021057, "learning_rate": 8.80496752412395e-05, "loss": 0.36021335124969484, "memory(GiB)": 61.91, "step": 26225, "token_acc": 0.8993506493506493, "train_speed(iter/s)": 1.461704 }, { "epoch": 1.1237736172400496, "grad_norm": 1.4267903566360474, "learning_rate": 8.804530889241984e-05, "loss": 0.28552565574645994, "memory(GiB)": 61.91, "step": 26230, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.461714 }, { "epoch": 1.1239878325692987, "grad_norm": 1.1366699934005737, "learning_rate": 8.804094185437373e-05, "loss": 0.10604041814804077, "memory(GiB)": 61.91, "step": 26235, "token_acc": 0.9742765273311897, "train_speed(iter/s)": 1.461731 }, { "epoch": 1.1242020478985477, "grad_norm": 1.6727384328842163, "learning_rate": 8.803657412718025e-05, "loss": 0.5167356014251709, "memory(GiB)": 61.91, "step": 26240, "token_acc": 0.8833922261484098, "train_speed(iter/s)": 1.461784 }, { "epoch": 1.1244162632277965, "grad_norm": 1.7886157035827637, "learning_rate": 8.803220571091857e-05, "loss": 0.38741886615753174, "memory(GiB)": 61.91, "step": 26245, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.461772 }, { "epoch": 1.1246304785570456, "grad_norm": 2.2191781997680664, "learning_rate": 8.80278366056678e-05, "loss": 0.3287331581115723, "memory(GiB)": 61.91, "step": 26250, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.461773 }, { "epoch": 1.1248446938862946, "grad_norm": 2.588775157928467, "learning_rate": 8.80234668115071e-05, "loss": 0.3286754608154297, "memory(GiB)": 61.91, "step": 26255, "token_acc": 0.9435483870967742, "train_speed(iter/s)": 1.46178 }, { "epoch": 1.1250589092155434, "grad_norm": 0.2267780303955078, "learning_rate": 8.801909632851561e-05, "loss": 0.24134862422943115, "memory(GiB)": 61.91, "step": 26260, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.461795 }, { "epoch": 1.1252731245447924, "grad_norm": 6.69584846496582, "learning_rate": 8.801472515677255e-05, "loss": 0.7560809135437012, "memory(GiB)": 61.91, "step": 26265, "token_acc": 0.8726708074534162, "train_speed(iter/s)": 1.461836 }, { "epoch": 1.1254873398740415, "grad_norm": 2.309577226638794, "learning_rate": 8.801035329635707e-05, "loss": 0.3805506229400635, "memory(GiB)": 61.91, "step": 26270, "token_acc": 0.9247648902821317, "train_speed(iter/s)": 1.461822 }, { "epoch": 1.1257015552032903, "grad_norm": 2.2896904945373535, "learning_rate": 8.800598074734839e-05, "loss": 0.23543791770935057, "memory(GiB)": 61.91, "step": 26275, "token_acc": 0.9405099150141643, "train_speed(iter/s)": 1.461811 }, { "epoch": 1.1259157705325393, "grad_norm": 2.063399314880371, "learning_rate": 8.800160750982572e-05, "loss": 0.35433077812194824, "memory(GiB)": 61.91, "step": 26280, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.461844 }, { "epoch": 1.1261299858617884, "grad_norm": 0.5752164721488953, "learning_rate": 8.799723358386826e-05, "loss": 0.19752790927886962, "memory(GiB)": 61.91, "step": 26285, "token_acc": 0.9543859649122807, "train_speed(iter/s)": 1.461853 }, { "epoch": 1.1263442011910372, "grad_norm": 4.012307643890381, "learning_rate": 8.79928589695553e-05, "loss": 0.8025521278381348, "memory(GiB)": 61.91, "step": 26290, "token_acc": 0.8689655172413793, "train_speed(iter/s)": 1.461876 }, { "epoch": 1.1265584165202862, "grad_norm": 4.477820873260498, "learning_rate": 8.798848366696604e-05, "loss": 0.4734935760498047, "memory(GiB)": 61.91, "step": 26295, "token_acc": 0.8969072164948454, "train_speed(iter/s)": 1.461867 }, { "epoch": 1.1267726318495352, "grad_norm": 6.467166900634766, "learning_rate": 8.798410767617977e-05, "loss": 0.26991212368011475, "memory(GiB)": 61.91, "step": 26300, "token_acc": 0.943609022556391, "train_speed(iter/s)": 1.461875 }, { "epoch": 1.126986847178784, "grad_norm": 3.980029582977295, "learning_rate": 8.797973099727575e-05, "loss": 0.45307326316833496, "memory(GiB)": 61.91, "step": 26305, "token_acc": 0.8867924528301887, "train_speed(iter/s)": 1.461881 }, { "epoch": 1.127201062508033, "grad_norm": 5.111351013183594, "learning_rate": 8.797535363033327e-05, "loss": 0.6769253253936768, "memory(GiB)": 61.91, "step": 26310, "token_acc": 0.8768656716417911, "train_speed(iter/s)": 1.461876 }, { "epoch": 1.1274152778372821, "grad_norm": 0.3300164043903351, "learning_rate": 8.797097557543166e-05, "loss": 0.18169888257980346, "memory(GiB)": 61.91, "step": 26315, "token_acc": 0.9680851063829787, "train_speed(iter/s)": 1.461874 }, { "epoch": 1.127629493166531, "grad_norm": 5.1741251945495605, "learning_rate": 8.796659683265019e-05, "loss": 0.5114869117736817, "memory(GiB)": 61.91, "step": 26320, "token_acc": 0.889273356401384, "train_speed(iter/s)": 1.461892 }, { "epoch": 1.12784370849578, "grad_norm": 1.7412159442901611, "learning_rate": 8.796221740206819e-05, "loss": 0.24938132762908935, "memory(GiB)": 61.91, "step": 26325, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.461898 }, { "epoch": 1.128057923825029, "grad_norm": 9.390755653381348, "learning_rate": 8.795783728376504e-05, "loss": 0.3866883277893066, "memory(GiB)": 61.91, "step": 26330, "token_acc": 0.9068100358422939, "train_speed(iter/s)": 1.46194 }, { "epoch": 1.1282721391542778, "grad_norm": 3.1858954429626465, "learning_rate": 8.795345647782002e-05, "loss": 0.33103716373443604, "memory(GiB)": 61.91, "step": 26335, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.461946 }, { "epoch": 1.1284863544835269, "grad_norm": 1.362095594406128, "learning_rate": 8.794907498431258e-05, "loss": 0.46091465950012206, "memory(GiB)": 61.91, "step": 26340, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.461947 }, { "epoch": 1.128700569812776, "grad_norm": 1.0109814405441284, "learning_rate": 8.794469280332203e-05, "loss": 0.17772690057754517, "memory(GiB)": 61.91, "step": 26345, "token_acc": 0.968944099378882, "train_speed(iter/s)": 1.461963 }, { "epoch": 1.1289147851420247, "grad_norm": 0.17211231589317322, "learning_rate": 8.794030993492777e-05, "loss": 0.29437663555145266, "memory(GiB)": 61.91, "step": 26350, "token_acc": 0.9509803921568627, "train_speed(iter/s)": 1.461981 }, { "epoch": 1.1291290004712737, "grad_norm": 3.646125078201294, "learning_rate": 8.79359263792092e-05, "loss": 0.2094897747039795, "memory(GiB)": 61.91, "step": 26355, "token_acc": 0.9516129032258065, "train_speed(iter/s)": 1.461986 }, { "epoch": 1.1293432158005228, "grad_norm": 1.7775609493255615, "learning_rate": 8.793154213624572e-05, "loss": 0.18389105796813965, "memory(GiB)": 61.91, "step": 26360, "token_acc": 0.947075208913649, "train_speed(iter/s)": 1.461982 }, { "epoch": 1.1295574311297716, "grad_norm": 3.1013858318328857, "learning_rate": 8.792715720611682e-05, "loss": 0.28110852241516116, "memory(GiB)": 61.91, "step": 26365, "token_acc": 0.9386973180076629, "train_speed(iter/s)": 1.461983 }, { "epoch": 1.1297716464590206, "grad_norm": 4.99979829788208, "learning_rate": 8.792277158890185e-05, "loss": 0.5510260581970214, "memory(GiB)": 61.91, "step": 26370, "token_acc": 0.8844884488448845, "train_speed(iter/s)": 1.461972 }, { "epoch": 1.1299858617882697, "grad_norm": 4.520094871520996, "learning_rate": 8.791838528468031e-05, "loss": 0.43732423782348634, "memory(GiB)": 61.91, "step": 26375, "token_acc": 0.9169960474308301, "train_speed(iter/s)": 1.461987 }, { "epoch": 1.1302000771175185, "grad_norm": 3.135410785675049, "learning_rate": 8.791399829353166e-05, "loss": 0.6960330009460449, "memory(GiB)": 61.91, "step": 26380, "token_acc": 0.8745980707395499, "train_speed(iter/s)": 1.461988 }, { "epoch": 1.1304142924467675, "grad_norm": 3.4536800384521484, "learning_rate": 8.790961061553535e-05, "loss": 0.7386884689331055, "memory(GiB)": 61.91, "step": 26385, "token_acc": 0.8490566037735849, "train_speed(iter/s)": 1.462008 }, { "epoch": 1.1306285077760165, "grad_norm": 4.701251029968262, "learning_rate": 8.79052222507709e-05, "loss": 0.34480297565460205, "memory(GiB)": 61.91, "step": 26390, "token_acc": 0.9388646288209607, "train_speed(iter/s)": 1.462008 }, { "epoch": 1.1308427231052653, "grad_norm": 4.125917434692383, "learning_rate": 8.790083319931776e-05, "loss": 0.33802318572998047, "memory(GiB)": 61.91, "step": 26395, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.462023 }, { "epoch": 1.1310569384345144, "grad_norm": 1.582581877708435, "learning_rate": 8.78964434612555e-05, "loss": 0.3007009029388428, "memory(GiB)": 61.91, "step": 26400, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.462021 }, { "epoch": 1.1312711537637634, "grad_norm": 5.054229259490967, "learning_rate": 8.789205303666361e-05, "loss": 0.36643755435943604, "memory(GiB)": 61.91, "step": 26405, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.462017 }, { "epoch": 1.1314853690930122, "grad_norm": 3.7279350757598877, "learning_rate": 8.788766192562162e-05, "loss": 0.48390746116638184, "memory(GiB)": 61.91, "step": 26410, "token_acc": 0.9177215189873418, "train_speed(iter/s)": 1.462024 }, { "epoch": 1.1316995844222613, "grad_norm": 5.66748046875, "learning_rate": 8.78832701282091e-05, "loss": 0.43849525451660154, "memory(GiB)": 61.91, "step": 26415, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.462035 }, { "epoch": 1.1319137997515103, "grad_norm": 5.298398017883301, "learning_rate": 8.787887764450561e-05, "loss": 0.4307565212249756, "memory(GiB)": 61.91, "step": 26420, "token_acc": 0.9119718309859155, "train_speed(iter/s)": 1.462046 }, { "epoch": 1.132128015080759, "grad_norm": 2.7949657440185547, "learning_rate": 8.787448447459073e-05, "loss": 0.3090996026992798, "memory(GiB)": 61.91, "step": 26425, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.462042 }, { "epoch": 1.1323422304100081, "grad_norm": 3.3758275508880615, "learning_rate": 8.787009061854401e-05, "loss": 0.41748714447021484, "memory(GiB)": 61.91, "step": 26430, "token_acc": 0.9101796407185628, "train_speed(iter/s)": 1.462044 }, { "epoch": 1.1325564457392572, "grad_norm": 4.111979961395264, "learning_rate": 8.786569607644509e-05, "loss": 0.6709149360656739, "memory(GiB)": 61.91, "step": 26435, "token_acc": 0.8595317725752508, "train_speed(iter/s)": 1.462065 }, { "epoch": 1.132770661068506, "grad_norm": 1.385623812675476, "learning_rate": 8.786130084837355e-05, "loss": 0.3760974645614624, "memory(GiB)": 61.91, "step": 26440, "token_acc": 0.9044368600682594, "train_speed(iter/s)": 1.462066 }, { "epoch": 1.132984876397755, "grad_norm": 5.607177257537842, "learning_rate": 8.785690493440904e-05, "loss": 0.2619167327880859, "memory(GiB)": 61.91, "step": 26445, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.462058 }, { "epoch": 1.133199091727004, "grad_norm": 5.212167739868164, "learning_rate": 8.78525083346312e-05, "loss": 0.2667985200881958, "memory(GiB)": 61.91, "step": 26450, "token_acc": 0.9486301369863014, "train_speed(iter/s)": 1.462072 }, { "epoch": 1.1334133070562529, "grad_norm": 1.5862174034118652, "learning_rate": 8.784811104911962e-05, "loss": 0.33330023288726807, "memory(GiB)": 61.91, "step": 26455, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.462076 }, { "epoch": 1.133627522385502, "grad_norm": 3.704350233078003, "learning_rate": 8.784371307795403e-05, "loss": 0.2546100616455078, "memory(GiB)": 61.91, "step": 26460, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.462098 }, { "epoch": 1.133841737714751, "grad_norm": 4.092438220977783, "learning_rate": 8.783931442121409e-05, "loss": 0.5389917850494385, "memory(GiB)": 61.91, "step": 26465, "token_acc": 0.8873720136518771, "train_speed(iter/s)": 1.46212 }, { "epoch": 1.1340559530439998, "grad_norm": 3.689357280731201, "learning_rate": 8.783491507897946e-05, "loss": 0.33945412635803224, "memory(GiB)": 61.91, "step": 26470, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.462142 }, { "epoch": 1.1342701683732488, "grad_norm": 3.49214506149292, "learning_rate": 8.783051505132984e-05, "loss": 0.35528762340545655, "memory(GiB)": 61.91, "step": 26475, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.462168 }, { "epoch": 1.1344843837024978, "grad_norm": 4.826119899749756, "learning_rate": 8.782611433834497e-05, "loss": 0.3468620300292969, "memory(GiB)": 61.91, "step": 26480, "token_acc": 0.9224489795918367, "train_speed(iter/s)": 1.462165 }, { "epoch": 1.1346985990317466, "grad_norm": 1.4491357803344727, "learning_rate": 8.782171294010455e-05, "loss": 0.31558690071105955, "memory(GiB)": 61.91, "step": 26485, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.462178 }, { "epoch": 1.1349128143609957, "grad_norm": 11.554035186767578, "learning_rate": 8.781731085668831e-05, "loss": 0.48640732765197753, "memory(GiB)": 61.91, "step": 26490, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.462175 }, { "epoch": 1.1351270296902447, "grad_norm": 6.421611785888672, "learning_rate": 8.781290808817602e-05, "loss": 0.8330686569213868, "memory(GiB)": 61.91, "step": 26495, "token_acc": 0.834920634920635, "train_speed(iter/s)": 1.46222 }, { "epoch": 1.1353412450194935, "grad_norm": 3.354543447494507, "learning_rate": 8.780850463464742e-05, "loss": 0.2645491361618042, "memory(GiB)": 61.91, "step": 26500, "token_acc": 0.9497041420118343, "train_speed(iter/s)": 1.462231 }, { "epoch": 1.1353412450194935, "eval_loss": 2.7438652515411377, "eval_runtime": 13.9044, "eval_samples_per_second": 7.192, "eval_steps_per_second": 7.192, "eval_token_acc": 0.40942562592047127, "step": 26500 }, { "epoch": 1.1355554603487426, "grad_norm": 4.273571968078613, "learning_rate": 8.78041004961823e-05, "loss": 0.12824413776397706, "memory(GiB)": 61.91, "step": 26505, "token_acc": 0.5770423991726991, "train_speed(iter/s)": 1.46105 }, { "epoch": 1.1357696756779916, "grad_norm": 1.6561557054519653, "learning_rate": 8.779969567286044e-05, "loss": 0.3181114435195923, "memory(GiB)": 61.91, "step": 26510, "token_acc": 0.9368770764119602, "train_speed(iter/s)": 1.461042 }, { "epoch": 1.1359838910072404, "grad_norm": 3.3794522285461426, "learning_rate": 8.779529016476163e-05, "loss": 0.55096435546875, "memory(GiB)": 61.91, "step": 26515, "token_acc": 0.8908450704225352, "train_speed(iter/s)": 1.461052 }, { "epoch": 1.1361981063364894, "grad_norm": 4.289788722991943, "learning_rate": 8.779088397196567e-05, "loss": 0.42473573684692384, "memory(GiB)": 61.91, "step": 26520, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.461049 }, { "epoch": 1.1364123216657385, "grad_norm": 5.276933670043945, "learning_rate": 8.778647709455241e-05, "loss": 0.4002220153808594, "memory(GiB)": 61.91, "step": 26525, "token_acc": 0.9037267080745341, "train_speed(iter/s)": 1.461044 }, { "epoch": 1.1366265369949873, "grad_norm": 2.022319793701172, "learning_rate": 8.778206953260166e-05, "loss": 0.40973901748657227, "memory(GiB)": 61.91, "step": 26530, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.461049 }, { "epoch": 1.1368407523242363, "grad_norm": 2.8799374103546143, "learning_rate": 8.77776612861933e-05, "loss": 0.248968505859375, "memory(GiB)": 61.91, "step": 26535, "token_acc": 0.924901185770751, "train_speed(iter/s)": 1.461031 }, { "epoch": 1.1370549676534853, "grad_norm": 3.935370445251465, "learning_rate": 8.777325235540715e-05, "loss": 0.5882336139678955, "memory(GiB)": 61.91, "step": 26540, "token_acc": 0.9003115264797508, "train_speed(iter/s)": 1.461056 }, { "epoch": 1.1372691829827342, "grad_norm": 2.681248426437378, "learning_rate": 8.776884274032309e-05, "loss": 0.5274367809295655, "memory(GiB)": 61.91, "step": 26545, "token_acc": 0.8996865203761756, "train_speed(iter/s)": 1.461061 }, { "epoch": 1.1374833983119832, "grad_norm": 3.3033900260925293, "learning_rate": 8.776443244102103e-05, "loss": 0.274369478225708, "memory(GiB)": 61.91, "step": 26550, "token_acc": 0.9049429657794676, "train_speed(iter/s)": 1.461078 }, { "epoch": 1.1376976136412322, "grad_norm": 0.2870866358280182, "learning_rate": 8.776002145758084e-05, "loss": 0.30403268337249756, "memory(GiB)": 61.91, "step": 26555, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.46107 }, { "epoch": 1.137911828970481, "grad_norm": 3.863036870956421, "learning_rate": 8.775560979008244e-05, "loss": 0.3843610525131226, "memory(GiB)": 61.91, "step": 26560, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.46111 }, { "epoch": 1.13812604429973, "grad_norm": 3.814878225326538, "learning_rate": 8.775119743860576e-05, "loss": 0.5964557647705078, "memory(GiB)": 61.91, "step": 26565, "token_acc": 0.8904593639575972, "train_speed(iter/s)": 1.461148 }, { "epoch": 1.1383402596289791, "grad_norm": 7.409132957458496, "learning_rate": 8.774678440323072e-05, "loss": 0.43335652351379395, "memory(GiB)": 61.91, "step": 26570, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.461151 }, { "epoch": 1.138554474958228, "grad_norm": 2.808820962905884, "learning_rate": 8.774237068403727e-05, "loss": 0.589351224899292, "memory(GiB)": 61.91, "step": 26575, "token_acc": 0.875886524822695, "train_speed(iter/s)": 1.461155 }, { "epoch": 1.138768690287477, "grad_norm": 6.124139785766602, "learning_rate": 8.773795628110539e-05, "loss": 0.608174180984497, "memory(GiB)": 61.91, "step": 26580, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.461144 }, { "epoch": 1.138982905616726, "grad_norm": 2.435736894607544, "learning_rate": 8.7733541194515e-05, "loss": 0.39284548759460447, "memory(GiB)": 61.91, "step": 26585, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.461144 }, { "epoch": 1.1391971209459748, "grad_norm": 5.412548542022705, "learning_rate": 8.772912542434613e-05, "loss": 0.7333405971527099, "memory(GiB)": 61.91, "step": 26590, "token_acc": 0.8576051779935275, "train_speed(iter/s)": 1.461181 }, { "epoch": 1.1394113362752238, "grad_norm": 3.823418378829956, "learning_rate": 8.772470897067877e-05, "loss": 0.1958150625228882, "memory(GiB)": 61.91, "step": 26595, "token_acc": 0.9457627118644067, "train_speed(iter/s)": 1.461176 }, { "epoch": 1.1396255516044729, "grad_norm": 1.6777039766311646, "learning_rate": 8.772029183359289e-05, "loss": 0.36051928997039795, "memory(GiB)": 61.91, "step": 26600, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.461184 }, { "epoch": 1.1398397669337217, "grad_norm": 6.240048885345459, "learning_rate": 8.771587401316856e-05, "loss": 0.4356318473815918, "memory(GiB)": 61.91, "step": 26605, "token_acc": 0.8880597014925373, "train_speed(iter/s)": 1.461173 }, { "epoch": 1.1400539822629707, "grad_norm": 0.8438418507575989, "learning_rate": 8.771145550948579e-05, "loss": 0.2984419584274292, "memory(GiB)": 61.91, "step": 26610, "token_acc": 0.926605504587156, "train_speed(iter/s)": 1.461174 }, { "epoch": 1.1402681975922198, "grad_norm": 6.933533191680908, "learning_rate": 8.770703632262463e-05, "loss": 0.3161618232727051, "memory(GiB)": 61.91, "step": 26615, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.461179 }, { "epoch": 1.1404824129214686, "grad_norm": 15.130616188049316, "learning_rate": 8.770261645266513e-05, "loss": 0.3326070547103882, "memory(GiB)": 61.91, "step": 26620, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.46118 }, { "epoch": 1.1406966282507176, "grad_norm": 4.683438301086426, "learning_rate": 8.769819589968737e-05, "loss": 0.6716950416564942, "memory(GiB)": 61.91, "step": 26625, "token_acc": 0.8548895899053628, "train_speed(iter/s)": 1.461219 }, { "epoch": 1.1409108435799666, "grad_norm": 4.932214260101318, "learning_rate": 8.76937746637714e-05, "loss": 0.5809123516082764, "memory(GiB)": 61.91, "step": 26630, "token_acc": 0.8966789667896679, "train_speed(iter/s)": 1.461204 }, { "epoch": 1.1411250589092155, "grad_norm": 3.569580078125, "learning_rate": 8.768935274499738e-05, "loss": 0.4189265251159668, "memory(GiB)": 61.91, "step": 26635, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.461195 }, { "epoch": 1.1413392742384645, "grad_norm": 5.013300895690918, "learning_rate": 8.768493014344536e-05, "loss": 0.6613543033599854, "memory(GiB)": 61.91, "step": 26640, "token_acc": 0.8796296296296297, "train_speed(iter/s)": 1.461189 }, { "epoch": 1.1415534895677135, "grad_norm": 7.03164005279541, "learning_rate": 8.768050685919548e-05, "loss": 0.3545989513397217, "memory(GiB)": 61.91, "step": 26645, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.461196 }, { "epoch": 1.1417677048969623, "grad_norm": 3.2936513423919678, "learning_rate": 8.767608289232787e-05, "loss": 0.41465468406677247, "memory(GiB)": 61.91, "step": 26650, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.461194 }, { "epoch": 1.1419819202262114, "grad_norm": 2.45546293258667, "learning_rate": 8.767165824292269e-05, "loss": 0.5031896591186523, "memory(GiB)": 61.91, "step": 26655, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.461234 }, { "epoch": 1.1421961355554604, "grad_norm": 4.533189296722412, "learning_rate": 8.766723291106007e-05, "loss": 0.4757691860198975, "memory(GiB)": 61.91, "step": 26660, "token_acc": 0.891156462585034, "train_speed(iter/s)": 1.461244 }, { "epoch": 1.1424103508847092, "grad_norm": 3.5602285861968994, "learning_rate": 8.76628068968202e-05, "loss": 0.6425958156585694, "memory(GiB)": 61.91, "step": 26665, "token_acc": 0.8527397260273972, "train_speed(iter/s)": 1.461326 }, { "epoch": 1.1426245662139582, "grad_norm": 2.2294564247131348, "learning_rate": 8.765838020028322e-05, "loss": 0.14861805438995362, "memory(GiB)": 61.91, "step": 26670, "token_acc": 0.9695817490494296, "train_speed(iter/s)": 1.461358 }, { "epoch": 1.1428387815432073, "grad_norm": 4.039392948150635, "learning_rate": 8.76539528215294e-05, "loss": 0.47201013565063477, "memory(GiB)": 61.91, "step": 26675, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.461381 }, { "epoch": 1.143052996872456, "grad_norm": 2.3085455894470215, "learning_rate": 8.764952476063887e-05, "loss": 0.45480642318725584, "memory(GiB)": 61.91, "step": 26680, "token_acc": 0.9080882352941176, "train_speed(iter/s)": 1.461385 }, { "epoch": 1.1432672122017051, "grad_norm": 2.8155980110168457, "learning_rate": 8.76450960176919e-05, "loss": 0.2605112552642822, "memory(GiB)": 61.91, "step": 26685, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.461388 }, { "epoch": 1.1434814275309542, "grad_norm": 2.4411351680755615, "learning_rate": 8.76406665927687e-05, "loss": 0.22828738689422606, "memory(GiB)": 61.91, "step": 26690, "token_acc": 0.9515570934256056, "train_speed(iter/s)": 1.461393 }, { "epoch": 1.143695642860203, "grad_norm": 3.1956703662872314, "learning_rate": 8.763623648594952e-05, "loss": 0.4074734687805176, "memory(GiB)": 61.91, "step": 26695, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.46139 }, { "epoch": 1.143909858189452, "grad_norm": 4.64717435836792, "learning_rate": 8.76318056973146e-05, "loss": 0.3331305503845215, "memory(GiB)": 61.91, "step": 26700, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.461379 }, { "epoch": 1.144124073518701, "grad_norm": 3.197009325027466, "learning_rate": 8.762737422694423e-05, "loss": 0.27103993892669676, "memory(GiB)": 61.91, "step": 26705, "token_acc": 0.9338235294117647, "train_speed(iter/s)": 1.461402 }, { "epoch": 1.1443382888479499, "grad_norm": 3.5140109062194824, "learning_rate": 8.762294207491865e-05, "loss": 0.4616459846496582, "memory(GiB)": 61.91, "step": 26710, "token_acc": 0.9119496855345912, "train_speed(iter/s)": 1.461398 }, { "epoch": 1.144552504177199, "grad_norm": 10.521489143371582, "learning_rate": 8.761850924131821e-05, "loss": 0.36064810752868653, "memory(GiB)": 61.91, "step": 26715, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.461414 }, { "epoch": 1.144766719506448, "grad_norm": 1.628326654434204, "learning_rate": 8.761407572622318e-05, "loss": 0.41204233169555665, "memory(GiB)": 61.91, "step": 26720, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.46142 }, { "epoch": 1.1449809348356967, "grad_norm": 2.1768410205841064, "learning_rate": 8.760964152971387e-05, "loss": 0.41211891174316406, "memory(GiB)": 61.91, "step": 26725, "token_acc": 0.8968253968253969, "train_speed(iter/s)": 1.461422 }, { "epoch": 1.1451951501649458, "grad_norm": 4.617672920227051, "learning_rate": 8.760520665187064e-05, "loss": 0.3817413091659546, "memory(GiB)": 61.91, "step": 26730, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.461434 }, { "epoch": 1.1454093654941948, "grad_norm": 0.8221848607063293, "learning_rate": 8.760077109277381e-05, "loss": 0.2713869333267212, "memory(GiB)": 61.91, "step": 26735, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.461442 }, { "epoch": 1.1456235808234436, "grad_norm": 2.6634857654571533, "learning_rate": 8.759633485250372e-05, "loss": 0.32805671691894533, "memory(GiB)": 61.91, "step": 26740, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.461467 }, { "epoch": 1.1458377961526927, "grad_norm": 5.354935169219971, "learning_rate": 8.759189793114077e-05, "loss": 0.5214598655700684, "memory(GiB)": 61.91, "step": 26745, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.461495 }, { "epoch": 1.1460520114819417, "grad_norm": 0.3496028780937195, "learning_rate": 8.758746032876534e-05, "loss": 0.2831345796585083, "memory(GiB)": 61.91, "step": 26750, "token_acc": 0.943217665615142, "train_speed(iter/s)": 1.461531 }, { "epoch": 1.1462662268111905, "grad_norm": 4.514954090118408, "learning_rate": 8.758302204545779e-05, "loss": 0.5584105491638184, "memory(GiB)": 61.91, "step": 26755, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.461531 }, { "epoch": 1.1464804421404395, "grad_norm": 1.5539497137069702, "learning_rate": 8.757858308129854e-05, "loss": 0.32678513526916503, "memory(GiB)": 61.91, "step": 26760, "token_acc": 0.9171974522292994, "train_speed(iter/s)": 1.461536 }, { "epoch": 1.1466946574696886, "grad_norm": 2.7008018493652344, "learning_rate": 8.7574143436368e-05, "loss": 0.3492297887802124, "memory(GiB)": 61.91, "step": 26765, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.461547 }, { "epoch": 1.1469088727989374, "grad_norm": 2.5524442195892334, "learning_rate": 8.756970311074663e-05, "loss": 0.4269843578338623, "memory(GiB)": 61.91, "step": 26770, "token_acc": 0.9094488188976378, "train_speed(iter/s)": 1.461554 }, { "epoch": 1.1471230881281864, "grad_norm": 2.3108484745025635, "learning_rate": 8.756526210451483e-05, "loss": 0.26482362747192384, "memory(GiB)": 61.91, "step": 26775, "token_acc": 0.9353846153846154, "train_speed(iter/s)": 1.461555 }, { "epoch": 1.1473373034574355, "grad_norm": 2.869001626968384, "learning_rate": 8.756082041775307e-05, "loss": 0.6392002582550049, "memory(GiB)": 61.91, "step": 26780, "token_acc": 0.871875, "train_speed(iter/s)": 1.461598 }, { "epoch": 1.1475515187866843, "grad_norm": 6.049563407897949, "learning_rate": 8.75563780505418e-05, "loss": 0.6253294467926025, "memory(GiB)": 61.91, "step": 26785, "token_acc": 0.8726591760299626, "train_speed(iter/s)": 1.461587 }, { "epoch": 1.1477657341159333, "grad_norm": 1.2965211868286133, "learning_rate": 8.755193500296152e-05, "loss": 0.3923739671707153, "memory(GiB)": 61.91, "step": 26790, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.461593 }, { "epoch": 1.1479799494451823, "grad_norm": 7.631797790527344, "learning_rate": 8.75474912750927e-05, "loss": 0.21072850227355958, "memory(GiB)": 61.91, "step": 26795, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.461596 }, { "epoch": 1.1481941647744311, "grad_norm": 2.7792611122131348, "learning_rate": 8.754304686701588e-05, "loss": 0.4612226963043213, "memory(GiB)": 61.91, "step": 26800, "token_acc": 0.8919667590027701, "train_speed(iter/s)": 1.461581 }, { "epoch": 1.1484083801036802, "grad_norm": 5.018490791320801, "learning_rate": 8.753860177881153e-05, "loss": 0.2552131414413452, "memory(GiB)": 61.91, "step": 26805, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.461585 }, { "epoch": 1.1486225954329292, "grad_norm": 4.857707977294922, "learning_rate": 8.753415601056019e-05, "loss": 0.338424825668335, "memory(GiB)": 61.91, "step": 26810, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.461578 }, { "epoch": 1.148836810762178, "grad_norm": 5.596978187561035, "learning_rate": 8.752970956234242e-05, "loss": 0.4507925987243652, "memory(GiB)": 61.91, "step": 26815, "token_acc": 0.9196141479099679, "train_speed(iter/s)": 1.461583 }, { "epoch": 1.149051026091427, "grad_norm": 2.386216878890991, "learning_rate": 8.752526243423874e-05, "loss": 0.4013469696044922, "memory(GiB)": 61.91, "step": 26820, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.461588 }, { "epoch": 1.149265241420676, "grad_norm": 2.2629246711730957, "learning_rate": 8.752081462632973e-05, "loss": 0.404984188079834, "memory(GiB)": 61.91, "step": 26825, "token_acc": 0.9, "train_speed(iter/s)": 1.461592 }, { "epoch": 1.149479456749925, "grad_norm": 2.845334529876709, "learning_rate": 8.751636613869599e-05, "loss": 0.2315990447998047, "memory(GiB)": 61.91, "step": 26830, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.461589 }, { "epoch": 1.149693672079174, "grad_norm": 3.614377975463867, "learning_rate": 8.751191697141805e-05, "loss": 0.2555378437042236, "memory(GiB)": 61.91, "step": 26835, "token_acc": 0.9553264604810997, "train_speed(iter/s)": 1.461612 }, { "epoch": 1.149907887408423, "grad_norm": 1.2070457935333252, "learning_rate": 8.750746712457656e-05, "loss": 0.4660207748413086, "memory(GiB)": 61.91, "step": 26840, "token_acc": 0.911042944785276, "train_speed(iter/s)": 1.461634 }, { "epoch": 1.1501221027376718, "grad_norm": 6.598074913024902, "learning_rate": 8.750301659825213e-05, "loss": 0.41602044105529784, "memory(GiB)": 61.91, "step": 26845, "token_acc": 0.8916967509025271, "train_speed(iter/s)": 1.461631 }, { "epoch": 1.1503363180669208, "grad_norm": 3.4315176010131836, "learning_rate": 8.749856539252537e-05, "loss": 0.35334115028381347, "memory(GiB)": 61.91, "step": 26850, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.461692 }, { "epoch": 1.1505505333961699, "grad_norm": 3.621868133544922, "learning_rate": 8.749411350747691e-05, "loss": 0.23037028312683105, "memory(GiB)": 61.91, "step": 26855, "token_acc": 0.9488636363636364, "train_speed(iter/s)": 1.461677 }, { "epoch": 1.1507647487254187, "grad_norm": 3.2805111408233643, "learning_rate": 8.74896609431874e-05, "loss": 0.44887447357177734, "memory(GiB)": 61.91, "step": 26860, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.461666 }, { "epoch": 1.1509789640546677, "grad_norm": 5.436990261077881, "learning_rate": 8.748520769973754e-05, "loss": 0.36088948249816893, "memory(GiB)": 61.91, "step": 26865, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.461659 }, { "epoch": 1.1511931793839167, "grad_norm": 1.8816511631011963, "learning_rate": 8.748075377720796e-05, "loss": 0.5005486965179443, "memory(GiB)": 61.91, "step": 26870, "token_acc": 0.8940809968847352, "train_speed(iter/s)": 1.461669 }, { "epoch": 1.1514073947131656, "grad_norm": 0.316930890083313, "learning_rate": 8.747629917567938e-05, "loss": 0.21820466518402098, "memory(GiB)": 61.91, "step": 26875, "token_acc": 0.9518072289156626, "train_speed(iter/s)": 1.461662 }, { "epoch": 1.1516216100424146, "grad_norm": 0.2483154833316803, "learning_rate": 8.747184389523249e-05, "loss": 0.3985395908355713, "memory(GiB)": 61.91, "step": 26880, "token_acc": 0.9189944134078212, "train_speed(iter/s)": 1.46168 }, { "epoch": 1.1518358253716636, "grad_norm": 2.2917938232421875, "learning_rate": 8.746738793594798e-05, "loss": 0.6126693248748779, "memory(GiB)": 61.91, "step": 26885, "token_acc": 0.8600583090379009, "train_speed(iter/s)": 1.461669 }, { "epoch": 1.1520500407009124, "grad_norm": 2.5798611640930176, "learning_rate": 8.746293129790659e-05, "loss": 0.6437596797943115, "memory(GiB)": 61.91, "step": 26890, "token_acc": 0.8637602179836512, "train_speed(iter/s)": 1.461658 }, { "epoch": 1.1522642560301615, "grad_norm": 2.5376362800598145, "learning_rate": 8.745847398118906e-05, "loss": 0.4123214244842529, "memory(GiB)": 61.91, "step": 26895, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.461679 }, { "epoch": 1.1524784713594105, "grad_norm": 5.213351249694824, "learning_rate": 8.745401598587614e-05, "loss": 0.5159083366394043, "memory(GiB)": 61.91, "step": 26900, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.461679 }, { "epoch": 1.1526926866886593, "grad_norm": 2.3423850536346436, "learning_rate": 8.744955731204858e-05, "loss": 0.29472510814666747, "memory(GiB)": 61.91, "step": 26905, "token_acc": 0.9442231075697212, "train_speed(iter/s)": 1.461685 }, { "epoch": 1.1529069020179084, "grad_norm": 0.2404128909111023, "learning_rate": 8.744509795978716e-05, "loss": 0.4204659938812256, "memory(GiB)": 61.91, "step": 26910, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.461665 }, { "epoch": 1.1531211173471574, "grad_norm": 3.8732352256774902, "learning_rate": 8.744063792917267e-05, "loss": 0.7047980785369873, "memory(GiB)": 61.91, "step": 26915, "token_acc": 0.8598130841121495, "train_speed(iter/s)": 1.461693 }, { "epoch": 1.1533353326764062, "grad_norm": 8.14934253692627, "learning_rate": 8.74361772202859e-05, "loss": 0.4716783046722412, "memory(GiB)": 61.91, "step": 26920, "token_acc": 0.8976897689768977, "train_speed(iter/s)": 1.46169 }, { "epoch": 1.1535495480056552, "grad_norm": 3.923034429550171, "learning_rate": 8.743171583320765e-05, "loss": 0.5041105270385742, "memory(GiB)": 61.91, "step": 26925, "token_acc": 0.89272030651341, "train_speed(iter/s)": 1.461681 }, { "epoch": 1.1537637633349043, "grad_norm": 3.038620710372925, "learning_rate": 8.742725376801877e-05, "loss": 0.49754629135131834, "memory(GiB)": 61.91, "step": 26930, "token_acc": 0.889273356401384, "train_speed(iter/s)": 1.461668 }, { "epoch": 1.1539779786641533, "grad_norm": 2.773599863052368, "learning_rate": 8.742279102480007e-05, "loss": 0.600126075744629, "memory(GiB)": 61.91, "step": 26935, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.461668 }, { "epoch": 1.1541921939934021, "grad_norm": 4.7869648933410645, "learning_rate": 8.74183276036324e-05, "loss": 0.6292204856872559, "memory(GiB)": 61.91, "step": 26940, "token_acc": 0.8664122137404581, "train_speed(iter/s)": 1.46169 }, { "epoch": 1.1544064093226512, "grad_norm": 3.7261319160461426, "learning_rate": 8.741386350459663e-05, "loss": 0.4502689361572266, "memory(GiB)": 61.91, "step": 26945, "token_acc": 0.8853046594982079, "train_speed(iter/s)": 1.46171 }, { "epoch": 1.1546206246519002, "grad_norm": 1.4275434017181396, "learning_rate": 8.740939872777363e-05, "loss": 0.3404550075531006, "memory(GiB)": 61.91, "step": 26950, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.461703 }, { "epoch": 1.154834839981149, "grad_norm": 1.018642783164978, "learning_rate": 8.740493327324425e-05, "loss": 0.3515539407730103, "memory(GiB)": 61.91, "step": 26955, "token_acc": 0.9274447949526814, "train_speed(iter/s)": 1.461741 }, { "epoch": 1.155049055310398, "grad_norm": 4.763855934143066, "learning_rate": 8.740046714108947e-05, "loss": 0.5544607162475585, "memory(GiB)": 61.91, "step": 26960, "token_acc": 0.8859060402684564, "train_speed(iter/s)": 1.46174 }, { "epoch": 1.155263270639647, "grad_norm": 2.9768805503845215, "learning_rate": 8.739600033139011e-05, "loss": 0.6591886043548584, "memory(GiB)": 61.91, "step": 26965, "token_acc": 0.8579710144927536, "train_speed(iter/s)": 1.461741 }, { "epoch": 1.1554774859688959, "grad_norm": 2.7662746906280518, "learning_rate": 8.739153284422712e-05, "loss": 0.3039403915405273, "memory(GiB)": 61.91, "step": 26970, "token_acc": 0.925531914893617, "train_speed(iter/s)": 1.461734 }, { "epoch": 1.155691701298145, "grad_norm": 4.4477763175964355, "learning_rate": 8.738706467968145e-05, "loss": 0.16495518684387206, "memory(GiB)": 61.91, "step": 26975, "token_acc": 0.9580152671755725, "train_speed(iter/s)": 1.461738 }, { "epoch": 1.155905916627394, "grad_norm": 4.663832664489746, "learning_rate": 8.738259583783401e-05, "loss": 0.4200472354888916, "memory(GiB)": 61.91, "step": 26980, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.461747 }, { "epoch": 1.1561201319566428, "grad_norm": 2.6624226570129395, "learning_rate": 8.737812631876581e-05, "loss": 0.27507672309875486, "memory(GiB)": 61.91, "step": 26985, "token_acc": 0.942652329749104, "train_speed(iter/s)": 1.461761 }, { "epoch": 1.1563343472858918, "grad_norm": 3.916761875152588, "learning_rate": 8.737365612255777e-05, "loss": 0.6516876220703125, "memory(GiB)": 61.91, "step": 26990, "token_acc": 0.8741007194244604, "train_speed(iter/s)": 1.461815 }, { "epoch": 1.1565485626151408, "grad_norm": 6.223006725311279, "learning_rate": 8.736918524929088e-05, "loss": 0.36646764278411864, "memory(GiB)": 61.91, "step": 26995, "token_acc": 0.9226190476190477, "train_speed(iter/s)": 1.461814 }, { "epoch": 1.1567627779443896, "grad_norm": 0.3463360071182251, "learning_rate": 8.736471369904616e-05, "loss": 0.15181455612182618, "memory(GiB)": 61.91, "step": 27000, "token_acc": 0.9666666666666667, "train_speed(iter/s)": 1.461822 }, { "epoch": 1.1567627779443896, "eval_loss": 2.429696559906006, "eval_runtime": 13.5526, "eval_samples_per_second": 7.379, "eval_steps_per_second": 7.379, "eval_token_acc": 0.43854748603351956, "step": 27000 }, { "epoch": 1.1569769932736387, "grad_norm": 2.5531721115112305, "learning_rate": 8.73602414719046e-05, "loss": 0.23930234909057618, "memory(GiB)": 61.91, "step": 27005, "token_acc": 0.5899705014749262, "train_speed(iter/s)": 1.46067 }, { "epoch": 1.1571912086028877, "grad_norm": 3.0251352787017822, "learning_rate": 8.735576856794722e-05, "loss": 0.37264466285705566, "memory(GiB)": 61.91, "step": 27010, "token_acc": 0.921161825726141, "train_speed(iter/s)": 1.460679 }, { "epoch": 1.1574054239321365, "grad_norm": 1.9043092727661133, "learning_rate": 8.735129498725504e-05, "loss": 0.36947178840637207, "memory(GiB)": 61.91, "step": 27015, "token_acc": 0.9226519337016574, "train_speed(iter/s)": 1.460681 }, { "epoch": 1.1576196392613856, "grad_norm": 2.872340202331543, "learning_rate": 8.734682072990912e-05, "loss": 0.4814347743988037, "memory(GiB)": 61.91, "step": 27020, "token_acc": 0.8964497041420119, "train_speed(iter/s)": 1.460692 }, { "epoch": 1.1578338545906346, "grad_norm": 1.1264070272445679, "learning_rate": 8.734234579599049e-05, "loss": 0.3190495252609253, "memory(GiB)": 61.91, "step": 27025, "token_acc": 0.9327217125382263, "train_speed(iter/s)": 1.460703 }, { "epoch": 1.1580480699198834, "grad_norm": 4.4127936363220215, "learning_rate": 8.733787018558026e-05, "loss": 0.34917254447937013, "memory(GiB)": 61.91, "step": 27030, "token_acc": 0.9006849315068494, "train_speed(iter/s)": 1.46073 }, { "epoch": 1.1582622852491324, "grad_norm": 7.701265335083008, "learning_rate": 8.733339389875947e-05, "loss": 0.29199419021606443, "memory(GiB)": 61.91, "step": 27035, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.460724 }, { "epoch": 1.1584765005783815, "grad_norm": 2.2245240211486816, "learning_rate": 8.732891693560926e-05, "loss": 0.3258677959442139, "memory(GiB)": 61.91, "step": 27040, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.460715 }, { "epoch": 1.1586907159076303, "grad_norm": 2.8822805881500244, "learning_rate": 8.732443929621067e-05, "loss": 0.2705258846282959, "memory(GiB)": 61.91, "step": 27045, "token_acc": 0.9475409836065574, "train_speed(iter/s)": 1.460721 }, { "epoch": 1.1589049312368793, "grad_norm": 3.4608795642852783, "learning_rate": 8.731996098064485e-05, "loss": 0.4080799579620361, "memory(GiB)": 61.91, "step": 27050, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.46078 }, { "epoch": 1.1591191465661284, "grad_norm": 4.119937896728516, "learning_rate": 8.731548198899295e-05, "loss": 0.44458322525024413, "memory(GiB)": 61.91, "step": 27055, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.46077 }, { "epoch": 1.1593333618953772, "grad_norm": 3.356840133666992, "learning_rate": 8.731100232133607e-05, "loss": 0.3473116397857666, "memory(GiB)": 61.91, "step": 27060, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.460772 }, { "epoch": 1.1595475772246262, "grad_norm": 3.8236172199249268, "learning_rate": 8.730652197775538e-05, "loss": 0.7024046897888183, "memory(GiB)": 61.91, "step": 27065, "token_acc": 0.8629629629629629, "train_speed(iter/s)": 1.460782 }, { "epoch": 1.1597617925538752, "grad_norm": 1.732124924659729, "learning_rate": 8.730204095833205e-05, "loss": 0.3555913925170898, "memory(GiB)": 61.91, "step": 27070, "token_acc": 0.9269005847953217, "train_speed(iter/s)": 1.460792 }, { "epoch": 1.159976007883124, "grad_norm": 4.695122241973877, "learning_rate": 8.729755926314727e-05, "loss": 0.3641330718994141, "memory(GiB)": 61.91, "step": 27075, "token_acc": 0.9108527131782945, "train_speed(iter/s)": 1.460816 }, { "epoch": 1.160190223212373, "grad_norm": 3.788940906524658, "learning_rate": 8.72930768922822e-05, "loss": 0.36044723987579347, "memory(GiB)": 61.91, "step": 27080, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.460799 }, { "epoch": 1.1604044385416221, "grad_norm": 0.6726099848747253, "learning_rate": 8.728859384581807e-05, "loss": 0.2768702030181885, "memory(GiB)": 61.91, "step": 27085, "token_acc": 0.9498525073746312, "train_speed(iter/s)": 1.460803 }, { "epoch": 1.160618653870871, "grad_norm": 1.037795901298523, "learning_rate": 8.728411012383609e-05, "loss": 0.36952362060546873, "memory(GiB)": 61.91, "step": 27090, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.460796 }, { "epoch": 1.16083286920012, "grad_norm": 1.9771710634231567, "learning_rate": 8.727962572641746e-05, "loss": 0.3629880905151367, "memory(GiB)": 61.91, "step": 27095, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.460785 }, { "epoch": 1.161047084529369, "grad_norm": 4.5945820808410645, "learning_rate": 8.727514065364344e-05, "loss": 0.4015196800231934, "memory(GiB)": 61.91, "step": 27100, "token_acc": 0.9085173501577287, "train_speed(iter/s)": 1.460795 }, { "epoch": 1.1612612998586178, "grad_norm": 13.400678634643555, "learning_rate": 8.72706549055953e-05, "loss": 0.6496493339538574, "memory(GiB)": 61.91, "step": 27105, "token_acc": 0.8860294117647058, "train_speed(iter/s)": 1.46081 }, { "epoch": 1.1614755151878668, "grad_norm": 1.4677107334136963, "learning_rate": 8.726616848235426e-05, "loss": 0.5809225082397461, "memory(GiB)": 61.91, "step": 27110, "token_acc": 0.8766666666666667, "train_speed(iter/s)": 1.460842 }, { "epoch": 1.1616897305171159, "grad_norm": 4.834376811981201, "learning_rate": 8.726168138400163e-05, "loss": 0.24409055709838867, "memory(GiB)": 61.91, "step": 27115, "token_acc": 0.9407114624505929, "train_speed(iter/s)": 1.460833 }, { "epoch": 1.1619039458463647, "grad_norm": 3.2266368865966797, "learning_rate": 8.725719361061868e-05, "loss": 0.2937380313873291, "memory(GiB)": 61.91, "step": 27120, "token_acc": 0.9455782312925171, "train_speed(iter/s)": 1.460826 }, { "epoch": 1.1621181611756137, "grad_norm": 6.295896053314209, "learning_rate": 8.725270516228674e-05, "loss": 0.47815933227539065, "memory(GiB)": 61.91, "step": 27125, "token_acc": 0.8850174216027874, "train_speed(iter/s)": 1.460828 }, { "epoch": 1.1623323765048628, "grad_norm": 6.782707214355469, "learning_rate": 8.724821603908708e-05, "loss": 0.48696441650390626, "memory(GiB)": 61.91, "step": 27130, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.460841 }, { "epoch": 1.1625465918341116, "grad_norm": 2.87168550491333, "learning_rate": 8.724372624110105e-05, "loss": 0.8753524780273437, "memory(GiB)": 61.91, "step": 27135, "token_acc": 0.8054607508532423, "train_speed(iter/s)": 1.460899 }, { "epoch": 1.1627608071633606, "grad_norm": 2.4264421463012695, "learning_rate": 8.723923576840997e-05, "loss": 0.42538604736328123, "memory(GiB)": 61.91, "step": 27140, "token_acc": 0.903010033444816, "train_speed(iter/s)": 1.460906 }, { "epoch": 1.1629750224926096, "grad_norm": 4.042639255523682, "learning_rate": 8.723474462109522e-05, "loss": 0.46027507781982424, "memory(GiB)": 61.91, "step": 27145, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.460893 }, { "epoch": 1.1631892378218585, "grad_norm": 3.2272777557373047, "learning_rate": 8.723025279923813e-05, "loss": 0.5300007343292237, "memory(GiB)": 61.91, "step": 27150, "token_acc": 0.8899082568807339, "train_speed(iter/s)": 1.460881 }, { "epoch": 1.1634034531511075, "grad_norm": 0.7378474473953247, "learning_rate": 8.722576030292009e-05, "loss": 0.28441083431243896, "memory(GiB)": 61.91, "step": 27155, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.46088 }, { "epoch": 1.1636176684803565, "grad_norm": 4.1307053565979, "learning_rate": 8.72212671322225e-05, "loss": 0.406647777557373, "memory(GiB)": 61.91, "step": 27160, "token_acc": 0.9180887372013652, "train_speed(iter/s)": 1.460872 }, { "epoch": 1.1638318838096053, "grad_norm": 3.850874900817871, "learning_rate": 8.72167732872267e-05, "loss": 0.45705575942993165, "memory(GiB)": 61.91, "step": 27165, "token_acc": 0.8849557522123894, "train_speed(iter/s)": 1.460857 }, { "epoch": 1.1640460991388544, "grad_norm": 2.1271936893463135, "learning_rate": 8.721227876801417e-05, "loss": 0.44733333587646484, "memory(GiB)": 61.91, "step": 27170, "token_acc": 0.8990228013029316, "train_speed(iter/s)": 1.460883 }, { "epoch": 1.1642603144681034, "grad_norm": 2.8834433555603027, "learning_rate": 8.72077835746663e-05, "loss": 0.42781639099121094, "memory(GiB)": 61.91, "step": 27175, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.460899 }, { "epoch": 1.1644745297973522, "grad_norm": 2.0410032272338867, "learning_rate": 8.720328770726452e-05, "loss": 0.2662254810333252, "memory(GiB)": 61.91, "step": 27180, "token_acc": 0.9377593360995851, "train_speed(iter/s)": 1.4609 }, { "epoch": 1.1646887451266013, "grad_norm": 2.0117454528808594, "learning_rate": 8.719879116589028e-05, "loss": 0.2769863367080688, "memory(GiB)": 61.91, "step": 27185, "token_acc": 0.941358024691358, "train_speed(iter/s)": 1.460933 }, { "epoch": 1.1649029604558503, "grad_norm": 2.7450499534606934, "learning_rate": 8.719429395062504e-05, "loss": 0.43062796592712405, "memory(GiB)": 61.91, "step": 27190, "token_acc": 0.9075144508670521, "train_speed(iter/s)": 1.460936 }, { "epoch": 1.165117175785099, "grad_norm": 5.097678184509277, "learning_rate": 8.718979606155029e-05, "loss": 0.5683262348175049, "memory(GiB)": 61.91, "step": 27195, "token_acc": 0.8861209964412812, "train_speed(iter/s)": 1.460913 }, { "epoch": 1.1653313911143481, "grad_norm": 1.399099349975586, "learning_rate": 8.718529749874749e-05, "loss": 0.3681401491165161, "memory(GiB)": 61.91, "step": 27200, "token_acc": 0.8895348837209303, "train_speed(iter/s)": 1.460959 }, { "epoch": 1.1655456064435972, "grad_norm": 3.62467885017395, "learning_rate": 8.718079826229813e-05, "loss": 0.27970669269561765, "memory(GiB)": 61.91, "step": 27205, "token_acc": 0.94921875, "train_speed(iter/s)": 1.460982 }, { "epoch": 1.165759821772846, "grad_norm": 4.3549604415893555, "learning_rate": 8.717629835228375e-05, "loss": 0.34541032314300535, "memory(GiB)": 61.91, "step": 27210, "token_acc": 0.9237804878048781, "train_speed(iter/s)": 1.460998 }, { "epoch": 1.165974037102095, "grad_norm": 5.398664951324463, "learning_rate": 8.717179776878583e-05, "loss": 0.5803122520446777, "memory(GiB)": 61.91, "step": 27215, "token_acc": 0.8854166666666666, "train_speed(iter/s)": 1.461013 }, { "epoch": 1.166188252431344, "grad_norm": 2.3413476943969727, "learning_rate": 8.716729651188595e-05, "loss": 0.35051331520080564, "memory(GiB)": 61.91, "step": 27220, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.461027 }, { "epoch": 1.1664024677605929, "grad_norm": 6.186840057373047, "learning_rate": 8.71627945816656e-05, "loss": 0.4927004337310791, "memory(GiB)": 61.91, "step": 27225, "token_acc": 0.9044585987261147, "train_speed(iter/s)": 1.461032 }, { "epoch": 1.166616683089842, "grad_norm": 3.859182596206665, "learning_rate": 8.715829197820639e-05, "loss": 0.34815685749053954, "memory(GiB)": 61.91, "step": 27230, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.461028 }, { "epoch": 1.166830898419091, "grad_norm": 3.5118043422698975, "learning_rate": 8.715378870158986e-05, "loss": 0.3174623966217041, "memory(GiB)": 61.91, "step": 27235, "token_acc": 0.9367088607594937, "train_speed(iter/s)": 1.461012 }, { "epoch": 1.1670451137483397, "grad_norm": 2.677747964859009, "learning_rate": 8.714928475189759e-05, "loss": 0.4030020236968994, "memory(GiB)": 61.91, "step": 27240, "token_acc": 0.9140625, "train_speed(iter/s)": 1.461013 }, { "epoch": 1.1672593290775888, "grad_norm": 3.561037302017212, "learning_rate": 8.714478012921117e-05, "loss": 0.26595358848571776, "memory(GiB)": 61.91, "step": 27245, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.461011 }, { "epoch": 1.1674735444068378, "grad_norm": 6.076673984527588, "learning_rate": 8.714027483361222e-05, "loss": 0.6540487289428711, "memory(GiB)": 61.91, "step": 27250, "token_acc": 0.8996763754045307, "train_speed(iter/s)": 1.461009 }, { "epoch": 1.1676877597360866, "grad_norm": 2.9037301540374756, "learning_rate": 8.713576886518235e-05, "loss": 0.5325719833374023, "memory(GiB)": 61.91, "step": 27255, "token_acc": 0.887240356083086, "train_speed(iter/s)": 1.461035 }, { "epoch": 1.1679019750653357, "grad_norm": 3.223959445953369, "learning_rate": 8.713126222400319e-05, "loss": 0.41504511833190916, "memory(GiB)": 61.91, "step": 27260, "token_acc": 0.9107692307692308, "train_speed(iter/s)": 1.461044 }, { "epoch": 1.1681161903945847, "grad_norm": 3.3443033695220947, "learning_rate": 8.712675491015638e-05, "loss": 0.19659769535064697, "memory(GiB)": 61.91, "step": 27265, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.461069 }, { "epoch": 1.1683304057238335, "grad_norm": 1.2430284023284912, "learning_rate": 8.712224692372358e-05, "loss": 0.1977284789085388, "memory(GiB)": 61.91, "step": 27270, "token_acc": 0.9461538461538461, "train_speed(iter/s)": 1.46107 }, { "epoch": 1.1685446210530825, "grad_norm": 7.776795387268066, "learning_rate": 8.711773826478646e-05, "loss": 0.5328941822052002, "memory(GiB)": 61.91, "step": 27275, "token_acc": 0.8884297520661157, "train_speed(iter/s)": 1.461084 }, { "epoch": 1.1687588363823316, "grad_norm": 2.1505463123321533, "learning_rate": 8.711322893342668e-05, "loss": 0.4903285980224609, "memory(GiB)": 61.91, "step": 27280, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.461103 }, { "epoch": 1.1689730517115804, "grad_norm": 4.274763107299805, "learning_rate": 8.710871892972595e-05, "loss": 0.5169817924499511, "memory(GiB)": 61.91, "step": 27285, "token_acc": 0.889261744966443, "train_speed(iter/s)": 1.461098 }, { "epoch": 1.1691872670408294, "grad_norm": 4.621191024780273, "learning_rate": 8.710420825376597e-05, "loss": 0.3140977621078491, "memory(GiB)": 61.91, "step": 27290, "token_acc": 0.9373134328358209, "train_speed(iter/s)": 1.461111 }, { "epoch": 1.1694014823700785, "grad_norm": 0.06240272894501686, "learning_rate": 8.709969690562845e-05, "loss": 0.41395859718322753, "memory(GiB)": 61.91, "step": 27295, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.461112 }, { "epoch": 1.1696156976993273, "grad_norm": 3.521989583969116, "learning_rate": 8.709518488539511e-05, "loss": 0.5750338554382324, "memory(GiB)": 61.91, "step": 27300, "token_acc": 0.8727810650887574, "train_speed(iter/s)": 1.461136 }, { "epoch": 1.1698299130285763, "grad_norm": 2.1408612728118896, "learning_rate": 8.709067219314771e-05, "loss": 0.27926547527313234, "memory(GiB)": 61.91, "step": 27305, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.461146 }, { "epoch": 1.1700441283578253, "grad_norm": 2.902564287185669, "learning_rate": 8.708615882896798e-05, "loss": 0.21940712928771972, "memory(GiB)": 61.91, "step": 27310, "token_acc": 0.9490196078431372, "train_speed(iter/s)": 1.461147 }, { "epoch": 1.1702583436870742, "grad_norm": 0.6718227863311768, "learning_rate": 8.708164479293767e-05, "loss": 0.2615920066833496, "memory(GiB)": 61.91, "step": 27315, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.461179 }, { "epoch": 1.1704725590163232, "grad_norm": 6.851076602935791, "learning_rate": 8.707713008513863e-05, "loss": 0.3633223056793213, "memory(GiB)": 61.91, "step": 27320, "token_acc": 0.9325396825396826, "train_speed(iter/s)": 1.461167 }, { "epoch": 1.1706867743455722, "grad_norm": 0.7372506856918335, "learning_rate": 8.707261470565256e-05, "loss": 0.2106724977493286, "memory(GiB)": 61.91, "step": 27325, "token_acc": 0.9513888888888888, "train_speed(iter/s)": 1.461154 }, { "epoch": 1.170900989674821, "grad_norm": 2.4116199016571045, "learning_rate": 8.706809865456131e-05, "loss": 0.18555760383605957, "memory(GiB)": 61.91, "step": 27330, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.46115 }, { "epoch": 1.17111520500407, "grad_norm": 7.784188747406006, "learning_rate": 8.706358193194668e-05, "loss": 0.33608691692352294, "memory(GiB)": 61.91, "step": 27335, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.461146 }, { "epoch": 1.171329420333319, "grad_norm": 0.6339495182037354, "learning_rate": 8.705906453789049e-05, "loss": 0.2623204946517944, "memory(GiB)": 61.91, "step": 27340, "token_acc": 0.9551282051282052, "train_speed(iter/s)": 1.461142 }, { "epoch": 1.171543635662568, "grad_norm": 1.8444576263427734, "learning_rate": 8.705454647247458e-05, "loss": 0.2400134563446045, "memory(GiB)": 61.91, "step": 27345, "token_acc": 0.9386973180076629, "train_speed(iter/s)": 1.461149 }, { "epoch": 1.171757850991817, "grad_norm": 4.573610782623291, "learning_rate": 8.705002773578081e-05, "loss": 0.47996368408203127, "memory(GiB)": 61.91, "step": 27350, "token_acc": 0.9100719424460432, "train_speed(iter/s)": 1.461165 }, { "epoch": 1.171972066321066, "grad_norm": 3.799398183822632, "learning_rate": 8.704550832789103e-05, "loss": 0.5838590621948242, "memory(GiB)": 61.91, "step": 27355, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.461172 }, { "epoch": 1.172186281650315, "grad_norm": 0.3304809331893921, "learning_rate": 8.704098824888711e-05, "loss": 0.22252418994903564, "memory(GiB)": 61.91, "step": 27360, "token_acc": 0.9622641509433962, "train_speed(iter/s)": 1.461184 }, { "epoch": 1.1724004969795638, "grad_norm": 0.31113743782043457, "learning_rate": 8.703646749885095e-05, "loss": 0.2334071636199951, "memory(GiB)": 61.91, "step": 27365, "token_acc": 0.9513677811550152, "train_speed(iter/s)": 1.461182 }, { "epoch": 1.1726147123088129, "grad_norm": 1.784250259399414, "learning_rate": 8.703194607786442e-05, "loss": 0.130288565158844, "memory(GiB)": 61.91, "step": 27370, "token_acc": 0.9777070063694268, "train_speed(iter/s)": 1.461192 }, { "epoch": 1.172828927638062, "grad_norm": 2.2480852603912354, "learning_rate": 8.702742398600947e-05, "loss": 0.4066332817077637, "memory(GiB)": 61.91, "step": 27375, "token_acc": 0.896875, "train_speed(iter/s)": 1.461199 }, { "epoch": 1.1730431429673107, "grad_norm": 3.3074028491973877, "learning_rate": 8.702290122336798e-05, "loss": 0.40485610961914065, "memory(GiB)": 61.91, "step": 27380, "token_acc": 0.925414364640884, "train_speed(iter/s)": 1.461182 }, { "epoch": 1.1732573582965597, "grad_norm": 4.337277889251709, "learning_rate": 8.701837779002192e-05, "loss": 0.3386176586151123, "memory(GiB)": 61.91, "step": 27385, "token_acc": 0.9255583126550868, "train_speed(iter/s)": 1.461192 }, { "epoch": 1.1734715736258088, "grad_norm": 1.7082427740097046, "learning_rate": 8.701385368605322e-05, "loss": 0.548438024520874, "memory(GiB)": 61.91, "step": 27390, "token_acc": 0.888135593220339, "train_speed(iter/s)": 1.461221 }, { "epoch": 1.1736857889550576, "grad_norm": 3.663837432861328, "learning_rate": 8.700932891154383e-05, "loss": 0.26578531265258787, "memory(GiB)": 61.91, "step": 27395, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.461224 }, { "epoch": 1.1739000042843066, "grad_norm": 9.805458068847656, "learning_rate": 8.700480346657574e-05, "loss": 0.6319806098937988, "memory(GiB)": 61.91, "step": 27400, "token_acc": 0.8783783783783784, "train_speed(iter/s)": 1.461216 }, { "epoch": 1.1741142196135557, "grad_norm": 2.616286516189575, "learning_rate": 8.70002773512309e-05, "loss": 0.4588280200958252, "memory(GiB)": 61.91, "step": 27405, "token_acc": 0.9015748031496063, "train_speed(iter/s)": 1.461212 }, { "epoch": 1.1743284349428045, "grad_norm": 5.402585983276367, "learning_rate": 8.699575056559134e-05, "loss": 0.38677082061767576, "memory(GiB)": 61.91, "step": 27410, "token_acc": 0.9221789883268483, "train_speed(iter/s)": 1.46123 }, { "epoch": 1.1745426502720535, "grad_norm": 5.67852258682251, "learning_rate": 8.699122310973907e-05, "loss": 0.6302347183227539, "memory(GiB)": 61.91, "step": 27415, "token_acc": 0.8524590163934426, "train_speed(iter/s)": 1.461251 }, { "epoch": 1.1747568656013025, "grad_norm": 3.1055490970611572, "learning_rate": 8.698669498375606e-05, "loss": 0.4468526840209961, "memory(GiB)": 61.91, "step": 27420, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.461257 }, { "epoch": 1.1749710809305514, "grad_norm": 0.04025399684906006, "learning_rate": 8.698216618772439e-05, "loss": 0.22625575065612794, "memory(GiB)": 61.91, "step": 27425, "token_acc": 0.9584569732937686, "train_speed(iter/s)": 1.461263 }, { "epoch": 1.1751852962598004, "grad_norm": 1.5601253509521484, "learning_rate": 8.69776367217261e-05, "loss": 0.28963565826416016, "memory(GiB)": 61.91, "step": 27430, "token_acc": 0.9536231884057971, "train_speed(iter/s)": 1.461306 }, { "epoch": 1.1753995115890494, "grad_norm": 3.0854616165161133, "learning_rate": 8.697310658584321e-05, "loss": 0.5477337837219238, "memory(GiB)": 61.91, "step": 27435, "token_acc": 0.8821752265861027, "train_speed(iter/s)": 1.461312 }, { "epoch": 1.1756137269182982, "grad_norm": 3.910092830657959, "learning_rate": 8.696857578015783e-05, "loss": 0.517252254486084, "memory(GiB)": 61.91, "step": 27440, "token_acc": 0.9185393258426966, "train_speed(iter/s)": 1.461337 }, { "epoch": 1.1758279422475473, "grad_norm": 3.5918619632720947, "learning_rate": 8.696404430475201e-05, "loss": 0.4231856822967529, "memory(GiB)": 61.91, "step": 27445, "token_acc": 0.9127725856697819, "train_speed(iter/s)": 1.461339 }, { "epoch": 1.1760421575767963, "grad_norm": 0.3004184365272522, "learning_rate": 8.695951215970785e-05, "loss": 0.9701239585876464, "memory(GiB)": 61.91, "step": 27450, "token_acc": 0.8612099644128114, "train_speed(iter/s)": 1.461324 }, { "epoch": 1.1762563729060451, "grad_norm": 1.1849474906921387, "learning_rate": 8.695497934510747e-05, "loss": 0.5460041046142579, "memory(GiB)": 61.91, "step": 27455, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.461354 }, { "epoch": 1.1764705882352942, "grad_norm": 0.43875399231910706, "learning_rate": 8.695044586103296e-05, "loss": 0.20521197319030762, "memory(GiB)": 61.91, "step": 27460, "token_acc": 0.9530201342281879, "train_speed(iter/s)": 1.461385 }, { "epoch": 1.1766848035645432, "grad_norm": 2.842956066131592, "learning_rate": 8.694591170756647e-05, "loss": 0.608389139175415, "memory(GiB)": 61.91, "step": 27465, "token_acc": 0.8745098039215686, "train_speed(iter/s)": 1.461379 }, { "epoch": 1.176899018893792, "grad_norm": 2.967393636703491, "learning_rate": 8.694137688479012e-05, "loss": 0.3976465940475464, "memory(GiB)": 61.91, "step": 27470, "token_acc": 0.9075907590759076, "train_speed(iter/s)": 1.461426 }, { "epoch": 1.177113234223041, "grad_norm": 0.30286338925361633, "learning_rate": 8.693684139278609e-05, "loss": 0.47690300941467284, "memory(GiB)": 61.91, "step": 27475, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.461436 }, { "epoch": 1.17732744955229, "grad_norm": 2.456118583679199, "learning_rate": 8.693230523163652e-05, "loss": 0.3592314004898071, "memory(GiB)": 61.91, "step": 27480, "token_acc": 0.8988326848249028, "train_speed(iter/s)": 1.461447 }, { "epoch": 1.1775416648815389, "grad_norm": 5.993458271026611, "learning_rate": 8.692776840142359e-05, "loss": 0.5350056648254394, "memory(GiB)": 61.91, "step": 27485, "token_acc": 0.891566265060241, "train_speed(iter/s)": 1.461423 }, { "epoch": 1.177755880210788, "grad_norm": 4.8261284828186035, "learning_rate": 8.692323090222949e-05, "loss": 0.3401159763336182, "memory(GiB)": 61.91, "step": 27490, "token_acc": 0.9127906976744186, "train_speed(iter/s)": 1.461415 }, { "epoch": 1.177970095540037, "grad_norm": 0.8125938773155212, "learning_rate": 8.691869273413645e-05, "loss": 0.32099523544311526, "memory(GiB)": 61.91, "step": 27495, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.461417 }, { "epoch": 1.1781843108692858, "grad_norm": 3.7261292934417725, "learning_rate": 8.691415389722663e-05, "loss": 0.3475987672805786, "memory(GiB)": 61.91, "step": 27500, "token_acc": 0.9098039215686274, "train_speed(iter/s)": 1.46142 }, { "epoch": 1.1781843108692858, "eval_loss": 2.6660988330841064, "eval_runtime": 13.8964, "eval_samples_per_second": 7.196, "eval_steps_per_second": 7.196, "eval_token_acc": 0.4216691068814056, "step": 27500 }, { "epoch": 1.1783985261985348, "grad_norm": 3.1181623935699463, "learning_rate": 8.690961439158229e-05, "loss": 0.3369602680206299, "memory(GiB)": 61.91, "step": 27505, "token_acc": 0.5661764705882353, "train_speed(iter/s)": 1.4603 }, { "epoch": 1.1786127415277838, "grad_norm": 0.273754358291626, "learning_rate": 8.690507421728566e-05, "loss": 0.1816296935081482, "memory(GiB)": 61.91, "step": 27510, "token_acc": 0.9610894941634242, "train_speed(iter/s)": 1.460318 }, { "epoch": 1.1788269568570326, "grad_norm": 2.5263724327087402, "learning_rate": 8.690053337441901e-05, "loss": 0.4082071304321289, "memory(GiB)": 61.91, "step": 27515, "token_acc": 0.9240506329113924, "train_speed(iter/s)": 1.46035 }, { "epoch": 1.1790411721862817, "grad_norm": 2.617877244949341, "learning_rate": 8.689599186306457e-05, "loss": 0.4701240539550781, "memory(GiB)": 61.91, "step": 27520, "token_acc": 0.8910505836575876, "train_speed(iter/s)": 1.460407 }, { "epoch": 1.1792553875155307, "grad_norm": 8.040634155273438, "learning_rate": 8.689144968330462e-05, "loss": 0.3190941095352173, "memory(GiB)": 61.91, "step": 27525, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.460408 }, { "epoch": 1.1794696028447795, "grad_norm": 2.0461626052856445, "learning_rate": 8.688690683522147e-05, "loss": 0.2766294479370117, "memory(GiB)": 61.91, "step": 27530, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.460404 }, { "epoch": 1.1796838181740286, "grad_norm": 1.9271962642669678, "learning_rate": 8.688236331889739e-05, "loss": 0.29475812911987304, "memory(GiB)": 61.91, "step": 27535, "token_acc": 0.9272151898734177, "train_speed(iter/s)": 1.460411 }, { "epoch": 1.1798980335032776, "grad_norm": 4.869226455688477, "learning_rate": 8.687781913441469e-05, "loss": 0.2279655933380127, "memory(GiB)": 61.91, "step": 27540, "token_acc": 0.9501915708812261, "train_speed(iter/s)": 1.460401 }, { "epoch": 1.1801122488325264, "grad_norm": 2.486436367034912, "learning_rate": 8.687327428185573e-05, "loss": 0.21123323440551758, "memory(GiB)": 61.91, "step": 27545, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.460433 }, { "epoch": 1.1803264641617754, "grad_norm": 3.5693612098693848, "learning_rate": 8.686872876130279e-05, "loss": 0.30232465267181396, "memory(GiB)": 61.91, "step": 27550, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.460448 }, { "epoch": 1.1805406794910245, "grad_norm": 3.338470458984375, "learning_rate": 8.686418257283824e-05, "loss": 0.4710539817810059, "memory(GiB)": 61.91, "step": 27555, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.460459 }, { "epoch": 1.1807548948202733, "grad_norm": 12.185235023498535, "learning_rate": 8.685963571654445e-05, "loss": 0.5314490318298339, "memory(GiB)": 61.91, "step": 27560, "token_acc": 0.9066147859922179, "train_speed(iter/s)": 1.46047 }, { "epoch": 1.1809691101495223, "grad_norm": 1.880761981010437, "learning_rate": 8.685508819250379e-05, "loss": 0.6484626293182373, "memory(GiB)": 61.91, "step": 27565, "token_acc": 0.8773234200743495, "train_speed(iter/s)": 1.460512 }, { "epoch": 1.1811833254787714, "grad_norm": 3.661987066268921, "learning_rate": 8.685054000079863e-05, "loss": 0.3060471534729004, "memory(GiB)": 61.91, "step": 27570, "token_acc": 0.9307958477508651, "train_speed(iter/s)": 1.460502 }, { "epoch": 1.1813975408080202, "grad_norm": 3.506565809249878, "learning_rate": 8.684599114151137e-05, "loss": 0.34361732006073, "memory(GiB)": 61.91, "step": 27575, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.460497 }, { "epoch": 1.1816117561372692, "grad_norm": 5.588808536529541, "learning_rate": 8.684144161472441e-05, "loss": 0.43086843490600585, "memory(GiB)": 61.91, "step": 27580, "token_acc": 0.9125, "train_speed(iter/s)": 1.460483 }, { "epoch": 1.1818259714665182, "grad_norm": 3.000349998474121, "learning_rate": 8.683689142052018e-05, "loss": 0.538430643081665, "memory(GiB)": 61.91, "step": 27585, "token_acc": 0.8848920863309353, "train_speed(iter/s)": 1.460492 }, { "epoch": 1.182040186795767, "grad_norm": 2.432335376739502, "learning_rate": 8.68323405589811e-05, "loss": 0.32781119346618653, "memory(GiB)": 61.91, "step": 27590, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.460492 }, { "epoch": 1.182254402125016, "grad_norm": 6.3943705558776855, "learning_rate": 8.682778903018962e-05, "loss": 0.4973437786102295, "memory(GiB)": 61.91, "step": 27595, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.460501 }, { "epoch": 1.1824686174542651, "grad_norm": 2.1886560916900635, "learning_rate": 8.68232368342282e-05, "loss": 0.5877338886260987, "memory(GiB)": 61.91, "step": 27600, "token_acc": 0.8711656441717791, "train_speed(iter/s)": 1.46049 }, { "epoch": 1.182682832783514, "grad_norm": 3.536951780319214, "learning_rate": 8.68186839711793e-05, "loss": 0.3492010831832886, "memory(GiB)": 61.91, "step": 27605, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.460498 }, { "epoch": 1.182897048112763, "grad_norm": 0.06246297433972359, "learning_rate": 8.68141304411254e-05, "loss": 0.1567667007446289, "memory(GiB)": 61.91, "step": 27610, "token_acc": 0.9681978798586572, "train_speed(iter/s)": 1.460498 }, { "epoch": 1.183111263442012, "grad_norm": 4.08609676361084, "learning_rate": 8.680957624414901e-05, "loss": 0.5827283382415771, "memory(GiB)": 61.91, "step": 27615, "token_acc": 0.8745519713261649, "train_speed(iter/s)": 1.460523 }, { "epoch": 1.1833254787712608, "grad_norm": 2.5387463569641113, "learning_rate": 8.68050213803326e-05, "loss": 0.2990507841110229, "memory(GiB)": 61.91, "step": 27620, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.460545 }, { "epoch": 1.1835396941005099, "grad_norm": 5.2335968017578125, "learning_rate": 8.68004658497587e-05, "loss": 0.4036684989929199, "memory(GiB)": 61.91, "step": 27625, "token_acc": 0.9169675090252708, "train_speed(iter/s)": 1.46054 }, { "epoch": 1.1837539094297589, "grad_norm": 4.865113258361816, "learning_rate": 8.679590965250984e-05, "loss": 0.4664731502532959, "memory(GiB)": 61.91, "step": 27630, "token_acc": 0.9043478260869565, "train_speed(iter/s)": 1.460517 }, { "epoch": 1.1839681247590077, "grad_norm": 4.395020484924316, "learning_rate": 8.679135278866855e-05, "loss": 0.36223368644714354, "memory(GiB)": 61.91, "step": 27635, "token_acc": 0.9192546583850931, "train_speed(iter/s)": 1.460508 }, { "epoch": 1.1841823400882567, "grad_norm": 2.5459609031677246, "learning_rate": 8.678679525831742e-05, "loss": 0.5691050052642822, "memory(GiB)": 61.91, "step": 27640, "token_acc": 0.9120521172638436, "train_speed(iter/s)": 1.460543 }, { "epoch": 1.1843965554175058, "grad_norm": 1.297463297843933, "learning_rate": 8.678223706153895e-05, "loss": 0.3210099220275879, "memory(GiB)": 61.91, "step": 27645, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.460546 }, { "epoch": 1.1846107707467546, "grad_norm": 3.25260066986084, "learning_rate": 8.677767819841577e-05, "loss": 0.45520620346069335, "memory(GiB)": 61.91, "step": 27650, "token_acc": 0.8993055555555556, "train_speed(iter/s)": 1.46057 }, { "epoch": 1.1848249860760036, "grad_norm": 1.494598150253296, "learning_rate": 8.677311866903044e-05, "loss": 0.36129810810089114, "memory(GiB)": 61.91, "step": 27655, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.46058 }, { "epoch": 1.1850392014052527, "grad_norm": 3.1212198734283447, "learning_rate": 8.676855847346559e-05, "loss": 0.38303992748260496, "memory(GiB)": 61.91, "step": 27660, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.460575 }, { "epoch": 1.1852534167345015, "grad_norm": 4.345556735992432, "learning_rate": 8.676399761180379e-05, "loss": 0.5732937812805176, "memory(GiB)": 61.91, "step": 27665, "token_acc": 0.8882521489971347, "train_speed(iter/s)": 1.460549 }, { "epoch": 1.1854676320637505, "grad_norm": 2.435948133468628, "learning_rate": 8.67594360841277e-05, "loss": 0.20181746482849122, "memory(GiB)": 61.91, "step": 27670, "token_acc": 0.9591078066914498, "train_speed(iter/s)": 1.460554 }, { "epoch": 1.1856818473929995, "grad_norm": 5.0014448165893555, "learning_rate": 8.675487389051993e-05, "loss": 0.3425534963607788, "memory(GiB)": 61.91, "step": 27675, "token_acc": 0.9358490566037736, "train_speed(iter/s)": 1.46054 }, { "epoch": 1.1858960627222483, "grad_norm": 2.6436309814453125, "learning_rate": 8.675031103106315e-05, "loss": 0.4479849815368652, "memory(GiB)": 61.91, "step": 27680, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.460561 }, { "epoch": 1.1861102780514974, "grad_norm": 1.7217724323272705, "learning_rate": 8.674574750584002e-05, "loss": 0.47141528129577637, "memory(GiB)": 61.91, "step": 27685, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.460581 }, { "epoch": 1.1863244933807464, "grad_norm": 5.502981662750244, "learning_rate": 8.674118331493317e-05, "loss": 0.47626380920410155, "memory(GiB)": 61.91, "step": 27690, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.460601 }, { "epoch": 1.1865387087099952, "grad_norm": 4.990617275238037, "learning_rate": 8.673661845842536e-05, "loss": 0.38213050365448, "memory(GiB)": 61.91, "step": 27695, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.46062 }, { "epoch": 1.1867529240392443, "grad_norm": 0.08761344105005264, "learning_rate": 8.673205293639921e-05, "loss": 0.2496865510940552, "memory(GiB)": 61.91, "step": 27700, "token_acc": 0.9482071713147411, "train_speed(iter/s)": 1.460618 }, { "epoch": 1.1869671393684933, "grad_norm": 1.7175565958023071, "learning_rate": 8.672748674893749e-05, "loss": 0.15277130603790284, "memory(GiB)": 61.91, "step": 27705, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.460618 }, { "epoch": 1.187181354697742, "grad_norm": 4.048750400543213, "learning_rate": 8.672291989612287e-05, "loss": 0.3579143524169922, "memory(GiB)": 61.91, "step": 27710, "token_acc": 0.9288135593220339, "train_speed(iter/s)": 1.460619 }, { "epoch": 1.1873955700269911, "grad_norm": 5.862030029296875, "learning_rate": 8.671835237803812e-05, "loss": 0.4447788238525391, "memory(GiB)": 61.91, "step": 27715, "token_acc": 0.910377358490566, "train_speed(iter/s)": 1.460647 }, { "epoch": 1.1876097853562402, "grad_norm": 3.9289848804473877, "learning_rate": 8.671378419476596e-05, "loss": 0.4060066699981689, "memory(GiB)": 61.91, "step": 27720, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.460642 }, { "epoch": 1.187824000685489, "grad_norm": 1.8064296245574951, "learning_rate": 8.670921534638918e-05, "loss": 0.3035606384277344, "memory(GiB)": 61.91, "step": 27725, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.460632 }, { "epoch": 1.188038216014738, "grad_norm": 3.9078991413116455, "learning_rate": 8.670464583299052e-05, "loss": 0.5067134857177734, "memory(GiB)": 61.91, "step": 27730, "token_acc": 0.8932038834951457, "train_speed(iter/s)": 1.46067 }, { "epoch": 1.188252431343987, "grad_norm": 3.5502026081085205, "learning_rate": 8.670007565465277e-05, "loss": 0.42167024612426757, "memory(GiB)": 61.91, "step": 27735, "token_acc": 0.9006849315068494, "train_speed(iter/s)": 1.460662 }, { "epoch": 1.1884666466732359, "grad_norm": 7.078333377838135, "learning_rate": 8.669550481145872e-05, "loss": 0.4318528175354004, "memory(GiB)": 61.91, "step": 27740, "token_acc": 0.8983739837398373, "train_speed(iter/s)": 1.460676 }, { "epoch": 1.188680862002485, "grad_norm": 5.993659019470215, "learning_rate": 8.669093330349118e-05, "loss": 0.6715028762817383, "memory(GiB)": 61.91, "step": 27745, "token_acc": 0.8685121107266436, "train_speed(iter/s)": 1.460669 }, { "epoch": 1.188895077331734, "grad_norm": 3.2523434162139893, "learning_rate": 8.668636113083296e-05, "loss": 0.2733804941177368, "memory(GiB)": 61.91, "step": 27750, "token_acc": 0.9307958477508651, "train_speed(iter/s)": 1.460659 }, { "epoch": 1.1891092926609828, "grad_norm": 2.3306844234466553, "learning_rate": 8.66817882935669e-05, "loss": 0.591513729095459, "memory(GiB)": 61.91, "step": 27755, "token_acc": 0.8844765342960289, "train_speed(iter/s)": 1.460666 }, { "epoch": 1.1893235079902318, "grad_norm": 2.3767032623291016, "learning_rate": 8.667721479177583e-05, "loss": 0.45267701148986816, "memory(GiB)": 61.91, "step": 27760, "token_acc": 0.8936781609195402, "train_speed(iter/s)": 1.460658 }, { "epoch": 1.1895377233194808, "grad_norm": 4.592023849487305, "learning_rate": 8.66726406255426e-05, "loss": 0.37990307807922363, "memory(GiB)": 61.91, "step": 27765, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.460694 }, { "epoch": 1.1897519386487296, "grad_norm": 3.88527250289917, "learning_rate": 8.66680657949501e-05, "loss": 0.4046185493469238, "memory(GiB)": 61.91, "step": 27770, "token_acc": 0.9149659863945578, "train_speed(iter/s)": 1.460712 }, { "epoch": 1.1899661539779787, "grad_norm": 4.222915172576904, "learning_rate": 8.666349030008118e-05, "loss": 0.31887116432189944, "memory(GiB)": 61.91, "step": 27775, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.460734 }, { "epoch": 1.1901803693072277, "grad_norm": 4.50861120223999, "learning_rate": 8.665891414101874e-05, "loss": 0.45573787689208983, "memory(GiB)": 61.91, "step": 27780, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.460728 }, { "epoch": 1.1903945846364765, "grad_norm": 1.6139092445373535, "learning_rate": 8.665433731784568e-05, "loss": 0.27750492095947266, "memory(GiB)": 61.91, "step": 27785, "token_acc": 0.9375, "train_speed(iter/s)": 1.460736 }, { "epoch": 1.1906087999657256, "grad_norm": 5.027646064758301, "learning_rate": 8.664975983064491e-05, "loss": 0.24301564693450928, "memory(GiB)": 61.91, "step": 27790, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.460726 }, { "epoch": 1.1908230152949746, "grad_norm": 2.637526273727417, "learning_rate": 8.664518167949937e-05, "loss": 0.27830004692077637, "memory(GiB)": 61.91, "step": 27795, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.460716 }, { "epoch": 1.1910372306242234, "grad_norm": 2.433516502380371, "learning_rate": 8.6640602864492e-05, "loss": 0.47333393096923826, "memory(GiB)": 61.91, "step": 27800, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.460721 }, { "epoch": 1.1912514459534724, "grad_norm": 1.6175190210342407, "learning_rate": 8.663602338570571e-05, "loss": 0.5544665336608887, "memory(GiB)": 61.91, "step": 27805, "token_acc": 0.8835616438356164, "train_speed(iter/s)": 1.460719 }, { "epoch": 1.1914656612827215, "grad_norm": 2.9511265754699707, "learning_rate": 8.663144324322351e-05, "loss": 0.253474760055542, "memory(GiB)": 61.91, "step": 27810, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.460714 }, { "epoch": 1.1916798766119703, "grad_norm": 6.227827072143555, "learning_rate": 8.662686243712834e-05, "loss": 0.3957479476928711, "memory(GiB)": 61.91, "step": 27815, "token_acc": 0.9133858267716536, "train_speed(iter/s)": 1.460711 }, { "epoch": 1.1918940919412193, "grad_norm": 0.07227258384227753, "learning_rate": 8.662228096750321e-05, "loss": 0.25143747329711913, "memory(GiB)": 61.91, "step": 27820, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.460713 }, { "epoch": 1.1921083072704683, "grad_norm": 5.512509346008301, "learning_rate": 8.661769883443109e-05, "loss": 0.5983704566955567, "memory(GiB)": 61.91, "step": 27825, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.460721 }, { "epoch": 1.1923225225997172, "grad_norm": 4.126183986663818, "learning_rate": 8.661311603799503e-05, "loss": 0.22532787322998046, "memory(GiB)": 61.91, "step": 27830, "token_acc": 0.9471544715447154, "train_speed(iter/s)": 1.460713 }, { "epoch": 1.1925367379289662, "grad_norm": 2.8886923789978027, "learning_rate": 8.660853257827801e-05, "loss": 0.3726520538330078, "memory(GiB)": 61.91, "step": 27835, "token_acc": 0.921875, "train_speed(iter/s)": 1.460732 }, { "epoch": 1.1927509532582152, "grad_norm": 7.896599769592285, "learning_rate": 8.660394845536309e-05, "loss": 0.2909923791885376, "memory(GiB)": 61.91, "step": 27840, "token_acc": 0.9241071428571429, "train_speed(iter/s)": 1.460736 }, { "epoch": 1.192965168587464, "grad_norm": 6.7992024421691895, "learning_rate": 8.659936366933328e-05, "loss": 0.8210404396057129, "memory(GiB)": 61.91, "step": 27845, "token_acc": 0.8227848101265823, "train_speed(iter/s)": 1.460731 }, { "epoch": 1.193179383916713, "grad_norm": 1.9434590339660645, "learning_rate": 8.659477822027169e-05, "loss": 0.28756487369537354, "memory(GiB)": 61.91, "step": 27850, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.460735 }, { "epoch": 1.1933935992459621, "grad_norm": 3.627007007598877, "learning_rate": 8.659019210826136e-05, "loss": 0.2786900997161865, "memory(GiB)": 61.91, "step": 27855, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.460737 }, { "epoch": 1.193607814575211, "grad_norm": 2.218353509902954, "learning_rate": 8.658560533338537e-05, "loss": 0.26366727352142333, "memory(GiB)": 61.91, "step": 27860, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.460769 }, { "epoch": 1.19382202990446, "grad_norm": 2.653010368347168, "learning_rate": 8.658101789572683e-05, "loss": 0.1858613133430481, "memory(GiB)": 61.91, "step": 27865, "token_acc": 0.9525547445255474, "train_speed(iter/s)": 1.460763 }, { "epoch": 1.194036245233709, "grad_norm": 3.2128870487213135, "learning_rate": 8.657642979536882e-05, "loss": 0.29883289337158203, "memory(GiB)": 61.91, "step": 27870, "token_acc": 0.9382716049382716, "train_speed(iter/s)": 1.460767 }, { "epoch": 1.1942504605629578, "grad_norm": 1.9711037874221802, "learning_rate": 8.657184103239448e-05, "loss": 0.47423410415649414, "memory(GiB)": 61.91, "step": 27875, "token_acc": 0.8978978978978979, "train_speed(iter/s)": 1.460758 }, { "epoch": 1.1944646758922068, "grad_norm": 4.6100053787231445, "learning_rate": 8.656725160688694e-05, "loss": 0.29145112037658694, "memory(GiB)": 61.91, "step": 27880, "token_acc": 0.9438596491228071, "train_speed(iter/s)": 1.460776 }, { "epoch": 1.1946788912214559, "grad_norm": 3.0555756092071533, "learning_rate": 8.656266151892932e-05, "loss": 0.47237286567687986, "memory(GiB)": 61.91, "step": 27885, "token_acc": 0.896969696969697, "train_speed(iter/s)": 1.460774 }, { "epoch": 1.1948931065507047, "grad_norm": 2.5839154720306396, "learning_rate": 8.655807076860481e-05, "loss": 0.33853778839111326, "memory(GiB)": 61.91, "step": 27890, "token_acc": 0.9258064516129032, "train_speed(iter/s)": 1.460772 }, { "epoch": 1.1951073218799537, "grad_norm": 4.189910411834717, "learning_rate": 8.655347935599653e-05, "loss": 0.43024392127990724, "memory(GiB)": 61.91, "step": 27895, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.46077 }, { "epoch": 1.1953215372092028, "grad_norm": 3.1846423149108887, "learning_rate": 8.654888728118769e-05, "loss": 0.21485917568206786, "memory(GiB)": 61.91, "step": 27900, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.460774 }, { "epoch": 1.1955357525384516, "grad_norm": 3.7577319145202637, "learning_rate": 8.654429454426148e-05, "loss": 0.6605565547943115, "memory(GiB)": 61.91, "step": 27905, "token_acc": 0.8533834586466166, "train_speed(iter/s)": 1.460769 }, { "epoch": 1.1957499678677006, "grad_norm": 1.9759801626205444, "learning_rate": 8.653970114530107e-05, "loss": 0.5945672035217285, "memory(GiB)": 61.91, "step": 27910, "token_acc": 0.8831168831168831, "train_speed(iter/s)": 1.460778 }, { "epoch": 1.1959641831969496, "grad_norm": 2.1129112243652344, "learning_rate": 8.65351070843897e-05, "loss": 0.30331013202667234, "memory(GiB)": 61.91, "step": 27915, "token_acc": 0.9397590361445783, "train_speed(iter/s)": 1.460777 }, { "epoch": 1.1961783985261984, "grad_norm": 2.9007315635681152, "learning_rate": 8.653051236161062e-05, "loss": 0.25067079067230225, "memory(GiB)": 61.91, "step": 27920, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.460769 }, { "epoch": 1.1963926138554475, "grad_norm": 3.4782657623291016, "learning_rate": 8.652591697704702e-05, "loss": 0.31124417781829833, "memory(GiB)": 61.91, "step": 27925, "token_acc": 0.9392857142857143, "train_speed(iter/s)": 1.460776 }, { "epoch": 1.1966068291846965, "grad_norm": 2.081183910369873, "learning_rate": 8.652132093078217e-05, "loss": 0.4089543342590332, "memory(GiB)": 61.91, "step": 27930, "token_acc": 0.9212328767123288, "train_speed(iter/s)": 1.460772 }, { "epoch": 1.1968210445139453, "grad_norm": 2.5384631156921387, "learning_rate": 8.651672422289934e-05, "loss": 0.30466084480285643, "memory(GiB)": 61.91, "step": 27935, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.460764 }, { "epoch": 1.1970352598431944, "grad_norm": 2.5552260875701904, "learning_rate": 8.65121268534818e-05, "loss": 0.43434958457946776, "memory(GiB)": 61.91, "step": 27940, "token_acc": 0.9065743944636678, "train_speed(iter/s)": 1.460761 }, { "epoch": 1.1972494751724434, "grad_norm": 5.045369625091553, "learning_rate": 8.650752882261282e-05, "loss": 0.40030813217163086, "memory(GiB)": 61.91, "step": 27945, "token_acc": 0.9109311740890689, "train_speed(iter/s)": 1.460753 }, { "epoch": 1.1974636905016922, "grad_norm": 1.5198307037353516, "learning_rate": 8.65029301303757e-05, "loss": 0.649674367904663, "memory(GiB)": 61.91, "step": 27950, "token_acc": 0.8700361010830325, "train_speed(iter/s)": 1.460748 }, { "epoch": 1.1976779058309412, "grad_norm": 3.5921056270599365, "learning_rate": 8.649833077685377e-05, "loss": 0.4389814376831055, "memory(GiB)": 61.91, "step": 27955, "token_acc": 0.9183006535947712, "train_speed(iter/s)": 1.460748 }, { "epoch": 1.1978921211601903, "grad_norm": 9.163003921508789, "learning_rate": 8.649373076213035e-05, "loss": 0.43363800048828127, "memory(GiB)": 61.91, "step": 27960, "token_acc": 0.90234375, "train_speed(iter/s)": 1.460765 }, { "epoch": 1.198106336489439, "grad_norm": 2.96183443069458, "learning_rate": 8.648913008628874e-05, "loss": 0.3425696134567261, "memory(GiB)": 61.91, "step": 27965, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.460798 }, { "epoch": 1.1983205518186881, "grad_norm": 11.292884826660156, "learning_rate": 8.648452874941232e-05, "loss": 0.49149255752563475, "memory(GiB)": 61.91, "step": 27970, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.460831 }, { "epoch": 1.1985347671479372, "grad_norm": 3.9998838901519775, "learning_rate": 8.647992675158443e-05, "loss": 0.25110797882080077, "memory(GiB)": 61.91, "step": 27975, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.460836 }, { "epoch": 1.198748982477186, "grad_norm": 3.9017081260681152, "learning_rate": 8.647532409288846e-05, "loss": 0.23569326400756835, "memory(GiB)": 61.91, "step": 27980, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.460844 }, { "epoch": 1.198963197806435, "grad_norm": 3.8430380821228027, "learning_rate": 8.647072077340776e-05, "loss": 0.2965246677398682, "memory(GiB)": 61.91, "step": 27985, "token_acc": 0.9369085173501577, "train_speed(iter/s)": 1.460849 }, { "epoch": 1.199177413135684, "grad_norm": 2.336280584335327, "learning_rate": 8.646611679322576e-05, "loss": 0.2888176441192627, "memory(GiB)": 61.91, "step": 27990, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.460856 }, { "epoch": 1.1993916284649329, "grad_norm": 3.4651288986206055, "learning_rate": 8.646151215242584e-05, "loss": 0.47187957763671873, "memory(GiB)": 61.91, "step": 27995, "token_acc": 0.89, "train_speed(iter/s)": 1.460877 }, { "epoch": 1.199605843794182, "grad_norm": 3.152578115463257, "learning_rate": 8.645690685109143e-05, "loss": 0.5831411361694336, "memory(GiB)": 61.91, "step": 28000, "token_acc": 0.8769716088328076, "train_speed(iter/s)": 1.460856 }, { "epoch": 1.199605843794182, "eval_loss": 2.5096592903137207, "eval_runtime": 13.683, "eval_samples_per_second": 7.308, "eval_steps_per_second": 7.308, "eval_token_acc": 0.45278450363196127, "step": 28000 }, { "epoch": 1.199820059123431, "grad_norm": 4.018642902374268, "learning_rate": 8.645230088930593e-05, "loss": 0.23248164653778075, "memory(GiB)": 61.91, "step": 28005, "token_acc": 0.5626767200754006, "train_speed(iter/s)": 1.459745 }, { "epoch": 1.2000342744526797, "grad_norm": 5.524142742156982, "learning_rate": 8.644769426715282e-05, "loss": 0.38020977973937986, "memory(GiB)": 61.91, "step": 28010, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.459755 }, { "epoch": 1.2002484897819288, "grad_norm": 2.946371555328369, "learning_rate": 8.644308698471553e-05, "loss": 0.3227513313293457, "memory(GiB)": 61.91, "step": 28015, "token_acc": 0.933933933933934, "train_speed(iter/s)": 1.459747 }, { "epoch": 1.2004627051111778, "grad_norm": 3.1455514430999756, "learning_rate": 8.643847904207755e-05, "loss": 0.3625441074371338, "memory(GiB)": 61.91, "step": 28020, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.459781 }, { "epoch": 1.2006769204404266, "grad_norm": 3.3631556034088135, "learning_rate": 8.643387043932232e-05, "loss": 0.3023067474365234, "memory(GiB)": 61.91, "step": 28025, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.459772 }, { "epoch": 1.2008911357696757, "grad_norm": 3.2706353664398193, "learning_rate": 8.642926117653336e-05, "loss": 0.19410618543624877, "memory(GiB)": 61.91, "step": 28030, "token_acc": 0.9597701149425287, "train_speed(iter/s)": 1.459763 }, { "epoch": 1.2011053510989247, "grad_norm": 6.316496849060059, "learning_rate": 8.642465125379415e-05, "loss": 0.47822070121765137, "memory(GiB)": 61.91, "step": 28035, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.459757 }, { "epoch": 1.2013195664281735, "grad_norm": 2.6856529712677, "learning_rate": 8.642004067118824e-05, "loss": 0.28473172187805174, "memory(GiB)": 61.91, "step": 28040, "token_acc": 0.9409722222222222, "train_speed(iter/s)": 1.459759 }, { "epoch": 1.2015337817574225, "grad_norm": 2.1148924827575684, "learning_rate": 8.64154294287991e-05, "loss": 0.3436261177062988, "memory(GiB)": 61.91, "step": 28045, "token_acc": 0.930635838150289, "train_speed(iter/s)": 1.459764 }, { "epoch": 1.2017479970866716, "grad_norm": 2.5761327743530273, "learning_rate": 8.64108175267103e-05, "loss": 0.445538330078125, "memory(GiB)": 61.91, "step": 28050, "token_acc": 0.8955823293172691, "train_speed(iter/s)": 1.459756 }, { "epoch": 1.2019622124159204, "grad_norm": 1.4132699966430664, "learning_rate": 8.64062049650054e-05, "loss": 0.11094833612442016, "memory(GiB)": 61.91, "step": 28055, "token_acc": 0.9711538461538461, "train_speed(iter/s)": 1.459757 }, { "epoch": 1.2021764277451694, "grad_norm": 2.2885355949401855, "learning_rate": 8.640159174376793e-05, "loss": 0.619700574874878, "memory(GiB)": 61.91, "step": 28060, "token_acc": 0.8593155893536122, "train_speed(iter/s)": 1.459765 }, { "epoch": 1.2023906430744185, "grad_norm": 0.5354423522949219, "learning_rate": 8.63969778630815e-05, "loss": 0.24304051399230958, "memory(GiB)": 61.91, "step": 28065, "token_acc": 0.945859872611465, "train_speed(iter/s)": 1.459771 }, { "epoch": 1.2026048584036673, "grad_norm": 5.000336170196533, "learning_rate": 8.639236332302964e-05, "loss": 0.5044078826904297, "memory(GiB)": 61.91, "step": 28070, "token_acc": 0.8973509933774835, "train_speed(iter/s)": 1.459768 }, { "epoch": 1.2028190737329163, "grad_norm": 2.0146937370300293, "learning_rate": 8.638774812369599e-05, "loss": 0.3279404640197754, "memory(GiB)": 61.91, "step": 28075, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.459807 }, { "epoch": 1.2030332890621653, "grad_norm": 3.6388721466064453, "learning_rate": 8.638313226516417e-05, "loss": 0.3655651569366455, "memory(GiB)": 61.91, "step": 28080, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.459821 }, { "epoch": 1.2032475043914141, "grad_norm": 3.607572317123413, "learning_rate": 8.637851574751776e-05, "loss": 0.500798511505127, "memory(GiB)": 61.91, "step": 28085, "token_acc": 0.8926174496644296, "train_speed(iter/s)": 1.459837 }, { "epoch": 1.2034617197206632, "grad_norm": 4.0003180503845215, "learning_rate": 8.637389857084041e-05, "loss": 0.467467212677002, "memory(GiB)": 61.91, "step": 28090, "token_acc": 0.8704318936877077, "train_speed(iter/s)": 1.459841 }, { "epoch": 1.2036759350499122, "grad_norm": 4.659339904785156, "learning_rate": 8.636928073521577e-05, "loss": 0.5863923549652099, "memory(GiB)": 61.91, "step": 28095, "token_acc": 0.8855218855218855, "train_speed(iter/s)": 1.45986 }, { "epoch": 1.203890150379161, "grad_norm": 5.063822269439697, "learning_rate": 8.63646622407275e-05, "loss": 0.4056821823120117, "memory(GiB)": 61.91, "step": 28100, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.459854 }, { "epoch": 1.20410436570841, "grad_norm": 6.201742172241211, "learning_rate": 8.636004308745925e-05, "loss": 0.3372138500213623, "memory(GiB)": 61.91, "step": 28105, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.459849 }, { "epoch": 1.204318581037659, "grad_norm": 1.7787158489227295, "learning_rate": 8.635542327549471e-05, "loss": 0.31066012382507324, "memory(GiB)": 61.91, "step": 28110, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.459848 }, { "epoch": 1.204532796366908, "grad_norm": 4.5597453117370605, "learning_rate": 8.635080280491758e-05, "loss": 0.24001779556274414, "memory(GiB)": 61.91, "step": 28115, "token_acc": 0.9477124183006536, "train_speed(iter/s)": 1.45987 }, { "epoch": 1.204747011696157, "grad_norm": 13.057588577270508, "learning_rate": 8.634618167581156e-05, "loss": 0.5345338821411133, "memory(GiB)": 61.91, "step": 28120, "token_acc": 0.8880866425992779, "train_speed(iter/s)": 1.459871 }, { "epoch": 1.204961227025406, "grad_norm": 3.2676684856414795, "learning_rate": 8.634155988826035e-05, "loss": 0.4874904632568359, "memory(GiB)": 61.91, "step": 28125, "token_acc": 0.9177631578947368, "train_speed(iter/s)": 1.459873 }, { "epoch": 1.2051754423546548, "grad_norm": 2.4030985832214355, "learning_rate": 8.633693744234771e-05, "loss": 0.23031084537506102, "memory(GiB)": 61.91, "step": 28130, "token_acc": 0.9471544715447154, "train_speed(iter/s)": 1.459877 }, { "epoch": 1.2053896576839038, "grad_norm": 4.123158931732178, "learning_rate": 8.633231433815735e-05, "loss": 0.7623964786529541, "memory(GiB)": 61.91, "step": 28135, "token_acc": 0.870253164556962, "train_speed(iter/s)": 1.459872 }, { "epoch": 1.2056038730131529, "grad_norm": 5.783171653747559, "learning_rate": 8.632769057577304e-05, "loss": 0.4329171657562256, "memory(GiB)": 61.91, "step": 28140, "token_acc": 0.9055793991416309, "train_speed(iter/s)": 1.45989 }, { "epoch": 1.2058180883424017, "grad_norm": 3.6788575649261475, "learning_rate": 8.632306615527853e-05, "loss": 0.2637831926345825, "memory(GiB)": 61.91, "step": 28145, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.459898 }, { "epoch": 1.2060323036716507, "grad_norm": 5.337653160095215, "learning_rate": 8.63184410767576e-05, "loss": 0.6391002178192139, "memory(GiB)": 61.91, "step": 28150, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.459898 }, { "epoch": 1.2062465190008997, "grad_norm": 3.700960636138916, "learning_rate": 8.631381534029404e-05, "loss": 0.38558814525604246, "memory(GiB)": 61.91, "step": 28155, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.459891 }, { "epoch": 1.2064607343301486, "grad_norm": 3.387779951095581, "learning_rate": 8.630918894597166e-05, "loss": 0.31129357814788816, "memory(GiB)": 61.91, "step": 28160, "token_acc": 0.9395161290322581, "train_speed(iter/s)": 1.459918 }, { "epoch": 1.2066749496593976, "grad_norm": 2.8449461460113525, "learning_rate": 8.630456189387426e-05, "loss": 0.24952354431152343, "memory(GiB)": 61.91, "step": 28165, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.459924 }, { "epoch": 1.2068891649886466, "grad_norm": 4.6224236488342285, "learning_rate": 8.629993418408566e-05, "loss": 0.5465425014495849, "memory(GiB)": 61.91, "step": 28170, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.459923 }, { "epoch": 1.2071033803178954, "grad_norm": 4.5857319831848145, "learning_rate": 8.62953058166897e-05, "loss": 0.35084664821624756, "memory(GiB)": 61.91, "step": 28175, "token_acc": 0.9383116883116883, "train_speed(iter/s)": 1.459948 }, { "epoch": 1.2073175956471445, "grad_norm": 3.8924574851989746, "learning_rate": 8.629067679177023e-05, "loss": 0.4107808589935303, "memory(GiB)": 61.91, "step": 28180, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.459929 }, { "epoch": 1.2075318109763935, "grad_norm": 2.2992641925811768, "learning_rate": 8.62860471094111e-05, "loss": 0.3128371715545654, "memory(GiB)": 61.91, "step": 28185, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.45994 }, { "epoch": 1.2077460263056423, "grad_norm": 2.859445095062256, "learning_rate": 8.62814167696962e-05, "loss": 0.3536311864852905, "memory(GiB)": 61.91, "step": 28190, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.459951 }, { "epoch": 1.2079602416348914, "grad_norm": 3.191948175430298, "learning_rate": 8.627678577270939e-05, "loss": 0.4690140724182129, "memory(GiB)": 61.91, "step": 28195, "token_acc": 0.8809523809523809, "train_speed(iter/s)": 1.45996 }, { "epoch": 1.2081744569641404, "grad_norm": 2.5329604148864746, "learning_rate": 8.627215411853459e-05, "loss": 0.42574052810668944, "memory(GiB)": 61.91, "step": 28200, "token_acc": 0.9137254901960784, "train_speed(iter/s)": 1.459976 }, { "epoch": 1.2083886722933892, "grad_norm": 2.932382822036743, "learning_rate": 8.626752180725568e-05, "loss": 0.4206682205200195, "memory(GiB)": 61.91, "step": 28205, "token_acc": 0.9134078212290503, "train_speed(iter/s)": 1.459993 }, { "epoch": 1.2086028876226382, "grad_norm": 3.857832193374634, "learning_rate": 8.626288883895659e-05, "loss": 0.41814308166503905, "memory(GiB)": 61.91, "step": 28210, "token_acc": 0.9033613445378151, "train_speed(iter/s)": 1.459972 }, { "epoch": 1.2088171029518873, "grad_norm": 2.0870256423950195, "learning_rate": 8.625825521372125e-05, "loss": 0.4810174942016602, "memory(GiB)": 61.91, "step": 28215, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.459962 }, { "epoch": 1.209031318281136, "grad_norm": 3.25201678276062, "learning_rate": 8.62536209316336e-05, "loss": 0.20692050457000732, "memory(GiB)": 61.91, "step": 28220, "token_acc": 0.9573170731707317, "train_speed(iter/s)": 1.459964 }, { "epoch": 1.2092455336103851, "grad_norm": 3.2210497856140137, "learning_rate": 8.624898599277762e-05, "loss": 0.25086545944213867, "memory(GiB)": 61.91, "step": 28225, "token_acc": 0.9455782312925171, "train_speed(iter/s)": 1.459955 }, { "epoch": 1.2094597489396341, "grad_norm": 2.4520580768585205, "learning_rate": 8.624435039723724e-05, "loss": 0.3127427339553833, "memory(GiB)": 61.91, "step": 28230, "token_acc": 0.9243697478991597, "train_speed(iter/s)": 1.460008 }, { "epoch": 1.209673964268883, "grad_norm": 0.32143110036849976, "learning_rate": 8.623971414509644e-05, "loss": 0.3568761348724365, "memory(GiB)": 61.91, "step": 28235, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.460033 }, { "epoch": 1.209888179598132, "grad_norm": 2.4051144123077393, "learning_rate": 8.623507723643924e-05, "loss": 0.40650715827941897, "memory(GiB)": 61.91, "step": 28240, "token_acc": 0.9140625, "train_speed(iter/s)": 1.460036 }, { "epoch": 1.210102394927381, "grad_norm": 4.323846340179443, "learning_rate": 8.623043967134963e-05, "loss": 0.511476469039917, "memory(GiB)": 65.35, "step": 28245, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.460033 }, { "epoch": 1.2103166102566298, "grad_norm": 3.9283812046051025, "learning_rate": 8.622580144991159e-05, "loss": 0.43967185020446775, "memory(GiB)": 65.35, "step": 28250, "token_acc": 0.9169675090252708, "train_speed(iter/s)": 1.460045 }, { "epoch": 1.2105308255858789, "grad_norm": 2.900285482406616, "learning_rate": 8.622116257220921e-05, "loss": 0.6238034248352051, "memory(GiB)": 65.35, "step": 28255, "token_acc": 0.8562091503267973, "train_speed(iter/s)": 1.460056 }, { "epoch": 1.210745040915128, "grad_norm": 2.0437841415405273, "learning_rate": 8.621652303832647e-05, "loss": 0.4387816429138184, "memory(GiB)": 65.35, "step": 28260, "token_acc": 0.9158576051779935, "train_speed(iter/s)": 1.46006 }, { "epoch": 1.2109592562443767, "grad_norm": 3.513793706893921, "learning_rate": 8.621188284834745e-05, "loss": 0.49291324615478516, "memory(GiB)": 65.35, "step": 28265, "token_acc": 0.9021739130434783, "train_speed(iter/s)": 1.460065 }, { "epoch": 1.2111734715736258, "grad_norm": 5.792081832885742, "learning_rate": 8.62072420023562e-05, "loss": 0.2662515640258789, "memory(GiB)": 65.35, "step": 28270, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.460079 }, { "epoch": 1.2113876869028748, "grad_norm": 3.331434965133667, "learning_rate": 8.620260050043678e-05, "loss": 0.3295966625213623, "memory(GiB)": 65.35, "step": 28275, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.460093 }, { "epoch": 1.2116019022321236, "grad_norm": 4.264649391174316, "learning_rate": 8.619795834267331e-05, "loss": 0.5271106719970703, "memory(GiB)": 65.35, "step": 28280, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.460145 }, { "epoch": 1.2118161175613726, "grad_norm": 2.189558744430542, "learning_rate": 8.619331552914987e-05, "loss": 0.3954843759536743, "memory(GiB)": 65.35, "step": 28285, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.460168 }, { "epoch": 1.2120303328906217, "grad_norm": 1.8964002132415771, "learning_rate": 8.618867205995056e-05, "loss": 0.32153000831604006, "memory(GiB)": 65.35, "step": 28290, "token_acc": 0.9312977099236641, "train_speed(iter/s)": 1.46016 }, { "epoch": 1.2122445482198707, "grad_norm": 3.0196118354797363, "learning_rate": 8.618402793515949e-05, "loss": 0.3545821666717529, "memory(GiB)": 65.35, "step": 28295, "token_acc": 0.9204892966360856, "train_speed(iter/s)": 1.460169 }, { "epoch": 1.2124587635491195, "grad_norm": 3.8822877407073975, "learning_rate": 8.617938315486084e-05, "loss": 0.5855368614196778, "memory(GiB)": 65.35, "step": 28300, "token_acc": 0.8971631205673759, "train_speed(iter/s)": 1.460175 }, { "epoch": 1.2126729788783686, "grad_norm": 2.030122995376587, "learning_rate": 8.617473771913871e-05, "loss": 0.37313051223754884, "memory(GiB)": 65.35, "step": 28305, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.460173 }, { "epoch": 1.2128871942076176, "grad_norm": 2.709770679473877, "learning_rate": 8.617009162807727e-05, "loss": 0.2883031368255615, "memory(GiB)": 65.35, "step": 28310, "token_acc": 0.9395770392749244, "train_speed(iter/s)": 1.460188 }, { "epoch": 1.2131014095368664, "grad_norm": 6.382560729980469, "learning_rate": 8.61654448817607e-05, "loss": 0.8410708427429199, "memory(GiB)": 65.35, "step": 28315, "token_acc": 0.8308157099697885, "train_speed(iter/s)": 1.460237 }, { "epoch": 1.2133156248661154, "grad_norm": 5.385939121246338, "learning_rate": 8.616079748027317e-05, "loss": 0.37341487407684326, "memory(GiB)": 65.35, "step": 28320, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.460219 }, { "epoch": 1.2135298401953645, "grad_norm": 1.8570003509521484, "learning_rate": 8.615614942369887e-05, "loss": 0.3430162906646729, "memory(GiB)": 65.35, "step": 28325, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.46023 }, { "epoch": 1.2137440555246133, "grad_norm": 3.420652389526367, "learning_rate": 8.6151500712122e-05, "loss": 0.6096773147583008, "memory(GiB)": 65.35, "step": 28330, "token_acc": 0.8674033149171271, "train_speed(iter/s)": 1.460267 }, { "epoch": 1.2139582708538623, "grad_norm": 6.012396812438965, "learning_rate": 8.614685134562679e-05, "loss": 0.9031429290771484, "memory(GiB)": 65.35, "step": 28335, "token_acc": 0.8303886925795053, "train_speed(iter/s)": 1.460253 }, { "epoch": 1.2141724861831114, "grad_norm": 3.566208839416504, "learning_rate": 8.614220132429746e-05, "loss": 0.3378458976745605, "memory(GiB)": 65.35, "step": 28340, "token_acc": 0.9280575539568345, "train_speed(iter/s)": 1.460246 }, { "epoch": 1.2143867015123602, "grad_norm": 3.81057071685791, "learning_rate": 8.613755064821824e-05, "loss": 0.3401157855987549, "memory(GiB)": 65.35, "step": 28345, "token_acc": 0.9206896551724137, "train_speed(iter/s)": 1.460231 }, { "epoch": 1.2146009168416092, "grad_norm": 2.6822686195373535, "learning_rate": 8.613289931747339e-05, "loss": 0.5216886043548584, "memory(GiB)": 65.35, "step": 28350, "token_acc": 0.9009584664536742, "train_speed(iter/s)": 1.460226 }, { "epoch": 1.2148151321708582, "grad_norm": 7.4567179679870605, "learning_rate": 8.612824733214717e-05, "loss": 0.6001446723937989, "memory(GiB)": 65.35, "step": 28355, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.460216 }, { "epoch": 1.215029347500107, "grad_norm": 5.147367000579834, "learning_rate": 8.612359469232387e-05, "loss": 0.3531834125518799, "memory(GiB)": 65.35, "step": 28360, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.46021 }, { "epoch": 1.215243562829356, "grad_norm": 6.37980842590332, "learning_rate": 8.611894139808776e-05, "loss": 0.21638710498809816, "memory(GiB)": 65.35, "step": 28365, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.460213 }, { "epoch": 1.2154577781586051, "grad_norm": 2.930772542953491, "learning_rate": 8.611428744952315e-05, "loss": 0.3490036725997925, "memory(GiB)": 65.35, "step": 28370, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.460221 }, { "epoch": 1.215671993487854, "grad_norm": 1.4674005508422852, "learning_rate": 8.610963284671433e-05, "loss": 0.40391831398010253, "memory(GiB)": 65.35, "step": 28375, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.460213 }, { "epoch": 1.215886208817103, "grad_norm": 1.4618442058563232, "learning_rate": 8.610497758974566e-05, "loss": 0.14197814464569092, "memory(GiB)": 65.35, "step": 28380, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.460224 }, { "epoch": 1.216100424146352, "grad_norm": 1.2743782997131348, "learning_rate": 8.610032167870144e-05, "loss": 0.45685505867004395, "memory(GiB)": 65.35, "step": 28385, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.460227 }, { "epoch": 1.2163146394756008, "grad_norm": 3.869875192642212, "learning_rate": 8.609566511366603e-05, "loss": 0.4434321403503418, "memory(GiB)": 65.35, "step": 28390, "token_acc": 0.8863636363636364, "train_speed(iter/s)": 1.46023 }, { "epoch": 1.2165288548048498, "grad_norm": 3.313890218734741, "learning_rate": 8.609100789472377e-05, "loss": 0.23853189945220948, "memory(GiB)": 65.35, "step": 28395, "token_acc": 0.9492957746478873, "train_speed(iter/s)": 1.460232 }, { "epoch": 1.2167430701340989, "grad_norm": 6.129242420196533, "learning_rate": 8.608635002195908e-05, "loss": 0.602261734008789, "memory(GiB)": 65.35, "step": 28400, "token_acc": 0.8767605633802817, "train_speed(iter/s)": 1.460235 }, { "epoch": 1.2169572854633477, "grad_norm": 2.5561740398406982, "learning_rate": 8.608169149545629e-05, "loss": 0.5529039859771728, "memory(GiB)": 65.35, "step": 28405, "token_acc": 0.8786885245901639, "train_speed(iter/s)": 1.460269 }, { "epoch": 1.2171715007925967, "grad_norm": 5.392173767089844, "learning_rate": 8.60770323152998e-05, "loss": 0.34566895961761473, "memory(GiB)": 65.35, "step": 28410, "token_acc": 0.934984520123839, "train_speed(iter/s)": 1.460269 }, { "epoch": 1.2173857161218458, "grad_norm": 3.6324377059936523, "learning_rate": 8.607237248157403e-05, "loss": 0.3127319097518921, "memory(GiB)": 65.35, "step": 28415, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.460261 }, { "epoch": 1.2175999314510946, "grad_norm": 2.924931764602661, "learning_rate": 8.606771199436341e-05, "loss": 0.35693798065185545, "memory(GiB)": 65.35, "step": 28420, "token_acc": 0.9197530864197531, "train_speed(iter/s)": 1.460257 }, { "epoch": 1.2178141467803436, "grad_norm": 3.2903215885162354, "learning_rate": 8.606305085375234e-05, "loss": 0.30550906658172605, "memory(GiB)": 65.35, "step": 28425, "token_acc": 0.9388379204892966, "train_speed(iter/s)": 1.460256 }, { "epoch": 1.2180283621095926, "grad_norm": 1.8877665996551514, "learning_rate": 8.605838905982526e-05, "loss": 0.3474111557006836, "memory(GiB)": 65.35, "step": 28430, "token_acc": 0.9118773946360154, "train_speed(iter/s)": 1.460235 }, { "epoch": 1.2182425774388415, "grad_norm": 3.0434253215789795, "learning_rate": 8.605372661266667e-05, "loss": 0.4959871292114258, "memory(GiB)": 65.35, "step": 28435, "token_acc": 0.9011976047904192, "train_speed(iter/s)": 1.460236 }, { "epoch": 1.2184567927680905, "grad_norm": 0.5078388452529907, "learning_rate": 8.604906351236097e-05, "loss": 0.49997587203979493, "memory(GiB)": 65.35, "step": 28440, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.460233 }, { "epoch": 1.2186710080973395, "grad_norm": 2.6899778842926025, "learning_rate": 8.604439975899269e-05, "loss": 0.545769739151001, "memory(GiB)": 65.35, "step": 28445, "token_acc": 0.8807692307692307, "train_speed(iter/s)": 1.460233 }, { "epoch": 1.2188852234265883, "grad_norm": 2.4926204681396484, "learning_rate": 8.60397353526463e-05, "loss": 0.3422074794769287, "memory(GiB)": 65.35, "step": 28450, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.460238 }, { "epoch": 1.2190994387558374, "grad_norm": 9.098722457885742, "learning_rate": 8.603507029340627e-05, "loss": 0.4513997554779053, "memory(GiB)": 65.35, "step": 28455, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.460241 }, { "epoch": 1.2193136540850864, "grad_norm": 0.38856807351112366, "learning_rate": 8.603040458135715e-05, "loss": 0.3014711618423462, "memory(GiB)": 65.35, "step": 28460, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.460256 }, { "epoch": 1.2195278694143352, "grad_norm": 5.859399318695068, "learning_rate": 8.602573821658345e-05, "loss": 0.2844375133514404, "memory(GiB)": 65.35, "step": 28465, "token_acc": 0.9479166666666666, "train_speed(iter/s)": 1.460245 }, { "epoch": 1.2197420847435843, "grad_norm": 2.3409230709075928, "learning_rate": 8.602107119916971e-05, "loss": 0.3645988702774048, "memory(GiB)": 65.35, "step": 28470, "token_acc": 0.9293286219081273, "train_speed(iter/s)": 1.460259 }, { "epoch": 1.2199563000728333, "grad_norm": 2.457714319229126, "learning_rate": 8.601640352920049e-05, "loss": 0.2600761651992798, "memory(GiB)": 65.35, "step": 28475, "token_acc": 0.9375, "train_speed(iter/s)": 1.460268 }, { "epoch": 1.220170515402082, "grad_norm": 3.0965373516082764, "learning_rate": 8.601173520676031e-05, "loss": 0.43578367233276366, "memory(GiB)": 70.5, "step": 28480, "token_acc": 0.9061371841155235, "train_speed(iter/s)": 1.46027 }, { "epoch": 1.2203847307313311, "grad_norm": 9.817384719848633, "learning_rate": 8.600706623193377e-05, "loss": 0.2756763219833374, "memory(GiB)": 70.5, "step": 28485, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.46029 }, { "epoch": 1.2205989460605802, "grad_norm": 0.3203599154949188, "learning_rate": 8.600239660480546e-05, "loss": 0.22174072265625, "memory(GiB)": 70.5, "step": 28490, "token_acc": 0.95625, "train_speed(iter/s)": 1.460284 }, { "epoch": 1.220813161389829, "grad_norm": 9.00351619720459, "learning_rate": 8.599772632545995e-05, "loss": 0.48617048263549806, "memory(GiB)": 70.5, "step": 28495, "token_acc": 0.8897435897435897, "train_speed(iter/s)": 1.460281 }, { "epoch": 1.221027376719078, "grad_norm": 0.043783530592918396, "learning_rate": 8.599305539398186e-05, "loss": 0.11486012935638427, "memory(GiB)": 70.5, "step": 28500, "token_acc": 0.9649805447470817, "train_speed(iter/s)": 1.460268 }, { "epoch": 1.221027376719078, "eval_loss": 2.6682300567626953, "eval_runtime": 13.7904, "eval_samples_per_second": 7.251, "eval_steps_per_second": 7.251, "eval_token_acc": 0.4305364511691884, "step": 28500 }, { "epoch": 1.221241592048327, "grad_norm": 2.6652050018310547, "learning_rate": 8.598838381045582e-05, "loss": 0.3963721990585327, "memory(GiB)": 70.5, "step": 28505, "token_acc": 0.5695952615992103, "train_speed(iter/s)": 1.459188 }, { "epoch": 1.2214558073775759, "grad_norm": 7.385496616363525, "learning_rate": 8.598371157496642e-05, "loss": 0.4642333984375, "memory(GiB)": 70.5, "step": 28510, "token_acc": 0.9007352941176471, "train_speed(iter/s)": 1.459177 }, { "epoch": 1.221670022706825, "grad_norm": 6.27261209487915, "learning_rate": 8.597903868759836e-05, "loss": 0.6061900138854981, "memory(GiB)": 70.5, "step": 28515, "token_acc": 0.8735632183908046, "train_speed(iter/s)": 1.459212 }, { "epoch": 1.221884238036074, "grad_norm": 5.944021224975586, "learning_rate": 8.597436514843625e-05, "loss": 0.4169018745422363, "memory(GiB)": 70.5, "step": 28520, "token_acc": 0.9149797570850202, "train_speed(iter/s)": 1.459217 }, { "epoch": 1.2220984533653227, "grad_norm": 4.3176798820495605, "learning_rate": 8.596969095756478e-05, "loss": 0.42948522567749026, "memory(GiB)": 70.5, "step": 28525, "token_acc": 0.9099099099099099, "train_speed(iter/s)": 1.459257 }, { "epoch": 1.2223126686945718, "grad_norm": 4.60506534576416, "learning_rate": 8.59650161150686e-05, "loss": 0.41767420768737795, "memory(GiB)": 70.5, "step": 28530, "token_acc": 0.9096774193548387, "train_speed(iter/s)": 1.459246 }, { "epoch": 1.2225268840238208, "grad_norm": 3.639427661895752, "learning_rate": 8.596034062103242e-05, "loss": 0.7455500602722168, "memory(GiB)": 70.5, "step": 28535, "token_acc": 0.85546875, "train_speed(iter/s)": 1.459241 }, { "epoch": 1.2227410993530696, "grad_norm": 1.5944561958312988, "learning_rate": 8.595566447554093e-05, "loss": 0.4830667972564697, "memory(GiB)": 70.5, "step": 28540, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.459232 }, { "epoch": 1.2229553146823187, "grad_norm": 2.8588011264801025, "learning_rate": 8.595098767867886e-05, "loss": 0.5178374767303466, "memory(GiB)": 70.5, "step": 28545, "token_acc": 0.8907849829351536, "train_speed(iter/s)": 1.459247 }, { "epoch": 1.2231695300115677, "grad_norm": 4.137483596801758, "learning_rate": 8.594631023053093e-05, "loss": 0.3964956760406494, "memory(GiB)": 70.5, "step": 28550, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.459252 }, { "epoch": 1.2233837453408165, "grad_norm": 3.0286550521850586, "learning_rate": 8.594163213118185e-05, "loss": 0.4017927646636963, "memory(GiB)": 70.5, "step": 28555, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.459266 }, { "epoch": 1.2235979606700655, "grad_norm": 2.804932117462158, "learning_rate": 8.593695338071639e-05, "loss": 0.21779294013977052, "memory(GiB)": 70.5, "step": 28560, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.459278 }, { "epoch": 1.2238121759993146, "grad_norm": 2.500014305114746, "learning_rate": 8.593227397921932e-05, "loss": 0.5392621994018555, "memory(GiB)": 70.5, "step": 28565, "token_acc": 0.8758620689655172, "train_speed(iter/s)": 1.459322 }, { "epoch": 1.2240263913285634, "grad_norm": 6.010717391967773, "learning_rate": 8.59275939267754e-05, "loss": 0.4046011447906494, "memory(GiB)": 70.5, "step": 28570, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.459323 }, { "epoch": 1.2242406066578124, "grad_norm": 2.4557695388793945, "learning_rate": 8.59229132234694e-05, "loss": 0.5093642234802246, "memory(GiB)": 70.5, "step": 28575, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.45935 }, { "epoch": 1.2244548219870615, "grad_norm": 4.539165496826172, "learning_rate": 8.591823186938614e-05, "loss": 0.38286681175231935, "memory(GiB)": 70.5, "step": 28580, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.459342 }, { "epoch": 1.2246690373163103, "grad_norm": 3.756697177886963, "learning_rate": 8.591354986461042e-05, "loss": 0.32461550235748293, "memory(GiB)": 70.5, "step": 28585, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.459341 }, { "epoch": 1.2248832526455593, "grad_norm": 3.549438953399658, "learning_rate": 8.590886720922704e-05, "loss": 0.3358589172363281, "memory(GiB)": 70.5, "step": 28590, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.459339 }, { "epoch": 1.2250974679748083, "grad_norm": 4.6078691482543945, "learning_rate": 8.590418390332084e-05, "loss": 0.43480691909790037, "memory(GiB)": 70.5, "step": 28595, "token_acc": 0.9041666666666667, "train_speed(iter/s)": 1.459339 }, { "epoch": 1.2253116833040572, "grad_norm": 4.199001789093018, "learning_rate": 8.589949994697669e-05, "loss": 0.4181361198425293, "memory(GiB)": 70.5, "step": 28600, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.459376 }, { "epoch": 1.2255258986333062, "grad_norm": 6.4896769523620605, "learning_rate": 8.589481534027941e-05, "loss": 0.31342453956604005, "memory(GiB)": 70.5, "step": 28605, "token_acc": 0.9272030651340997, "train_speed(iter/s)": 1.459364 }, { "epoch": 1.2257401139625552, "grad_norm": 8.924944877624512, "learning_rate": 8.589013008331388e-05, "loss": 0.43149623870849607, "memory(GiB)": 70.5, "step": 28610, "token_acc": 0.8901960784313725, "train_speed(iter/s)": 1.459385 }, { "epoch": 1.225954329291804, "grad_norm": 1.7413734197616577, "learning_rate": 8.588544417616497e-05, "loss": 0.3726505756378174, "memory(GiB)": 70.5, "step": 28615, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.459372 }, { "epoch": 1.226168544621053, "grad_norm": 2.79915714263916, "learning_rate": 8.588075761891758e-05, "loss": 0.2623190402984619, "memory(GiB)": 70.5, "step": 28620, "token_acc": 0.9392712550607287, "train_speed(iter/s)": 1.459341 }, { "epoch": 1.226382759950302, "grad_norm": 2.4305312633514404, "learning_rate": 8.587607041165662e-05, "loss": 0.42760567665100097, "memory(GiB)": 70.5, "step": 28625, "token_acc": 0.9087719298245615, "train_speed(iter/s)": 1.459344 }, { "epoch": 1.226596975279551, "grad_norm": 3.396895408630371, "learning_rate": 8.587138255446698e-05, "loss": 0.547541332244873, "memory(GiB)": 70.5, "step": 28630, "token_acc": 0.875968992248062, "train_speed(iter/s)": 1.459338 }, { "epoch": 1.2268111906088, "grad_norm": 4.583094120025635, "learning_rate": 8.586669404743359e-05, "loss": 0.72474365234375, "memory(GiB)": 70.5, "step": 28635, "token_acc": 0.8436482084690554, "train_speed(iter/s)": 1.459363 }, { "epoch": 1.227025405938049, "grad_norm": 5.1589035987854, "learning_rate": 8.586200489064138e-05, "loss": 0.405120325088501, "memory(GiB)": 70.5, "step": 28640, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.459366 }, { "epoch": 1.2272396212672978, "grad_norm": 3.3551888465881348, "learning_rate": 8.585731508417533e-05, "loss": 0.354505729675293, "memory(GiB)": 70.5, "step": 28645, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.459373 }, { "epoch": 1.2274538365965468, "grad_norm": 3.5921969413757324, "learning_rate": 8.585262462812038e-05, "loss": 0.392541241645813, "memory(GiB)": 70.5, "step": 28650, "token_acc": 0.9023569023569024, "train_speed(iter/s)": 1.459358 }, { "epoch": 1.2276680519257959, "grad_norm": 3.168592691421509, "learning_rate": 8.584793352256149e-05, "loss": 0.3253945827484131, "memory(GiB)": 70.5, "step": 28655, "token_acc": 0.9227642276422764, "train_speed(iter/s)": 1.459359 }, { "epoch": 1.2278822672550447, "grad_norm": 3.7773995399475098, "learning_rate": 8.584324176758367e-05, "loss": 0.3911292552947998, "memory(GiB)": 70.5, "step": 28660, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.459357 }, { "epoch": 1.2280964825842937, "grad_norm": 3.424224376678467, "learning_rate": 8.583854936327189e-05, "loss": 0.5317602157592773, "memory(GiB)": 70.5, "step": 28665, "token_acc": 0.9, "train_speed(iter/s)": 1.459364 }, { "epoch": 1.2283106979135427, "grad_norm": 2.6607155799865723, "learning_rate": 8.583385630971118e-05, "loss": 0.4934698104858398, "memory(GiB)": 70.5, "step": 28670, "token_acc": 0.8859060402684564, "train_speed(iter/s)": 1.459397 }, { "epoch": 1.2285249132427916, "grad_norm": 3.4515416622161865, "learning_rate": 8.582916260698655e-05, "loss": 0.3314671039581299, "memory(GiB)": 70.5, "step": 28675, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.459406 }, { "epoch": 1.2287391285720406, "grad_norm": 1.3812919855117798, "learning_rate": 8.582446825518302e-05, "loss": 0.4162896156311035, "memory(GiB)": 70.5, "step": 28680, "token_acc": 0.8992248062015504, "train_speed(iter/s)": 1.459407 }, { "epoch": 1.2289533439012896, "grad_norm": 5.83089542388916, "learning_rate": 8.581977325438564e-05, "loss": 0.35457663536071776, "memory(GiB)": 70.5, "step": 28685, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.459439 }, { "epoch": 1.2291675592305384, "grad_norm": 3.297933578491211, "learning_rate": 8.581507760467945e-05, "loss": 0.2458139181137085, "memory(GiB)": 70.5, "step": 28690, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.45945 }, { "epoch": 1.2293817745597875, "grad_norm": 3.665548801422119, "learning_rate": 8.581038130614957e-05, "loss": 0.3766390323638916, "memory(GiB)": 70.5, "step": 28695, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.459453 }, { "epoch": 1.2295959898890365, "grad_norm": 1.5152146816253662, "learning_rate": 8.580568435888102e-05, "loss": 0.5002099990844726, "memory(GiB)": 70.5, "step": 28700, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.459433 }, { "epoch": 1.2298102052182853, "grad_norm": 1.8952580690383911, "learning_rate": 8.580098676295891e-05, "loss": 0.6652101993560791, "memory(GiB)": 70.5, "step": 28705, "token_acc": 0.8769230769230769, "train_speed(iter/s)": 1.459482 }, { "epoch": 1.2300244205475344, "grad_norm": 3.0868444442749023, "learning_rate": 8.579628851846834e-05, "loss": 0.1136460542678833, "memory(GiB)": 70.5, "step": 28710, "token_acc": 0.9641434262948207, "train_speed(iter/s)": 1.459476 }, { "epoch": 1.2302386358767834, "grad_norm": 1.3543386459350586, "learning_rate": 8.579158962549443e-05, "loss": 0.4275153636932373, "memory(GiB)": 70.5, "step": 28715, "token_acc": 0.9051724137931034, "train_speed(iter/s)": 1.459472 }, { "epoch": 1.2304528512060324, "grad_norm": 4.341468334197998, "learning_rate": 8.578689008412229e-05, "loss": 0.47058582305908203, "memory(GiB)": 70.5, "step": 28720, "token_acc": 0.8961538461538462, "train_speed(iter/s)": 1.459491 }, { "epoch": 1.2306670665352812, "grad_norm": 2.261753559112549, "learning_rate": 8.578218989443706e-05, "loss": 0.32040581703186033, "memory(GiB)": 70.5, "step": 28725, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.459504 }, { "epoch": 1.2308812818645303, "grad_norm": 7.691011905670166, "learning_rate": 8.577748905652389e-05, "loss": 0.27357323169708253, "memory(GiB)": 70.5, "step": 28730, "token_acc": 0.9375, "train_speed(iter/s)": 1.459549 }, { "epoch": 1.2310954971937793, "grad_norm": 4.0928754806518555, "learning_rate": 8.577278757046797e-05, "loss": 0.43462166786193845, "memory(GiB)": 70.5, "step": 28735, "token_acc": 0.8834355828220859, "train_speed(iter/s)": 1.459551 }, { "epoch": 1.2313097125230281, "grad_norm": 0.43395888805389404, "learning_rate": 8.576808543635443e-05, "loss": 0.21653640270233154, "memory(GiB)": 70.5, "step": 28740, "token_acc": 0.9552238805970149, "train_speed(iter/s)": 1.459548 }, { "epoch": 1.2315239278522772, "grad_norm": 4.157932281494141, "learning_rate": 8.576338265426846e-05, "loss": 0.6290278911590577, "memory(GiB)": 70.5, "step": 28745, "token_acc": 0.8810975609756098, "train_speed(iter/s)": 1.459567 }, { "epoch": 1.2317381431815262, "grad_norm": 5.734848499298096, "learning_rate": 8.575867922429529e-05, "loss": 0.5779965877532959, "memory(GiB)": 70.5, "step": 28750, "token_acc": 0.8736462093862816, "train_speed(iter/s)": 1.459584 }, { "epoch": 1.231952358510775, "grad_norm": 4.653557300567627, "learning_rate": 8.575397514652008e-05, "loss": 0.35430202484130857, "memory(GiB)": 70.5, "step": 28755, "token_acc": 0.9102167182662538, "train_speed(iter/s)": 1.459589 }, { "epoch": 1.232166573840024, "grad_norm": 6.49091100692749, "learning_rate": 8.574927042102807e-05, "loss": 0.41132383346557616, "memory(GiB)": 70.5, "step": 28760, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.459617 }, { "epoch": 1.232380789169273, "grad_norm": 0.679785966873169, "learning_rate": 8.574456504790451e-05, "loss": 0.19811904430389404, "memory(GiB)": 70.5, "step": 28765, "token_acc": 0.94140625, "train_speed(iter/s)": 1.459653 }, { "epoch": 1.2325950044985219, "grad_norm": 2.063711643218994, "learning_rate": 8.57398590272346e-05, "loss": 0.36643757820129397, "memory(GiB)": 70.5, "step": 28770, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.459686 }, { "epoch": 1.232809219827771, "grad_norm": 1.8696587085723877, "learning_rate": 8.573515235910364e-05, "loss": 0.3570316553115845, "memory(GiB)": 70.5, "step": 28775, "token_acc": 0.9171597633136095, "train_speed(iter/s)": 1.459691 }, { "epoch": 1.23302343515702, "grad_norm": 2.8457818031311035, "learning_rate": 8.573044504359686e-05, "loss": 0.45464282035827636, "memory(GiB)": 70.5, "step": 28780, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.459713 }, { "epoch": 1.2332376504862688, "grad_norm": 4.043572902679443, "learning_rate": 8.572573708079954e-05, "loss": 0.5563277721405029, "memory(GiB)": 70.5, "step": 28785, "token_acc": 0.8991097922848664, "train_speed(iter/s)": 1.459727 }, { "epoch": 1.2334518658155178, "grad_norm": 7.788190841674805, "learning_rate": 8.5721028470797e-05, "loss": 0.25124990940093994, "memory(GiB)": 70.5, "step": 28790, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.459724 }, { "epoch": 1.2336660811447668, "grad_norm": 8.407608985900879, "learning_rate": 8.571631921367451e-05, "loss": 0.38669581413269044, "memory(GiB)": 70.5, "step": 28795, "token_acc": 0.94375, "train_speed(iter/s)": 1.459725 }, { "epoch": 1.2338802964740156, "grad_norm": 7.932103633880615, "learning_rate": 8.571160930951738e-05, "loss": 0.5228333473205566, "memory(GiB)": 70.5, "step": 28800, "token_acc": 0.8858131487889274, "train_speed(iter/s)": 1.459732 }, { "epoch": 1.2340945118032647, "grad_norm": 2.0269482135772705, "learning_rate": 8.570689875841095e-05, "loss": 0.2339808464050293, "memory(GiB)": 70.5, "step": 28805, "token_acc": 0.9537953795379538, "train_speed(iter/s)": 1.459724 }, { "epoch": 1.2343087271325137, "grad_norm": 2.3556950092315674, "learning_rate": 8.570218756044058e-05, "loss": 0.2837003469467163, "memory(GiB)": 70.5, "step": 28810, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.459722 }, { "epoch": 1.2345229424617625, "grad_norm": 3.0436832904815674, "learning_rate": 8.569747571569157e-05, "loss": 0.46598243713378906, "memory(GiB)": 70.5, "step": 28815, "token_acc": 0.910958904109589, "train_speed(iter/s)": 1.45975 }, { "epoch": 1.2347371577910116, "grad_norm": 1.9308842420578003, "learning_rate": 8.569276322424931e-05, "loss": 0.227305006980896, "memory(GiB)": 70.5, "step": 28820, "token_acc": 0.948339483394834, "train_speed(iter/s)": 1.45975 }, { "epoch": 1.2349513731202606, "grad_norm": 3.580707311630249, "learning_rate": 8.568805008619916e-05, "loss": 0.6920265674591064, "memory(GiB)": 70.5, "step": 28825, "token_acc": 0.8604651162790697, "train_speed(iter/s)": 1.459774 }, { "epoch": 1.2351655884495094, "grad_norm": 5.626827239990234, "learning_rate": 8.568333630162649e-05, "loss": 0.4729918956756592, "memory(GiB)": 70.5, "step": 28830, "token_acc": 0.8960573476702509, "train_speed(iter/s)": 1.459767 }, { "epoch": 1.2353798037787584, "grad_norm": 5.439419269561768, "learning_rate": 8.567862187061673e-05, "loss": 0.36155588626861573, "memory(GiB)": 70.5, "step": 28835, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.459767 }, { "epoch": 1.2355940191080075, "grad_norm": 3.7669289112091064, "learning_rate": 8.567390679325525e-05, "loss": 0.24959793090820312, "memory(GiB)": 70.5, "step": 28840, "token_acc": 0.9423728813559322, "train_speed(iter/s)": 1.459761 }, { "epoch": 1.2358082344372563, "grad_norm": 1.7373647689819336, "learning_rate": 8.56691910696275e-05, "loss": 0.44876947402954104, "memory(GiB)": 70.5, "step": 28845, "token_acc": 0.9102564102564102, "train_speed(iter/s)": 1.459759 }, { "epoch": 1.2360224497665053, "grad_norm": 2.7433671951293945, "learning_rate": 8.566447469981888e-05, "loss": 0.4350141525268555, "memory(GiB)": 70.5, "step": 28850, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.459774 }, { "epoch": 1.2362366650957544, "grad_norm": 1.865427017211914, "learning_rate": 8.565975768391484e-05, "loss": 0.24694678783416749, "memory(GiB)": 70.5, "step": 28855, "token_acc": 0.9484126984126984, "train_speed(iter/s)": 1.459781 }, { "epoch": 1.2364508804250032, "grad_norm": 3.5901381969451904, "learning_rate": 8.565504002200084e-05, "loss": 0.2563082218170166, "memory(GiB)": 70.5, "step": 28860, "token_acc": 0.94, "train_speed(iter/s)": 1.459786 }, { "epoch": 1.2366650957542522, "grad_norm": 4.55043888092041, "learning_rate": 8.565032171416236e-05, "loss": 0.447910737991333, "memory(GiB)": 70.5, "step": 28865, "token_acc": 0.887459807073955, "train_speed(iter/s)": 1.459781 }, { "epoch": 1.2368793110835012, "grad_norm": 0.5246461033821106, "learning_rate": 8.564560276048483e-05, "loss": 0.1848062038421631, "memory(GiB)": 70.5, "step": 28870, "token_acc": 0.9501915708812261, "train_speed(iter/s)": 1.459781 }, { "epoch": 1.23709352641275, "grad_norm": 1.4550343751907349, "learning_rate": 8.56408831610538e-05, "loss": 0.21954522132873536, "memory(GiB)": 70.5, "step": 28875, "token_acc": 0.9527272727272728, "train_speed(iter/s)": 1.459775 }, { "epoch": 1.237307741741999, "grad_norm": 1.5454086065292358, "learning_rate": 8.563616291595473e-05, "loss": 0.29167561531066893, "memory(GiB)": 70.5, "step": 28880, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.459768 }, { "epoch": 1.2375219570712481, "grad_norm": 3.651888370513916, "learning_rate": 8.563144202527312e-05, "loss": 0.38002190589904783, "memory(GiB)": 70.5, "step": 28885, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.459757 }, { "epoch": 1.237736172400497, "grad_norm": 3.1474192142486572, "learning_rate": 8.562672048909453e-05, "loss": 0.5399748325347901, "memory(GiB)": 70.5, "step": 28890, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.459778 }, { "epoch": 1.237950387729746, "grad_norm": 7.90514612197876, "learning_rate": 8.562199830750447e-05, "loss": 0.4144142150878906, "memory(GiB)": 70.5, "step": 28895, "token_acc": 0.9025157232704403, "train_speed(iter/s)": 1.459784 }, { "epoch": 1.238164603058995, "grad_norm": 4.611439228057861, "learning_rate": 8.561727548058849e-05, "loss": 0.5246325492858886, "memory(GiB)": 70.5, "step": 28900, "token_acc": 0.8940677966101694, "train_speed(iter/s)": 1.459792 }, { "epoch": 1.2383788183882438, "grad_norm": 2.7660176753997803, "learning_rate": 8.561255200843216e-05, "loss": 0.38808698654174806, "memory(GiB)": 70.5, "step": 28905, "token_acc": 0.9100719424460432, "train_speed(iter/s)": 1.459796 }, { "epoch": 1.2385930337174929, "grad_norm": 4.510849952697754, "learning_rate": 8.560782789112105e-05, "loss": 0.4601875305175781, "memory(GiB)": 70.5, "step": 28910, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.459799 }, { "epoch": 1.2388072490467419, "grad_norm": 4.3433661460876465, "learning_rate": 8.560310312874074e-05, "loss": 0.4888437271118164, "memory(GiB)": 70.5, "step": 28915, "token_acc": 0.9118773946360154, "train_speed(iter/s)": 1.459801 }, { "epoch": 1.2390214643759907, "grad_norm": 4.092221736907959, "learning_rate": 8.55983777213768e-05, "loss": 0.5752559661865234, "memory(GiB)": 70.5, "step": 28920, "token_acc": 0.9035087719298246, "train_speed(iter/s)": 1.459796 }, { "epoch": 1.2392356797052397, "grad_norm": 3.211754083633423, "learning_rate": 8.559365166911486e-05, "loss": 0.5797932624816895, "memory(GiB)": 70.5, "step": 28925, "token_acc": 0.8930041152263375, "train_speed(iter/s)": 1.459831 }, { "epoch": 1.2394498950344888, "grad_norm": 4.0878987312316895, "learning_rate": 8.558892497204052e-05, "loss": 0.2980156660079956, "memory(GiB)": 70.5, "step": 28930, "token_acc": 0.9392857142857143, "train_speed(iter/s)": 1.459842 }, { "epoch": 1.2396641103637376, "grad_norm": 3.2938647270202637, "learning_rate": 8.558419763023944e-05, "loss": 0.2715883255004883, "memory(GiB)": 70.5, "step": 28935, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.459828 }, { "epoch": 1.2398783256929866, "grad_norm": 1.4768784046173096, "learning_rate": 8.557946964379723e-05, "loss": 0.27364349365234375, "memory(GiB)": 70.5, "step": 28940, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.459824 }, { "epoch": 1.2400925410222357, "grad_norm": 2.780395746231079, "learning_rate": 8.557474101279955e-05, "loss": 0.7205383777618408, "memory(GiB)": 70.5, "step": 28945, "token_acc": 0.8473520249221184, "train_speed(iter/s)": 1.459823 }, { "epoch": 1.2403067563514845, "grad_norm": 1.0614615678787231, "learning_rate": 8.557001173733206e-05, "loss": 0.29847934246063235, "memory(GiB)": 70.5, "step": 28950, "token_acc": 0.9294871794871795, "train_speed(iter/s)": 1.459823 }, { "epoch": 1.2405209716807335, "grad_norm": 1.1671044826507568, "learning_rate": 8.556528181748044e-05, "loss": 0.304517126083374, "memory(GiB)": 70.5, "step": 28955, "token_acc": 0.9376947040498442, "train_speed(iter/s)": 1.459842 }, { "epoch": 1.2407351870099825, "grad_norm": 2.3180508613586426, "learning_rate": 8.556055125333039e-05, "loss": 0.31635441780090334, "memory(GiB)": 70.5, "step": 28960, "token_acc": 0.9475308641975309, "train_speed(iter/s)": 1.45985 }, { "epoch": 1.2409494023392313, "grad_norm": 2.8442888259887695, "learning_rate": 8.555582004496758e-05, "loss": 0.4768516540527344, "memory(GiB)": 70.5, "step": 28965, "token_acc": 0.8869257950530035, "train_speed(iter/s)": 1.459837 }, { "epoch": 1.2411636176684804, "grad_norm": 1.021620273590088, "learning_rate": 8.555108819247774e-05, "loss": 0.14426627159118652, "memory(GiB)": 70.5, "step": 28970, "token_acc": 0.9570957095709571, "train_speed(iter/s)": 1.459821 }, { "epoch": 1.2413778329977294, "grad_norm": 1.8862017393112183, "learning_rate": 8.55463556959466e-05, "loss": 0.40767779350280764, "memory(GiB)": 70.5, "step": 28975, "token_acc": 0.9025974025974026, "train_speed(iter/s)": 1.459804 }, { "epoch": 1.2415920483269782, "grad_norm": 5.05383825302124, "learning_rate": 8.554162255545987e-05, "loss": 0.7440976142883301, "memory(GiB)": 70.5, "step": 28980, "token_acc": 0.8355263157894737, "train_speed(iter/s)": 1.459801 }, { "epoch": 1.2418062636562273, "grad_norm": 4.222079753875732, "learning_rate": 8.55368887711033e-05, "loss": 0.5529540538787842, "memory(GiB)": 70.5, "step": 28985, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.459816 }, { "epoch": 1.2420204789854763, "grad_norm": 0.13829588890075684, "learning_rate": 8.553215434296268e-05, "loss": 0.3693918943405151, "memory(GiB)": 70.5, "step": 28990, "token_acc": 0.937007874015748, "train_speed(iter/s)": 1.459829 }, { "epoch": 1.242234694314725, "grad_norm": 6.722729682922363, "learning_rate": 8.552741927112373e-05, "loss": 0.30765609741210936, "memory(GiB)": 70.5, "step": 28995, "token_acc": 0.932, "train_speed(iter/s)": 1.45984 }, { "epoch": 1.2424489096439741, "grad_norm": 1.3156976699829102, "learning_rate": 8.552268355567226e-05, "loss": 0.3905435800552368, "memory(GiB)": 70.5, "step": 29000, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.459862 }, { "epoch": 1.2424489096439741, "eval_loss": 2.8005011081695557, "eval_runtime": 12.3422, "eval_samples_per_second": 8.102, "eval_steps_per_second": 8.102, "eval_token_acc": 0.4096209912536443, "step": 29000 }, { "epoch": 1.2426631249732232, "grad_norm": 0.11432226002216339, "learning_rate": 8.551794719669405e-05, "loss": 0.50243239402771, "memory(GiB)": 70.5, "step": 29005, "token_acc": 0.5523906408952187, "train_speed(iter/s)": 1.458859 }, { "epoch": 1.242877340302472, "grad_norm": 5.912202835083008, "learning_rate": 8.551321019427491e-05, "loss": 0.5213563442230225, "memory(GiB)": 70.5, "step": 29010, "token_acc": 0.8925619834710744, "train_speed(iter/s)": 1.458851 }, { "epoch": 1.243091555631721, "grad_norm": 5.640886306762695, "learning_rate": 8.550847254850065e-05, "loss": 0.344775652885437, "memory(GiB)": 70.5, "step": 29015, "token_acc": 0.9197530864197531, "train_speed(iter/s)": 1.458866 }, { "epoch": 1.24330577096097, "grad_norm": 1.5810576677322388, "learning_rate": 8.550373425945711e-05, "loss": 0.23305246829986573, "memory(GiB)": 70.5, "step": 29020, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.458903 }, { "epoch": 1.2435199862902189, "grad_norm": 2.995321035385132, "learning_rate": 8.549899532723011e-05, "loss": 0.47314157485961916, "memory(GiB)": 70.5, "step": 29025, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.458917 }, { "epoch": 1.243734201619468, "grad_norm": 1.5212292671203613, "learning_rate": 8.549425575190551e-05, "loss": 0.31415791511535646, "memory(GiB)": 70.5, "step": 29030, "token_acc": 0.93359375, "train_speed(iter/s)": 1.458923 }, { "epoch": 1.243948416948717, "grad_norm": 2.8490805625915527, "learning_rate": 8.548951553356917e-05, "loss": 0.31001482009887693, "memory(GiB)": 70.5, "step": 29035, "token_acc": 0.9288389513108615, "train_speed(iter/s)": 1.458932 }, { "epoch": 1.2441626322779658, "grad_norm": 3.2684969902038574, "learning_rate": 8.548477467230697e-05, "loss": 0.30856592655181886, "memory(GiB)": 70.5, "step": 29040, "token_acc": 0.9399293286219081, "train_speed(iter/s)": 1.458928 }, { "epoch": 1.2443768476072148, "grad_norm": 3.5933282375335693, "learning_rate": 8.548003316820478e-05, "loss": 0.6146927833557129, "memory(GiB)": 70.5, "step": 29045, "token_acc": 0.8644688644688645, "train_speed(iter/s)": 1.458931 }, { "epoch": 1.2445910629364638, "grad_norm": 7.600386619567871, "learning_rate": 8.547529102134852e-05, "loss": 0.6400364875793457, "memory(GiB)": 70.5, "step": 29050, "token_acc": 0.8664688427299704, "train_speed(iter/s)": 1.45896 }, { "epoch": 1.2448052782657126, "grad_norm": 0.9201118350028992, "learning_rate": 8.547054823182408e-05, "loss": 0.18789709806442262, "memory(GiB)": 70.5, "step": 29055, "token_acc": 0.9568627450980393, "train_speed(iter/s)": 1.458961 }, { "epoch": 1.2450194935949617, "grad_norm": 1.9765459299087524, "learning_rate": 8.546580479971737e-05, "loss": 0.2538090705871582, "memory(GiB)": 70.5, "step": 29060, "token_acc": 0.9606299212598425, "train_speed(iter/s)": 1.458952 }, { "epoch": 1.2452337089242107, "grad_norm": 2.0024800300598145, "learning_rate": 8.546106072511435e-05, "loss": 0.30450661182403566, "memory(GiB)": 70.5, "step": 29065, "token_acc": 0.9401993355481728, "train_speed(iter/s)": 1.458961 }, { "epoch": 1.2454479242534595, "grad_norm": 0.07608669996261597, "learning_rate": 8.545631600810094e-05, "loss": 0.17100688219070434, "memory(GiB)": 70.5, "step": 29070, "token_acc": 0.9568627450980393, "train_speed(iter/s)": 1.458965 }, { "epoch": 1.2456621395827085, "grad_norm": 2.5897634029388428, "learning_rate": 8.545157064876311e-05, "loss": 0.42132954597473143, "memory(GiB)": 70.5, "step": 29075, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.458962 }, { "epoch": 1.2458763549119576, "grad_norm": 3.237631320953369, "learning_rate": 8.544682464718684e-05, "loss": 0.30410854816436766, "memory(GiB)": 70.5, "step": 29080, "token_acc": 0.9291666666666667, "train_speed(iter/s)": 1.458963 }, { "epoch": 1.2460905702412064, "grad_norm": 1.7259944677352905, "learning_rate": 8.544207800345808e-05, "loss": 0.25592570304870604, "memory(GiB)": 70.5, "step": 29085, "token_acc": 0.9501466275659824, "train_speed(iter/s)": 1.458963 }, { "epoch": 1.2463047855704554, "grad_norm": 3.6296374797821045, "learning_rate": 8.54373307176628e-05, "loss": 0.5873505115509033, "memory(GiB)": 70.5, "step": 29090, "token_acc": 0.8605577689243028, "train_speed(iter/s)": 1.458968 }, { "epoch": 1.2465190008997045, "grad_norm": 3.0311009883880615, "learning_rate": 8.543258278988706e-05, "loss": 0.1008824348449707, "memory(GiB)": 70.5, "step": 29095, "token_acc": 0.9738805970149254, "train_speed(iter/s)": 1.458973 }, { "epoch": 1.2467332162289533, "grad_norm": 1.3565449714660645, "learning_rate": 8.542783422021684e-05, "loss": 0.29799978733062743, "memory(GiB)": 70.5, "step": 29100, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.458972 }, { "epoch": 1.2469474315582023, "grad_norm": 5.559714317321777, "learning_rate": 8.542308500873817e-05, "loss": 0.4301851749420166, "memory(GiB)": 70.5, "step": 29105, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.458988 }, { "epoch": 1.2471616468874513, "grad_norm": 1.1793007850646973, "learning_rate": 8.541833515553707e-05, "loss": 0.36353268623352053, "memory(GiB)": 70.5, "step": 29110, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.458988 }, { "epoch": 1.2473758622167002, "grad_norm": 3.349144458770752, "learning_rate": 8.541358466069962e-05, "loss": 0.47623772621154786, "memory(GiB)": 70.5, "step": 29115, "token_acc": 0.9082278481012658, "train_speed(iter/s)": 1.458989 }, { "epoch": 1.2475900775459492, "grad_norm": 7.387825012207031, "learning_rate": 8.540883352431186e-05, "loss": 0.7970787048339844, "memory(GiB)": 70.5, "step": 29120, "token_acc": 0.8632218844984803, "train_speed(iter/s)": 1.458998 }, { "epoch": 1.2478042928751982, "grad_norm": 0.4633074402809143, "learning_rate": 8.540408174645986e-05, "loss": 0.13221513032913207, "memory(GiB)": 70.5, "step": 29125, "token_acc": 0.9748427672955975, "train_speed(iter/s)": 1.459009 }, { "epoch": 1.248018508204447, "grad_norm": 5.0791215896606445, "learning_rate": 8.539932932722971e-05, "loss": 0.43569259643554686, "memory(GiB)": 70.5, "step": 29130, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.459025 }, { "epoch": 1.248232723533696, "grad_norm": 2.352419376373291, "learning_rate": 8.539457626670752e-05, "loss": 0.2878647565841675, "memory(GiB)": 70.5, "step": 29135, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.45903 }, { "epoch": 1.248446938862945, "grad_norm": 7.328221321105957, "learning_rate": 8.538982256497937e-05, "loss": 0.4525442123413086, "memory(GiB)": 70.5, "step": 29140, "token_acc": 0.8904109589041096, "train_speed(iter/s)": 1.459023 }, { "epoch": 1.248661154192194, "grad_norm": 0.08371718227863312, "learning_rate": 8.538506822213136e-05, "loss": 0.2914890289306641, "memory(GiB)": 70.5, "step": 29145, "token_acc": 0.9456869009584664, "train_speed(iter/s)": 1.459044 }, { "epoch": 1.248875369521443, "grad_norm": 2.1332688331604004, "learning_rate": 8.538031323824967e-05, "loss": 0.4658027172088623, "memory(GiB)": 70.5, "step": 29150, "token_acc": 0.8775510204081632, "train_speed(iter/s)": 1.459063 }, { "epoch": 1.249089584850692, "grad_norm": 3.8882882595062256, "learning_rate": 8.537555761342039e-05, "loss": 0.42287864685058596, "memory(GiB)": 70.5, "step": 29155, "token_acc": 0.9051724137931034, "train_speed(iter/s)": 1.459055 }, { "epoch": 1.2493038001799408, "grad_norm": 3.24094295501709, "learning_rate": 8.537080134772973e-05, "loss": 0.7032517433166504, "memory(GiB)": 70.5, "step": 29160, "token_acc": 0.8562691131498471, "train_speed(iter/s)": 1.459076 }, { "epoch": 1.2495180155091898, "grad_norm": 6.988615989685059, "learning_rate": 8.536604444126382e-05, "loss": 0.8840079307556152, "memory(GiB)": 70.5, "step": 29165, "token_acc": 0.8611111111111112, "train_speed(iter/s)": 1.459161 }, { "epoch": 1.2497322308384389, "grad_norm": 4.241008758544922, "learning_rate": 8.536128689410882e-05, "loss": 0.6901953220367432, "memory(GiB)": 70.5, "step": 29170, "token_acc": 0.8466666666666667, "train_speed(iter/s)": 1.459172 }, { "epoch": 1.2499464461676877, "grad_norm": 1.4506499767303467, "learning_rate": 8.535652870635094e-05, "loss": 0.30791759490966797, "memory(GiB)": 70.5, "step": 29175, "token_acc": 0.9337349397590361, "train_speed(iter/s)": 1.459166 }, { "epoch": 1.2501606614969367, "grad_norm": 5.401695251464844, "learning_rate": 8.535176987807639e-05, "loss": 0.47107744216918945, "memory(GiB)": 70.5, "step": 29180, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.459175 }, { "epoch": 1.2503748768261858, "grad_norm": 6.9021477699279785, "learning_rate": 8.534701040937136e-05, "loss": 0.27175335884094237, "memory(GiB)": 70.5, "step": 29185, "token_acc": 0.9475806451612904, "train_speed(iter/s)": 1.459204 }, { "epoch": 1.2505890921554346, "grad_norm": 2.8939247131347656, "learning_rate": 8.534225030032208e-05, "loss": 0.26791355609893797, "memory(GiB)": 70.5, "step": 29190, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.459192 }, { "epoch": 1.2508033074846836, "grad_norm": 2.8367764949798584, "learning_rate": 8.533748955101477e-05, "loss": 0.6467954635620117, "memory(GiB)": 70.5, "step": 29195, "token_acc": 0.8658536585365854, "train_speed(iter/s)": 1.459215 }, { "epoch": 1.2510175228139326, "grad_norm": 1.7617906332015991, "learning_rate": 8.533272816153571e-05, "loss": 0.34679675102233887, "memory(GiB)": 70.5, "step": 29200, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.459229 }, { "epoch": 1.2512317381431814, "grad_norm": 4.4903106689453125, "learning_rate": 8.532796613197112e-05, "loss": 0.2919837474822998, "memory(GiB)": 70.5, "step": 29205, "token_acc": 0.9438596491228071, "train_speed(iter/s)": 1.45922 }, { "epoch": 1.2514459534724305, "grad_norm": 6.98383903503418, "learning_rate": 8.532320346240728e-05, "loss": 0.5498651981353759, "memory(GiB)": 70.5, "step": 29210, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.45922 }, { "epoch": 1.2516601688016795, "grad_norm": 2.8354265689849854, "learning_rate": 8.531844015293047e-05, "loss": 0.274582839012146, "memory(GiB)": 70.5, "step": 29215, "token_acc": 0.9261538461538461, "train_speed(iter/s)": 1.459236 }, { "epoch": 1.2518743841309283, "grad_norm": 4.92308235168457, "learning_rate": 8.531367620362699e-05, "loss": 0.343812894821167, "memory(GiB)": 70.5, "step": 29220, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.459236 }, { "epoch": 1.2520885994601774, "grad_norm": 4.177879810333252, "learning_rate": 8.530891161458314e-05, "loss": 0.3351206541061401, "memory(GiB)": 70.5, "step": 29225, "token_acc": 0.9297124600638977, "train_speed(iter/s)": 1.459243 }, { "epoch": 1.2523028147894264, "grad_norm": 4.229502201080322, "learning_rate": 8.530414638588525e-05, "loss": 0.4089400291442871, "memory(GiB)": 70.5, "step": 29230, "token_acc": 0.9240924092409241, "train_speed(iter/s)": 1.459257 }, { "epoch": 1.2525170301186752, "grad_norm": 2.7000114917755127, "learning_rate": 8.529938051761961e-05, "loss": 0.22326862812042236, "memory(GiB)": 70.5, "step": 29235, "token_acc": 0.952755905511811, "train_speed(iter/s)": 1.459249 }, { "epoch": 1.2527312454479242, "grad_norm": 6.276409149169922, "learning_rate": 8.529461400987258e-05, "loss": 0.639458703994751, "memory(GiB)": 70.5, "step": 29240, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.45928 }, { "epoch": 1.2529454607771733, "grad_norm": 2.294243574142456, "learning_rate": 8.528984686273051e-05, "loss": 0.4025286674499512, "memory(GiB)": 70.5, "step": 29245, "token_acc": 0.905511811023622, "train_speed(iter/s)": 1.459278 }, { "epoch": 1.253159676106422, "grad_norm": 2.1983466148376465, "learning_rate": 8.528507907627977e-05, "loss": 0.3566704988479614, "memory(GiB)": 70.5, "step": 29250, "token_acc": 0.9404761904761905, "train_speed(iter/s)": 1.459275 }, { "epoch": 1.2533738914356711, "grad_norm": 1.5105807781219482, "learning_rate": 8.528031065060669e-05, "loss": 0.4867263317108154, "memory(GiB)": 70.5, "step": 29255, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.4593 }, { "epoch": 1.2535881067649202, "grad_norm": 2.4161007404327393, "learning_rate": 8.527554158579772e-05, "loss": 0.4373313426971436, "memory(GiB)": 70.5, "step": 29260, "token_acc": 0.9075907590759076, "train_speed(iter/s)": 1.459312 }, { "epoch": 1.253802322094169, "grad_norm": 3.5348730087280273, "learning_rate": 8.527077188193921e-05, "loss": 0.3148198127746582, "memory(GiB)": 70.5, "step": 29265, "token_acc": 0.928, "train_speed(iter/s)": 1.459352 }, { "epoch": 1.254016537423418, "grad_norm": 7.50300407409668, "learning_rate": 8.52660015391176e-05, "loss": 0.23862810134887696, "memory(GiB)": 70.5, "step": 29270, "token_acc": 0.9414225941422594, "train_speed(iter/s)": 1.459362 }, { "epoch": 1.254230752752667, "grad_norm": 4.476847171783447, "learning_rate": 8.526123055741926e-05, "loss": 0.2558876991271973, "memory(GiB)": 70.5, "step": 29275, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.459352 }, { "epoch": 1.2544449680819159, "grad_norm": 6.159602165222168, "learning_rate": 8.525645893693067e-05, "loss": 0.25457963943481443, "memory(GiB)": 70.5, "step": 29280, "token_acc": 0.9605734767025089, "train_speed(iter/s)": 1.459385 }, { "epoch": 1.254659183411165, "grad_norm": 6.141406536102295, "learning_rate": 8.525168667773824e-05, "loss": 0.6891416549682617, "memory(GiB)": 70.5, "step": 29285, "token_acc": 0.855457227138643, "train_speed(iter/s)": 1.45938 }, { "epoch": 1.254873398740414, "grad_norm": 4.8232502937316895, "learning_rate": 8.524691377992844e-05, "loss": 0.6522525310516357, "memory(GiB)": 70.5, "step": 29290, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.459383 }, { "epoch": 1.2550876140696627, "grad_norm": 3.1136302947998047, "learning_rate": 8.524214024358775e-05, "loss": 0.6452937126159668, "memory(GiB)": 70.5, "step": 29295, "token_acc": 0.8687258687258688, "train_speed(iter/s)": 1.459397 }, { "epoch": 1.2553018293989118, "grad_norm": 0.552302360534668, "learning_rate": 8.523736606880261e-05, "loss": 0.24634690284729005, "memory(GiB)": 70.5, "step": 29300, "token_acc": 0.9548387096774194, "train_speed(iter/s)": 1.459409 }, { "epoch": 1.2555160447281608, "grad_norm": 1.2489570379257202, "learning_rate": 8.523259125565954e-05, "loss": 0.40707855224609374, "memory(GiB)": 70.5, "step": 29305, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.459434 }, { "epoch": 1.2557302600574096, "grad_norm": 5.957675933837891, "learning_rate": 8.522781580424502e-05, "loss": 0.4742734909057617, "memory(GiB)": 70.5, "step": 29310, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.459465 }, { "epoch": 1.2559444753866587, "grad_norm": 2.38187837600708, "learning_rate": 8.522303971464557e-05, "loss": 0.12131955623626708, "memory(GiB)": 70.5, "step": 29315, "token_acc": 0.9745454545454545, "train_speed(iter/s)": 1.459459 }, { "epoch": 1.2561586907159077, "grad_norm": 4.360453128814697, "learning_rate": 8.521826298694773e-05, "loss": 0.6178698539733887, "memory(GiB)": 70.5, "step": 29320, "token_acc": 0.8704318936877077, "train_speed(iter/s)": 1.459457 }, { "epoch": 1.2563729060451565, "grad_norm": 2.035855293273926, "learning_rate": 8.5213485621238e-05, "loss": 0.39890825748443604, "memory(GiB)": 70.5, "step": 29325, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.459453 }, { "epoch": 1.2565871213744055, "grad_norm": 4.895684242248535, "learning_rate": 8.520870761760296e-05, "loss": 0.6082334041595459, "memory(GiB)": 70.5, "step": 29330, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.459445 }, { "epoch": 1.2568013367036546, "grad_norm": 2.4869489669799805, "learning_rate": 8.520392897612913e-05, "loss": 0.29045541286468507, "memory(GiB)": 70.5, "step": 29335, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.45945 }, { "epoch": 1.2570155520329034, "grad_norm": 3.417677879333496, "learning_rate": 8.519914969690313e-05, "loss": 0.26786007881164553, "memory(GiB)": 70.5, "step": 29340, "token_acc": 0.9543859649122807, "train_speed(iter/s)": 1.459456 }, { "epoch": 1.2572297673621524, "grad_norm": 6.659523963928223, "learning_rate": 8.51943697800115e-05, "loss": 0.5537979125976562, "memory(GiB)": 70.5, "step": 29345, "token_acc": 0.8876404494382022, "train_speed(iter/s)": 1.459451 }, { "epoch": 1.2574439826914015, "grad_norm": 2.5123026371002197, "learning_rate": 8.518958922554085e-05, "loss": 0.26626837253570557, "memory(GiB)": 70.5, "step": 29350, "token_acc": 0.9395770392749244, "train_speed(iter/s)": 1.459445 }, { "epoch": 1.2576581980206503, "grad_norm": 3.2117135524749756, "learning_rate": 8.518480803357778e-05, "loss": 0.3281700611114502, "memory(GiB)": 70.5, "step": 29355, "token_acc": 0.9306122448979591, "train_speed(iter/s)": 1.459452 }, { "epoch": 1.2578724133498993, "grad_norm": 2.199904203414917, "learning_rate": 8.51800262042089e-05, "loss": 0.3753015995025635, "memory(GiB)": 70.5, "step": 29360, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.459501 }, { "epoch": 1.2580866286791483, "grad_norm": 4.344563961029053, "learning_rate": 8.517524373752083e-05, "loss": 0.6609188556671143, "memory(GiB)": 70.5, "step": 29365, "token_acc": 0.8461538461538461, "train_speed(iter/s)": 1.459503 }, { "epoch": 1.2583008440083971, "grad_norm": 6.746652603149414, "learning_rate": 8.517046063360026e-05, "loss": 0.5054292678833008, "memory(GiB)": 70.5, "step": 29370, "token_acc": 0.8825214899713467, "train_speed(iter/s)": 1.459522 }, { "epoch": 1.2585150593376462, "grad_norm": 2.8431644439697266, "learning_rate": 8.516567689253378e-05, "loss": 0.35375218391418456, "memory(GiB)": 70.5, "step": 29375, "token_acc": 0.916, "train_speed(iter/s)": 1.45955 }, { "epoch": 1.2587292746668952, "grad_norm": 2.975360155105591, "learning_rate": 8.516089251440809e-05, "loss": 0.24523301124572755, "memory(GiB)": 70.5, "step": 29380, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.459555 }, { "epoch": 1.258943489996144, "grad_norm": 2.5990257263183594, "learning_rate": 8.515610749930982e-05, "loss": 0.35393433570861815, "memory(GiB)": 70.5, "step": 29385, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.459554 }, { "epoch": 1.259157705325393, "grad_norm": 2.165531635284424, "learning_rate": 8.51513218473257e-05, "loss": 0.3198702812194824, "memory(GiB)": 70.5, "step": 29390, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.459566 }, { "epoch": 1.259371920654642, "grad_norm": 1.1080985069274902, "learning_rate": 8.514653555854242e-05, "loss": 0.2838580131530762, "memory(GiB)": 70.5, "step": 29395, "token_acc": 0.9425675675675675, "train_speed(iter/s)": 1.459566 }, { "epoch": 1.259586135983891, "grad_norm": 3.015970230102539, "learning_rate": 8.514174863304667e-05, "loss": 0.2974837064743042, "memory(GiB)": 70.5, "step": 29400, "token_acc": 0.9409937888198758, "train_speed(iter/s)": 1.459566 }, { "epoch": 1.25980035131314, "grad_norm": 7.240182876586914, "learning_rate": 8.513696107092517e-05, "loss": 0.5272724151611328, "memory(GiB)": 70.5, "step": 29405, "token_acc": 0.867109634551495, "train_speed(iter/s)": 1.459592 }, { "epoch": 1.260014566642389, "grad_norm": 3.594452142715454, "learning_rate": 8.513217287226466e-05, "loss": 0.2696810245513916, "memory(GiB)": 70.5, "step": 29410, "token_acc": 0.9465648854961832, "train_speed(iter/s)": 1.459592 }, { "epoch": 1.2602287819716378, "grad_norm": 6.8041462898254395, "learning_rate": 8.51273840371519e-05, "loss": 0.8743310928344726, "memory(GiB)": 70.5, "step": 29415, "token_acc": 0.8248175182481752, "train_speed(iter/s)": 1.459609 }, { "epoch": 1.2604429973008868, "grad_norm": 2.0887532234191895, "learning_rate": 8.512259456567362e-05, "loss": 0.26936898231506345, "memory(GiB)": 70.5, "step": 29420, "token_acc": 0.9103139013452914, "train_speed(iter/s)": 1.45961 }, { "epoch": 1.2606572126301359, "grad_norm": 3.885690450668335, "learning_rate": 8.511780445791659e-05, "loss": 0.26320586204528806, "memory(GiB)": 70.5, "step": 29425, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.459621 }, { "epoch": 1.2608714279593847, "grad_norm": 2.9421584606170654, "learning_rate": 8.511301371396758e-05, "loss": 0.305998420715332, "memory(GiB)": 70.5, "step": 29430, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.459617 }, { "epoch": 1.2610856432886337, "grad_norm": 7.47238826751709, "learning_rate": 8.510822233391338e-05, "loss": 0.21636428833007812, "memory(GiB)": 70.5, "step": 29435, "token_acc": 0.9566473988439307, "train_speed(iter/s)": 1.459618 }, { "epoch": 1.2612998586178827, "grad_norm": 2.947373390197754, "learning_rate": 8.510343031784082e-05, "loss": 0.4619601726531982, "memory(GiB)": 70.5, "step": 29440, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.45962 }, { "epoch": 1.2615140739471316, "grad_norm": 2.7564048767089844, "learning_rate": 8.509863766583668e-05, "loss": 0.4633053779602051, "memory(GiB)": 70.5, "step": 29445, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.459635 }, { "epoch": 1.2617282892763806, "grad_norm": 3.020763397216797, "learning_rate": 8.50938443779878e-05, "loss": 0.2540282249450684, "memory(GiB)": 70.5, "step": 29450, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.459651 }, { "epoch": 1.2619425046056296, "grad_norm": 4.042263031005859, "learning_rate": 8.5089050454381e-05, "loss": 0.6614347457885742, "memory(GiB)": 70.5, "step": 29455, "token_acc": 0.8635014836795252, "train_speed(iter/s)": 1.459655 }, { "epoch": 1.2621567199348784, "grad_norm": 0.20716750621795654, "learning_rate": 8.508425589510314e-05, "loss": 0.09188072085380554, "memory(GiB)": 70.5, "step": 29460, "token_acc": 0.9809885931558935, "train_speed(iter/s)": 1.459647 }, { "epoch": 1.2623709352641275, "grad_norm": 1.8272584676742554, "learning_rate": 8.507946070024108e-05, "loss": 0.23726961612701417, "memory(GiB)": 70.5, "step": 29465, "token_acc": 0.9336283185840708, "train_speed(iter/s)": 1.459646 }, { "epoch": 1.2625851505933765, "grad_norm": 4.413135051727295, "learning_rate": 8.507466486988168e-05, "loss": 0.41762652397155764, "memory(GiB)": 70.5, "step": 29470, "token_acc": 0.8984375, "train_speed(iter/s)": 1.459645 }, { "epoch": 1.2627993659226253, "grad_norm": 6.202095985412598, "learning_rate": 8.50698684041118e-05, "loss": 0.9201351165771484, "memory(GiB)": 70.5, "step": 29475, "token_acc": 0.8146718146718147, "train_speed(iter/s)": 1.459653 }, { "epoch": 1.2630135812518744, "grad_norm": 5.074501991271973, "learning_rate": 8.506507130301837e-05, "loss": 0.6145877838134766, "memory(GiB)": 70.5, "step": 29480, "token_acc": 0.8772455089820359, "train_speed(iter/s)": 1.459638 }, { "epoch": 1.2632277965811234, "grad_norm": 8.206456184387207, "learning_rate": 8.50602735666883e-05, "loss": 0.3873254060745239, "memory(GiB)": 70.5, "step": 29485, "token_acc": 0.9407114624505929, "train_speed(iter/s)": 1.459661 }, { "epoch": 1.2634420119103722, "grad_norm": 3.1469428539276123, "learning_rate": 8.505547519520845e-05, "loss": 0.3197284698486328, "memory(GiB)": 70.5, "step": 29490, "token_acc": 0.934640522875817, "train_speed(iter/s)": 1.459657 }, { "epoch": 1.2636562272396212, "grad_norm": 2.79191255569458, "learning_rate": 8.50506761886658e-05, "loss": 0.3960960865020752, "memory(GiB)": 70.5, "step": 29495, "token_acc": 0.9162995594713657, "train_speed(iter/s)": 1.459671 }, { "epoch": 1.2638704425688703, "grad_norm": 5.558347702026367, "learning_rate": 8.504587654714727e-05, "loss": 0.6469968318939209, "memory(GiB)": 70.5, "step": 29500, "token_acc": 0.8817204301075269, "train_speed(iter/s)": 1.459672 }, { "epoch": 1.2638704425688703, "eval_loss": 2.3853721618652344, "eval_runtime": 13.2949, "eval_samples_per_second": 7.522, "eval_steps_per_second": 7.522, "eval_token_acc": 0.46907993966817496, "step": 29500 }, { "epoch": 1.264084657898119, "grad_norm": 3.5258467197418213, "learning_rate": 8.50410762707398e-05, "loss": 0.21535353660583495, "memory(GiB)": 70.5, "step": 29505, "token_acc": 0.6139240506329114, "train_speed(iter/s)": 1.458634 }, { "epoch": 1.2642988732273681, "grad_norm": 3.423760414123535, "learning_rate": 8.503627535953039e-05, "loss": 0.4491128921508789, "memory(GiB)": 70.5, "step": 29510, "token_acc": 0.8981132075471698, "train_speed(iter/s)": 1.458635 }, { "epoch": 1.2645130885566171, "grad_norm": 3.715054988861084, "learning_rate": 8.503147381360595e-05, "loss": 0.42610626220703124, "memory(GiB)": 70.5, "step": 29515, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.458638 }, { "epoch": 1.264727303885866, "grad_norm": 5.445971488952637, "learning_rate": 8.502667163305353e-05, "loss": 0.42958989143371584, "memory(GiB)": 70.5, "step": 29520, "token_acc": 0.9198813056379822, "train_speed(iter/s)": 1.458642 }, { "epoch": 1.264941519215115, "grad_norm": 2.743868827819824, "learning_rate": 8.502186881796008e-05, "loss": 0.44974508285522463, "memory(GiB)": 70.5, "step": 29525, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.45864 }, { "epoch": 1.265155734544364, "grad_norm": 3.1829349994659424, "learning_rate": 8.501706536841263e-05, "loss": 0.601075267791748, "memory(GiB)": 70.5, "step": 29530, "token_acc": 0.88, "train_speed(iter/s)": 1.458657 }, { "epoch": 1.2653699498736128, "grad_norm": 3.351846694946289, "learning_rate": 8.501226128449818e-05, "loss": 0.5974397659301758, "memory(GiB)": 70.5, "step": 29535, "token_acc": 0.8676470588235294, "train_speed(iter/s)": 1.45865 }, { "epoch": 1.2655841652028619, "grad_norm": 7.076048851013184, "learning_rate": 8.500745656630379e-05, "loss": 0.4228181838989258, "memory(GiB)": 70.5, "step": 29540, "token_acc": 0.9136690647482014, "train_speed(iter/s)": 1.45864 }, { "epoch": 1.265798380532111, "grad_norm": 3.8224270343780518, "learning_rate": 8.500265121391649e-05, "loss": 0.19450316429138184, "memory(GiB)": 70.5, "step": 29545, "token_acc": 0.9505703422053232, "train_speed(iter/s)": 1.458642 }, { "epoch": 1.2660125958613597, "grad_norm": 1.7245527505874634, "learning_rate": 8.499784522742329e-05, "loss": 0.3539534568786621, "memory(GiB)": 70.5, "step": 29550, "token_acc": 0.9298780487804879, "train_speed(iter/s)": 1.45866 }, { "epoch": 1.2662268111906088, "grad_norm": 3.6926724910736084, "learning_rate": 8.499303860691132e-05, "loss": 0.4948283195495605, "memory(GiB)": 70.5, "step": 29555, "token_acc": 0.9006849315068494, "train_speed(iter/s)": 1.458676 }, { "epoch": 1.2664410265198578, "grad_norm": 1.224912166595459, "learning_rate": 8.498823135246763e-05, "loss": 0.5429718971252442, "memory(GiB)": 70.5, "step": 29560, "token_acc": 0.8782051282051282, "train_speed(iter/s)": 1.458678 }, { "epoch": 1.2666552418491066, "grad_norm": 1.8287054300308228, "learning_rate": 8.49834234641793e-05, "loss": 0.29946820735931395, "memory(GiB)": 70.5, "step": 29565, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.458682 }, { "epoch": 1.2668694571783556, "grad_norm": 3.513446569442749, "learning_rate": 8.497861494213344e-05, "loss": 0.5439561367034912, "memory(GiB)": 70.5, "step": 29570, "token_acc": 0.8914728682170543, "train_speed(iter/s)": 1.458692 }, { "epoch": 1.2670836725076047, "grad_norm": 4.542169094085693, "learning_rate": 8.497380578641717e-05, "loss": 0.4530160903930664, "memory(GiB)": 70.5, "step": 29575, "token_acc": 0.8931750741839762, "train_speed(iter/s)": 1.45869 }, { "epoch": 1.2672978878368535, "grad_norm": 4.6029372215271, "learning_rate": 8.496899599711759e-05, "loss": 0.4029345989227295, "memory(GiB)": 70.5, "step": 29580, "token_acc": 0.905511811023622, "train_speed(iter/s)": 1.458695 }, { "epoch": 1.2675121031661025, "grad_norm": 8.973332405090332, "learning_rate": 8.496418557432183e-05, "loss": 0.4158797264099121, "memory(GiB)": 70.5, "step": 29585, "token_acc": 0.9101796407185628, "train_speed(iter/s)": 1.458707 }, { "epoch": 1.2677263184953516, "grad_norm": 3.049168825149536, "learning_rate": 8.495937451811706e-05, "loss": 0.651026439666748, "memory(GiB)": 70.5, "step": 29590, "token_acc": 0.8719346049046321, "train_speed(iter/s)": 1.458701 }, { "epoch": 1.2679405338246004, "grad_norm": 2.0089340209960938, "learning_rate": 8.495456282859043e-05, "loss": 0.5116514205932617, "memory(GiB)": 70.5, "step": 29595, "token_acc": 0.8662420382165605, "train_speed(iter/s)": 1.458706 }, { "epoch": 1.2681547491538494, "grad_norm": 3.671867609024048, "learning_rate": 8.494975050582909e-05, "loss": 0.4380197525024414, "memory(GiB)": 70.5, "step": 29600, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.458713 }, { "epoch": 1.2683689644830984, "grad_norm": 4.57921838760376, "learning_rate": 8.494493754992026e-05, "loss": 0.5987775802612305, "memory(GiB)": 70.5, "step": 29605, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.458733 }, { "epoch": 1.2685831798123473, "grad_norm": 2.2865347862243652, "learning_rate": 8.49401239609511e-05, "loss": 0.334675931930542, "memory(GiB)": 70.5, "step": 29610, "token_acc": 0.925764192139738, "train_speed(iter/s)": 1.458734 }, { "epoch": 1.2687973951415963, "grad_norm": 1.6028703451156616, "learning_rate": 8.49353097390088e-05, "loss": 0.5188379764556885, "memory(GiB)": 70.5, "step": 29615, "token_acc": 0.9184397163120568, "train_speed(iter/s)": 1.458759 }, { "epoch": 1.2690116104708453, "grad_norm": 2.378448486328125, "learning_rate": 8.493049488418061e-05, "loss": 0.27522637844085696, "memory(GiB)": 70.5, "step": 29620, "token_acc": 0.9575971731448764, "train_speed(iter/s)": 1.458774 }, { "epoch": 1.2692258258000941, "grad_norm": 4.79465913772583, "learning_rate": 8.492567939655371e-05, "loss": 0.5007132530212403, "memory(GiB)": 70.5, "step": 29625, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.45879 }, { "epoch": 1.2694400411293432, "grad_norm": 5.266905784606934, "learning_rate": 8.49208632762154e-05, "loss": 0.33909246921539304, "memory(GiB)": 70.5, "step": 29630, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.458792 }, { "epoch": 1.2696542564585922, "grad_norm": 2.0880701541900635, "learning_rate": 8.491604652325287e-05, "loss": 0.34532716274261477, "memory(GiB)": 70.5, "step": 29635, "token_acc": 0.9190031152647975, "train_speed(iter/s)": 1.458796 }, { "epoch": 1.269868471787841, "grad_norm": 5.966197490692139, "learning_rate": 8.491122913775342e-05, "loss": 0.5256233215332031, "memory(GiB)": 70.5, "step": 29640, "token_acc": 0.9105691056910569, "train_speed(iter/s)": 1.458781 }, { "epoch": 1.27008268711709, "grad_norm": 3.2529380321502686, "learning_rate": 8.490641111980429e-05, "loss": 0.5330324172973633, "memory(GiB)": 70.5, "step": 29645, "token_acc": 0.8730769230769231, "train_speed(iter/s)": 1.458787 }, { "epoch": 1.270296902446339, "grad_norm": 5.165452003479004, "learning_rate": 8.490159246949278e-05, "loss": 0.5016083717346191, "memory(GiB)": 70.5, "step": 29650, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.458794 }, { "epoch": 1.270511117775588, "grad_norm": 1.6002442836761475, "learning_rate": 8.489677318690619e-05, "loss": 0.26929244995117185, "memory(GiB)": 70.5, "step": 29655, "token_acc": 0.944954128440367, "train_speed(iter/s)": 1.458804 }, { "epoch": 1.270725333104837, "grad_norm": 3.590688943862915, "learning_rate": 8.48919532721318e-05, "loss": 0.3608463525772095, "memory(GiB)": 70.5, "step": 29660, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.458811 }, { "epoch": 1.270939548434086, "grad_norm": 7.401162624359131, "learning_rate": 8.488713272525696e-05, "loss": 0.14221543073654175, "memory(GiB)": 70.5, "step": 29665, "token_acc": 0.9666666666666667, "train_speed(iter/s)": 1.458817 }, { "epoch": 1.2711537637633348, "grad_norm": 5.671383380889893, "learning_rate": 8.488231154636899e-05, "loss": 0.43341598510742185, "memory(GiB)": 70.5, "step": 29670, "token_acc": 0.9252336448598131, "train_speed(iter/s)": 1.458851 }, { "epoch": 1.2713679790925838, "grad_norm": 2.6080334186553955, "learning_rate": 8.487748973555523e-05, "loss": 0.20556674003601075, "memory(GiB)": 70.5, "step": 29675, "token_acc": 0.932, "train_speed(iter/s)": 1.458869 }, { "epoch": 1.2715821944218328, "grad_norm": 3.6386735439300537, "learning_rate": 8.487266729290299e-05, "loss": 0.3783675193786621, "memory(GiB)": 70.5, "step": 29680, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.458871 }, { "epoch": 1.2717964097510817, "grad_norm": 2.6982903480529785, "learning_rate": 8.48678442184997e-05, "loss": 0.1514630913734436, "memory(GiB)": 70.5, "step": 29685, "token_acc": 0.972, "train_speed(iter/s)": 1.458881 }, { "epoch": 1.2720106250803307, "grad_norm": 2.801332712173462, "learning_rate": 8.48630205124327e-05, "loss": 0.27791943550109866, "memory(GiB)": 70.5, "step": 29690, "token_acc": 0.9612903225806452, "train_speed(iter/s)": 1.458913 }, { "epoch": 1.2722248404095797, "grad_norm": 2.981699228286743, "learning_rate": 8.485819617478936e-05, "loss": 0.261728310585022, "memory(GiB)": 70.5, "step": 29695, "token_acc": 0.9352941176470588, "train_speed(iter/s)": 1.45891 }, { "epoch": 1.2724390557388285, "grad_norm": 4.557619571685791, "learning_rate": 8.485337120565712e-05, "loss": 0.5662907600402832, "memory(GiB)": 70.5, "step": 29700, "token_acc": 0.8851851851851852, "train_speed(iter/s)": 1.458912 }, { "epoch": 1.2726532710680776, "grad_norm": 8.436476707458496, "learning_rate": 8.484854560512335e-05, "loss": 0.7185813903808593, "memory(GiB)": 70.5, "step": 29705, "token_acc": 0.8618181818181818, "train_speed(iter/s)": 1.458912 }, { "epoch": 1.2728674863973266, "grad_norm": 2.366124391555786, "learning_rate": 8.48437193732755e-05, "loss": 0.16902703046798706, "memory(GiB)": 70.5, "step": 29710, "token_acc": 0.9700374531835206, "train_speed(iter/s)": 1.458925 }, { "epoch": 1.2730817017265754, "grad_norm": 5.958586692810059, "learning_rate": 8.483889251020097e-05, "loss": 0.7384626388549804, "memory(GiB)": 70.5, "step": 29715, "token_acc": 0.8576512455516014, "train_speed(iter/s)": 1.458924 }, { "epoch": 1.2732959170558245, "grad_norm": 3.1885979175567627, "learning_rate": 8.483406501598724e-05, "loss": 0.514025354385376, "memory(GiB)": 70.5, "step": 29720, "token_acc": 0.8934707903780069, "train_speed(iter/s)": 1.458914 }, { "epoch": 1.2735101323850735, "grad_norm": 5.713798999786377, "learning_rate": 8.482923689072173e-05, "loss": 0.383162784576416, "memory(GiB)": 70.5, "step": 29725, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.458913 }, { "epoch": 1.2737243477143223, "grad_norm": 3.1353511810302734, "learning_rate": 8.482440813449193e-05, "loss": 0.19593286514282227, "memory(GiB)": 70.5, "step": 29730, "token_acc": 0.9448818897637795, "train_speed(iter/s)": 1.458903 }, { "epoch": 1.2739385630435713, "grad_norm": 4.115242004394531, "learning_rate": 8.481957874738529e-05, "loss": 0.45981249809265134, "memory(GiB)": 70.5, "step": 29735, "token_acc": 0.8840125391849529, "train_speed(iter/s)": 1.458903 }, { "epoch": 1.2741527783728204, "grad_norm": 1.1208608150482178, "learning_rate": 8.481474872948933e-05, "loss": 0.3527961730957031, "memory(GiB)": 70.5, "step": 29740, "token_acc": 0.9316770186335404, "train_speed(iter/s)": 1.458906 }, { "epoch": 1.2743669937020692, "grad_norm": 6.6403489112854, "learning_rate": 8.480991808089156e-05, "loss": 0.49266824722290037, "memory(GiB)": 70.5, "step": 29745, "token_acc": 0.8881118881118881, "train_speed(iter/s)": 1.45893 }, { "epoch": 1.2745812090313182, "grad_norm": 0.20183557271957397, "learning_rate": 8.480508680167945e-05, "loss": 0.57802414894104, "memory(GiB)": 70.5, "step": 29750, "token_acc": 0.8660436137071651, "train_speed(iter/s)": 1.45893 }, { "epoch": 1.2747954243605673, "grad_norm": 7.588120460510254, "learning_rate": 8.480025489194055e-05, "loss": 0.29415862560272216, "memory(GiB)": 70.5, "step": 29755, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.458937 }, { "epoch": 1.275009639689816, "grad_norm": 0.6383906006813049, "learning_rate": 8.479542235176236e-05, "loss": 0.15490446090698243, "memory(GiB)": 70.5, "step": 29760, "token_acc": 0.9585798816568047, "train_speed(iter/s)": 1.458926 }, { "epoch": 1.275223855019065, "grad_norm": 4.747074127197266, "learning_rate": 8.479058918123248e-05, "loss": 0.473863410949707, "memory(GiB)": 70.5, "step": 29765, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.458957 }, { "epoch": 1.2754380703483141, "grad_norm": 2.325996160507202, "learning_rate": 8.478575538043843e-05, "loss": 0.4399857044219971, "memory(GiB)": 70.5, "step": 29770, "token_acc": 0.898989898989899, "train_speed(iter/s)": 1.458957 }, { "epoch": 1.2756522856775632, "grad_norm": 4.548655986785889, "learning_rate": 8.47809209494678e-05, "loss": 0.703969669342041, "memory(GiB)": 70.5, "step": 29775, "token_acc": 0.849112426035503, "train_speed(iter/s)": 1.459004 }, { "epoch": 1.275866501006812, "grad_norm": 2.850931406021118, "learning_rate": 8.477608588840815e-05, "loss": 0.5269589424133301, "memory(GiB)": 70.5, "step": 29780, "token_acc": 0.8808664259927798, "train_speed(iter/s)": 1.458993 }, { "epoch": 1.276080716336061, "grad_norm": 3.2139194011688232, "learning_rate": 8.477125019734709e-05, "loss": 0.4174196720123291, "memory(GiB)": 70.5, "step": 29785, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.458989 }, { "epoch": 1.27629493166531, "grad_norm": 1.4902491569519043, "learning_rate": 8.476641387637221e-05, "loss": 0.6477405548095703, "memory(GiB)": 70.5, "step": 29790, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.458985 }, { "epoch": 1.2765091469945589, "grad_norm": 1.6624939441680908, "learning_rate": 8.476157692557112e-05, "loss": 0.3559281349182129, "memory(GiB)": 70.5, "step": 29795, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.458986 }, { "epoch": 1.276723362323808, "grad_norm": 4.922333717346191, "learning_rate": 8.475673934503147e-05, "loss": 0.5401047706604004, "memory(GiB)": 70.5, "step": 29800, "token_acc": 0.8980891719745223, "train_speed(iter/s)": 1.45897 }, { "epoch": 1.276937577653057, "grad_norm": 1.4910491704940796, "learning_rate": 8.475190113484088e-05, "loss": 0.2811305999755859, "memory(GiB)": 70.5, "step": 29805, "token_acc": 0.940625, "train_speed(iter/s)": 1.458981 }, { "epoch": 1.2771517929823057, "grad_norm": 2.8214588165283203, "learning_rate": 8.4747062295087e-05, "loss": 0.5604674816131592, "memory(GiB)": 70.5, "step": 29810, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.459004 }, { "epoch": 1.2773660083115548, "grad_norm": 5.1869611740112305, "learning_rate": 8.474222282585749e-05, "loss": 0.6191688537597656, "memory(GiB)": 70.5, "step": 29815, "token_acc": 0.8615819209039548, "train_speed(iter/s)": 1.45899 }, { "epoch": 1.2775802236408038, "grad_norm": 2.96712064743042, "learning_rate": 8.473738272724001e-05, "loss": 0.6996460437774659, "memory(GiB)": 70.5, "step": 29820, "token_acc": 0.8691275167785235, "train_speed(iter/s)": 1.458984 }, { "epoch": 1.2777944389700526, "grad_norm": 3.6422829627990723, "learning_rate": 8.473254199932227e-05, "loss": 0.342163610458374, "memory(GiB)": 70.5, "step": 29825, "token_acc": 0.916030534351145, "train_speed(iter/s)": 1.458986 }, { "epoch": 1.2780086542993017, "grad_norm": 3.816073417663574, "learning_rate": 8.472770064219196e-05, "loss": 0.2063524007797241, "memory(GiB)": 70.5, "step": 29830, "token_acc": 0.9603658536585366, "train_speed(iter/s)": 1.458985 }, { "epoch": 1.2782228696285507, "grad_norm": 4.224072456359863, "learning_rate": 8.472285865593674e-05, "loss": 0.4125386714935303, "memory(GiB)": 70.5, "step": 29835, "token_acc": 0.9027237354085603, "train_speed(iter/s)": 1.458982 }, { "epoch": 1.2784370849577995, "grad_norm": 3.3599400520324707, "learning_rate": 8.47180160406444e-05, "loss": 0.5086872577667236, "memory(GiB)": 70.5, "step": 29840, "token_acc": 0.8866666666666667, "train_speed(iter/s)": 1.458983 }, { "epoch": 1.2786513002870485, "grad_norm": 1.791751503944397, "learning_rate": 8.471317279640261e-05, "loss": 0.29446992874145506, "memory(GiB)": 70.5, "step": 29845, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.458977 }, { "epoch": 1.2788655156162976, "grad_norm": 4.830417633056641, "learning_rate": 8.470832892329912e-05, "loss": 0.4301201820373535, "memory(GiB)": 70.5, "step": 29850, "token_acc": 0.9238095238095239, "train_speed(iter/s)": 1.458965 }, { "epoch": 1.2790797309455464, "grad_norm": 4.122400760650635, "learning_rate": 8.470348442142172e-05, "loss": 0.44718027114868164, "memory(GiB)": 70.5, "step": 29855, "token_acc": 0.9198717948717948, "train_speed(iter/s)": 1.458967 }, { "epoch": 1.2792939462747954, "grad_norm": 2.8930602073669434, "learning_rate": 8.469863929085813e-05, "loss": 0.5242234230041504, "memory(GiB)": 70.5, "step": 29860, "token_acc": 0.8766233766233766, "train_speed(iter/s)": 1.45894 }, { "epoch": 1.2795081616040445, "grad_norm": 2.5271313190460205, "learning_rate": 8.469379353169615e-05, "loss": 0.38395450115203855, "memory(GiB)": 70.5, "step": 29865, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.458968 }, { "epoch": 1.2797223769332933, "grad_norm": 6.374747276306152, "learning_rate": 8.468894714402356e-05, "loss": 0.6272805213928223, "memory(GiB)": 70.5, "step": 29870, "token_acc": 0.8759398496240601, "train_speed(iter/s)": 1.458989 }, { "epoch": 1.2799365922625423, "grad_norm": 2.678518533706665, "learning_rate": 8.468410012792813e-05, "loss": 0.29465909004211427, "memory(GiB)": 70.5, "step": 29875, "token_acc": 0.9375, "train_speed(iter/s)": 1.458976 }, { "epoch": 1.2801508075917913, "grad_norm": 7.466035842895508, "learning_rate": 8.467925248349771e-05, "loss": 0.2940552234649658, "memory(GiB)": 70.5, "step": 29880, "token_acc": 0.9396551724137931, "train_speed(iter/s)": 1.458969 }, { "epoch": 1.2803650229210402, "grad_norm": 1.88372004032135, "learning_rate": 8.467440421082008e-05, "loss": 0.34623758792877196, "memory(GiB)": 70.5, "step": 29885, "token_acc": 0.9250814332247557, "train_speed(iter/s)": 1.458958 }, { "epoch": 1.2805792382502892, "grad_norm": 4.124139785766602, "learning_rate": 8.466955530998311e-05, "loss": 0.2557192802429199, "memory(GiB)": 70.5, "step": 29890, "token_acc": 0.9300911854103343, "train_speed(iter/s)": 1.458944 }, { "epoch": 1.2807934535795382, "grad_norm": 0.5330995321273804, "learning_rate": 8.466470578107462e-05, "loss": 0.5040007591247558, "memory(GiB)": 70.5, "step": 29895, "token_acc": 0.8838709677419355, "train_speed(iter/s)": 1.458947 }, { "epoch": 1.281007668908787, "grad_norm": 6.731649398803711, "learning_rate": 8.465985562418244e-05, "loss": 0.501140022277832, "memory(GiB)": 70.5, "step": 29900, "token_acc": 0.9050632911392406, "train_speed(iter/s)": 1.458938 }, { "epoch": 1.281221884238036, "grad_norm": 3.6900737285614014, "learning_rate": 8.465500483939449e-05, "loss": 0.31390197277069093, "memory(GiB)": 70.5, "step": 29905, "token_acc": 0.9174917491749175, "train_speed(iter/s)": 1.458938 }, { "epoch": 1.281436099567285, "grad_norm": 2.3455448150634766, "learning_rate": 8.465015342679861e-05, "loss": 0.4726511001586914, "memory(GiB)": 70.5, "step": 29910, "token_acc": 0.9106628242074928, "train_speed(iter/s)": 1.458936 }, { "epoch": 1.281650314896534, "grad_norm": 10.730294227600098, "learning_rate": 8.46453013864827e-05, "loss": 0.4314899444580078, "memory(GiB)": 70.5, "step": 29915, "token_acc": 0.9012738853503185, "train_speed(iter/s)": 1.458986 }, { "epoch": 1.281864530225783, "grad_norm": 3.045428514480591, "learning_rate": 8.464044871853465e-05, "loss": 0.48629183769226075, "memory(GiB)": 70.5, "step": 29920, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.458984 }, { "epoch": 1.282078745555032, "grad_norm": 0.24822835624217987, "learning_rate": 8.463559542304238e-05, "loss": 0.42037043571472166, "memory(GiB)": 70.5, "step": 29925, "token_acc": 0.9244712990936556, "train_speed(iter/s)": 1.459011 }, { "epoch": 1.2822929608842808, "grad_norm": 10.605915069580078, "learning_rate": 8.46307415000938e-05, "loss": 0.5231534481048584, "memory(GiB)": 70.5, "step": 29930, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.459005 }, { "epoch": 1.2825071762135298, "grad_norm": 1.670084834098816, "learning_rate": 8.462588694977686e-05, "loss": 0.25115835666656494, "memory(GiB)": 70.5, "step": 29935, "token_acc": 0.9388489208633094, "train_speed(iter/s)": 1.459037 }, { "epoch": 1.2827213915427789, "grad_norm": 2.6144778728485107, "learning_rate": 8.46210317721795e-05, "loss": 0.20669689178466796, "memory(GiB)": 70.5, "step": 29940, "token_acc": 0.9465408805031447, "train_speed(iter/s)": 1.459029 }, { "epoch": 1.282935606872028, "grad_norm": 3.8312618732452393, "learning_rate": 8.461617596738967e-05, "loss": 0.5742983341217041, "memory(GiB)": 70.5, "step": 29945, "token_acc": 0.8757961783439491, "train_speed(iter/s)": 1.459027 }, { "epoch": 1.2831498222012767, "grad_norm": 0.6021116375923157, "learning_rate": 8.461131953549532e-05, "loss": 0.35267813205718995, "memory(GiB)": 70.5, "step": 29950, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.459036 }, { "epoch": 1.2833640375305257, "grad_norm": 5.433487415313721, "learning_rate": 8.460646247658446e-05, "loss": 0.6305763721466064, "memory(GiB)": 70.5, "step": 29955, "token_acc": 0.8698630136986302, "train_speed(iter/s)": 1.459029 }, { "epoch": 1.2835782528597748, "grad_norm": 9.005967140197754, "learning_rate": 8.460160479074508e-05, "loss": 0.5377079486846924, "memory(GiB)": 70.5, "step": 29960, "token_acc": 0.8582089552238806, "train_speed(iter/s)": 1.459054 }, { "epoch": 1.2837924681890236, "grad_norm": 4.968935489654541, "learning_rate": 8.459674647806517e-05, "loss": 0.3803140163421631, "memory(GiB)": 70.5, "step": 29965, "token_acc": 0.9211956521739131, "train_speed(iter/s)": 1.459051 }, { "epoch": 1.2840066835182726, "grad_norm": 2.7944939136505127, "learning_rate": 8.459188753863273e-05, "loss": 0.5167969226837158, "memory(GiB)": 70.5, "step": 29970, "token_acc": 0.9020979020979021, "train_speed(iter/s)": 1.459067 }, { "epoch": 1.2842208988475217, "grad_norm": 3.64066219329834, "learning_rate": 8.458702797253581e-05, "loss": 0.4003900051116943, "memory(GiB)": 70.5, "step": 29975, "token_acc": 0.946058091286307, "train_speed(iter/s)": 1.459084 }, { "epoch": 1.2844351141767705, "grad_norm": 1.681719183921814, "learning_rate": 8.458216777986243e-05, "loss": 0.3096207618713379, "memory(GiB)": 70.5, "step": 29980, "token_acc": 0.9173228346456693, "train_speed(iter/s)": 1.459109 }, { "epoch": 1.2846493295060195, "grad_norm": 1.9459717273712158, "learning_rate": 8.457730696070063e-05, "loss": 0.5179637908935547, "memory(GiB)": 70.5, "step": 29985, "token_acc": 0.9252336448598131, "train_speed(iter/s)": 1.459133 }, { "epoch": 1.2848635448352685, "grad_norm": 3.676410436630249, "learning_rate": 8.457244551513848e-05, "loss": 0.24379475116729737, "memory(GiB)": 70.5, "step": 29990, "token_acc": 0.9517543859649122, "train_speed(iter/s)": 1.459108 }, { "epoch": 1.2850777601645174, "grad_norm": 2.706596612930298, "learning_rate": 8.456758344326406e-05, "loss": 0.2934009552001953, "memory(GiB)": 70.5, "step": 29995, "token_acc": 0.9328358208955224, "train_speed(iter/s)": 1.459114 }, { "epoch": 1.2852919754937664, "grad_norm": 2.5565052032470703, "learning_rate": 8.456272074516542e-05, "loss": 0.17023004293441774, "memory(GiB)": 70.5, "step": 30000, "token_acc": 0.9574468085106383, "train_speed(iter/s)": 1.459114 }, { "epoch": 1.2852919754937664, "eval_loss": 2.5184383392333984, "eval_runtime": 12.996, "eval_samples_per_second": 7.695, "eval_steps_per_second": 7.695, "eval_token_acc": 0.40242261103633914, "step": 30000 }, { "epoch": 1.2855061908230154, "grad_norm": 3.9537088871002197, "learning_rate": 8.45578574209307e-05, "loss": 0.39363799095153806, "memory(GiB)": 70.5, "step": 30005, "token_acc": 0.5246727089627392, "train_speed(iter/s)": 1.458123 }, { "epoch": 1.2857204061522642, "grad_norm": 3.2739484310150146, "learning_rate": 8.455299347064795e-05, "loss": 0.23785266876220704, "memory(GiB)": 70.5, "step": 30010, "token_acc": 0.9471544715447154, "train_speed(iter/s)": 1.458125 }, { "epoch": 1.2859346214815133, "grad_norm": 6.409919261932373, "learning_rate": 8.454812889440533e-05, "loss": 0.4956367969512939, "memory(GiB)": 70.5, "step": 30015, "token_acc": 0.896797153024911, "train_speed(iter/s)": 1.458119 }, { "epoch": 1.2861488368107623, "grad_norm": 2.0644640922546387, "learning_rate": 8.454326369229094e-05, "loss": 0.28067281246185305, "memory(GiB)": 70.5, "step": 30020, "token_acc": 0.9502923976608187, "train_speed(iter/s)": 1.458127 }, { "epoch": 1.2863630521400111, "grad_norm": 2.044461250305176, "learning_rate": 8.45383978643929e-05, "loss": 0.2562739372253418, "memory(GiB)": 70.5, "step": 30025, "token_acc": 0.9518900343642611, "train_speed(iter/s)": 1.458132 }, { "epoch": 1.2865772674692602, "grad_norm": 2.226694345474243, "learning_rate": 8.453353141079941e-05, "loss": 0.3011181831359863, "memory(GiB)": 70.5, "step": 30030, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.458132 }, { "epoch": 1.2867914827985092, "grad_norm": 4.527383804321289, "learning_rate": 8.452866433159859e-05, "loss": 0.4662955284118652, "memory(GiB)": 70.5, "step": 30035, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.458135 }, { "epoch": 1.287005698127758, "grad_norm": 6.652449607849121, "learning_rate": 8.452379662687864e-05, "loss": 0.6231932640075684, "memory(GiB)": 70.5, "step": 30040, "token_acc": 0.8850174216027874, "train_speed(iter/s)": 1.458165 }, { "epoch": 1.287219913457007, "grad_norm": 4.222189426422119, "learning_rate": 8.45189282967277e-05, "loss": 0.6096114158630371, "memory(GiB)": 70.5, "step": 30045, "token_acc": 0.9039735099337748, "train_speed(iter/s)": 1.458189 }, { "epoch": 1.287434128786256, "grad_norm": 6.116891860961914, "learning_rate": 8.451405934123402e-05, "loss": 0.33343169689178465, "memory(GiB)": 70.5, "step": 30050, "token_acc": 0.9331306990881459, "train_speed(iter/s)": 1.458214 }, { "epoch": 1.2876483441155049, "grad_norm": 0.03571288660168648, "learning_rate": 8.450918976048577e-05, "loss": 0.3327286720275879, "memory(GiB)": 70.5, "step": 30055, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.458213 }, { "epoch": 1.287862559444754, "grad_norm": 2.856956720352173, "learning_rate": 8.450431955457118e-05, "loss": 0.340993595123291, "memory(GiB)": 70.5, "step": 30060, "token_acc": 0.9243697478991597, "train_speed(iter/s)": 1.458212 }, { "epoch": 1.288076774774003, "grad_norm": 3.268655776977539, "learning_rate": 8.449944872357845e-05, "loss": 0.26356987953186034, "memory(GiB)": 70.5, "step": 30065, "token_acc": 0.9510204081632653, "train_speed(iter/s)": 1.458214 }, { "epoch": 1.2882909901032518, "grad_norm": 0.598157525062561, "learning_rate": 8.449457726759586e-05, "loss": 0.392649245262146, "memory(GiB)": 70.5, "step": 30070, "token_acc": 0.9171974522292994, "train_speed(iter/s)": 1.458223 }, { "epoch": 1.2885052054325008, "grad_norm": 4.578696250915527, "learning_rate": 8.448970518671165e-05, "loss": 0.5073931694030762, "memory(GiB)": 70.5, "step": 30075, "token_acc": 0.8779661016949153, "train_speed(iter/s)": 1.458249 }, { "epoch": 1.2887194207617498, "grad_norm": 6.804533004760742, "learning_rate": 8.448483248101408e-05, "loss": 0.5093812942504883, "memory(GiB)": 70.5, "step": 30080, "token_acc": 0.8858024691358025, "train_speed(iter/s)": 1.458251 }, { "epoch": 1.2889336360909986, "grad_norm": 4.5196614265441895, "learning_rate": 8.447995915059142e-05, "loss": 0.5194606781005859, "memory(GiB)": 70.5, "step": 30085, "token_acc": 0.8903225806451613, "train_speed(iter/s)": 1.458251 }, { "epoch": 1.2891478514202477, "grad_norm": 3.8750336170196533, "learning_rate": 8.447508519553194e-05, "loss": 0.4613044261932373, "memory(GiB)": 70.5, "step": 30090, "token_acc": 0.8953846153846153, "train_speed(iter/s)": 1.458257 }, { "epoch": 1.2893620667494967, "grad_norm": 2.771576404571533, "learning_rate": 8.447021061592396e-05, "loss": 0.2743196725845337, "memory(GiB)": 70.5, "step": 30095, "token_acc": 0.9486404833836858, "train_speed(iter/s)": 1.458263 }, { "epoch": 1.2895762820787455, "grad_norm": 2.6836910247802734, "learning_rate": 8.446533541185577e-05, "loss": 0.265791916847229, "memory(GiB)": 70.5, "step": 30100, "token_acc": 0.9462809917355371, "train_speed(iter/s)": 1.458278 }, { "epoch": 1.2897904974079946, "grad_norm": 4.909562110900879, "learning_rate": 8.44604595834157e-05, "loss": 0.28582425117492677, "memory(GiB)": 70.5, "step": 30105, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.458274 }, { "epoch": 1.2900047127372436, "grad_norm": 2.0642929077148438, "learning_rate": 8.445558313069208e-05, "loss": 0.24840967655181884, "memory(GiB)": 70.5, "step": 30110, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.458303 }, { "epoch": 1.2902189280664924, "grad_norm": 2.361539125442505, "learning_rate": 8.445070605377326e-05, "loss": 0.5139287948608399, "memory(GiB)": 70.5, "step": 30115, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.458316 }, { "epoch": 1.2904331433957414, "grad_norm": 2.4001433849334717, "learning_rate": 8.444582835274758e-05, "loss": 0.23269104957580566, "memory(GiB)": 70.5, "step": 30120, "token_acc": 0.9559322033898305, "train_speed(iter/s)": 1.45833 }, { "epoch": 1.2906473587249905, "grad_norm": 3.326000452041626, "learning_rate": 8.444095002770341e-05, "loss": 0.3852069616317749, "memory(GiB)": 70.5, "step": 30125, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.45832 }, { "epoch": 1.2908615740542393, "grad_norm": 3.9654464721679688, "learning_rate": 8.44360710787291e-05, "loss": 0.5478158950805664, "memory(GiB)": 70.5, "step": 30130, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.458313 }, { "epoch": 1.2910757893834883, "grad_norm": 2.166776657104492, "learning_rate": 8.443119150591309e-05, "loss": 0.23938210010528566, "memory(GiB)": 70.5, "step": 30135, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.458313 }, { "epoch": 1.2912900047127374, "grad_norm": 4.00125789642334, "learning_rate": 8.442631130934372e-05, "loss": 0.3719663381576538, "memory(GiB)": 70.5, "step": 30140, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.458299 }, { "epoch": 1.2915042200419862, "grad_norm": 0.9174326062202454, "learning_rate": 8.442143048910944e-05, "loss": 0.20450005531311036, "memory(GiB)": 70.5, "step": 30145, "token_acc": 0.9629629629629629, "train_speed(iter/s)": 1.458304 }, { "epoch": 1.2917184353712352, "grad_norm": 1.8179670572280884, "learning_rate": 8.441654904529866e-05, "loss": 0.3477299690246582, "memory(GiB)": 70.5, "step": 30150, "token_acc": 0.9150943396226415, "train_speed(iter/s)": 1.458318 }, { "epoch": 1.2919326507004842, "grad_norm": 1.4222861528396606, "learning_rate": 8.441166697799981e-05, "loss": 0.3365692377090454, "memory(GiB)": 70.5, "step": 30155, "token_acc": 0.9418604651162791, "train_speed(iter/s)": 1.458317 }, { "epoch": 1.292146866029733, "grad_norm": 2.249558210372925, "learning_rate": 8.440678428730131e-05, "loss": 0.5029243469238281, "memory(GiB)": 70.5, "step": 30160, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.458338 }, { "epoch": 1.292361081358982, "grad_norm": 2.347201108932495, "learning_rate": 8.440190097329165e-05, "loss": 0.4322988510131836, "memory(GiB)": 70.5, "step": 30165, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.458344 }, { "epoch": 1.2925752966882311, "grad_norm": 6.547861099243164, "learning_rate": 8.439701703605929e-05, "loss": 0.5659078121185303, "memory(GiB)": 70.5, "step": 30170, "token_acc": 0.889261744966443, "train_speed(iter/s)": 1.458356 }, { "epoch": 1.29278951201748, "grad_norm": 2.9590468406677246, "learning_rate": 8.439213247569269e-05, "loss": 0.4748394012451172, "memory(GiB)": 70.5, "step": 30175, "token_acc": 0.9082278481012658, "train_speed(iter/s)": 1.458352 }, { "epoch": 1.293003727346729, "grad_norm": 2.951038122177124, "learning_rate": 8.438724729228034e-05, "loss": 0.41629915237426757, "memory(GiB)": 70.5, "step": 30180, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.458349 }, { "epoch": 1.293217942675978, "grad_norm": 6.607094764709473, "learning_rate": 8.438236148591076e-05, "loss": 0.30472211837768554, "memory(GiB)": 70.5, "step": 30185, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.458351 }, { "epoch": 1.2934321580052268, "grad_norm": 3.5958123207092285, "learning_rate": 8.437747505667243e-05, "loss": 0.4924447059631348, "memory(GiB)": 70.5, "step": 30190, "token_acc": 0.8975409836065574, "train_speed(iter/s)": 1.458384 }, { "epoch": 1.2936463733344759, "grad_norm": 1.4519730806350708, "learning_rate": 8.437258800465392e-05, "loss": 0.4160431385040283, "memory(GiB)": 70.5, "step": 30195, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.458385 }, { "epoch": 1.2938605886637249, "grad_norm": 2.4063892364501953, "learning_rate": 8.436770032994372e-05, "loss": 0.2876838207244873, "memory(GiB)": 70.5, "step": 30200, "token_acc": 0.9501779359430605, "train_speed(iter/s)": 1.45839 }, { "epoch": 1.2940748039929737, "grad_norm": 2.394120454788208, "learning_rate": 8.436281203263039e-05, "loss": 0.36110696792602537, "memory(GiB)": 70.5, "step": 30205, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.458396 }, { "epoch": 1.2942890193222227, "grad_norm": 5.880767822265625, "learning_rate": 8.435792311280248e-05, "loss": 0.49774909019470215, "memory(GiB)": 70.5, "step": 30210, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.458384 }, { "epoch": 1.2945032346514718, "grad_norm": 2.962679862976074, "learning_rate": 8.435303357054857e-05, "loss": 0.26512560844421384, "memory(GiB)": 70.5, "step": 30215, "token_acc": 0.9315589353612167, "train_speed(iter/s)": 1.458374 }, { "epoch": 1.2947174499807206, "grad_norm": 3.3766441345214844, "learning_rate": 8.434814340595722e-05, "loss": 0.335724925994873, "memory(GiB)": 70.5, "step": 30220, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.458372 }, { "epoch": 1.2949316653099696, "grad_norm": 3.475271463394165, "learning_rate": 8.434325261911705e-05, "loss": 0.428848934173584, "memory(GiB)": 70.5, "step": 30225, "token_acc": 0.9182879377431906, "train_speed(iter/s)": 1.458403 }, { "epoch": 1.2951458806392186, "grad_norm": 2.8398513793945312, "learning_rate": 8.433836121011662e-05, "loss": 0.457656192779541, "memory(GiB)": 70.5, "step": 30230, "token_acc": 0.9051094890510949, "train_speed(iter/s)": 1.458435 }, { "epoch": 1.2953600959684675, "grad_norm": 3.9846298694610596, "learning_rate": 8.43334691790446e-05, "loss": 0.24355754852294922, "memory(GiB)": 70.5, "step": 30235, "token_acc": 0.9453924914675768, "train_speed(iter/s)": 1.458438 }, { "epoch": 1.2955743112977165, "grad_norm": 1.7453004121780396, "learning_rate": 8.432857652598954e-05, "loss": 0.33660039901733396, "memory(GiB)": 70.5, "step": 30240, "token_acc": 0.9296875, "train_speed(iter/s)": 1.458468 }, { "epoch": 1.2957885266269655, "grad_norm": 2.7953476905822754, "learning_rate": 8.432368325104014e-05, "loss": 0.2706213712692261, "memory(GiB)": 70.5, "step": 30245, "token_acc": 0.946058091286307, "train_speed(iter/s)": 1.458453 }, { "epoch": 1.2960027419562143, "grad_norm": 2.485037088394165, "learning_rate": 8.4318789354285e-05, "loss": 0.5242500305175781, "memory(GiB)": 70.5, "step": 30250, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.458449 }, { "epoch": 1.2962169572854634, "grad_norm": 5.447127342224121, "learning_rate": 8.43138948358128e-05, "loss": 0.5402061462402343, "memory(GiB)": 70.5, "step": 30255, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.458481 }, { "epoch": 1.2964311726147124, "grad_norm": 4.394353866577148, "learning_rate": 8.43089996957122e-05, "loss": 0.4694994926452637, "memory(GiB)": 70.5, "step": 30260, "token_acc": 0.9045936395759717, "train_speed(iter/s)": 1.458484 }, { "epoch": 1.2966453879439612, "grad_norm": 6.264091968536377, "learning_rate": 8.43041039340719e-05, "loss": 0.5922162532806396, "memory(GiB)": 70.5, "step": 30265, "token_acc": 0.8694158075601375, "train_speed(iter/s)": 1.458504 }, { "epoch": 1.2968596032732103, "grad_norm": 2.6779918670654297, "learning_rate": 8.429920755098058e-05, "loss": 0.24519219398498535, "memory(GiB)": 70.5, "step": 30270, "token_acc": 0.9422492401215805, "train_speed(iter/s)": 1.458525 }, { "epoch": 1.2970738186024593, "grad_norm": 0.16881245374679565, "learning_rate": 8.429431054652695e-05, "loss": 0.3606379985809326, "memory(GiB)": 70.5, "step": 30275, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.45851 }, { "epoch": 1.297288033931708, "grad_norm": 5.221207618713379, "learning_rate": 8.428941292079968e-05, "loss": 0.3302126884460449, "memory(GiB)": 70.5, "step": 30280, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.458497 }, { "epoch": 1.2975022492609571, "grad_norm": 4.075576305389404, "learning_rate": 8.428451467388756e-05, "loss": 0.44627747535705564, "memory(GiB)": 70.5, "step": 30285, "token_acc": 0.8844984802431611, "train_speed(iter/s)": 1.458511 }, { "epoch": 1.2977164645902062, "grad_norm": 0.16002584993839264, "learning_rate": 8.427961580587927e-05, "loss": 0.2184974193572998, "memory(GiB)": 70.5, "step": 30290, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.458561 }, { "epoch": 1.297930679919455, "grad_norm": 4.323823928833008, "learning_rate": 8.427471631686359e-05, "loss": 0.728772783279419, "memory(GiB)": 70.5, "step": 30295, "token_acc": 0.8644688644688645, "train_speed(iter/s)": 1.458561 }, { "epoch": 1.298144895248704, "grad_norm": 4.32131290435791, "learning_rate": 8.426981620692929e-05, "loss": 0.42799954414367675, "memory(GiB)": 70.5, "step": 30300, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.458562 }, { "epoch": 1.298359110577953, "grad_norm": 3.029005289077759, "learning_rate": 8.426491547616511e-05, "loss": 0.42690377235412597, "memory(GiB)": 70.5, "step": 30305, "token_acc": 0.9023569023569024, "train_speed(iter/s)": 1.458559 }, { "epoch": 1.2985733259072019, "grad_norm": 3.5467302799224854, "learning_rate": 8.426001412465984e-05, "loss": 0.3418227195739746, "memory(GiB)": 70.5, "step": 30310, "token_acc": 0.9295302013422819, "train_speed(iter/s)": 1.458566 }, { "epoch": 1.298787541236451, "grad_norm": 2.7245285511016846, "learning_rate": 8.425511215250226e-05, "loss": 0.18051953315734864, "memory(GiB)": 70.5, "step": 30315, "token_acc": 0.9537366548042705, "train_speed(iter/s)": 1.458569 }, { "epoch": 1.2990017565657, "grad_norm": 3.9731051921844482, "learning_rate": 8.425020955978122e-05, "loss": 0.24331314563751222, "memory(GiB)": 70.5, "step": 30320, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.458566 }, { "epoch": 1.2992159718949488, "grad_norm": 2.7079808712005615, "learning_rate": 8.424530634658549e-05, "loss": 0.1700899839401245, "memory(GiB)": 70.5, "step": 30325, "token_acc": 0.9589552238805971, "train_speed(iter/s)": 1.458575 }, { "epoch": 1.2994301872241978, "grad_norm": 6.110294818878174, "learning_rate": 8.424040251300392e-05, "loss": 0.5049260139465332, "memory(GiB)": 70.5, "step": 30330, "token_acc": 0.8736462093862816, "train_speed(iter/s)": 1.458584 }, { "epoch": 1.2996444025534468, "grad_norm": 0.578482449054718, "learning_rate": 8.423549805912532e-05, "loss": 0.3013183832168579, "memory(GiB)": 70.5, "step": 30335, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.458582 }, { "epoch": 1.2998586178826956, "grad_norm": 2.968851327896118, "learning_rate": 8.423059298503857e-05, "loss": 0.4921926498413086, "memory(GiB)": 70.5, "step": 30340, "token_acc": 0.9054545454545454, "train_speed(iter/s)": 1.458584 }, { "epoch": 1.3000728332119447, "grad_norm": 3.3922643661499023, "learning_rate": 8.422568729083251e-05, "loss": 0.4118156433105469, "memory(GiB)": 70.5, "step": 30345, "token_acc": 0.9124293785310734, "train_speed(iter/s)": 1.458584 }, { "epoch": 1.3002870485411937, "grad_norm": 2.8108761310577393, "learning_rate": 8.422078097659602e-05, "loss": 0.35151607990264894, "memory(GiB)": 70.5, "step": 30350, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.458613 }, { "epoch": 1.3005012638704425, "grad_norm": 2.362638235092163, "learning_rate": 8.4215874042418e-05, "loss": 0.600164270401001, "memory(GiB)": 70.5, "step": 30355, "token_acc": 0.8787878787878788, "train_speed(iter/s)": 1.458643 }, { "epoch": 1.3007154791996915, "grad_norm": 4.584272384643555, "learning_rate": 8.421096648838728e-05, "loss": 0.4759012222290039, "memory(GiB)": 70.5, "step": 30360, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.458649 }, { "epoch": 1.3009296945289406, "grad_norm": 8.29997730255127, "learning_rate": 8.420605831459285e-05, "loss": 0.5823267936706543, "memory(GiB)": 70.5, "step": 30365, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.45868 }, { "epoch": 1.3011439098581894, "grad_norm": 2.0527522563934326, "learning_rate": 8.420114952112358e-05, "loss": 0.3540160655975342, "memory(GiB)": 70.5, "step": 30370, "token_acc": 0.9111969111969112, "train_speed(iter/s)": 1.45869 }, { "epoch": 1.3013581251874384, "grad_norm": 1.040458083152771, "learning_rate": 8.419624010806839e-05, "loss": 0.21936850547790526, "memory(GiB)": 70.5, "step": 30375, "token_acc": 0.9518900343642611, "train_speed(iter/s)": 1.4587 }, { "epoch": 1.3015723405166875, "grad_norm": 0.35792046785354614, "learning_rate": 8.419133007551626e-05, "loss": 0.5241628646850586, "memory(GiB)": 70.5, "step": 30380, "token_acc": 0.8543046357615894, "train_speed(iter/s)": 1.458686 }, { "epoch": 1.3017865558459363, "grad_norm": 5.423676013946533, "learning_rate": 8.418641942355608e-05, "loss": 0.5394115447998047, "memory(GiB)": 70.5, "step": 30385, "token_acc": 0.8915254237288136, "train_speed(iter/s)": 1.458677 }, { "epoch": 1.3020007711751853, "grad_norm": 1.758417010307312, "learning_rate": 8.418150815227686e-05, "loss": 0.4969788074493408, "memory(GiB)": 70.5, "step": 30390, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.458673 }, { "epoch": 1.3022149865044343, "grad_norm": 4.46324348449707, "learning_rate": 8.417659626176754e-05, "loss": 0.6770334720611573, "memory(GiB)": 70.5, "step": 30395, "token_acc": 0.841726618705036, "train_speed(iter/s)": 1.458694 }, { "epoch": 1.3024292018336832, "grad_norm": 1.4313844442367554, "learning_rate": 8.417168375211713e-05, "loss": 0.37269864082336424, "memory(GiB)": 70.5, "step": 30400, "token_acc": 0.9261538461538461, "train_speed(iter/s)": 1.458728 }, { "epoch": 1.3026434171629322, "grad_norm": 4.043055057525635, "learning_rate": 8.41667706234146e-05, "loss": 0.48621292114257814, "memory(GiB)": 70.5, "step": 30405, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.45875 }, { "epoch": 1.3028576324921812, "grad_norm": 3.5361063480377197, "learning_rate": 8.416185687574898e-05, "loss": 0.3554396629333496, "memory(GiB)": 70.5, "step": 30410, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.458758 }, { "epoch": 1.30307184782143, "grad_norm": 2.7981958389282227, "learning_rate": 8.415694250920927e-05, "loss": 0.47729835510253904, "memory(GiB)": 70.5, "step": 30415, "token_acc": 0.8735632183908046, "train_speed(iter/s)": 1.458744 }, { "epoch": 1.303286063150679, "grad_norm": 4.137937068939209, "learning_rate": 8.415202752388451e-05, "loss": 0.3707114696502686, "memory(GiB)": 70.5, "step": 30420, "token_acc": 0.9158576051779935, "train_speed(iter/s)": 1.458761 }, { "epoch": 1.303500278479928, "grad_norm": 7.761385917663574, "learning_rate": 8.414711191986374e-05, "loss": 0.6717432975769043, "memory(GiB)": 70.5, "step": 30425, "token_acc": 0.8634686346863468, "train_speed(iter/s)": 1.458764 }, { "epoch": 1.303714493809177, "grad_norm": 9.8212890625, "learning_rate": 8.4142195697236e-05, "loss": 0.46683692932128906, "memory(GiB)": 70.5, "step": 30430, "token_acc": 0.8717948717948718, "train_speed(iter/s)": 1.458753 }, { "epoch": 1.303928709138426, "grad_norm": 1.6040866374969482, "learning_rate": 8.413727885609036e-05, "loss": 0.19434595108032227, "memory(GiB)": 70.5, "step": 30435, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.458756 }, { "epoch": 1.304142924467675, "grad_norm": 2.940934181213379, "learning_rate": 8.413236139651588e-05, "loss": 0.6646674156188965, "memory(GiB)": 70.5, "step": 30440, "token_acc": 0.8469387755102041, "train_speed(iter/s)": 1.45878 }, { "epoch": 1.3043571397969238, "grad_norm": 4.450944900512695, "learning_rate": 8.412744331860167e-05, "loss": 0.5437392234802246, "memory(GiB)": 70.5, "step": 30445, "token_acc": 0.8673835125448028, "train_speed(iter/s)": 1.458785 }, { "epoch": 1.3045713551261728, "grad_norm": 2.9924278259277344, "learning_rate": 8.412252462243679e-05, "loss": 0.5618436336517334, "memory(GiB)": 70.5, "step": 30450, "token_acc": 0.8916083916083916, "train_speed(iter/s)": 1.458771 }, { "epoch": 1.3047855704554219, "grad_norm": 2.0216455459594727, "learning_rate": 8.411760530811038e-05, "loss": 0.16362122297286988, "memory(GiB)": 70.5, "step": 30455, "token_acc": 0.9553903345724907, "train_speed(iter/s)": 1.458767 }, { "epoch": 1.3049997857846707, "grad_norm": 4.945137023925781, "learning_rate": 8.411268537571154e-05, "loss": 0.47379121780395506, "memory(GiB)": 70.5, "step": 30460, "token_acc": 0.8765822784810127, "train_speed(iter/s)": 1.458793 }, { "epoch": 1.3052140011139197, "grad_norm": 4.6641058921813965, "learning_rate": 8.410776482532941e-05, "loss": 0.6435734748840332, "memory(GiB)": 70.5, "step": 30465, "token_acc": 0.8725761772853186, "train_speed(iter/s)": 1.458854 }, { "epoch": 1.3054282164431688, "grad_norm": 1.9916435480117798, "learning_rate": 8.410284365705311e-05, "loss": 0.32862350940704343, "memory(GiB)": 70.5, "step": 30470, "token_acc": 0.9191176470588235, "train_speed(iter/s)": 1.45885 }, { "epoch": 1.3056424317724176, "grad_norm": 1.024082064628601, "learning_rate": 8.409792187097183e-05, "loss": 0.3867784023284912, "memory(GiB)": 70.5, "step": 30475, "token_acc": 0.9161073825503355, "train_speed(iter/s)": 1.458853 }, { "epoch": 1.3058566471016666, "grad_norm": 3.5441548824310303, "learning_rate": 8.409299946717469e-05, "loss": 0.4745145797729492, "memory(GiB)": 70.5, "step": 30480, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.458836 }, { "epoch": 1.3060708624309156, "grad_norm": 6.967602729797363, "learning_rate": 8.408807644575089e-05, "loss": 0.4131251335144043, "memory(GiB)": 70.5, "step": 30485, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.458837 }, { "epoch": 1.3062850777601644, "grad_norm": 6.003180027008057, "learning_rate": 8.40831528067896e-05, "loss": 0.37790670394897463, "memory(GiB)": 70.5, "step": 30490, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.458831 }, { "epoch": 1.3064992930894135, "grad_norm": 2.4738855361938477, "learning_rate": 8.407822855038004e-05, "loss": 0.3080177068710327, "memory(GiB)": 70.5, "step": 30495, "token_acc": 0.9341317365269461, "train_speed(iter/s)": 1.458828 }, { "epoch": 1.3067135084186625, "grad_norm": 3.8551712036132812, "learning_rate": 8.40733036766114e-05, "loss": 0.2368237018585205, "memory(GiB)": 70.5, "step": 30500, "token_acc": 0.9625468164794008, "train_speed(iter/s)": 1.458815 }, { "epoch": 1.3067135084186625, "eval_loss": 2.7180280685424805, "eval_runtime": 13.2681, "eval_samples_per_second": 7.537, "eval_steps_per_second": 7.537, "eval_token_acc": 0.41127922971114167, "step": 30500 }, { "epoch": 1.3069277237479113, "grad_norm": 0.2290612757205963, "learning_rate": 8.406837818557289e-05, "loss": 0.195368230342865, "memory(GiB)": 70.5, "step": 30505, "token_acc": 0.584045584045584, "train_speed(iter/s)": 1.457825 }, { "epoch": 1.3071419390771604, "grad_norm": 5.452404975891113, "learning_rate": 8.406345207735375e-05, "loss": 0.22805421352386473, "memory(GiB)": 70.5, "step": 30510, "token_acc": 0.9498327759197325, "train_speed(iter/s)": 1.457826 }, { "epoch": 1.3073561544064094, "grad_norm": 1.7528338432312012, "learning_rate": 8.405852535204323e-05, "loss": 0.42523751258850095, "memory(GiB)": 70.5, "step": 30515, "token_acc": 0.9098039215686274, "train_speed(iter/s)": 1.457827 }, { "epoch": 1.3075703697356582, "grad_norm": 3.702613115310669, "learning_rate": 8.405359800973056e-05, "loss": 0.5885271072387696, "memory(GiB)": 70.5, "step": 30520, "token_acc": 0.8731884057971014, "train_speed(iter/s)": 1.457828 }, { "epoch": 1.3077845850649072, "grad_norm": 2.2801413536071777, "learning_rate": 8.404867005050504e-05, "loss": 0.47568812370300295, "memory(GiB)": 70.5, "step": 30525, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.457829 }, { "epoch": 1.3079988003941563, "grad_norm": 7.602977752685547, "learning_rate": 8.404374147445591e-05, "loss": 0.8110660552978516, "memory(GiB)": 70.5, "step": 30530, "token_acc": 0.8431372549019608, "train_speed(iter/s)": 1.457835 }, { "epoch": 1.308213015723405, "grad_norm": 4.43577241897583, "learning_rate": 8.403881228167245e-05, "loss": 0.3486222267150879, "memory(GiB)": 70.5, "step": 30535, "token_acc": 0.9429657794676806, "train_speed(iter/s)": 1.45788 }, { "epoch": 1.3084272310526541, "grad_norm": 7.934449672698975, "learning_rate": 8.4033882472244e-05, "loss": 0.21192939281463624, "memory(GiB)": 70.5, "step": 30540, "token_acc": 0.9577464788732394, "train_speed(iter/s)": 1.457885 }, { "epoch": 1.3086414463819032, "grad_norm": 3.385495662689209, "learning_rate": 8.402895204625982e-05, "loss": 0.5312220573425293, "memory(GiB)": 70.5, "step": 30545, "token_acc": 0.8971061093247589, "train_speed(iter/s)": 1.457896 }, { "epoch": 1.308855661711152, "grad_norm": 2.196964740753174, "learning_rate": 8.402402100380927e-05, "loss": 0.27678232192993163, "memory(GiB)": 70.5, "step": 30550, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.457896 }, { "epoch": 1.309069877040401, "grad_norm": 0.5141097903251648, "learning_rate": 8.401908934498166e-05, "loss": 0.2499398708343506, "memory(GiB)": 70.5, "step": 30555, "token_acc": 0.9358490566037736, "train_speed(iter/s)": 1.4579 }, { "epoch": 1.30928409236965, "grad_norm": 4.675086498260498, "learning_rate": 8.401415706986634e-05, "loss": 0.5774421691894531, "memory(GiB)": 70.5, "step": 30560, "token_acc": 0.8821428571428571, "train_speed(iter/s)": 1.457918 }, { "epoch": 1.3094983076988989, "grad_norm": 2.5438196659088135, "learning_rate": 8.400922417855265e-05, "loss": 0.32752408981323244, "memory(GiB)": 70.5, "step": 30565, "token_acc": 0.9488188976377953, "train_speed(iter/s)": 1.457912 }, { "epoch": 1.309712523028148, "grad_norm": 4.085822105407715, "learning_rate": 8.400429067112996e-05, "loss": 0.5730610847473144, "memory(GiB)": 70.5, "step": 30570, "token_acc": 0.9049295774647887, "train_speed(iter/s)": 1.457901 }, { "epoch": 1.309926738357397, "grad_norm": 2.1819255352020264, "learning_rate": 8.399935654768765e-05, "loss": 0.33213019371032715, "memory(GiB)": 70.5, "step": 30575, "token_acc": 0.9335180055401662, "train_speed(iter/s)": 1.457908 }, { "epoch": 1.3101409536866457, "grad_norm": 1.7883703708648682, "learning_rate": 8.399442180831509e-05, "loss": 0.22753021717071534, "memory(GiB)": 70.5, "step": 30580, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.457909 }, { "epoch": 1.3103551690158948, "grad_norm": 5.068688869476318, "learning_rate": 8.398948645310169e-05, "loss": 0.3215725183486938, "memory(GiB)": 70.5, "step": 30585, "token_acc": 0.948905109489051, "train_speed(iter/s)": 1.457912 }, { "epoch": 1.3105693843451438, "grad_norm": 1.910747766494751, "learning_rate": 8.398455048213687e-05, "loss": 0.46162939071655273, "memory(GiB)": 70.5, "step": 30590, "token_acc": 0.92, "train_speed(iter/s)": 1.457904 }, { "epoch": 1.3107835996743926, "grad_norm": 2.113567590713501, "learning_rate": 8.397961389551003e-05, "loss": 0.1303043007850647, "memory(GiB)": 70.5, "step": 30595, "token_acc": 0.9623824451410659, "train_speed(iter/s)": 1.457897 }, { "epoch": 1.3109978150036417, "grad_norm": 1.9809762239456177, "learning_rate": 8.39746766933106e-05, "loss": 0.4700284004211426, "memory(GiB)": 70.5, "step": 30600, "token_acc": 0.9323943661971831, "train_speed(iter/s)": 1.4579 }, { "epoch": 1.3112120303328907, "grad_norm": 2.181337833404541, "learning_rate": 8.396973887562806e-05, "loss": 0.5080564498901368, "memory(GiB)": 70.5, "step": 30605, "token_acc": 0.8847457627118644, "train_speed(iter/s)": 1.457922 }, { "epoch": 1.3114262456621395, "grad_norm": 2.1526846885681152, "learning_rate": 8.396480044255181e-05, "loss": 0.3235430955886841, "memory(GiB)": 70.5, "step": 30610, "token_acc": 0.9202453987730062, "train_speed(iter/s)": 1.457927 }, { "epoch": 1.3116404609913885, "grad_norm": 10.975157737731934, "learning_rate": 8.395986139417135e-05, "loss": 0.5667816638946533, "memory(GiB)": 70.5, "step": 30615, "token_acc": 0.8485804416403786, "train_speed(iter/s)": 1.457949 }, { "epoch": 1.3118546763206376, "grad_norm": 5.004209995269775, "learning_rate": 8.395492173057613e-05, "loss": 0.41092872619628906, "memory(GiB)": 70.5, "step": 30620, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.457937 }, { "epoch": 1.3120688916498864, "grad_norm": 2.533482551574707, "learning_rate": 8.394998145185566e-05, "loss": 0.4664000511169434, "memory(GiB)": 70.5, "step": 30625, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.457948 }, { "epoch": 1.3122831069791354, "grad_norm": 7.046419143676758, "learning_rate": 8.394504055809944e-05, "loss": 0.6424255847930909, "memory(GiB)": 70.5, "step": 30630, "token_acc": 0.8897058823529411, "train_speed(iter/s)": 1.457953 }, { "epoch": 1.3124973223083845, "grad_norm": 3.992307186126709, "learning_rate": 8.394009904939695e-05, "loss": 0.3167220115661621, "memory(GiB)": 70.5, "step": 30635, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.457959 }, { "epoch": 1.3127115376376333, "grad_norm": 5.89705753326416, "learning_rate": 8.393515692583774e-05, "loss": 0.5647421836853027, "memory(GiB)": 70.5, "step": 30640, "token_acc": 0.8943661971830986, "train_speed(iter/s)": 1.457954 }, { "epoch": 1.3129257529668823, "grad_norm": 3.079371213912964, "learning_rate": 8.393021418751132e-05, "loss": 0.43929319381713866, "memory(GiB)": 70.5, "step": 30645, "token_acc": 0.9025974025974026, "train_speed(iter/s)": 1.457967 }, { "epoch": 1.3131399682961313, "grad_norm": 2.573155641555786, "learning_rate": 8.392527083450723e-05, "loss": 0.5921721458435059, "memory(GiB)": 70.5, "step": 30650, "token_acc": 0.8774193548387097, "train_speed(iter/s)": 1.457951 }, { "epoch": 1.3133541836253801, "grad_norm": 1.6000055074691772, "learning_rate": 8.392032686691504e-05, "loss": 0.44088406562805177, "memory(GiB)": 70.5, "step": 30655, "token_acc": 0.9060402684563759, "train_speed(iter/s)": 1.457973 }, { "epoch": 1.3135683989546292, "grad_norm": 2.0228655338287354, "learning_rate": 8.391538228482432e-05, "loss": 0.18300763368606568, "memory(GiB)": 70.5, "step": 30660, "token_acc": 0.9574468085106383, "train_speed(iter/s)": 1.457963 }, { "epoch": 1.3137826142838782, "grad_norm": 0.5937181115150452, "learning_rate": 8.391043708832463e-05, "loss": 0.38955249786376955, "memory(GiB)": 70.5, "step": 30665, "token_acc": 0.9241379310344827, "train_speed(iter/s)": 1.457986 }, { "epoch": 1.313996829613127, "grad_norm": 1.7209419012069702, "learning_rate": 8.390549127750558e-05, "loss": 0.4701199531555176, "memory(GiB)": 70.5, "step": 30670, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.45798 }, { "epoch": 1.314211044942376, "grad_norm": 1.6279237270355225, "learning_rate": 8.390054485245672e-05, "loss": 0.33775112628936765, "memory(GiB)": 70.5, "step": 30675, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.457975 }, { "epoch": 1.314425260271625, "grad_norm": 4.888917446136475, "learning_rate": 8.38955978132677e-05, "loss": 0.7157889366149902, "memory(GiB)": 70.5, "step": 30680, "token_acc": 0.8713826366559485, "train_speed(iter/s)": 1.457961 }, { "epoch": 1.314639475600874, "grad_norm": 2.5322399139404297, "learning_rate": 8.389065016002812e-05, "loss": 0.8392210960388183, "memory(GiB)": 70.5, "step": 30685, "token_acc": 0.8566775244299675, "train_speed(iter/s)": 1.457963 }, { "epoch": 1.314853690930123, "grad_norm": 3.339062213897705, "learning_rate": 8.388570189282765e-05, "loss": 0.3544142246246338, "memory(GiB)": 70.5, "step": 30690, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.457971 }, { "epoch": 1.315067906259372, "grad_norm": 1.5267716646194458, "learning_rate": 8.388075301175586e-05, "loss": 0.2738867521286011, "memory(GiB)": 70.5, "step": 30695, "token_acc": 0.9377431906614786, "train_speed(iter/s)": 1.458011 }, { "epoch": 1.3152821215886208, "grad_norm": 1.2500563859939575, "learning_rate": 8.387580351690248e-05, "loss": 0.35848250389099123, "memory(GiB)": 70.5, "step": 30700, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.45804 }, { "epoch": 1.3154963369178698, "grad_norm": 6.868077754974365, "learning_rate": 8.387085340835713e-05, "loss": 0.4126345157623291, "memory(GiB)": 70.5, "step": 30705, "token_acc": 0.8922413793103449, "train_speed(iter/s)": 1.458053 }, { "epoch": 1.3157105522471189, "grad_norm": 1.7125130891799927, "learning_rate": 8.38659026862095e-05, "loss": 0.3762622356414795, "memory(GiB)": 70.5, "step": 30710, "token_acc": 0.9325513196480938, "train_speed(iter/s)": 1.458057 }, { "epoch": 1.3159247675763677, "grad_norm": 2.1199522018432617, "learning_rate": 8.386095135054927e-05, "loss": 0.4955745220184326, "memory(GiB)": 70.5, "step": 30715, "token_acc": 0.8963210702341137, "train_speed(iter/s)": 1.458074 }, { "epoch": 1.3161389829056167, "grad_norm": 2.2059290409088135, "learning_rate": 8.385599940146613e-05, "loss": 0.4038558483123779, "memory(GiB)": 70.5, "step": 30720, "token_acc": 0.9267399267399268, "train_speed(iter/s)": 1.458072 }, { "epoch": 1.3163531982348657, "grad_norm": 2.4012398719787598, "learning_rate": 8.385104683904983e-05, "loss": 0.40550832748413085, "memory(GiB)": 70.5, "step": 30725, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.458067 }, { "epoch": 1.3165674135641146, "grad_norm": 1.8482091426849365, "learning_rate": 8.384609366339003e-05, "loss": 0.2018254280090332, "memory(GiB)": 70.5, "step": 30730, "token_acc": 0.9563492063492064, "train_speed(iter/s)": 1.458066 }, { "epoch": 1.3167816288933636, "grad_norm": 2.3505613803863525, "learning_rate": 8.38411398745765e-05, "loss": 0.33086423873901366, "memory(GiB)": 70.5, "step": 30735, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.458076 }, { "epoch": 1.3169958442226126, "grad_norm": 4.5795440673828125, "learning_rate": 8.383618547269901e-05, "loss": 0.6249231338500977, "memory(GiB)": 70.5, "step": 30740, "token_acc": 0.8536585365853658, "train_speed(iter/s)": 1.458099 }, { "epoch": 1.3172100595518614, "grad_norm": 1.5227638483047485, "learning_rate": 8.383123045784724e-05, "loss": 0.3347670555114746, "memory(GiB)": 70.5, "step": 30745, "token_acc": 0.9343283582089552, "train_speed(iter/s)": 1.458099 }, { "epoch": 1.3174242748811105, "grad_norm": 0.68864506483078, "learning_rate": 8.382627483011101e-05, "loss": 0.32859828472137453, "memory(GiB)": 70.5, "step": 30750, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.458115 }, { "epoch": 1.3176384902103595, "grad_norm": 3.5946247577667236, "learning_rate": 8.382131858958009e-05, "loss": 0.3848949670791626, "memory(GiB)": 70.5, "step": 30755, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.458131 }, { "epoch": 1.3178527055396083, "grad_norm": 3.2810187339782715, "learning_rate": 8.381636173634425e-05, "loss": 0.6838945388793946, "memory(GiB)": 70.5, "step": 30760, "token_acc": 0.8646153846153846, "train_speed(iter/s)": 1.458153 }, { "epoch": 1.3180669208688574, "grad_norm": 5.365682125091553, "learning_rate": 8.381140427049328e-05, "loss": 0.4156081199645996, "memory(GiB)": 70.5, "step": 30765, "token_acc": 0.9196141479099679, "train_speed(iter/s)": 1.458146 }, { "epoch": 1.3182811361981064, "grad_norm": 2.449557065963745, "learning_rate": 8.380644619211702e-05, "loss": 0.3608639478683472, "memory(GiB)": 70.5, "step": 30770, "token_acc": 0.9375, "train_speed(iter/s)": 1.458139 }, { "epoch": 1.3184953515273552, "grad_norm": 4.697730541229248, "learning_rate": 8.380148750130525e-05, "loss": 0.2753224611282349, "memory(GiB)": 70.5, "step": 30775, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.458142 }, { "epoch": 1.3187095668566042, "grad_norm": 4.082615375518799, "learning_rate": 8.379652819814785e-05, "loss": 0.26561665534973145, "memory(GiB)": 70.5, "step": 30780, "token_acc": 0.9331103678929766, "train_speed(iter/s)": 1.458149 }, { "epoch": 1.3189237821858533, "grad_norm": 3.2127087116241455, "learning_rate": 8.379156828273463e-05, "loss": 0.2877392530441284, "memory(GiB)": 70.5, "step": 30785, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.458139 }, { "epoch": 1.319137997515102, "grad_norm": 4.3650383949279785, "learning_rate": 8.378660775515546e-05, "loss": 0.5391834735870361, "memory(GiB)": 70.5, "step": 30790, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.458163 }, { "epoch": 1.3193522128443511, "grad_norm": 5.907803058624268, "learning_rate": 8.378164661550019e-05, "loss": 0.4515474796295166, "memory(GiB)": 70.5, "step": 30795, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.458186 }, { "epoch": 1.3195664281736001, "grad_norm": 5.914632797241211, "learning_rate": 8.37766848638587e-05, "loss": 0.6256268978118896, "memory(GiB)": 70.5, "step": 30800, "token_acc": 0.8675496688741722, "train_speed(iter/s)": 1.45819 }, { "epoch": 1.319780643502849, "grad_norm": 6.9973955154418945, "learning_rate": 8.377172250032086e-05, "loss": 0.7283639430999755, "memory(GiB)": 70.5, "step": 30805, "token_acc": 0.8657243816254417, "train_speed(iter/s)": 1.45817 }, { "epoch": 1.319994858832098, "grad_norm": 1.9494221210479736, "learning_rate": 8.376675952497661e-05, "loss": 0.4046614646911621, "memory(GiB)": 70.5, "step": 30810, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.458178 }, { "epoch": 1.320209074161347, "grad_norm": 1.9118719100952148, "learning_rate": 8.376179593791585e-05, "loss": 0.33115034103393554, "memory(GiB)": 70.5, "step": 30815, "token_acc": 0.9313304721030042, "train_speed(iter/s)": 1.458214 }, { "epoch": 1.3204232894905958, "grad_norm": 8.964221000671387, "learning_rate": 8.375683173922846e-05, "loss": 0.3202383995056152, "memory(GiB)": 70.5, "step": 30820, "token_acc": 0.9240506329113924, "train_speed(iter/s)": 1.458226 }, { "epoch": 1.3206375048198449, "grad_norm": 3.874807357788086, "learning_rate": 8.37518669290044e-05, "loss": 0.5685357093811035, "memory(GiB)": 70.5, "step": 30825, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.458221 }, { "epoch": 1.320851720149094, "grad_norm": 5.041698932647705, "learning_rate": 8.374690150733362e-05, "loss": 0.22869770526885985, "memory(GiB)": 70.5, "step": 30830, "token_acc": 0.949685534591195, "train_speed(iter/s)": 1.458239 }, { "epoch": 1.3210659354783427, "grad_norm": 5.513265609741211, "learning_rate": 8.374193547430605e-05, "loss": 0.46187586784362794, "memory(GiB)": 70.5, "step": 30835, "token_acc": 0.8902439024390244, "train_speed(iter/s)": 1.458242 }, { "epoch": 1.3212801508075918, "grad_norm": 0.31715497374534607, "learning_rate": 8.373696883001166e-05, "loss": 0.32266685962677, "memory(GiB)": 70.5, "step": 30840, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.458242 }, { "epoch": 1.3214943661368408, "grad_norm": 4.98422384262085, "learning_rate": 8.373200157454044e-05, "loss": 0.4395422458648682, "memory(GiB)": 70.5, "step": 30845, "token_acc": 0.92, "train_speed(iter/s)": 1.458246 }, { "epoch": 1.3217085814660896, "grad_norm": 1.3079220056533813, "learning_rate": 8.372703370798239e-05, "loss": 0.4414348125457764, "memory(GiB)": 70.5, "step": 30850, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.458257 }, { "epoch": 1.3219227967953386, "grad_norm": 2.9921865463256836, "learning_rate": 8.372206523042746e-05, "loss": 0.34229743480682373, "memory(GiB)": 70.5, "step": 30855, "token_acc": 0.9297124600638977, "train_speed(iter/s)": 1.458264 }, { "epoch": 1.3221370121245877, "grad_norm": 4.221546649932861, "learning_rate": 8.371709614196568e-05, "loss": 0.4637269973754883, "memory(GiB)": 70.5, "step": 30860, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.458256 }, { "epoch": 1.3223512274538365, "grad_norm": 3.2667622566223145, "learning_rate": 8.371212644268709e-05, "loss": 0.5565732479095459, "memory(GiB)": 70.5, "step": 30865, "token_acc": 0.8687943262411347, "train_speed(iter/s)": 1.458255 }, { "epoch": 1.3225654427830855, "grad_norm": 5.440427303314209, "learning_rate": 8.370715613268171e-05, "loss": 0.31175923347473145, "memory(GiB)": 70.5, "step": 30870, "token_acc": 0.9221311475409836, "train_speed(iter/s)": 1.458262 }, { "epoch": 1.3227796581123346, "grad_norm": 2.585902452468872, "learning_rate": 8.370218521203956e-05, "loss": 0.3628865718841553, "memory(GiB)": 70.5, "step": 30875, "token_acc": 0.9360902255639098, "train_speed(iter/s)": 1.458277 }, { "epoch": 1.3229938734415834, "grad_norm": 0.5103923082351685, "learning_rate": 8.369721368085072e-05, "loss": 0.2007852554321289, "memory(GiB)": 70.5, "step": 30880, "token_acc": 0.9440559440559441, "train_speed(iter/s)": 1.458283 }, { "epoch": 1.3232080887708324, "grad_norm": 1.8377292156219482, "learning_rate": 8.369224153920523e-05, "loss": 0.4679257869720459, "memory(GiB)": 70.5, "step": 30885, "token_acc": 0.9049429657794676, "train_speed(iter/s)": 1.458338 }, { "epoch": 1.3234223041000814, "grad_norm": 3.2367069721221924, "learning_rate": 8.36872687871932e-05, "loss": 0.3671630859375, "memory(GiB)": 70.5, "step": 30890, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.458343 }, { "epoch": 1.3236365194293302, "grad_norm": 0.22071129083633423, "learning_rate": 8.368229542490468e-05, "loss": 0.38025579452514646, "memory(GiB)": 70.5, "step": 30895, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.458348 }, { "epoch": 1.3238507347585793, "grad_norm": 2.7525722980499268, "learning_rate": 8.36773214524298e-05, "loss": 0.38388690948486326, "memory(GiB)": 70.5, "step": 30900, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.45837 }, { "epoch": 1.3240649500878283, "grad_norm": 9.98099422454834, "learning_rate": 8.367234686985862e-05, "loss": 0.4624149322509766, "memory(GiB)": 70.5, "step": 30905, "token_acc": 0.9053030303030303, "train_speed(iter/s)": 1.458398 }, { "epoch": 1.3242791654170771, "grad_norm": 6.414818286895752, "learning_rate": 8.366737167728132e-05, "loss": 0.49415788650512693, "memory(GiB)": 70.5, "step": 30910, "token_acc": 0.8826979472140762, "train_speed(iter/s)": 1.458392 }, { "epoch": 1.3244933807463262, "grad_norm": 0.5998333096504211, "learning_rate": 8.366239587478797e-05, "loss": 0.31247615814208984, "memory(GiB)": 70.5, "step": 30915, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.458396 }, { "epoch": 1.3247075960755752, "grad_norm": 1.994537115097046, "learning_rate": 8.365741946246875e-05, "loss": 0.3893040657043457, "memory(GiB)": 70.5, "step": 30920, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.458403 }, { "epoch": 1.324921811404824, "grad_norm": 5.3230719566345215, "learning_rate": 8.365244244041382e-05, "loss": 0.6933044910430908, "memory(GiB)": 70.5, "step": 30925, "token_acc": 0.8705501618122977, "train_speed(iter/s)": 1.458435 }, { "epoch": 1.325136026734073, "grad_norm": 6.182384490966797, "learning_rate": 8.364746480871331e-05, "loss": 0.31999707221984863, "memory(GiB)": 70.5, "step": 30930, "token_acc": 0.9457364341085271, "train_speed(iter/s)": 1.458433 }, { "epoch": 1.325350242063322, "grad_norm": 2.2937557697296143, "learning_rate": 8.364248656745741e-05, "loss": 0.4266510963439941, "memory(GiB)": 70.5, "step": 30935, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.458434 }, { "epoch": 1.325564457392571, "grad_norm": 3.4184892177581787, "learning_rate": 8.36375077167363e-05, "loss": 0.29435684680938723, "memory(GiB)": 70.5, "step": 30940, "token_acc": 0.9363957597173145, "train_speed(iter/s)": 1.458446 }, { "epoch": 1.32577867272182, "grad_norm": 6.2364397048950195, "learning_rate": 8.36325282566402e-05, "loss": 0.6456130981445313, "memory(GiB)": 70.5, "step": 30945, "token_acc": 0.883495145631068, "train_speed(iter/s)": 1.458472 }, { "epoch": 1.325992888051069, "grad_norm": 3.632054090499878, "learning_rate": 8.36275481872593e-05, "loss": 0.12550405263900757, "memory(GiB)": 70.5, "step": 30950, "token_acc": 0.9735849056603774, "train_speed(iter/s)": 1.458478 }, { "epoch": 1.3262071033803178, "grad_norm": 2.485821485519409, "learning_rate": 8.362256750868382e-05, "loss": 0.18398520946502686, "memory(GiB)": 70.5, "step": 30955, "token_acc": 0.9748201438848921, "train_speed(iter/s)": 1.458483 }, { "epoch": 1.3264213187095668, "grad_norm": 0.036505550146102905, "learning_rate": 8.361758622100399e-05, "loss": 0.38934500217437745, "memory(GiB)": 70.5, "step": 30960, "token_acc": 0.91875, "train_speed(iter/s)": 1.458499 }, { "epoch": 1.3266355340388158, "grad_norm": 3.1628026962280273, "learning_rate": 8.361260432431003e-05, "loss": 0.2537261962890625, "memory(GiB)": 70.5, "step": 30965, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.458501 }, { "epoch": 1.3268497493680647, "grad_norm": 5.105327129364014, "learning_rate": 8.360762181869224e-05, "loss": 0.7402002334594726, "memory(GiB)": 70.5, "step": 30970, "token_acc": 0.85625, "train_speed(iter/s)": 1.458518 }, { "epoch": 1.3270639646973137, "grad_norm": 5.838544845581055, "learning_rate": 8.360263870424083e-05, "loss": 0.4067881107330322, "memory(GiB)": 70.5, "step": 30975, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.458518 }, { "epoch": 1.3272781800265627, "grad_norm": 3.496830940246582, "learning_rate": 8.359765498104612e-05, "loss": 0.3674940586090088, "memory(GiB)": 70.5, "step": 30980, "token_acc": 0.9316770186335404, "train_speed(iter/s)": 1.458514 }, { "epoch": 1.3274923953558115, "grad_norm": 3.0507278442382812, "learning_rate": 8.359267064919836e-05, "loss": 0.5726096153259277, "memory(GiB)": 70.5, "step": 30985, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.458516 }, { "epoch": 1.3277066106850606, "grad_norm": 2.910489320755005, "learning_rate": 8.358768570878786e-05, "loss": 0.09157218933105468, "memory(GiB)": 70.5, "step": 30990, "token_acc": 0.9775280898876404, "train_speed(iter/s)": 1.458547 }, { "epoch": 1.3279208260143096, "grad_norm": 3.381350040435791, "learning_rate": 8.358270015990492e-05, "loss": 0.5638362407684326, "memory(GiB)": 70.5, "step": 30995, "token_acc": 0.8754098360655738, "train_speed(iter/s)": 1.458547 }, { "epoch": 1.3281350413435584, "grad_norm": 1.1046626567840576, "learning_rate": 8.357771400263988e-05, "loss": 0.5038112163543701, "memory(GiB)": 70.5, "step": 31000, "token_acc": 0.8803088803088803, "train_speed(iter/s)": 1.458551 }, { "epoch": 1.3281350413435584, "eval_loss": 2.7662265300750732, "eval_runtime": 13.1503, "eval_samples_per_second": 7.604, "eval_steps_per_second": 7.604, "eval_token_acc": 0.3953804347826087, "step": 31000 }, { "epoch": 1.3283492566728075, "grad_norm": 2.2119672298431396, "learning_rate": 8.357272723708303e-05, "loss": 0.3419873476028442, "memory(GiB)": 70.5, "step": 31005, "token_acc": 0.5401960784313725, "train_speed(iter/s)": 1.457622 }, { "epoch": 1.3285634720020565, "grad_norm": 2.207697868347168, "learning_rate": 8.356773986332475e-05, "loss": 0.2734371662139893, "memory(GiB)": 70.5, "step": 31010, "token_acc": 0.9532163742690059, "train_speed(iter/s)": 1.457619 }, { "epoch": 1.3287776873313053, "grad_norm": 4.529321670532227, "learning_rate": 8.356275188145537e-05, "loss": 0.5452449321746826, "memory(GiB)": 70.5, "step": 31015, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.457601 }, { "epoch": 1.3289919026605543, "grad_norm": 3.078209638595581, "learning_rate": 8.355776329156525e-05, "loss": 0.3211393356323242, "memory(GiB)": 70.5, "step": 31020, "token_acc": 0.9368770764119602, "train_speed(iter/s)": 1.457593 }, { "epoch": 1.3292061179898034, "grad_norm": 3.767778158187866, "learning_rate": 8.355277409374477e-05, "loss": 0.4477419853210449, "memory(GiB)": 70.5, "step": 31025, "token_acc": 0.8970588235294118, "train_speed(iter/s)": 1.457619 }, { "epoch": 1.3294203333190522, "grad_norm": 6.755476474761963, "learning_rate": 8.354778428808431e-05, "loss": 0.5544904708862305, "memory(GiB)": 70.5, "step": 31030, "token_acc": 0.891566265060241, "train_speed(iter/s)": 1.457658 }, { "epoch": 1.3296345486483012, "grad_norm": 5.012375354766846, "learning_rate": 8.354279387467427e-05, "loss": 0.5890726089477539, "memory(GiB)": 70.5, "step": 31035, "token_acc": 0.8802588996763754, "train_speed(iter/s)": 1.457684 }, { "epoch": 1.3298487639775503, "grad_norm": 4.865383625030518, "learning_rate": 8.353780285360505e-05, "loss": 0.425032901763916, "memory(GiB)": 70.5, "step": 31040, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.457686 }, { "epoch": 1.330062979306799, "grad_norm": 2.505467653274536, "learning_rate": 8.353281122496708e-05, "loss": 0.26324920654296874, "memory(GiB)": 70.5, "step": 31045, "token_acc": 0.9501779359430605, "train_speed(iter/s)": 1.457681 }, { "epoch": 1.330277194636048, "grad_norm": 5.424981117248535, "learning_rate": 8.352781898885076e-05, "loss": 0.4077505111694336, "memory(GiB)": 70.5, "step": 31050, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.457683 }, { "epoch": 1.3304914099652971, "grad_norm": 6.503383159637451, "learning_rate": 8.352282614534657e-05, "loss": 0.4809126377105713, "memory(GiB)": 70.5, "step": 31055, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.457706 }, { "epoch": 1.330705625294546, "grad_norm": 2.2178032398223877, "learning_rate": 8.351783269454493e-05, "loss": 0.2834794998168945, "memory(GiB)": 70.5, "step": 31060, "token_acc": 0.944954128440367, "train_speed(iter/s)": 1.457708 }, { "epoch": 1.330919840623795, "grad_norm": 2.656898021697998, "learning_rate": 8.35128386365363e-05, "loss": 0.37271573543548586, "memory(GiB)": 70.5, "step": 31065, "token_acc": 0.9184397163120568, "train_speed(iter/s)": 1.457711 }, { "epoch": 1.331134055953044, "grad_norm": 5.984844207763672, "learning_rate": 8.350784397141116e-05, "loss": 0.18597898483276368, "memory(GiB)": 70.5, "step": 31070, "token_acc": 0.968503937007874, "train_speed(iter/s)": 1.457711 }, { "epoch": 1.3313482712822928, "grad_norm": 1.086124062538147, "learning_rate": 8.350284869926e-05, "loss": 0.06085920333862305, "memory(GiB)": 70.5, "step": 31075, "token_acc": 0.9844236760124611, "train_speed(iter/s)": 1.457711 }, { "epoch": 1.3315624866115419, "grad_norm": 2.3821704387664795, "learning_rate": 8.349785282017332e-05, "loss": 0.8004261016845703, "memory(GiB)": 70.5, "step": 31080, "token_acc": 0.8117647058823529, "train_speed(iter/s)": 1.457749 }, { "epoch": 1.331776701940791, "grad_norm": 1.6684410572052002, "learning_rate": 8.34928563342416e-05, "loss": 0.16533632278442384, "memory(GiB)": 70.5, "step": 31085, "token_acc": 0.9665551839464883, "train_speed(iter/s)": 1.457754 }, { "epoch": 1.3319909172700397, "grad_norm": 4.104395866394043, "learning_rate": 8.348785924155536e-05, "loss": 0.3593613862991333, "memory(GiB)": 70.5, "step": 31090, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.457753 }, { "epoch": 1.3322051325992887, "grad_norm": 4.032218933105469, "learning_rate": 8.348286154220514e-05, "loss": 0.3442162275314331, "memory(GiB)": 70.5, "step": 31095, "token_acc": 0.9432314410480349, "train_speed(iter/s)": 1.457753 }, { "epoch": 1.3324193479285378, "grad_norm": 3.7357561588287354, "learning_rate": 8.347786323628149e-05, "loss": 0.225404691696167, "memory(GiB)": 70.5, "step": 31100, "token_acc": 0.950920245398773, "train_speed(iter/s)": 1.457767 }, { "epoch": 1.3326335632577866, "grad_norm": 4.419565200805664, "learning_rate": 8.347286432387493e-05, "loss": 0.5715919017791748, "memory(GiB)": 70.5, "step": 31105, "token_acc": 0.8710691823899371, "train_speed(iter/s)": 1.457755 }, { "epoch": 1.3328477785870356, "grad_norm": 2.893514394760132, "learning_rate": 8.346786480507603e-05, "loss": 0.38820934295654297, "memory(GiB)": 70.5, "step": 31110, "token_acc": 0.9395465994962217, "train_speed(iter/s)": 1.457744 }, { "epoch": 1.3330619939162847, "grad_norm": 1.741992712020874, "learning_rate": 8.346286467997536e-05, "loss": 0.10230168104171752, "memory(GiB)": 70.5, "step": 31115, "token_acc": 0.9734513274336283, "train_speed(iter/s)": 1.457769 }, { "epoch": 1.3332762092455335, "grad_norm": 2.6172544956207275, "learning_rate": 8.345786394866352e-05, "loss": 0.27806408405303956, "memory(GiB)": 70.5, "step": 31120, "token_acc": 0.9522184300341296, "train_speed(iter/s)": 1.457773 }, { "epoch": 1.3334904245747825, "grad_norm": 0.2575242519378662, "learning_rate": 8.345286261123108e-05, "loss": 0.20925917625427246, "memory(GiB)": 70.5, "step": 31125, "token_acc": 0.9634146341463414, "train_speed(iter/s)": 1.457802 }, { "epoch": 1.3337046399040315, "grad_norm": 1.6595304012298584, "learning_rate": 8.344786066776865e-05, "loss": 0.18914788961410522, "memory(GiB)": 70.5, "step": 31130, "token_acc": 0.9562043795620438, "train_speed(iter/s)": 1.457816 }, { "epoch": 1.3339188552332806, "grad_norm": 4.688130855560303, "learning_rate": 8.344285811836686e-05, "loss": 0.37704341411590575, "memory(GiB)": 70.5, "step": 31135, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.457827 }, { "epoch": 1.3341330705625294, "grad_norm": 4.107277870178223, "learning_rate": 8.34378549631163e-05, "loss": 0.3210508108139038, "memory(GiB)": 70.5, "step": 31140, "token_acc": 0.9233576642335767, "train_speed(iter/s)": 1.457828 }, { "epoch": 1.3343472858917784, "grad_norm": 5.844002723693848, "learning_rate": 8.343285120210764e-05, "loss": 0.4314523696899414, "memory(GiB)": 70.5, "step": 31145, "token_acc": 0.9151515151515152, "train_speed(iter/s)": 1.457826 }, { "epoch": 1.3345615012210275, "grad_norm": 2.506340742111206, "learning_rate": 8.342784683543152e-05, "loss": 0.27683162689208984, "memory(GiB)": 70.5, "step": 31150, "token_acc": 0.9246031746031746, "train_speed(iter/s)": 1.457833 }, { "epoch": 1.3347757165502763, "grad_norm": 3.7729432582855225, "learning_rate": 8.342284186317862e-05, "loss": 0.4621148109436035, "memory(GiB)": 70.5, "step": 31155, "token_acc": 0.907258064516129, "train_speed(iter/s)": 1.457827 }, { "epoch": 1.3349899318795253, "grad_norm": 5.993578910827637, "learning_rate": 8.341783628543956e-05, "loss": 0.6883987426757813, "memory(GiB)": 70.5, "step": 31160, "token_acc": 0.8709677419354839, "train_speed(iter/s)": 1.457839 }, { "epoch": 1.3352041472087743, "grad_norm": 6.005300521850586, "learning_rate": 8.341283010230506e-05, "loss": 0.39380812644958496, "memory(GiB)": 70.5, "step": 31165, "token_acc": 0.9003831417624522, "train_speed(iter/s)": 1.457873 }, { "epoch": 1.3354183625380232, "grad_norm": 2.190178632736206, "learning_rate": 8.340782331386578e-05, "loss": 0.3699225425720215, "memory(GiB)": 70.5, "step": 31170, "token_acc": 0.911042944785276, "train_speed(iter/s)": 1.457873 }, { "epoch": 1.3356325778672722, "grad_norm": 2.6624016761779785, "learning_rate": 8.340281592021246e-05, "loss": 0.3809111833572388, "memory(GiB)": 70.5, "step": 31175, "token_acc": 0.9147286821705426, "train_speed(iter/s)": 1.457884 }, { "epoch": 1.3358467931965212, "grad_norm": 2.8840959072113037, "learning_rate": 8.339780792143578e-05, "loss": 0.40105252265930175, "memory(GiB)": 70.5, "step": 31180, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.457872 }, { "epoch": 1.33606100852577, "grad_norm": 6.561142921447754, "learning_rate": 8.33927993176265e-05, "loss": 0.3412734031677246, "memory(GiB)": 70.5, "step": 31185, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.457873 }, { "epoch": 1.336275223855019, "grad_norm": 2.5268890857696533, "learning_rate": 8.338779010887532e-05, "loss": 0.18282386064529418, "memory(GiB)": 70.5, "step": 31190, "token_acc": 0.9640287769784173, "train_speed(iter/s)": 1.457876 }, { "epoch": 1.336489439184268, "grad_norm": 4.930416584014893, "learning_rate": 8.3382780295273e-05, "loss": 0.3477349758148193, "memory(GiB)": 70.5, "step": 31195, "token_acc": 0.926530612244898, "train_speed(iter/s)": 1.457872 }, { "epoch": 1.336703654513517, "grad_norm": 1.3177157640457153, "learning_rate": 8.337776987691031e-05, "loss": 0.3792971134185791, "memory(GiB)": 70.5, "step": 31200, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.457876 }, { "epoch": 1.336917869842766, "grad_norm": 4.684971332550049, "learning_rate": 8.3372758853878e-05, "loss": 0.454222297668457, "memory(GiB)": 70.5, "step": 31205, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.457913 }, { "epoch": 1.337132085172015, "grad_norm": 4.186781883239746, "learning_rate": 8.336774722626685e-05, "loss": 0.25101385116577146, "memory(GiB)": 70.5, "step": 31210, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.457931 }, { "epoch": 1.3373463005012638, "grad_norm": 2.652313470840454, "learning_rate": 8.336273499416767e-05, "loss": 0.5430817127227783, "memory(GiB)": 70.5, "step": 31215, "token_acc": 0.8957528957528957, "train_speed(iter/s)": 1.457941 }, { "epoch": 1.3375605158305128, "grad_norm": 5.613762378692627, "learning_rate": 8.335772215767125e-05, "loss": 0.6481259346008301, "memory(GiB)": 70.5, "step": 31220, "token_acc": 0.8717948717948718, "train_speed(iter/s)": 1.45796 }, { "epoch": 1.3377747311597619, "grad_norm": 3.5432753562927246, "learning_rate": 8.33527087168684e-05, "loss": 0.41013035774230955, "memory(GiB)": 70.5, "step": 31225, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.457953 }, { "epoch": 1.3379889464890107, "grad_norm": 3.396841049194336, "learning_rate": 8.334769467184992e-05, "loss": 0.3944345474243164, "memory(GiB)": 70.5, "step": 31230, "token_acc": 0.9372822299651568, "train_speed(iter/s)": 1.457982 }, { "epoch": 1.3382031618182597, "grad_norm": 1.5638725757598877, "learning_rate": 8.334268002270671e-05, "loss": 0.3274683475494385, "memory(GiB)": 70.5, "step": 31235, "token_acc": 0.9188311688311688, "train_speed(iter/s)": 1.458002 }, { "epoch": 1.3384173771475087, "grad_norm": 0.6727954745292664, "learning_rate": 8.333766476952955e-05, "loss": 0.14229387044906616, "memory(GiB)": 70.5, "step": 31240, "token_acc": 0.9528301886792453, "train_speed(iter/s)": 1.45801 }, { "epoch": 1.3386315924767576, "grad_norm": 4.660986423492432, "learning_rate": 8.333264891240933e-05, "loss": 0.47110638618469236, "memory(GiB)": 70.5, "step": 31245, "token_acc": 0.8961038961038961, "train_speed(iter/s)": 1.458011 }, { "epoch": 1.3388458078060066, "grad_norm": 7.255104064941406, "learning_rate": 8.33276324514369e-05, "loss": 0.47139577865600585, "memory(GiB)": 70.5, "step": 31250, "token_acc": 0.9067524115755627, "train_speed(iter/s)": 1.458042 }, { "epoch": 1.3390600231352556, "grad_norm": 2.8038973808288574, "learning_rate": 8.332261538670313e-05, "loss": 0.40904793739318845, "memory(GiB)": 70.5, "step": 31255, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.45805 }, { "epoch": 1.3392742384645044, "grad_norm": 4.339719772338867, "learning_rate": 8.331759771829893e-05, "loss": 0.3278007507324219, "memory(GiB)": 70.5, "step": 31260, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.458069 }, { "epoch": 1.3394884537937535, "grad_norm": 4.128951549530029, "learning_rate": 8.331257944631521e-05, "loss": 0.46600685119628904, "memory(GiB)": 70.5, "step": 31265, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.458069 }, { "epoch": 1.3397026691230025, "grad_norm": 8.069190979003906, "learning_rate": 8.330756057084284e-05, "loss": 0.27997579574584963, "memory(GiB)": 70.5, "step": 31270, "token_acc": 0.956, "train_speed(iter/s)": 1.458059 }, { "epoch": 1.3399168844522513, "grad_norm": 4.666238784790039, "learning_rate": 8.330254109197279e-05, "loss": 0.4550352096557617, "memory(GiB)": 70.5, "step": 31275, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.458079 }, { "epoch": 1.3401310997815004, "grad_norm": 6.417453289031982, "learning_rate": 8.329752100979595e-05, "loss": 0.390762186050415, "memory(GiB)": 70.5, "step": 31280, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.458078 }, { "epoch": 1.3403453151107494, "grad_norm": 5.8136396408081055, "learning_rate": 8.329250032440329e-05, "loss": 0.5129222869873047, "memory(GiB)": 70.5, "step": 31285, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.458089 }, { "epoch": 1.3405595304399982, "grad_norm": 2.4792592525482178, "learning_rate": 8.328747903588575e-05, "loss": 0.5206907272338868, "memory(GiB)": 70.5, "step": 31290, "token_acc": 0.9114391143911439, "train_speed(iter/s)": 1.458084 }, { "epoch": 1.3407737457692472, "grad_norm": 1.6315717697143555, "learning_rate": 8.32824571443343e-05, "loss": 0.21797757148742675, "memory(GiB)": 70.5, "step": 31295, "token_acc": 0.954225352112676, "train_speed(iter/s)": 1.458108 }, { "epoch": 1.3409879610984963, "grad_norm": 5.144736289978027, "learning_rate": 8.327743464983992e-05, "loss": 0.6422752857208252, "memory(GiB)": 70.5, "step": 31300, "token_acc": 0.8045602605863192, "train_speed(iter/s)": 1.458106 }, { "epoch": 1.3412021764277453, "grad_norm": 4.302720546722412, "learning_rate": 8.327241155249361e-05, "loss": 0.5841968536376954, "memory(GiB)": 70.5, "step": 31305, "token_acc": 0.875, "train_speed(iter/s)": 1.458102 }, { "epoch": 1.3414163917569941, "grad_norm": 6.465277671813965, "learning_rate": 8.326738785238633e-05, "loss": 0.3523401260375977, "memory(GiB)": 70.5, "step": 31310, "token_acc": 0.9246376811594202, "train_speed(iter/s)": 1.458105 }, { "epoch": 1.3416306070862432, "grad_norm": 2.071016311645508, "learning_rate": 8.326236354960912e-05, "loss": 0.3168393611907959, "memory(GiB)": 70.5, "step": 31315, "token_acc": 0.9275766016713092, "train_speed(iter/s)": 1.458102 }, { "epoch": 1.3418448224154922, "grad_norm": 3.1681230068206787, "learning_rate": 8.325733864425299e-05, "loss": 0.17941938638687133, "memory(GiB)": 70.5, "step": 31320, "token_acc": 0.9602888086642599, "train_speed(iter/s)": 1.4581 }, { "epoch": 1.342059037744741, "grad_norm": 2.9571008682250977, "learning_rate": 8.325231313640898e-05, "loss": 0.2989517688751221, "memory(GiB)": 70.5, "step": 31325, "token_acc": 0.9246987951807228, "train_speed(iter/s)": 1.458107 }, { "epoch": 1.34227325307399, "grad_norm": 2.9613218307495117, "learning_rate": 8.32472870261681e-05, "loss": 0.48579721450805663, "memory(GiB)": 70.5, "step": 31330, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.458097 }, { "epoch": 1.342487468403239, "grad_norm": 3.142446756362915, "learning_rate": 8.324226031362146e-05, "loss": 0.4617640495300293, "memory(GiB)": 70.5, "step": 31335, "token_acc": 0.8953068592057761, "train_speed(iter/s)": 1.45813 }, { "epoch": 1.3427016837324879, "grad_norm": 5.625137805938721, "learning_rate": 8.323723299886008e-05, "loss": 0.5173349857330323, "memory(GiB)": 70.5, "step": 31340, "token_acc": 0.8675496688741722, "train_speed(iter/s)": 1.458137 }, { "epoch": 1.342915899061737, "grad_norm": 0.6058438420295715, "learning_rate": 8.323220508197505e-05, "loss": 0.18857717514038086, "memory(GiB)": 70.5, "step": 31345, "token_acc": 0.967948717948718, "train_speed(iter/s)": 1.458142 }, { "epoch": 1.343130114390986, "grad_norm": 1.8134223222732544, "learning_rate": 8.322717656305744e-05, "loss": 0.3202333927154541, "memory(GiB)": 70.5, "step": 31350, "token_acc": 0.9375, "train_speed(iter/s)": 1.458144 }, { "epoch": 1.3433443297202348, "grad_norm": 1.989943504333496, "learning_rate": 8.322214744219837e-05, "loss": 0.3360640048980713, "memory(GiB)": 70.5, "step": 31355, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.458141 }, { "epoch": 1.3435585450494838, "grad_norm": 2.905843734741211, "learning_rate": 8.32171177194889e-05, "loss": 0.4420632362365723, "memory(GiB)": 70.5, "step": 31360, "token_acc": 0.8916967509025271, "train_speed(iter/s)": 1.458146 }, { "epoch": 1.3437727603787328, "grad_norm": 1.2550536394119263, "learning_rate": 8.32120873950202e-05, "loss": 0.3534934282302856, "memory(GiB)": 70.5, "step": 31365, "token_acc": 0.940149625935162, "train_speed(iter/s)": 1.458151 }, { "epoch": 1.3439869757079816, "grad_norm": 0.3865787386894226, "learning_rate": 8.320705646888341e-05, "loss": 0.13963112831115723, "memory(GiB)": 70.5, "step": 31370, "token_acc": 0.9764309764309764, "train_speed(iter/s)": 1.45815 }, { "epoch": 1.3442011910372307, "grad_norm": 0.6689972281455994, "learning_rate": 8.320202494116961e-05, "loss": 0.16998995542526246, "memory(GiB)": 70.5, "step": 31375, "token_acc": 0.9541284403669725, "train_speed(iter/s)": 1.458153 }, { "epoch": 1.3444154063664797, "grad_norm": 1.5665984153747559, "learning_rate": 8.319699281196999e-05, "loss": 0.5258900165557862, "memory(GiB)": 70.5, "step": 31380, "token_acc": 0.875, "train_speed(iter/s)": 1.45817 }, { "epoch": 1.3446296216957285, "grad_norm": 0.07869243621826172, "learning_rate": 8.319196008137571e-05, "loss": 0.26584973335266116, "memory(GiB)": 70.5, "step": 31385, "token_acc": 0.93359375, "train_speed(iter/s)": 1.458173 }, { "epoch": 1.3448438370249776, "grad_norm": 6.08672571182251, "learning_rate": 8.318692674947793e-05, "loss": 0.39982969760894777, "memory(GiB)": 70.5, "step": 31390, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.458185 }, { "epoch": 1.3450580523542266, "grad_norm": 3.1844992637634277, "learning_rate": 8.318189281636784e-05, "loss": 0.3577871799468994, "memory(GiB)": 70.5, "step": 31395, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.45818 }, { "epoch": 1.3452722676834754, "grad_norm": 4.300228595733643, "learning_rate": 8.317685828213664e-05, "loss": 0.40508365631103516, "memory(GiB)": 70.5, "step": 31400, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.458213 }, { "epoch": 1.3454864830127244, "grad_norm": 2.142059326171875, "learning_rate": 8.317182314687551e-05, "loss": 0.5300819396972656, "memory(GiB)": 70.5, "step": 31405, "token_acc": 0.9039145907473309, "train_speed(iter/s)": 1.458226 }, { "epoch": 1.3457006983419735, "grad_norm": 4.371211528778076, "learning_rate": 8.316678741067571e-05, "loss": 0.3056309700012207, "memory(GiB)": 70.5, "step": 31410, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.458244 }, { "epoch": 1.3459149136712223, "grad_norm": 4.26999044418335, "learning_rate": 8.316175107362842e-05, "loss": 0.2888308525085449, "memory(GiB)": 70.5, "step": 31415, "token_acc": 0.9363057324840764, "train_speed(iter/s)": 1.45824 }, { "epoch": 1.3461291290004713, "grad_norm": 1.964132308959961, "learning_rate": 8.315671413582493e-05, "loss": 0.22614424228668212, "memory(GiB)": 70.5, "step": 31420, "token_acc": 0.9447852760736196, "train_speed(iter/s)": 1.458241 }, { "epoch": 1.3463433443297204, "grad_norm": 1.5403454303741455, "learning_rate": 8.315167659735646e-05, "loss": 0.4862874984741211, "memory(GiB)": 70.5, "step": 31425, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.458227 }, { "epoch": 1.3465575596589692, "grad_norm": 0.5216339230537415, "learning_rate": 8.314663845831425e-05, "loss": 0.2609166145324707, "memory(GiB)": 70.5, "step": 31430, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.458231 }, { "epoch": 1.3467717749882182, "grad_norm": 5.359578609466553, "learning_rate": 8.314159971878958e-05, "loss": 0.575745677947998, "memory(GiB)": 70.5, "step": 31435, "token_acc": 0.8754325259515571, "train_speed(iter/s)": 1.458259 }, { "epoch": 1.3469859903174672, "grad_norm": 4.854711055755615, "learning_rate": 8.313656037887377e-05, "loss": 0.575189733505249, "memory(GiB)": 70.5, "step": 31440, "token_acc": 0.8708609271523179, "train_speed(iter/s)": 1.458256 }, { "epoch": 1.347200205646716, "grad_norm": 6.071609973907471, "learning_rate": 8.313152043865806e-05, "loss": 0.6041620254516602, "memory(GiB)": 70.5, "step": 31445, "token_acc": 0.8666666666666667, "train_speed(iter/s)": 1.458272 }, { "epoch": 1.347414420975965, "grad_norm": 0.6513997316360474, "learning_rate": 8.312647989823379e-05, "loss": 0.2755934715270996, "memory(GiB)": 70.5, "step": 31450, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.458301 }, { "epoch": 1.3476286363052141, "grad_norm": 0.27853065729141235, "learning_rate": 8.312143875769227e-05, "loss": 0.21753389835357667, "memory(GiB)": 70.5, "step": 31455, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.458317 }, { "epoch": 1.347842851634463, "grad_norm": 5.064693927764893, "learning_rate": 8.31163970171248e-05, "loss": 0.3999814987182617, "memory(GiB)": 70.5, "step": 31460, "token_acc": 0.9044585987261147, "train_speed(iter/s)": 1.45835 }, { "epoch": 1.348057066963712, "grad_norm": 3.371999979019165, "learning_rate": 8.311135467662275e-05, "loss": 0.4224726676940918, "memory(GiB)": 70.5, "step": 31465, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.458365 }, { "epoch": 1.348271282292961, "grad_norm": 2.594611883163452, "learning_rate": 8.310631173627743e-05, "loss": 0.4168051242828369, "memory(GiB)": 70.5, "step": 31470, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.458377 }, { "epoch": 1.3484854976222098, "grad_norm": 4.097012996673584, "learning_rate": 8.310126819618023e-05, "loss": 0.33737137317657473, "memory(GiB)": 70.5, "step": 31475, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.458384 }, { "epoch": 1.3486997129514589, "grad_norm": 3.00378155708313, "learning_rate": 8.30962240564225e-05, "loss": 0.3993062973022461, "memory(GiB)": 70.5, "step": 31480, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.45837 }, { "epoch": 1.3489139282807079, "grad_norm": 1.8676398992538452, "learning_rate": 8.309117931709563e-05, "loss": 0.3838521957397461, "memory(GiB)": 70.5, "step": 31485, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.458385 }, { "epoch": 1.3491281436099567, "grad_norm": 1.0610096454620361, "learning_rate": 8.3086133978291e-05, "loss": 0.5443651676177979, "memory(GiB)": 70.5, "step": 31490, "token_acc": 0.9014598540145985, "train_speed(iter/s)": 1.458361 }, { "epoch": 1.3493423589392057, "grad_norm": 2.2066187858581543, "learning_rate": 8.308108804009999e-05, "loss": 0.3407717704772949, "memory(GiB)": 70.5, "step": 31495, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.458356 }, { "epoch": 1.3495565742684548, "grad_norm": 2.3297457695007324, "learning_rate": 8.307604150261408e-05, "loss": 0.4819674491882324, "memory(GiB)": 70.5, "step": 31500, "token_acc": 0.9013157894736842, "train_speed(iter/s)": 1.458363 }, { "epoch": 1.3495565742684548, "eval_loss": 2.611281156539917, "eval_runtime": 13.214, "eval_samples_per_second": 7.568, "eval_steps_per_second": 7.568, "eval_token_acc": 0.40583554376657827, "step": 31500 }, { "epoch": 1.3497707895977036, "grad_norm": 5.228427886962891, "learning_rate": 8.307099436592463e-05, "loss": 0.48831920623779296, "memory(GiB)": 70.5, "step": 31505, "token_acc": 0.5429650613786591, "train_speed(iter/s)": 1.457402 }, { "epoch": 1.3499850049269526, "grad_norm": 1.4234580993652344, "learning_rate": 8.306594663012308e-05, "loss": 0.11884136199951172, "memory(GiB)": 70.5, "step": 31510, "token_acc": 0.972809667673716, "train_speed(iter/s)": 1.457401 }, { "epoch": 1.3501992202562016, "grad_norm": 3.767691135406494, "learning_rate": 8.306089829530092e-05, "loss": 0.29082186222076417, "memory(GiB)": 70.5, "step": 31515, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.457406 }, { "epoch": 1.3504134355854505, "grad_norm": 3.699087381362915, "learning_rate": 8.305584936154956e-05, "loss": 0.46152620315551757, "memory(GiB)": 70.5, "step": 31520, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.457414 }, { "epoch": 1.3506276509146995, "grad_norm": 0.24070784449577332, "learning_rate": 8.305079982896047e-05, "loss": 0.27644929885864256, "memory(GiB)": 70.5, "step": 31525, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.457394 }, { "epoch": 1.3508418662439485, "grad_norm": 3.2839231491088867, "learning_rate": 8.304574969762515e-05, "loss": 0.33754887580871584, "memory(GiB)": 70.5, "step": 31530, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.457398 }, { "epoch": 1.3510560815731973, "grad_norm": 2.043254852294922, "learning_rate": 8.304069896763506e-05, "loss": 0.2933093547821045, "memory(GiB)": 70.5, "step": 31535, "token_acc": 0.9221789883268483, "train_speed(iter/s)": 1.457405 }, { "epoch": 1.3512702969024464, "grad_norm": 2.2559332847595215, "learning_rate": 8.303564763908173e-05, "loss": 0.2991332054138184, "memory(GiB)": 70.5, "step": 31540, "token_acc": 0.9448818897637795, "train_speed(iter/s)": 1.45741 }, { "epoch": 1.3514845122316954, "grad_norm": 8.778142929077148, "learning_rate": 8.303059571205665e-05, "loss": 0.6723471641540527, "memory(GiB)": 70.5, "step": 31545, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.457436 }, { "epoch": 1.3516987275609442, "grad_norm": 3.4278905391693115, "learning_rate": 8.302554318665135e-05, "loss": 0.31026382446289064, "memory(GiB)": 70.5, "step": 31550, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.457463 }, { "epoch": 1.3519129428901933, "grad_norm": 0.6479233503341675, "learning_rate": 8.302049006295734e-05, "loss": 0.45173215866088867, "memory(GiB)": 70.5, "step": 31555, "token_acc": 0.8970099667774086, "train_speed(iter/s)": 1.45746 }, { "epoch": 1.3521271582194423, "grad_norm": 5.213809967041016, "learning_rate": 8.301543634106617e-05, "loss": 0.42549972534179686, "memory(GiB)": 70.5, "step": 31560, "token_acc": 0.9206798866855525, "train_speed(iter/s)": 1.45746 }, { "epoch": 1.352341373548691, "grad_norm": 3.4815118312835693, "learning_rate": 8.301038202106942e-05, "loss": 0.37028701305389405, "memory(GiB)": 70.5, "step": 31565, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.457455 }, { "epoch": 1.3525555888779401, "grad_norm": 3.293849229812622, "learning_rate": 8.300532710305862e-05, "loss": 0.4224871635437012, "memory(GiB)": 70.5, "step": 31570, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.457464 }, { "epoch": 1.3527698042071892, "grad_norm": 4.519466400146484, "learning_rate": 8.300027158712537e-05, "loss": 0.532077407836914, "memory(GiB)": 70.5, "step": 31575, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.457482 }, { "epoch": 1.352984019536438, "grad_norm": 4.007935047149658, "learning_rate": 8.299521547336123e-05, "loss": 0.5078132629394532, "memory(GiB)": 70.5, "step": 31580, "token_acc": 0.8869257950530035, "train_speed(iter/s)": 1.457472 }, { "epoch": 1.353198234865687, "grad_norm": 5.995413303375244, "learning_rate": 8.299015876185782e-05, "loss": 0.6253188133239747, "memory(GiB)": 70.5, "step": 31585, "token_acc": 0.8641114982578397, "train_speed(iter/s)": 1.457462 }, { "epoch": 1.353412450194936, "grad_norm": 3.1256814002990723, "learning_rate": 8.298510145270673e-05, "loss": 0.268801212310791, "memory(GiB)": 70.5, "step": 31590, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.457452 }, { "epoch": 1.3536266655241849, "grad_norm": 6.689597129821777, "learning_rate": 8.298004354599959e-05, "loss": 0.5257444381713867, "memory(GiB)": 70.5, "step": 31595, "token_acc": 0.8912280701754386, "train_speed(iter/s)": 1.457451 }, { "epoch": 1.353840880853434, "grad_norm": 4.428259372711182, "learning_rate": 8.297498504182802e-05, "loss": 0.18505666255950928, "memory(GiB)": 70.5, "step": 31600, "token_acc": 0.964, "train_speed(iter/s)": 1.457472 }, { "epoch": 1.354055096182683, "grad_norm": 1.1759752035140991, "learning_rate": 8.296992594028366e-05, "loss": 0.27304530143737793, "memory(GiB)": 70.5, "step": 31605, "token_acc": 0.9283489096573209, "train_speed(iter/s)": 1.457489 }, { "epoch": 1.3542693115119318, "grad_norm": 3.2038490772247314, "learning_rate": 8.296486624145819e-05, "loss": 0.6789849281311036, "memory(GiB)": 70.5, "step": 31610, "token_acc": 0.858433734939759, "train_speed(iter/s)": 1.457476 }, { "epoch": 1.3544835268411808, "grad_norm": 2.664546489715576, "learning_rate": 8.295980594544321e-05, "loss": 0.518758487701416, "memory(GiB)": 70.5, "step": 31615, "token_acc": 0.891156462585034, "train_speed(iter/s)": 1.45748 }, { "epoch": 1.3546977421704298, "grad_norm": 1.9108167886734009, "learning_rate": 8.295474505233042e-05, "loss": 0.24088516235351562, "memory(GiB)": 70.5, "step": 31620, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.457474 }, { "epoch": 1.3549119574996786, "grad_norm": 3.443537950515747, "learning_rate": 8.294968356221155e-05, "loss": 0.22837305068969727, "memory(GiB)": 70.5, "step": 31625, "token_acc": 0.9470198675496688, "train_speed(iter/s)": 1.457492 }, { "epoch": 1.3551261728289277, "grad_norm": 6.281049728393555, "learning_rate": 8.294462147517823e-05, "loss": 0.4173742294311523, "memory(GiB)": 70.5, "step": 31630, "token_acc": 0.9105960264900662, "train_speed(iter/s)": 1.4575 }, { "epoch": 1.3553403881581767, "grad_norm": 3.81531023979187, "learning_rate": 8.293955879132219e-05, "loss": 0.5457525253295898, "memory(GiB)": 70.5, "step": 31635, "token_acc": 0.8797250859106529, "train_speed(iter/s)": 1.457517 }, { "epoch": 1.3555546034874255, "grad_norm": 2.270087957382202, "learning_rate": 8.293449551073513e-05, "loss": 0.40015544891357424, "memory(GiB)": 70.5, "step": 31640, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.457521 }, { "epoch": 1.3557688188166745, "grad_norm": 2.0096659660339355, "learning_rate": 8.29294316335088e-05, "loss": 0.537608528137207, "memory(GiB)": 70.5, "step": 31645, "token_acc": 0.907051282051282, "train_speed(iter/s)": 1.457508 }, { "epoch": 1.3559830341459236, "grad_norm": 5.371156215667725, "learning_rate": 8.292436715973493e-05, "loss": 0.3092637062072754, "memory(GiB)": 70.5, "step": 31650, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.457542 }, { "epoch": 1.3561972494751724, "grad_norm": 5.551722049713135, "learning_rate": 8.291930208950525e-05, "loss": 0.3300652027130127, "memory(GiB)": 70.5, "step": 31655, "token_acc": 0.9369085173501577, "train_speed(iter/s)": 1.457539 }, { "epoch": 1.3564114648044214, "grad_norm": 4.3979997634887695, "learning_rate": 8.291423642291153e-05, "loss": 0.462493896484375, "memory(GiB)": 70.5, "step": 31660, "token_acc": 0.8880866425992779, "train_speed(iter/s)": 1.457542 }, { "epoch": 1.3566256801336705, "grad_norm": 1.6741498708724976, "learning_rate": 8.290917016004556e-05, "loss": 0.2721752882003784, "memory(GiB)": 70.5, "step": 31665, "token_acc": 0.9298245614035088, "train_speed(iter/s)": 1.457545 }, { "epoch": 1.3568398954629193, "grad_norm": 3.1074466705322266, "learning_rate": 8.29041033009991e-05, "loss": 0.3269950866699219, "memory(GiB)": 70.5, "step": 31670, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.45758 }, { "epoch": 1.3570541107921683, "grad_norm": 5.817209720611572, "learning_rate": 8.289903584586394e-05, "loss": 0.4599153995513916, "memory(GiB)": 70.5, "step": 31675, "token_acc": 0.9, "train_speed(iter/s)": 1.457571 }, { "epoch": 1.3572683261214173, "grad_norm": 2.1849780082702637, "learning_rate": 8.289396779473188e-05, "loss": 0.3695369720458984, "memory(GiB)": 70.5, "step": 31680, "token_acc": 0.9205776173285198, "train_speed(iter/s)": 1.457552 }, { "epoch": 1.3574825414506662, "grad_norm": 2.574631690979004, "learning_rate": 8.288889914769473e-05, "loss": 0.4390693664550781, "memory(GiB)": 70.5, "step": 31685, "token_acc": 0.9106628242074928, "train_speed(iter/s)": 1.457551 }, { "epoch": 1.3576967567799152, "grad_norm": 3.294745683670044, "learning_rate": 8.288382990484432e-05, "loss": 0.45723567008972166, "memory(GiB)": 70.5, "step": 31690, "token_acc": 0.8741721854304636, "train_speed(iter/s)": 1.457569 }, { "epoch": 1.3579109721091642, "grad_norm": 1.9312301874160767, "learning_rate": 8.287876006627248e-05, "loss": 0.3131565570831299, "memory(GiB)": 70.5, "step": 31695, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.457563 }, { "epoch": 1.358125187438413, "grad_norm": 1.923977017402649, "learning_rate": 8.287368963207107e-05, "loss": 0.5107987880706787, "memory(GiB)": 70.5, "step": 31700, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.45757 }, { "epoch": 1.358339402767662, "grad_norm": 4.459475994110107, "learning_rate": 8.286861860233194e-05, "loss": 0.4275771141052246, "memory(GiB)": 70.5, "step": 31705, "token_acc": 0.919732441471572, "train_speed(iter/s)": 1.457563 }, { "epoch": 1.358553618096911, "grad_norm": 2.336400270462036, "learning_rate": 8.286354697714695e-05, "loss": 0.4818380355834961, "memory(GiB)": 70.5, "step": 31710, "token_acc": 0.8877551020408163, "train_speed(iter/s)": 1.457562 }, { "epoch": 1.35876783342616, "grad_norm": 5.0402960777282715, "learning_rate": 8.285847475660797e-05, "loss": 0.21220428943634034, "memory(GiB)": 70.5, "step": 31715, "token_acc": 0.9574468085106383, "train_speed(iter/s)": 1.45755 }, { "epoch": 1.358982048755409, "grad_norm": 3.147918939590454, "learning_rate": 8.285340194080688e-05, "loss": 0.5331828117370605, "memory(GiB)": 70.5, "step": 31720, "token_acc": 0.8618421052631579, "train_speed(iter/s)": 1.457539 }, { "epoch": 1.359196264084658, "grad_norm": 0.23215453326702118, "learning_rate": 8.284832852983562e-05, "loss": 0.2112562894821167, "memory(GiB)": 70.5, "step": 31725, "token_acc": 0.9572649572649573, "train_speed(iter/s)": 1.457554 }, { "epoch": 1.3594104794139068, "grad_norm": 0.9198223948478699, "learning_rate": 8.284325452378606e-05, "loss": 0.18191972970962525, "memory(GiB)": 70.5, "step": 31730, "token_acc": 0.9608540925266904, "train_speed(iter/s)": 1.457581 }, { "epoch": 1.3596246947431558, "grad_norm": 2.4925503730773926, "learning_rate": 8.283817992275014e-05, "loss": 0.3063460350036621, "memory(GiB)": 70.5, "step": 31735, "token_acc": 0.9191176470588235, "train_speed(iter/s)": 1.45759 }, { "epoch": 1.3598389100724049, "grad_norm": 0.47195398807525635, "learning_rate": 8.28331047268198e-05, "loss": 0.20419020652770997, "memory(GiB)": 70.5, "step": 31740, "token_acc": 0.9467213114754098, "train_speed(iter/s)": 1.457585 }, { "epoch": 1.3600531254016537, "grad_norm": 2.1890647411346436, "learning_rate": 8.282802893608694e-05, "loss": 0.20753872394561768, "memory(GiB)": 70.5, "step": 31745, "token_acc": 0.9548387096774194, "train_speed(iter/s)": 1.457579 }, { "epoch": 1.3602673407309027, "grad_norm": 6.001003742218018, "learning_rate": 8.282295255064356e-05, "loss": 0.5342288017272949, "memory(GiB)": 70.5, "step": 31750, "token_acc": 0.8952702702702703, "train_speed(iter/s)": 1.457593 }, { "epoch": 1.3604815560601518, "grad_norm": 2.137545585632324, "learning_rate": 8.28178755705816e-05, "loss": 0.5405715942382813, "memory(GiB)": 70.5, "step": 31755, "token_acc": 0.9039735099337748, "train_speed(iter/s)": 1.457588 }, { "epoch": 1.3606957713894006, "grad_norm": 5.605576992034912, "learning_rate": 8.281279799599303e-05, "loss": 0.4166828155517578, "memory(GiB)": 70.5, "step": 31760, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.457583 }, { "epoch": 1.3609099867186496, "grad_norm": 0.8982571363449097, "learning_rate": 8.280771982696985e-05, "loss": 0.2926686525344849, "memory(GiB)": 70.5, "step": 31765, "token_acc": 0.9348659003831418, "train_speed(iter/s)": 1.457587 }, { "epoch": 1.3611242020478986, "grad_norm": 2.336404323577881, "learning_rate": 8.280264106360405e-05, "loss": 0.2748521327972412, "memory(GiB)": 70.5, "step": 31770, "token_acc": 0.9377431906614786, "train_speed(iter/s)": 1.457593 }, { "epoch": 1.3613384173771474, "grad_norm": 2.7605977058410645, "learning_rate": 8.279756170598764e-05, "loss": 0.2712565422058105, "memory(GiB)": 70.5, "step": 31775, "token_acc": 0.9418282548476454, "train_speed(iter/s)": 1.457599 }, { "epoch": 1.3615526327063965, "grad_norm": 3.5356578826904297, "learning_rate": 8.279248175421264e-05, "loss": 0.38242621421813966, "memory(GiB)": 70.5, "step": 31780, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.45761 }, { "epoch": 1.3617668480356455, "grad_norm": 2.8994650840759277, "learning_rate": 8.278740120837106e-05, "loss": 0.2880394458770752, "memory(GiB)": 70.5, "step": 31785, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.457608 }, { "epoch": 1.3619810633648943, "grad_norm": 4.904291152954102, "learning_rate": 8.278232006855495e-05, "loss": 0.5209952354431152, "memory(GiB)": 70.5, "step": 31790, "token_acc": 0.8860294117647058, "train_speed(iter/s)": 1.457622 }, { "epoch": 1.3621952786941434, "grad_norm": 7.806861400604248, "learning_rate": 8.277723833485635e-05, "loss": 0.5902299404144287, "memory(GiB)": 70.5, "step": 31795, "token_acc": 0.8775510204081632, "train_speed(iter/s)": 1.457617 }, { "epoch": 1.3624094940233924, "grad_norm": 1.4798176288604736, "learning_rate": 8.277215600736734e-05, "loss": 0.27154500484466554, "memory(GiB)": 70.5, "step": 31800, "token_acc": 0.9299363057324841, "train_speed(iter/s)": 1.457622 }, { "epoch": 1.3626237093526412, "grad_norm": 2.861999273300171, "learning_rate": 8.276707308617999e-05, "loss": 0.43672518730163573, "memory(GiB)": 70.5, "step": 31805, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.457622 }, { "epoch": 1.3628379246818902, "grad_norm": 0.6378884315490723, "learning_rate": 8.276198957138636e-05, "loss": 0.2236936330795288, "memory(GiB)": 70.5, "step": 31810, "token_acc": 0.948051948051948, "train_speed(iter/s)": 1.457638 }, { "epoch": 1.3630521400111393, "grad_norm": 2.2827675342559814, "learning_rate": 8.275690546307854e-05, "loss": 0.42516460418701174, "memory(GiB)": 70.5, "step": 31815, "token_acc": 0.9088397790055248, "train_speed(iter/s)": 1.457644 }, { "epoch": 1.363266355340388, "grad_norm": 1.6200697422027588, "learning_rate": 8.275182076134869e-05, "loss": 0.4541475296020508, "memory(GiB)": 70.5, "step": 31820, "token_acc": 0.9029126213592233, "train_speed(iter/s)": 1.457686 }, { "epoch": 1.3634805706696371, "grad_norm": 5.8344950675964355, "learning_rate": 8.274673546628886e-05, "loss": 0.7373260021209717, "memory(GiB)": 70.5, "step": 31825, "token_acc": 0.8520710059171598, "train_speed(iter/s)": 1.457694 }, { "epoch": 1.3636947859988862, "grad_norm": 4.6076884269714355, "learning_rate": 8.27416495779912e-05, "loss": 0.39714808464050294, "memory(GiB)": 70.5, "step": 31830, "token_acc": 0.9036827195467422, "train_speed(iter/s)": 1.45769 }, { "epoch": 1.363909001328135, "grad_norm": 1.6688635349273682, "learning_rate": 8.273656309654785e-05, "loss": 0.29014604091644286, "memory(GiB)": 70.5, "step": 31835, "token_acc": 0.943217665615142, "train_speed(iter/s)": 1.457689 }, { "epoch": 1.364123216657384, "grad_norm": 6.867883682250977, "learning_rate": 8.273147602205094e-05, "loss": 0.1330750584602356, "memory(GiB)": 70.5, "step": 31840, "token_acc": 0.974025974025974, "train_speed(iter/s)": 1.457679 }, { "epoch": 1.364337431986633, "grad_norm": 2.486879587173462, "learning_rate": 8.272638835459263e-05, "loss": 0.4693158149719238, "memory(GiB)": 70.5, "step": 31845, "token_acc": 0.9, "train_speed(iter/s)": 1.457693 }, { "epoch": 1.3645516473158819, "grad_norm": 1.7542009353637695, "learning_rate": 8.272130009426511e-05, "loss": 0.49951767921447754, "memory(GiB)": 70.5, "step": 31850, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.457729 }, { "epoch": 1.364765862645131, "grad_norm": 3.818143844604492, "learning_rate": 8.271621124116055e-05, "loss": 0.23757975101470946, "memory(GiB)": 70.5, "step": 31855, "token_acc": 0.9577922077922078, "train_speed(iter/s)": 1.457735 }, { "epoch": 1.36498007797438, "grad_norm": 3.4566218852996826, "learning_rate": 8.271112179537113e-05, "loss": 0.25226621627807616, "memory(GiB)": 70.5, "step": 31860, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.457726 }, { "epoch": 1.3651942933036287, "grad_norm": 2.7064688205718994, "learning_rate": 8.270603175698904e-05, "loss": 0.2932211637496948, "memory(GiB)": 70.5, "step": 31865, "token_acc": 0.9260450160771704, "train_speed(iter/s)": 1.457706 }, { "epoch": 1.3654085086328778, "grad_norm": 5.313023090362549, "learning_rate": 8.270094112610651e-05, "loss": 0.33902697563171386, "memory(GiB)": 70.5, "step": 31870, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.457695 }, { "epoch": 1.3656227239621268, "grad_norm": 4.478115081787109, "learning_rate": 8.269584990281575e-05, "loss": 0.36807637214660643, "memory(GiB)": 70.5, "step": 31875, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.457685 }, { "epoch": 1.3658369392913756, "grad_norm": 3.70460844039917, "learning_rate": 8.269075808720901e-05, "loss": 0.4497077941894531, "memory(GiB)": 70.5, "step": 31880, "token_acc": 0.8839285714285714, "train_speed(iter/s)": 1.457687 }, { "epoch": 1.3660511546206247, "grad_norm": 3.599642515182495, "learning_rate": 8.268566567937851e-05, "loss": 0.4753710746765137, "memory(GiB)": 70.5, "step": 31885, "token_acc": 0.9072463768115943, "train_speed(iter/s)": 1.457712 }, { "epoch": 1.3662653699498737, "grad_norm": 1.5511904954910278, "learning_rate": 8.268057267941652e-05, "loss": 0.22208127975463868, "memory(GiB)": 70.5, "step": 31890, "token_acc": 0.9537366548042705, "train_speed(iter/s)": 1.457707 }, { "epoch": 1.3664795852791225, "grad_norm": 0.3646186888217926, "learning_rate": 8.267547908741529e-05, "loss": 0.3991560459136963, "memory(GiB)": 70.5, "step": 31895, "token_acc": 0.9182389937106918, "train_speed(iter/s)": 1.45773 }, { "epoch": 1.3666938006083715, "grad_norm": 3.52769136428833, "learning_rate": 8.267038490346712e-05, "loss": 0.13142645359039307, "memory(GiB)": 70.5, "step": 31900, "token_acc": 0.9708333333333333, "train_speed(iter/s)": 1.45774 }, { "epoch": 1.3669080159376206, "grad_norm": 3.0482499599456787, "learning_rate": 8.266529012766428e-05, "loss": 0.30567524433135984, "memory(GiB)": 70.5, "step": 31905, "token_acc": 0.9348534201954397, "train_speed(iter/s)": 1.457751 }, { "epoch": 1.3671222312668694, "grad_norm": 1.7586435079574585, "learning_rate": 8.266019476009905e-05, "loss": 0.3112502574920654, "memory(GiB)": 70.5, "step": 31910, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.457755 }, { "epoch": 1.3673364465961184, "grad_norm": 3.4042930603027344, "learning_rate": 8.265509880086376e-05, "loss": 0.5230031967163086, "memory(GiB)": 70.5, "step": 31915, "token_acc": 0.89937106918239, "train_speed(iter/s)": 1.457749 }, { "epoch": 1.3675506619253675, "grad_norm": 3.1301589012145996, "learning_rate": 8.265000225005073e-05, "loss": 0.3028783082962036, "memory(GiB)": 70.5, "step": 31920, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.457746 }, { "epoch": 1.3677648772546163, "grad_norm": 3.2562994956970215, "learning_rate": 8.264490510775227e-05, "loss": 0.34783596992492677, "memory(GiB)": 70.5, "step": 31925, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.457744 }, { "epoch": 1.3679790925838653, "grad_norm": 2.54302716255188, "learning_rate": 8.263980737406073e-05, "loss": 0.3944389343261719, "memory(GiB)": 70.5, "step": 31930, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.457742 }, { "epoch": 1.3681933079131143, "grad_norm": 3.174320936203003, "learning_rate": 8.263470904906849e-05, "loss": 0.22257516384124756, "memory(GiB)": 70.5, "step": 31935, "token_acc": 0.9490445859872612, "train_speed(iter/s)": 1.457755 }, { "epoch": 1.3684075232423631, "grad_norm": 5.440356731414795, "learning_rate": 8.262961013286785e-05, "loss": 0.38632588386535643, "memory(GiB)": 70.5, "step": 31940, "token_acc": 0.9403341288782816, "train_speed(iter/s)": 1.457763 }, { "epoch": 1.3686217385716122, "grad_norm": 3.0923423767089844, "learning_rate": 8.262451062555123e-05, "loss": 0.34702956676483154, "memory(GiB)": 70.5, "step": 31945, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.457757 }, { "epoch": 1.3688359539008612, "grad_norm": 4.387852191925049, "learning_rate": 8.261941052721098e-05, "loss": 0.5697942733764648, "memory(GiB)": 70.5, "step": 31950, "token_acc": 0.8717105263157895, "train_speed(iter/s)": 1.457748 }, { "epoch": 1.36905016923011, "grad_norm": 3.1159541606903076, "learning_rate": 8.261430983793952e-05, "loss": 0.3980264186859131, "memory(GiB)": 70.5, "step": 31955, "token_acc": 0.9018181818181819, "train_speed(iter/s)": 1.457778 }, { "epoch": 1.369264384559359, "grad_norm": 3.072955369949341, "learning_rate": 8.260920855782924e-05, "loss": 0.36126999855041503, "memory(GiB)": 70.5, "step": 31960, "token_acc": 0.9168975069252078, "train_speed(iter/s)": 1.457787 }, { "epoch": 1.369478599888608, "grad_norm": 4.613468647003174, "learning_rate": 8.260410668697255e-05, "loss": 0.2875772476196289, "memory(GiB)": 70.5, "step": 31965, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.457842 }, { "epoch": 1.369692815217857, "grad_norm": 3.6807515621185303, "learning_rate": 8.25990042254619e-05, "loss": 0.5986684799194336, "memory(GiB)": 70.5, "step": 31970, "token_acc": 0.8910256410256411, "train_speed(iter/s)": 1.457825 }, { "epoch": 1.369907030547106, "grad_norm": 5.805131912231445, "learning_rate": 8.25939011733897e-05, "loss": 0.5248695373535156, "memory(GiB)": 70.5, "step": 31975, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.45786 }, { "epoch": 1.370121245876355, "grad_norm": 5.289968967437744, "learning_rate": 8.25887975308484e-05, "loss": 0.37045552730560305, "memory(GiB)": 70.5, "step": 31980, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.457874 }, { "epoch": 1.3703354612056038, "grad_norm": 4.060341835021973, "learning_rate": 8.258369329793046e-05, "loss": 0.3497091054916382, "memory(GiB)": 70.5, "step": 31985, "token_acc": 0.9152542372881356, "train_speed(iter/s)": 1.457892 }, { "epoch": 1.3705496765348528, "grad_norm": 7.005445957183838, "learning_rate": 8.257858847472836e-05, "loss": 0.7233686447143555, "memory(GiB)": 70.5, "step": 31990, "token_acc": 0.8518518518518519, "train_speed(iter/s)": 1.457889 }, { "epoch": 1.3707638918641019, "grad_norm": 2.3483924865722656, "learning_rate": 8.257348306133457e-05, "loss": 0.4229994297027588, "memory(GiB)": 70.5, "step": 31995, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.457919 }, { "epoch": 1.3709781071933507, "grad_norm": 4.431223392486572, "learning_rate": 8.256837705784157e-05, "loss": 0.2819665431976318, "memory(GiB)": 70.5, "step": 32000, "token_acc": 0.9420849420849421, "train_speed(iter/s)": 1.457896 }, { "epoch": 1.3709781071933507, "eval_loss": 2.532160520553589, "eval_runtime": 13.176, "eval_samples_per_second": 7.59, "eval_steps_per_second": 7.59, "eval_token_acc": 0.44297082228116713, "step": 32000 }, { "epoch": 1.3711923225225997, "grad_norm": 2.0553698539733887, "learning_rate": 8.256327046434187e-05, "loss": 0.3000528812408447, "memory(GiB)": 70.5, "step": 32005, "token_acc": 0.5786407766990291, "train_speed(iter/s)": 1.45695 }, { "epoch": 1.3714065378518487, "grad_norm": 7.197695255279541, "learning_rate": 8.255816328092798e-05, "loss": 0.4443856716156006, "memory(GiB)": 70.5, "step": 32010, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.456957 }, { "epoch": 1.3716207531810976, "grad_norm": 4.224639892578125, "learning_rate": 8.255305550769242e-05, "loss": 0.37266175746917723, "memory(GiB)": 70.5, "step": 32015, "token_acc": 0.9315068493150684, "train_speed(iter/s)": 1.456972 }, { "epoch": 1.3718349685103466, "grad_norm": 2.7306935787200928, "learning_rate": 8.254794714472771e-05, "loss": 0.4203244686126709, "memory(GiB)": 70.5, "step": 32020, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.456969 }, { "epoch": 1.3720491838395956, "grad_norm": 1.7140921354293823, "learning_rate": 8.254283819212641e-05, "loss": 0.54891357421875, "memory(GiB)": 70.5, "step": 32025, "token_acc": 0.8659420289855072, "train_speed(iter/s)": 1.456969 }, { "epoch": 1.3722633991688444, "grad_norm": 2.77667498588562, "learning_rate": 8.253772864998108e-05, "loss": 0.1545766830444336, "memory(GiB)": 70.5, "step": 32030, "token_acc": 0.9612903225806452, "train_speed(iter/s)": 1.456994 }, { "epoch": 1.3724776144980935, "grad_norm": 7.833395004272461, "learning_rate": 8.253261851838426e-05, "loss": 0.5221186637878418, "memory(GiB)": 70.5, "step": 32035, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.456986 }, { "epoch": 1.3726918298273425, "grad_norm": 6.054584980010986, "learning_rate": 8.252750779742855e-05, "loss": 0.33981420993804934, "memory(GiB)": 70.5, "step": 32040, "token_acc": 0.925, "train_speed(iter/s)": 1.45698 }, { "epoch": 1.3729060451565913, "grad_norm": 2.408048152923584, "learning_rate": 8.252239648720652e-05, "loss": 0.5078291416168212, "memory(GiB)": 70.5, "step": 32045, "token_acc": 0.9058171745152355, "train_speed(iter/s)": 1.456987 }, { "epoch": 1.3731202604858403, "grad_norm": 1.74077308177948, "learning_rate": 8.251728458781077e-05, "loss": 0.2731207609176636, "memory(GiB)": 70.5, "step": 32050, "token_acc": 0.9535714285714286, "train_speed(iter/s)": 1.457004 }, { "epoch": 1.3733344758150894, "grad_norm": 1.6594849824905396, "learning_rate": 8.25121720993339e-05, "loss": 0.1647382855415344, "memory(GiB)": 70.5, "step": 32055, "token_acc": 0.9667774086378738, "train_speed(iter/s)": 1.457033 }, { "epoch": 1.3735486911443382, "grad_norm": 2.4297006130218506, "learning_rate": 8.250705902186853e-05, "loss": 0.2204202175140381, "memory(GiB)": 70.5, "step": 32060, "token_acc": 0.958904109589041, "train_speed(iter/s)": 1.45704 }, { "epoch": 1.3737629064735872, "grad_norm": 3.165316343307495, "learning_rate": 8.250194535550731e-05, "loss": 0.7310371398925781, "memory(GiB)": 70.5, "step": 32065, "token_acc": 0.82421875, "train_speed(iter/s)": 1.457068 }, { "epoch": 1.3739771218028363, "grad_norm": 6.0668745040893555, "learning_rate": 8.249683110034283e-05, "loss": 0.2654466390609741, "memory(GiB)": 70.5, "step": 32070, "token_acc": 0.942652329749104, "train_speed(iter/s)": 1.457092 }, { "epoch": 1.374191337132085, "grad_norm": 0.18777452409267426, "learning_rate": 8.249171625646779e-05, "loss": 0.13462849855422973, "memory(GiB)": 70.5, "step": 32075, "token_acc": 0.9717514124293786, "train_speed(iter/s)": 1.457087 }, { "epoch": 1.3744055524613341, "grad_norm": 3.8863282203674316, "learning_rate": 8.248660082397484e-05, "loss": 0.26862483024597167, "memory(GiB)": 70.5, "step": 32080, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.457091 }, { "epoch": 1.3746197677905831, "grad_norm": 2.524604558944702, "learning_rate": 8.248148480295662e-05, "loss": 0.21852569580078124, "memory(GiB)": 70.5, "step": 32085, "token_acc": 0.9594594594594594, "train_speed(iter/s)": 1.457087 }, { "epoch": 1.374833983119832, "grad_norm": 3.3840060234069824, "learning_rate": 8.247636819350584e-05, "loss": 0.37594149112701414, "memory(GiB)": 70.5, "step": 32090, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.457083 }, { "epoch": 1.375048198449081, "grad_norm": 7.884053707122803, "learning_rate": 8.24712509957152e-05, "loss": 0.36580533981323243, "memory(GiB)": 70.5, "step": 32095, "token_acc": 0.9176829268292683, "train_speed(iter/s)": 1.457109 }, { "epoch": 1.37526241377833, "grad_norm": 1.3437950611114502, "learning_rate": 8.246613320967737e-05, "loss": 0.3781019687652588, "memory(GiB)": 70.5, "step": 32100, "token_acc": 0.9155844155844156, "train_speed(iter/s)": 1.457137 }, { "epoch": 1.3754766291075788, "grad_norm": 4.883720397949219, "learning_rate": 8.246101483548508e-05, "loss": 0.8176294326782226, "memory(GiB)": 70.5, "step": 32105, "token_acc": 0.8492647058823529, "train_speed(iter/s)": 1.457163 }, { "epoch": 1.3756908444368279, "grad_norm": 3.8525595664978027, "learning_rate": 8.245589587323106e-05, "loss": 0.5294348239898682, "memory(GiB)": 70.5, "step": 32110, "token_acc": 0.8925619834710744, "train_speed(iter/s)": 1.45717 }, { "epoch": 1.375905059766077, "grad_norm": 6.001770496368408, "learning_rate": 8.245077632300803e-05, "loss": 0.3820990562438965, "memory(GiB)": 70.5, "step": 32115, "token_acc": 0.9189944134078212, "train_speed(iter/s)": 1.457186 }, { "epoch": 1.3761192750953257, "grad_norm": 0.11563972383737564, "learning_rate": 8.244565618490876e-05, "loss": 0.23394761085510254, "memory(GiB)": 70.5, "step": 32120, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.457182 }, { "epoch": 1.3763334904245748, "grad_norm": 1.7527168989181519, "learning_rate": 8.2440535459026e-05, "loss": 0.3112152338027954, "memory(GiB)": 70.5, "step": 32125, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.457176 }, { "epoch": 1.3765477057538238, "grad_norm": 3.780912399291992, "learning_rate": 8.24354141454525e-05, "loss": 0.31051657199859617, "memory(GiB)": 70.5, "step": 32130, "token_acc": 0.92018779342723, "train_speed(iter/s)": 1.457181 }, { "epoch": 1.3767619210830726, "grad_norm": 4.370469570159912, "learning_rate": 8.243029224428104e-05, "loss": 0.5296236991882324, "memory(GiB)": 70.5, "step": 32135, "token_acc": 0.8970189701897019, "train_speed(iter/s)": 1.457166 }, { "epoch": 1.3769761364123216, "grad_norm": 2.074307441711426, "learning_rate": 8.242516975560441e-05, "loss": 0.174199640750885, "memory(GiB)": 70.5, "step": 32140, "token_acc": 0.9373219373219374, "train_speed(iter/s)": 1.457153 }, { "epoch": 1.3771903517415707, "grad_norm": 5.273651599884033, "learning_rate": 8.242004667951542e-05, "loss": 0.6477722644805908, "memory(GiB)": 70.5, "step": 32145, "token_acc": 0.8717105263157895, "train_speed(iter/s)": 1.457156 }, { "epoch": 1.3774045670708195, "grad_norm": 1.6734651327133179, "learning_rate": 8.241492301610688e-05, "loss": 0.15907740592956543, "memory(GiB)": 70.5, "step": 32150, "token_acc": 0.9663299663299664, "train_speed(iter/s)": 1.457163 }, { "epoch": 1.3776187824000685, "grad_norm": 3.82840633392334, "learning_rate": 8.240979876547159e-05, "loss": 0.4010054588317871, "memory(GiB)": 70.5, "step": 32155, "token_acc": 0.9276729559748428, "train_speed(iter/s)": 1.457172 }, { "epoch": 1.3778329977293176, "grad_norm": 2.002253770828247, "learning_rate": 8.240467392770239e-05, "loss": 0.43529567718505857, "memory(GiB)": 70.5, "step": 32160, "token_acc": 0.8934707903780069, "train_speed(iter/s)": 1.45718 }, { "epoch": 1.3780472130585664, "grad_norm": 3.6323773860931396, "learning_rate": 8.239954850289215e-05, "loss": 0.3604447364807129, "memory(GiB)": 70.5, "step": 32165, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.457194 }, { "epoch": 1.3782614283878154, "grad_norm": 3.106717586517334, "learning_rate": 8.239442249113366e-05, "loss": 0.6518895149230957, "memory(GiB)": 70.5, "step": 32170, "token_acc": 0.8515151515151516, "train_speed(iter/s)": 1.457175 }, { "epoch": 1.3784756437170644, "grad_norm": 2.947605609893799, "learning_rate": 8.238929589251984e-05, "loss": 0.4439565658569336, "memory(GiB)": 70.5, "step": 32175, "token_acc": 0.9221183800623053, "train_speed(iter/s)": 1.457179 }, { "epoch": 1.3786898590463132, "grad_norm": 2.819342613220215, "learning_rate": 8.238416870714354e-05, "loss": 0.28369951248168945, "memory(GiB)": 70.5, "step": 32180, "token_acc": 0.9351851851851852, "train_speed(iter/s)": 1.457191 }, { "epoch": 1.3789040743755623, "grad_norm": 1.3997077941894531, "learning_rate": 8.237904093509763e-05, "loss": 0.29419400691986086, "memory(GiB)": 70.5, "step": 32185, "token_acc": 0.9380281690140845, "train_speed(iter/s)": 1.45722 }, { "epoch": 1.3791182897048113, "grad_norm": 3.8012571334838867, "learning_rate": 8.237391257647503e-05, "loss": 0.5045494079589844, "memory(GiB)": 70.5, "step": 32190, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.457228 }, { "epoch": 1.3793325050340601, "grad_norm": 4.8689351081848145, "learning_rate": 8.236878363136864e-05, "loss": 0.5186985015869141, "memory(GiB)": 70.5, "step": 32195, "token_acc": 0.8694158075601375, "train_speed(iter/s)": 1.45723 }, { "epoch": 1.3795467203633092, "grad_norm": 3.273545265197754, "learning_rate": 8.236365409987136e-05, "loss": 0.3814507484436035, "memory(GiB)": 70.5, "step": 32200, "token_acc": 0.9208860759493671, "train_speed(iter/s)": 1.457231 }, { "epoch": 1.3797609356925582, "grad_norm": 3.7436485290527344, "learning_rate": 8.235852398207613e-05, "loss": 0.2816474437713623, "memory(GiB)": 70.5, "step": 32205, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.457248 }, { "epoch": 1.379975151021807, "grad_norm": 3.266841173171997, "learning_rate": 8.235339327807588e-05, "loss": 0.25040698051452637, "memory(GiB)": 70.5, "step": 32210, "token_acc": 0.9682539682539683, "train_speed(iter/s)": 1.457245 }, { "epoch": 1.380189366351056, "grad_norm": 2.162195920944214, "learning_rate": 8.234826198796357e-05, "loss": 0.3920745849609375, "memory(GiB)": 70.5, "step": 32215, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.457249 }, { "epoch": 1.380403581680305, "grad_norm": 5.555407524108887, "learning_rate": 8.234313011183215e-05, "loss": 0.5098192691802979, "memory(GiB)": 70.5, "step": 32220, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.457254 }, { "epoch": 1.380617797009554, "grad_norm": 5.773783206939697, "learning_rate": 8.233799764977457e-05, "loss": 0.38539412021636965, "memory(GiB)": 70.5, "step": 32225, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.457269 }, { "epoch": 1.380832012338803, "grad_norm": 1.874683141708374, "learning_rate": 8.233286460188385e-05, "loss": 0.24951043128967285, "memory(GiB)": 70.5, "step": 32230, "token_acc": 0.94, "train_speed(iter/s)": 1.457269 }, { "epoch": 1.381046227668052, "grad_norm": 4.714671611785889, "learning_rate": 8.232773096825295e-05, "loss": 0.6903680801391602, "memory(GiB)": 70.5, "step": 32235, "token_acc": 0.8545454545454545, "train_speed(iter/s)": 1.457276 }, { "epoch": 1.3812604429973008, "grad_norm": 5.8517255783081055, "learning_rate": 8.232259674897486e-05, "loss": 0.4654864311218262, "memory(GiB)": 70.5, "step": 32240, "token_acc": 0.8803418803418803, "train_speed(iter/s)": 1.457321 }, { "epoch": 1.3814746583265498, "grad_norm": 2.907517433166504, "learning_rate": 8.231746194414262e-05, "loss": 0.4842668533325195, "memory(GiB)": 70.5, "step": 32245, "token_acc": 0.8949416342412452, "train_speed(iter/s)": 1.457315 }, { "epoch": 1.3816888736557988, "grad_norm": 2.86401629447937, "learning_rate": 8.231232655384924e-05, "loss": 0.2550722599029541, "memory(GiB)": 70.5, "step": 32250, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.45731 }, { "epoch": 1.3819030889850477, "grad_norm": 0.7355995774269104, "learning_rate": 8.230719057818776e-05, "loss": 0.1644014835357666, "memory(GiB)": 70.5, "step": 32255, "token_acc": 0.966996699669967, "train_speed(iter/s)": 1.457322 }, { "epoch": 1.3821173043142967, "grad_norm": 2.1709253787994385, "learning_rate": 8.230205401725122e-05, "loss": 0.3010568141937256, "memory(GiB)": 70.5, "step": 32260, "token_acc": 0.9423868312757202, "train_speed(iter/s)": 1.45734 }, { "epoch": 1.3823315196435457, "grad_norm": 5.618229866027832, "learning_rate": 8.229691687113266e-05, "loss": 0.3503615379333496, "memory(GiB)": 70.5, "step": 32265, "token_acc": 0.9280575539568345, "train_speed(iter/s)": 1.457338 }, { "epoch": 1.3825457349727945, "grad_norm": 1.3078358173370361, "learning_rate": 8.229177913992515e-05, "loss": 0.2029940128326416, "memory(GiB)": 70.5, "step": 32270, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.457339 }, { "epoch": 1.3827599503020436, "grad_norm": 2.388556957244873, "learning_rate": 8.228664082372177e-05, "loss": 0.31212592124938965, "memory(GiB)": 70.5, "step": 32275, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.45733 }, { "epoch": 1.3829741656312926, "grad_norm": 5.729292869567871, "learning_rate": 8.22815019226156e-05, "loss": 0.1592286705970764, "memory(GiB)": 70.5, "step": 32280, "token_acc": 0.9606299212598425, "train_speed(iter/s)": 1.457345 }, { "epoch": 1.3831883809605414, "grad_norm": 2.128621816635132, "learning_rate": 8.227636243669976e-05, "loss": 0.37635436058044436, "memory(GiB)": 70.5, "step": 32285, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.45734 }, { "epoch": 1.3834025962897905, "grad_norm": 2.9642577171325684, "learning_rate": 8.227122236606733e-05, "loss": 0.6915351867675781, "memory(GiB)": 70.5, "step": 32290, "token_acc": 0.8745762711864407, "train_speed(iter/s)": 1.457333 }, { "epoch": 1.3836168116190395, "grad_norm": 2.712683916091919, "learning_rate": 8.226608171081143e-05, "loss": 0.24480268955230713, "memory(GiB)": 70.5, "step": 32295, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.457342 }, { "epoch": 1.3838310269482883, "grad_norm": 4.460596084594727, "learning_rate": 8.22609404710252e-05, "loss": 0.6994988441467285, "memory(GiB)": 70.5, "step": 32300, "token_acc": 0.8127090301003345, "train_speed(iter/s)": 1.45738 }, { "epoch": 1.3840452422775373, "grad_norm": 5.2526373863220215, "learning_rate": 8.225579864680175e-05, "loss": 0.281787633895874, "memory(GiB)": 70.5, "step": 32305, "token_acc": 0.9471947194719472, "train_speed(iter/s)": 1.457437 }, { "epoch": 1.3842594576067864, "grad_norm": 4.060218811035156, "learning_rate": 8.225065623823427e-05, "loss": 0.31798791885375977, "memory(GiB)": 70.5, "step": 32310, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.457446 }, { "epoch": 1.3844736729360352, "grad_norm": 3.5831298828125, "learning_rate": 8.224551324541588e-05, "loss": 0.3508453845977783, "memory(GiB)": 70.5, "step": 32315, "token_acc": 0.9131832797427653, "train_speed(iter/s)": 1.45745 }, { "epoch": 1.3846878882652842, "grad_norm": 5.700521945953369, "learning_rate": 8.224036966843978e-05, "loss": 0.8032633781433105, "memory(GiB)": 70.5, "step": 32320, "token_acc": 0.8576051779935275, "train_speed(iter/s)": 1.457456 }, { "epoch": 1.3849021035945333, "grad_norm": 3.8296563625335693, "learning_rate": 8.223522550739913e-05, "loss": 0.8110624313354492, "memory(GiB)": 70.5, "step": 32325, "token_acc": 0.8308823529411765, "train_speed(iter/s)": 1.457508 }, { "epoch": 1.385116318923782, "grad_norm": 2.8741486072540283, "learning_rate": 8.223008076238714e-05, "loss": 0.3082327127456665, "memory(GiB)": 70.5, "step": 32330, "token_acc": 0.9313432835820895, "train_speed(iter/s)": 1.457501 }, { "epoch": 1.385330534253031, "grad_norm": 1.2491176128387451, "learning_rate": 8.2224935433497e-05, "loss": 0.22350528240203857, "memory(GiB)": 70.5, "step": 32335, "token_acc": 0.9295774647887324, "train_speed(iter/s)": 1.457503 }, { "epoch": 1.3855447495822801, "grad_norm": 6.132023334503174, "learning_rate": 8.221978952082192e-05, "loss": 0.41538128852844236, "memory(GiB)": 70.5, "step": 32340, "token_acc": 0.9158249158249159, "train_speed(iter/s)": 1.457517 }, { "epoch": 1.385758964911529, "grad_norm": 2.58559513092041, "learning_rate": 8.221464302445514e-05, "loss": 0.45170011520385744, "memory(GiB)": 70.5, "step": 32345, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.457504 }, { "epoch": 1.385973180240778, "grad_norm": 0.6782761812210083, "learning_rate": 8.220949594448987e-05, "loss": 0.12675669193267822, "memory(GiB)": 70.5, "step": 32350, "token_acc": 0.9607250755287009, "train_speed(iter/s)": 1.457494 }, { "epoch": 1.386187395570027, "grad_norm": 7.788651466369629, "learning_rate": 8.220434828101937e-05, "loss": 0.3112801551818848, "memory(GiB)": 70.5, "step": 32355, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.45751 }, { "epoch": 1.3864016108992758, "grad_norm": 1.8518377542495728, "learning_rate": 8.219920003413691e-05, "loss": 0.29560849666595457, "memory(GiB)": 70.5, "step": 32360, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.457516 }, { "epoch": 1.3866158262285249, "grad_norm": 2.37375807762146, "learning_rate": 8.219405120393571e-05, "loss": 0.1750762939453125, "memory(GiB)": 70.5, "step": 32365, "token_acc": 0.962457337883959, "train_speed(iter/s)": 1.45752 }, { "epoch": 1.386830041557774, "grad_norm": 0.31941303610801697, "learning_rate": 8.218890179050908e-05, "loss": 0.16724418401718139, "memory(GiB)": 70.5, "step": 32370, "token_acc": 0.9664429530201343, "train_speed(iter/s)": 1.457516 }, { "epoch": 1.3870442568870227, "grad_norm": 4.626865386962891, "learning_rate": 8.21837517939503e-05, "loss": 0.32498321533203123, "memory(GiB)": 70.5, "step": 32375, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.457513 }, { "epoch": 1.3872584722162717, "grad_norm": 5.046666145324707, "learning_rate": 8.217860121435267e-05, "loss": 0.6435388565063477, "memory(GiB)": 70.5, "step": 32380, "token_acc": 0.8662420382165605, "train_speed(iter/s)": 1.457515 }, { "epoch": 1.3874726875455208, "grad_norm": 2.330810546875, "learning_rate": 8.217345005180949e-05, "loss": 0.3567062854766846, "memory(GiB)": 70.5, "step": 32385, "token_acc": 0.9331476323119777, "train_speed(iter/s)": 1.457509 }, { "epoch": 1.3876869028747696, "grad_norm": 3.0437517166137695, "learning_rate": 8.216829830641408e-05, "loss": 0.2532304286956787, "memory(GiB)": 70.5, "step": 32390, "token_acc": 0.9513274336283186, "train_speed(iter/s)": 1.45752 }, { "epoch": 1.3879011182040186, "grad_norm": 1.2395830154418945, "learning_rate": 8.216314597825976e-05, "loss": 0.2379124641418457, "memory(GiB)": 70.5, "step": 32395, "token_acc": 0.948051948051948, "train_speed(iter/s)": 1.457539 }, { "epoch": 1.3881153335332677, "grad_norm": 0.2521931827068329, "learning_rate": 8.215799306743989e-05, "loss": 0.415833854675293, "memory(GiB)": 70.5, "step": 32400, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.45754 }, { "epoch": 1.3883295488625165, "grad_norm": 5.775968074798584, "learning_rate": 8.21528395740478e-05, "loss": 0.3893868923187256, "memory(GiB)": 70.5, "step": 32405, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.457548 }, { "epoch": 1.3885437641917655, "grad_norm": 2.998814582824707, "learning_rate": 8.214768549817687e-05, "loss": 0.25188887119293213, "memory(GiB)": 70.5, "step": 32410, "token_acc": 0.9536423841059603, "train_speed(iter/s)": 1.457566 }, { "epoch": 1.3887579795210145, "grad_norm": 1.9365471601486206, "learning_rate": 8.214253083992046e-05, "loss": 0.5819493770599365, "memory(GiB)": 70.5, "step": 32415, "token_acc": 0.8777173913043478, "train_speed(iter/s)": 1.457599 }, { "epoch": 1.3889721948502634, "grad_norm": 3.5333220958709717, "learning_rate": 8.213737559937195e-05, "loss": 0.18404618501663209, "memory(GiB)": 70.5, "step": 32420, "token_acc": 0.9616613418530351, "train_speed(iter/s)": 1.45761 }, { "epoch": 1.3891864101795124, "grad_norm": 3.850618600845337, "learning_rate": 8.213221977662473e-05, "loss": 0.5739490032196045, "memory(GiB)": 70.5, "step": 32425, "token_acc": 0.8721311475409836, "train_speed(iter/s)": 1.457646 }, { "epoch": 1.3894006255087614, "grad_norm": 3.344301700592041, "learning_rate": 8.212706337177221e-05, "loss": 0.4605151653289795, "memory(GiB)": 70.5, "step": 32430, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.457663 }, { "epoch": 1.3896148408380102, "grad_norm": 0.8902029395103455, "learning_rate": 8.21219063849078e-05, "loss": 0.3421536207199097, "memory(GiB)": 70.5, "step": 32435, "token_acc": 0.9312714776632303, "train_speed(iter/s)": 1.45768 }, { "epoch": 1.3898290561672593, "grad_norm": 2.6746323108673096, "learning_rate": 8.211674881612492e-05, "loss": 0.33944244384765626, "memory(GiB)": 70.5, "step": 32440, "token_acc": 0.9343065693430657, "train_speed(iter/s)": 1.457688 }, { "epoch": 1.3900432714965083, "grad_norm": 0.15770979225635529, "learning_rate": 8.211159066551701e-05, "loss": 0.4866902828216553, "memory(GiB)": 70.5, "step": 32445, "token_acc": 0.8978328173374613, "train_speed(iter/s)": 1.457704 }, { "epoch": 1.3902574868257571, "grad_norm": 3.7505221366882324, "learning_rate": 8.210643193317751e-05, "loss": 0.3109349489212036, "memory(GiB)": 70.5, "step": 32450, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.457717 }, { "epoch": 1.3904717021550062, "grad_norm": 2.0908260345458984, "learning_rate": 8.210127261919987e-05, "loss": 0.28436739444732667, "memory(GiB)": 70.5, "step": 32455, "token_acc": 0.9288888888888889, "train_speed(iter/s)": 1.457717 }, { "epoch": 1.3906859174842552, "grad_norm": 2.5882954597473145, "learning_rate": 8.209611272367758e-05, "loss": 0.2871634721755981, "memory(GiB)": 70.5, "step": 32460, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.457705 }, { "epoch": 1.390900132813504, "grad_norm": 6.337554931640625, "learning_rate": 8.209095224670409e-05, "loss": 0.2924015522003174, "memory(GiB)": 70.5, "step": 32465, "token_acc": 0.9338235294117647, "train_speed(iter/s)": 1.457694 }, { "epoch": 1.391114348142753, "grad_norm": 3.0511271953582764, "learning_rate": 8.208579118837289e-05, "loss": 0.35691592693328855, "memory(GiB)": 70.5, "step": 32470, "token_acc": 0.9360902255639098, "train_speed(iter/s)": 1.457698 }, { "epoch": 1.391328563472002, "grad_norm": 2.0219345092773438, "learning_rate": 8.208062954877749e-05, "loss": 0.38450539112091064, "memory(GiB)": 70.5, "step": 32475, "token_acc": 0.9449152542372882, "train_speed(iter/s)": 1.457705 }, { "epoch": 1.3915427788012509, "grad_norm": 3.6567561626434326, "learning_rate": 8.207546732801139e-05, "loss": 0.1799038529396057, "memory(GiB)": 70.5, "step": 32480, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.45771 }, { "epoch": 1.3917569941305, "grad_norm": 2.849738597869873, "learning_rate": 8.207030452616811e-05, "loss": 0.5542884349822998, "memory(GiB)": 70.5, "step": 32485, "token_acc": 0.873015873015873, "train_speed(iter/s)": 1.457722 }, { "epoch": 1.391971209459749, "grad_norm": 3.5457189083099365, "learning_rate": 8.206514114334119e-05, "loss": 0.5737681865692139, "memory(GiB)": 70.5, "step": 32490, "token_acc": 0.8851851851851852, "train_speed(iter/s)": 1.457713 }, { "epoch": 1.392185424788998, "grad_norm": 2.6888997554779053, "learning_rate": 8.205997717962415e-05, "loss": 0.344900107383728, "memory(GiB)": 70.5, "step": 32495, "token_acc": 0.9296875, "train_speed(iter/s)": 1.457716 }, { "epoch": 1.3923996401182468, "grad_norm": 4.539571762084961, "learning_rate": 8.205481263511054e-05, "loss": 0.4072887420654297, "memory(GiB)": 70.5, "step": 32500, "token_acc": 0.912751677852349, "train_speed(iter/s)": 1.457729 }, { "epoch": 1.3923996401182468, "eval_loss": 2.4475386142730713, "eval_runtime": 13.3317, "eval_samples_per_second": 7.501, "eval_steps_per_second": 7.501, "eval_token_acc": 0.4547945205479452, "step": 32500 }, { "epoch": 1.3926138554474958, "grad_norm": 2.547537088394165, "learning_rate": 8.204964750989393e-05, "loss": 0.3446155071258545, "memory(GiB)": 70.5, "step": 32505, "token_acc": 0.5884086444007859, "train_speed(iter/s)": 1.456829 }, { "epoch": 1.3928280707767449, "grad_norm": 2.6049139499664307, "learning_rate": 8.204448180406789e-05, "loss": 0.34813761711120605, "memory(GiB)": 70.5, "step": 32510, "token_acc": 0.9375, "train_speed(iter/s)": 1.456854 }, { "epoch": 1.3930422861059937, "grad_norm": 3.0669353008270264, "learning_rate": 8.203931551772602e-05, "loss": 0.5219156742095947, "memory(GiB)": 70.5, "step": 32515, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.456856 }, { "epoch": 1.3932565014352427, "grad_norm": 1.457571029663086, "learning_rate": 8.203414865096188e-05, "loss": 0.7406195640563965, "memory(GiB)": 70.5, "step": 32520, "token_acc": 0.8715596330275229, "train_speed(iter/s)": 1.456862 }, { "epoch": 1.3934707167644917, "grad_norm": 4.963481903076172, "learning_rate": 8.20289812038691e-05, "loss": 0.33655457496643065, "memory(GiB)": 70.5, "step": 32525, "token_acc": 0.9294117647058824, "train_speed(iter/s)": 1.45689 }, { "epoch": 1.3936849320937406, "grad_norm": 3.9330601692199707, "learning_rate": 8.202381317654126e-05, "loss": 0.25961782932281496, "memory(GiB)": 70.5, "step": 32530, "token_acc": 0.9558823529411765, "train_speed(iter/s)": 1.456891 }, { "epoch": 1.3938991474229896, "grad_norm": 4.296410083770752, "learning_rate": 8.201864456907203e-05, "loss": 0.49450244903564455, "memory(GiB)": 70.5, "step": 32535, "token_acc": 0.8954703832752613, "train_speed(iter/s)": 1.456892 }, { "epoch": 1.3941133627522386, "grad_norm": 0.4593694508075714, "learning_rate": 8.201347538155499e-05, "loss": 0.3442013502120972, "memory(GiB)": 70.5, "step": 32540, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.456874 }, { "epoch": 1.3943275780814874, "grad_norm": 2.6535565853118896, "learning_rate": 8.200830561408382e-05, "loss": 0.39291086196899416, "memory(GiB)": 70.5, "step": 32545, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.456875 }, { "epoch": 1.3945417934107365, "grad_norm": 2.3668134212493896, "learning_rate": 8.200313526675218e-05, "loss": 0.2432154893875122, "memory(GiB)": 70.5, "step": 32550, "token_acc": 0.9462025316455697, "train_speed(iter/s)": 1.456875 }, { "epoch": 1.3947560087399855, "grad_norm": 2.0684783458709717, "learning_rate": 8.199796433965373e-05, "loss": 0.28412318229675293, "memory(GiB)": 70.5, "step": 32555, "token_acc": 0.921161825726141, "train_speed(iter/s)": 1.456883 }, { "epoch": 1.3949702240692343, "grad_norm": 3.1799514293670654, "learning_rate": 8.19927928328821e-05, "loss": 0.49066872596740724, "memory(GiB)": 70.5, "step": 32560, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.456902 }, { "epoch": 1.3951844393984834, "grad_norm": 1.3796772956848145, "learning_rate": 8.198762074653104e-05, "loss": 0.24728062152862548, "memory(GiB)": 70.5, "step": 32565, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.456903 }, { "epoch": 1.3953986547277324, "grad_norm": 3.4443199634552, "learning_rate": 8.198244808069424e-05, "loss": 0.5234926223754883, "memory(GiB)": 70.5, "step": 32570, "token_acc": 0.8806451612903226, "train_speed(iter/s)": 1.456899 }, { "epoch": 1.3956128700569812, "grad_norm": 1.9202876091003418, "learning_rate": 8.197727483546539e-05, "loss": 0.43076291084289553, "memory(GiB)": 70.5, "step": 32575, "token_acc": 0.9027355623100304, "train_speed(iter/s)": 1.456922 }, { "epoch": 1.3958270853862302, "grad_norm": 4.18393087387085, "learning_rate": 8.197210101093817e-05, "loss": 0.3352519989013672, "memory(GiB)": 70.5, "step": 32580, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.456924 }, { "epoch": 1.3960413007154793, "grad_norm": 0.8290443420410156, "learning_rate": 8.196692660720638e-05, "loss": 0.2951753854751587, "memory(GiB)": 70.5, "step": 32585, "token_acc": 0.9272151898734177, "train_speed(iter/s)": 1.456946 }, { "epoch": 1.396255516044728, "grad_norm": 1.6768499612808228, "learning_rate": 8.196175162436371e-05, "loss": 0.4942878246307373, "memory(GiB)": 70.5, "step": 32590, "token_acc": 0.8982456140350877, "train_speed(iter/s)": 1.45696 }, { "epoch": 1.3964697313739771, "grad_norm": 4.803144931793213, "learning_rate": 8.195657606250393e-05, "loss": 0.290220308303833, "memory(GiB)": 70.5, "step": 32595, "token_acc": 0.9420731707317073, "train_speed(iter/s)": 1.456975 }, { "epoch": 1.3966839467032262, "grad_norm": 20.355358123779297, "learning_rate": 8.195139992172081e-05, "loss": 0.6406887531280517, "memory(GiB)": 70.5, "step": 32600, "token_acc": 0.8266666666666667, "train_speed(iter/s)": 1.456998 }, { "epoch": 1.396898162032475, "grad_norm": 1.9052033424377441, "learning_rate": 8.19462232021081e-05, "loss": 0.20948572158813478, "memory(GiB)": 70.5, "step": 32605, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.457012 }, { "epoch": 1.397112377361724, "grad_norm": 2.730332374572754, "learning_rate": 8.19410459037596e-05, "loss": 0.425843334197998, "memory(GiB)": 70.5, "step": 32610, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.457006 }, { "epoch": 1.397326592690973, "grad_norm": 0.7484063506126404, "learning_rate": 8.193586802676907e-05, "loss": 0.3086252689361572, "memory(GiB)": 70.5, "step": 32615, "token_acc": 0.926605504587156, "train_speed(iter/s)": 1.456999 }, { "epoch": 1.3975408080202218, "grad_norm": 0.9649738073348999, "learning_rate": 8.193068957123034e-05, "loss": 0.5715395450592041, "memory(GiB)": 70.5, "step": 32620, "token_acc": 0.8875379939209727, "train_speed(iter/s)": 1.456988 }, { "epoch": 1.3977550233494709, "grad_norm": 4.378729343414307, "learning_rate": 8.192551053723721e-05, "loss": 0.5624303817749023, "memory(GiB)": 70.5, "step": 32625, "token_acc": 0.8666666666666667, "train_speed(iter/s)": 1.456977 }, { "epoch": 1.39796923867872, "grad_norm": 2.1343581676483154, "learning_rate": 8.192033092488351e-05, "loss": 0.403475284576416, "memory(GiB)": 70.5, "step": 32630, "token_acc": 0.9022556390977443, "train_speed(iter/s)": 1.456981 }, { "epoch": 1.3981834540079687, "grad_norm": 2.8757150173187256, "learning_rate": 8.191515073426309e-05, "loss": 0.40824112892150877, "memory(GiB)": 70.5, "step": 32635, "token_acc": 0.9141104294478528, "train_speed(iter/s)": 1.456979 }, { "epoch": 1.3983976693372178, "grad_norm": 3.9283571243286133, "learning_rate": 8.190996996546975e-05, "loss": 0.3777120113372803, "memory(GiB)": 70.5, "step": 32640, "token_acc": 0.9124579124579124, "train_speed(iter/s)": 1.456973 }, { "epoch": 1.3986118846664668, "grad_norm": 3.996260404586792, "learning_rate": 8.19047886185974e-05, "loss": 0.5046377182006836, "memory(GiB)": 70.5, "step": 32645, "token_acc": 0.8975265017667845, "train_speed(iter/s)": 1.45698 }, { "epoch": 1.3988260999957156, "grad_norm": 4.874833583831787, "learning_rate": 8.189960669373987e-05, "loss": 0.4222135543823242, "memory(GiB)": 70.5, "step": 32650, "token_acc": 0.9053497942386831, "train_speed(iter/s)": 1.457007 }, { "epoch": 1.3990403153249646, "grad_norm": 3.726600170135498, "learning_rate": 8.189442419099101e-05, "loss": 0.8931366920471191, "memory(GiB)": 70.5, "step": 32655, "token_acc": 0.8311258278145696, "train_speed(iter/s)": 1.457024 }, { "epoch": 1.3992545306542137, "grad_norm": 1.6954745054244995, "learning_rate": 8.188924111044476e-05, "loss": 0.07632827758789062, "memory(GiB)": 70.5, "step": 32660, "token_acc": 0.9758064516129032, "train_speed(iter/s)": 1.457038 }, { "epoch": 1.3994687459834627, "grad_norm": 3.0792980194091797, "learning_rate": 8.188405745219498e-05, "loss": 0.34496588706970216, "memory(GiB)": 70.5, "step": 32665, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.457018 }, { "epoch": 1.3996829613127115, "grad_norm": 6.477052688598633, "learning_rate": 8.18788732163356e-05, "loss": 0.25321207046508787, "memory(GiB)": 70.5, "step": 32670, "token_acc": 0.9327354260089686, "train_speed(iter/s)": 1.457026 }, { "epoch": 1.3998971766419606, "grad_norm": 2.38881516456604, "learning_rate": 8.187368840296052e-05, "loss": 0.3770682096481323, "memory(GiB)": 70.5, "step": 32675, "token_acc": 0.9101449275362319, "train_speed(iter/s)": 1.45703 }, { "epoch": 1.4001113919712096, "grad_norm": 0.14079424738883972, "learning_rate": 8.186850301216368e-05, "loss": 0.20249123573303224, "memory(GiB)": 70.5, "step": 32680, "token_acc": 0.9578544061302682, "train_speed(iter/s)": 1.457041 }, { "epoch": 1.4003256073004584, "grad_norm": 3.4032864570617676, "learning_rate": 8.186331704403902e-05, "loss": 0.4315175533294678, "memory(GiB)": 70.5, "step": 32685, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.457041 }, { "epoch": 1.4005398226297074, "grad_norm": 1.13609778881073, "learning_rate": 8.185813049868045e-05, "loss": 0.23426132202148436, "memory(GiB)": 70.5, "step": 32690, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.457065 }, { "epoch": 1.4007540379589565, "grad_norm": 1.9269152879714966, "learning_rate": 8.185294337618198e-05, "loss": 0.30211195945739744, "memory(GiB)": 70.5, "step": 32695, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.457089 }, { "epoch": 1.4009682532882053, "grad_norm": 3.021683931350708, "learning_rate": 8.184775567663759e-05, "loss": 0.6100616931915284, "memory(GiB)": 70.5, "step": 32700, "token_acc": 0.8937007874015748, "train_speed(iter/s)": 1.457096 }, { "epoch": 1.4011824686174543, "grad_norm": 3.20383882522583, "learning_rate": 8.18425674001412e-05, "loss": 0.27007977962493895, "memory(GiB)": 70.5, "step": 32705, "token_acc": 0.9345238095238095, "train_speed(iter/s)": 1.45712 }, { "epoch": 1.4013966839467034, "grad_norm": 2.496783494949341, "learning_rate": 8.183737854678684e-05, "loss": 0.21017627716064452, "memory(GiB)": 70.5, "step": 32710, "token_acc": 0.9603960396039604, "train_speed(iter/s)": 1.457122 }, { "epoch": 1.4016108992759522, "grad_norm": 6.275294303894043, "learning_rate": 8.183218911666849e-05, "loss": 0.4323254585266113, "memory(GiB)": 70.5, "step": 32715, "token_acc": 0.9068825910931174, "train_speed(iter/s)": 1.457142 }, { "epoch": 1.4018251146052012, "grad_norm": 3.208155393600464, "learning_rate": 8.182699910988018e-05, "loss": 0.5796047687530518, "memory(GiB)": 70.5, "step": 32720, "token_acc": 0.8581081081081081, "train_speed(iter/s)": 1.457182 }, { "epoch": 1.4020393299344502, "grad_norm": 2.095874071121216, "learning_rate": 8.182180852651592e-05, "loss": 0.37890212535858153, "memory(GiB)": 70.5, "step": 32725, "token_acc": 0.9281914893617021, "train_speed(iter/s)": 1.457183 }, { "epoch": 1.402253545263699, "grad_norm": 4.813762187957764, "learning_rate": 8.181661736666974e-05, "loss": 0.4610123634338379, "memory(GiB)": 70.5, "step": 32730, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.457204 }, { "epoch": 1.402467760592948, "grad_norm": 3.4712541103363037, "learning_rate": 8.181142563043572e-05, "loss": 0.3525123357772827, "memory(GiB)": 70.5, "step": 32735, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.457202 }, { "epoch": 1.4026819759221971, "grad_norm": 4.331108093261719, "learning_rate": 8.180623331790785e-05, "loss": 0.42989673614501955, "memory(GiB)": 70.5, "step": 32740, "token_acc": 0.8975265017667845, "train_speed(iter/s)": 1.457224 }, { "epoch": 1.402896191251446, "grad_norm": 0.2908916473388672, "learning_rate": 8.180104042918025e-05, "loss": 0.4041594982147217, "memory(GiB)": 70.5, "step": 32745, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.457226 }, { "epoch": 1.403110406580695, "grad_norm": 4.711930751800537, "learning_rate": 8.179584696434696e-05, "loss": 0.42483906745910643, "memory(GiB)": 70.5, "step": 32750, "token_acc": 0.9067164179104478, "train_speed(iter/s)": 1.457209 }, { "epoch": 1.403324621909944, "grad_norm": 2.454829216003418, "learning_rate": 8.179065292350208e-05, "loss": 0.29645090103149413, "memory(GiB)": 70.5, "step": 32755, "token_acc": 0.9365994236311239, "train_speed(iter/s)": 1.457214 }, { "epoch": 1.4035388372391928, "grad_norm": 2.4790024757385254, "learning_rate": 8.178545830673969e-05, "loss": 0.19789780378341676, "memory(GiB)": 70.5, "step": 32760, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.457208 }, { "epoch": 1.4037530525684419, "grad_norm": 1.606726884841919, "learning_rate": 8.178026311415392e-05, "loss": 0.383199667930603, "memory(GiB)": 70.5, "step": 32765, "token_acc": 0.9241379310344827, "train_speed(iter/s)": 1.457213 }, { "epoch": 1.4039672678976909, "grad_norm": 2.795623779296875, "learning_rate": 8.177506734583886e-05, "loss": 0.21650247573852538, "memory(GiB)": 70.5, "step": 32770, "token_acc": 0.9396825396825397, "train_speed(iter/s)": 1.45722 }, { "epoch": 1.4041814832269397, "grad_norm": 3.936260223388672, "learning_rate": 8.176987100188865e-05, "loss": 0.6849263191223145, "memory(GiB)": 70.5, "step": 32775, "token_acc": 0.8766666666666667, "train_speed(iter/s)": 1.457232 }, { "epoch": 1.4043956985561887, "grad_norm": 3.421041965484619, "learning_rate": 8.176467408239743e-05, "loss": 0.3952858686447144, "memory(GiB)": 70.5, "step": 32780, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.457224 }, { "epoch": 1.4046099138854378, "grad_norm": 4.288143634796143, "learning_rate": 8.175947658745934e-05, "loss": 0.35706090927124023, "memory(GiB)": 70.5, "step": 32785, "token_acc": 0.9104938271604939, "train_speed(iter/s)": 1.457221 }, { "epoch": 1.4048241292146866, "grad_norm": 6.147327899932861, "learning_rate": 8.175427851716855e-05, "loss": 0.6862931251525879, "memory(GiB)": 70.5, "step": 32790, "token_acc": 0.8494983277591973, "train_speed(iter/s)": 1.45724 }, { "epoch": 1.4050383445439356, "grad_norm": 3.724022626876831, "learning_rate": 8.17490798716192e-05, "loss": 0.5375057697296143, "memory(GiB)": 70.5, "step": 32795, "token_acc": 0.875, "train_speed(iter/s)": 1.457235 }, { "epoch": 1.4052525598731846, "grad_norm": 3.0461032390594482, "learning_rate": 8.174388065090548e-05, "loss": 0.5028678894042968, "memory(GiB)": 70.5, "step": 32800, "token_acc": 0.8745644599303136, "train_speed(iter/s)": 1.457257 }, { "epoch": 1.4054667752024335, "grad_norm": 2.680044174194336, "learning_rate": 8.17386808551216e-05, "loss": 0.2678628921508789, "memory(GiB)": 70.5, "step": 32805, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.457253 }, { "epoch": 1.4056809905316825, "grad_norm": 4.957310199737549, "learning_rate": 8.173348048436174e-05, "loss": 0.39209814071655275, "memory(GiB)": 70.5, "step": 32810, "token_acc": 0.9212598425196851, "train_speed(iter/s)": 1.457245 }, { "epoch": 1.4058952058609315, "grad_norm": 2.253152847290039, "learning_rate": 8.17282795387201e-05, "loss": 0.2749802589416504, "memory(GiB)": 70.5, "step": 32815, "token_acc": 0.9377593360995851, "train_speed(iter/s)": 1.457244 }, { "epoch": 1.4061094211901803, "grad_norm": 1.0963762998580933, "learning_rate": 8.172307801829093e-05, "loss": 0.33878130912780763, "memory(GiB)": 70.5, "step": 32820, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.457243 }, { "epoch": 1.4063236365194294, "grad_norm": 3.238893508911133, "learning_rate": 8.171787592316842e-05, "loss": 0.25013983249664307, "memory(GiB)": 70.5, "step": 32825, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.457245 }, { "epoch": 1.4065378518486784, "grad_norm": 2.2937510013580322, "learning_rate": 8.171267325344685e-05, "loss": 0.46530561447143554, "memory(GiB)": 70.5, "step": 32830, "token_acc": 0.9212598425196851, "train_speed(iter/s)": 1.457254 }, { "epoch": 1.4067520671779272, "grad_norm": 3.8610758781433105, "learning_rate": 8.170747000922045e-05, "loss": 0.596051549911499, "memory(GiB)": 70.5, "step": 32835, "token_acc": 0.8657243816254417, "train_speed(iter/s)": 1.457237 }, { "epoch": 1.4069662825071763, "grad_norm": 1.3271775245666504, "learning_rate": 8.17022661905835e-05, "loss": 0.3353103160858154, "memory(GiB)": 70.5, "step": 32840, "token_acc": 0.9382352941176471, "train_speed(iter/s)": 1.457241 }, { "epoch": 1.4071804978364253, "grad_norm": 2.799544334411621, "learning_rate": 8.169706179763023e-05, "loss": 0.5643760204315186, "memory(GiB)": 70.5, "step": 32845, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.457255 }, { "epoch": 1.407394713165674, "grad_norm": 4.37346076965332, "learning_rate": 8.169185683045498e-05, "loss": 0.471873140335083, "memory(GiB)": 70.5, "step": 32850, "token_acc": 0.8832684824902723, "train_speed(iter/s)": 1.457259 }, { "epoch": 1.4076089284949231, "grad_norm": 4.481375217437744, "learning_rate": 8.168665128915201e-05, "loss": 0.3180342197418213, "memory(GiB)": 70.5, "step": 32855, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.457257 }, { "epoch": 1.4078231438241722, "grad_norm": 2.1614296436309814, "learning_rate": 8.168144517381562e-05, "loss": 0.26809492111206057, "memory(GiB)": 70.5, "step": 32860, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.457292 }, { "epoch": 1.408037359153421, "grad_norm": 0.4920479357242584, "learning_rate": 8.167623848454014e-05, "loss": 0.12593677043914794, "memory(GiB)": 70.5, "step": 32865, "token_acc": 0.9607142857142857, "train_speed(iter/s)": 1.457294 }, { "epoch": 1.40825157448267, "grad_norm": 3.4773716926574707, "learning_rate": 8.167103122141988e-05, "loss": 0.3188636064529419, "memory(GiB)": 70.5, "step": 32870, "token_acc": 0.9369627507163324, "train_speed(iter/s)": 1.457311 }, { "epoch": 1.408465789811919, "grad_norm": 2.781749963760376, "learning_rate": 8.166582338454917e-05, "loss": 0.3772738456726074, "memory(GiB)": 70.5, "step": 32875, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.457307 }, { "epoch": 1.4086800051411679, "grad_norm": 0.3788779079914093, "learning_rate": 8.166061497402236e-05, "loss": 0.49984011650085447, "memory(GiB)": 70.5, "step": 32880, "token_acc": 0.89272030651341, "train_speed(iter/s)": 1.45733 }, { "epoch": 1.408894220470417, "grad_norm": 1.7798092365264893, "learning_rate": 8.165540598993383e-05, "loss": 0.36661269664764407, "memory(GiB)": 70.5, "step": 32885, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.457315 }, { "epoch": 1.409108435799666, "grad_norm": 3.2622766494750977, "learning_rate": 8.165019643237792e-05, "loss": 0.36964504718780516, "memory(GiB)": 70.5, "step": 32890, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.457327 }, { "epoch": 1.4093226511289147, "grad_norm": 5.36054801940918, "learning_rate": 8.164498630144901e-05, "loss": 0.3593886375427246, "memory(GiB)": 70.5, "step": 32895, "token_acc": 0.9036544850498339, "train_speed(iter/s)": 1.457317 }, { "epoch": 1.4095368664581638, "grad_norm": 4.430522441864014, "learning_rate": 8.163977559724147e-05, "loss": 0.341241455078125, "memory(GiB)": 70.5, "step": 32900, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.457329 }, { "epoch": 1.4097510817874128, "grad_norm": 4.38557767868042, "learning_rate": 8.163456431984975e-05, "loss": 0.4161365985870361, "memory(GiB)": 70.5, "step": 32905, "token_acc": 0.9245901639344263, "train_speed(iter/s)": 1.457318 }, { "epoch": 1.4099652971166616, "grad_norm": 0.1920361965894699, "learning_rate": 8.162935246936819e-05, "loss": 0.16424822807312012, "memory(GiB)": 70.5, "step": 32910, "token_acc": 0.9806949806949807, "train_speed(iter/s)": 1.457324 }, { "epoch": 1.4101795124459107, "grad_norm": 6.170867919921875, "learning_rate": 8.162414004589126e-05, "loss": 0.39278039932250974, "memory(GiB)": 70.5, "step": 32915, "token_acc": 0.9018181818181819, "train_speed(iter/s)": 1.457318 }, { "epoch": 1.4103937277751597, "grad_norm": 0.5748429894447327, "learning_rate": 8.161892704951334e-05, "loss": 0.2674154996871948, "memory(GiB)": 70.5, "step": 32920, "token_acc": 0.9277566539923955, "train_speed(iter/s)": 1.457322 }, { "epoch": 1.4106079431044085, "grad_norm": 3.261650323867798, "learning_rate": 8.161371348032893e-05, "loss": 0.5922637462615967, "memory(GiB)": 70.5, "step": 32925, "token_acc": 0.8685258964143426, "train_speed(iter/s)": 1.45731 }, { "epoch": 1.4108221584336575, "grad_norm": 2.762840747833252, "learning_rate": 8.160849933843241e-05, "loss": 0.6604074954986572, "memory(GiB)": 70.5, "step": 32930, "token_acc": 0.8538205980066446, "train_speed(iter/s)": 1.457311 }, { "epoch": 1.4110363737629066, "grad_norm": 2.7803502082824707, "learning_rate": 8.16032846239183e-05, "loss": 0.5291987895965576, "memory(GiB)": 70.5, "step": 32935, "token_acc": 0.8869257950530035, "train_speed(iter/s)": 1.457314 }, { "epoch": 1.4112505890921554, "grad_norm": 3.3659377098083496, "learning_rate": 8.159806933688105e-05, "loss": 0.37460989952087403, "memory(GiB)": 70.5, "step": 32940, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.457309 }, { "epoch": 1.4114648044214044, "grad_norm": 2.013261318206787, "learning_rate": 8.159285347741513e-05, "loss": 0.48163065910339353, "memory(GiB)": 70.5, "step": 32945, "token_acc": 0.9033232628398792, "train_speed(iter/s)": 1.457331 }, { "epoch": 1.4116790197506535, "grad_norm": 2.01153826713562, "learning_rate": 8.1587637045615e-05, "loss": 0.34597163200378417, "memory(GiB)": 70.5, "step": 32950, "token_acc": 0.9206896551724137, "train_speed(iter/s)": 1.457347 }, { "epoch": 1.4118932350799023, "grad_norm": 4.56410551071167, "learning_rate": 8.158242004157522e-05, "loss": 0.3247312784194946, "memory(GiB)": 70.5, "step": 32955, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.457343 }, { "epoch": 1.4121074504091513, "grad_norm": 0.25424665212631226, "learning_rate": 8.157720246539026e-05, "loss": 0.6132190704345704, "memory(GiB)": 70.5, "step": 32960, "token_acc": 0.8770226537216829, "train_speed(iter/s)": 1.457344 }, { "epoch": 1.4123216657384003, "grad_norm": 0.7736417651176453, "learning_rate": 8.157198431715466e-05, "loss": 0.523277759552002, "memory(GiB)": 70.5, "step": 32965, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.457337 }, { "epoch": 1.4125358810676492, "grad_norm": 2.38248872756958, "learning_rate": 8.156676559696294e-05, "loss": 0.4416013717651367, "memory(GiB)": 70.5, "step": 32970, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.457333 }, { "epoch": 1.4127500963968982, "grad_norm": 2.447136163711548, "learning_rate": 8.156154630490968e-05, "loss": 0.35826570987701417, "memory(GiB)": 70.5, "step": 32975, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.457344 }, { "epoch": 1.4129643117261472, "grad_norm": 2.8321547508239746, "learning_rate": 8.155632644108937e-05, "loss": 0.43501968383789064, "memory(GiB)": 70.5, "step": 32980, "token_acc": 0.9296875, "train_speed(iter/s)": 1.457349 }, { "epoch": 1.413178527055396, "grad_norm": 0.6830782294273376, "learning_rate": 8.155110600559661e-05, "loss": 0.17997997999191284, "memory(GiB)": 70.5, "step": 32985, "token_acc": 0.9556962025316456, "train_speed(iter/s)": 1.457358 }, { "epoch": 1.413392742384645, "grad_norm": 3.60085129737854, "learning_rate": 8.154588499852598e-05, "loss": 0.45759124755859376, "memory(GiB)": 70.5, "step": 32990, "token_acc": 0.9182389937106918, "train_speed(iter/s)": 1.457375 }, { "epoch": 1.413606957713894, "grad_norm": 3.2108826637268066, "learning_rate": 8.154066341997204e-05, "loss": 0.26685659885406493, "memory(GiB)": 70.5, "step": 32995, "token_acc": 0.9391304347826087, "train_speed(iter/s)": 1.457372 }, { "epoch": 1.413821173043143, "grad_norm": 2.6011962890625, "learning_rate": 8.15354412700294e-05, "loss": 0.2644827842712402, "memory(GiB)": 70.5, "step": 33000, "token_acc": 0.94140625, "train_speed(iter/s)": 1.457368 }, { "epoch": 1.413821173043143, "eval_loss": 2.6718151569366455, "eval_runtime": 13.4231, "eval_samples_per_second": 7.45, "eval_steps_per_second": 7.45, "eval_token_acc": 0.43631778058007564, "step": 33000 }, { "epoch": 1.414035388372392, "grad_norm": 2.6278209686279297, "learning_rate": 8.153021854879266e-05, "loss": 0.20426197052001954, "memory(GiB)": 70.5, "step": 33005, "token_acc": 0.5555555555555556, "train_speed(iter/s)": 1.456462 }, { "epoch": 1.414249603701641, "grad_norm": 4.254202365875244, "learning_rate": 8.152499525635643e-05, "loss": 0.8451970100402832, "memory(GiB)": 70.5, "step": 33010, "token_acc": 0.8474025974025974, "train_speed(iter/s)": 1.456459 }, { "epoch": 1.4144638190308898, "grad_norm": 3.861721992492676, "learning_rate": 8.151977139281534e-05, "loss": 0.560482120513916, "memory(GiB)": 70.5, "step": 33015, "token_acc": 0.8633093525179856, "train_speed(iter/s)": 1.456447 }, { "epoch": 1.4146780343601388, "grad_norm": 4.984864711761475, "learning_rate": 8.151454695826402e-05, "loss": 0.5471345901489257, "memory(GiB)": 70.5, "step": 33020, "token_acc": 0.8766666666666667, "train_speed(iter/s)": 1.456446 }, { "epoch": 1.4148922496893879, "grad_norm": 1.5953147411346436, "learning_rate": 8.150932195279711e-05, "loss": 0.2694020986557007, "memory(GiB)": 70.5, "step": 33025, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.456456 }, { "epoch": 1.4151064650186367, "grad_norm": 1.8826793432235718, "learning_rate": 8.150409637650928e-05, "loss": 0.27998788356781007, "memory(GiB)": 70.5, "step": 33030, "token_acc": 0.9461279461279462, "train_speed(iter/s)": 1.45646 }, { "epoch": 1.4153206803478857, "grad_norm": 0.56926429271698, "learning_rate": 8.14988702294952e-05, "loss": 0.3492791414260864, "memory(GiB)": 70.5, "step": 33035, "token_acc": 0.92, "train_speed(iter/s)": 1.456458 }, { "epoch": 1.4155348956771348, "grad_norm": 2.54703426361084, "learning_rate": 8.149364351184953e-05, "loss": 0.42732667922973633, "memory(GiB)": 70.5, "step": 33040, "token_acc": 0.8958990536277602, "train_speed(iter/s)": 1.456448 }, { "epoch": 1.4157491110063836, "grad_norm": 0.4269501566886902, "learning_rate": 8.148841622366699e-05, "loss": 0.38133552074432375, "memory(GiB)": 70.5, "step": 33045, "token_acc": 0.9058441558441559, "train_speed(iter/s)": 1.45645 }, { "epoch": 1.4159633263356326, "grad_norm": 4.726733207702637, "learning_rate": 8.148318836504221e-05, "loss": 0.6491533279418945, "memory(GiB)": 70.5, "step": 33050, "token_acc": 0.8691275167785235, "train_speed(iter/s)": 1.456442 }, { "epoch": 1.4161775416648816, "grad_norm": 4.171665668487549, "learning_rate": 8.147795993606997e-05, "loss": 0.36583409309387205, "memory(GiB)": 70.5, "step": 33055, "token_acc": 0.9283018867924528, "train_speed(iter/s)": 1.456456 }, { "epoch": 1.4163917569941304, "grad_norm": 3.8137905597686768, "learning_rate": 8.147273093684493e-05, "loss": 0.5125202178955078, "memory(GiB)": 70.5, "step": 33060, "token_acc": 0.9003021148036254, "train_speed(iter/s)": 1.456466 }, { "epoch": 1.4166059723233795, "grad_norm": 3.9197323322296143, "learning_rate": 8.146750136746187e-05, "loss": 0.36158835887908936, "memory(GiB)": 70.5, "step": 33065, "token_acc": 0.9239543726235742, "train_speed(iter/s)": 1.456476 }, { "epoch": 1.4168201876526285, "grad_norm": 3.2603602409362793, "learning_rate": 8.14622712280155e-05, "loss": 0.28153579235076903, "memory(GiB)": 70.5, "step": 33070, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.456495 }, { "epoch": 1.4170344029818773, "grad_norm": 4.496239185333252, "learning_rate": 8.145704051860056e-05, "loss": 0.5580110549926758, "memory(GiB)": 70.5, "step": 33075, "token_acc": 0.875, "train_speed(iter/s)": 1.456495 }, { "epoch": 1.4172486183111264, "grad_norm": 1.9401124715805054, "learning_rate": 8.145180923931184e-05, "loss": 0.3889272212982178, "memory(GiB)": 70.5, "step": 33080, "token_acc": 0.9299610894941635, "train_speed(iter/s)": 1.456511 }, { "epoch": 1.4174628336403754, "grad_norm": 3.6620824337005615, "learning_rate": 8.144657739024408e-05, "loss": 0.4959586143493652, "memory(GiB)": 70.5, "step": 33085, "token_acc": 0.9018867924528302, "train_speed(iter/s)": 1.456542 }, { "epoch": 1.4176770489696242, "grad_norm": 3.2700705528259277, "learning_rate": 8.144134497149207e-05, "loss": 0.1465437889099121, "memory(GiB)": 70.5, "step": 33090, "token_acc": 0.9710144927536232, "train_speed(iter/s)": 1.45654 }, { "epoch": 1.4178912642988732, "grad_norm": 4.308524131774902, "learning_rate": 8.14361119831506e-05, "loss": 0.31584725379943845, "memory(GiB)": 70.5, "step": 33095, "token_acc": 0.9261538461538461, "train_speed(iter/s)": 1.456532 }, { "epoch": 1.4181054796281223, "grad_norm": 5.512431621551514, "learning_rate": 8.143087842531447e-05, "loss": 0.41625151634216306, "memory(GiB)": 70.5, "step": 33100, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.456558 }, { "epoch": 1.418319694957371, "grad_norm": 2.57415509223938, "learning_rate": 8.142564429807851e-05, "loss": 0.4085883617401123, "memory(GiB)": 70.5, "step": 33105, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.456555 }, { "epoch": 1.4185339102866201, "grad_norm": 11.143917083740234, "learning_rate": 8.142040960153749e-05, "loss": 0.2902386665344238, "memory(GiB)": 70.5, "step": 33110, "token_acc": 0.9305019305019305, "train_speed(iter/s)": 1.456543 }, { "epoch": 1.4187481256158692, "grad_norm": 2.3700687885284424, "learning_rate": 8.141517433578632e-05, "loss": 0.4735823631286621, "memory(GiB)": 70.5, "step": 33115, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.45655 }, { "epoch": 1.418962340945118, "grad_norm": 3.43365740776062, "learning_rate": 8.140993850091977e-05, "loss": 0.4420041084289551, "memory(GiB)": 70.5, "step": 33120, "token_acc": 0.9112627986348123, "train_speed(iter/s)": 1.456547 }, { "epoch": 1.419176556274367, "grad_norm": 5.041223049163818, "learning_rate": 8.140470209703273e-05, "loss": 0.24570038318634033, "memory(GiB)": 70.5, "step": 33125, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.456568 }, { "epoch": 1.419390771603616, "grad_norm": 2.8809797763824463, "learning_rate": 8.139946512422003e-05, "loss": 0.4264055252075195, "memory(GiB)": 70.5, "step": 33130, "token_acc": 0.9198813056379822, "train_speed(iter/s)": 1.456561 }, { "epoch": 1.4196049869328649, "grad_norm": 1.9259886741638184, "learning_rate": 8.139422758257658e-05, "loss": 0.6078725337982178, "memory(GiB)": 70.5, "step": 33135, "token_acc": 0.8630952380952381, "train_speed(iter/s)": 1.456539 }, { "epoch": 1.4198192022621139, "grad_norm": 4.0213799476623535, "learning_rate": 8.138898947219724e-05, "loss": 0.44325838088989256, "memory(GiB)": 70.5, "step": 33140, "token_acc": 0.9115646258503401, "train_speed(iter/s)": 1.456543 }, { "epoch": 1.420033417591363, "grad_norm": 2.6651525497436523, "learning_rate": 8.138375079317693e-05, "loss": 0.3328728437423706, "memory(GiB)": 70.5, "step": 33145, "token_acc": 0.9527272727272728, "train_speed(iter/s)": 1.456537 }, { "epoch": 1.4202476329206117, "grad_norm": 1.1902437210083008, "learning_rate": 8.137851154561051e-05, "loss": 0.21490805149078368, "memory(GiB)": 70.5, "step": 33150, "token_acc": 0.9541666666666667, "train_speed(iter/s)": 1.456545 }, { "epoch": 1.4204618482498608, "grad_norm": 2.616215229034424, "learning_rate": 8.137327172959294e-05, "loss": 0.4864184379577637, "memory(GiB)": 70.5, "step": 33155, "token_acc": 0.8955223880597015, "train_speed(iter/s)": 1.456571 }, { "epoch": 1.4206760635791098, "grad_norm": 1.6101315021514893, "learning_rate": 8.136803134521912e-05, "loss": 0.4900513648986816, "memory(GiB)": 70.5, "step": 33160, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.456566 }, { "epoch": 1.4208902789083586, "grad_norm": 2.3107597827911377, "learning_rate": 8.136279039258397e-05, "loss": 0.3838516712188721, "memory(GiB)": 70.5, "step": 33165, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.456578 }, { "epoch": 1.4211044942376077, "grad_norm": 4.1138715744018555, "learning_rate": 8.135754887178246e-05, "loss": 0.6679475784301758, "memory(GiB)": 70.5, "step": 33170, "token_acc": 0.8185328185328186, "train_speed(iter/s)": 1.45661 }, { "epoch": 1.4213187095668567, "grad_norm": 4.796355247497559, "learning_rate": 8.135230678290953e-05, "loss": 0.42113866806030276, "memory(GiB)": 70.5, "step": 33175, "token_acc": 0.9153605015673981, "train_speed(iter/s)": 1.456603 }, { "epoch": 1.4215329248961055, "grad_norm": 2.813556671142578, "learning_rate": 8.134706412606014e-05, "loss": 0.35226848125457766, "memory(GiB)": 70.5, "step": 33180, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.456598 }, { "epoch": 1.4217471402253545, "grad_norm": 2.6552600860595703, "learning_rate": 8.134182090132931e-05, "loss": 0.3917742490768433, "memory(GiB)": 70.5, "step": 33185, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.456594 }, { "epoch": 1.4219613555546036, "grad_norm": 5.132579803466797, "learning_rate": 8.133657710881197e-05, "loss": 0.5608185291290283, "memory(GiB)": 70.5, "step": 33190, "token_acc": 0.8910505836575876, "train_speed(iter/s)": 1.456614 }, { "epoch": 1.4221755708838524, "grad_norm": 7.820512771606445, "learning_rate": 8.133133274860316e-05, "loss": 0.4693275451660156, "memory(GiB)": 70.5, "step": 33195, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.456643 }, { "epoch": 1.4223897862131014, "grad_norm": 4.697415828704834, "learning_rate": 8.132608782079785e-05, "loss": 0.350005054473877, "memory(GiB)": 70.5, "step": 33200, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.456648 }, { "epoch": 1.4226040015423504, "grad_norm": 1.7253355979919434, "learning_rate": 8.132084232549107e-05, "loss": 0.4151186466217041, "memory(GiB)": 70.5, "step": 33205, "token_acc": 0.8832116788321168, "train_speed(iter/s)": 1.456653 }, { "epoch": 1.4228182168715993, "grad_norm": 3.0471253395080566, "learning_rate": 8.131559626277784e-05, "loss": 0.4329203128814697, "memory(GiB)": 70.5, "step": 33210, "token_acc": 0.8944281524926686, "train_speed(iter/s)": 1.456668 }, { "epoch": 1.4230324322008483, "grad_norm": 4.059229850769043, "learning_rate": 8.131034963275324e-05, "loss": 0.4346205711364746, "memory(GiB)": 70.5, "step": 33215, "token_acc": 0.9147540983606557, "train_speed(iter/s)": 1.456668 }, { "epoch": 1.4232466475300973, "grad_norm": 0.4543757736682892, "learning_rate": 8.130510243551227e-05, "loss": 0.2613328218460083, "memory(GiB)": 70.5, "step": 33220, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.456677 }, { "epoch": 1.4234608628593461, "grad_norm": 3.9995007514953613, "learning_rate": 8.129985467115e-05, "loss": 0.2922301054000854, "memory(GiB)": 70.5, "step": 33225, "token_acc": 0.9331210191082803, "train_speed(iter/s)": 1.456681 }, { "epoch": 1.4236750781885952, "grad_norm": 3.3030245304107666, "learning_rate": 8.12946063397615e-05, "loss": 0.33501811027526857, "memory(GiB)": 70.5, "step": 33230, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.456703 }, { "epoch": 1.4238892935178442, "grad_norm": 2.9223225116729736, "learning_rate": 8.128935744144186e-05, "loss": 0.3627391576766968, "memory(GiB)": 70.5, "step": 33235, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.456689 }, { "epoch": 1.424103508847093, "grad_norm": 4.958597183227539, "learning_rate": 8.128410797628615e-05, "loss": 0.4882349014282227, "memory(GiB)": 70.5, "step": 33240, "token_acc": 0.9020979020979021, "train_speed(iter/s)": 1.456687 }, { "epoch": 1.424317724176342, "grad_norm": 6.08347225189209, "learning_rate": 8.127885794438947e-05, "loss": 0.45880517959594724, "memory(GiB)": 70.5, "step": 33245, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.456693 }, { "epoch": 1.424531939505591, "grad_norm": 3.7669527530670166, "learning_rate": 8.127360734584695e-05, "loss": 0.5617469787597656, "memory(GiB)": 70.5, "step": 33250, "token_acc": 0.881159420289855, "train_speed(iter/s)": 1.456688 }, { "epoch": 1.42474615483484, "grad_norm": 3.922896146774292, "learning_rate": 8.126835618075368e-05, "loss": 0.18497157096862793, "memory(GiB)": 70.5, "step": 33255, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.456683 }, { "epoch": 1.424960370164089, "grad_norm": 1.4744842052459717, "learning_rate": 8.126310444920482e-05, "loss": 0.4293792724609375, "memory(GiB)": 70.5, "step": 33260, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.456687 }, { "epoch": 1.425174585493338, "grad_norm": 3.2344110012054443, "learning_rate": 8.12578521512955e-05, "loss": 0.2593980312347412, "memory(GiB)": 70.5, "step": 33265, "token_acc": 0.9387096774193548, "train_speed(iter/s)": 1.456691 }, { "epoch": 1.4253888008225868, "grad_norm": 4.012906551361084, "learning_rate": 8.125259928712085e-05, "loss": 0.3547673225402832, "memory(GiB)": 70.5, "step": 33270, "token_acc": 0.9300411522633745, "train_speed(iter/s)": 1.456698 }, { "epoch": 1.4256030161518358, "grad_norm": 5.272249698638916, "learning_rate": 8.124734585677606e-05, "loss": 0.5169068336486816, "memory(GiB)": 70.5, "step": 33275, "token_acc": 0.8929663608562691, "train_speed(iter/s)": 1.456694 }, { "epoch": 1.4258172314810849, "grad_norm": 0.06140856444835663, "learning_rate": 8.124209186035627e-05, "loss": 0.3069688558578491, "memory(GiB)": 70.5, "step": 33280, "token_acc": 0.937984496124031, "train_speed(iter/s)": 1.456671 }, { "epoch": 1.4260314468103337, "grad_norm": 0.9538911581039429, "learning_rate": 8.12368372979567e-05, "loss": 0.32826344966888427, "memory(GiB)": 70.5, "step": 33285, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.456661 }, { "epoch": 1.4262456621395827, "grad_norm": 0.8089777827262878, "learning_rate": 8.12315821696725e-05, "loss": 0.2702775001525879, "memory(GiB)": 70.5, "step": 33290, "token_acc": 0.9405204460966543, "train_speed(iter/s)": 1.456657 }, { "epoch": 1.4264598774688317, "grad_norm": 3.5258893966674805, "learning_rate": 8.122632647559891e-05, "loss": 0.3091012239456177, "memory(GiB)": 70.5, "step": 33295, "token_acc": 0.9305135951661632, "train_speed(iter/s)": 1.456654 }, { "epoch": 1.4266740927980806, "grad_norm": 3.185558795928955, "learning_rate": 8.122107021583112e-05, "loss": 0.498584508895874, "memory(GiB)": 70.5, "step": 33300, "token_acc": 0.8811881188118812, "train_speed(iter/s)": 1.456646 }, { "epoch": 1.4268883081273296, "grad_norm": 2.993779420852661, "learning_rate": 8.121581339046433e-05, "loss": 0.5780401229858398, "memory(GiB)": 70.5, "step": 33305, "token_acc": 0.8575949367088608, "train_speed(iter/s)": 1.456659 }, { "epoch": 1.4271025234565786, "grad_norm": 2.867169141769409, "learning_rate": 8.121055599959382e-05, "loss": 0.5301659584045411, "memory(GiB)": 70.5, "step": 33310, "token_acc": 0.8807017543859649, "train_speed(iter/s)": 1.456653 }, { "epoch": 1.4273167387858274, "grad_norm": 4.870437145233154, "learning_rate": 8.120529804331482e-05, "loss": 0.5494884490966797, "memory(GiB)": 70.5, "step": 33315, "token_acc": 0.8863636363636364, "train_speed(iter/s)": 1.456648 }, { "epoch": 1.4275309541150765, "grad_norm": 0.18666736781597137, "learning_rate": 8.120003952172255e-05, "loss": 0.14410074949264526, "memory(GiB)": 70.5, "step": 33320, "token_acc": 0.9715302491103203, "train_speed(iter/s)": 1.456649 }, { "epoch": 1.4277451694443255, "grad_norm": 3.3726439476013184, "learning_rate": 8.119478043491232e-05, "loss": 0.2851261138916016, "memory(GiB)": 70.5, "step": 33325, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.456639 }, { "epoch": 1.4279593847735743, "grad_norm": 5.166472911834717, "learning_rate": 8.118952078297936e-05, "loss": 0.443618631362915, "memory(GiB)": 70.5, "step": 33330, "token_acc": 0.9075144508670521, "train_speed(iter/s)": 1.456654 }, { "epoch": 1.4281736001028233, "grad_norm": 3.4901366233825684, "learning_rate": 8.118426056601897e-05, "loss": 0.5122519493103027, "memory(GiB)": 70.5, "step": 33335, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.456647 }, { "epoch": 1.4283878154320724, "grad_norm": 3.2845053672790527, "learning_rate": 8.117899978412646e-05, "loss": 0.649764108657837, "memory(GiB)": 70.5, "step": 33340, "token_acc": 0.8607594936708861, "train_speed(iter/s)": 1.456673 }, { "epoch": 1.4286020307613212, "grad_norm": 8.528438568115234, "learning_rate": 8.11737384373971e-05, "loss": 0.2757920742034912, "memory(GiB)": 70.5, "step": 33345, "token_acc": 0.9477911646586346, "train_speed(iter/s)": 1.456668 }, { "epoch": 1.4288162460905702, "grad_norm": 2.513190984725952, "learning_rate": 8.116847652592626e-05, "loss": 0.2550288915634155, "memory(GiB)": 70.5, "step": 33350, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.456673 }, { "epoch": 1.4290304614198193, "grad_norm": 4.860166549682617, "learning_rate": 8.11632140498092e-05, "loss": 0.541877555847168, "memory(GiB)": 70.5, "step": 33355, "token_acc": 0.8908450704225352, "train_speed(iter/s)": 1.456653 }, { "epoch": 1.429244676749068, "grad_norm": 4.662451267242432, "learning_rate": 8.115795100914129e-05, "loss": 0.424450159072876, "memory(GiB)": 70.5, "step": 33360, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.456639 }, { "epoch": 1.4294588920783171, "grad_norm": 0.7202897667884827, "learning_rate": 8.115268740401787e-05, "loss": 0.33549063205718993, "memory(GiB)": 70.5, "step": 33365, "token_acc": 0.9293286219081273, "train_speed(iter/s)": 1.456627 }, { "epoch": 1.4296731074075661, "grad_norm": 4.812852382659912, "learning_rate": 8.114742323453431e-05, "loss": 0.3460377216339111, "memory(GiB)": 70.5, "step": 33370, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.456629 }, { "epoch": 1.429887322736815, "grad_norm": 0.2262609899044037, "learning_rate": 8.114215850078595e-05, "loss": 0.35448408126831055, "memory(GiB)": 70.5, "step": 33375, "token_acc": 0.9186046511627907, "train_speed(iter/s)": 1.456639 }, { "epoch": 1.430101538066064, "grad_norm": 2.881704807281494, "learning_rate": 8.113689320286817e-05, "loss": 0.40473618507385256, "memory(GiB)": 70.5, "step": 33380, "token_acc": 0.9227642276422764, "train_speed(iter/s)": 1.456655 }, { "epoch": 1.430315753395313, "grad_norm": 3.7898647785186768, "learning_rate": 8.113162734087636e-05, "loss": 0.44742231369018554, "memory(GiB)": 70.5, "step": 33385, "token_acc": 0.8840579710144928, "train_speed(iter/s)": 1.456687 }, { "epoch": 1.4305299687245618, "grad_norm": 8.647343635559082, "learning_rate": 8.112636091490591e-05, "loss": 0.29366703033447267, "memory(GiB)": 70.5, "step": 33390, "token_acc": 0.9212121212121213, "train_speed(iter/s)": 1.456707 }, { "epoch": 1.4307441840538109, "grad_norm": 5.721720218658447, "learning_rate": 8.112109392505225e-05, "loss": 0.29133455753326415, "memory(GiB)": 70.5, "step": 33395, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.45671 }, { "epoch": 1.43095839938306, "grad_norm": 2.0793070793151855, "learning_rate": 8.111582637141078e-05, "loss": 0.39633493423461913, "memory(GiB)": 70.5, "step": 33400, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.456724 }, { "epoch": 1.4311726147123087, "grad_norm": 3.1133646965026855, "learning_rate": 8.11105582540769e-05, "loss": 0.3854947090148926, "memory(GiB)": 70.5, "step": 33405, "token_acc": 0.9211356466876972, "train_speed(iter/s)": 1.456731 }, { "epoch": 1.4313868300415578, "grad_norm": 2.0668551921844482, "learning_rate": 8.11052895731461e-05, "loss": 0.2726698637008667, "memory(GiB)": 70.5, "step": 33410, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.456749 }, { "epoch": 1.4316010453708068, "grad_norm": 4.300293445587158, "learning_rate": 8.11000203287138e-05, "loss": 0.265415358543396, "memory(GiB)": 70.5, "step": 33415, "token_acc": 0.9387096774193548, "train_speed(iter/s)": 1.456775 }, { "epoch": 1.4318152607000556, "grad_norm": 1.9050976037979126, "learning_rate": 8.109475052087543e-05, "loss": 0.3740545272827148, "memory(GiB)": 70.5, "step": 33420, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.456781 }, { "epoch": 1.4320294760293046, "grad_norm": 5.535609245300293, "learning_rate": 8.108948014972652e-05, "loss": 0.6599618911743164, "memory(GiB)": 70.5, "step": 33425, "token_acc": 0.8582375478927203, "train_speed(iter/s)": 1.456788 }, { "epoch": 1.4322436913585537, "grad_norm": 2.5224719047546387, "learning_rate": 8.108420921536248e-05, "loss": 0.31258816719055177, "memory(GiB)": 70.5, "step": 33430, "token_acc": 0.934375, "train_speed(iter/s)": 1.456799 }, { "epoch": 1.4324579066878025, "grad_norm": 2.7757503986358643, "learning_rate": 8.107893771787885e-05, "loss": 0.2998582601547241, "memory(GiB)": 70.5, "step": 33435, "token_acc": 0.9287925696594427, "train_speed(iter/s)": 1.456804 }, { "epoch": 1.4326721220170515, "grad_norm": 2.9733505249023438, "learning_rate": 8.107366565737112e-05, "loss": 0.3693912267684937, "memory(GiB)": 70.5, "step": 33440, "token_acc": 0.9252336448598131, "train_speed(iter/s)": 1.456815 }, { "epoch": 1.4328863373463006, "grad_norm": 6.551183223724365, "learning_rate": 8.106839303393476e-05, "loss": 0.47687854766845705, "memory(GiB)": 70.5, "step": 33445, "token_acc": 0.8895522388059701, "train_speed(iter/s)": 1.456823 }, { "epoch": 1.4331005526755494, "grad_norm": 12.49964714050293, "learning_rate": 8.106311984766535e-05, "loss": 0.5982562065124511, "memory(GiB)": 70.5, "step": 33450, "token_acc": 0.8603174603174604, "train_speed(iter/s)": 1.45683 }, { "epoch": 1.4333147680047984, "grad_norm": 2.6840474605560303, "learning_rate": 8.105784609865835e-05, "loss": 0.3816049814224243, "memory(GiB)": 70.5, "step": 33455, "token_acc": 0.9157894736842105, "train_speed(iter/s)": 1.45683 }, { "epoch": 1.4335289833340474, "grad_norm": 3.297771692276001, "learning_rate": 8.105257178700935e-05, "loss": 0.2723667621612549, "memory(GiB)": 70.5, "step": 33460, "token_acc": 0.9430379746835443, "train_speed(iter/s)": 1.456837 }, { "epoch": 1.4337431986632962, "grad_norm": 0.8123674392700195, "learning_rate": 8.104729691281387e-05, "loss": 0.13467657566070557, "memory(GiB)": 70.5, "step": 33465, "token_acc": 0.9692982456140351, "train_speed(iter/s)": 1.456827 }, { "epoch": 1.4339574139925453, "grad_norm": 5.575889587402344, "learning_rate": 8.10420214761675e-05, "loss": 0.6048870086669922, "memory(GiB)": 70.5, "step": 33470, "token_acc": 0.8701298701298701, "train_speed(iter/s)": 1.456814 }, { "epoch": 1.4341716293217943, "grad_norm": 4.257656097412109, "learning_rate": 8.103674547716577e-05, "loss": 0.4463034152984619, "memory(GiB)": 70.5, "step": 33475, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.456816 }, { "epoch": 1.4343858446510431, "grad_norm": 9.683680534362793, "learning_rate": 8.10314689159043e-05, "loss": 0.2747670650482178, "memory(GiB)": 70.5, "step": 33480, "token_acc": 0.9328358208955224, "train_speed(iter/s)": 1.456841 }, { "epoch": 1.4346000599802922, "grad_norm": 2.2066428661346436, "learning_rate": 8.102619179247866e-05, "loss": 0.2818820238113403, "memory(GiB)": 70.5, "step": 33485, "token_acc": 0.9533333333333334, "train_speed(iter/s)": 1.456846 }, { "epoch": 1.4348142753095412, "grad_norm": 3.746452808380127, "learning_rate": 8.102091410698445e-05, "loss": 0.5236235618591308, "memory(GiB)": 70.5, "step": 33490, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.456861 }, { "epoch": 1.43502849063879, "grad_norm": 3.1337101459503174, "learning_rate": 8.101563585951728e-05, "loss": 0.4516003131866455, "memory(GiB)": 70.5, "step": 33495, "token_acc": 0.8962962962962963, "train_speed(iter/s)": 1.456873 }, { "epoch": 1.435242705968039, "grad_norm": 4.902766227722168, "learning_rate": 8.101035705017277e-05, "loss": 0.7123597145080567, "memory(GiB)": 70.5, "step": 33500, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.456895 }, { "epoch": 1.435242705968039, "eval_loss": 2.7286128997802734, "eval_runtime": 13.505, "eval_samples_per_second": 7.405, "eval_steps_per_second": 7.405, "eval_token_acc": 0.42005813953488375, "step": 33500 }, { "epoch": 1.435456921297288, "grad_norm": 0.7341150045394897, "learning_rate": 8.100507767904653e-05, "loss": 0.24185795783996583, "memory(GiB)": 70.5, "step": 33505, "token_acc": 0.5766871165644172, "train_speed(iter/s)": 1.455999 }, { "epoch": 1.435671136626537, "grad_norm": 1.1361342668533325, "learning_rate": 8.099979774623425e-05, "loss": 0.22224001884460448, "memory(GiB)": 70.5, "step": 33510, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.455997 }, { "epoch": 1.435885351955786, "grad_norm": 2.02335262298584, "learning_rate": 8.099451725183154e-05, "loss": 0.4811833381652832, "memory(GiB)": 70.5, "step": 33515, "token_acc": 0.8729508196721312, "train_speed(iter/s)": 1.456008 }, { "epoch": 1.436099567285035, "grad_norm": 5.73859167098999, "learning_rate": 8.098923619593409e-05, "loss": 0.35887322425842283, "memory(GiB)": 70.5, "step": 33520, "token_acc": 0.9388379204892966, "train_speed(iter/s)": 1.456002 }, { "epoch": 1.4363137826142838, "grad_norm": 6.555582046508789, "learning_rate": 8.098395457863755e-05, "loss": 0.7895328521728515, "memory(GiB)": 70.5, "step": 33525, "token_acc": 0.8498293515358362, "train_speed(iter/s)": 1.455996 }, { "epoch": 1.4365279979435328, "grad_norm": 6.1171770095825195, "learning_rate": 8.097867240003761e-05, "loss": 0.34111192226409914, "memory(GiB)": 70.5, "step": 33530, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.456013 }, { "epoch": 1.4367422132727818, "grad_norm": 3.1095619201660156, "learning_rate": 8.097338966022993e-05, "loss": 0.27434310913085935, "memory(GiB)": 70.5, "step": 33535, "token_acc": 0.9564032697547684, "train_speed(iter/s)": 1.45601 }, { "epoch": 1.4369564286020307, "grad_norm": 3.659109354019165, "learning_rate": 8.096810635931026e-05, "loss": 0.20297684669494628, "memory(GiB)": 70.5, "step": 33540, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.456012 }, { "epoch": 1.4371706439312797, "grad_norm": 2.7900099754333496, "learning_rate": 8.09628224973743e-05, "loss": 0.48868231773376464, "memory(GiB)": 70.5, "step": 33545, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.45602 }, { "epoch": 1.4373848592605287, "grad_norm": 2.8152525424957275, "learning_rate": 8.095753807451777e-05, "loss": 0.28028852939605714, "memory(GiB)": 70.5, "step": 33550, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.456028 }, { "epoch": 1.4375990745897775, "grad_norm": 1.67078697681427, "learning_rate": 8.095225309083638e-05, "loss": 0.3844248533248901, "memory(GiB)": 70.5, "step": 33555, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.45603 }, { "epoch": 1.4378132899190266, "grad_norm": 1.7426589727401733, "learning_rate": 8.094696754642591e-05, "loss": 0.3112859487533569, "memory(GiB)": 70.5, "step": 33560, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.456032 }, { "epoch": 1.4380275052482756, "grad_norm": 3.569446325302124, "learning_rate": 8.094168144138208e-05, "loss": 0.44565935134887696, "memory(GiB)": 70.5, "step": 33565, "token_acc": 0.8878205128205128, "train_speed(iter/s)": 1.456025 }, { "epoch": 1.4382417205775244, "grad_norm": 3.342331886291504, "learning_rate": 8.093639477580066e-05, "loss": 0.4537001132965088, "memory(GiB)": 70.5, "step": 33570, "token_acc": 0.8912280701754386, "train_speed(iter/s)": 1.456037 }, { "epoch": 1.4384559359067735, "grad_norm": 3.325000524520874, "learning_rate": 8.093110754977744e-05, "loss": 0.39541327953338623, "memory(GiB)": 70.5, "step": 33575, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.456036 }, { "epoch": 1.4386701512360225, "grad_norm": 9.145515441894531, "learning_rate": 8.092581976340819e-05, "loss": 0.5258680820465088, "memory(GiB)": 70.5, "step": 33580, "token_acc": 0.8525641025641025, "train_speed(iter/s)": 1.456071 }, { "epoch": 1.4388843665652713, "grad_norm": 3.780531406402588, "learning_rate": 8.09205314167887e-05, "loss": 0.23837792873382568, "memory(GiB)": 70.5, "step": 33585, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.456067 }, { "epoch": 1.4390985818945203, "grad_norm": 4.677660942077637, "learning_rate": 8.091524251001476e-05, "loss": 0.3866374731063843, "memory(GiB)": 70.5, "step": 33590, "token_acc": 0.9330708661417323, "train_speed(iter/s)": 1.456076 }, { "epoch": 1.4393127972237694, "grad_norm": 0.9472354650497437, "learning_rate": 8.090995304318224e-05, "loss": 0.2745575189590454, "memory(GiB)": 70.5, "step": 33595, "token_acc": 0.9501915708812261, "train_speed(iter/s)": 1.456076 }, { "epoch": 1.4395270125530182, "grad_norm": 3.2153518199920654, "learning_rate": 8.090466301638688e-05, "loss": 0.23937056064605713, "memory(GiB)": 70.5, "step": 33600, "token_acc": 0.953125, "train_speed(iter/s)": 1.456104 }, { "epoch": 1.4397412278822672, "grad_norm": 2.8721976280212402, "learning_rate": 8.089937242972459e-05, "loss": 0.27277967929840086, "memory(GiB)": 70.5, "step": 33605, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.456107 }, { "epoch": 1.4399554432115163, "grad_norm": 3.668402671813965, "learning_rate": 8.089408128329118e-05, "loss": 0.22743620872497558, "memory(GiB)": 70.5, "step": 33610, "token_acc": 0.9336283185840708, "train_speed(iter/s)": 1.456099 }, { "epoch": 1.440169658540765, "grad_norm": 4.01747465133667, "learning_rate": 8.088878957718249e-05, "loss": 0.2968353509902954, "memory(GiB)": 70.5, "step": 33615, "token_acc": 0.945054945054945, "train_speed(iter/s)": 1.456098 }, { "epoch": 1.440383873870014, "grad_norm": 1.306504487991333, "learning_rate": 8.088349731149441e-05, "loss": 0.17838187217712403, "memory(GiB)": 70.5, "step": 33620, "token_acc": 0.9587020648967551, "train_speed(iter/s)": 1.456105 }, { "epoch": 1.4405980891992631, "grad_norm": 4.575624942779541, "learning_rate": 8.087820448632282e-05, "loss": 0.5588573455810547, "memory(GiB)": 70.5, "step": 33625, "token_acc": 0.8634686346863468, "train_speed(iter/s)": 1.456122 }, { "epoch": 1.440812304528512, "grad_norm": 5.0717573165893555, "learning_rate": 8.087291110176355e-05, "loss": 0.5713727951049805, "memory(GiB)": 70.5, "step": 33630, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.456128 }, { "epoch": 1.441026519857761, "grad_norm": 0.8173125982284546, "learning_rate": 8.086761715791258e-05, "loss": 0.27412593364715576, "memory(GiB)": 70.5, "step": 33635, "token_acc": 0.9322709163346613, "train_speed(iter/s)": 1.45613 }, { "epoch": 1.44124073518701, "grad_norm": 2.2936458587646484, "learning_rate": 8.086232265486576e-05, "loss": 0.4175689220428467, "memory(GiB)": 70.5, "step": 33640, "token_acc": 0.9174917491749175, "train_speed(iter/s)": 1.456131 }, { "epoch": 1.4414549505162588, "grad_norm": 5.615164279937744, "learning_rate": 8.085702759271899e-05, "loss": 0.37200117111206055, "memory(GiB)": 70.5, "step": 33645, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.456127 }, { "epoch": 1.4416691658455079, "grad_norm": 2.73396635055542, "learning_rate": 8.085173197156824e-05, "loss": 0.1378488063812256, "memory(GiB)": 70.5, "step": 33650, "token_acc": 0.965034965034965, "train_speed(iter/s)": 1.456125 }, { "epoch": 1.441883381174757, "grad_norm": 3.242422103881836, "learning_rate": 8.084643579150941e-05, "loss": 0.34174256324768065, "memory(GiB)": 70.5, "step": 33655, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.456115 }, { "epoch": 1.4420975965040057, "grad_norm": 0.9703684449195862, "learning_rate": 8.084113905263846e-05, "loss": 0.6047898769378662, "memory(GiB)": 70.5, "step": 33660, "token_acc": 0.910394265232975, "train_speed(iter/s)": 1.456114 }, { "epoch": 1.4423118118332547, "grad_norm": 2.6682159900665283, "learning_rate": 8.083584175505134e-05, "loss": 0.37187535762786866, "memory(GiB)": 70.5, "step": 33665, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.456139 }, { "epoch": 1.4425260271625038, "grad_norm": 4.066932201385498, "learning_rate": 8.083054389884405e-05, "loss": 0.3636352777481079, "memory(GiB)": 70.5, "step": 33670, "token_acc": 0.9399293286219081, "train_speed(iter/s)": 1.456177 }, { "epoch": 1.4427402424917526, "grad_norm": 1.84194815158844, "learning_rate": 8.082524548411252e-05, "loss": 0.16236608028411864, "memory(GiB)": 70.5, "step": 33675, "token_acc": 0.9674267100977199, "train_speed(iter/s)": 1.456177 }, { "epoch": 1.4429544578210016, "grad_norm": 1.6201813220977783, "learning_rate": 8.081994651095273e-05, "loss": 0.13873189687728882, "memory(GiB)": 70.5, "step": 33680, "token_acc": 0.9667774086378738, "train_speed(iter/s)": 1.456183 }, { "epoch": 1.4431686731502507, "grad_norm": 1.030038833618164, "learning_rate": 8.081464697946072e-05, "loss": 0.5362287998199463, "memory(GiB)": 70.5, "step": 33685, "token_acc": 0.8728323699421965, "train_speed(iter/s)": 1.456173 }, { "epoch": 1.4433828884794995, "grad_norm": 1.8253769874572754, "learning_rate": 8.080934688973248e-05, "loss": 0.24431936740875243, "memory(GiB)": 70.5, "step": 33690, "token_acc": 0.949685534591195, "train_speed(iter/s)": 1.456177 }, { "epoch": 1.4435971038087485, "grad_norm": 2.6905486583709717, "learning_rate": 8.0804046241864e-05, "loss": 0.4290763378143311, "memory(GiB)": 70.5, "step": 33695, "token_acc": 0.90234375, "train_speed(iter/s)": 1.456189 }, { "epoch": 1.4438113191379975, "grad_norm": 2.474961996078491, "learning_rate": 8.079874503595133e-05, "loss": 0.18717256784439087, "memory(GiB)": 70.5, "step": 33700, "token_acc": 0.9554896142433235, "train_speed(iter/s)": 1.456193 }, { "epoch": 1.4440255344672464, "grad_norm": 4.682316780090332, "learning_rate": 8.079344327209051e-05, "loss": 0.381260085105896, "memory(GiB)": 70.5, "step": 33705, "token_acc": 0.9235474006116208, "train_speed(iter/s)": 1.456194 }, { "epoch": 1.4442397497964954, "grad_norm": 5.860673904418945, "learning_rate": 8.078814095037758e-05, "loss": 0.38004865646362307, "memory(GiB)": 70.5, "step": 33710, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.456201 }, { "epoch": 1.4444539651257444, "grad_norm": 3.749908208847046, "learning_rate": 8.078283807090858e-05, "loss": 0.4054743766784668, "memory(GiB)": 70.5, "step": 33715, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.456211 }, { "epoch": 1.4446681804549932, "grad_norm": 8.038121223449707, "learning_rate": 8.077753463377962e-05, "loss": 0.586198091506958, "memory(GiB)": 70.5, "step": 33720, "token_acc": 0.8722044728434505, "train_speed(iter/s)": 1.456254 }, { "epoch": 1.4448823957842423, "grad_norm": 4.293758869171143, "learning_rate": 8.07722306390867e-05, "loss": 0.235148286819458, "memory(GiB)": 70.5, "step": 33725, "token_acc": 0.9438943894389439, "train_speed(iter/s)": 1.456257 }, { "epoch": 1.4450966111134913, "grad_norm": 2.3492302894592285, "learning_rate": 8.0766926086926e-05, "loss": 0.40038213729858396, "memory(GiB)": 70.5, "step": 33730, "token_acc": 0.912621359223301, "train_speed(iter/s)": 1.456254 }, { "epoch": 1.4453108264427401, "grad_norm": 3.4287564754486084, "learning_rate": 8.076162097739356e-05, "loss": 0.5315575122833252, "memory(GiB)": 70.5, "step": 33735, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.456252 }, { "epoch": 1.4455250417719891, "grad_norm": 4.09869384765625, "learning_rate": 8.07563153105855e-05, "loss": 0.35046753883361814, "memory(GiB)": 70.5, "step": 33740, "token_acc": 0.9134615384615384, "train_speed(iter/s)": 1.456247 }, { "epoch": 1.4457392571012382, "grad_norm": 3.457951784133911, "learning_rate": 8.075100908659793e-05, "loss": 0.5925482749938965, "memory(GiB)": 70.5, "step": 33745, "token_acc": 0.9053030303030303, "train_speed(iter/s)": 1.456246 }, { "epoch": 1.445953472430487, "grad_norm": 4.101690292358398, "learning_rate": 8.074570230552698e-05, "loss": 0.41150741577148436, "memory(GiB)": 70.5, "step": 33750, "token_acc": 0.913312693498452, "train_speed(iter/s)": 1.45625 }, { "epoch": 1.446167687759736, "grad_norm": 4.513691425323486, "learning_rate": 8.074039496746881e-05, "loss": 0.3407751083374023, "memory(GiB)": 70.5, "step": 33755, "token_acc": 0.9148148148148149, "train_speed(iter/s)": 1.456278 }, { "epoch": 1.446381903088985, "grad_norm": 5.3126115798950195, "learning_rate": 8.073508707251953e-05, "loss": 0.2110121250152588, "memory(GiB)": 70.5, "step": 33760, "token_acc": 0.9511278195488722, "train_speed(iter/s)": 1.45628 }, { "epoch": 1.4465961184182339, "grad_norm": 5.123061656951904, "learning_rate": 8.072977862077532e-05, "loss": 0.3695167303085327, "memory(GiB)": 70.5, "step": 33765, "token_acc": 0.9182879377431906, "train_speed(iter/s)": 1.456267 }, { "epoch": 1.446810333747483, "grad_norm": 3.7912518978118896, "learning_rate": 8.072446961233236e-05, "loss": 0.2515966176986694, "memory(GiB)": 70.5, "step": 33770, "token_acc": 0.9456066945606695, "train_speed(iter/s)": 1.45626 }, { "epoch": 1.447024549076732, "grad_norm": 0.5082888007164001, "learning_rate": 8.07191600472868e-05, "loss": 0.47006540298461913, "memory(GiB)": 70.5, "step": 33775, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.456283 }, { "epoch": 1.4472387644059808, "grad_norm": 0.20798468589782715, "learning_rate": 8.071384992573482e-05, "loss": 0.2752924919128418, "memory(GiB)": 70.5, "step": 33780, "token_acc": 0.9570200573065902, "train_speed(iter/s)": 1.456275 }, { "epoch": 1.4474529797352298, "grad_norm": 1.769346833229065, "learning_rate": 8.070853924777266e-05, "loss": 0.5852202415466309, "memory(GiB)": 70.5, "step": 33785, "token_acc": 0.8782287822878229, "train_speed(iter/s)": 1.456281 }, { "epoch": 1.4476671950644788, "grad_norm": 1.904242992401123, "learning_rate": 8.070322801349649e-05, "loss": 0.24972946643829347, "memory(GiB)": 70.5, "step": 33790, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.456295 }, { "epoch": 1.4478814103937276, "grad_norm": 0.2259232997894287, "learning_rate": 8.069791622300255e-05, "loss": 0.31232903003692625, "memory(GiB)": 70.5, "step": 33795, "token_acc": 0.9470404984423676, "train_speed(iter/s)": 1.456289 }, { "epoch": 1.4480956257229767, "grad_norm": 1.5060168504714966, "learning_rate": 8.069260387638705e-05, "loss": 0.49539756774902344, "memory(GiB)": 70.5, "step": 33800, "token_acc": 0.896, "train_speed(iter/s)": 1.456278 }, { "epoch": 1.4483098410522257, "grad_norm": 3.077190637588501, "learning_rate": 8.068729097374626e-05, "loss": 0.1791548490524292, "memory(GiB)": 70.5, "step": 33805, "token_acc": 0.9635761589403974, "train_speed(iter/s)": 1.456282 }, { "epoch": 1.4485240563814745, "grad_norm": 1.4435697793960571, "learning_rate": 8.068197751517638e-05, "loss": 0.27458696365356444, "memory(GiB)": 70.5, "step": 33810, "token_acc": 0.93, "train_speed(iter/s)": 1.456291 }, { "epoch": 1.4487382717107236, "grad_norm": 6.262722015380859, "learning_rate": 8.067666350077372e-05, "loss": 0.6963133811950684, "memory(GiB)": 70.5, "step": 33815, "token_acc": 0.8611111111111112, "train_speed(iter/s)": 1.456289 }, { "epoch": 1.4489524870399726, "grad_norm": 6.317347526550293, "learning_rate": 8.067134893063452e-05, "loss": 0.5266135215759278, "memory(GiB)": 70.5, "step": 33820, "token_acc": 0.9016949152542373, "train_speed(iter/s)": 1.4563 }, { "epoch": 1.4491667023692214, "grad_norm": 4.713284015655518, "learning_rate": 8.066603380485505e-05, "loss": 0.47337779998779295, "memory(GiB)": 70.5, "step": 33825, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.456319 }, { "epoch": 1.4493809176984704, "grad_norm": 2.6152102947235107, "learning_rate": 8.066071812353162e-05, "loss": 0.5634575366973877, "memory(GiB)": 70.5, "step": 33830, "token_acc": 0.8714285714285714, "train_speed(iter/s)": 1.456345 }, { "epoch": 1.4495951330277195, "grad_norm": 3.727640390396118, "learning_rate": 8.06554018867605e-05, "loss": 0.5154270172119141, "memory(GiB)": 70.5, "step": 33835, "token_acc": 0.8899082568807339, "train_speed(iter/s)": 1.456332 }, { "epoch": 1.4498093483569683, "grad_norm": 3.7586169242858887, "learning_rate": 8.065008509463805e-05, "loss": 0.5000697612762451, "memory(GiB)": 70.5, "step": 33840, "token_acc": 0.8825503355704698, "train_speed(iter/s)": 1.456323 }, { "epoch": 1.4500235636862173, "grad_norm": 6.398438930511475, "learning_rate": 8.064476774726054e-05, "loss": 0.24309883117675782, "memory(GiB)": 70.5, "step": 33845, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.456353 }, { "epoch": 1.4502377790154664, "grad_norm": 2.481947183609009, "learning_rate": 8.063944984472431e-05, "loss": 0.35377686023712157, "memory(GiB)": 70.5, "step": 33850, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.456356 }, { "epoch": 1.4504519943447154, "grad_norm": 2.986947536468506, "learning_rate": 8.063413138712572e-05, "loss": 0.44171648025512694, "memory(GiB)": 70.5, "step": 33855, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.456377 }, { "epoch": 1.4506662096739642, "grad_norm": 2.8188722133636475, "learning_rate": 8.062881237456108e-05, "loss": 0.3406226634979248, "memory(GiB)": 70.5, "step": 33860, "token_acc": 0.9153846153846154, "train_speed(iter/s)": 1.45638 }, { "epoch": 1.4508804250032132, "grad_norm": 1.6642826795578003, "learning_rate": 8.06234928071268e-05, "loss": 0.34090945720672605, "memory(GiB)": 70.5, "step": 33865, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.456382 }, { "epoch": 1.4510946403324623, "grad_norm": 4.7979512214660645, "learning_rate": 8.061817268491919e-05, "loss": 0.5060721397399902, "memory(GiB)": 70.5, "step": 33870, "token_acc": 0.8910256410256411, "train_speed(iter/s)": 1.456387 }, { "epoch": 1.451308855661711, "grad_norm": 1.118110179901123, "learning_rate": 8.061285200803467e-05, "loss": 0.13907543420791627, "memory(GiB)": 70.5, "step": 33875, "token_acc": 0.9739413680781759, "train_speed(iter/s)": 1.456389 }, { "epoch": 1.4515230709909601, "grad_norm": 3.856687068939209, "learning_rate": 8.060753077656964e-05, "loss": 0.21549863815307618, "memory(GiB)": 70.5, "step": 33880, "token_acc": 0.9429824561403509, "train_speed(iter/s)": 1.456398 }, { "epoch": 1.4517372863202092, "grad_norm": 3.5219266414642334, "learning_rate": 8.060220899062045e-05, "loss": 0.4658827781677246, "memory(GiB)": 70.5, "step": 33885, "token_acc": 0.9061371841155235, "train_speed(iter/s)": 1.456416 }, { "epoch": 1.451951501649458, "grad_norm": 3.036120653152466, "learning_rate": 8.059688665028355e-05, "loss": 0.48525371551513674, "memory(GiB)": 70.5, "step": 33890, "token_acc": 0.9122257053291536, "train_speed(iter/s)": 1.456436 }, { "epoch": 1.452165716978707, "grad_norm": 10.285018920898438, "learning_rate": 8.059156375565535e-05, "loss": 0.4781638145446777, "memory(GiB)": 70.5, "step": 33895, "token_acc": 0.8975265017667845, "train_speed(iter/s)": 1.45643 }, { "epoch": 1.452379932307956, "grad_norm": 1.9121507406234741, "learning_rate": 8.058624030683226e-05, "loss": 0.35310540199279783, "memory(GiB)": 70.5, "step": 33900, "token_acc": 0.92, "train_speed(iter/s)": 1.456428 }, { "epoch": 1.4525941476372048, "grad_norm": 2.0887839794158936, "learning_rate": 8.058091630391076e-05, "loss": 0.4727607250213623, "memory(GiB)": 70.5, "step": 33905, "token_acc": 0.900709219858156, "train_speed(iter/s)": 1.456435 }, { "epoch": 1.4528083629664539, "grad_norm": 2.384114980697632, "learning_rate": 8.057559174698725e-05, "loss": 0.22896895408630372, "memory(GiB)": 70.5, "step": 33910, "token_acc": 0.9504132231404959, "train_speed(iter/s)": 1.45645 }, { "epoch": 1.453022578295703, "grad_norm": 4.441871166229248, "learning_rate": 8.05702666361582e-05, "loss": 0.4489466190338135, "memory(GiB)": 70.5, "step": 33915, "token_acc": 0.8972222222222223, "train_speed(iter/s)": 1.45646 }, { "epoch": 1.4532367936249517, "grad_norm": 3.364086627960205, "learning_rate": 8.056494097152013e-05, "loss": 0.16212034225463867, "memory(GiB)": 70.5, "step": 33920, "token_acc": 0.9584487534626038, "train_speed(iter/s)": 1.456488 }, { "epoch": 1.4534510089542008, "grad_norm": 2.1551027297973633, "learning_rate": 8.055961475316947e-05, "loss": 0.32392706871032717, "memory(GiB)": 70.5, "step": 33925, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.456504 }, { "epoch": 1.4536652242834498, "grad_norm": 3.1070685386657715, "learning_rate": 8.055428798120272e-05, "loss": 0.41553196907043455, "memory(GiB)": 70.5, "step": 33930, "token_acc": 0.9057377049180327, "train_speed(iter/s)": 1.456509 }, { "epoch": 1.4538794396126986, "grad_norm": 1.7730708122253418, "learning_rate": 8.054896065571638e-05, "loss": 0.41185603141784666, "memory(GiB)": 70.5, "step": 33935, "token_acc": 0.9149659863945578, "train_speed(iter/s)": 1.456502 }, { "epoch": 1.4540936549419476, "grad_norm": 3.3240432739257812, "learning_rate": 8.054363277680695e-05, "loss": 0.5764716625213623, "memory(GiB)": 70.5, "step": 33940, "token_acc": 0.8745387453874539, "train_speed(iter/s)": 1.456496 }, { "epoch": 1.4543078702711967, "grad_norm": 3.289592742919922, "learning_rate": 8.053830434457097e-05, "loss": 0.33661391735076907, "memory(GiB)": 70.5, "step": 33945, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.456499 }, { "epoch": 1.4545220856004455, "grad_norm": 1.9082207679748535, "learning_rate": 8.053297535910496e-05, "loss": 0.4861290454864502, "memory(GiB)": 70.5, "step": 33950, "token_acc": 0.8908554572271387, "train_speed(iter/s)": 1.456526 }, { "epoch": 1.4547363009296945, "grad_norm": 6.217767238616943, "learning_rate": 8.052764582050544e-05, "loss": 0.355702543258667, "memory(GiB)": 70.5, "step": 33955, "token_acc": 0.9300911854103343, "train_speed(iter/s)": 1.456546 }, { "epoch": 1.4549505162589436, "grad_norm": 0.4065174460411072, "learning_rate": 8.052231572886902e-05, "loss": 0.7254111289978027, "memory(GiB)": 70.5, "step": 33960, "token_acc": 0.8414239482200647, "train_speed(iter/s)": 1.45657 }, { "epoch": 1.4551647315881924, "grad_norm": 4.500709056854248, "learning_rate": 8.051698508429219e-05, "loss": 0.6248703479766846, "memory(GiB)": 70.5, "step": 33965, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.456572 }, { "epoch": 1.4553789469174414, "grad_norm": 4.220986843109131, "learning_rate": 8.051165388687154e-05, "loss": 0.32059297561645506, "memory(GiB)": 70.5, "step": 33970, "token_acc": 0.9367088607594937, "train_speed(iter/s)": 1.456581 }, { "epoch": 1.4555931622466904, "grad_norm": 6.0977630615234375, "learning_rate": 8.050632213670368e-05, "loss": 0.3015800714492798, "memory(GiB)": 70.5, "step": 33975, "token_acc": 0.9291666666666667, "train_speed(iter/s)": 1.456578 }, { "epoch": 1.4558073775759393, "grad_norm": 3.6212270259857178, "learning_rate": 8.050098983388516e-05, "loss": 0.55916748046875, "memory(GiB)": 70.5, "step": 33980, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.45659 }, { "epoch": 1.4560215929051883, "grad_norm": 2.174985885620117, "learning_rate": 8.04956569785126e-05, "loss": 0.4853366851806641, "memory(GiB)": 70.5, "step": 33985, "token_acc": 0.8875, "train_speed(iter/s)": 1.45658 }, { "epoch": 1.4562358082344373, "grad_norm": 2.618180274963379, "learning_rate": 8.049032357068263e-05, "loss": 0.2355872869491577, "memory(GiB)": 70.5, "step": 33990, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.456584 }, { "epoch": 1.4564500235636861, "grad_norm": 2.1335854530334473, "learning_rate": 8.048498961049182e-05, "loss": 0.3524254560470581, "memory(GiB)": 70.5, "step": 33995, "token_acc": 0.9323076923076923, "train_speed(iter/s)": 1.456597 }, { "epoch": 1.4566642388929352, "grad_norm": 1.0111279487609863, "learning_rate": 8.047965509803684e-05, "loss": 0.23416492938995362, "memory(GiB)": 70.5, "step": 34000, "token_acc": 0.9348659003831418, "train_speed(iter/s)": 1.456593 }, { "epoch": 1.4566642388929352, "eval_loss": 2.601572275161743, "eval_runtime": 13.9194, "eval_samples_per_second": 7.184, "eval_steps_per_second": 7.184, "eval_token_acc": 0.42913385826771655, "step": 34000 }, { "epoch": 1.4568784542221842, "grad_norm": 0.41749656200408936, "learning_rate": 8.04743200334143e-05, "loss": 0.17872778177261353, "memory(GiB)": 70.5, "step": 34005, "token_acc": 0.580765639589169, "train_speed(iter/s)": 1.455671 }, { "epoch": 1.457092669551433, "grad_norm": 3.2983219623565674, "learning_rate": 8.046898441672087e-05, "loss": 0.36252493858337403, "memory(GiB)": 70.5, "step": 34010, "token_acc": 0.914364640883978, "train_speed(iter/s)": 1.455672 }, { "epoch": 1.457306884880682, "grad_norm": 6.534317493438721, "learning_rate": 8.046364824805321e-05, "loss": 0.33112258911132814, "memory(GiB)": 70.5, "step": 34015, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.455673 }, { "epoch": 1.457521100209931, "grad_norm": 0.44917964935302734, "learning_rate": 8.045831152750798e-05, "loss": 0.46323180198669434, "memory(GiB)": 70.5, "step": 34020, "token_acc": 0.8861209964412812, "train_speed(iter/s)": 1.455692 }, { "epoch": 1.4577353155391801, "grad_norm": 4.399231433868408, "learning_rate": 8.045297425518188e-05, "loss": 0.5865123748779297, "memory(GiB)": 70.5, "step": 34025, "token_acc": 0.8814229249011858, "train_speed(iter/s)": 1.455694 }, { "epoch": 1.457949530868429, "grad_norm": 3.736008882522583, "learning_rate": 8.044763643117157e-05, "loss": 0.5234577655792236, "memory(GiB)": 70.5, "step": 34030, "token_acc": 0.8774373259052924, "train_speed(iter/s)": 1.455691 }, { "epoch": 1.458163746197678, "grad_norm": 2.9257113933563232, "learning_rate": 8.044229805557377e-05, "loss": 0.29869282245635986, "memory(GiB)": 70.5, "step": 34035, "token_acc": 0.9533333333333334, "train_speed(iter/s)": 1.4557 }, { "epoch": 1.458377961526927, "grad_norm": 3.0756301879882812, "learning_rate": 8.043695912848517e-05, "loss": 0.1576961874961853, "memory(GiB)": 70.5, "step": 34040, "token_acc": 0.9685314685314685, "train_speed(iter/s)": 1.455703 }, { "epoch": 1.4585921768561758, "grad_norm": 1.9987317323684692, "learning_rate": 8.043161965000252e-05, "loss": 0.29141194820404054, "memory(GiB)": 70.5, "step": 34045, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.455728 }, { "epoch": 1.4588063921854248, "grad_norm": 7.0272417068481445, "learning_rate": 8.042627962022252e-05, "loss": 0.23022122383117677, "memory(GiB)": 70.5, "step": 34050, "token_acc": 0.9568627450980393, "train_speed(iter/s)": 1.455728 }, { "epoch": 1.4590206075146739, "grad_norm": 2.486330509185791, "learning_rate": 8.042093903924193e-05, "loss": 0.4942915916442871, "memory(GiB)": 70.5, "step": 34055, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.455739 }, { "epoch": 1.4592348228439227, "grad_norm": 1.610487937927246, "learning_rate": 8.04155979071575e-05, "loss": 0.1617902398109436, "memory(GiB)": 70.5, "step": 34060, "token_acc": 0.9522184300341296, "train_speed(iter/s)": 1.455739 }, { "epoch": 1.4594490381731717, "grad_norm": 3.2824339866638184, "learning_rate": 8.041025622406596e-05, "loss": 0.8372491836547852, "memory(GiB)": 70.5, "step": 34065, "token_acc": 0.8539325842696629, "train_speed(iter/s)": 1.455739 }, { "epoch": 1.4596632535024208, "grad_norm": 2.5210161209106445, "learning_rate": 8.040491399006412e-05, "loss": 0.35575523376464846, "memory(GiB)": 70.5, "step": 34070, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.455741 }, { "epoch": 1.4598774688316696, "grad_norm": 4.050143241882324, "learning_rate": 8.039957120524872e-05, "loss": 0.6006979942321777, "memory(GiB)": 70.5, "step": 34075, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.455758 }, { "epoch": 1.4600916841609186, "grad_norm": 2.5118024349212646, "learning_rate": 8.03942278697166e-05, "loss": 0.6056692123413085, "memory(GiB)": 70.5, "step": 34080, "token_acc": 0.8706896551724138, "train_speed(iter/s)": 1.455767 }, { "epoch": 1.4603058994901676, "grad_norm": 2.609204053878784, "learning_rate": 8.03888839835645e-05, "loss": 0.19291507005691527, "memory(GiB)": 70.5, "step": 34085, "token_acc": 0.9770114942528736, "train_speed(iter/s)": 1.455761 }, { "epoch": 1.4605201148194165, "grad_norm": 2.0175817012786865, "learning_rate": 8.038353954688928e-05, "loss": 0.33805534839630125, "memory(GiB)": 70.5, "step": 34090, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.455758 }, { "epoch": 1.4607343301486655, "grad_norm": 3.790681838989258, "learning_rate": 8.037819455978774e-05, "loss": 0.2518023490905762, "memory(GiB)": 70.5, "step": 34095, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.455754 }, { "epoch": 1.4609485454779145, "grad_norm": 3.736747980117798, "learning_rate": 8.037284902235669e-05, "loss": 0.5506904125213623, "memory(GiB)": 70.5, "step": 34100, "token_acc": 0.8930817610062893, "train_speed(iter/s)": 1.455752 }, { "epoch": 1.4611627608071633, "grad_norm": 1.6684175729751587, "learning_rate": 8.036750293469302e-05, "loss": 0.6316912174224854, "memory(GiB)": 70.5, "step": 34105, "token_acc": 0.8557046979865772, "train_speed(iter/s)": 1.455749 }, { "epoch": 1.4613769761364124, "grad_norm": 5.3617353439331055, "learning_rate": 8.036215629689352e-05, "loss": 0.4730494976043701, "memory(GiB)": 70.5, "step": 34110, "token_acc": 0.8726114649681529, "train_speed(iter/s)": 1.455746 }, { "epoch": 1.4615911914656614, "grad_norm": 5.4479827880859375, "learning_rate": 8.035680910905508e-05, "loss": 0.34221060276031495, "memory(GiB)": 70.5, "step": 34115, "token_acc": 0.9361022364217252, "train_speed(iter/s)": 1.455743 }, { "epoch": 1.4618054067949102, "grad_norm": 2.494231939315796, "learning_rate": 8.035146137127458e-05, "loss": 0.5201244354248047, "memory(GiB)": 70.5, "step": 34120, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.45574 }, { "epoch": 1.4620196221241593, "grad_norm": 2.545955181121826, "learning_rate": 8.034611308364888e-05, "loss": 0.3916786670684814, "memory(GiB)": 70.5, "step": 34125, "token_acc": 0.9039735099337748, "train_speed(iter/s)": 1.455737 }, { "epoch": 1.4622338374534083, "grad_norm": 0.07731367647647858, "learning_rate": 8.034076424627486e-05, "loss": 0.4990840435028076, "memory(GiB)": 70.5, "step": 34130, "token_acc": 0.9184952978056427, "train_speed(iter/s)": 1.455733 }, { "epoch": 1.462448052782657, "grad_norm": 3.391986131668091, "learning_rate": 8.033541485924945e-05, "loss": 0.44433937072753904, "memory(GiB)": 70.5, "step": 34135, "token_acc": 0.9023569023569024, "train_speed(iter/s)": 1.455738 }, { "epoch": 1.4626622681119061, "grad_norm": 4.060154438018799, "learning_rate": 8.033006492266952e-05, "loss": 1.1214964866638184, "memory(GiB)": 70.5, "step": 34140, "token_acc": 0.7723880597014925, "train_speed(iter/s)": 1.455722 }, { "epoch": 1.4628764834411552, "grad_norm": 0.905083417892456, "learning_rate": 8.032471443663203e-05, "loss": 0.22432501316070558, "memory(GiB)": 70.5, "step": 34145, "token_acc": 0.9541984732824428, "train_speed(iter/s)": 1.455735 }, { "epoch": 1.463090698770404, "grad_norm": 2.68670916557312, "learning_rate": 8.03193634012339e-05, "loss": 0.3040745258331299, "memory(GiB)": 70.5, "step": 34150, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.455736 }, { "epoch": 1.463304914099653, "grad_norm": 2.8615806102752686, "learning_rate": 8.031401181657206e-05, "loss": 0.45002002716064454, "memory(GiB)": 70.5, "step": 34155, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.455756 }, { "epoch": 1.463519129428902, "grad_norm": 2.4223999977111816, "learning_rate": 8.030865968274344e-05, "loss": 0.26651768684387206, "memory(GiB)": 70.5, "step": 34160, "token_acc": 0.9377431906614786, "train_speed(iter/s)": 1.455738 }, { "epoch": 1.4637333447581509, "grad_norm": 2.721541404724121, "learning_rate": 8.030330699984503e-05, "loss": 0.39132323265075686, "memory(GiB)": 70.5, "step": 34165, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.455738 }, { "epoch": 1.4639475600874, "grad_norm": 1.632399320602417, "learning_rate": 8.029795376797377e-05, "loss": 0.4704144954681396, "memory(GiB)": 70.5, "step": 34170, "token_acc": 0.8951612903225806, "train_speed(iter/s)": 1.455727 }, { "epoch": 1.464161775416649, "grad_norm": 6.5857038497924805, "learning_rate": 8.029259998722667e-05, "loss": 0.39681520462036135, "memory(GiB)": 70.5, "step": 34175, "token_acc": 0.9191176470588235, "train_speed(iter/s)": 1.455731 }, { "epoch": 1.4643759907458977, "grad_norm": 3.0407252311706543, "learning_rate": 8.028724565770071e-05, "loss": 0.1535636305809021, "memory(GiB)": 70.5, "step": 34180, "token_acc": 0.9618055555555556, "train_speed(iter/s)": 1.455754 }, { "epoch": 1.4645902060751468, "grad_norm": 2.4290568828582764, "learning_rate": 8.028189077949288e-05, "loss": 0.48316278457641604, "memory(GiB)": 70.5, "step": 34185, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.45575 }, { "epoch": 1.4648044214043958, "grad_norm": 3.5231173038482666, "learning_rate": 8.027653535270019e-05, "loss": 0.29424688816070554, "memory(GiB)": 70.5, "step": 34190, "token_acc": 0.9556451612903226, "train_speed(iter/s)": 1.455756 }, { "epoch": 1.4650186367336446, "grad_norm": 4.568315029144287, "learning_rate": 8.027117937741966e-05, "loss": 0.24543819427490235, "memory(GiB)": 70.5, "step": 34195, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.455766 }, { "epoch": 1.4652328520628937, "grad_norm": 0.6251782178878784, "learning_rate": 8.026582285374832e-05, "loss": 0.13764736652374268, "memory(GiB)": 70.5, "step": 34200, "token_acc": 0.952191235059761, "train_speed(iter/s)": 1.45577 }, { "epoch": 1.4654470673921427, "grad_norm": 14.693560600280762, "learning_rate": 8.02604657817832e-05, "loss": 0.5181086540222168, "memory(GiB)": 70.5, "step": 34205, "token_acc": 0.884375, "train_speed(iter/s)": 1.455785 }, { "epoch": 1.4656612827213915, "grad_norm": 4.400154113769531, "learning_rate": 8.025510816162137e-05, "loss": 0.35047101974487305, "memory(GiB)": 70.5, "step": 34210, "token_acc": 0.9094650205761317, "train_speed(iter/s)": 1.455796 }, { "epoch": 1.4658754980506405, "grad_norm": 7.262684345245361, "learning_rate": 8.024974999335985e-05, "loss": 0.32107431888580323, "memory(GiB)": 70.5, "step": 34215, "token_acc": 0.9305135951661632, "train_speed(iter/s)": 1.455811 }, { "epoch": 1.4660897133798896, "grad_norm": 10.849763870239258, "learning_rate": 8.024439127709575e-05, "loss": 0.4774023056030273, "memory(GiB)": 70.5, "step": 34220, "token_acc": 0.88, "train_speed(iter/s)": 1.455826 }, { "epoch": 1.4663039287091384, "grad_norm": 4.269772052764893, "learning_rate": 8.023903201292613e-05, "loss": 0.437786340713501, "memory(GiB)": 70.5, "step": 34225, "token_acc": 0.9176029962546817, "train_speed(iter/s)": 1.455847 }, { "epoch": 1.4665181440383874, "grad_norm": 7.961014270782471, "learning_rate": 8.023367220094809e-05, "loss": 0.4049995422363281, "memory(GiB)": 70.5, "step": 34230, "token_acc": 0.8996138996138996, "train_speed(iter/s)": 1.45587 }, { "epoch": 1.4667323593676365, "grad_norm": 3.0303330421447754, "learning_rate": 8.02283118412587e-05, "loss": 0.23601167201995848, "memory(GiB)": 70.5, "step": 34235, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.455875 }, { "epoch": 1.4669465746968853, "grad_norm": 3.911341667175293, "learning_rate": 8.022295093395509e-05, "loss": 0.27957513332366946, "memory(GiB)": 70.5, "step": 34240, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.455871 }, { "epoch": 1.4671607900261343, "grad_norm": 0.7173221111297607, "learning_rate": 8.021758947913436e-05, "loss": 0.24948368072509766, "memory(GiB)": 70.5, "step": 34245, "token_acc": 0.9420849420849421, "train_speed(iter/s)": 1.45587 }, { "epoch": 1.4673750053553833, "grad_norm": 4.11175012588501, "learning_rate": 8.021222747689367e-05, "loss": 0.3677948474884033, "memory(GiB)": 70.5, "step": 34250, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.455883 }, { "epoch": 1.4675892206846322, "grad_norm": 4.071471691131592, "learning_rate": 8.020686492733011e-05, "loss": 0.5275938987731934, "memory(GiB)": 70.5, "step": 34255, "token_acc": 0.8817204301075269, "train_speed(iter/s)": 1.455899 }, { "epoch": 1.4678034360138812, "grad_norm": 4.465188980102539, "learning_rate": 8.02015018305409e-05, "loss": 0.41013541221618655, "memory(GiB)": 70.5, "step": 34260, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.455913 }, { "epoch": 1.4680176513431302, "grad_norm": 2.4505345821380615, "learning_rate": 8.019613818662312e-05, "loss": 0.25667057037353513, "memory(GiB)": 70.5, "step": 34265, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.455926 }, { "epoch": 1.468231866672379, "grad_norm": 2.071429491043091, "learning_rate": 8.019077399567398e-05, "loss": 0.5072190761566162, "memory(GiB)": 70.5, "step": 34270, "token_acc": 0.9111747851002865, "train_speed(iter/s)": 1.455941 }, { "epoch": 1.468446082001628, "grad_norm": 1.7226234674453735, "learning_rate": 8.018540925779064e-05, "loss": 0.19298069477081298, "memory(GiB)": 70.5, "step": 34275, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.455939 }, { "epoch": 1.468660297330877, "grad_norm": 4.2554497718811035, "learning_rate": 8.018004397307031e-05, "loss": 0.5054027557373046, "memory(GiB)": 70.5, "step": 34280, "token_acc": 0.8865546218487395, "train_speed(iter/s)": 1.455947 }, { "epoch": 1.468874512660126, "grad_norm": 3.044149160385132, "learning_rate": 8.017467814161015e-05, "loss": 0.5528375625610351, "memory(GiB)": 70.5, "step": 34285, "token_acc": 0.8817204301075269, "train_speed(iter/s)": 1.455965 }, { "epoch": 1.469088727989375, "grad_norm": 2.327265739440918, "learning_rate": 8.016931176350741e-05, "loss": 0.4748426914215088, "memory(GiB)": 70.5, "step": 34290, "token_acc": 0.8961538461538462, "train_speed(iter/s)": 1.455976 }, { "epoch": 1.469302943318624, "grad_norm": 5.198019981384277, "learning_rate": 8.016394483885928e-05, "loss": 0.5041196823120118, "memory(GiB)": 70.5, "step": 34295, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.455992 }, { "epoch": 1.4695171586478728, "grad_norm": 4.337114334106445, "learning_rate": 8.015857736776299e-05, "loss": 0.34467158317565916, "memory(GiB)": 70.5, "step": 34300, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.455983 }, { "epoch": 1.4697313739771218, "grad_norm": 2.8301098346710205, "learning_rate": 8.015320935031579e-05, "loss": 0.26375856399536135, "memory(GiB)": 70.5, "step": 34305, "token_acc": 0.9486166007905138, "train_speed(iter/s)": 1.455987 }, { "epoch": 1.4699455893063709, "grad_norm": 3.719467878341675, "learning_rate": 8.014784078661491e-05, "loss": 0.43950886726379396, "memory(GiB)": 70.5, "step": 34310, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.456015 }, { "epoch": 1.4701598046356197, "grad_norm": 2.125335693359375, "learning_rate": 8.01424716767576e-05, "loss": 0.43907828330993653, "memory(GiB)": 70.5, "step": 34315, "token_acc": 0.9203821656050956, "train_speed(iter/s)": 1.456017 }, { "epoch": 1.4703740199648687, "grad_norm": 5.201113700866699, "learning_rate": 8.013710202084115e-05, "loss": 0.17896703481674195, "memory(GiB)": 70.5, "step": 34320, "token_acc": 0.963855421686747, "train_speed(iter/s)": 1.456027 }, { "epoch": 1.4705882352941178, "grad_norm": 2.7227799892425537, "learning_rate": 8.013173181896283e-05, "loss": 0.31815528869628906, "memory(GiB)": 70.5, "step": 34325, "token_acc": 0.921875, "train_speed(iter/s)": 1.456047 }, { "epoch": 1.4708024506233666, "grad_norm": 4.75642204284668, "learning_rate": 8.012636107121992e-05, "loss": 0.715931510925293, "memory(GiB)": 70.5, "step": 34330, "token_acc": 0.8333333333333334, "train_speed(iter/s)": 1.456037 }, { "epoch": 1.4710166659526156, "grad_norm": 3.902169704437256, "learning_rate": 8.012098977770971e-05, "loss": 0.4021204948425293, "memory(GiB)": 70.5, "step": 34335, "token_acc": 0.919732441471572, "train_speed(iter/s)": 1.456042 }, { "epoch": 1.4712308812818646, "grad_norm": 4.6738691329956055, "learning_rate": 8.011561793852953e-05, "loss": 0.3668795585632324, "memory(GiB)": 70.5, "step": 34340, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.456044 }, { "epoch": 1.4714450966111134, "grad_norm": 2.244853973388672, "learning_rate": 8.011024555377667e-05, "loss": 0.2986437797546387, "memory(GiB)": 70.5, "step": 34345, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.456048 }, { "epoch": 1.4716593119403625, "grad_norm": 5.788876533508301, "learning_rate": 8.010487262354847e-05, "loss": 0.46922645568847654, "memory(GiB)": 70.5, "step": 34350, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.456066 }, { "epoch": 1.4718735272696115, "grad_norm": 4.036468982696533, "learning_rate": 8.009949914794226e-05, "loss": 0.3079704284667969, "memory(GiB)": 70.5, "step": 34355, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.456079 }, { "epoch": 1.4720877425988603, "grad_norm": 1.7323062419891357, "learning_rate": 8.009412512705539e-05, "loss": 0.3098806858062744, "memory(GiB)": 70.5, "step": 34360, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.456086 }, { "epoch": 1.4723019579281094, "grad_norm": 3.3007707595825195, "learning_rate": 8.00887505609852e-05, "loss": 0.3483920097351074, "memory(GiB)": 70.5, "step": 34365, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.45609 }, { "epoch": 1.4725161732573584, "grad_norm": 2.9709227085113525, "learning_rate": 8.008337544982909e-05, "loss": 0.4676970481872559, "memory(GiB)": 70.5, "step": 34370, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.456091 }, { "epoch": 1.4727303885866072, "grad_norm": 5.496466636657715, "learning_rate": 8.00779997936844e-05, "loss": 0.4057310581207275, "memory(GiB)": 70.5, "step": 34375, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.456083 }, { "epoch": 1.4729446039158562, "grad_norm": 5.1204657554626465, "learning_rate": 8.007262359264852e-05, "loss": 0.47240619659423827, "memory(GiB)": 70.5, "step": 34380, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.456091 }, { "epoch": 1.4731588192451053, "grad_norm": 8.59643840789795, "learning_rate": 8.006724684681888e-05, "loss": 0.25269179344177245, "memory(GiB)": 70.5, "step": 34385, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.456084 }, { "epoch": 1.473373034574354, "grad_norm": 5.346027851104736, "learning_rate": 8.006186955629282e-05, "loss": 0.2623694658279419, "memory(GiB)": 70.5, "step": 34390, "token_acc": 0.9561128526645768, "train_speed(iter/s)": 1.456074 }, { "epoch": 1.4735872499036031, "grad_norm": 3.726494789123535, "learning_rate": 8.005649172116782e-05, "loss": 0.2986262559890747, "memory(GiB)": 70.5, "step": 34395, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.456077 }, { "epoch": 1.4738014652328522, "grad_norm": 2.894200325012207, "learning_rate": 8.005111334154127e-05, "loss": 0.334486198425293, "memory(GiB)": 70.5, "step": 34400, "token_acc": 0.9052287581699346, "train_speed(iter/s)": 1.456078 }, { "epoch": 1.474015680562101, "grad_norm": 1.852386474609375, "learning_rate": 8.004573441751062e-05, "loss": 0.3135518550872803, "memory(GiB)": 70.5, "step": 34405, "token_acc": 0.9375, "train_speed(iter/s)": 1.456082 }, { "epoch": 1.47422989589135, "grad_norm": 4.654975414276123, "learning_rate": 8.00403549491733e-05, "loss": 0.32889623641967775, "memory(GiB)": 70.5, "step": 34410, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.456085 }, { "epoch": 1.474444111220599, "grad_norm": 2.6114704608917236, "learning_rate": 8.003497493662678e-05, "loss": 0.47225289344787597, "memory(GiB)": 70.5, "step": 34415, "token_acc": 0.8818181818181818, "train_speed(iter/s)": 1.45608 }, { "epoch": 1.4746583265498479, "grad_norm": 4.303999900817871, "learning_rate": 8.002959437996849e-05, "loss": 0.4675889015197754, "memory(GiB)": 70.5, "step": 34420, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.456081 }, { "epoch": 1.4748725418790969, "grad_norm": 2.311577320098877, "learning_rate": 8.002421327929592e-05, "loss": 0.5382297515869141, "memory(GiB)": 70.5, "step": 34425, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.45608 }, { "epoch": 1.475086757208346, "grad_norm": 3.1905484199523926, "learning_rate": 8.001883163470661e-05, "loss": 0.4651947975158691, "memory(GiB)": 70.5, "step": 34430, "token_acc": 0.8869047619047619, "train_speed(iter/s)": 1.456071 }, { "epoch": 1.4753009725375947, "grad_norm": 2.740889549255371, "learning_rate": 8.001344944629796e-05, "loss": 0.5708941459655762, "memory(GiB)": 70.5, "step": 34435, "token_acc": 0.871900826446281, "train_speed(iter/s)": 1.456073 }, { "epoch": 1.4755151878668438, "grad_norm": 1.6054861545562744, "learning_rate": 8.000806671416754e-05, "loss": 0.2751967430114746, "memory(GiB)": 70.5, "step": 34440, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.456076 }, { "epoch": 1.4757294031960928, "grad_norm": 2.8797290325164795, "learning_rate": 8.000268343841283e-05, "loss": 0.7470288276672363, "memory(GiB)": 70.5, "step": 34445, "token_acc": 0.8481012658227848, "train_speed(iter/s)": 1.456071 }, { "epoch": 1.4759436185253416, "grad_norm": 3.1505727767944336, "learning_rate": 7.999729961913139e-05, "loss": 0.3576786518096924, "memory(GiB)": 70.5, "step": 34450, "token_acc": 0.9171974522292994, "train_speed(iter/s)": 1.456083 }, { "epoch": 1.4761578338545907, "grad_norm": 4.0858235359191895, "learning_rate": 7.999191525642069e-05, "loss": 0.42720851898193357, "memory(GiB)": 70.5, "step": 34455, "token_acc": 0.8984375, "train_speed(iter/s)": 1.456108 }, { "epoch": 1.4763720491838397, "grad_norm": 3.0896081924438477, "learning_rate": 7.998653035037834e-05, "loss": 0.3956362962722778, "memory(GiB)": 70.5, "step": 34460, "token_acc": 0.924812030075188, "train_speed(iter/s)": 1.456112 }, { "epoch": 1.4765862645130885, "grad_norm": 1.4863277673721313, "learning_rate": 7.998114490110185e-05, "loss": 0.4323300838470459, "memory(GiB)": 70.5, "step": 34465, "token_acc": 0.9177631578947368, "train_speed(iter/s)": 1.456113 }, { "epoch": 1.4768004798423375, "grad_norm": 1.2755409479141235, "learning_rate": 7.997575890868879e-05, "loss": 0.22938215732574463, "memory(GiB)": 70.5, "step": 34470, "token_acc": 0.9522058823529411, "train_speed(iter/s)": 1.456123 }, { "epoch": 1.4770146951715866, "grad_norm": 4.107542514801025, "learning_rate": 7.997037237323675e-05, "loss": 0.7939009666442871, "memory(GiB)": 70.5, "step": 34475, "token_acc": 0.8401253918495298, "train_speed(iter/s)": 1.456122 }, { "epoch": 1.4772289105008354, "grad_norm": 3.335815906524658, "learning_rate": 7.99649852948433e-05, "loss": 0.2630064249038696, "memory(GiB)": 70.5, "step": 34480, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.456124 }, { "epoch": 1.4774431258300844, "grad_norm": 7.950351715087891, "learning_rate": 7.995959767360604e-05, "loss": 0.4760587215423584, "memory(GiB)": 70.5, "step": 34485, "token_acc": 0.8840125391849529, "train_speed(iter/s)": 1.456158 }, { "epoch": 1.4776573411593334, "grad_norm": 0.551179826259613, "learning_rate": 7.995420950962254e-05, "loss": 0.08005102276802063, "memory(GiB)": 70.5, "step": 34490, "token_acc": 0.9771241830065359, "train_speed(iter/s)": 1.456171 }, { "epoch": 1.4778715564885823, "grad_norm": 1.5179733037948608, "learning_rate": 7.994882080299044e-05, "loss": 0.258227801322937, "memory(GiB)": 70.5, "step": 34495, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.45619 }, { "epoch": 1.4780857718178313, "grad_norm": 1.8089371919631958, "learning_rate": 7.994343155380737e-05, "loss": 0.19598482847213744, "memory(GiB)": 70.5, "step": 34500, "token_acc": 0.9563758389261745, "train_speed(iter/s)": 1.456182 }, { "epoch": 1.4780857718178313, "eval_loss": 2.572617292404175, "eval_runtime": 13.7902, "eval_samples_per_second": 7.252, "eval_steps_per_second": 7.252, "eval_token_acc": 0.4323962516733601, "step": 34500 }, { "epoch": 1.4782999871470803, "grad_norm": 3.152817726135254, "learning_rate": 7.993804176217093e-05, "loss": 0.30916714668273926, "memory(GiB)": 70.5, "step": 34505, "token_acc": 0.5600393700787402, "train_speed(iter/s)": 1.455306 }, { "epoch": 1.4785142024763291, "grad_norm": 2.332034111022949, "learning_rate": 7.993265142817881e-05, "loss": 0.23485217094421387, "memory(GiB)": 70.5, "step": 34510, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.455306 }, { "epoch": 1.4787284178055782, "grad_norm": 2.7114343643188477, "learning_rate": 7.99272605519286e-05, "loss": 0.6535956382751464, "memory(GiB)": 70.5, "step": 34515, "token_acc": 0.8802816901408451, "train_speed(iter/s)": 1.455299 }, { "epoch": 1.4789426331348272, "grad_norm": 1.531501293182373, "learning_rate": 7.9921869133518e-05, "loss": 0.3484785079956055, "memory(GiB)": 70.5, "step": 34520, "token_acc": 0.9271356783919598, "train_speed(iter/s)": 1.455292 }, { "epoch": 1.479156848464076, "grad_norm": 7.195781230926514, "learning_rate": 7.991647717304467e-05, "loss": 0.5732417106628418, "memory(GiB)": 70.5, "step": 34525, "token_acc": 0.8839285714285714, "train_speed(iter/s)": 1.455299 }, { "epoch": 1.479371063793325, "grad_norm": 3.3123085498809814, "learning_rate": 7.99110846706063e-05, "loss": 0.34392752647399905, "memory(GiB)": 70.5, "step": 34530, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.455305 }, { "epoch": 1.479585279122574, "grad_norm": 0.47233447432518005, "learning_rate": 7.990569162630057e-05, "loss": 0.4097888469696045, "memory(GiB)": 70.5, "step": 34535, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.455301 }, { "epoch": 1.479799494451823, "grad_norm": 10.455978393554688, "learning_rate": 7.990029804022518e-05, "loss": 0.4500273704528809, "memory(GiB)": 70.5, "step": 34540, "token_acc": 0.9015384615384615, "train_speed(iter/s)": 1.455317 }, { "epoch": 1.480013709781072, "grad_norm": 2.4203858375549316, "learning_rate": 7.989490391247784e-05, "loss": 0.502648401260376, "memory(GiB)": 70.5, "step": 34545, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.45533 }, { "epoch": 1.480227925110321, "grad_norm": 5.394595623016357, "learning_rate": 7.988950924315628e-05, "loss": 0.5659533977508545, "memory(GiB)": 70.5, "step": 34550, "token_acc": 0.9012738853503185, "train_speed(iter/s)": 1.45534 }, { "epoch": 1.4804421404395698, "grad_norm": 5.444210529327393, "learning_rate": 7.988411403235823e-05, "loss": 0.2576026439666748, "memory(GiB)": 70.5, "step": 34555, "token_acc": 0.939209726443769, "train_speed(iter/s)": 1.455343 }, { "epoch": 1.4806563557688188, "grad_norm": 4.045653343200684, "learning_rate": 7.987871828018141e-05, "loss": 0.2961733818054199, "memory(GiB)": 70.5, "step": 34560, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.455364 }, { "epoch": 1.4808705710980679, "grad_norm": 4.370522499084473, "learning_rate": 7.987332198672356e-05, "loss": 0.579848051071167, "memory(GiB)": 70.5, "step": 34565, "token_acc": 0.9013605442176871, "train_speed(iter/s)": 1.455367 }, { "epoch": 1.4810847864273167, "grad_norm": 2.1375274658203125, "learning_rate": 7.986792515208248e-05, "loss": 0.5427759170532227, "memory(GiB)": 70.5, "step": 34570, "token_acc": 0.8912280701754386, "train_speed(iter/s)": 1.455372 }, { "epoch": 1.4812990017565657, "grad_norm": 3.0748205184936523, "learning_rate": 7.986252777635592e-05, "loss": 0.2778775691986084, "memory(GiB)": 70.5, "step": 34575, "token_acc": 0.9395770392749244, "train_speed(iter/s)": 1.45537 }, { "epoch": 1.4815132170858147, "grad_norm": 2.057950019836426, "learning_rate": 7.985712985964164e-05, "loss": 0.21636452674865722, "memory(GiB)": 70.5, "step": 34580, "token_acc": 0.9525862068965517, "train_speed(iter/s)": 1.45537 }, { "epoch": 1.4817274324150636, "grad_norm": 2.962069034576416, "learning_rate": 7.985173140203745e-05, "loss": 0.32089412212371826, "memory(GiB)": 70.5, "step": 34585, "token_acc": 0.9457627118644067, "train_speed(iter/s)": 1.455358 }, { "epoch": 1.4819416477443126, "grad_norm": 2.8226966857910156, "learning_rate": 7.984633240364116e-05, "loss": 0.20073516368865968, "memory(GiB)": 70.5, "step": 34590, "token_acc": 0.9504950495049505, "train_speed(iter/s)": 1.455367 }, { "epoch": 1.4821558630735616, "grad_norm": 4.0740742683410645, "learning_rate": 7.984093286455055e-05, "loss": 0.39075565338134766, "memory(GiB)": 70.5, "step": 34595, "token_acc": 0.9169811320754717, "train_speed(iter/s)": 1.45537 }, { "epoch": 1.4823700784028104, "grad_norm": 7.664714336395264, "learning_rate": 7.983553278486344e-05, "loss": 0.26865553855895996, "memory(GiB)": 70.5, "step": 34600, "token_acc": 0.9483282674772037, "train_speed(iter/s)": 1.455372 }, { "epoch": 1.4825842937320595, "grad_norm": 2.8765006065368652, "learning_rate": 7.983013216467768e-05, "loss": 0.46701898574829104, "memory(GiB)": 70.5, "step": 34605, "token_acc": 0.9052287581699346, "train_speed(iter/s)": 1.455367 }, { "epoch": 1.4827985090613085, "grad_norm": 3.7161755561828613, "learning_rate": 7.982473100409107e-05, "loss": 0.43094382286071775, "memory(GiB)": 70.5, "step": 34610, "token_acc": 0.9191176470588235, "train_speed(iter/s)": 1.455365 }, { "epoch": 1.4830127243905573, "grad_norm": 5.613711357116699, "learning_rate": 7.981932930320149e-05, "loss": 0.7136321067810059, "memory(GiB)": 70.5, "step": 34615, "token_acc": 0.8797653958944281, "train_speed(iter/s)": 1.455356 }, { "epoch": 1.4832269397198063, "grad_norm": 9.230955123901367, "learning_rate": 7.98139270621068e-05, "loss": 0.3828678369522095, "memory(GiB)": 70.5, "step": 34620, "token_acc": 0.9176954732510288, "train_speed(iter/s)": 1.455364 }, { "epoch": 1.4834411550490554, "grad_norm": 5.9153900146484375, "learning_rate": 7.980852428090484e-05, "loss": 0.6031455039978028, "memory(GiB)": 70.5, "step": 34625, "token_acc": 0.887240356083086, "train_speed(iter/s)": 1.455364 }, { "epoch": 1.4836553703783042, "grad_norm": 2.6147186756134033, "learning_rate": 7.980312095969351e-05, "loss": 0.3120787858963013, "memory(GiB)": 70.5, "step": 34630, "token_acc": 0.9376770538243626, "train_speed(iter/s)": 1.455374 }, { "epoch": 1.4838695857075532, "grad_norm": 4.275772571563721, "learning_rate": 7.979771709857066e-05, "loss": 0.515981388092041, "memory(GiB)": 70.5, "step": 34635, "token_acc": 0.8875379939209727, "train_speed(iter/s)": 1.455381 }, { "epoch": 1.4840838010368023, "grad_norm": 3.9116296768188477, "learning_rate": 7.979231269763425e-05, "loss": 0.4480780601501465, "memory(GiB)": 70.5, "step": 34640, "token_acc": 0.9087301587301587, "train_speed(iter/s)": 1.455378 }, { "epoch": 1.484298016366051, "grad_norm": 3.784015655517578, "learning_rate": 7.978690775698213e-05, "loss": 0.31779568195343016, "memory(GiB)": 70.5, "step": 34645, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.455377 }, { "epoch": 1.4845122316953, "grad_norm": 2.6359078884124756, "learning_rate": 7.978150227671223e-05, "loss": 0.2216705560684204, "memory(GiB)": 70.5, "step": 34650, "token_acc": 0.9425675675675675, "train_speed(iter/s)": 1.455368 }, { "epoch": 1.4847264470245491, "grad_norm": 1.6860268115997314, "learning_rate": 7.977609625692248e-05, "loss": 0.2877696990966797, "memory(GiB)": 70.5, "step": 34655, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.455372 }, { "epoch": 1.484940662353798, "grad_norm": 2.730039358139038, "learning_rate": 7.977068969771083e-05, "loss": 0.17357180118560792, "memory(GiB)": 70.5, "step": 34660, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.455361 }, { "epoch": 1.485154877683047, "grad_norm": 3.5059573650360107, "learning_rate": 7.97652825991752e-05, "loss": 0.6232457160949707, "memory(GiB)": 70.5, "step": 34665, "token_acc": 0.85, "train_speed(iter/s)": 1.455353 }, { "epoch": 1.485369093012296, "grad_norm": 9.588077545166016, "learning_rate": 7.975987496141354e-05, "loss": 0.42986283302307127, "memory(GiB)": 70.5, "step": 34670, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.455354 }, { "epoch": 1.4855833083415448, "grad_norm": 3.9572174549102783, "learning_rate": 7.975446678452384e-05, "loss": 0.5414800643920898, "memory(GiB)": 70.5, "step": 34675, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.455359 }, { "epoch": 1.4857975236707939, "grad_norm": 1.549605131149292, "learning_rate": 7.974905806860407e-05, "loss": 0.4609567165374756, "memory(GiB)": 70.5, "step": 34680, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.455363 }, { "epoch": 1.486011739000043, "grad_norm": 5.116684436798096, "learning_rate": 7.974364881375218e-05, "loss": 0.27497570514678954, "memory(GiB)": 70.5, "step": 34685, "token_acc": 0.9392712550607287, "train_speed(iter/s)": 1.455361 }, { "epoch": 1.4862259543292917, "grad_norm": 1.9800137281417847, "learning_rate": 7.97382390200662e-05, "loss": 0.3450393438339233, "memory(GiB)": 70.5, "step": 34690, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.455354 }, { "epoch": 1.4864401696585408, "grad_norm": 1.7100807428359985, "learning_rate": 7.973282868764413e-05, "loss": 0.1449721336364746, "memory(GiB)": 70.5, "step": 34695, "token_acc": 0.9651898734177216, "train_speed(iter/s)": 1.455364 }, { "epoch": 1.4866543849877898, "grad_norm": 2.8473026752471924, "learning_rate": 7.972741781658399e-05, "loss": 0.3445319175720215, "memory(GiB)": 70.5, "step": 34700, "token_acc": 0.9254385964912281, "train_speed(iter/s)": 1.455393 }, { "epoch": 1.4868686003170386, "grad_norm": 1.7406667470932007, "learning_rate": 7.972200640698377e-05, "loss": 0.41650710105895994, "memory(GiB)": 70.5, "step": 34705, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.455404 }, { "epoch": 1.4870828156462876, "grad_norm": 1.8872872591018677, "learning_rate": 7.971659445894152e-05, "loss": 0.4181836605072021, "memory(GiB)": 70.5, "step": 34710, "token_acc": 0.9067524115755627, "train_speed(iter/s)": 1.455433 }, { "epoch": 1.4872970309755367, "grad_norm": 1.21588933467865, "learning_rate": 7.97111819725553e-05, "loss": 0.48754420280456545, "memory(GiB)": 70.5, "step": 34715, "token_acc": 0.8909657320872274, "train_speed(iter/s)": 1.455439 }, { "epoch": 1.4875112463047855, "grad_norm": 1.6119019985198975, "learning_rate": 7.970576894792314e-05, "loss": 0.5611565589904786, "memory(GiB)": 70.5, "step": 34720, "token_acc": 0.9044368600682594, "train_speed(iter/s)": 1.455492 }, { "epoch": 1.4877254616340345, "grad_norm": 3.5177226066589355, "learning_rate": 7.97003553851431e-05, "loss": 0.29432153701782227, "memory(GiB)": 70.5, "step": 34725, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.455496 }, { "epoch": 1.4879396769632836, "grad_norm": 6.4547648429870605, "learning_rate": 7.969494128431327e-05, "loss": 0.5225250244140625, "memory(GiB)": 70.5, "step": 34730, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.455493 }, { "epoch": 1.4881538922925324, "grad_norm": 5.4001240730285645, "learning_rate": 7.968952664553172e-05, "loss": 0.416029691696167, "memory(GiB)": 70.5, "step": 34735, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.455493 }, { "epoch": 1.4883681076217814, "grad_norm": 0.7814594507217407, "learning_rate": 7.968411146889656e-05, "loss": 0.18059191703796387, "memory(GiB)": 70.5, "step": 34740, "token_acc": 0.9506172839506173, "train_speed(iter/s)": 1.455499 }, { "epoch": 1.4885823229510304, "grad_norm": 2.5539026260375977, "learning_rate": 7.967869575450587e-05, "loss": 0.3254230260848999, "memory(GiB)": 70.5, "step": 34745, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.4555 }, { "epoch": 1.4887965382802792, "grad_norm": 2.5035183429718018, "learning_rate": 7.967327950245775e-05, "loss": 0.3686449766159058, "memory(GiB)": 70.5, "step": 34750, "token_acc": 0.9065743944636678, "train_speed(iter/s)": 1.455523 }, { "epoch": 1.4890107536095283, "grad_norm": 4.379935264587402, "learning_rate": 7.966786271285034e-05, "loss": 0.3011446714401245, "memory(GiB)": 70.5, "step": 34755, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.455524 }, { "epoch": 1.4892249689387773, "grad_norm": 5.6381707191467285, "learning_rate": 7.966244538578177e-05, "loss": 0.37162957191467283, "memory(GiB)": 70.5, "step": 34760, "token_acc": 0.9311926605504587, "train_speed(iter/s)": 1.455542 }, { "epoch": 1.4894391842680261, "grad_norm": 3.1649868488311768, "learning_rate": 7.965702752135018e-05, "loss": 0.4054431438446045, "memory(GiB)": 70.5, "step": 34765, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.45555 }, { "epoch": 1.4896533995972752, "grad_norm": 3.2943999767303467, "learning_rate": 7.965160911965371e-05, "loss": 0.2526509761810303, "memory(GiB)": 70.5, "step": 34770, "token_acc": 0.9453924914675768, "train_speed(iter/s)": 1.455578 }, { "epoch": 1.4898676149265242, "grad_norm": 2.593472957611084, "learning_rate": 7.964619018079054e-05, "loss": 0.19202609062194825, "memory(GiB)": 70.5, "step": 34775, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.455597 }, { "epoch": 1.490081830255773, "grad_norm": 2.1867942810058594, "learning_rate": 7.964077070485881e-05, "loss": 0.3064500093460083, "memory(GiB)": 70.5, "step": 34780, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.455591 }, { "epoch": 1.490296045585022, "grad_norm": 3.6007838249206543, "learning_rate": 7.963535069195671e-05, "loss": 0.6767500877380371, "memory(GiB)": 70.5, "step": 34785, "token_acc": 0.8840125391849529, "train_speed(iter/s)": 1.455595 }, { "epoch": 1.490510260914271, "grad_norm": 4.200218677520752, "learning_rate": 7.962993014218243e-05, "loss": 0.3755157709121704, "memory(GiB)": 70.5, "step": 34790, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.45559 }, { "epoch": 1.49072447624352, "grad_norm": 2.2946081161499023, "learning_rate": 7.962450905563418e-05, "loss": 0.229477596282959, "memory(GiB)": 70.5, "step": 34795, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.455596 }, { "epoch": 1.490938691572769, "grad_norm": 4.832059860229492, "learning_rate": 7.961908743241016e-05, "loss": 0.6557494163513183, "memory(GiB)": 70.5, "step": 34800, "token_acc": 0.8641509433962264, "train_speed(iter/s)": 1.455599 }, { "epoch": 1.491152906902018, "grad_norm": 2.223418951034546, "learning_rate": 7.961366527260858e-05, "loss": 0.41901235580444335, "memory(GiB)": 70.5, "step": 34805, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.455605 }, { "epoch": 1.4913671222312668, "grad_norm": 2.4195876121520996, "learning_rate": 7.960824257632768e-05, "loss": 0.3150567770004272, "memory(GiB)": 70.5, "step": 34810, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.455604 }, { "epoch": 1.4915813375605158, "grad_norm": 2.0361597537994385, "learning_rate": 7.960281934366568e-05, "loss": 0.37796289920806886, "memory(GiB)": 70.5, "step": 34815, "token_acc": 0.8929663608562691, "train_speed(iter/s)": 1.455594 }, { "epoch": 1.4917955528897648, "grad_norm": 3.4438858032226562, "learning_rate": 7.959739557472085e-05, "loss": 0.4256147384643555, "memory(GiB)": 70.5, "step": 34820, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.455587 }, { "epoch": 1.4920097682190137, "grad_norm": 3.8510522842407227, "learning_rate": 7.959197126959142e-05, "loss": 0.5506343364715576, "memory(GiB)": 70.5, "step": 34825, "token_acc": 0.8956228956228957, "train_speed(iter/s)": 1.455596 }, { "epoch": 1.4922239835482627, "grad_norm": 2.8670833110809326, "learning_rate": 7.958654642837569e-05, "loss": 0.31556239128112795, "memory(GiB)": 70.5, "step": 34830, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.455619 }, { "epoch": 1.4924381988775117, "grad_norm": 2.105057716369629, "learning_rate": 7.95811210511719e-05, "loss": 0.2528392791748047, "memory(GiB)": 70.5, "step": 34835, "token_acc": 0.9343283582089552, "train_speed(iter/s)": 1.455637 }, { "epoch": 1.4926524142067605, "grad_norm": 8.94981575012207, "learning_rate": 7.957569513807836e-05, "loss": 0.35908732414245603, "memory(GiB)": 70.5, "step": 34840, "token_acc": 0.9025787965616046, "train_speed(iter/s)": 1.455636 }, { "epoch": 1.4928666295360096, "grad_norm": 0.24711477756500244, "learning_rate": 7.957026868919334e-05, "loss": 0.3258531093597412, "memory(GiB)": 70.5, "step": 34845, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.455637 }, { "epoch": 1.4930808448652586, "grad_norm": 2.3716554641723633, "learning_rate": 7.95648417046152e-05, "loss": 0.3224760055541992, "memory(GiB)": 70.5, "step": 34850, "token_acc": 0.9274924471299094, "train_speed(iter/s)": 1.455638 }, { "epoch": 1.4932950601945074, "grad_norm": 3.5311226844787598, "learning_rate": 7.955941418444221e-05, "loss": 0.44638991355895996, "memory(GiB)": 70.5, "step": 34855, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.455635 }, { "epoch": 1.4935092755237565, "grad_norm": 3.0320498943328857, "learning_rate": 7.955398612877269e-05, "loss": 0.4340227127075195, "memory(GiB)": 70.5, "step": 34860, "token_acc": 0.9067524115755627, "train_speed(iter/s)": 1.45564 }, { "epoch": 1.4937234908530055, "grad_norm": 3.836120843887329, "learning_rate": 7.954855753770499e-05, "loss": 0.36961755752563474, "memory(GiB)": 70.5, "step": 34865, "token_acc": 0.9305019305019305, "train_speed(iter/s)": 1.455651 }, { "epoch": 1.4939377061822543, "grad_norm": 1.972031593322754, "learning_rate": 7.954312841133744e-05, "loss": 0.20979886054992675, "memory(GiB)": 70.5, "step": 34870, "token_acc": 0.9593023255813954, "train_speed(iter/s)": 1.455643 }, { "epoch": 1.4941519215115033, "grad_norm": 4.425050258636475, "learning_rate": 7.953769874976842e-05, "loss": 0.4355788230895996, "memory(GiB)": 70.5, "step": 34875, "token_acc": 0.9028213166144201, "train_speed(iter/s)": 1.45565 }, { "epoch": 1.4943661368407524, "grad_norm": 2.626718521118164, "learning_rate": 7.953226855309628e-05, "loss": 0.37184414863586424, "memory(GiB)": 70.5, "step": 34880, "token_acc": 0.9361022364217252, "train_speed(iter/s)": 1.455654 }, { "epoch": 1.4945803521700012, "grad_norm": 5.104114055633545, "learning_rate": 7.952683782141939e-05, "loss": 0.29866900444030764, "memory(GiB)": 70.5, "step": 34885, "token_acc": 0.944078947368421, "train_speed(iter/s)": 1.455652 }, { "epoch": 1.4947945674992502, "grad_norm": 3.8287811279296875, "learning_rate": 7.952140655483613e-05, "loss": 0.5451876640319824, "memory(GiB)": 70.5, "step": 34890, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.455673 }, { "epoch": 1.4950087828284992, "grad_norm": 1.4760717153549194, "learning_rate": 7.95159747534449e-05, "loss": 0.6439196586608886, "memory(GiB)": 70.5, "step": 34895, "token_acc": 0.8972602739726028, "train_speed(iter/s)": 1.4557 }, { "epoch": 1.495222998157748, "grad_norm": 4.603642463684082, "learning_rate": 7.95105424173441e-05, "loss": 0.12953004837036133, "memory(GiB)": 70.5, "step": 34900, "token_acc": 0.9680851063829787, "train_speed(iter/s)": 1.455703 }, { "epoch": 1.495437213486997, "grad_norm": 2.1553494930267334, "learning_rate": 7.950510954663213e-05, "loss": 0.28015623092651365, "memory(GiB)": 70.5, "step": 34905, "token_acc": 0.9539007092198581, "train_speed(iter/s)": 1.455702 }, { "epoch": 1.4956514288162461, "grad_norm": 1.8937830924987793, "learning_rate": 7.949967614140744e-05, "loss": 0.5176852226257325, "memory(GiB)": 70.5, "step": 34910, "token_acc": 0.8745387453874539, "train_speed(iter/s)": 1.455735 }, { "epoch": 1.495865644145495, "grad_norm": 2.214625120162964, "learning_rate": 7.949424220176843e-05, "loss": 0.42978787422180176, "memory(GiB)": 70.5, "step": 34915, "token_acc": 0.9212328767123288, "train_speed(iter/s)": 1.455731 }, { "epoch": 1.496079859474744, "grad_norm": 1.3858954906463623, "learning_rate": 7.948880772781356e-05, "loss": 0.7034565448760987, "memory(GiB)": 70.5, "step": 34920, "token_acc": 0.8819444444444444, "train_speed(iter/s)": 1.455762 }, { "epoch": 1.496294074803993, "grad_norm": 2.4119479656219482, "learning_rate": 7.948337271964128e-05, "loss": 0.3316943168640137, "memory(GiB)": 70.5, "step": 34925, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.455774 }, { "epoch": 1.4965082901332418, "grad_norm": 3.018850326538086, "learning_rate": 7.947793717735003e-05, "loss": 0.2877993106842041, "memory(GiB)": 70.5, "step": 34930, "token_acc": 0.9201277955271565, "train_speed(iter/s)": 1.455771 }, { "epoch": 1.4967225054624909, "grad_norm": 3.9138572216033936, "learning_rate": 7.947250110103832e-05, "loss": 0.4417302131652832, "memory(GiB)": 70.5, "step": 34935, "token_acc": 0.9336099585062241, "train_speed(iter/s)": 1.455766 }, { "epoch": 1.49693672079174, "grad_norm": 0.16046546399593353, "learning_rate": 7.946706449080459e-05, "loss": 0.34991495609283446, "memory(GiB)": 70.5, "step": 34940, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.455789 }, { "epoch": 1.4971509361209887, "grad_norm": 5.34853982925415, "learning_rate": 7.946162734674734e-05, "loss": 0.6305210590362549, "memory(GiB)": 70.5, "step": 34945, "token_acc": 0.8526315789473684, "train_speed(iter/s)": 1.455799 }, { "epoch": 1.4973651514502377, "grad_norm": 2.871464490890503, "learning_rate": 7.945618966896508e-05, "loss": 0.15445680618286134, "memory(GiB)": 70.5, "step": 34950, "token_acc": 0.975609756097561, "train_speed(iter/s)": 1.455793 }, { "epoch": 1.4975793667794868, "grad_norm": 2.802719831466675, "learning_rate": 7.945075145755632e-05, "loss": 0.4820714950561523, "memory(GiB)": 70.5, "step": 34955, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.455804 }, { "epoch": 1.4977935821087356, "grad_norm": 7.055443286895752, "learning_rate": 7.944531271261955e-05, "loss": 0.30111088752746584, "memory(GiB)": 70.5, "step": 34960, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.455823 }, { "epoch": 1.4980077974379846, "grad_norm": 5.359797954559326, "learning_rate": 7.943987343425335e-05, "loss": 0.6861623287200928, "memory(GiB)": 70.5, "step": 34965, "token_acc": 0.89272030651341, "train_speed(iter/s)": 1.455831 }, { "epoch": 1.4982220127672337, "grad_norm": 2.3562827110290527, "learning_rate": 7.94344336225562e-05, "loss": 0.31561856269836425, "memory(GiB)": 70.5, "step": 34970, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.455845 }, { "epoch": 1.4984362280964825, "grad_norm": 1.470863938331604, "learning_rate": 7.942899327762668e-05, "loss": 0.33873226642608645, "memory(GiB)": 70.5, "step": 34975, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.455842 }, { "epoch": 1.4986504434257315, "grad_norm": 5.859305381774902, "learning_rate": 7.942355239956332e-05, "loss": 0.25083630084991454, "memory(GiB)": 70.5, "step": 34980, "token_acc": 0.9506578947368421, "train_speed(iter/s)": 1.455832 }, { "epoch": 1.4988646587549805, "grad_norm": 1.2638671398162842, "learning_rate": 7.941811098846472e-05, "loss": 0.27797105312347414, "memory(GiB)": 70.5, "step": 34985, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.455843 }, { "epoch": 1.4990788740842294, "grad_norm": 3.3670403957366943, "learning_rate": 7.941266904442945e-05, "loss": 0.38451566696166994, "memory(GiB)": 70.5, "step": 34990, "token_acc": 0.9053627760252366, "train_speed(iter/s)": 1.455846 }, { "epoch": 1.4992930894134784, "grad_norm": 3.5657553672790527, "learning_rate": 7.94072265675561e-05, "loss": 0.2448596477508545, "memory(GiB)": 70.5, "step": 34995, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.455844 }, { "epoch": 1.4995073047427274, "grad_norm": 1.9308090209960938, "learning_rate": 7.940178355794324e-05, "loss": 0.33464617729187013, "memory(GiB)": 70.5, "step": 35000, "token_acc": 0.9227941176470589, "train_speed(iter/s)": 1.455838 }, { "epoch": 1.4995073047427274, "eval_loss": 2.6519699096679688, "eval_runtime": 13.0337, "eval_samples_per_second": 7.672, "eval_steps_per_second": 7.672, "eval_token_acc": 0.42160737812911725, "step": 35000 }, { "epoch": 1.4997215200719762, "grad_norm": 3.875047206878662, "learning_rate": 7.93963400156895e-05, "loss": 0.39771366119384766, "memory(GiB)": 70.5, "step": 35005, "token_acc": 0.5579847908745247, "train_speed(iter/s)": 1.455028 }, { "epoch": 1.4999357354012253, "grad_norm": 4.273797512054443, "learning_rate": 7.939089594089347e-05, "loss": 0.5219860076904297, "memory(GiB)": 70.5, "step": 35010, "token_acc": 0.8724035608308606, "train_speed(iter/s)": 1.455015 }, { "epoch": 1.5001499507304743, "grad_norm": 2.9751172065734863, "learning_rate": 7.93854513336538e-05, "loss": 0.5277973175048828, "memory(GiB)": 70.5, "step": 35015, "token_acc": 0.8816568047337278, "train_speed(iter/s)": 1.455014 }, { "epoch": 1.5003641660597231, "grad_norm": 5.770832061767578, "learning_rate": 7.93800061940691e-05, "loss": 0.28682751655578614, "memory(GiB)": 70.5, "step": 35020, "token_acc": 0.9303135888501742, "train_speed(iter/s)": 1.455063 }, { "epoch": 1.5005783813889721, "grad_norm": 3.4294238090515137, "learning_rate": 7.937456052223804e-05, "loss": 0.36414761543273927, "memory(GiB)": 70.5, "step": 35025, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.455068 }, { "epoch": 1.5007925967182212, "grad_norm": 1.8764411211013794, "learning_rate": 7.936911431825926e-05, "loss": 0.5128159999847413, "memory(GiB)": 70.5, "step": 35030, "token_acc": 0.899641577060932, "train_speed(iter/s)": 1.455081 }, { "epoch": 1.50100681204747, "grad_norm": 3.0010619163513184, "learning_rate": 7.936366758223142e-05, "loss": 0.3104084014892578, "memory(GiB)": 70.5, "step": 35035, "token_acc": 0.9172185430463576, "train_speed(iter/s)": 1.455081 }, { "epoch": 1.501221027376719, "grad_norm": 2.6885361671447754, "learning_rate": 7.935822031425319e-05, "loss": 0.27163870334625245, "memory(GiB)": 70.5, "step": 35040, "token_acc": 0.9416342412451362, "train_speed(iter/s)": 1.455091 }, { "epoch": 1.501435242705968, "grad_norm": 1.3410334587097168, "learning_rate": 7.935277251442326e-05, "loss": 0.29900684356689455, "memory(GiB)": 70.5, "step": 35045, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.455088 }, { "epoch": 1.5016494580352169, "grad_norm": 4.9979400634765625, "learning_rate": 7.934732418284035e-05, "loss": 0.667492151260376, "memory(GiB)": 70.5, "step": 35050, "token_acc": 0.8594771241830066, "train_speed(iter/s)": 1.455075 }, { "epoch": 1.501863673364466, "grad_norm": 3.8195979595184326, "learning_rate": 7.93418753196031e-05, "loss": 0.36385226249694824, "memory(GiB)": 70.5, "step": 35055, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.455084 }, { "epoch": 1.502077888693715, "grad_norm": 5.3410868644714355, "learning_rate": 7.933642592481026e-05, "loss": 0.5010695934295655, "memory(GiB)": 70.5, "step": 35060, "token_acc": 0.90625, "train_speed(iter/s)": 1.455072 }, { "epoch": 1.5022921040229638, "grad_norm": 5.960295677185059, "learning_rate": 7.933097599856054e-05, "loss": 0.7856390953063965, "memory(GiB)": 70.5, "step": 35065, "token_acc": 0.8599348534201955, "train_speed(iter/s)": 1.455067 }, { "epoch": 1.5025063193522128, "grad_norm": 3.2431347370147705, "learning_rate": 7.932552554095268e-05, "loss": 0.3218043565750122, "memory(GiB)": 70.5, "step": 35070, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.45506 }, { "epoch": 1.5027205346814618, "grad_norm": 0.7311701774597168, "learning_rate": 7.932007455208542e-05, "loss": 0.47867765426635744, "memory(GiB)": 70.5, "step": 35075, "token_acc": 0.8952380952380953, "train_speed(iter/s)": 1.455093 }, { "epoch": 1.5029347500107106, "grad_norm": 3.4093806743621826, "learning_rate": 7.93146230320575e-05, "loss": 0.24455370903015136, "memory(GiB)": 70.5, "step": 35080, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.455095 }, { "epoch": 1.5031489653399597, "grad_norm": 3.133664608001709, "learning_rate": 7.93091709809677e-05, "loss": 0.5241455554962158, "memory(GiB)": 70.5, "step": 35085, "token_acc": 0.8713450292397661, "train_speed(iter/s)": 1.455095 }, { "epoch": 1.5033631806692087, "grad_norm": 4.236396789550781, "learning_rate": 7.930371839891475e-05, "loss": 0.3708005666732788, "memory(GiB)": 70.5, "step": 35090, "token_acc": 0.9294117647058824, "train_speed(iter/s)": 1.455079 }, { "epoch": 1.5035773959984575, "grad_norm": 6.616097450256348, "learning_rate": 7.929826528599746e-05, "loss": 0.5404676914215087, "memory(GiB)": 70.5, "step": 35095, "token_acc": 0.8923611111111112, "train_speed(iter/s)": 1.455083 }, { "epoch": 1.5037916113277066, "grad_norm": 0.6624740362167358, "learning_rate": 7.929281164231461e-05, "loss": 0.13406115770339966, "memory(GiB)": 70.5, "step": 35100, "token_acc": 0.956953642384106, "train_speed(iter/s)": 1.455084 }, { "epoch": 1.5040058266569556, "grad_norm": 2.599402904510498, "learning_rate": 7.928735746796501e-05, "loss": 0.36077821254730225, "memory(GiB)": 70.5, "step": 35105, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.455092 }, { "epoch": 1.5042200419862044, "grad_norm": 3.0226314067840576, "learning_rate": 7.928190276304744e-05, "loss": 0.423538064956665, "memory(GiB)": 70.5, "step": 35110, "token_acc": 0.89171974522293, "train_speed(iter/s)": 1.455085 }, { "epoch": 1.5044342573154537, "grad_norm": 3.2212231159210205, "learning_rate": 7.927644752766074e-05, "loss": 0.2901317596435547, "memory(GiB)": 70.5, "step": 35115, "token_acc": 0.9373219373219374, "train_speed(iter/s)": 1.455081 }, { "epoch": 1.5046484726447025, "grad_norm": 6.285838603973389, "learning_rate": 7.927099176190374e-05, "loss": 0.4977982521057129, "memory(GiB)": 70.5, "step": 35120, "token_acc": 0.8875739644970414, "train_speed(iter/s)": 1.455084 }, { "epoch": 1.5048626879739513, "grad_norm": 3.065859079360962, "learning_rate": 7.926553546587525e-05, "loss": 0.2471409559249878, "memory(GiB)": 70.5, "step": 35125, "token_acc": 0.9530201342281879, "train_speed(iter/s)": 1.455106 }, { "epoch": 1.5050769033032005, "grad_norm": 5.862625598907471, "learning_rate": 7.926007863967413e-05, "loss": 0.9141420364379883, "memory(GiB)": 70.5, "step": 35130, "token_acc": 0.8495934959349594, "train_speed(iter/s)": 1.455133 }, { "epoch": 1.5052911186324494, "grad_norm": 5.510425567626953, "learning_rate": 7.925462128339925e-05, "loss": 0.6017920494079589, "memory(GiB)": 70.5, "step": 35135, "token_acc": 0.8489425981873112, "train_speed(iter/s)": 1.455164 }, { "epoch": 1.5055053339616982, "grad_norm": 2.6765687465667725, "learning_rate": 7.924916339714945e-05, "loss": 0.3271677255630493, "memory(GiB)": 70.5, "step": 35140, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.455167 }, { "epoch": 1.5057195492909474, "grad_norm": 2.6980955600738525, "learning_rate": 7.924370498102363e-05, "loss": 0.12745318412780762, "memory(GiB)": 70.5, "step": 35145, "token_acc": 0.9752650176678446, "train_speed(iter/s)": 1.455179 }, { "epoch": 1.5059337646201962, "grad_norm": 5.171628952026367, "learning_rate": 7.923824603512065e-05, "loss": 0.3162405252456665, "memory(GiB)": 70.5, "step": 35150, "token_acc": 0.9288389513108615, "train_speed(iter/s)": 1.455179 }, { "epoch": 1.506147979949445, "grad_norm": 4.254791259765625, "learning_rate": 7.923278655953943e-05, "loss": 0.47292318344116213, "memory(GiB)": 70.5, "step": 35155, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.455161 }, { "epoch": 1.5063621952786943, "grad_norm": 1.5798839330673218, "learning_rate": 7.922732655437884e-05, "loss": 0.34424710273742676, "memory(GiB)": 70.5, "step": 35160, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.455145 }, { "epoch": 1.5065764106079431, "grad_norm": 4.43233060836792, "learning_rate": 7.922186601973782e-05, "loss": 0.2906650066375732, "memory(GiB)": 70.5, "step": 35165, "token_acc": 0.9049586776859504, "train_speed(iter/s)": 1.455148 }, { "epoch": 1.506790625937192, "grad_norm": 3.89077091217041, "learning_rate": 7.921640495571529e-05, "loss": 0.4097257137298584, "memory(GiB)": 70.5, "step": 35170, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.455168 }, { "epoch": 1.5070048412664412, "grad_norm": 6.8977789878845215, "learning_rate": 7.921094336241017e-05, "loss": 0.35735454559326174, "memory(GiB)": 70.5, "step": 35175, "token_acc": 0.9301587301587302, "train_speed(iter/s)": 1.455163 }, { "epoch": 1.50721905659569, "grad_norm": 1.6764369010925293, "learning_rate": 7.920548123992142e-05, "loss": 0.6757354259490966, "memory(GiB)": 70.5, "step": 35180, "token_acc": 0.8383458646616542, "train_speed(iter/s)": 1.45515 }, { "epoch": 1.5074332719249388, "grad_norm": 4.542453765869141, "learning_rate": 7.920001858834796e-05, "loss": 0.4430511951446533, "memory(GiB)": 70.5, "step": 35185, "token_acc": 0.912751677852349, "train_speed(iter/s)": 1.455146 }, { "epoch": 1.507647487254188, "grad_norm": 3.901641607284546, "learning_rate": 7.91945554077888e-05, "loss": 0.6272119998931884, "memory(GiB)": 70.5, "step": 35190, "token_acc": 0.8370607028753994, "train_speed(iter/s)": 1.455147 }, { "epoch": 1.5078617025834369, "grad_norm": 7.588398456573486, "learning_rate": 7.918909169834285e-05, "loss": 0.3477802753448486, "memory(GiB)": 70.5, "step": 35195, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.455132 }, { "epoch": 1.5080759179126857, "grad_norm": 0.2545609772205353, "learning_rate": 7.918362746010914e-05, "loss": 0.389675498008728, "memory(GiB)": 70.5, "step": 35200, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.455135 }, { "epoch": 1.508290133241935, "grad_norm": 7.011326313018799, "learning_rate": 7.917816269318665e-05, "loss": 0.7333412647247315, "memory(GiB)": 70.5, "step": 35205, "token_acc": 0.8464163822525598, "train_speed(iter/s)": 1.455138 }, { "epoch": 1.5085043485711838, "grad_norm": 2.0750815868377686, "learning_rate": 7.917269739767434e-05, "loss": 0.2329253911972046, "memory(GiB)": 70.5, "step": 35210, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.455124 }, { "epoch": 1.5087185639004326, "grad_norm": 4.031038284301758, "learning_rate": 7.916723157367129e-05, "loss": 0.4592780590057373, "memory(GiB)": 70.5, "step": 35215, "token_acc": 0.895397489539749, "train_speed(iter/s)": 1.45513 }, { "epoch": 1.5089327792296818, "grad_norm": 5.118336200714111, "learning_rate": 7.916176522127645e-05, "loss": 0.29103913307189944, "memory(GiB)": 70.5, "step": 35220, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.455125 }, { "epoch": 1.5091469945589306, "grad_norm": 4.378943920135498, "learning_rate": 7.91562983405889e-05, "loss": 0.3534574508666992, "memory(GiB)": 70.5, "step": 35225, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.45512 }, { "epoch": 1.5093612098881795, "grad_norm": 2.923551321029663, "learning_rate": 7.915083093170764e-05, "loss": 0.4809094429016113, "memory(GiB)": 70.5, "step": 35230, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.45512 }, { "epoch": 1.5095754252174287, "grad_norm": 1.2721339464187622, "learning_rate": 7.914536299473173e-05, "loss": 0.36352336406707764, "memory(GiB)": 70.5, "step": 35235, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.455124 }, { "epoch": 1.5097896405466775, "grad_norm": 3.755457639694214, "learning_rate": 7.913989452976023e-05, "loss": 0.5431818008422852, "memory(GiB)": 70.5, "step": 35240, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.455111 }, { "epoch": 1.5100038558759263, "grad_norm": 0.03166729584336281, "learning_rate": 7.913442553689221e-05, "loss": 0.12240279912948608, "memory(GiB)": 70.5, "step": 35245, "token_acc": 0.9792387543252595, "train_speed(iter/s)": 1.455114 }, { "epoch": 1.5102180712051756, "grad_norm": 3.718273401260376, "learning_rate": 7.912895601622675e-05, "loss": 0.3755930185317993, "memory(GiB)": 70.5, "step": 35250, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.455132 }, { "epoch": 1.5104322865344244, "grad_norm": 2.8416459560394287, "learning_rate": 7.912348596786291e-05, "loss": 0.4017319202423096, "memory(GiB)": 70.5, "step": 35255, "token_acc": 0.9107142857142857, "train_speed(iter/s)": 1.455137 }, { "epoch": 1.5106465018636732, "grad_norm": 5.6943678855896, "learning_rate": 7.91180153918998e-05, "loss": 0.4003596305847168, "memory(GiB)": 70.5, "step": 35260, "token_acc": 0.9149560117302052, "train_speed(iter/s)": 1.455154 }, { "epoch": 1.5108607171929225, "grad_norm": 5.183603763580322, "learning_rate": 7.911254428843654e-05, "loss": 0.4677162170410156, "memory(GiB)": 70.5, "step": 35265, "token_acc": 0.8860294117647058, "train_speed(iter/s)": 1.455152 }, { "epoch": 1.5110749325221713, "grad_norm": 4.569490909576416, "learning_rate": 7.910707265757222e-05, "loss": 0.6049376964569092, "memory(GiB)": 70.5, "step": 35270, "token_acc": 0.8621908127208481, "train_speed(iter/s)": 1.455139 }, { "epoch": 1.51128914785142, "grad_norm": 0.1802099645137787, "learning_rate": 7.910160049940598e-05, "loss": 0.28168301582336425, "memory(GiB)": 70.5, "step": 35275, "token_acc": 0.9447852760736196, "train_speed(iter/s)": 1.455149 }, { "epoch": 1.5115033631806694, "grad_norm": 0.1856347769498825, "learning_rate": 7.909612781403695e-05, "loss": 0.19219032526016236, "memory(GiB)": 70.5, "step": 35280, "token_acc": 0.975975975975976, "train_speed(iter/s)": 1.455154 }, { "epoch": 1.5117175785099182, "grad_norm": 2.595827579498291, "learning_rate": 7.909065460156427e-05, "loss": 0.2370211362838745, "memory(GiB)": 70.5, "step": 35285, "token_acc": 0.9375, "train_speed(iter/s)": 1.45517 }, { "epoch": 1.511931793839167, "grad_norm": 4.154641151428223, "learning_rate": 7.90851808620871e-05, "loss": 0.47914953231811525, "memory(GiB)": 70.5, "step": 35290, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.455162 }, { "epoch": 1.5121460091684162, "grad_norm": 8.493849754333496, "learning_rate": 7.907970659570457e-05, "loss": 0.2674244403839111, "memory(GiB)": 70.5, "step": 35295, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.455151 }, { "epoch": 1.512360224497665, "grad_norm": 2.3543434143066406, "learning_rate": 7.90742318025159e-05, "loss": 0.33461756706237794, "memory(GiB)": 70.5, "step": 35300, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.455154 }, { "epoch": 1.5125744398269139, "grad_norm": 4.833373546600342, "learning_rate": 7.906875648262022e-05, "loss": 0.250572943687439, "memory(GiB)": 70.5, "step": 35305, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.455161 }, { "epoch": 1.5127886551561631, "grad_norm": 0.799879789352417, "learning_rate": 7.906328063611677e-05, "loss": 0.32836503982543946, "memory(GiB)": 70.5, "step": 35310, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.455159 }, { "epoch": 1.513002870485412, "grad_norm": 3.6472413539886475, "learning_rate": 7.905780426310472e-05, "loss": 0.46687607765197753, "memory(GiB)": 70.5, "step": 35315, "token_acc": 0.8847736625514403, "train_speed(iter/s)": 1.455148 }, { "epoch": 1.5132170858146607, "grad_norm": 3.4826138019561768, "learning_rate": 7.905232736368328e-05, "loss": 0.3200789451599121, "memory(GiB)": 70.5, "step": 35320, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.455167 }, { "epoch": 1.51343130114391, "grad_norm": 3.362610340118408, "learning_rate": 7.90468499379517e-05, "loss": 0.5096293449401855, "memory(GiB)": 70.5, "step": 35325, "token_acc": 0.8879781420765027, "train_speed(iter/s)": 1.455164 }, { "epoch": 1.5136455164731588, "grad_norm": 9.29800796508789, "learning_rate": 7.904137198600917e-05, "loss": 0.3758652687072754, "memory(GiB)": 70.5, "step": 35330, "token_acc": 0.8895705521472392, "train_speed(iter/s)": 1.455151 }, { "epoch": 1.5138597318024076, "grad_norm": 3.5959177017211914, "learning_rate": 7.903589350795495e-05, "loss": 0.36475019454956054, "memory(GiB)": 70.5, "step": 35335, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.455151 }, { "epoch": 1.5140739471316569, "grad_norm": 5.585933685302734, "learning_rate": 7.903041450388828e-05, "loss": 0.2079242467880249, "memory(GiB)": 70.5, "step": 35340, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.455152 }, { "epoch": 1.5142881624609057, "grad_norm": 5.9693603515625, "learning_rate": 7.902493497390842e-05, "loss": 0.30733940601348875, "memory(GiB)": 70.5, "step": 35345, "token_acc": 0.941747572815534, "train_speed(iter/s)": 1.455148 }, { "epoch": 1.5145023777901545, "grad_norm": 3.898280382156372, "learning_rate": 7.901945491811462e-05, "loss": 0.3990234375, "memory(GiB)": 70.5, "step": 35350, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.455166 }, { "epoch": 1.5147165931194038, "grad_norm": 3.072967767715454, "learning_rate": 7.901397433660617e-05, "loss": 0.23954453468322753, "memory(GiB)": 70.5, "step": 35355, "token_acc": 0.9509433962264151, "train_speed(iter/s)": 1.45517 }, { "epoch": 1.5149308084486526, "grad_norm": 3.192047119140625, "learning_rate": 7.900849322948239e-05, "loss": 0.27290844917297363, "memory(GiB)": 70.5, "step": 35360, "token_acc": 0.9296636085626911, "train_speed(iter/s)": 1.455183 }, { "epoch": 1.5151450237779014, "grad_norm": 2.843644618988037, "learning_rate": 7.900301159684251e-05, "loss": 0.3199605464935303, "memory(GiB)": 70.5, "step": 35365, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.45519 }, { "epoch": 1.5153592391071506, "grad_norm": 3.9578211307525635, "learning_rate": 7.899752943878589e-05, "loss": 0.2850840091705322, "memory(GiB)": 70.5, "step": 35370, "token_acc": 0.9528985507246377, "train_speed(iter/s)": 1.455188 }, { "epoch": 1.5155734544363995, "grad_norm": 0.17286980152130127, "learning_rate": 7.89920467554118e-05, "loss": 0.1774985671043396, "memory(GiB)": 70.5, "step": 35375, "token_acc": 0.9576271186440678, "train_speed(iter/s)": 1.455193 }, { "epoch": 1.5157876697656483, "grad_norm": 3.993929147720337, "learning_rate": 7.898656354681961e-05, "loss": 0.4229094982147217, "memory(GiB)": 70.5, "step": 35380, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.455194 }, { "epoch": 1.5160018850948975, "grad_norm": 1.6528170108795166, "learning_rate": 7.89810798131086e-05, "loss": 0.05916541814804077, "memory(GiB)": 70.5, "step": 35385, "token_acc": 0.9766536964980544, "train_speed(iter/s)": 1.455213 }, { "epoch": 1.5162161004241463, "grad_norm": 4.6038336753845215, "learning_rate": 7.897559555437817e-05, "loss": 0.21145401000976563, "memory(GiB)": 70.5, "step": 35390, "token_acc": 0.9575163398692811, "train_speed(iter/s)": 1.455215 }, { "epoch": 1.5164303157533952, "grad_norm": 9.570601463317871, "learning_rate": 7.897011077072763e-05, "loss": 0.4185389518737793, "memory(GiB)": 70.5, "step": 35395, "token_acc": 0.9081632653061225, "train_speed(iter/s)": 1.455237 }, { "epoch": 1.5166445310826444, "grad_norm": 4.495016574859619, "learning_rate": 7.896462546225637e-05, "loss": 0.5254689693450928, "memory(GiB)": 70.5, "step": 35400, "token_acc": 0.8737201365187713, "train_speed(iter/s)": 1.455225 }, { "epoch": 1.5168587464118932, "grad_norm": 2.9343042373657227, "learning_rate": 7.895913962906374e-05, "loss": 0.2579438924789429, "memory(GiB)": 70.5, "step": 35405, "token_acc": 0.9518072289156626, "train_speed(iter/s)": 1.455224 }, { "epoch": 1.517072961741142, "grad_norm": 2.9900834560394287, "learning_rate": 7.895365327124912e-05, "loss": 0.2568881273269653, "memory(GiB)": 70.5, "step": 35410, "token_acc": 0.9430051813471503, "train_speed(iter/s)": 1.455231 }, { "epoch": 1.5172871770703913, "grad_norm": 3.561184883117676, "learning_rate": 7.894816638891194e-05, "loss": 0.441438627243042, "memory(GiB)": 70.5, "step": 35415, "token_acc": 0.9172185430463576, "train_speed(iter/s)": 1.455234 }, { "epoch": 1.51750139239964, "grad_norm": 1.9225085973739624, "learning_rate": 7.894267898215155e-05, "loss": 0.2996860980987549, "memory(GiB)": 70.5, "step": 35420, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.455233 }, { "epoch": 1.517715607728889, "grad_norm": 4.6699676513671875, "learning_rate": 7.893719105106739e-05, "loss": 0.31643147468566896, "memory(GiB)": 70.5, "step": 35425, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.455227 }, { "epoch": 1.5179298230581382, "grad_norm": 4.843156337738037, "learning_rate": 7.893170259575886e-05, "loss": 0.38980774879455565, "memory(GiB)": 70.5, "step": 35430, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.455219 }, { "epoch": 1.518144038387387, "grad_norm": 2.838552236557007, "learning_rate": 7.892621361632539e-05, "loss": 0.19291067123413086, "memory(GiB)": 70.5, "step": 35435, "token_acc": 0.9576923076923077, "train_speed(iter/s)": 1.455214 }, { "epoch": 1.5183582537166358, "grad_norm": 3.2316274642944336, "learning_rate": 7.892072411286644e-05, "loss": 0.35353827476501465, "memory(GiB)": 70.5, "step": 35440, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.455232 }, { "epoch": 1.518572469045885, "grad_norm": 4.179715633392334, "learning_rate": 7.891523408548143e-05, "loss": 0.24557342529296874, "memory(GiB)": 70.5, "step": 35445, "token_acc": 0.9307359307359307, "train_speed(iter/s)": 1.455253 }, { "epoch": 1.5187866843751339, "grad_norm": 2.3284826278686523, "learning_rate": 7.890974353426984e-05, "loss": 0.4084358215332031, "memory(GiB)": 70.5, "step": 35450, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.455264 }, { "epoch": 1.519000899704383, "grad_norm": 3.604797601699829, "learning_rate": 7.890425245933113e-05, "loss": 0.6232208728790283, "memory(GiB)": 70.5, "step": 35455, "token_acc": 0.8779661016949153, "train_speed(iter/s)": 1.455264 }, { "epoch": 1.519215115033632, "grad_norm": 4.275146007537842, "learning_rate": 7.889876086076475e-05, "loss": 0.4677189826965332, "memory(GiB)": 70.5, "step": 35460, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.45526 }, { "epoch": 1.5194293303628807, "grad_norm": 2.002976417541504, "learning_rate": 7.889326873867022e-05, "loss": 0.2061852216720581, "memory(GiB)": 70.5, "step": 35465, "token_acc": 0.9440559440559441, "train_speed(iter/s)": 1.45526 }, { "epoch": 1.5196435456921298, "grad_norm": 6.290721416473389, "learning_rate": 7.888777609314704e-05, "loss": 0.5458513259887695, "memory(GiB)": 70.5, "step": 35470, "token_acc": 0.8896797153024911, "train_speed(iter/s)": 1.455255 }, { "epoch": 1.5198577610213788, "grad_norm": 2.611698865890503, "learning_rate": 7.888228292429468e-05, "loss": 0.23593780994415284, "memory(GiB)": 70.5, "step": 35475, "token_acc": 0.9337748344370861, "train_speed(iter/s)": 1.455276 }, { "epoch": 1.5200719763506276, "grad_norm": 3.377850294113159, "learning_rate": 7.887678923221267e-05, "loss": 0.39882447719573977, "memory(GiB)": 70.5, "step": 35480, "token_acc": 0.9176136363636364, "train_speed(iter/s)": 1.455281 }, { "epoch": 1.5202861916798767, "grad_norm": 5.842268466949463, "learning_rate": 7.887129501700055e-05, "loss": 0.31245424747467043, "memory(GiB)": 70.5, "step": 35485, "token_acc": 0.934984520123839, "train_speed(iter/s)": 1.455291 }, { "epoch": 1.5205004070091257, "grad_norm": 0.152810201048851, "learning_rate": 7.886580027875782e-05, "loss": 0.4031352519989014, "memory(GiB)": 70.5, "step": 35490, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.455292 }, { "epoch": 1.5207146223383745, "grad_norm": 1.9831500053405762, "learning_rate": 7.886030501758404e-05, "loss": 0.2632662534713745, "memory(GiB)": 70.5, "step": 35495, "token_acc": 0.953307392996109, "train_speed(iter/s)": 1.455286 }, { "epoch": 1.5209288376676235, "grad_norm": 1.7557613849639893, "learning_rate": 7.885480923357878e-05, "loss": 0.2536567211151123, "memory(GiB)": 70.5, "step": 35500, "token_acc": 0.9574468085106383, "train_speed(iter/s)": 1.455285 }, { "epoch": 1.5209288376676235, "eval_loss": 2.5869197845458984, "eval_runtime": 13.8187, "eval_samples_per_second": 7.237, "eval_steps_per_second": 7.237, "eval_token_acc": 0.424, "step": 35500 }, { "epoch": 1.5211430529968726, "grad_norm": 1.7950143814086914, "learning_rate": 7.884931292684157e-05, "loss": 0.3926915884017944, "memory(GiB)": 70.5, "step": 35505, "token_acc": 0.5651769087523277, "train_speed(iter/s)": 1.454416 }, { "epoch": 1.5213572683261214, "grad_norm": 1.6639857292175293, "learning_rate": 7.884381609747198e-05, "loss": 0.419342565536499, "memory(GiB)": 70.5, "step": 35510, "token_acc": 0.9010238907849829, "train_speed(iter/s)": 1.454411 }, { "epoch": 1.5215714836553704, "grad_norm": 2.761254072189331, "learning_rate": 7.883831874556962e-05, "loss": 0.27197585105895994, "memory(GiB)": 70.5, "step": 35515, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.454396 }, { "epoch": 1.5217856989846195, "grad_norm": 2.748790740966797, "learning_rate": 7.883282087123407e-05, "loss": 0.5017034530639648, "memory(GiB)": 70.5, "step": 35520, "token_acc": 0.89453125, "train_speed(iter/s)": 1.454324 }, { "epoch": 1.5219999143138683, "grad_norm": 0.2744179964065552, "learning_rate": 7.88273224745649e-05, "loss": 0.2637808322906494, "memory(GiB)": 70.5, "step": 35525, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.45432 }, { "epoch": 1.5222141296431173, "grad_norm": 3.3327066898345947, "learning_rate": 7.882182355566177e-05, "loss": 0.3286318778991699, "memory(GiB)": 70.5, "step": 35530, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.454321 }, { "epoch": 1.5224283449723663, "grad_norm": 2.2406444549560547, "learning_rate": 7.881632411462424e-05, "loss": 0.24324989318847656, "memory(GiB)": 70.5, "step": 35535, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.454327 }, { "epoch": 1.5226425603016152, "grad_norm": 4.349251747131348, "learning_rate": 7.881082415155198e-05, "loss": 0.37428712844848633, "memory(GiB)": 70.5, "step": 35540, "token_acc": 0.9107692307692308, "train_speed(iter/s)": 1.454305 }, { "epoch": 1.5228567756308642, "grad_norm": 5.963205814361572, "learning_rate": 7.880532366654462e-05, "loss": 0.7305412769317627, "memory(GiB)": 70.5, "step": 35545, "token_acc": 0.8163265306122449, "train_speed(iter/s)": 1.454302 }, { "epoch": 1.5230709909601132, "grad_norm": 2.8718483448028564, "learning_rate": 7.879982265970178e-05, "loss": 0.3297011613845825, "memory(GiB)": 70.5, "step": 35550, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.454317 }, { "epoch": 1.523285206289362, "grad_norm": 1.885535478591919, "learning_rate": 7.879432113112316e-05, "loss": 0.29150776863098143, "memory(GiB)": 70.5, "step": 35555, "token_acc": 0.9494584837545126, "train_speed(iter/s)": 1.454316 }, { "epoch": 1.523499421618611, "grad_norm": 5.392949104309082, "learning_rate": 7.87888190809084e-05, "loss": 0.6229614734649658, "memory(GiB)": 70.5, "step": 35560, "token_acc": 0.8732876712328768, "train_speed(iter/s)": 1.454308 }, { "epoch": 1.52371363694786, "grad_norm": 5.446480751037598, "learning_rate": 7.878331650915716e-05, "loss": 0.32566978931427004, "memory(GiB)": 70.5, "step": 35565, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.454297 }, { "epoch": 1.523927852277109, "grad_norm": 3.696805715560913, "learning_rate": 7.877781341596915e-05, "loss": 0.4704679012298584, "memory(GiB)": 70.5, "step": 35570, "token_acc": 0.8977272727272727, "train_speed(iter/s)": 1.454307 }, { "epoch": 1.524142067606358, "grad_norm": 4.328130722045898, "learning_rate": 7.877230980144404e-05, "loss": 0.27168750762939453, "memory(GiB)": 70.5, "step": 35575, "token_acc": 0.9484126984126984, "train_speed(iter/s)": 1.454313 }, { "epoch": 1.524356282935607, "grad_norm": 3.516188621520996, "learning_rate": 7.876680566568157e-05, "loss": 0.4140778541564941, "memory(GiB)": 70.5, "step": 35580, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.454308 }, { "epoch": 1.5245704982648558, "grad_norm": 1.8971318006515503, "learning_rate": 7.876130100878142e-05, "loss": 0.35277752876281737, "memory(GiB)": 70.5, "step": 35585, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.454304 }, { "epoch": 1.5247847135941048, "grad_norm": 6.512512683868408, "learning_rate": 7.875579583084331e-05, "loss": 0.5542943000793457, "memory(GiB)": 70.5, "step": 35590, "token_acc": 0.9025974025974026, "train_speed(iter/s)": 1.454349 }, { "epoch": 1.5249989289233539, "grad_norm": 3.26126766204834, "learning_rate": 7.8750290131967e-05, "loss": 0.37936155796051024, "memory(GiB)": 70.5, "step": 35595, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.454345 }, { "epoch": 1.5252131442526027, "grad_norm": 6.493283748626709, "learning_rate": 7.874478391225221e-05, "loss": 0.4745510101318359, "memory(GiB)": 70.5, "step": 35600, "token_acc": 0.8881118881118881, "train_speed(iter/s)": 1.454378 }, { "epoch": 1.5254273595818517, "grad_norm": 4.33941125869751, "learning_rate": 7.87392771717987e-05, "loss": 0.28192505836486814, "memory(GiB)": 70.5, "step": 35605, "token_acc": 0.941717791411043, "train_speed(iter/s)": 1.454382 }, { "epoch": 1.5256415749111008, "grad_norm": 2.306516647338867, "learning_rate": 7.873376991070623e-05, "loss": 0.29585824012756345, "memory(GiB)": 70.5, "step": 35610, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.454403 }, { "epoch": 1.5258557902403496, "grad_norm": 6.04729700088501, "learning_rate": 7.872826212907453e-05, "loss": 0.4343556880950928, "memory(GiB)": 70.5, "step": 35615, "token_acc": 0.8976377952755905, "train_speed(iter/s)": 1.454408 }, { "epoch": 1.5260700055695986, "grad_norm": 3.8437130451202393, "learning_rate": 7.872275382700345e-05, "loss": 0.3954968214035034, "memory(GiB)": 70.5, "step": 35620, "token_acc": 0.9029126213592233, "train_speed(iter/s)": 1.454422 }, { "epoch": 1.5262842208988476, "grad_norm": 5.214289665222168, "learning_rate": 7.871724500459272e-05, "loss": 0.4889078617095947, "memory(GiB)": 70.5, "step": 35625, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.454413 }, { "epoch": 1.5264984362280964, "grad_norm": 4.324461460113525, "learning_rate": 7.871173566194217e-05, "loss": 0.35028629302978515, "memory(GiB)": 70.5, "step": 35630, "token_acc": 0.946058091286307, "train_speed(iter/s)": 1.454419 }, { "epoch": 1.5267126515573455, "grad_norm": 2.2253122329711914, "learning_rate": 7.870622579915158e-05, "loss": 0.5961165428161621, "memory(GiB)": 70.5, "step": 35635, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.454427 }, { "epoch": 1.5269268668865945, "grad_norm": 5.25931978225708, "learning_rate": 7.870071541632078e-05, "loss": 0.4913150787353516, "memory(GiB)": 70.5, "step": 35640, "token_acc": 0.9051724137931034, "train_speed(iter/s)": 1.454428 }, { "epoch": 1.5271410822158433, "grad_norm": 2.159656286239624, "learning_rate": 7.869520451354961e-05, "loss": 0.3891474962234497, "memory(GiB)": 70.5, "step": 35645, "token_acc": 0.9208860759493671, "train_speed(iter/s)": 1.454422 }, { "epoch": 1.5273552975450924, "grad_norm": 3.995546817779541, "learning_rate": 7.868969309093788e-05, "loss": 0.41381030082702636, "memory(GiB)": 70.5, "step": 35650, "token_acc": 0.9122137404580153, "train_speed(iter/s)": 1.454434 }, { "epoch": 1.5275695128743414, "grad_norm": 3.6339025497436523, "learning_rate": 7.868418114858545e-05, "loss": 0.32793848514556884, "memory(GiB)": 70.5, "step": 35655, "token_acc": 0.926530612244898, "train_speed(iter/s)": 1.454439 }, { "epoch": 1.5277837282035902, "grad_norm": 2.573699474334717, "learning_rate": 7.867866868659218e-05, "loss": 0.22508327960968016, "memory(GiB)": 70.5, "step": 35660, "token_acc": 0.9492753623188406, "train_speed(iter/s)": 1.454433 }, { "epoch": 1.5279979435328392, "grad_norm": 3.572819709777832, "learning_rate": 7.867315570505792e-05, "loss": 0.36694087982177737, "memory(GiB)": 70.5, "step": 35665, "token_acc": 0.9323529411764706, "train_speed(iter/s)": 1.454436 }, { "epoch": 1.5282121588620883, "grad_norm": 2.6973936557769775, "learning_rate": 7.866764220408255e-05, "loss": 0.5283097267150879, "memory(GiB)": 70.5, "step": 35670, "token_acc": 0.899390243902439, "train_speed(iter/s)": 1.454431 }, { "epoch": 1.528426374191337, "grad_norm": 2.6575043201446533, "learning_rate": 7.866212818376594e-05, "loss": 0.18520693778991698, "memory(GiB)": 70.5, "step": 35675, "token_acc": 0.9587155963302753, "train_speed(iter/s)": 1.454436 }, { "epoch": 1.5286405895205861, "grad_norm": 4.662832260131836, "learning_rate": 7.8656613644208e-05, "loss": 0.6513695240020752, "memory(GiB)": 70.5, "step": 35680, "token_acc": 0.890282131661442, "train_speed(iter/s)": 1.454472 }, { "epoch": 1.5288548048498352, "grad_norm": 3.538331985473633, "learning_rate": 7.865109858550862e-05, "loss": 0.5083042621612549, "memory(GiB)": 70.5, "step": 35685, "token_acc": 0.8813559322033898, "train_speed(iter/s)": 1.454495 }, { "epoch": 1.529069020179084, "grad_norm": 6.289170265197754, "learning_rate": 7.86455830077677e-05, "loss": 0.6487184047698975, "memory(GiB)": 70.5, "step": 35690, "token_acc": 0.8603174603174604, "train_speed(iter/s)": 1.454496 }, { "epoch": 1.529283235508333, "grad_norm": 1.8637338876724243, "learning_rate": 7.864006691108519e-05, "loss": 0.3687494516372681, "memory(GiB)": 70.5, "step": 35695, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.454497 }, { "epoch": 1.529497450837582, "grad_norm": 2.35659122467041, "learning_rate": 7.863455029556098e-05, "loss": 0.4288932800292969, "memory(GiB)": 70.5, "step": 35700, "token_acc": 0.8986928104575164, "train_speed(iter/s)": 1.454503 }, { "epoch": 1.5297116661668309, "grad_norm": 2.96008038520813, "learning_rate": 7.862903316129504e-05, "loss": 0.49300575256347656, "memory(GiB)": 70.5, "step": 35705, "token_acc": 0.9008264462809917, "train_speed(iter/s)": 1.454506 }, { "epoch": 1.5299258814960799, "grad_norm": 0.17014656960964203, "learning_rate": 7.862351550838732e-05, "loss": 0.5109248638153077, "memory(GiB)": 70.5, "step": 35710, "token_acc": 0.9127906976744186, "train_speed(iter/s)": 1.454513 }, { "epoch": 1.530140096825329, "grad_norm": 3.3138997554779053, "learning_rate": 7.861799733693775e-05, "loss": 0.2746044397354126, "memory(GiB)": 70.5, "step": 35715, "token_acc": 0.9379562043795621, "train_speed(iter/s)": 1.454498 }, { "epoch": 1.5303543121545777, "grad_norm": 6.1462483406066895, "learning_rate": 7.86124786470463e-05, "loss": 0.5388484954833984, "memory(GiB)": 70.5, "step": 35720, "token_acc": 0.8859934853420195, "train_speed(iter/s)": 1.454496 }, { "epoch": 1.5305685274838268, "grad_norm": 2.9319067001342773, "learning_rate": 7.8606959438813e-05, "loss": 0.12769680023193358, "memory(GiB)": 70.5, "step": 35725, "token_acc": 0.9761904761904762, "train_speed(iter/s)": 1.454501 }, { "epoch": 1.5307827428130758, "grad_norm": 2.3843541145324707, "learning_rate": 7.860143971233776e-05, "loss": 0.5081887245178223, "memory(GiB)": 70.5, "step": 35730, "token_acc": 0.8980392156862745, "train_speed(iter/s)": 1.454503 }, { "epoch": 1.5309969581423246, "grad_norm": 0.7897017598152161, "learning_rate": 7.859591946772062e-05, "loss": 0.22180466651916503, "memory(GiB)": 70.5, "step": 35735, "token_acc": 0.9587628865979382, "train_speed(iter/s)": 1.4545 }, { "epoch": 1.5312111734715736, "grad_norm": 2.0018341541290283, "learning_rate": 7.859039870506158e-05, "loss": 0.44312586784362795, "memory(GiB)": 70.5, "step": 35740, "token_acc": 0.9059561128526645, "train_speed(iter/s)": 1.454509 }, { "epoch": 1.5314253888008227, "grad_norm": 6.255533218383789, "learning_rate": 7.858487742446063e-05, "loss": 0.3016995429992676, "memory(GiB)": 70.5, "step": 35745, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.454503 }, { "epoch": 1.5316396041300715, "grad_norm": 1.8422714471817017, "learning_rate": 7.857935562601783e-05, "loss": 0.6305891990661621, "memory(GiB)": 70.5, "step": 35750, "token_acc": 0.8716417910447761, "train_speed(iter/s)": 1.454522 }, { "epoch": 1.5318538194593205, "grad_norm": 3.7248027324676514, "learning_rate": 7.857383330983319e-05, "loss": 0.4698214054107666, "memory(GiB)": 70.5, "step": 35755, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.454541 }, { "epoch": 1.5320680347885696, "grad_norm": 5.276378631591797, "learning_rate": 7.856831047600674e-05, "loss": 0.5537188529968262, "memory(GiB)": 70.5, "step": 35760, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.454557 }, { "epoch": 1.5322822501178184, "grad_norm": 4.1958441734313965, "learning_rate": 7.856278712463856e-05, "loss": 0.44065265655517577, "memory(GiB)": 70.5, "step": 35765, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.454558 }, { "epoch": 1.5324964654470674, "grad_norm": 1.812787413597107, "learning_rate": 7.855726325582869e-05, "loss": 0.342342734336853, "memory(GiB)": 70.5, "step": 35770, "token_acc": 0.9122137404580153, "train_speed(iter/s)": 1.454568 }, { "epoch": 1.5327106807763164, "grad_norm": 2.895254373550415, "learning_rate": 7.855173886967722e-05, "loss": 0.5885238170623779, "memory(GiB)": 70.5, "step": 35775, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.454604 }, { "epoch": 1.5329248961055653, "grad_norm": 2.623488187789917, "learning_rate": 7.854621396628421e-05, "loss": 0.3065093994140625, "memory(GiB)": 70.5, "step": 35780, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.454622 }, { "epoch": 1.5331391114348143, "grad_norm": 3.567392110824585, "learning_rate": 7.854068854574976e-05, "loss": 0.15203816890716554, "memory(GiB)": 70.5, "step": 35785, "token_acc": 0.9736842105263158, "train_speed(iter/s)": 1.454623 }, { "epoch": 1.5333533267640633, "grad_norm": 9.753863334655762, "learning_rate": 7.853516260817395e-05, "loss": 0.9470474243164062, "memory(GiB)": 70.5, "step": 35790, "token_acc": 0.7840236686390533, "train_speed(iter/s)": 1.454604 }, { "epoch": 1.5335675420933121, "grad_norm": 5.180292129516602, "learning_rate": 7.85296361536569e-05, "loss": 0.5246982574462891, "memory(GiB)": 70.5, "step": 35795, "token_acc": 0.8807692307692307, "train_speed(iter/s)": 1.454601 }, { "epoch": 1.5337817574225612, "grad_norm": 10.85864543914795, "learning_rate": 7.852410918229874e-05, "loss": 0.680355453491211, "memory(GiB)": 70.5, "step": 35800, "token_acc": 0.8600682593856656, "train_speed(iter/s)": 1.454614 }, { "epoch": 1.5339959727518102, "grad_norm": 2.3720128536224365, "learning_rate": 7.851858169419959e-05, "loss": 0.26025469303131105, "memory(GiB)": 70.5, "step": 35805, "token_acc": 0.9372822299651568, "train_speed(iter/s)": 1.454618 }, { "epoch": 1.534210188081059, "grad_norm": 8.34424877166748, "learning_rate": 7.851305368945956e-05, "loss": 0.47978363037109373, "memory(GiB)": 70.5, "step": 35810, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.454637 }, { "epoch": 1.534424403410308, "grad_norm": 2.0183446407318115, "learning_rate": 7.850752516817883e-05, "loss": 0.25615825653076174, "memory(GiB)": 70.5, "step": 35815, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.454642 }, { "epoch": 1.534638618739557, "grad_norm": 2.8009073734283447, "learning_rate": 7.850199613045754e-05, "loss": 0.3213240146636963, "memory(GiB)": 70.5, "step": 35820, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.454654 }, { "epoch": 1.534852834068806, "grad_norm": 3.79608416557312, "learning_rate": 7.849646657639585e-05, "loss": 0.34736127853393556, "memory(GiB)": 70.5, "step": 35825, "token_acc": 0.9258160237388724, "train_speed(iter/s)": 1.454656 }, { "epoch": 1.535067049398055, "grad_norm": 5.199913501739502, "learning_rate": 7.849093650609392e-05, "loss": 0.478715181350708, "memory(GiB)": 70.5, "step": 35830, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.454674 }, { "epoch": 1.535281264727304, "grad_norm": 3.6183464527130127, "learning_rate": 7.848540591965197e-05, "loss": 0.39489316940307617, "memory(GiB)": 70.5, "step": 35835, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.454694 }, { "epoch": 1.5354954800565528, "grad_norm": 2.389488697052002, "learning_rate": 7.847987481717017e-05, "loss": 0.563099479675293, "memory(GiB)": 70.5, "step": 35840, "token_acc": 0.9012738853503185, "train_speed(iter/s)": 1.454703 }, { "epoch": 1.5357096953858018, "grad_norm": 4.252538204193115, "learning_rate": 7.847434319874871e-05, "loss": 0.6009092330932617, "memory(GiB)": 70.5, "step": 35845, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.454714 }, { "epoch": 1.5359239107150509, "grad_norm": 3.097358465194702, "learning_rate": 7.846881106448781e-05, "loss": 0.32625794410705566, "memory(GiB)": 70.5, "step": 35850, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.454708 }, { "epoch": 1.5361381260442997, "grad_norm": 1.9700125455856323, "learning_rate": 7.846327841448771e-05, "loss": 0.23868610858917236, "memory(GiB)": 70.5, "step": 35855, "token_acc": 0.944078947368421, "train_speed(iter/s)": 1.45471 }, { "epoch": 1.5363523413735487, "grad_norm": 4.818068027496338, "learning_rate": 7.84577452488486e-05, "loss": 0.41489486694335936, "memory(GiB)": 70.5, "step": 35860, "token_acc": 0.889261744966443, "train_speed(iter/s)": 1.454712 }, { "epoch": 1.5365665567027977, "grad_norm": 2.347099542617798, "learning_rate": 7.845221156767076e-05, "loss": 0.15153591632843016, "memory(GiB)": 70.5, "step": 35865, "token_acc": 0.9576547231270358, "train_speed(iter/s)": 1.454724 }, { "epoch": 1.5367807720320465, "grad_norm": 2.6575143337249756, "learning_rate": 7.84466773710544e-05, "loss": 0.36276266574859617, "memory(GiB)": 70.5, "step": 35870, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.45472 }, { "epoch": 1.5369949873612956, "grad_norm": 2.7446796894073486, "learning_rate": 7.844114265909979e-05, "loss": 0.5376415729522706, "memory(GiB)": 70.5, "step": 35875, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.454725 }, { "epoch": 1.5372092026905446, "grad_norm": 0.6968674659729004, "learning_rate": 7.843560743190721e-05, "loss": 0.3510243892669678, "memory(GiB)": 70.5, "step": 35880, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.454744 }, { "epoch": 1.5374234180197934, "grad_norm": 3.253208875656128, "learning_rate": 7.843007168957693e-05, "loss": 0.2683022737503052, "memory(GiB)": 70.5, "step": 35885, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.454738 }, { "epoch": 1.5376376333490425, "grad_norm": 5.895941734313965, "learning_rate": 7.842453543220924e-05, "loss": 0.6346344470977783, "memory(GiB)": 70.5, "step": 35890, "token_acc": 0.8547008547008547, "train_speed(iter/s)": 1.45476 }, { "epoch": 1.5378518486782915, "grad_norm": 4.327568054199219, "learning_rate": 7.841899865990441e-05, "loss": 0.4758888244628906, "memory(GiB)": 70.5, "step": 35895, "token_acc": 0.9219653179190751, "train_speed(iter/s)": 1.454773 }, { "epoch": 1.5380660640075403, "grad_norm": 3.3087432384490967, "learning_rate": 7.841346137276276e-05, "loss": 0.27228970527648927, "memory(GiB)": 70.5, "step": 35900, "token_acc": 0.9243697478991597, "train_speed(iter/s)": 1.454767 }, { "epoch": 1.5382802793367893, "grad_norm": 0.46942636370658875, "learning_rate": 7.84079235708846e-05, "loss": 0.23606908321380615, "memory(GiB)": 70.5, "step": 35905, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.454769 }, { "epoch": 1.5384944946660384, "grad_norm": 2.974081516265869, "learning_rate": 7.840238525437027e-05, "loss": 0.36043787002563477, "memory(GiB)": 70.5, "step": 35910, "token_acc": 0.9079497907949791, "train_speed(iter/s)": 1.454764 }, { "epoch": 1.5387087099952872, "grad_norm": 3.7242274284362793, "learning_rate": 7.839684642332008e-05, "loss": 0.3619387626647949, "memory(GiB)": 70.5, "step": 35915, "token_acc": 0.9054545454545454, "train_speed(iter/s)": 1.454757 }, { "epoch": 1.5389229253245362, "grad_norm": 0.4083687365055084, "learning_rate": 7.839130707783438e-05, "loss": 0.21530675888061523, "memory(GiB)": 70.5, "step": 35920, "token_acc": 0.952076677316294, "train_speed(iter/s)": 1.454761 }, { "epoch": 1.5391371406537853, "grad_norm": 6.69803524017334, "learning_rate": 7.838576721801351e-05, "loss": 0.38328778743743896, "memory(GiB)": 70.5, "step": 35925, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.454758 }, { "epoch": 1.539351355983034, "grad_norm": 5.55518102645874, "learning_rate": 7.838022684395786e-05, "loss": 0.18129570484161378, "memory(GiB)": 70.5, "step": 35930, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.454779 }, { "epoch": 1.539565571312283, "grad_norm": 1.131928563117981, "learning_rate": 7.837468595576777e-05, "loss": 0.3629601001739502, "memory(GiB)": 70.5, "step": 35935, "token_acc": 0.9272151898734177, "train_speed(iter/s)": 1.45478 }, { "epoch": 1.5397797866415321, "grad_norm": 2.2748425006866455, "learning_rate": 7.836914455354362e-05, "loss": 0.4428565979003906, "memory(GiB)": 70.5, "step": 35940, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.454775 }, { "epoch": 1.539994001970781, "grad_norm": 1.4993411302566528, "learning_rate": 7.83636026373858e-05, "loss": 0.5685799598693848, "memory(GiB)": 70.5, "step": 35945, "token_acc": 0.906832298136646, "train_speed(iter/s)": 1.454782 }, { "epoch": 1.54020821730003, "grad_norm": 3.3303754329681396, "learning_rate": 7.835806020739472e-05, "loss": 0.3156765937805176, "memory(GiB)": 70.5, "step": 35950, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.454806 }, { "epoch": 1.540422432629279, "grad_norm": 7.6702494621276855, "learning_rate": 7.835251726367078e-05, "loss": 0.4070268630981445, "memory(GiB)": 70.5, "step": 35955, "token_acc": 0.9181034482758621, "train_speed(iter/s)": 1.454821 }, { "epoch": 1.5406366479585278, "grad_norm": 3.7111363410949707, "learning_rate": 7.834697380631438e-05, "loss": 0.27619335651397703, "memory(GiB)": 70.5, "step": 35960, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.45483 }, { "epoch": 1.5408508632877769, "grad_norm": 0.3124500811100006, "learning_rate": 7.834142983542597e-05, "loss": 0.4045262336730957, "memory(GiB)": 70.5, "step": 35965, "token_acc": 0.8952380952380953, "train_speed(iter/s)": 1.454851 }, { "epoch": 1.541065078617026, "grad_norm": 3.3677101135253906, "learning_rate": 7.833588535110597e-05, "loss": 0.3124194622039795, "memory(GiB)": 70.5, "step": 35970, "token_acc": 0.9205020920502092, "train_speed(iter/s)": 1.45485 }, { "epoch": 1.5412792939462747, "grad_norm": 4.6144819259643555, "learning_rate": 7.833034035345483e-05, "loss": 0.6550428867340088, "memory(GiB)": 70.5, "step": 35975, "token_acc": 0.8615384615384616, "train_speed(iter/s)": 1.454849 }, { "epoch": 1.5414935092755238, "grad_norm": 3.0141894817352295, "learning_rate": 7.832479484257298e-05, "loss": 0.3450296878814697, "memory(GiB)": 70.5, "step": 35980, "token_acc": 0.9245901639344263, "train_speed(iter/s)": 1.454857 }, { "epoch": 1.5417077246047728, "grad_norm": 2.9122722148895264, "learning_rate": 7.831924881856092e-05, "loss": 0.3575387477874756, "memory(GiB)": 70.5, "step": 35985, "token_acc": 0.9292307692307692, "train_speed(iter/s)": 1.45484 }, { "epoch": 1.5419219399340216, "grad_norm": 3.3025293350219727, "learning_rate": 7.831370228151909e-05, "loss": 0.3050563812255859, "memory(GiB)": 70.5, "step": 35990, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.454869 }, { "epoch": 1.5421361552632706, "grad_norm": 2.133110284805298, "learning_rate": 7.830815523154799e-05, "loss": 0.35400824546813964, "memory(GiB)": 70.5, "step": 35995, "token_acc": 0.9343283582089552, "train_speed(iter/s)": 1.454899 }, { "epoch": 1.5423503705925197, "grad_norm": 5.6972880363464355, "learning_rate": 7.83026076687481e-05, "loss": 0.4009350299835205, "memory(GiB)": 70.5, "step": 36000, "token_acc": 0.8861209964412812, "train_speed(iter/s)": 1.454904 }, { "epoch": 1.5423503705925197, "eval_loss": 2.6697757244110107, "eval_runtime": 13.9267, "eval_samples_per_second": 7.18, "eval_steps_per_second": 7.18, "eval_token_acc": 0.427307206068268, "step": 36000 }, { "epoch": 1.5425645859217685, "grad_norm": 5.239740371704102, "learning_rate": 7.829705959321993e-05, "loss": 0.5152445793151855, "memory(GiB)": 70.5, "step": 36005, "token_acc": 0.5663716814159292, "train_speed(iter/s)": 1.454014 }, { "epoch": 1.5427788012510175, "grad_norm": 2.935922145843506, "learning_rate": 7.829151100506396e-05, "loss": 0.3559565544128418, "memory(GiB)": 70.5, "step": 36010, "token_acc": 0.9037037037037037, "train_speed(iter/s)": 1.454014 }, { "epoch": 1.5429930165802666, "grad_norm": 9.57572078704834, "learning_rate": 7.828596190438075e-05, "loss": 0.3869513511657715, "memory(GiB)": 70.5, "step": 36015, "token_acc": 0.9042145593869731, "train_speed(iter/s)": 1.45406 }, { "epoch": 1.5432072319095154, "grad_norm": 9.264183044433594, "learning_rate": 7.828041229127079e-05, "loss": 0.24991273880004883, "memory(GiB)": 70.5, "step": 36020, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.454082 }, { "epoch": 1.5434214472387644, "grad_norm": 1.2252815961837769, "learning_rate": 7.827486216583465e-05, "loss": 0.43653383255004885, "memory(GiB)": 70.5, "step": 36025, "token_acc": 0.9030303030303031, "train_speed(iter/s)": 1.454079 }, { "epoch": 1.5436356625680134, "grad_norm": 0.5626189112663269, "learning_rate": 7.826931152817283e-05, "loss": 0.10504354238510132, "memory(GiB)": 70.5, "step": 36030, "token_acc": 0.958041958041958, "train_speed(iter/s)": 1.454092 }, { "epoch": 1.5438498778972622, "grad_norm": 5.1091814041137695, "learning_rate": 7.826376037838594e-05, "loss": 0.28080196380615235, "memory(GiB)": 70.5, "step": 36035, "token_acc": 0.9341317365269461, "train_speed(iter/s)": 1.454103 }, { "epoch": 1.5440640932265113, "grad_norm": 4.632138252258301, "learning_rate": 7.82582087165745e-05, "loss": 0.6149715900421142, "memory(GiB)": 70.5, "step": 36040, "token_acc": 0.8557377049180328, "train_speed(iter/s)": 1.454124 }, { "epoch": 1.5442783085557603, "grad_norm": 4.848214149475098, "learning_rate": 7.82526565428391e-05, "loss": 0.487932014465332, "memory(GiB)": 70.5, "step": 36045, "token_acc": 0.915129151291513, "train_speed(iter/s)": 1.454126 }, { "epoch": 1.5444925238850091, "grad_norm": 4.0383501052856445, "learning_rate": 7.824710385728033e-05, "loss": 0.44413051605224607, "memory(GiB)": 70.5, "step": 36050, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.45412 }, { "epoch": 1.5447067392142582, "grad_norm": 14.12353801727295, "learning_rate": 7.824155065999879e-05, "loss": 0.6255293846130371, "memory(GiB)": 70.5, "step": 36055, "token_acc": 0.8659420289855072, "train_speed(iter/s)": 1.454141 }, { "epoch": 1.5449209545435072, "grad_norm": 1.7216993570327759, "learning_rate": 7.823599695109504e-05, "loss": 0.5278702735900879, "memory(GiB)": 70.5, "step": 36060, "token_acc": 0.8808664259927798, "train_speed(iter/s)": 1.454142 }, { "epoch": 1.545135169872756, "grad_norm": 7.22850227355957, "learning_rate": 7.823044273066975e-05, "loss": 0.7018226146697998, "memory(GiB)": 70.5, "step": 36065, "token_acc": 0.8295454545454546, "train_speed(iter/s)": 1.454156 }, { "epoch": 1.545349385202005, "grad_norm": 2.6711509227752686, "learning_rate": 7.822488799882348e-05, "loss": 0.3063040256500244, "memory(GiB)": 70.5, "step": 36070, "token_acc": 0.9400630914826499, "train_speed(iter/s)": 1.45416 }, { "epoch": 1.545563600531254, "grad_norm": 2.268249988555908, "learning_rate": 7.821933275565691e-05, "loss": 0.4234921455383301, "memory(GiB)": 70.5, "step": 36075, "token_acc": 0.9216300940438872, "train_speed(iter/s)": 1.454154 }, { "epoch": 1.545777815860503, "grad_norm": 4.090318202972412, "learning_rate": 7.821377700127063e-05, "loss": 0.5019046306610108, "memory(GiB)": 70.5, "step": 36080, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.454152 }, { "epoch": 1.545992031189752, "grad_norm": 2.2411880493164062, "learning_rate": 7.820822073576534e-05, "loss": 0.2503363847732544, "memory(GiB)": 70.5, "step": 36085, "token_acc": 0.9363636363636364, "train_speed(iter/s)": 1.454138 }, { "epoch": 1.546206246519001, "grad_norm": 4.749850273132324, "learning_rate": 7.820266395924164e-05, "loss": 0.40471768379211426, "memory(GiB)": 70.5, "step": 36090, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.454136 }, { "epoch": 1.5464204618482498, "grad_norm": 2.7579171657562256, "learning_rate": 7.819710667180024e-05, "loss": 0.2979117870330811, "memory(GiB)": 70.5, "step": 36095, "token_acc": 0.937888198757764, "train_speed(iter/s)": 1.454134 }, { "epoch": 1.5466346771774988, "grad_norm": 3.797621250152588, "learning_rate": 7.819154887354182e-05, "loss": 0.47805142402648926, "memory(GiB)": 70.5, "step": 36100, "token_acc": 0.8960244648318043, "train_speed(iter/s)": 1.454127 }, { "epoch": 1.5468488925067478, "grad_norm": 4.037912845611572, "learning_rate": 7.818599056456701e-05, "loss": 0.3473518371582031, "memory(GiB)": 70.5, "step": 36105, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.454172 }, { "epoch": 1.5470631078359967, "grad_norm": 2.621657371520996, "learning_rate": 7.818043174497656e-05, "loss": 0.5513715744018555, "memory(GiB)": 70.5, "step": 36110, "token_acc": 0.8926174496644296, "train_speed(iter/s)": 1.454177 }, { "epoch": 1.5472773231652457, "grad_norm": 5.324483871459961, "learning_rate": 7.817487241487115e-05, "loss": 0.5657027721405029, "memory(GiB)": 70.5, "step": 36115, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.454205 }, { "epoch": 1.5474915384944947, "grad_norm": 3.4712021350860596, "learning_rate": 7.816931257435151e-05, "loss": 0.4774474143981934, "memory(GiB)": 70.5, "step": 36120, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.454207 }, { "epoch": 1.5477057538237435, "grad_norm": 5.096844673156738, "learning_rate": 7.816375222351833e-05, "loss": 0.46150927543640136, "memory(GiB)": 70.5, "step": 36125, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.454225 }, { "epoch": 1.5479199691529926, "grad_norm": 2.1582937240600586, "learning_rate": 7.815819136247238e-05, "loss": 0.17468669414520263, "memory(GiB)": 70.5, "step": 36130, "token_acc": 0.946031746031746, "train_speed(iter/s)": 1.454226 }, { "epoch": 1.5481341844822416, "grad_norm": 2.23860764503479, "learning_rate": 7.815262999131436e-05, "loss": 0.26679871082305906, "memory(GiB)": 70.5, "step": 36135, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.454231 }, { "epoch": 1.5483483998114904, "grad_norm": 2.96530818939209, "learning_rate": 7.814706811014504e-05, "loss": 0.7650543689727783, "memory(GiB)": 70.5, "step": 36140, "token_acc": 0.8554216867469879, "train_speed(iter/s)": 1.45424 }, { "epoch": 1.5485626151407395, "grad_norm": 1.6170456409454346, "learning_rate": 7.814150571906517e-05, "loss": 0.40576887130737305, "memory(GiB)": 70.5, "step": 36145, "token_acc": 0.9006622516556292, "train_speed(iter/s)": 1.454248 }, { "epoch": 1.5487768304699885, "grad_norm": 4.57145357131958, "learning_rate": 7.813594281817555e-05, "loss": 0.24271225929260254, "memory(GiB)": 70.5, "step": 36150, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.454259 }, { "epoch": 1.5489910457992373, "grad_norm": 4.674867630004883, "learning_rate": 7.813037940757692e-05, "loss": 0.6397357940673828, "memory(GiB)": 70.5, "step": 36155, "token_acc": 0.8721311475409836, "train_speed(iter/s)": 1.454255 }, { "epoch": 1.5492052611284863, "grad_norm": 3.404175043106079, "learning_rate": 7.812481548737007e-05, "loss": 0.5701194763183594, "memory(GiB)": 70.5, "step": 36160, "token_acc": 0.8803680981595092, "train_speed(iter/s)": 1.45426 }, { "epoch": 1.5494194764577354, "grad_norm": 2.87174391746521, "learning_rate": 7.811925105765581e-05, "loss": 0.43214855194091795, "memory(GiB)": 70.5, "step": 36165, "token_acc": 0.9265536723163842, "train_speed(iter/s)": 1.454262 }, { "epoch": 1.5496336917869842, "grad_norm": 8.763022422790527, "learning_rate": 7.811368611853493e-05, "loss": 0.3442173957824707, "memory(GiB)": 70.5, "step": 36170, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.454262 }, { "epoch": 1.5498479071162332, "grad_norm": 6.462432861328125, "learning_rate": 7.810812067010827e-05, "loss": 0.47447566986083983, "memory(GiB)": 70.5, "step": 36175, "token_acc": 0.8904109589041096, "train_speed(iter/s)": 1.454283 }, { "epoch": 1.5500621224454822, "grad_norm": 4.167203426361084, "learning_rate": 7.810255471247663e-05, "loss": 0.6184118270874024, "memory(GiB)": 70.5, "step": 36180, "token_acc": 0.853035143769968, "train_speed(iter/s)": 1.454294 }, { "epoch": 1.550276337774731, "grad_norm": 2.9044504165649414, "learning_rate": 7.809698824574085e-05, "loss": 0.4331668853759766, "memory(GiB)": 70.5, "step": 36185, "token_acc": 0.9081967213114754, "train_speed(iter/s)": 1.454294 }, { "epoch": 1.55049055310398, "grad_norm": 2.5598108768463135, "learning_rate": 7.809142127000176e-05, "loss": 0.393706226348877, "memory(GiB)": 70.5, "step": 36190, "token_acc": 0.9002932551319648, "train_speed(iter/s)": 1.454289 }, { "epoch": 1.5507047684332291, "grad_norm": 2.5497894287109375, "learning_rate": 7.808585378536024e-05, "loss": 0.8420129776000976, "memory(GiB)": 70.5, "step": 36195, "token_acc": 0.839541547277937, "train_speed(iter/s)": 1.45431 }, { "epoch": 1.550918983762478, "grad_norm": 2.6351492404937744, "learning_rate": 7.808028579191711e-05, "loss": 0.44678621292114257, "memory(GiB)": 70.5, "step": 36200, "token_acc": 0.8882521489971347, "train_speed(iter/s)": 1.454322 }, { "epoch": 1.551133199091727, "grad_norm": 3.8940088748931885, "learning_rate": 7.807471728977327e-05, "loss": 0.3911684513092041, "memory(GiB)": 70.5, "step": 36205, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.454319 }, { "epoch": 1.551347414420976, "grad_norm": 1.2570109367370605, "learning_rate": 7.80691482790296e-05, "loss": 0.29116158485412597, "memory(GiB)": 70.5, "step": 36210, "token_acc": 0.9522388059701492, "train_speed(iter/s)": 1.454338 }, { "epoch": 1.5515616297502248, "grad_norm": 3.4537084102630615, "learning_rate": 7.806357875978698e-05, "loss": 0.5447785377502441, "memory(GiB)": 70.5, "step": 36215, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.454348 }, { "epoch": 1.5517758450794739, "grad_norm": 3.8484201431274414, "learning_rate": 7.805800873214628e-05, "loss": 0.4511068820953369, "memory(GiB)": 70.5, "step": 36220, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.454356 }, { "epoch": 1.551990060408723, "grad_norm": 2.7567474842071533, "learning_rate": 7.805243819620845e-05, "loss": 0.3805284261703491, "memory(GiB)": 70.5, "step": 36225, "token_acc": 0.9317269076305221, "train_speed(iter/s)": 1.454361 }, { "epoch": 1.5522042757379717, "grad_norm": 3.567854642868042, "learning_rate": 7.80468671520744e-05, "loss": 0.5062446117401123, "memory(GiB)": 70.5, "step": 36230, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.454372 }, { "epoch": 1.5524184910672207, "grad_norm": 4.975484848022461, "learning_rate": 7.804129559984501e-05, "loss": 0.4080010414123535, "memory(GiB)": 70.5, "step": 36235, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.454371 }, { "epoch": 1.5526327063964698, "grad_norm": 3.11242938041687, "learning_rate": 7.803572353962126e-05, "loss": 0.43349604606628417, "memory(GiB)": 70.5, "step": 36240, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.454369 }, { "epoch": 1.5528469217257186, "grad_norm": 1.9767661094665527, "learning_rate": 7.80301509715041e-05, "loss": 0.17776111364364625, "memory(GiB)": 70.5, "step": 36245, "token_acc": 0.9697986577181208, "train_speed(iter/s)": 1.454368 }, { "epoch": 1.5530611370549676, "grad_norm": 3.5211634635925293, "learning_rate": 7.802457789559443e-05, "loss": 0.41927084922790525, "memory(GiB)": 70.5, "step": 36250, "token_acc": 0.9016949152542373, "train_speed(iter/s)": 1.454399 }, { "epoch": 1.5532753523842167, "grad_norm": 4.1401214599609375, "learning_rate": 7.801900431199325e-05, "loss": 0.416552734375, "memory(GiB)": 70.5, "step": 36255, "token_acc": 0.9135802469135802, "train_speed(iter/s)": 1.454391 }, { "epoch": 1.5534895677134655, "grad_norm": 3.116847038269043, "learning_rate": 7.801343022080152e-05, "loss": 0.3938605546951294, "memory(GiB)": 70.5, "step": 36260, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.454396 }, { "epoch": 1.5537037830427145, "grad_norm": 3.276296615600586, "learning_rate": 7.800785562212025e-05, "loss": 0.3343381881713867, "memory(GiB)": 70.5, "step": 36265, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.454398 }, { "epoch": 1.5539179983719635, "grad_norm": 3.8224740028381348, "learning_rate": 7.800228051605036e-05, "loss": 0.23956751823425293, "memory(GiB)": 70.5, "step": 36270, "token_acc": 0.9442508710801394, "train_speed(iter/s)": 1.454392 }, { "epoch": 1.5541322137012124, "grad_norm": 1.5373085737228394, "learning_rate": 7.799670490269291e-05, "loss": 0.40029587745666506, "memory(GiB)": 70.5, "step": 36275, "token_acc": 0.9051094890510949, "train_speed(iter/s)": 1.454439 }, { "epoch": 1.5543464290304614, "grad_norm": 3.5768401622772217, "learning_rate": 7.799112878214889e-05, "loss": 0.3137021064758301, "memory(GiB)": 70.5, "step": 36280, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.454439 }, { "epoch": 1.5545606443597104, "grad_norm": 3.3509116172790527, "learning_rate": 7.798555215451932e-05, "loss": 0.5813887596130372, "memory(GiB)": 70.5, "step": 36285, "token_acc": 0.8816793893129771, "train_speed(iter/s)": 1.454437 }, { "epoch": 1.5547748596889592, "grad_norm": 3.4941985607147217, "learning_rate": 7.797997501990522e-05, "loss": 0.36936259269714355, "memory(GiB)": 70.5, "step": 36290, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.454453 }, { "epoch": 1.5549890750182083, "grad_norm": 2.1556644439697266, "learning_rate": 7.797439737840761e-05, "loss": 0.2528573751449585, "memory(GiB)": 70.5, "step": 36295, "token_acc": 0.9418604651162791, "train_speed(iter/s)": 1.454474 }, { "epoch": 1.5552032903474573, "grad_norm": 2.60627818107605, "learning_rate": 7.796881923012755e-05, "loss": 0.32078597545623777, "memory(GiB)": 70.5, "step": 36300, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.454478 }, { "epoch": 1.5554175056767061, "grad_norm": 1.8509081602096558, "learning_rate": 7.796324057516611e-05, "loss": 0.27310791015625, "memory(GiB)": 70.5, "step": 36305, "token_acc": 0.945859872611465, "train_speed(iter/s)": 1.454476 }, { "epoch": 1.5556317210059551, "grad_norm": 4.919260501861572, "learning_rate": 7.795766141362432e-05, "loss": 0.4655631065368652, "memory(GiB)": 70.5, "step": 36310, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.45449 }, { "epoch": 1.5558459363352042, "grad_norm": 2.4618003368377686, "learning_rate": 7.795208174560326e-05, "loss": 0.3608003377914429, "memory(GiB)": 70.5, "step": 36315, "token_acc": 0.9058441558441559, "train_speed(iter/s)": 1.454499 }, { "epoch": 1.556060151664453, "grad_norm": 1.4996848106384277, "learning_rate": 7.794650157120405e-05, "loss": 0.42133469581604005, "memory(GiB)": 70.5, "step": 36320, "token_acc": 0.9058823529411765, "train_speed(iter/s)": 1.454494 }, { "epoch": 1.556274366993702, "grad_norm": 4.420621395111084, "learning_rate": 7.794092089052774e-05, "loss": 0.43471298217773435, "memory(GiB)": 70.5, "step": 36325, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.454495 }, { "epoch": 1.556488582322951, "grad_norm": 0.7371699810028076, "learning_rate": 7.793533970367542e-05, "loss": 0.311870813369751, "memory(GiB)": 70.5, "step": 36330, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.454519 }, { "epoch": 1.5567027976521999, "grad_norm": 2.631833076477051, "learning_rate": 7.792975801074822e-05, "loss": 0.4421239852905273, "memory(GiB)": 70.5, "step": 36335, "token_acc": 0.9087947882736156, "train_speed(iter/s)": 1.454525 }, { "epoch": 1.556917012981449, "grad_norm": 0.8346519470214844, "learning_rate": 7.792417581184725e-05, "loss": 0.3035343885421753, "memory(GiB)": 70.5, "step": 36340, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.454535 }, { "epoch": 1.557131228310698, "grad_norm": 2.0706257820129395, "learning_rate": 7.791859310707366e-05, "loss": 0.43399686813354493, "memory(GiB)": 70.5, "step": 36345, "token_acc": 0.9147540983606557, "train_speed(iter/s)": 1.454552 }, { "epoch": 1.5573454436399468, "grad_norm": 3.8850839138031006, "learning_rate": 7.791300989652853e-05, "loss": 0.558219051361084, "memory(GiB)": 70.5, "step": 36350, "token_acc": 0.8923611111111112, "train_speed(iter/s)": 1.454551 }, { "epoch": 1.5575596589691958, "grad_norm": 3.017068386077881, "learning_rate": 7.790742618031307e-05, "loss": 0.2758305072784424, "memory(GiB)": 70.5, "step": 36355, "token_acc": 0.9510204081632653, "train_speed(iter/s)": 1.454556 }, { "epoch": 1.5577738742984448, "grad_norm": 1.3091943264007568, "learning_rate": 7.79018419585284e-05, "loss": 0.535820198059082, "memory(GiB)": 70.5, "step": 36360, "token_acc": 0.889261744966443, "train_speed(iter/s)": 1.454554 }, { "epoch": 1.5579880896276936, "grad_norm": 0.874298632144928, "learning_rate": 7.78962572312757e-05, "loss": 0.3848033666610718, "memory(GiB)": 70.5, "step": 36365, "token_acc": 0.9064516129032258, "train_speed(iter/s)": 1.454556 }, { "epoch": 1.5582023049569427, "grad_norm": 6.296496868133545, "learning_rate": 7.78906719986561e-05, "loss": 0.5448535919189453, "memory(GiB)": 70.5, "step": 36370, "token_acc": 0.8993288590604027, "train_speed(iter/s)": 1.454554 }, { "epoch": 1.5584165202861917, "grad_norm": 2.8890931606292725, "learning_rate": 7.788508626077084e-05, "loss": 0.41826977729797366, "memory(GiB)": 70.5, "step": 36375, "token_acc": 0.896414342629482, "train_speed(iter/s)": 1.454562 }, { "epoch": 1.5586307356154405, "grad_norm": 4.405234336853027, "learning_rate": 7.787950001772108e-05, "loss": 0.6959373950958252, "memory(GiB)": 70.5, "step": 36380, "token_acc": 0.8740740740740741, "train_speed(iter/s)": 1.454607 }, { "epoch": 1.5588449509446896, "grad_norm": 5.102669715881348, "learning_rate": 7.7873913269608e-05, "loss": 0.6946154117584229, "memory(GiB)": 70.5, "step": 36385, "token_acc": 0.8426573426573427, "train_speed(iter/s)": 1.454602 }, { "epoch": 1.5590591662739386, "grad_norm": 13.7695894241333, "learning_rate": 7.786832601653287e-05, "loss": 0.6742988586425781, "memory(GiB)": 70.5, "step": 36390, "token_acc": 0.8419452887537994, "train_speed(iter/s)": 1.454617 }, { "epoch": 1.5592733816031874, "grad_norm": 1.6646808385849, "learning_rate": 7.786273825859684e-05, "loss": 0.2737318515777588, "memory(GiB)": 70.5, "step": 36395, "token_acc": 0.9470198675496688, "train_speed(iter/s)": 1.454617 }, { "epoch": 1.5594875969324364, "grad_norm": 0.9568140506744385, "learning_rate": 7.78571499959012e-05, "loss": 0.33028061389923097, "memory(GiB)": 70.5, "step": 36400, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.454615 }, { "epoch": 1.5597018122616855, "grad_norm": 2.8860816955566406, "learning_rate": 7.785156122854713e-05, "loss": 0.4498604774475098, "memory(GiB)": 70.5, "step": 36405, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.454637 }, { "epoch": 1.5599160275909343, "grad_norm": 3.583556652069092, "learning_rate": 7.784597195663593e-05, "loss": 0.2724164485931396, "memory(GiB)": 70.5, "step": 36410, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.454639 }, { "epoch": 1.5601302429201833, "grad_norm": 5.97880744934082, "learning_rate": 7.78403821802688e-05, "loss": 0.5229710578918457, "memory(GiB)": 70.5, "step": 36415, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.454649 }, { "epoch": 1.5603444582494324, "grad_norm": 3.830686092376709, "learning_rate": 7.783479189954704e-05, "loss": 0.3746919631958008, "memory(GiB)": 70.5, "step": 36420, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.454677 }, { "epoch": 1.5605586735786812, "grad_norm": 5.155292987823486, "learning_rate": 7.78292011145719e-05, "loss": 0.6791773319244385, "memory(GiB)": 70.5, "step": 36425, "token_acc": 0.8635014836795252, "train_speed(iter/s)": 1.454682 }, { "epoch": 1.5607728889079302, "grad_norm": 2.1027987003326416, "learning_rate": 7.782360982544469e-05, "loss": 0.14515501260757446, "memory(GiB)": 70.5, "step": 36430, "token_acc": 0.967032967032967, "train_speed(iter/s)": 1.454691 }, { "epoch": 1.5609871042371792, "grad_norm": 5.071035385131836, "learning_rate": 7.781801803226669e-05, "loss": 0.5013799667358398, "memory(GiB)": 70.5, "step": 36435, "token_acc": 0.9063545150501672, "train_speed(iter/s)": 1.454701 }, { "epoch": 1.561201319566428, "grad_norm": 3.096414804458618, "learning_rate": 7.781242573513918e-05, "loss": 0.19520857334136962, "memory(GiB)": 70.5, "step": 36440, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.454716 }, { "epoch": 1.561415534895677, "grad_norm": 3.6122395992279053, "learning_rate": 7.780683293416352e-05, "loss": 0.4689614295959473, "memory(GiB)": 70.5, "step": 36445, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.454713 }, { "epoch": 1.5616297502249261, "grad_norm": 4.255952835083008, "learning_rate": 7.780123962944097e-05, "loss": 0.6169033050537109, "memory(GiB)": 70.5, "step": 36450, "token_acc": 0.8631578947368421, "train_speed(iter/s)": 1.454727 }, { "epoch": 1.561843965554175, "grad_norm": 0.5142369270324707, "learning_rate": 7.779564582107289e-05, "loss": 0.3948946237564087, "memory(GiB)": 70.5, "step": 36455, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.454746 }, { "epoch": 1.562058180883424, "grad_norm": 5.285428047180176, "learning_rate": 7.779005150916061e-05, "loss": 0.4597489833831787, "memory(GiB)": 70.5, "step": 36460, "token_acc": 0.9087719298245615, "train_speed(iter/s)": 1.45476 }, { "epoch": 1.562272396212673, "grad_norm": 2.782583713531494, "learning_rate": 7.778445669380548e-05, "loss": 0.39028382301330566, "memory(GiB)": 70.5, "step": 36465, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.454764 }, { "epoch": 1.5624866115419218, "grad_norm": 1.5237678289413452, "learning_rate": 7.777886137510885e-05, "loss": 0.3997316598892212, "memory(GiB)": 70.5, "step": 36470, "token_acc": 0.9177215189873418, "train_speed(iter/s)": 1.454756 }, { "epoch": 1.562700826871171, "grad_norm": 0.9177229404449463, "learning_rate": 7.777326555317209e-05, "loss": 0.16102582216262817, "memory(GiB)": 70.5, "step": 36475, "token_acc": 0.9531772575250836, "train_speed(iter/s)": 1.454765 }, { "epoch": 1.5629150422004199, "grad_norm": 2.8454699516296387, "learning_rate": 7.776766922809658e-05, "loss": 0.3524693489074707, "memory(GiB)": 70.5, "step": 36480, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.454761 }, { "epoch": 1.5631292575296687, "grad_norm": 0.19773553311824799, "learning_rate": 7.776207239998368e-05, "loss": 0.35891284942626955, "memory(GiB)": 70.5, "step": 36485, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.454792 }, { "epoch": 1.563343472858918, "grad_norm": 2.6666691303253174, "learning_rate": 7.775647506893481e-05, "loss": 0.21894803047180175, "memory(GiB)": 70.5, "step": 36490, "token_acc": 0.9504950495049505, "train_speed(iter/s)": 1.4548 }, { "epoch": 1.5635576881881668, "grad_norm": 4.044371128082275, "learning_rate": 7.775087723505133e-05, "loss": 0.2720160007476807, "memory(GiB)": 70.5, "step": 36495, "token_acc": 0.9554655870445344, "train_speed(iter/s)": 1.454807 }, { "epoch": 1.5637719035174156, "grad_norm": 2.738236427307129, "learning_rate": 7.774527889843471e-05, "loss": 0.155863356590271, "memory(GiB)": 70.5, "step": 36500, "token_acc": 0.9596273291925466, "train_speed(iter/s)": 1.454808 }, { "epoch": 1.5637719035174156, "eval_loss": 2.5543763637542725, "eval_runtime": 12.8001, "eval_samples_per_second": 7.812, "eval_steps_per_second": 7.812, "eval_token_acc": 0.42052023121387283, "step": 36500 }, { "epoch": 1.5639861188466648, "grad_norm": 4.420803546905518, "learning_rate": 7.773968005918631e-05, "loss": 0.3197226047515869, "memory(GiB)": 70.5, "step": 36505, "token_acc": 0.5788423153692615, "train_speed(iter/s)": 1.454006 }, { "epoch": 1.5642003341759136, "grad_norm": 3.495655059814453, "learning_rate": 7.77340807174076e-05, "loss": 0.5543100357055664, "memory(GiB)": 70.5, "step": 36510, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.453997 }, { "epoch": 1.5644145495051625, "grad_norm": 4.529973983764648, "learning_rate": 7.77284808732e-05, "loss": 0.6035037040710449, "memory(GiB)": 70.5, "step": 36515, "token_acc": 0.8835616438356164, "train_speed(iter/s)": 1.454008 }, { "epoch": 1.5646287648344117, "grad_norm": 2.9526963233947754, "learning_rate": 7.772288052666494e-05, "loss": 0.2590616464614868, "memory(GiB)": 70.5, "step": 36520, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.454015 }, { "epoch": 1.5648429801636605, "grad_norm": 0.8661471605300903, "learning_rate": 7.771727967790393e-05, "loss": 0.37381136417388916, "memory(GiB)": 70.5, "step": 36525, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.454036 }, { "epoch": 1.5650571954929093, "grad_norm": 2.957026958465576, "learning_rate": 7.771167832701835e-05, "loss": 0.4842185974121094, "memory(GiB)": 70.5, "step": 36530, "token_acc": 0.8885135135135135, "train_speed(iter/s)": 1.454033 }, { "epoch": 1.5652714108221586, "grad_norm": 4.008723258972168, "learning_rate": 7.770607647410975e-05, "loss": 0.281404447555542, "memory(GiB)": 70.5, "step": 36535, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.454028 }, { "epoch": 1.5654856261514074, "grad_norm": 2.701366662979126, "learning_rate": 7.770047411927958e-05, "loss": 0.21988861560821532, "memory(GiB)": 70.5, "step": 36540, "token_acc": 0.9529411764705882, "train_speed(iter/s)": 1.454026 }, { "epoch": 1.5656998414806562, "grad_norm": 3.523251533508301, "learning_rate": 7.769487126262934e-05, "loss": 0.42205057144165037, "memory(GiB)": 70.5, "step": 36545, "token_acc": 0.9159663865546218, "train_speed(iter/s)": 1.454026 }, { "epoch": 1.5659140568099055, "grad_norm": 3.467878580093384, "learning_rate": 7.768926790426052e-05, "loss": 0.36420636177062987, "memory(GiB)": 70.5, "step": 36550, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.454029 }, { "epoch": 1.5661282721391543, "grad_norm": 1.4716466665267944, "learning_rate": 7.768366404427464e-05, "loss": 0.32560033798217775, "memory(GiB)": 70.5, "step": 36555, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.454048 }, { "epoch": 1.566342487468403, "grad_norm": 2.346137762069702, "learning_rate": 7.767805968277322e-05, "loss": 0.33389060497283934, "memory(GiB)": 70.5, "step": 36560, "token_acc": 0.9287749287749287, "train_speed(iter/s)": 1.454056 }, { "epoch": 1.5665567027976524, "grad_norm": 2.5806424617767334, "learning_rate": 7.767245481985777e-05, "loss": 0.4736124038696289, "memory(GiB)": 70.5, "step": 36565, "token_acc": 0.8970099667774086, "train_speed(iter/s)": 1.454053 }, { "epoch": 1.5667709181269012, "grad_norm": 3.8582937717437744, "learning_rate": 7.766684945562986e-05, "loss": 0.4689964771270752, "memory(GiB)": 70.5, "step": 36570, "token_acc": 0.9147540983606557, "train_speed(iter/s)": 1.454053 }, { "epoch": 1.56698513345615, "grad_norm": 1.568804383277893, "learning_rate": 7.7661243590191e-05, "loss": 0.27380564212799074, "memory(GiB)": 70.5, "step": 36575, "token_acc": 0.934984520123839, "train_speed(iter/s)": 1.454051 }, { "epoch": 1.5671993487853992, "grad_norm": 5.017545223236084, "learning_rate": 7.765563722364278e-05, "loss": 0.5468060970306396, "memory(GiB)": 70.5, "step": 36580, "token_acc": 0.8903225806451613, "train_speed(iter/s)": 1.45405 }, { "epoch": 1.567413564114648, "grad_norm": 1.9472720623016357, "learning_rate": 7.765003035608676e-05, "loss": 0.7502296924591064, "memory(GiB)": 70.5, "step": 36585, "token_acc": 0.8401162790697675, "train_speed(iter/s)": 1.454079 }, { "epoch": 1.5676277794438969, "grad_norm": 2.047034740447998, "learning_rate": 7.764442298762448e-05, "loss": 0.2969480514526367, "memory(GiB)": 70.5, "step": 36590, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.454106 }, { "epoch": 1.5678419947731461, "grad_norm": 4.170193672180176, "learning_rate": 7.763881511835754e-05, "loss": 0.36360821723937986, "memory(GiB)": 70.5, "step": 36595, "token_acc": 0.9137380191693291, "train_speed(iter/s)": 1.454132 }, { "epoch": 1.568056210102395, "grad_norm": 3.6162683963775635, "learning_rate": 7.763320674838756e-05, "loss": 0.7616177558898926, "memory(GiB)": 70.5, "step": 36600, "token_acc": 0.839041095890411, "train_speed(iter/s)": 1.454134 }, { "epoch": 1.5682704254316437, "grad_norm": 4.194642066955566, "learning_rate": 7.76275978778161e-05, "loss": 0.3937236785888672, "memory(GiB)": 70.5, "step": 36605, "token_acc": 0.8838709677419355, "train_speed(iter/s)": 1.454135 }, { "epoch": 1.568484640760893, "grad_norm": 4.089443206787109, "learning_rate": 7.762198850674478e-05, "loss": 0.33149266242980957, "memory(GiB)": 70.5, "step": 36610, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.454118 }, { "epoch": 1.5686988560901418, "grad_norm": 2.996403694152832, "learning_rate": 7.761637863527524e-05, "loss": 0.45551071166992185, "memory(GiB)": 70.5, "step": 36615, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.454122 }, { "epoch": 1.5689130714193906, "grad_norm": 1.3688077926635742, "learning_rate": 7.761076826350911e-05, "loss": 0.18887970447540284, "memory(GiB)": 70.5, "step": 36620, "token_acc": 0.9608433734939759, "train_speed(iter/s)": 1.454126 }, { "epoch": 1.5691272867486399, "grad_norm": 3.3028149604797363, "learning_rate": 7.760515739154798e-05, "loss": 0.43480720520019533, "memory(GiB)": 70.5, "step": 36625, "token_acc": 0.9073482428115016, "train_speed(iter/s)": 1.454157 }, { "epoch": 1.5693415020778887, "grad_norm": 4.979658126831055, "learning_rate": 7.759954601949355e-05, "loss": 0.6113897323608398, "memory(GiB)": 70.5, "step": 36630, "token_acc": 0.8842975206611571, "train_speed(iter/s)": 1.454161 }, { "epoch": 1.5695557174071375, "grad_norm": 3.066324234008789, "learning_rate": 7.759393414744747e-05, "loss": 0.4681064605712891, "memory(GiB)": 70.5, "step": 36635, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.45417 }, { "epoch": 1.5697699327363868, "grad_norm": 4.284884929656982, "learning_rate": 7.758832177551136e-05, "loss": 0.3060285568237305, "memory(GiB)": 70.5, "step": 36640, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.454168 }, { "epoch": 1.5699841480656356, "grad_norm": 1.100051760673523, "learning_rate": 7.758270890378691e-05, "loss": 0.26844723224639894, "memory(GiB)": 70.5, "step": 36645, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.454162 }, { "epoch": 1.5701983633948844, "grad_norm": 2.8638267517089844, "learning_rate": 7.757709553237584e-05, "loss": 0.483125114440918, "memory(GiB)": 70.5, "step": 36650, "token_acc": 0.882943143812709, "train_speed(iter/s)": 1.454179 }, { "epoch": 1.5704125787241336, "grad_norm": 0.3367224335670471, "learning_rate": 7.757148166137981e-05, "loss": 0.12546881437301635, "memory(GiB)": 70.5, "step": 36655, "token_acc": 0.9688715953307393, "train_speed(iter/s)": 1.454183 }, { "epoch": 1.5706267940533825, "grad_norm": 1.609082818031311, "learning_rate": 7.756586729090052e-05, "loss": 0.22737441062927247, "memory(GiB)": 70.5, "step": 36660, "token_acc": 0.9577039274924471, "train_speed(iter/s)": 1.454203 }, { "epoch": 1.5708410093826313, "grad_norm": 3.175541400909424, "learning_rate": 7.756025242103969e-05, "loss": 0.42074995040893554, "memory(GiB)": 70.5, "step": 36665, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.45419 }, { "epoch": 1.5710552247118805, "grad_norm": 4.515352725982666, "learning_rate": 7.755463705189902e-05, "loss": 0.38086528778076173, "memory(GiB)": 70.5, "step": 36670, "token_acc": 0.9430604982206405, "train_speed(iter/s)": 1.454197 }, { "epoch": 1.5712694400411293, "grad_norm": 4.397727966308594, "learning_rate": 7.754902118358027e-05, "loss": 0.5132431507110595, "memory(GiB)": 70.5, "step": 36675, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.4542 }, { "epoch": 1.5714836553703782, "grad_norm": 3.5554099082946777, "learning_rate": 7.754340481618514e-05, "loss": 0.48854818344116213, "memory(GiB)": 70.5, "step": 36680, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.454225 }, { "epoch": 1.5716978706996274, "grad_norm": 0.835174024105072, "learning_rate": 7.753778794981541e-05, "loss": 0.3758122444152832, "memory(GiB)": 70.5, "step": 36685, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.454269 }, { "epoch": 1.5719120860288762, "grad_norm": 2.2143070697784424, "learning_rate": 7.753217058457281e-05, "loss": 0.6729938507080078, "memory(GiB)": 70.5, "step": 36690, "token_acc": 0.8771929824561403, "train_speed(iter/s)": 1.454291 }, { "epoch": 1.572126301358125, "grad_norm": 5.808877944946289, "learning_rate": 7.75265527205591e-05, "loss": 0.4284354686737061, "memory(GiB)": 70.5, "step": 36695, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.4543 }, { "epoch": 1.5723405166873743, "grad_norm": 2.8686156272888184, "learning_rate": 7.752093435787611e-05, "loss": 0.27477684020996096, "memory(GiB)": 70.5, "step": 36700, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.454326 }, { "epoch": 1.572554732016623, "grad_norm": 2.669867515563965, "learning_rate": 7.751531549662553e-05, "loss": 0.12873380184173583, "memory(GiB)": 70.5, "step": 36705, "token_acc": 0.9651567944250871, "train_speed(iter/s)": 1.454337 }, { "epoch": 1.572768947345872, "grad_norm": 5.601002216339111, "learning_rate": 7.750969613690923e-05, "loss": 0.46732702255249026, "memory(GiB)": 70.5, "step": 36710, "token_acc": 0.882943143812709, "train_speed(iter/s)": 1.454332 }, { "epoch": 1.5729831626751212, "grad_norm": 3.487166404724121, "learning_rate": 7.750407627882897e-05, "loss": 0.5069211006164551, "memory(GiB)": 70.5, "step": 36715, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.454347 }, { "epoch": 1.57319737800437, "grad_norm": 2.4968149662017822, "learning_rate": 7.749845592248654e-05, "loss": 0.22866981029510497, "memory(GiB)": 70.5, "step": 36720, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.454354 }, { "epoch": 1.5734115933336188, "grad_norm": 6.314023971557617, "learning_rate": 7.749283506798382e-05, "loss": 0.47907023429870604, "memory(GiB)": 70.5, "step": 36725, "token_acc": 0.8885017421602788, "train_speed(iter/s)": 1.45436 }, { "epoch": 1.573625808662868, "grad_norm": 3.5807621479034424, "learning_rate": 7.748721371542258e-05, "loss": 0.7759534358978272, "memory(GiB)": 70.5, "step": 36730, "token_acc": 0.8373702422145328, "train_speed(iter/s)": 1.454365 }, { "epoch": 1.5738400239921169, "grad_norm": 1.5508182048797607, "learning_rate": 7.748159186490469e-05, "loss": 0.6031177520751954, "memory(GiB)": 70.5, "step": 36735, "token_acc": 0.8932038834951457, "train_speed(iter/s)": 1.454393 }, { "epoch": 1.5740542393213657, "grad_norm": 5.0413079261779785, "learning_rate": 7.747596951653198e-05, "loss": 0.4092289924621582, "memory(GiB)": 70.5, "step": 36740, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.454395 }, { "epoch": 1.574268454650615, "grad_norm": 9.61226749420166, "learning_rate": 7.747034667040632e-05, "loss": 0.3099703788757324, "memory(GiB)": 70.5, "step": 36745, "token_acc": 0.9331210191082803, "train_speed(iter/s)": 1.454388 }, { "epoch": 1.5744826699798637, "grad_norm": 4.7528557777404785, "learning_rate": 7.746472332662955e-05, "loss": 0.48466243743896487, "memory(GiB)": 70.5, "step": 36750, "token_acc": 0.8905660377358491, "train_speed(iter/s)": 1.454376 }, { "epoch": 1.5746968853091126, "grad_norm": 2.3106472492218018, "learning_rate": 7.745909948530355e-05, "loss": 0.3965918779373169, "memory(GiB)": 70.5, "step": 36755, "token_acc": 0.9133858267716536, "train_speed(iter/s)": 1.45439 }, { "epoch": 1.5749111006383618, "grad_norm": 4.704562664031982, "learning_rate": 7.745347514653021e-05, "loss": 0.5643089294433594, "memory(GiB)": 70.5, "step": 36760, "token_acc": 0.8926174496644296, "train_speed(iter/s)": 1.454394 }, { "epoch": 1.5751253159676106, "grad_norm": 6.662149906158447, "learning_rate": 7.74478503104114e-05, "loss": 0.4410027027130127, "memory(GiB)": 70.5, "step": 36765, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.454403 }, { "epoch": 1.5753395312968594, "grad_norm": 1.7428761720657349, "learning_rate": 7.744222497704904e-05, "loss": 0.26643369197845457, "memory(GiB)": 70.5, "step": 36770, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.454402 }, { "epoch": 1.5755537466261087, "grad_norm": 1.1253341436386108, "learning_rate": 7.743659914654504e-05, "loss": 0.4179983615875244, "memory(GiB)": 70.5, "step": 36775, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.454449 }, { "epoch": 1.5757679619553575, "grad_norm": 3.26057505607605, "learning_rate": 7.743097281900131e-05, "loss": 0.30688393115997314, "memory(GiB)": 70.5, "step": 36780, "token_acc": 0.9316239316239316, "train_speed(iter/s)": 1.45445 }, { "epoch": 1.5759821772846063, "grad_norm": 4.82668924331665, "learning_rate": 7.742534599451978e-05, "loss": 0.4580327033996582, "memory(GiB)": 70.5, "step": 36785, "token_acc": 0.9059561128526645, "train_speed(iter/s)": 1.454451 }, { "epoch": 1.5761963926138556, "grad_norm": 8.320390701293945, "learning_rate": 7.741971867320237e-05, "loss": 0.3194576740264893, "memory(GiB)": 70.5, "step": 36790, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.454461 }, { "epoch": 1.5764106079431044, "grad_norm": 3.0592775344848633, "learning_rate": 7.741409085515103e-05, "loss": 0.4348561763763428, "memory(GiB)": 70.5, "step": 36795, "token_acc": 0.9195046439628483, "train_speed(iter/s)": 1.454484 }, { "epoch": 1.5766248232723532, "grad_norm": 5.863743305206299, "learning_rate": 7.740846254046772e-05, "loss": 0.46227273941040037, "memory(GiB)": 70.5, "step": 36800, "token_acc": 0.885558583106267, "train_speed(iter/s)": 1.45448 }, { "epoch": 1.5768390386016025, "grad_norm": 3.1722664833068848, "learning_rate": 7.74028337292544e-05, "loss": 0.3443412780761719, "memory(GiB)": 70.5, "step": 36805, "token_acc": 0.9105691056910569, "train_speed(iter/s)": 1.454504 }, { "epoch": 1.5770532539308513, "grad_norm": 2.0868642330169678, "learning_rate": 7.739720442161305e-05, "loss": 0.3294799566268921, "memory(GiB)": 70.5, "step": 36810, "token_acc": 0.9416666666666667, "train_speed(iter/s)": 1.454505 }, { "epoch": 1.5772674692601003, "grad_norm": 2.3819057941436768, "learning_rate": 7.739157461764564e-05, "loss": 0.3276965618133545, "memory(GiB)": 70.5, "step": 36815, "token_acc": 0.9249146757679181, "train_speed(iter/s)": 1.454499 }, { "epoch": 1.5774816845893493, "grad_norm": 2.577277421951294, "learning_rate": 7.738594431745415e-05, "loss": 0.4886923313140869, "memory(GiB)": 70.5, "step": 36820, "token_acc": 0.9, "train_speed(iter/s)": 1.45449 }, { "epoch": 1.5776958999185982, "grad_norm": 2.4339098930358887, "learning_rate": 7.738031352114059e-05, "loss": 0.29069504737854, "memory(GiB)": 70.5, "step": 36825, "token_acc": 0.9409722222222222, "train_speed(iter/s)": 1.454492 }, { "epoch": 1.5779101152478472, "grad_norm": 1.9424128532409668, "learning_rate": 7.737468222880697e-05, "loss": 0.3784750461578369, "memory(GiB)": 70.5, "step": 36830, "token_acc": 0.9379562043795621, "train_speed(iter/s)": 1.454505 }, { "epoch": 1.5781243305770962, "grad_norm": 2.704270601272583, "learning_rate": 7.736905044055531e-05, "loss": 0.5383882999420166, "memory(GiB)": 70.5, "step": 36835, "token_acc": 0.8810975609756098, "train_speed(iter/s)": 1.454507 }, { "epoch": 1.578338545906345, "grad_norm": 2.080890417098999, "learning_rate": 7.736341815648761e-05, "loss": 0.5548263549804687, "memory(GiB)": 70.5, "step": 36840, "token_acc": 0.8862876254180602, "train_speed(iter/s)": 1.454517 }, { "epoch": 1.578552761235594, "grad_norm": 4.558860778808594, "learning_rate": 7.735778537670594e-05, "loss": 0.23902630805969238, "memory(GiB)": 70.5, "step": 36845, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.454497 }, { "epoch": 1.578766976564843, "grad_norm": 1.6771494150161743, "learning_rate": 7.73521521013123e-05, "loss": 0.4006399631500244, "memory(GiB)": 70.5, "step": 36850, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.454502 }, { "epoch": 1.578981191894092, "grad_norm": 4.552356719970703, "learning_rate": 7.734651833040879e-05, "loss": 0.5575076103210449, "memory(GiB)": 70.5, "step": 36855, "token_acc": 0.8844765342960289, "train_speed(iter/s)": 1.454504 }, { "epoch": 1.579195407223341, "grad_norm": 0.39587974548339844, "learning_rate": 7.734088406409744e-05, "loss": 0.2517223596572876, "memory(GiB)": 70.5, "step": 36860, "token_acc": 0.9421487603305785, "train_speed(iter/s)": 1.454495 }, { "epoch": 1.57940962255259, "grad_norm": 2.313175916671753, "learning_rate": 7.733524930248032e-05, "loss": 0.37199270725250244, "memory(GiB)": 70.5, "step": 36865, "token_acc": 0.9186046511627907, "train_speed(iter/s)": 1.454488 }, { "epoch": 1.5796238378818388, "grad_norm": 5.152317523956299, "learning_rate": 7.732961404565953e-05, "loss": 0.5372009754180909, "memory(GiB)": 70.5, "step": 36870, "token_acc": 0.8875, "train_speed(iter/s)": 1.454515 }, { "epoch": 1.5798380532110878, "grad_norm": 3.4797351360321045, "learning_rate": 7.732397829373713e-05, "loss": 0.32226197719573973, "memory(GiB)": 70.5, "step": 36875, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.45451 }, { "epoch": 1.5800522685403369, "grad_norm": 5.358000755310059, "learning_rate": 7.731834204681522e-05, "loss": 0.45491971969604494, "memory(GiB)": 70.5, "step": 36880, "token_acc": 0.8908450704225352, "train_speed(iter/s)": 1.454493 }, { "epoch": 1.5802664838695857, "grad_norm": 1.3006434440612793, "learning_rate": 7.731270530499595e-05, "loss": 0.3652102708816528, "memory(GiB)": 70.5, "step": 36885, "token_acc": 0.9277456647398844, "train_speed(iter/s)": 1.45453 }, { "epoch": 1.5804806991988347, "grad_norm": 3.3118090629577637, "learning_rate": 7.73070680683814e-05, "loss": 0.28890693187713623, "memory(GiB)": 70.5, "step": 36890, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.45454 }, { "epoch": 1.5806949145280837, "grad_norm": 7.370466709136963, "learning_rate": 7.730143033707367e-05, "loss": 0.32303605079650877, "memory(GiB)": 70.5, "step": 36895, "token_acc": 0.9367088607594937, "train_speed(iter/s)": 1.454543 }, { "epoch": 1.5809091298573326, "grad_norm": 2.6381149291992188, "learning_rate": 7.729579211117492e-05, "loss": 0.7614893436431884, "memory(GiB)": 70.5, "step": 36900, "token_acc": 0.8629629629629629, "train_speed(iter/s)": 1.454546 }, { "epoch": 1.5811233451865816, "grad_norm": 5.104997634887695, "learning_rate": 7.729015339078731e-05, "loss": 0.6287915229797363, "memory(GiB)": 70.5, "step": 36905, "token_acc": 0.8643533123028391, "train_speed(iter/s)": 1.454538 }, { "epoch": 1.5813375605158306, "grad_norm": 2.2143518924713135, "learning_rate": 7.728451417601297e-05, "loss": 0.34490416049957273, "memory(GiB)": 70.5, "step": 36910, "token_acc": 0.9388489208633094, "train_speed(iter/s)": 1.454537 }, { "epoch": 1.5815517758450794, "grad_norm": 3.0949225425720215, "learning_rate": 7.727887446695405e-05, "loss": 0.38772077560424806, "memory(GiB)": 70.5, "step": 36915, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.454546 }, { "epoch": 1.5817659911743285, "grad_norm": 2.3842453956604004, "learning_rate": 7.727323426371272e-05, "loss": 0.4028726577758789, "memory(GiB)": 70.5, "step": 36920, "token_acc": 0.8968253968253969, "train_speed(iter/s)": 1.454548 }, { "epoch": 1.5819802065035775, "grad_norm": 3.6364121437072754, "learning_rate": 7.726759356639119e-05, "loss": 0.4171422004699707, "memory(GiB)": 70.5, "step": 36925, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.454555 }, { "epoch": 1.5821944218328263, "grad_norm": 1.8620144128799438, "learning_rate": 7.726195237509162e-05, "loss": 0.249276065826416, "memory(GiB)": 70.5, "step": 36930, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.454578 }, { "epoch": 1.5824086371620754, "grad_norm": 2.4810047149658203, "learning_rate": 7.725631068991618e-05, "loss": 0.20338354110717774, "memory(GiB)": 70.5, "step": 36935, "token_acc": 0.9503105590062112, "train_speed(iter/s)": 1.454583 }, { "epoch": 1.5826228524913244, "grad_norm": 1.6666045188903809, "learning_rate": 7.725066851096714e-05, "loss": 0.30616815090179444, "memory(GiB)": 70.5, "step": 36940, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.454593 }, { "epoch": 1.5828370678205732, "grad_norm": 0.5798658728599548, "learning_rate": 7.724502583834665e-05, "loss": 0.24896905422210694, "memory(GiB)": 70.5, "step": 36945, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.454599 }, { "epoch": 1.5830512831498222, "grad_norm": 3.920891761779785, "learning_rate": 7.723938267215698e-05, "loss": 0.36342692375183105, "memory(GiB)": 70.5, "step": 36950, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.454614 }, { "epoch": 1.5832654984790713, "grad_norm": 4.433283805847168, "learning_rate": 7.723373901250032e-05, "loss": 0.3037583827972412, "memory(GiB)": 70.5, "step": 36955, "token_acc": 0.9452554744525548, "train_speed(iter/s)": 1.454613 }, { "epoch": 1.58347971380832, "grad_norm": 2.499624013900757, "learning_rate": 7.722809485947895e-05, "loss": 0.1523704171180725, "memory(GiB)": 70.5, "step": 36960, "token_acc": 0.9652777777777778, "train_speed(iter/s)": 1.454659 }, { "epoch": 1.5836939291375691, "grad_norm": 2.7976315021514893, "learning_rate": 7.72224502131951e-05, "loss": 0.26560053825378416, "memory(GiB)": 70.5, "step": 36965, "token_acc": 0.9375, "train_speed(iter/s)": 1.454668 }, { "epoch": 1.5839081444668182, "grad_norm": 3.803255796432495, "learning_rate": 7.721680507375102e-05, "loss": 0.2537947654724121, "memory(GiB)": 70.5, "step": 36970, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.45468 }, { "epoch": 1.584122359796067, "grad_norm": 2.0332295894622803, "learning_rate": 7.721115944124897e-05, "loss": 0.4416079521179199, "memory(GiB)": 70.5, "step": 36975, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.454695 }, { "epoch": 1.584336575125316, "grad_norm": 5.381776332855225, "learning_rate": 7.720551331579126e-05, "loss": 0.6436289310455322, "memory(GiB)": 70.5, "step": 36980, "token_acc": 0.8707482993197279, "train_speed(iter/s)": 1.454703 }, { "epoch": 1.584550790454565, "grad_norm": 1.9788399934768677, "learning_rate": 7.719986669748013e-05, "loss": 0.2776284694671631, "memory(GiB)": 70.5, "step": 36985, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.454708 }, { "epoch": 1.5847650057838139, "grad_norm": 1.8548954725265503, "learning_rate": 7.719421958641794e-05, "loss": 0.4675287246704102, "memory(GiB)": 70.5, "step": 36990, "token_acc": 0.8961937716262975, "train_speed(iter/s)": 1.454716 }, { "epoch": 1.5849792211130629, "grad_norm": 2.8586103916168213, "learning_rate": 7.718857198270692e-05, "loss": 0.16915388107299806, "memory(GiB)": 70.5, "step": 36995, "token_acc": 0.9692307692307692, "train_speed(iter/s)": 1.454722 }, { "epoch": 1.585193436442312, "grad_norm": 0.18709133565425873, "learning_rate": 7.718292388644943e-05, "loss": 0.21957578659057617, "memory(GiB)": 70.5, "step": 37000, "token_acc": 0.9442508710801394, "train_speed(iter/s)": 1.454718 }, { "epoch": 1.585193436442312, "eval_loss": 2.5093507766723633, "eval_runtime": 13.8314, "eval_samples_per_second": 7.23, "eval_steps_per_second": 7.23, "eval_token_acc": 0.4493333333333333, "step": 37000 }, { "epoch": 1.5854076517715607, "grad_norm": 2.6122567653656006, "learning_rate": 7.717727529774777e-05, "loss": 0.2959511518478394, "memory(GiB)": 70.5, "step": 37005, "token_acc": 0.5853423336547734, "train_speed(iter/s)": 1.453879 }, { "epoch": 1.5856218671008098, "grad_norm": 3.06193470954895, "learning_rate": 7.717162621670427e-05, "loss": 0.5425867080688477, "memory(GiB)": 70.5, "step": 37010, "token_acc": 0.8745519713261649, "train_speed(iter/s)": 1.453884 }, { "epoch": 1.5858360824300588, "grad_norm": 2.145775318145752, "learning_rate": 7.716597664342127e-05, "loss": 0.37511434555053713, "memory(GiB)": 70.5, "step": 37015, "token_acc": 0.9194029850746268, "train_speed(iter/s)": 1.45389 }, { "epoch": 1.5860502977593076, "grad_norm": 3.736370325088501, "learning_rate": 7.716032657800113e-05, "loss": 0.3960052251815796, "memory(GiB)": 70.5, "step": 37020, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.453924 }, { "epoch": 1.5862645130885566, "grad_norm": 4.411503791809082, "learning_rate": 7.715467602054618e-05, "loss": 0.5068892955780029, "memory(GiB)": 70.5, "step": 37025, "token_acc": 0.9100719424460432, "train_speed(iter/s)": 1.453947 }, { "epoch": 1.5864787284178057, "grad_norm": 2.059856414794922, "learning_rate": 7.714902497115881e-05, "loss": 0.14202333688735963, "memory(GiB)": 70.5, "step": 37030, "token_acc": 0.9733333333333334, "train_speed(iter/s)": 1.453964 }, { "epoch": 1.5866929437470545, "grad_norm": 5.056396961212158, "learning_rate": 7.714337342994139e-05, "loss": 0.23669066429138183, "memory(GiB)": 70.5, "step": 37035, "token_acc": 0.9546827794561934, "train_speed(iter/s)": 1.453992 }, { "epoch": 1.5869071590763035, "grad_norm": 4.887988090515137, "learning_rate": 7.71377213969963e-05, "loss": 0.37900686264038086, "memory(GiB)": 70.5, "step": 37040, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.45401 }, { "epoch": 1.5871213744055526, "grad_norm": 6.060805320739746, "learning_rate": 7.713206887242592e-05, "loss": 0.36185245513916015, "memory(GiB)": 70.5, "step": 37045, "token_acc": 0.9289940828402367, "train_speed(iter/s)": 1.454006 }, { "epoch": 1.5873355897348014, "grad_norm": 2.791309118270874, "learning_rate": 7.712641585633265e-05, "loss": 0.267950177192688, "memory(GiB)": 70.5, "step": 37050, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.454008 }, { "epoch": 1.5875498050640504, "grad_norm": 3.1183907985687256, "learning_rate": 7.712076234881893e-05, "loss": 0.176956844329834, "memory(GiB)": 70.5, "step": 37055, "token_acc": 0.9559322033898305, "train_speed(iter/s)": 1.454012 }, { "epoch": 1.5877640203932994, "grad_norm": 4.314916610717773, "learning_rate": 7.711510834998714e-05, "loss": 0.34834673404693606, "memory(GiB)": 70.5, "step": 37060, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.454007 }, { "epoch": 1.5879782357225483, "grad_norm": 2.7228939533233643, "learning_rate": 7.710945385993975e-05, "loss": 0.32642488479614257, "memory(GiB)": 70.5, "step": 37065, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.454016 }, { "epoch": 1.5881924510517973, "grad_norm": 1.983752965927124, "learning_rate": 7.710379887877917e-05, "loss": 0.4358389854431152, "memory(GiB)": 70.5, "step": 37070, "token_acc": 0.9096774193548387, "train_speed(iter/s)": 1.454054 }, { "epoch": 1.5884066663810463, "grad_norm": 2.470871925354004, "learning_rate": 7.709814340660784e-05, "loss": 0.3816145181655884, "memory(GiB)": 70.5, "step": 37075, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.454056 }, { "epoch": 1.5886208817102951, "grad_norm": 2.412447452545166, "learning_rate": 7.709248744352822e-05, "loss": 0.22291176319122313, "memory(GiB)": 70.5, "step": 37080, "token_acc": 0.9531914893617022, "train_speed(iter/s)": 1.454054 }, { "epoch": 1.5888350970395442, "grad_norm": 4.94849967956543, "learning_rate": 7.708683098964275e-05, "loss": 0.4905003547668457, "memory(GiB)": 70.5, "step": 37085, "token_acc": 0.9240924092409241, "train_speed(iter/s)": 1.454074 }, { "epoch": 1.5890493123687932, "grad_norm": 3.166104793548584, "learning_rate": 7.708117404505397e-05, "loss": 0.33607795238494875, "memory(GiB)": 70.5, "step": 37090, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.454081 }, { "epoch": 1.589263527698042, "grad_norm": 0.1753639280796051, "learning_rate": 7.707551660986429e-05, "loss": 0.5423746585845948, "memory(GiB)": 70.5, "step": 37095, "token_acc": 0.89419795221843, "train_speed(iter/s)": 1.454102 }, { "epoch": 1.589477743027291, "grad_norm": 10.513298034667969, "learning_rate": 7.706985868417624e-05, "loss": 0.277066707611084, "memory(GiB)": 70.5, "step": 37100, "token_acc": 0.9407894736842105, "train_speed(iter/s)": 1.454127 }, { "epoch": 1.58969195835654, "grad_norm": 4.092499256134033, "learning_rate": 7.706420026809232e-05, "loss": 0.4056910514831543, "memory(GiB)": 70.5, "step": 37105, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.454125 }, { "epoch": 1.589906173685789, "grad_norm": 3.348649263381958, "learning_rate": 7.7058541361715e-05, "loss": 0.367364239692688, "memory(GiB)": 70.5, "step": 37110, "token_acc": 0.9409937888198758, "train_speed(iter/s)": 1.454125 }, { "epoch": 1.590120389015038, "grad_norm": 1.697424054145813, "learning_rate": 7.705288196514682e-05, "loss": 0.33192944526672363, "memory(GiB)": 70.5, "step": 37115, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.454135 }, { "epoch": 1.590334604344287, "grad_norm": 2.2756450176239014, "learning_rate": 7.70472220784903e-05, "loss": 0.26935040950775146, "memory(GiB)": 70.5, "step": 37120, "token_acc": 0.939297124600639, "train_speed(iter/s)": 1.454118 }, { "epoch": 1.5905488196735358, "grad_norm": 6.034268856048584, "learning_rate": 7.704156170184801e-05, "loss": 0.22858881950378418, "memory(GiB)": 70.5, "step": 37125, "token_acc": 0.9623287671232876, "train_speed(iter/s)": 1.454114 }, { "epoch": 1.5907630350027848, "grad_norm": 0.08892310410737991, "learning_rate": 7.703590083532244e-05, "loss": 0.1931756019592285, "memory(GiB)": 70.5, "step": 37130, "token_acc": 0.9602649006622517, "train_speed(iter/s)": 1.454116 }, { "epoch": 1.5909772503320339, "grad_norm": 2.9974923133850098, "learning_rate": 7.703023947901618e-05, "loss": 0.4670853137969971, "memory(GiB)": 70.5, "step": 37135, "token_acc": 0.8768115942028986, "train_speed(iter/s)": 1.454127 }, { "epoch": 1.5911914656612827, "grad_norm": 5.887843608856201, "learning_rate": 7.702457763303177e-05, "loss": 0.4195298671722412, "memory(GiB)": 70.5, "step": 37140, "token_acc": 0.9314641744548287, "train_speed(iter/s)": 1.454146 }, { "epoch": 1.5914056809905317, "grad_norm": 1.0789650678634644, "learning_rate": 7.701891529747178e-05, "loss": 0.348801326751709, "memory(GiB)": 70.5, "step": 37145, "token_acc": 0.9140127388535032, "train_speed(iter/s)": 1.454176 }, { "epoch": 1.5916198963197807, "grad_norm": 2.101348876953125, "learning_rate": 7.70132524724388e-05, "loss": 0.5561594486236572, "memory(GiB)": 70.5, "step": 37150, "token_acc": 0.8914473684210527, "train_speed(iter/s)": 1.454183 }, { "epoch": 1.5918341116490295, "grad_norm": 3.0121443271636963, "learning_rate": 7.70075891580354e-05, "loss": 0.46001472473144533, "memory(GiB)": 70.5, "step": 37155, "token_acc": 0.8993288590604027, "train_speed(iter/s)": 1.454206 }, { "epoch": 1.5920483269782786, "grad_norm": 7.537271022796631, "learning_rate": 7.70019253543642e-05, "loss": 0.4933182716369629, "memory(GiB)": 70.5, "step": 37160, "token_acc": 0.8901515151515151, "train_speed(iter/s)": 1.454203 }, { "epoch": 1.5922625423075276, "grad_norm": 4.990788459777832, "learning_rate": 7.699626106152778e-05, "loss": 0.5775738716125488, "memory(GiB)": 70.5, "step": 37165, "token_acc": 0.8731343283582089, "train_speed(iter/s)": 1.45422 }, { "epoch": 1.5924767576367764, "grad_norm": 3.4840259552001953, "learning_rate": 7.699059627962877e-05, "loss": 0.29656119346618653, "memory(GiB)": 70.5, "step": 37170, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.454219 }, { "epoch": 1.5926909729660255, "grad_norm": 5.0967631340026855, "learning_rate": 7.698493100876979e-05, "loss": 0.27429900169372556, "memory(GiB)": 70.5, "step": 37175, "token_acc": 0.932806324110672, "train_speed(iter/s)": 1.45424 }, { "epoch": 1.5929051882952745, "grad_norm": 5.554698467254639, "learning_rate": 7.697926524905348e-05, "loss": 0.4832024097442627, "memory(GiB)": 70.5, "step": 37180, "token_acc": 0.9042553191489362, "train_speed(iter/s)": 1.454261 }, { "epoch": 1.5931194036245233, "grad_norm": 4.62334680557251, "learning_rate": 7.697359900058245e-05, "loss": 0.4683672904968262, "memory(GiB)": 70.5, "step": 37185, "token_acc": 0.9053627760252366, "train_speed(iter/s)": 1.454251 }, { "epoch": 1.5933336189537723, "grad_norm": 1.493914246559143, "learning_rate": 7.696793226345939e-05, "loss": 0.33725364208221437, "memory(GiB)": 70.5, "step": 37190, "token_acc": 0.9400630914826499, "train_speed(iter/s)": 1.454255 }, { "epoch": 1.5935478342830214, "grad_norm": 4.738956451416016, "learning_rate": 7.696226503778694e-05, "loss": 0.3871596097946167, "memory(GiB)": 70.5, "step": 37195, "token_acc": 0.9051094890510949, "train_speed(iter/s)": 1.454256 }, { "epoch": 1.5937620496122702, "grad_norm": 4.814446926116943, "learning_rate": 7.695659732366774e-05, "loss": 0.38423259258270265, "memory(GiB)": 70.5, "step": 37200, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.454247 }, { "epoch": 1.5939762649415192, "grad_norm": 2.7234976291656494, "learning_rate": 7.695092912120452e-05, "loss": 0.37540245056152344, "memory(GiB)": 70.5, "step": 37205, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.454239 }, { "epoch": 1.5941904802707683, "grad_norm": 0.10289990156888962, "learning_rate": 7.694526043049995e-05, "loss": 0.27220566272735597, "memory(GiB)": 70.5, "step": 37210, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.454244 }, { "epoch": 1.594404695600017, "grad_norm": 2.6588332653045654, "learning_rate": 7.693959125165666e-05, "loss": 0.3108238458633423, "memory(GiB)": 70.5, "step": 37215, "token_acc": 0.9492753623188406, "train_speed(iter/s)": 1.454246 }, { "epoch": 1.594618910929266, "grad_norm": 5.351898193359375, "learning_rate": 7.693392158477745e-05, "loss": 0.43889904022216797, "memory(GiB)": 70.5, "step": 37220, "token_acc": 0.9208860759493671, "train_speed(iter/s)": 1.454257 }, { "epoch": 1.5948331262585151, "grad_norm": 2.3978075981140137, "learning_rate": 7.692825142996498e-05, "loss": 0.45203776359558107, "memory(GiB)": 70.5, "step": 37225, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.454259 }, { "epoch": 1.595047341587764, "grad_norm": 3.464595079421997, "learning_rate": 7.692258078732196e-05, "loss": 0.5638110160827636, "memory(GiB)": 70.5, "step": 37230, "token_acc": 0.8742138364779874, "train_speed(iter/s)": 1.454266 }, { "epoch": 1.595261556917013, "grad_norm": 5.012094020843506, "learning_rate": 7.69169096569511e-05, "loss": 0.22778966426849365, "memory(GiB)": 70.5, "step": 37235, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.454264 }, { "epoch": 1.595475772246262, "grad_norm": 2.4361305236816406, "learning_rate": 7.691123803895523e-05, "loss": 0.39712700843811033, "memory(GiB)": 70.5, "step": 37240, "token_acc": 0.9262820512820513, "train_speed(iter/s)": 1.454269 }, { "epoch": 1.5956899875755108, "grad_norm": 4.067086219787598, "learning_rate": 7.6905565933437e-05, "loss": 0.5266332149505615, "memory(GiB)": 70.5, "step": 37245, "token_acc": 0.8910505836575876, "train_speed(iter/s)": 1.45427 }, { "epoch": 1.5959042029047599, "grad_norm": 1.1261216402053833, "learning_rate": 7.689989334049923e-05, "loss": 0.32628684043884276, "memory(GiB)": 70.5, "step": 37250, "token_acc": 0.934156378600823, "train_speed(iter/s)": 1.454299 }, { "epoch": 1.596118418234009, "grad_norm": 4.933315753936768, "learning_rate": 7.689422026024464e-05, "loss": 0.40663795471191405, "memory(GiB)": 70.5, "step": 37255, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.454305 }, { "epoch": 1.5963326335632577, "grad_norm": 2.1603317260742188, "learning_rate": 7.688854669277604e-05, "loss": 0.3903364896774292, "memory(GiB)": 70.5, "step": 37260, "token_acc": 0.9191176470588235, "train_speed(iter/s)": 1.45433 }, { "epoch": 1.5965468488925068, "grad_norm": 4.590950012207031, "learning_rate": 7.688287263819617e-05, "loss": 0.5727517604827881, "memory(GiB)": 70.5, "step": 37265, "token_acc": 0.8938356164383562, "train_speed(iter/s)": 1.454363 }, { "epoch": 1.5967610642217558, "grad_norm": 3.752255439758301, "learning_rate": 7.687719809660785e-05, "loss": 0.5431558609008789, "memory(GiB)": 70.5, "step": 37270, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.454354 }, { "epoch": 1.5969752795510046, "grad_norm": 3.580660343170166, "learning_rate": 7.687152306811388e-05, "loss": 0.48426337242126466, "memory(GiB)": 70.5, "step": 37275, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.454345 }, { "epoch": 1.5971894948802536, "grad_norm": 8.552793502807617, "learning_rate": 7.686584755281708e-05, "loss": 0.3983913421630859, "memory(GiB)": 70.5, "step": 37280, "token_acc": 0.9186046511627907, "train_speed(iter/s)": 1.454346 }, { "epoch": 1.5974037102095027, "grad_norm": 3.9192192554473877, "learning_rate": 7.686017155082021e-05, "loss": 0.44617486000061035, "memory(GiB)": 70.5, "step": 37285, "token_acc": 0.8885017421602788, "train_speed(iter/s)": 1.454343 }, { "epoch": 1.5976179255387515, "grad_norm": 2.3547403812408447, "learning_rate": 7.685449506222616e-05, "loss": 0.32305049896240234, "memory(GiB)": 70.5, "step": 37290, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.454341 }, { "epoch": 1.5978321408680005, "grad_norm": 0.7786230444908142, "learning_rate": 7.684881808713774e-05, "loss": 0.28272984027862547, "memory(GiB)": 70.5, "step": 37295, "token_acc": 0.9507042253521126, "train_speed(iter/s)": 1.454347 }, { "epoch": 1.5980463561972496, "grad_norm": 6.493716716766357, "learning_rate": 7.684314062565779e-05, "loss": 0.41569061279296876, "memory(GiB)": 70.5, "step": 37300, "token_acc": 0.9129032258064517, "train_speed(iter/s)": 1.454352 }, { "epoch": 1.5982605715264984, "grad_norm": 3.0482430458068848, "learning_rate": 7.683746267788916e-05, "loss": 0.2121281623840332, "memory(GiB)": 70.5, "step": 37305, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.454358 }, { "epoch": 1.5984747868557474, "grad_norm": 4.170345306396484, "learning_rate": 7.683178424393472e-05, "loss": 0.4609241962432861, "memory(GiB)": 70.5, "step": 37310, "token_acc": 0.91015625, "train_speed(iter/s)": 1.454362 }, { "epoch": 1.5986890021849964, "grad_norm": 6.205911159515381, "learning_rate": 7.682610532389734e-05, "loss": 0.343643045425415, "memory(GiB)": 70.5, "step": 37315, "token_acc": 0.9455128205128205, "train_speed(iter/s)": 1.454398 }, { "epoch": 1.5989032175142452, "grad_norm": 5.0939178466796875, "learning_rate": 7.68204259178799e-05, "loss": 0.6013216018676758, "memory(GiB)": 70.5, "step": 37320, "token_acc": 0.8506944444444444, "train_speed(iter/s)": 1.454416 }, { "epoch": 1.5991174328434943, "grad_norm": 3.3394062519073486, "learning_rate": 7.681474602598529e-05, "loss": 0.6271971702575684, "memory(GiB)": 70.5, "step": 37325, "token_acc": 0.8923076923076924, "train_speed(iter/s)": 1.454425 }, { "epoch": 1.5993316481727433, "grad_norm": 2.4549131393432617, "learning_rate": 7.68090656483164e-05, "loss": 0.4400663375854492, "memory(GiB)": 70.5, "step": 37330, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.45443 }, { "epoch": 1.5995458635019921, "grad_norm": 3.73154878616333, "learning_rate": 7.680338478497613e-05, "loss": 0.29880218505859374, "memory(GiB)": 70.5, "step": 37335, "token_acc": 0.9294478527607362, "train_speed(iter/s)": 1.454427 }, { "epoch": 1.5997600788312412, "grad_norm": 5.750152111053467, "learning_rate": 7.679770343606741e-05, "loss": 0.35526208877563475, "memory(GiB)": 70.5, "step": 37340, "token_acc": 0.9233576642335767, "train_speed(iter/s)": 1.45443 }, { "epoch": 1.5999742941604902, "grad_norm": 4.6616435050964355, "learning_rate": 7.679202160169314e-05, "loss": 0.4279648780822754, "memory(GiB)": 70.5, "step": 37345, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.454437 }, { "epoch": 1.600188509489739, "grad_norm": 4.141043663024902, "learning_rate": 7.678633928195628e-05, "loss": 0.39908905029296876, "memory(GiB)": 70.5, "step": 37350, "token_acc": 0.9316770186335404, "train_speed(iter/s)": 1.454439 }, { "epoch": 1.600402724818988, "grad_norm": 5.180839538574219, "learning_rate": 7.678065647695975e-05, "loss": 0.24632973670959474, "memory(GiB)": 70.5, "step": 37355, "token_acc": 0.9592476489028213, "train_speed(iter/s)": 1.454446 }, { "epoch": 1.600616940148237, "grad_norm": 6.219354629516602, "learning_rate": 7.67749731868065e-05, "loss": 0.27548387050628664, "memory(GiB)": 70.5, "step": 37360, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.454452 }, { "epoch": 1.600831155477486, "grad_norm": 1.9620840549468994, "learning_rate": 7.676928941159951e-05, "loss": 0.1831890821456909, "memory(GiB)": 70.5, "step": 37365, "token_acc": 0.9739413680781759, "train_speed(iter/s)": 1.454451 }, { "epoch": 1.601045370806735, "grad_norm": 5.080473899841309, "learning_rate": 7.676360515144172e-05, "loss": 0.44533228874206543, "memory(GiB)": 70.5, "step": 37370, "token_acc": 0.9181286549707602, "train_speed(iter/s)": 1.454457 }, { "epoch": 1.601259586135984, "grad_norm": 8.502824783325195, "learning_rate": 7.675792040643611e-05, "loss": 0.4102786064147949, "memory(GiB)": 70.5, "step": 37375, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.454469 }, { "epoch": 1.6014738014652328, "grad_norm": 3.133580446243286, "learning_rate": 7.675223517668569e-05, "loss": 0.26208689212799074, "memory(GiB)": 70.5, "step": 37380, "token_acc": 0.9533333333333334, "train_speed(iter/s)": 1.454492 }, { "epoch": 1.6016880167944818, "grad_norm": 0.38771048188209534, "learning_rate": 7.67465494622934e-05, "loss": 0.338036036491394, "memory(GiB)": 70.5, "step": 37385, "token_acc": 0.9161290322580645, "train_speed(iter/s)": 1.454487 }, { "epoch": 1.6019022321237308, "grad_norm": 7.811864376068115, "learning_rate": 7.67408632633623e-05, "loss": 0.5765791893005371, "memory(GiB)": 70.5, "step": 37390, "token_acc": 0.89419795221843, "train_speed(iter/s)": 1.4545 }, { "epoch": 1.6021164474529797, "grad_norm": 0.1598006933927536, "learning_rate": 7.673517657999538e-05, "loss": 0.3441760540008545, "memory(GiB)": 70.5, "step": 37395, "token_acc": 0.9418604651162791, "train_speed(iter/s)": 1.454516 }, { "epoch": 1.6023306627822287, "grad_norm": 5.082267761230469, "learning_rate": 7.672948941229565e-05, "loss": 0.3922961950302124, "memory(GiB)": 70.5, "step": 37400, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.454517 }, { "epoch": 1.6025448781114777, "grad_norm": 5.969120502471924, "learning_rate": 7.672380176036615e-05, "loss": 0.3394594430923462, "memory(GiB)": 70.5, "step": 37405, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.454529 }, { "epoch": 1.6027590934407265, "grad_norm": 9.667455673217773, "learning_rate": 7.671811362430992e-05, "loss": 0.8676422119140625, "memory(GiB)": 70.5, "step": 37410, "token_acc": 0.8, "train_speed(iter/s)": 1.454565 }, { "epoch": 1.6029733087699756, "grad_norm": 4.427768707275391, "learning_rate": 7.671242500422998e-05, "loss": 0.48247947692871096, "memory(GiB)": 70.5, "step": 37415, "token_acc": 0.8968481375358166, "train_speed(iter/s)": 1.454582 }, { "epoch": 1.6031875240992246, "grad_norm": 1.385407567024231, "learning_rate": 7.670673590022939e-05, "loss": 0.3051556825637817, "memory(GiB)": 70.5, "step": 37420, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.454597 }, { "epoch": 1.6034017394284734, "grad_norm": 3.5645432472229004, "learning_rate": 7.670104631241126e-05, "loss": 0.3544075727462769, "memory(GiB)": 70.5, "step": 37425, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.4546 }, { "epoch": 1.6036159547577225, "grad_norm": 0.09369220584630966, "learning_rate": 7.66953562408786e-05, "loss": 0.22742824554443358, "memory(GiB)": 70.5, "step": 37430, "token_acc": 0.94140625, "train_speed(iter/s)": 1.454595 }, { "epoch": 1.6038301700869715, "grad_norm": 3.3523478507995605, "learning_rate": 7.668966568573455e-05, "loss": 0.44794559478759766, "memory(GiB)": 70.5, "step": 37435, "token_acc": 0.9161073825503355, "train_speed(iter/s)": 1.454594 }, { "epoch": 1.6040443854162203, "grad_norm": 5.034569263458252, "learning_rate": 7.668397464708214e-05, "loss": 0.40445985794067385, "memory(GiB)": 70.5, "step": 37440, "token_acc": 0.9502762430939227, "train_speed(iter/s)": 1.454602 }, { "epoch": 1.6042586007454693, "grad_norm": 6.4519243240356445, "learning_rate": 7.667828312502452e-05, "loss": 0.6108265876770019, "memory(GiB)": 70.5, "step": 37445, "token_acc": 0.8765957446808511, "train_speed(iter/s)": 1.454612 }, { "epoch": 1.6044728160747184, "grad_norm": 3.4254605770111084, "learning_rate": 7.667259111966476e-05, "loss": 0.3674298286437988, "memory(GiB)": 70.5, "step": 37450, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.454624 }, { "epoch": 1.6046870314039672, "grad_norm": 1.9440044164657593, "learning_rate": 7.6666898631106e-05, "loss": 0.7334197044372559, "memory(GiB)": 70.5, "step": 37455, "token_acc": 0.8544891640866873, "train_speed(iter/s)": 1.454634 }, { "epoch": 1.6049012467332162, "grad_norm": 5.247393608093262, "learning_rate": 7.666120565945135e-05, "loss": 0.521531629562378, "memory(GiB)": 70.5, "step": 37460, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.454643 }, { "epoch": 1.6051154620624652, "grad_norm": 2.6927545070648193, "learning_rate": 7.665551220480395e-05, "loss": 0.20566399097442628, "memory(GiB)": 70.5, "step": 37465, "token_acc": 0.9603960396039604, "train_speed(iter/s)": 1.454645 }, { "epoch": 1.605329677391714, "grad_norm": 4.854351997375488, "learning_rate": 7.664981826726695e-05, "loss": 0.41524429321289064, "memory(GiB)": 70.5, "step": 37470, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.454634 }, { "epoch": 1.605543892720963, "grad_norm": 0.1138262078166008, "learning_rate": 7.664412384694348e-05, "loss": 0.37047591209411623, "memory(GiB)": 70.5, "step": 37475, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.454631 }, { "epoch": 1.6057581080502121, "grad_norm": 4.108814239501953, "learning_rate": 7.663842894393672e-05, "loss": 0.5383580207824707, "memory(GiB)": 70.5, "step": 37480, "token_acc": 0.8833922261484098, "train_speed(iter/s)": 1.454623 }, { "epoch": 1.605972323379461, "grad_norm": 3.2910053730010986, "learning_rate": 7.663273355834984e-05, "loss": 0.27167391777038574, "memory(GiB)": 70.5, "step": 37485, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.45462 }, { "epoch": 1.60618653870871, "grad_norm": 3.665292263031006, "learning_rate": 7.662703769028599e-05, "loss": 0.350124454498291, "memory(GiB)": 70.5, "step": 37490, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.454638 }, { "epoch": 1.606400754037959, "grad_norm": 3.441387891769409, "learning_rate": 7.662134133984838e-05, "loss": 0.4553953170776367, "memory(GiB)": 70.5, "step": 37495, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.454658 }, { "epoch": 1.6066149693672078, "grad_norm": 1.9772539138793945, "learning_rate": 7.66156445071402e-05, "loss": 0.28888707160949706, "memory(GiB)": 70.5, "step": 37500, "token_acc": 0.9427480916030534, "train_speed(iter/s)": 1.454664 }, { "epoch": 1.6066149693672078, "eval_loss": 2.6213738918304443, "eval_runtime": 13.2501, "eval_samples_per_second": 7.547, "eval_steps_per_second": 7.547, "eval_token_acc": 0.4352112676056338, "step": 37500 }, { "epoch": 1.6068291846964569, "grad_norm": 3.7117514610290527, "learning_rate": 7.660994719226464e-05, "loss": 0.14478888511657714, "memory(GiB)": 70.5, "step": 37505, "token_acc": 0.5845070422535211, "train_speed(iter/s)": 1.45387 }, { "epoch": 1.607043400025706, "grad_norm": 7.208720684051514, "learning_rate": 7.660424939532494e-05, "loss": 0.5574752807617187, "memory(GiB)": 70.5, "step": 37510, "token_acc": 0.8847457627118644, "train_speed(iter/s)": 1.453872 }, { "epoch": 1.6072576153549547, "grad_norm": 2.2574856281280518, "learning_rate": 7.65985511164243e-05, "loss": 0.5612183570861816, "memory(GiB)": 70.5, "step": 37515, "token_acc": 0.8984962406015038, "train_speed(iter/s)": 1.453885 }, { "epoch": 1.6074718306842037, "grad_norm": 3.9117438793182373, "learning_rate": 7.659285235566596e-05, "loss": 0.41473846435546874, "memory(GiB)": 70.5, "step": 37520, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.453913 }, { "epoch": 1.6076860460134528, "grad_norm": 5.201493263244629, "learning_rate": 7.658715311315314e-05, "loss": 0.4456812381744385, "memory(GiB)": 70.5, "step": 37525, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.453955 }, { "epoch": 1.6079002613427016, "grad_norm": 3.9496724605560303, "learning_rate": 7.658145338898912e-05, "loss": 0.34386818408966063, "memory(GiB)": 70.5, "step": 37530, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.453964 }, { "epoch": 1.6081144766719506, "grad_norm": 3.0138895511627197, "learning_rate": 7.657575318327712e-05, "loss": 0.2972557067871094, "memory(GiB)": 70.5, "step": 37535, "token_acc": 0.9356223175965666, "train_speed(iter/s)": 1.453967 }, { "epoch": 1.6083286920011997, "grad_norm": 3.788080930709839, "learning_rate": 7.657005249612044e-05, "loss": 0.4375565528869629, "memory(GiB)": 70.5, "step": 37540, "token_acc": 0.9053254437869822, "train_speed(iter/s)": 1.453976 }, { "epoch": 1.6085429073304485, "grad_norm": 1.9006835222244263, "learning_rate": 7.65643513276223e-05, "loss": 0.3102485418319702, "memory(GiB)": 70.5, "step": 37545, "token_acc": 0.9138461538461539, "train_speed(iter/s)": 1.453978 }, { "epoch": 1.6087571226596975, "grad_norm": 0.6651418209075928, "learning_rate": 7.655864967788605e-05, "loss": 0.4542214870452881, "memory(GiB)": 70.5, "step": 37550, "token_acc": 0.9019073569482289, "train_speed(iter/s)": 1.453971 }, { "epoch": 1.6089713379889465, "grad_norm": 11.028507232666016, "learning_rate": 7.655294754701494e-05, "loss": 0.40111703872680665, "memory(GiB)": 70.5, "step": 37555, "token_acc": 0.9073033707865169, "train_speed(iter/s)": 1.453975 }, { "epoch": 1.6091855533181953, "grad_norm": 5.319486618041992, "learning_rate": 7.654724493511227e-05, "loss": 0.4236593246459961, "memory(GiB)": 70.5, "step": 37560, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.453971 }, { "epoch": 1.6093997686474444, "grad_norm": 1.9395896196365356, "learning_rate": 7.654154184228137e-05, "loss": 0.2471756935119629, "memory(GiB)": 70.5, "step": 37565, "token_acc": 0.9554655870445344, "train_speed(iter/s)": 1.453972 }, { "epoch": 1.6096139839766934, "grad_norm": 3.991408109664917, "learning_rate": 7.653583826862552e-05, "loss": 0.8429314613342285, "memory(GiB)": 70.5, "step": 37570, "token_acc": 0.8159340659340659, "train_speed(iter/s)": 1.45398 }, { "epoch": 1.6098281993059422, "grad_norm": 2.3975439071655273, "learning_rate": 7.653013421424806e-05, "loss": 0.2887016773223877, "memory(GiB)": 70.5, "step": 37575, "token_acc": 0.9323308270676691, "train_speed(iter/s)": 1.453976 }, { "epoch": 1.6100424146351913, "grad_norm": 4.1680707931518555, "learning_rate": 7.652442967925236e-05, "loss": 0.38659398555755614, "memory(GiB)": 70.5, "step": 37580, "token_acc": 0.9137254901960784, "train_speed(iter/s)": 1.453961 }, { "epoch": 1.6102566299644403, "grad_norm": 3.391021966934204, "learning_rate": 7.651872466374172e-05, "loss": 0.31133222579956055, "memory(GiB)": 70.5, "step": 37585, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.453964 }, { "epoch": 1.6104708452936891, "grad_norm": 1.452420949935913, "learning_rate": 7.65130191678195e-05, "loss": 0.27758288383483887, "memory(GiB)": 70.5, "step": 37590, "token_acc": 0.9298245614035088, "train_speed(iter/s)": 1.453956 }, { "epoch": 1.6106850606229381, "grad_norm": 2.9681694507598877, "learning_rate": 7.650731319158908e-05, "loss": 0.23826954364776612, "memory(GiB)": 70.5, "step": 37595, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.453953 }, { "epoch": 1.6108992759521872, "grad_norm": 3.7478830814361572, "learning_rate": 7.650160673515381e-05, "loss": 0.35993428230285646, "memory(GiB)": 70.5, "step": 37600, "token_acc": 0.9043478260869565, "train_speed(iter/s)": 1.453969 }, { "epoch": 1.611113491281436, "grad_norm": 1.8884690999984741, "learning_rate": 7.649589979861706e-05, "loss": 0.3835011005401611, "memory(GiB)": 70.5, "step": 37605, "token_acc": 0.908284023668639, "train_speed(iter/s)": 1.453965 }, { "epoch": 1.611327706610685, "grad_norm": 0.15771430730819702, "learning_rate": 7.649019238208226e-05, "loss": 0.5206642150878906, "memory(GiB)": 70.5, "step": 37610, "token_acc": 0.8802228412256268, "train_speed(iter/s)": 1.453984 }, { "epoch": 1.611541921939934, "grad_norm": 2.891481876373291, "learning_rate": 7.648448448565275e-05, "loss": 0.3461515188217163, "memory(GiB)": 70.5, "step": 37615, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.453991 }, { "epoch": 1.6117561372691829, "grad_norm": 2.160290479660034, "learning_rate": 7.647877610943197e-05, "loss": 0.4433790683746338, "memory(GiB)": 70.5, "step": 37620, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.454001 }, { "epoch": 1.611970352598432, "grad_norm": 4.0464677810668945, "learning_rate": 7.647306725352332e-05, "loss": 0.7809590816497802, "memory(GiB)": 70.5, "step": 37625, "token_acc": 0.8501529051987767, "train_speed(iter/s)": 1.454002 }, { "epoch": 1.612184567927681, "grad_norm": 5.028346538543701, "learning_rate": 7.646735791803024e-05, "loss": 0.4439197540283203, "memory(GiB)": 70.5, "step": 37630, "token_acc": 0.9105691056910569, "train_speed(iter/s)": 1.454042 }, { "epoch": 1.6123987832569298, "grad_norm": 0.2830718159675598, "learning_rate": 7.646164810305611e-05, "loss": 0.24535493850708007, "memory(GiB)": 70.5, "step": 37635, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.454063 }, { "epoch": 1.6126129985861788, "grad_norm": 4.4128098487854, "learning_rate": 7.645593780870442e-05, "loss": 0.2245401620864868, "memory(GiB)": 70.5, "step": 37640, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.454063 }, { "epoch": 1.6128272139154278, "grad_norm": 1.551724910736084, "learning_rate": 7.645022703507858e-05, "loss": 0.2268974542617798, "memory(GiB)": 70.5, "step": 37645, "token_acc": 0.9587628865979382, "train_speed(iter/s)": 1.454071 }, { "epoch": 1.6130414292446766, "grad_norm": 2.3517956733703613, "learning_rate": 7.64445157822821e-05, "loss": 0.3146965026855469, "memory(GiB)": 70.5, "step": 37650, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.454063 }, { "epoch": 1.6132556445739257, "grad_norm": 3.532388925552368, "learning_rate": 7.643880405041838e-05, "loss": 0.576822566986084, "memory(GiB)": 70.5, "step": 37655, "token_acc": 0.8976897689768977, "train_speed(iter/s)": 1.454067 }, { "epoch": 1.6134698599031747, "grad_norm": 2.6905925273895264, "learning_rate": 7.643309183959094e-05, "loss": 0.3704824924468994, "memory(GiB)": 70.5, "step": 37660, "token_acc": 0.9148264984227129, "train_speed(iter/s)": 1.454072 }, { "epoch": 1.6136840752324235, "grad_norm": 2.3528013229370117, "learning_rate": 7.642737914990324e-05, "loss": 0.2914103984832764, "memory(GiB)": 70.5, "step": 37665, "token_acc": 0.9498432601880877, "train_speed(iter/s)": 1.454075 }, { "epoch": 1.6138982905616726, "grad_norm": 4.881906986236572, "learning_rate": 7.642166598145876e-05, "loss": 0.4254015922546387, "memory(GiB)": 70.5, "step": 37670, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.454095 }, { "epoch": 1.6141125058909216, "grad_norm": 4.1083760261535645, "learning_rate": 7.641595233436104e-05, "loss": 0.5240877628326416, "memory(GiB)": 70.5, "step": 37675, "token_acc": 0.8620689655172413, "train_speed(iter/s)": 1.454093 }, { "epoch": 1.6143267212201704, "grad_norm": 2.2683796882629395, "learning_rate": 7.641023820871353e-05, "loss": 0.36909098625183107, "memory(GiB)": 70.5, "step": 37680, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.454103 }, { "epoch": 1.6145409365494194, "grad_norm": 5.482154846191406, "learning_rate": 7.640452360461982e-05, "loss": 0.2244722843170166, "memory(GiB)": 70.5, "step": 37685, "token_acc": 0.9567567567567568, "train_speed(iter/s)": 1.454094 }, { "epoch": 1.6147551518786685, "grad_norm": 4.067728519439697, "learning_rate": 7.639880852218338e-05, "loss": 0.7357173442840577, "memory(GiB)": 70.5, "step": 37690, "token_acc": 0.8786885245901639, "train_speed(iter/s)": 1.45409 }, { "epoch": 1.6149693672079173, "grad_norm": 0.21969729661941528, "learning_rate": 7.639309296150775e-05, "loss": 0.2428257703781128, "memory(GiB)": 70.5, "step": 37695, "token_acc": 0.9435215946843853, "train_speed(iter/s)": 1.454098 }, { "epoch": 1.6151835825371663, "grad_norm": 2.1986746788024902, "learning_rate": 7.638737692269649e-05, "loss": 0.331666374206543, "memory(GiB)": 70.5, "step": 37700, "token_acc": 0.9251968503937008, "train_speed(iter/s)": 1.454103 }, { "epoch": 1.6153977978664154, "grad_norm": 4.583387851715088, "learning_rate": 7.638166040585314e-05, "loss": 0.3511821746826172, "memory(GiB)": 70.5, "step": 37705, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.454085 }, { "epoch": 1.6156120131956642, "grad_norm": 4.342747688293457, "learning_rate": 7.637594341108127e-05, "loss": 0.36246724128723146, "memory(GiB)": 70.5, "step": 37710, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.454073 }, { "epoch": 1.6158262285249132, "grad_norm": 3.4778549671173096, "learning_rate": 7.637022593848444e-05, "loss": 0.25449771881103517, "memory(GiB)": 70.5, "step": 37715, "token_acc": 0.9528619528619529, "train_speed(iter/s)": 1.454077 }, { "epoch": 1.6160404438541622, "grad_norm": 1.70010244846344, "learning_rate": 7.636450798816624e-05, "loss": 0.8022019386291503, "memory(GiB)": 70.5, "step": 37720, "token_acc": 0.823170731707317, "train_speed(iter/s)": 1.454084 }, { "epoch": 1.616254659183411, "grad_norm": 3.1738128662109375, "learning_rate": 7.635878956023023e-05, "loss": 0.4699068546295166, "memory(GiB)": 70.5, "step": 37725, "token_acc": 0.9023746701846965, "train_speed(iter/s)": 1.45409 }, { "epoch": 1.61646887451266, "grad_norm": 3.142544984817505, "learning_rate": 7.635307065478003e-05, "loss": 0.6921768665313721, "memory(GiB)": 70.5, "step": 37730, "token_acc": 0.89, "train_speed(iter/s)": 1.454089 }, { "epoch": 1.6166830898419091, "grad_norm": 3.6165707111358643, "learning_rate": 7.634735127191922e-05, "loss": 0.2899024963378906, "memory(GiB)": 70.5, "step": 37735, "token_acc": 0.937984496124031, "train_speed(iter/s)": 1.454086 }, { "epoch": 1.616897305171158, "grad_norm": 2.9155468940734863, "learning_rate": 7.634163141175144e-05, "loss": 0.44092841148376466, "memory(GiB)": 70.5, "step": 37740, "token_acc": 0.9206896551724137, "train_speed(iter/s)": 1.454101 }, { "epoch": 1.617111520500407, "grad_norm": 4.961300373077393, "learning_rate": 7.633591107438029e-05, "loss": 0.5979938983917237, "memory(GiB)": 70.5, "step": 37745, "token_acc": 0.8586206896551725, "train_speed(iter/s)": 1.454129 }, { "epoch": 1.617325735829656, "grad_norm": 2.0657103061676025, "learning_rate": 7.63301902599094e-05, "loss": 0.33704380989074706, "memory(GiB)": 70.5, "step": 37750, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.454129 }, { "epoch": 1.6175399511589048, "grad_norm": 3.163822650909424, "learning_rate": 7.632446896844243e-05, "loss": 0.452362585067749, "memory(GiB)": 70.5, "step": 37755, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.454149 }, { "epoch": 1.6177541664881538, "grad_norm": 0.16533279418945312, "learning_rate": 7.631874720008301e-05, "loss": 0.28867924213409424, "memory(GiB)": 70.5, "step": 37760, "token_acc": 0.9380530973451328, "train_speed(iter/s)": 1.454157 }, { "epoch": 1.6179683818174029, "grad_norm": 4.522668838500977, "learning_rate": 7.631302495493478e-05, "loss": 0.7350180149078369, "memory(GiB)": 70.5, "step": 37765, "token_acc": 0.8671328671328671, "train_speed(iter/s)": 1.454188 }, { "epoch": 1.6181825971466517, "grad_norm": 2.720524311065674, "learning_rate": 7.630730223310143e-05, "loss": 0.48704843521118163, "memory(GiB)": 70.5, "step": 37770, "token_acc": 0.8709677419354839, "train_speed(iter/s)": 1.454187 }, { "epoch": 1.6183968124759007, "grad_norm": 19.10087013244629, "learning_rate": 7.630157903468663e-05, "loss": 0.4147953510284424, "memory(GiB)": 70.5, "step": 37775, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.454185 }, { "epoch": 1.6186110278051498, "grad_norm": 2.6473727226257324, "learning_rate": 7.629585535979402e-05, "loss": 0.3544682741165161, "memory(GiB)": 70.5, "step": 37780, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.454178 }, { "epoch": 1.6188252431343986, "grad_norm": 5.14130973815918, "learning_rate": 7.629013120852736e-05, "loss": 0.3449560642242432, "memory(GiB)": 70.5, "step": 37785, "token_acc": 0.9402390438247012, "train_speed(iter/s)": 1.454189 }, { "epoch": 1.6190394584636476, "grad_norm": 0.42839938402175903, "learning_rate": 7.628440658099029e-05, "loss": 0.3175839900970459, "memory(GiB)": 70.5, "step": 37790, "token_acc": 0.9317269076305221, "train_speed(iter/s)": 1.454184 }, { "epoch": 1.6192536737928966, "grad_norm": 4.656126976013184, "learning_rate": 7.627868147728654e-05, "loss": 0.355401611328125, "memory(GiB)": 70.5, "step": 37795, "token_acc": 0.9133858267716536, "train_speed(iter/s)": 1.454187 }, { "epoch": 1.6194678891221455, "grad_norm": 3.5066802501678467, "learning_rate": 7.627295589751982e-05, "loss": 0.32614750862121583, "memory(GiB)": 70.5, "step": 37800, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.454194 }, { "epoch": 1.6196821044513945, "grad_norm": 3.358826160430908, "learning_rate": 7.626722984179387e-05, "loss": 0.6405954837799073, "memory(GiB)": 70.5, "step": 37805, "token_acc": 0.8653061224489796, "train_speed(iter/s)": 1.454193 }, { "epoch": 1.6198963197806435, "grad_norm": 3.051621913909912, "learning_rate": 7.62615033102124e-05, "loss": 0.42200260162353515, "memory(GiB)": 70.5, "step": 37810, "token_acc": 0.89272030651341, "train_speed(iter/s)": 1.45423 }, { "epoch": 1.6201105351098923, "grad_norm": 7.306730270385742, "learning_rate": 7.625577630287918e-05, "loss": 0.5019565105438233, "memory(GiB)": 70.5, "step": 37815, "token_acc": 0.9006622516556292, "train_speed(iter/s)": 1.454239 }, { "epoch": 1.6203247504391414, "grad_norm": 4.179375648498535, "learning_rate": 7.625004881989792e-05, "loss": 0.35704717636108396, "memory(GiB)": 70.5, "step": 37820, "token_acc": 0.9282868525896414, "train_speed(iter/s)": 1.45426 }, { "epoch": 1.6205389657683904, "grad_norm": 5.639708042144775, "learning_rate": 7.624432086137241e-05, "loss": 0.2504983901977539, "memory(GiB)": 70.5, "step": 37825, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.454254 }, { "epoch": 1.6207531810976392, "grad_norm": 0.4896070957183838, "learning_rate": 7.623859242740642e-05, "loss": 0.20821313858032225, "memory(GiB)": 70.5, "step": 37830, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.454275 }, { "epoch": 1.6209673964268885, "grad_norm": 4.1158294677734375, "learning_rate": 7.62328635181037e-05, "loss": 0.6273265838623047, "memory(GiB)": 70.5, "step": 37835, "token_acc": 0.8664383561643836, "train_speed(iter/s)": 1.454285 }, { "epoch": 1.6211816117561373, "grad_norm": 3.9534482955932617, "learning_rate": 7.622713413356806e-05, "loss": 0.3820387601852417, "memory(GiB)": 70.5, "step": 37840, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.454278 }, { "epoch": 1.621395827085386, "grad_norm": 2.231717586517334, "learning_rate": 7.622140427390327e-05, "loss": 0.12245986461639405, "memory(GiB)": 70.5, "step": 37845, "token_acc": 0.963265306122449, "train_speed(iter/s)": 1.45428 }, { "epoch": 1.6216100424146354, "grad_norm": 6.779424667358398, "learning_rate": 7.621567393921315e-05, "loss": 0.49419193267822265, "memory(GiB)": 70.5, "step": 37850, "token_acc": 0.89568345323741, "train_speed(iter/s)": 1.454311 }, { "epoch": 1.6218242577438842, "grad_norm": 3.227741003036499, "learning_rate": 7.62099431296015e-05, "loss": 0.5980219364166259, "memory(GiB)": 70.5, "step": 37855, "token_acc": 0.8757961783439491, "train_speed(iter/s)": 1.454349 }, { "epoch": 1.622038473073133, "grad_norm": 4.2890801429748535, "learning_rate": 7.620421184517216e-05, "loss": 0.37161931991577146, "memory(GiB)": 70.5, "step": 37860, "token_acc": 0.9126506024096386, "train_speed(iter/s)": 1.454343 }, { "epoch": 1.6222526884023822, "grad_norm": 3.4675405025482178, "learning_rate": 7.61984800860289e-05, "loss": 0.33632922172546387, "memory(GiB)": 70.5, "step": 37865, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.454343 }, { "epoch": 1.622466903731631, "grad_norm": 3.4323666095733643, "learning_rate": 7.619274785227564e-05, "loss": 0.660981559753418, "memory(GiB)": 70.5, "step": 37870, "token_acc": 0.8844765342960289, "train_speed(iter/s)": 1.45435 }, { "epoch": 1.6226811190608799, "grad_norm": 3.46372652053833, "learning_rate": 7.618701514401618e-05, "loss": 0.3125059366226196, "memory(GiB)": 70.5, "step": 37875, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.454352 }, { "epoch": 1.6228953343901291, "grad_norm": 1.7139395475387573, "learning_rate": 7.618128196135435e-05, "loss": 0.4035337448120117, "memory(GiB)": 70.5, "step": 37880, "token_acc": 0.916923076923077, "train_speed(iter/s)": 1.454352 }, { "epoch": 1.623109549719378, "grad_norm": 7.536229610443115, "learning_rate": 7.617554830439406e-05, "loss": 0.46647844314575193, "memory(GiB)": 70.5, "step": 37885, "token_acc": 0.9136690647482014, "train_speed(iter/s)": 1.454347 }, { "epoch": 1.6233237650486267, "grad_norm": 2.5002620220184326, "learning_rate": 7.616981417323914e-05, "loss": 0.3683753490447998, "memory(GiB)": 70.5, "step": 37890, "token_acc": 0.9126984126984127, "train_speed(iter/s)": 1.454332 }, { "epoch": 1.623537980377876, "grad_norm": 0.1116921454668045, "learning_rate": 7.61640795679935e-05, "loss": 0.1374591112136841, "memory(GiB)": 70.5, "step": 37895, "token_acc": 0.972027972027972, "train_speed(iter/s)": 1.454327 }, { "epoch": 1.6237521957071248, "grad_norm": 5.157870769500732, "learning_rate": 7.6158344488761e-05, "loss": 0.574320650100708, "memory(GiB)": 70.5, "step": 37900, "token_acc": 0.8788732394366198, "train_speed(iter/s)": 1.454357 }, { "epoch": 1.6239664110363736, "grad_norm": 4.687005519866943, "learning_rate": 7.615260893564556e-05, "loss": 0.24017701148986817, "memory(GiB)": 70.5, "step": 37905, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.454383 }, { "epoch": 1.6241806263656229, "grad_norm": 5.2685418128967285, "learning_rate": 7.614687290875107e-05, "loss": 0.4989461898803711, "memory(GiB)": 70.5, "step": 37910, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.45438 }, { "epoch": 1.6243948416948717, "grad_norm": 0.45512935519218445, "learning_rate": 7.614113640818145e-05, "loss": 0.5065241813659668, "memory(GiB)": 70.5, "step": 37915, "token_acc": 0.889589905362776, "train_speed(iter/s)": 1.454401 }, { "epoch": 1.6246090570241205, "grad_norm": 2.1794214248657227, "learning_rate": 7.613539943404064e-05, "loss": 0.3726815223693848, "memory(GiB)": 70.5, "step": 37920, "token_acc": 0.8884892086330936, "train_speed(iter/s)": 1.454437 }, { "epoch": 1.6248232723533698, "grad_norm": 1.9052610397338867, "learning_rate": 7.612966198643254e-05, "loss": 0.23051273822784424, "memory(GiB)": 70.5, "step": 37925, "token_acc": 0.9476923076923077, "train_speed(iter/s)": 1.454425 }, { "epoch": 1.6250374876826186, "grad_norm": 4.108259677886963, "learning_rate": 7.612392406546109e-05, "loss": 0.3285454273223877, "memory(GiB)": 70.5, "step": 37930, "token_acc": 0.909433962264151, "train_speed(iter/s)": 1.454419 }, { "epoch": 1.6252517030118674, "grad_norm": 5.586935997009277, "learning_rate": 7.611818567123025e-05, "loss": 0.6497061729431153, "memory(GiB)": 70.5, "step": 37935, "token_acc": 0.8975155279503105, "train_speed(iter/s)": 1.454425 }, { "epoch": 1.6254659183411166, "grad_norm": 1.969165563583374, "learning_rate": 7.6112446803844e-05, "loss": 0.5042243480682373, "memory(GiB)": 70.5, "step": 37940, "token_acc": 0.8945686900958466, "train_speed(iter/s)": 1.454422 }, { "epoch": 1.6256801336703655, "grad_norm": 7.697597980499268, "learning_rate": 7.610670746340626e-05, "loss": 0.6519233703613281, "memory(GiB)": 70.5, "step": 37945, "token_acc": 0.8868613138686131, "train_speed(iter/s)": 1.454428 }, { "epoch": 1.6258943489996143, "grad_norm": 4.097588062286377, "learning_rate": 7.610096765002104e-05, "loss": 0.536610460281372, "memory(GiB)": 70.5, "step": 37950, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.454421 }, { "epoch": 1.6261085643288635, "grad_norm": 2.0949721336364746, "learning_rate": 7.609522736379229e-05, "loss": 0.18149760961532593, "memory(GiB)": 70.5, "step": 37955, "token_acc": 0.9521739130434783, "train_speed(iter/s)": 1.454452 }, { "epoch": 1.6263227796581123, "grad_norm": 1.2974863052368164, "learning_rate": 7.608948660482403e-05, "loss": 0.09434347748756408, "memory(GiB)": 70.5, "step": 37960, "token_acc": 0.9758064516129032, "train_speed(iter/s)": 1.454451 }, { "epoch": 1.6265369949873612, "grad_norm": 1.9426285028457642, "learning_rate": 7.608374537322024e-05, "loss": 0.4842409133911133, "memory(GiB)": 70.5, "step": 37965, "token_acc": 0.893687707641196, "train_speed(iter/s)": 1.454475 }, { "epoch": 1.6267512103166104, "grad_norm": 3.198206663131714, "learning_rate": 7.607800366908493e-05, "loss": 0.25907316207885744, "memory(GiB)": 70.5, "step": 37970, "token_acc": 0.9336099585062241, "train_speed(iter/s)": 1.454487 }, { "epoch": 1.6269654256458592, "grad_norm": 3.8946285247802734, "learning_rate": 7.607226149252212e-05, "loss": 0.6782419681549072, "memory(GiB)": 70.5, "step": 37975, "token_acc": 0.8771331058020477, "train_speed(iter/s)": 1.454495 }, { "epoch": 1.627179640975108, "grad_norm": 8.331330299377441, "learning_rate": 7.606651884363585e-05, "loss": 0.48900723457336426, "memory(GiB)": 70.5, "step": 37980, "token_acc": 0.909433962264151, "train_speed(iter/s)": 1.454499 }, { "epoch": 1.6273938563043573, "grad_norm": 5.281151294708252, "learning_rate": 7.606077572253012e-05, "loss": 0.3269219875335693, "memory(GiB)": 70.5, "step": 37985, "token_acc": 0.9291666666666667, "train_speed(iter/s)": 1.454501 }, { "epoch": 1.627608071633606, "grad_norm": 5.23884391784668, "learning_rate": 7.6055032129309e-05, "loss": 0.4917795658111572, "memory(GiB)": 70.5, "step": 37990, "token_acc": 0.9033149171270718, "train_speed(iter/s)": 1.454508 }, { "epoch": 1.627822286962855, "grad_norm": 5.504274845123291, "learning_rate": 7.604928806407654e-05, "loss": 0.6063360691070556, "memory(GiB)": 70.5, "step": 37995, "token_acc": 0.8724137931034482, "train_speed(iter/s)": 1.454512 }, { "epoch": 1.6280365022921042, "grad_norm": 1.2238829135894775, "learning_rate": 7.604354352693677e-05, "loss": 0.26202094554901123, "memory(GiB)": 70.5, "step": 38000, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.454512 }, { "epoch": 1.6280365022921042, "eval_loss": 2.618666172027588, "eval_runtime": 13.8293, "eval_samples_per_second": 7.231, "eval_steps_per_second": 7.231, "eval_token_acc": 0.46808510638297873, "step": 38000 }, { "epoch": 1.628250717621353, "grad_norm": 3.164980173110962, "learning_rate": 7.60377985179938e-05, "loss": 0.818605613708496, "memory(GiB)": 70.5, "step": 38005, "token_acc": 0.5677233429394812, "train_speed(iter/s)": 1.453687 }, { "epoch": 1.6284649329506018, "grad_norm": 2.563304901123047, "learning_rate": 7.603205303735166e-05, "loss": 0.27171764373779295, "memory(GiB)": 70.5, "step": 38010, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.453702 }, { "epoch": 1.628679148279851, "grad_norm": 5.925216197967529, "learning_rate": 7.602630708511448e-05, "loss": 0.5393668174743652, "memory(GiB)": 70.5, "step": 38015, "token_acc": 0.8911764705882353, "train_speed(iter/s)": 1.453725 }, { "epoch": 1.6288933636090999, "grad_norm": 3.6248176097869873, "learning_rate": 7.602056066138633e-05, "loss": 0.38050169944763185, "memory(GiB)": 70.5, "step": 38020, "token_acc": 0.9246861924686193, "train_speed(iter/s)": 1.453741 }, { "epoch": 1.6291075789383487, "grad_norm": 1.3170783519744873, "learning_rate": 7.601481376627131e-05, "loss": 0.217291522026062, "memory(GiB)": 70.5, "step": 38025, "token_acc": 0.9620689655172414, "train_speed(iter/s)": 1.453745 }, { "epoch": 1.629321794267598, "grad_norm": 4.660367965698242, "learning_rate": 7.600906639987352e-05, "loss": 0.5121158599853516, "memory(GiB)": 70.5, "step": 38030, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.453742 }, { "epoch": 1.6295360095968467, "grad_norm": 3.4207565784454346, "learning_rate": 7.600331856229712e-05, "loss": 0.17249406576156617, "memory(GiB)": 70.5, "step": 38035, "token_acc": 0.9609375, "train_speed(iter/s)": 1.453752 }, { "epoch": 1.6297502249260956, "grad_norm": 3.5987019538879395, "learning_rate": 7.59975702536462e-05, "loss": 0.2819300889968872, "memory(GiB)": 70.5, "step": 38040, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.453767 }, { "epoch": 1.6299644402553448, "grad_norm": 3.6765236854553223, "learning_rate": 7.599182147402491e-05, "loss": 0.593369436264038, "memory(GiB)": 70.5, "step": 38045, "token_acc": 0.8963210702341137, "train_speed(iter/s)": 1.453773 }, { "epoch": 1.6301786555845936, "grad_norm": 3.010119915008545, "learning_rate": 7.598607222353739e-05, "loss": 0.3234145402908325, "memory(GiB)": 70.5, "step": 38050, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.453779 }, { "epoch": 1.6303928709138424, "grad_norm": 3.272035837173462, "learning_rate": 7.598032250228779e-05, "loss": 0.2442007303237915, "memory(GiB)": 70.5, "step": 38055, "token_acc": 0.9439252336448598, "train_speed(iter/s)": 1.453784 }, { "epoch": 1.6306070862430917, "grad_norm": 2.553511619567871, "learning_rate": 7.597457231038028e-05, "loss": 0.18855559825897217, "memory(GiB)": 70.5, "step": 38060, "token_acc": 0.973568281938326, "train_speed(iter/s)": 1.45379 }, { "epoch": 1.6308213015723405, "grad_norm": 0.6616986393928528, "learning_rate": 7.596882164791903e-05, "loss": 0.18458636999130248, "memory(GiB)": 70.5, "step": 38065, "token_acc": 0.9669421487603306, "train_speed(iter/s)": 1.453793 }, { "epoch": 1.6310355169015893, "grad_norm": 3.753936767578125, "learning_rate": 7.596307051500821e-05, "loss": 0.47066287994384765, "memory(GiB)": 70.5, "step": 38070, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.4538 }, { "epoch": 1.6312497322308386, "grad_norm": 5.636535167694092, "learning_rate": 7.595731891175202e-05, "loss": 0.4342973232269287, "memory(GiB)": 70.5, "step": 38075, "token_acc": 0.92, "train_speed(iter/s)": 1.453794 }, { "epoch": 1.6314639475600874, "grad_norm": 3.2227559089660645, "learning_rate": 7.595156683825463e-05, "loss": 0.4038394927978516, "memory(GiB)": 70.5, "step": 38080, "token_acc": 0.9158878504672897, "train_speed(iter/s)": 1.453789 }, { "epoch": 1.6316781628893362, "grad_norm": 1.570601224899292, "learning_rate": 7.594581429462026e-05, "loss": 0.5166860103607178, "memory(GiB)": 70.5, "step": 38085, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.453801 }, { "epoch": 1.6318923782185855, "grad_norm": 2.6652562618255615, "learning_rate": 7.594006128095314e-05, "loss": 0.44013190269470215, "memory(GiB)": 70.5, "step": 38090, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.453833 }, { "epoch": 1.6321065935478343, "grad_norm": 4.828022003173828, "learning_rate": 7.593430779735749e-05, "loss": 0.4906179904937744, "memory(GiB)": 70.5, "step": 38095, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.453835 }, { "epoch": 1.632320808877083, "grad_norm": 2.2190043926239014, "learning_rate": 7.592855384393752e-05, "loss": 0.2931745290756226, "memory(GiB)": 70.5, "step": 38100, "token_acc": 0.9453125, "train_speed(iter/s)": 1.45383 }, { "epoch": 1.6325350242063323, "grad_norm": 2.9868898391723633, "learning_rate": 7.592279942079746e-05, "loss": 0.48761844635009766, "memory(GiB)": 70.5, "step": 38105, "token_acc": 0.9169811320754717, "train_speed(iter/s)": 1.453844 }, { "epoch": 1.6327492395355812, "grad_norm": 1.0777318477630615, "learning_rate": 7.591704452804157e-05, "loss": 0.3800934314727783, "memory(GiB)": 70.5, "step": 38110, "token_acc": 0.9307958477508651, "train_speed(iter/s)": 1.45384 }, { "epoch": 1.63296345486483, "grad_norm": 3.789748430252075, "learning_rate": 7.591128916577412e-05, "loss": 0.2737541198730469, "memory(GiB)": 70.5, "step": 38115, "token_acc": 0.9442379182156134, "train_speed(iter/s)": 1.453849 }, { "epoch": 1.6331776701940792, "grad_norm": 2.705845832824707, "learning_rate": 7.590553333409934e-05, "loss": 0.3847815990447998, "memory(GiB)": 70.5, "step": 38120, "token_acc": 0.8915254237288136, "train_speed(iter/s)": 1.453848 }, { "epoch": 1.633391885523328, "grad_norm": 8.308585166931152, "learning_rate": 7.589977703312152e-05, "loss": 0.3806962490081787, "memory(GiB)": 70.5, "step": 38125, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.453874 }, { "epoch": 1.6336061008525768, "grad_norm": 3.8529138565063477, "learning_rate": 7.589402026294497e-05, "loss": 0.4840544700622559, "memory(GiB)": 70.5, "step": 38130, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.453908 }, { "epoch": 1.633820316181826, "grad_norm": 3.728288412094116, "learning_rate": 7.588826302367394e-05, "loss": 0.45669145584106446, "memory(GiB)": 70.5, "step": 38135, "token_acc": 0.889589905362776, "train_speed(iter/s)": 1.45391 }, { "epoch": 1.634034531511075, "grad_norm": 0.3206925094127655, "learning_rate": 7.588250531541274e-05, "loss": 0.673977279663086, "memory(GiB)": 70.5, "step": 38140, "token_acc": 0.8651685393258427, "train_speed(iter/s)": 1.45391 }, { "epoch": 1.6342487468403237, "grad_norm": 2.3020901679992676, "learning_rate": 7.587674713826567e-05, "loss": 0.2534902811050415, "memory(GiB)": 70.5, "step": 38145, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.453908 }, { "epoch": 1.634462962169573, "grad_norm": 2.828505754470825, "learning_rate": 7.587098849233707e-05, "loss": 0.7244701385498047, "memory(GiB)": 70.5, "step": 38150, "token_acc": 0.8703703703703703, "train_speed(iter/s)": 1.453927 }, { "epoch": 1.6346771774988218, "grad_norm": 3.583791732788086, "learning_rate": 7.586522937773123e-05, "loss": 0.35256237983703614, "memory(GiB)": 70.5, "step": 38155, "token_acc": 0.9184397163120568, "train_speed(iter/s)": 1.453922 }, { "epoch": 1.6348913928280706, "grad_norm": 4.189933776855469, "learning_rate": 7.58594697945525e-05, "loss": 0.41320204734802246, "memory(GiB)": 70.5, "step": 38160, "token_acc": 0.9076433121019108, "train_speed(iter/s)": 1.453922 }, { "epoch": 1.6351056081573199, "grad_norm": 2.0813405513763428, "learning_rate": 7.585370974290521e-05, "loss": 0.6524993419647217, "memory(GiB)": 70.5, "step": 38165, "token_acc": 0.8664383561643836, "train_speed(iter/s)": 1.453918 }, { "epoch": 1.6353198234865687, "grad_norm": 2.727020502090454, "learning_rate": 7.584794922289371e-05, "loss": 0.38351917266845703, "memory(GiB)": 70.5, "step": 38170, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.453921 }, { "epoch": 1.6355340388158177, "grad_norm": 2.9053800106048584, "learning_rate": 7.584218823462238e-05, "loss": 0.43259191513061523, "memory(GiB)": 70.5, "step": 38175, "token_acc": 0.8933333333333333, "train_speed(iter/s)": 1.453908 }, { "epoch": 1.6357482541450667, "grad_norm": 3.180591106414795, "learning_rate": 7.583642677819557e-05, "loss": 0.6998558044433594, "memory(GiB)": 70.5, "step": 38180, "token_acc": 0.8384615384615385, "train_speed(iter/s)": 1.453907 }, { "epoch": 1.6359624694743156, "grad_norm": 2.307499885559082, "learning_rate": 7.583066485371764e-05, "loss": 0.4517496585845947, "memory(GiB)": 70.5, "step": 38185, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.453912 }, { "epoch": 1.6361766848035646, "grad_norm": 6.725872993469238, "learning_rate": 7.582490246129299e-05, "loss": 0.5836194515228271, "memory(GiB)": 70.5, "step": 38190, "token_acc": 0.8597122302158273, "train_speed(iter/s)": 1.453953 }, { "epoch": 1.6363909001328136, "grad_norm": 2.1687161922454834, "learning_rate": 7.5819139601026e-05, "loss": 0.4989126205444336, "memory(GiB)": 70.5, "step": 38195, "token_acc": 0.8804347826086957, "train_speed(iter/s)": 1.453955 }, { "epoch": 1.6366051154620624, "grad_norm": 1.3478420972824097, "learning_rate": 7.581337627302107e-05, "loss": 0.36559062004089354, "memory(GiB)": 70.5, "step": 38200, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.453971 }, { "epoch": 1.6368193307913115, "grad_norm": 1.8576080799102783, "learning_rate": 7.580761247738264e-05, "loss": 0.342420768737793, "memory(GiB)": 70.5, "step": 38205, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.453985 }, { "epoch": 1.6370335461205605, "grad_norm": 1.905734658241272, "learning_rate": 7.580184821421508e-05, "loss": 0.3409430980682373, "memory(GiB)": 70.5, "step": 38210, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.453979 }, { "epoch": 1.6372477614498093, "grad_norm": 2.799816131591797, "learning_rate": 7.579608348362284e-05, "loss": 0.2954440116882324, "memory(GiB)": 70.5, "step": 38215, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.453979 }, { "epoch": 1.6374619767790584, "grad_norm": 2.233968496322632, "learning_rate": 7.579031828571035e-05, "loss": 0.5467352390289306, "memory(GiB)": 70.5, "step": 38220, "token_acc": 0.8933333333333333, "train_speed(iter/s)": 1.453972 }, { "epoch": 1.6376761921083074, "grad_norm": 1.777005672454834, "learning_rate": 7.578455262058204e-05, "loss": 0.3842761754989624, "memory(GiB)": 70.5, "step": 38225, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.454001 }, { "epoch": 1.6378904074375562, "grad_norm": 0.8787346482276917, "learning_rate": 7.577878648834241e-05, "loss": 0.14938100576400756, "memory(GiB)": 70.5, "step": 38230, "token_acc": 0.9594594594594594, "train_speed(iter/s)": 1.45401 }, { "epoch": 1.6381046227668052, "grad_norm": 0.13895325362682343, "learning_rate": 7.577301988909583e-05, "loss": 0.41216015815734863, "memory(GiB)": 70.5, "step": 38235, "token_acc": 0.925531914893617, "train_speed(iter/s)": 1.454008 }, { "epoch": 1.6383188380960543, "grad_norm": 2.844888687133789, "learning_rate": 7.576725282294683e-05, "loss": 0.185227370262146, "memory(GiB)": 70.5, "step": 38240, "token_acc": 0.9461538461538461, "train_speed(iter/s)": 1.454004 }, { "epoch": 1.638533053425303, "grad_norm": 3.494873523712158, "learning_rate": 7.576148528999989e-05, "loss": 0.40305075645446775, "memory(GiB)": 70.5, "step": 38245, "token_acc": 0.9171122994652406, "train_speed(iter/s)": 1.453998 }, { "epoch": 1.6387472687545521, "grad_norm": 5.101650238037109, "learning_rate": 7.575571729035948e-05, "loss": 0.747618293762207, "memory(GiB)": 70.5, "step": 38250, "token_acc": 0.8314285714285714, "train_speed(iter/s)": 1.454005 }, { "epoch": 1.6389614840838012, "grad_norm": 0.36021658778190613, "learning_rate": 7.574994882413007e-05, "loss": 0.175684916973114, "memory(GiB)": 70.5, "step": 38255, "token_acc": 0.9572368421052632, "train_speed(iter/s)": 1.454005 }, { "epoch": 1.63917569941305, "grad_norm": 2.51971697807312, "learning_rate": 7.57441798914162e-05, "loss": 0.2797213077545166, "memory(GiB)": 70.5, "step": 38260, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.454012 }, { "epoch": 1.639389914742299, "grad_norm": 5.283199787139893, "learning_rate": 7.573841049232236e-05, "loss": 0.2378298282623291, "memory(GiB)": 70.5, "step": 38265, "token_acc": 0.9498207885304659, "train_speed(iter/s)": 1.454021 }, { "epoch": 1.639604130071548, "grad_norm": 10.811509132385254, "learning_rate": 7.573264062695304e-05, "loss": 0.566251564025879, "memory(GiB)": 70.5, "step": 38270, "token_acc": 0.889763779527559, "train_speed(iter/s)": 1.454044 }, { "epoch": 1.6398183454007969, "grad_norm": 5.619171142578125, "learning_rate": 7.572687029541283e-05, "loss": 0.6572329521179199, "memory(GiB)": 70.5, "step": 38275, "token_acc": 0.867595818815331, "train_speed(iter/s)": 1.454061 }, { "epoch": 1.6400325607300459, "grad_norm": 0.7863997220993042, "learning_rate": 7.572109949780624e-05, "loss": 0.3620820283889771, "memory(GiB)": 70.5, "step": 38280, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.454056 }, { "epoch": 1.640246776059295, "grad_norm": 0.7230288982391357, "learning_rate": 7.571532823423777e-05, "loss": 0.31575038433074953, "memory(GiB)": 70.5, "step": 38285, "token_acc": 0.9294117647058824, "train_speed(iter/s)": 1.454053 }, { "epoch": 1.6404609913885437, "grad_norm": 4.259191036224365, "learning_rate": 7.570955650481202e-05, "loss": 0.3459575653076172, "memory(GiB)": 70.5, "step": 38290, "token_acc": 0.9585798816568047, "train_speed(iter/s)": 1.454067 }, { "epoch": 1.6406752067177928, "grad_norm": 2.842249870300293, "learning_rate": 7.570378430963355e-05, "loss": 0.5846884250640869, "memory(GiB)": 70.5, "step": 38295, "token_acc": 0.8770491803278688, "train_speed(iter/s)": 1.454092 }, { "epoch": 1.6408894220470418, "grad_norm": 2.0601823329925537, "learning_rate": 7.56980116488069e-05, "loss": 0.2235400676727295, "memory(GiB)": 70.5, "step": 38300, "token_acc": 0.956989247311828, "train_speed(iter/s)": 1.454096 }, { "epoch": 1.6411036373762906, "grad_norm": 4.43905782699585, "learning_rate": 7.569223852243666e-05, "loss": 0.4106001853942871, "memory(GiB)": 70.5, "step": 38305, "token_acc": 0.8978723404255319, "train_speed(iter/s)": 1.4541 }, { "epoch": 1.6413178527055396, "grad_norm": 4.102894306182861, "learning_rate": 7.568646493062742e-05, "loss": 0.308023476600647, "memory(GiB)": 70.5, "step": 38310, "token_acc": 0.9401709401709402, "train_speed(iter/s)": 1.454112 }, { "epoch": 1.6415320680347887, "grad_norm": 2.173065423965454, "learning_rate": 7.568069087348377e-05, "loss": 0.47310519218444824, "memory(GiB)": 70.5, "step": 38315, "token_acc": 0.9044585987261147, "train_speed(iter/s)": 1.454111 }, { "epoch": 1.6417462833640375, "grad_norm": 7.0421600341796875, "learning_rate": 7.567491635111033e-05, "loss": 0.3369940757751465, "memory(GiB)": 70.5, "step": 38320, "token_acc": 0.9419354838709677, "train_speed(iter/s)": 1.454136 }, { "epoch": 1.6419604986932865, "grad_norm": 4.985114574432373, "learning_rate": 7.566914136361168e-05, "loss": 0.5918572902679443, "memory(GiB)": 70.5, "step": 38325, "token_acc": 0.8924302788844621, "train_speed(iter/s)": 1.454166 }, { "epoch": 1.6421747140225356, "grad_norm": 3.9462966918945312, "learning_rate": 7.566336591109245e-05, "loss": 0.3606222629547119, "memory(GiB)": 70.5, "step": 38330, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.454179 }, { "epoch": 1.6423889293517844, "grad_norm": 0.7937378287315369, "learning_rate": 7.565758999365728e-05, "loss": 0.30115694999694825, "memory(GiB)": 70.5, "step": 38335, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.454173 }, { "epoch": 1.6426031446810334, "grad_norm": 3.567183017730713, "learning_rate": 7.56518136114108e-05, "loss": 0.453032112121582, "memory(GiB)": 70.5, "step": 38340, "token_acc": 0.9009009009009009, "train_speed(iter/s)": 1.454185 }, { "epoch": 1.6428173600102824, "grad_norm": 5.3094072341918945, "learning_rate": 7.564603676445765e-05, "loss": 0.42542123794555664, "memory(GiB)": 70.5, "step": 38345, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.454188 }, { "epoch": 1.6430315753395313, "grad_norm": 3.874988079071045, "learning_rate": 7.56402594529025e-05, "loss": 0.5836525917053222, "memory(GiB)": 70.5, "step": 38350, "token_acc": 0.8761609907120743, "train_speed(iter/s)": 1.45419 }, { "epoch": 1.6432457906687803, "grad_norm": 1.0546319484710693, "learning_rate": 7.563448167684996e-05, "loss": 0.297449517250061, "memory(GiB)": 70.5, "step": 38355, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.454189 }, { "epoch": 1.6434600059980293, "grad_norm": 3.3667306900024414, "learning_rate": 7.562870343640478e-05, "loss": 0.40724549293518064, "memory(GiB)": 70.5, "step": 38360, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.454192 }, { "epoch": 1.6436742213272781, "grad_norm": 2.8190903663635254, "learning_rate": 7.562292473167158e-05, "loss": 0.3946200132369995, "memory(GiB)": 70.5, "step": 38365, "token_acc": 0.9, "train_speed(iter/s)": 1.454187 }, { "epoch": 1.6438884366565272, "grad_norm": 4.701895236968994, "learning_rate": 7.561714556275505e-05, "loss": 0.2959238052368164, "memory(GiB)": 70.5, "step": 38370, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.454194 }, { "epoch": 1.6441026519857762, "grad_norm": 3.545989513397217, "learning_rate": 7.561136592975993e-05, "loss": 0.3039505958557129, "memory(GiB)": 70.5, "step": 38375, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.454197 }, { "epoch": 1.644316867315025, "grad_norm": 0.4073001742362976, "learning_rate": 7.560558583279085e-05, "loss": 0.47083077430725095, "memory(GiB)": 70.5, "step": 38380, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.454212 }, { "epoch": 1.644531082644274, "grad_norm": 2.973123073577881, "learning_rate": 7.559980527195259e-05, "loss": 0.42217416763305665, "memory(GiB)": 70.5, "step": 38385, "token_acc": 0.907185628742515, "train_speed(iter/s)": 1.454224 }, { "epoch": 1.644745297973523, "grad_norm": 3.4900007247924805, "learning_rate": 7.559402424734982e-05, "loss": 0.5608039379119873, "memory(GiB)": 70.5, "step": 38390, "token_acc": 0.8843283582089553, "train_speed(iter/s)": 1.454225 }, { "epoch": 1.644959513302772, "grad_norm": 2.3944942951202393, "learning_rate": 7.558824275908732e-05, "loss": 0.35646567344665525, "memory(GiB)": 70.5, "step": 38395, "token_acc": 0.9196428571428571, "train_speed(iter/s)": 1.454225 }, { "epoch": 1.645173728632021, "grad_norm": 1.4959086179733276, "learning_rate": 7.558246080726978e-05, "loss": 0.6009365081787109, "memory(GiB)": 70.5, "step": 38400, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.454214 }, { "epoch": 1.64538794396127, "grad_norm": 6.791683673858643, "learning_rate": 7.557667839200198e-05, "loss": 0.401853609085083, "memory(GiB)": 70.5, "step": 38405, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.454214 }, { "epoch": 1.6456021592905188, "grad_norm": 1.759587049484253, "learning_rate": 7.557089551338865e-05, "loss": 0.48642721176147463, "memory(GiB)": 70.5, "step": 38410, "token_acc": 0.8927444794952681, "train_speed(iter/s)": 1.454231 }, { "epoch": 1.6458163746197678, "grad_norm": 5.766465187072754, "learning_rate": 7.556511217153455e-05, "loss": 0.35099310874938966, "memory(GiB)": 70.5, "step": 38415, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.45424 }, { "epoch": 1.6460305899490169, "grad_norm": 3.5238499641418457, "learning_rate": 7.555932836654447e-05, "loss": 0.3429140090942383, "memory(GiB)": 70.5, "step": 38420, "token_acc": 0.9256198347107438, "train_speed(iter/s)": 1.454252 }, { "epoch": 1.6462448052782657, "grad_norm": 1.2856647968292236, "learning_rate": 7.555354409852318e-05, "loss": 0.3364109992980957, "memory(GiB)": 70.5, "step": 38425, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.454268 }, { "epoch": 1.6464590206075147, "grad_norm": 3.1325929164886475, "learning_rate": 7.554775936757545e-05, "loss": 0.35254631042480467, "memory(GiB)": 70.5, "step": 38430, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.45427 }, { "epoch": 1.6466732359367637, "grad_norm": 1.4733004570007324, "learning_rate": 7.554197417380613e-05, "loss": 0.19906388521194457, "memory(GiB)": 70.5, "step": 38435, "token_acc": 0.9590443686006825, "train_speed(iter/s)": 1.454268 }, { "epoch": 1.6468874512660125, "grad_norm": 3.4945132732391357, "learning_rate": 7.553618851731996e-05, "loss": 0.19119019508361818, "memory(GiB)": 70.5, "step": 38440, "token_acc": 0.9550561797752809, "train_speed(iter/s)": 1.454273 }, { "epoch": 1.6471016665952616, "grad_norm": 3.0398356914520264, "learning_rate": 7.553040239822179e-05, "loss": 0.6341733932495117, "memory(GiB)": 70.5, "step": 38445, "token_acc": 0.8592057761732852, "train_speed(iter/s)": 1.454277 }, { "epoch": 1.6473158819245106, "grad_norm": 2.800849676132202, "learning_rate": 7.552461581661643e-05, "loss": 0.33096773624420167, "memory(GiB)": 70.5, "step": 38450, "token_acc": 0.9116465863453815, "train_speed(iter/s)": 1.454272 }, { "epoch": 1.6475300972537594, "grad_norm": 2.305391788482666, "learning_rate": 7.55188287726087e-05, "loss": 0.549852705001831, "memory(GiB)": 70.5, "step": 38455, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.454268 }, { "epoch": 1.6477443125830085, "grad_norm": 7.325887680053711, "learning_rate": 7.551304126630345e-05, "loss": 0.18303894996643066, "memory(GiB)": 70.5, "step": 38460, "token_acc": 0.9574468085106383, "train_speed(iter/s)": 1.454288 }, { "epoch": 1.6479585279122575, "grad_norm": 5.319623947143555, "learning_rate": 7.550725329780555e-05, "loss": 0.6041553497314454, "memory(GiB)": 70.5, "step": 38465, "token_acc": 0.8768115942028986, "train_speed(iter/s)": 1.454286 }, { "epoch": 1.6481727432415063, "grad_norm": 4.5755181312561035, "learning_rate": 7.550146486721981e-05, "loss": 0.5711180210113526, "memory(GiB)": 70.5, "step": 38470, "token_acc": 0.8935361216730038, "train_speed(iter/s)": 1.454303 }, { "epoch": 1.6483869585707553, "grad_norm": 3.156907796859741, "learning_rate": 7.54956759746511e-05, "loss": 0.32408857345581055, "memory(GiB)": 70.5, "step": 38475, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.454306 }, { "epoch": 1.6486011739000044, "grad_norm": 2.298086404800415, "learning_rate": 7.548988662020433e-05, "loss": 0.2526059627532959, "memory(GiB)": 70.5, "step": 38480, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.454301 }, { "epoch": 1.6488153892292532, "grad_norm": 4.067742347717285, "learning_rate": 7.548409680398433e-05, "loss": 0.38907537460327146, "memory(GiB)": 70.5, "step": 38485, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.454294 }, { "epoch": 1.6490296045585022, "grad_norm": 4.530467987060547, "learning_rate": 7.547830652609601e-05, "loss": 0.2087712049484253, "memory(GiB)": 70.5, "step": 38490, "token_acc": 0.959866220735786, "train_speed(iter/s)": 1.454294 }, { "epoch": 1.6492438198877513, "grad_norm": 4.858787536621094, "learning_rate": 7.547251578664427e-05, "loss": 0.4739786148071289, "memory(GiB)": 70.5, "step": 38495, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.454285 }, { "epoch": 1.649458035217, "grad_norm": 2.6708269119262695, "learning_rate": 7.546672458573402e-05, "loss": 0.2591583490371704, "memory(GiB)": 70.5, "step": 38500, "token_acc": 0.9307958477508651, "train_speed(iter/s)": 1.454288 }, { "epoch": 1.649458035217, "eval_loss": 2.3321945667266846, "eval_runtime": 13.408, "eval_samples_per_second": 7.458, "eval_steps_per_second": 7.458, "eval_token_acc": 0.4635854341736695, "step": 38500 }, { "epoch": 1.649672250546249, "grad_norm": 4.68212366104126, "learning_rate": 7.546093292347016e-05, "loss": 0.5445551872253418, "memory(GiB)": 70.5, "step": 38505, "token_acc": 0.5928085519922255, "train_speed(iter/s)": 1.453511 }, { "epoch": 1.6498864658754981, "grad_norm": 2.6649184226989746, "learning_rate": 7.545514079995762e-05, "loss": 0.2340101718902588, "memory(GiB)": 70.5, "step": 38510, "token_acc": 0.9537953795379538, "train_speed(iter/s)": 1.453504 }, { "epoch": 1.650100681204747, "grad_norm": 5.1951680183410645, "learning_rate": 7.544934821530132e-05, "loss": 0.8254538536071777, "memory(GiB)": 70.5, "step": 38515, "token_acc": 0.8305555555555556, "train_speed(iter/s)": 1.453499 }, { "epoch": 1.650314896533996, "grad_norm": 4.127137184143066, "learning_rate": 7.544355516960621e-05, "loss": 0.1590348482131958, "memory(GiB)": 70.5, "step": 38520, "token_acc": 0.9621848739495799, "train_speed(iter/s)": 1.453492 }, { "epoch": 1.650529111863245, "grad_norm": 0.4799747169017792, "learning_rate": 7.543776166297723e-05, "loss": 0.19725366830825805, "memory(GiB)": 70.5, "step": 38525, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.453495 }, { "epoch": 1.6507433271924938, "grad_norm": 0.6031202077865601, "learning_rate": 7.543196769551931e-05, "loss": 0.22146286964416503, "memory(GiB)": 70.5, "step": 38530, "token_acc": 0.9518518518518518, "train_speed(iter/s)": 1.453517 }, { "epoch": 1.6509575425217429, "grad_norm": 4.103658676147461, "learning_rate": 7.542617326733747e-05, "loss": 0.44726061820983887, "memory(GiB)": 70.5, "step": 38535, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.45352 }, { "epoch": 1.651171757850992, "grad_norm": 3.1850693225860596, "learning_rate": 7.542037837853664e-05, "loss": 0.5195112228393555, "memory(GiB)": 70.5, "step": 38540, "token_acc": 0.8745387453874539, "train_speed(iter/s)": 1.453523 }, { "epoch": 1.6513859731802407, "grad_norm": 3.854196310043335, "learning_rate": 7.541458302922179e-05, "loss": 0.32059330940246583, "memory(GiB)": 70.5, "step": 38545, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.453515 }, { "epoch": 1.6516001885094898, "grad_norm": 0.5097566246986389, "learning_rate": 7.540878721949796e-05, "loss": 0.2844571113586426, "memory(GiB)": 70.5, "step": 38550, "token_acc": 0.9329268292682927, "train_speed(iter/s)": 1.453504 }, { "epoch": 1.6518144038387388, "grad_norm": 0.2812715172767639, "learning_rate": 7.54029909494701e-05, "loss": 0.2984903573989868, "memory(GiB)": 70.5, "step": 38555, "token_acc": 0.9528619528619529, "train_speed(iter/s)": 1.453513 }, { "epoch": 1.6520286191679876, "grad_norm": 3.417607307434082, "learning_rate": 7.539719421924322e-05, "loss": 0.16840600967407227, "memory(GiB)": 70.5, "step": 38560, "token_acc": 0.9745222929936306, "train_speed(iter/s)": 1.453515 }, { "epoch": 1.6522428344972366, "grad_norm": 1.0868128538131714, "learning_rate": 7.539139702892235e-05, "loss": 0.5110331535339355, "memory(GiB)": 70.5, "step": 38565, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.453516 }, { "epoch": 1.6524570498264857, "grad_norm": 4.273051738739014, "learning_rate": 7.538559937861251e-05, "loss": 0.4871467113494873, "memory(GiB)": 70.5, "step": 38570, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.453509 }, { "epoch": 1.6526712651557345, "grad_norm": 2.8501720428466797, "learning_rate": 7.53798012684187e-05, "loss": 0.3508581638336182, "memory(GiB)": 70.5, "step": 38575, "token_acc": 0.9221556886227545, "train_speed(iter/s)": 1.453554 }, { "epoch": 1.6528854804849835, "grad_norm": 0.5452368259429932, "learning_rate": 7.537400269844601e-05, "loss": 0.6056573390960693, "memory(GiB)": 70.5, "step": 38580, "token_acc": 0.8701298701298701, "train_speed(iter/s)": 1.453548 }, { "epoch": 1.6530996958142326, "grad_norm": 2.897977828979492, "learning_rate": 7.536820366879946e-05, "loss": 0.4371019840240479, "memory(GiB)": 70.5, "step": 38585, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.453546 }, { "epoch": 1.6533139111434814, "grad_norm": 5.757652282714844, "learning_rate": 7.536240417958409e-05, "loss": 0.8579480171203613, "memory(GiB)": 70.5, "step": 38590, "token_acc": 0.8221574344023324, "train_speed(iter/s)": 1.453567 }, { "epoch": 1.6535281264727304, "grad_norm": 1.4865413904190063, "learning_rate": 7.535660423090498e-05, "loss": 0.21690652370452881, "memory(GiB)": 70.5, "step": 38595, "token_acc": 0.9495798319327731, "train_speed(iter/s)": 1.453586 }, { "epoch": 1.6537423418019794, "grad_norm": 5.762242794036865, "learning_rate": 7.535080382286718e-05, "loss": 0.8427379608154297, "memory(GiB)": 70.5, "step": 38600, "token_acc": 0.8216374269005848, "train_speed(iter/s)": 1.453587 }, { "epoch": 1.6539565571312282, "grad_norm": 2.9482669830322266, "learning_rate": 7.534500295557581e-05, "loss": 0.3367419958114624, "memory(GiB)": 70.5, "step": 38605, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.453596 }, { "epoch": 1.6541707724604773, "grad_norm": 4.330409526824951, "learning_rate": 7.533920162913592e-05, "loss": 0.5594541549682617, "memory(GiB)": 70.5, "step": 38610, "token_acc": 0.8776758409785933, "train_speed(iter/s)": 1.453597 }, { "epoch": 1.6543849877897263, "grad_norm": 4.750876426696777, "learning_rate": 7.533339984365265e-05, "loss": 0.1747591257095337, "memory(GiB)": 70.5, "step": 38615, "token_acc": 0.9522184300341296, "train_speed(iter/s)": 1.453609 }, { "epoch": 1.6545992031189751, "grad_norm": 0.10537248104810715, "learning_rate": 7.532759759923105e-05, "loss": 0.3384012460708618, "memory(GiB)": 70.5, "step": 38620, "token_acc": 0.9045801526717557, "train_speed(iter/s)": 1.453617 }, { "epoch": 1.6548134184482242, "grad_norm": 0.6922722458839417, "learning_rate": 7.532179489597626e-05, "loss": 0.3148325443267822, "memory(GiB)": 70.5, "step": 38625, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.453635 }, { "epoch": 1.6550276337774732, "grad_norm": 4.000414848327637, "learning_rate": 7.531599173399342e-05, "loss": 0.5421735763549804, "memory(GiB)": 70.5, "step": 38630, "token_acc": 0.8868613138686131, "train_speed(iter/s)": 1.45364 }, { "epoch": 1.655241849106722, "grad_norm": 4.181220054626465, "learning_rate": 7.531018811338764e-05, "loss": 0.21601476669311523, "memory(GiB)": 70.5, "step": 38635, "token_acc": 0.9554140127388535, "train_speed(iter/s)": 1.453683 }, { "epoch": 1.655456064435971, "grad_norm": 1.908170223236084, "learning_rate": 7.530438403426403e-05, "loss": 0.3450536966323853, "memory(GiB)": 70.5, "step": 38640, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.453691 }, { "epoch": 1.65567027976522, "grad_norm": 3.4259235858917236, "learning_rate": 7.529857949672778e-05, "loss": 0.4013663291931152, "memory(GiB)": 70.5, "step": 38645, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.453689 }, { "epoch": 1.6558844950944689, "grad_norm": 4.786943435668945, "learning_rate": 7.529277450088405e-05, "loss": 0.39964559078216555, "memory(GiB)": 70.5, "step": 38650, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.453691 }, { "epoch": 1.656098710423718, "grad_norm": 10.099268913269043, "learning_rate": 7.528696904683797e-05, "loss": 0.32553131580352784, "memory(GiB)": 70.5, "step": 38655, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.453707 }, { "epoch": 1.656312925752967, "grad_norm": 5.135870456695557, "learning_rate": 7.528116313469473e-05, "loss": 0.5264992713928223, "memory(GiB)": 70.5, "step": 38660, "token_acc": 0.8952380952380953, "train_speed(iter/s)": 1.453718 }, { "epoch": 1.6565271410822158, "grad_norm": 3.052278995513916, "learning_rate": 7.52753567645595e-05, "loss": 0.5625455856323243, "memory(GiB)": 70.5, "step": 38665, "token_acc": 0.9037900874635568, "train_speed(iter/s)": 1.45372 }, { "epoch": 1.6567413564114648, "grad_norm": 3.7880072593688965, "learning_rate": 7.526954993653747e-05, "loss": 0.4740262508392334, "memory(GiB)": 70.5, "step": 38670, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.453747 }, { "epoch": 1.6569555717407138, "grad_norm": 3.2718188762664795, "learning_rate": 7.526374265073384e-05, "loss": 0.40797224044799807, "memory(GiB)": 70.5, "step": 38675, "token_acc": 0.9046052631578947, "train_speed(iter/s)": 1.453758 }, { "epoch": 1.6571697870699627, "grad_norm": 0.848301351070404, "learning_rate": 7.525793490725381e-05, "loss": 0.37417576313018797, "memory(GiB)": 70.5, "step": 38680, "token_acc": 0.925, "train_speed(iter/s)": 1.453752 }, { "epoch": 1.6573840023992117, "grad_norm": 3.583139181137085, "learning_rate": 7.525212670620261e-05, "loss": 0.32303447723388673, "memory(GiB)": 70.5, "step": 38685, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.453757 }, { "epoch": 1.6575982177284607, "grad_norm": 2.5477402210235596, "learning_rate": 7.524631804768543e-05, "loss": 0.3436806440353394, "memory(GiB)": 70.5, "step": 38690, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.453753 }, { "epoch": 1.6578124330577095, "grad_norm": 1.208693027496338, "learning_rate": 7.524050893180752e-05, "loss": 0.18663153648376465, "memory(GiB)": 70.5, "step": 38695, "token_acc": 0.9605263157894737, "train_speed(iter/s)": 1.453753 }, { "epoch": 1.6580266483869586, "grad_norm": 7.315008640289307, "learning_rate": 7.523469935867411e-05, "loss": 0.3595290184020996, "memory(GiB)": 70.5, "step": 38700, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.453749 }, { "epoch": 1.6582408637162076, "grad_norm": 0.033851224929094315, "learning_rate": 7.522888932839045e-05, "loss": 0.3448637962341309, "memory(GiB)": 70.5, "step": 38705, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.453772 }, { "epoch": 1.6584550790454564, "grad_norm": 0.713729739189148, "learning_rate": 7.52230788410618e-05, "loss": 0.4341418266296387, "memory(GiB)": 70.5, "step": 38710, "token_acc": 0.9006622516556292, "train_speed(iter/s)": 1.453786 }, { "epoch": 1.6586692943747054, "grad_norm": 2.0917844772338867, "learning_rate": 7.52172678967934e-05, "loss": 0.39523792266845703, "memory(GiB)": 70.5, "step": 38715, "token_acc": 0.9009009009009009, "train_speed(iter/s)": 1.453788 }, { "epoch": 1.6588835097039545, "grad_norm": 1.0861198902130127, "learning_rate": 7.521145649569054e-05, "loss": 0.435145902633667, "memory(GiB)": 70.5, "step": 38720, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.453812 }, { "epoch": 1.6590977250332033, "grad_norm": 4.323198318481445, "learning_rate": 7.520564463785851e-05, "loss": 0.44858102798461913, "memory(GiB)": 70.5, "step": 38725, "token_acc": 0.8970588235294118, "train_speed(iter/s)": 1.453821 }, { "epoch": 1.6593119403624523, "grad_norm": 2.486497640609741, "learning_rate": 7.519983232340258e-05, "loss": 0.44878296852111815, "memory(GiB)": 70.5, "step": 38730, "token_acc": 0.9111969111969112, "train_speed(iter/s)": 1.45381 }, { "epoch": 1.6595261556917014, "grad_norm": 4.549025058746338, "learning_rate": 7.519401955242803e-05, "loss": 0.47341384887695315, "memory(GiB)": 70.5, "step": 38735, "token_acc": 0.9118541033434651, "train_speed(iter/s)": 1.45382 }, { "epoch": 1.6597403710209502, "grad_norm": 3.384404182434082, "learning_rate": 7.518820632504021e-05, "loss": 0.8117670059204102, "memory(GiB)": 70.5, "step": 38740, "token_acc": 0.8338658146964856, "train_speed(iter/s)": 1.453836 }, { "epoch": 1.6599545863501992, "grad_norm": 4.9207658767700195, "learning_rate": 7.518239264134439e-05, "loss": 0.6647111892700195, "memory(GiB)": 70.5, "step": 38745, "token_acc": 0.8618421052631579, "train_speed(iter/s)": 1.45385 }, { "epoch": 1.6601688016794482, "grad_norm": 5.026841163635254, "learning_rate": 7.51765785014459e-05, "loss": 0.38076305389404297, "memory(GiB)": 70.5, "step": 38750, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.453839 }, { "epoch": 1.660383017008697, "grad_norm": 2.2934048175811768, "learning_rate": 7.517076390545007e-05, "loss": 0.3756966829299927, "memory(GiB)": 70.5, "step": 38755, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.453845 }, { "epoch": 1.660597232337946, "grad_norm": 4.30295467376709, "learning_rate": 7.516494885346223e-05, "loss": 0.4604618072509766, "memory(GiB)": 70.5, "step": 38760, "token_acc": 0.90234375, "train_speed(iter/s)": 1.453852 }, { "epoch": 1.6608114476671951, "grad_norm": 3.5741310119628906, "learning_rate": 7.515913334558778e-05, "loss": 0.2389530658721924, "memory(GiB)": 70.5, "step": 38765, "token_acc": 0.9603174603174603, "train_speed(iter/s)": 1.453842 }, { "epoch": 1.661025662996444, "grad_norm": 1.6019359827041626, "learning_rate": 7.515331738193199e-05, "loss": 0.3282909870147705, "memory(GiB)": 70.5, "step": 38770, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.45384 }, { "epoch": 1.661239878325693, "grad_norm": 6.546706676483154, "learning_rate": 7.514750096260026e-05, "loss": 0.4659979820251465, "memory(GiB)": 70.5, "step": 38775, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.453848 }, { "epoch": 1.661454093654942, "grad_norm": 17.345909118652344, "learning_rate": 7.514168408769798e-05, "loss": 0.45112180709838867, "memory(GiB)": 70.5, "step": 38780, "token_acc": 0.9283387622149837, "train_speed(iter/s)": 1.453869 }, { "epoch": 1.6616683089841908, "grad_norm": 5.541643142700195, "learning_rate": 7.513586675733049e-05, "loss": 0.22703137397766113, "memory(GiB)": 70.5, "step": 38785, "token_acc": 0.9543568464730291, "train_speed(iter/s)": 1.453865 }, { "epoch": 1.6618825243134399, "grad_norm": 4.856726169586182, "learning_rate": 7.51300489716032e-05, "loss": 0.3588895082473755, "memory(GiB)": 70.5, "step": 38790, "token_acc": 0.9129032258064517, "train_speed(iter/s)": 1.453858 }, { "epoch": 1.662096739642689, "grad_norm": 2.6571197509765625, "learning_rate": 7.51242307306215e-05, "loss": 0.40015287399291993, "memory(GiB)": 70.5, "step": 38795, "token_acc": 0.9198717948717948, "train_speed(iter/s)": 1.453867 }, { "epoch": 1.6623109549719377, "grad_norm": 3.580622434616089, "learning_rate": 7.511841203449079e-05, "loss": 0.5351943016052246, "memory(GiB)": 70.5, "step": 38800, "token_acc": 0.8863636363636364, "train_speed(iter/s)": 1.453864 }, { "epoch": 1.6625251703011867, "grad_norm": 3.4387123584747314, "learning_rate": 7.511259288331649e-05, "loss": 0.4170957088470459, "memory(GiB)": 70.5, "step": 38805, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.453852 }, { "epoch": 1.6627393856304358, "grad_norm": 5.506420612335205, "learning_rate": 7.510677327720401e-05, "loss": 0.5638983726501465, "memory(GiB)": 70.5, "step": 38810, "token_acc": 0.8882521489971347, "train_speed(iter/s)": 1.453859 }, { "epoch": 1.6629536009596846, "grad_norm": 0.7646521925926208, "learning_rate": 7.510095321625878e-05, "loss": 0.4073202610015869, "memory(GiB)": 70.5, "step": 38815, "token_acc": 0.92, "train_speed(iter/s)": 1.453888 }, { "epoch": 1.6631678162889336, "grad_norm": 6.763670444488525, "learning_rate": 7.509513270058624e-05, "loss": 0.3360013008117676, "memory(GiB)": 70.5, "step": 38820, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.453916 }, { "epoch": 1.6633820316181827, "grad_norm": 2.165602922439575, "learning_rate": 7.50893117302918e-05, "loss": 0.3527958631515503, "memory(GiB)": 70.5, "step": 38825, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.453913 }, { "epoch": 1.6635962469474315, "grad_norm": 2.403409957885742, "learning_rate": 7.508349030548099e-05, "loss": 0.680927848815918, "memory(GiB)": 70.5, "step": 38830, "token_acc": 0.8639240506329114, "train_speed(iter/s)": 1.453908 }, { "epoch": 1.6638104622766805, "grad_norm": 2.5226809978485107, "learning_rate": 7.507766842625918e-05, "loss": 0.18951188325881957, "memory(GiB)": 70.5, "step": 38835, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.453912 }, { "epoch": 1.6640246776059295, "grad_norm": 2.142721176147461, "learning_rate": 7.50718460927319e-05, "loss": 0.23062634468078613, "memory(GiB)": 70.5, "step": 38840, "token_acc": 0.9577464788732394, "train_speed(iter/s)": 1.453911 }, { "epoch": 1.6642388929351783, "grad_norm": 0.7766073942184448, "learning_rate": 7.506602330500462e-05, "loss": 0.3739809513092041, "memory(GiB)": 70.5, "step": 38845, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.453924 }, { "epoch": 1.6644531082644274, "grad_norm": 5.874261856079102, "learning_rate": 7.506020006318279e-05, "loss": 0.18852248191833496, "memory(GiB)": 70.5, "step": 38850, "token_acc": 0.9442508710801394, "train_speed(iter/s)": 1.453926 }, { "epoch": 1.6646673235936764, "grad_norm": 2.336003541946411, "learning_rate": 7.505437636737196e-05, "loss": 0.27014381885528566, "memory(GiB)": 70.5, "step": 38855, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.453925 }, { "epoch": 1.6648815389229252, "grad_norm": 4.5091376304626465, "learning_rate": 7.504855221767757e-05, "loss": 0.592491626739502, "memory(GiB)": 70.5, "step": 38860, "token_acc": 0.8745874587458746, "train_speed(iter/s)": 1.453927 }, { "epoch": 1.6650957542521743, "grad_norm": 3.8419971466064453, "learning_rate": 7.504272761420517e-05, "loss": 0.36207945346832277, "memory(GiB)": 70.5, "step": 38865, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.453926 }, { "epoch": 1.6653099695814233, "grad_norm": 2.734929084777832, "learning_rate": 7.503690255706026e-05, "loss": 0.4039429187774658, "memory(GiB)": 70.5, "step": 38870, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.453927 }, { "epoch": 1.6655241849106721, "grad_norm": 4.514584064483643, "learning_rate": 7.503107704634838e-05, "loss": 0.5432223320007324, "memory(GiB)": 70.5, "step": 38875, "token_acc": 0.8862275449101796, "train_speed(iter/s)": 1.453932 }, { "epoch": 1.6657384002399211, "grad_norm": 0.04410155117511749, "learning_rate": 7.502525108217506e-05, "loss": 0.4591036796569824, "memory(GiB)": 70.5, "step": 38880, "token_acc": 0.930327868852459, "train_speed(iter/s)": 1.453955 }, { "epoch": 1.6659526155691702, "grad_norm": 1.4571642875671387, "learning_rate": 7.501942466464584e-05, "loss": 0.22933506965637207, "memory(GiB)": 70.5, "step": 38885, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.453958 }, { "epoch": 1.666166830898419, "grad_norm": 2.247299909591675, "learning_rate": 7.501359779386627e-05, "loss": 0.4754020690917969, "memory(GiB)": 70.5, "step": 38890, "token_acc": 0.8801498127340824, "train_speed(iter/s)": 1.453961 }, { "epoch": 1.666381046227668, "grad_norm": 1.7490583658218384, "learning_rate": 7.500777046994192e-05, "loss": 0.35605993270874026, "memory(GiB)": 70.5, "step": 38895, "token_acc": 0.9356913183279743, "train_speed(iter/s)": 1.453969 }, { "epoch": 1.666595261556917, "grad_norm": 5.637236595153809, "learning_rate": 7.500194269297833e-05, "loss": 0.32269964218139646, "memory(GiB)": 70.5, "step": 38900, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.453968 }, { "epoch": 1.6668094768861659, "grad_norm": 1.5545852184295654, "learning_rate": 7.49961144630811e-05, "loss": 0.527290678024292, "memory(GiB)": 70.5, "step": 38905, "token_acc": 0.8900343642611683, "train_speed(iter/s)": 1.453962 }, { "epoch": 1.667023692215415, "grad_norm": 1.4292106628417969, "learning_rate": 7.499028578035581e-05, "loss": 0.33984692096710206, "memory(GiB)": 70.5, "step": 38910, "token_acc": 0.9379084967320261, "train_speed(iter/s)": 1.453963 }, { "epoch": 1.667237907544664, "grad_norm": 5.771944046020508, "learning_rate": 7.498445664490807e-05, "loss": 0.49729170799255373, "memory(GiB)": 70.5, "step": 38915, "token_acc": 0.9028213166144201, "train_speed(iter/s)": 1.453982 }, { "epoch": 1.6674521228739128, "grad_norm": 1.9439548254013062, "learning_rate": 7.497862705684345e-05, "loss": 0.24865286350250243, "memory(GiB)": 70.5, "step": 38920, "token_acc": 0.9395973154362416, "train_speed(iter/s)": 1.453987 }, { "epoch": 1.6676663382031618, "grad_norm": 4.031322956085205, "learning_rate": 7.497279701626756e-05, "loss": 0.2507979393005371, "memory(GiB)": 70.5, "step": 38925, "token_acc": 0.9283018867924528, "train_speed(iter/s)": 1.453987 }, { "epoch": 1.6678805535324108, "grad_norm": 1.2504323720932007, "learning_rate": 7.496696652328603e-05, "loss": 0.3519754886627197, "memory(GiB)": 70.5, "step": 38930, "token_acc": 0.9096774193548387, "train_speed(iter/s)": 1.453975 }, { "epoch": 1.6680947688616596, "grad_norm": 0.18572726845741272, "learning_rate": 7.496113557800446e-05, "loss": 0.23576593399047852, "memory(GiB)": 70.5, "step": 38935, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.453989 }, { "epoch": 1.6683089841909087, "grad_norm": 5.557038307189941, "learning_rate": 7.495530418052855e-05, "loss": 0.2639147758483887, "memory(GiB)": 70.5, "step": 38940, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.453992 }, { "epoch": 1.6685231995201577, "grad_norm": 2.786146879196167, "learning_rate": 7.494947233096385e-05, "loss": 0.3865020990371704, "memory(GiB)": 70.5, "step": 38945, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.454007 }, { "epoch": 1.6687374148494065, "grad_norm": 3.198312520980835, "learning_rate": 7.494364002941608e-05, "loss": 0.3998744010925293, "memory(GiB)": 70.5, "step": 38950, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.454005 }, { "epoch": 1.6689516301786556, "grad_norm": 1.382652997970581, "learning_rate": 7.493780727599086e-05, "loss": 0.5028648376464844, "memory(GiB)": 70.5, "step": 38955, "token_acc": 0.8719723183391004, "train_speed(iter/s)": 1.454012 }, { "epoch": 1.6691658455079046, "grad_norm": 8.177180290222168, "learning_rate": 7.493197407079389e-05, "loss": 0.5266335487365723, "memory(GiB)": 70.5, "step": 38960, "token_acc": 0.871875, "train_speed(iter/s)": 1.453998 }, { "epoch": 1.6693800608371534, "grad_norm": 3.4273643493652344, "learning_rate": 7.492614041393079e-05, "loss": 0.4031225204467773, "memory(GiB)": 70.5, "step": 38965, "token_acc": 0.9087136929460581, "train_speed(iter/s)": 1.453983 }, { "epoch": 1.6695942761664024, "grad_norm": 1.431535005569458, "learning_rate": 7.492030630550728e-05, "loss": 0.42275471687316896, "memory(GiB)": 70.5, "step": 38970, "token_acc": 0.915068493150685, "train_speed(iter/s)": 1.453996 }, { "epoch": 1.6698084914956515, "grad_norm": 3.804105281829834, "learning_rate": 7.491447174562906e-05, "loss": 0.38226981163024903, "memory(GiB)": 70.5, "step": 38975, "token_acc": 0.9192825112107623, "train_speed(iter/s)": 1.45403 }, { "epoch": 1.6700227068249003, "grad_norm": 3.4452786445617676, "learning_rate": 7.49086367344018e-05, "loss": 0.39681105613708495, "memory(GiB)": 70.5, "step": 38980, "token_acc": 0.8977272727272727, "train_speed(iter/s)": 1.454035 }, { "epoch": 1.6702369221541493, "grad_norm": 2.6674563884735107, "learning_rate": 7.490280127193122e-05, "loss": 0.3235613346099854, "memory(GiB)": 70.5, "step": 38985, "token_acc": 0.9378531073446328, "train_speed(iter/s)": 1.454024 }, { "epoch": 1.6704511374833984, "grad_norm": 5.0682878494262695, "learning_rate": 7.489696535832305e-05, "loss": 0.18958487510681152, "memory(GiB)": 70.5, "step": 38990, "token_acc": 0.959409594095941, "train_speed(iter/s)": 1.454032 }, { "epoch": 1.6706653528126472, "grad_norm": 4.926344871520996, "learning_rate": 7.489112899368298e-05, "loss": 0.8912738800048828, "memory(GiB)": 70.5, "step": 38995, "token_acc": 0.793939393939394, "train_speed(iter/s)": 1.454054 }, { "epoch": 1.6708795681418962, "grad_norm": 2.4149560928344727, "learning_rate": 7.488529217811676e-05, "loss": 0.24499359130859374, "memory(GiB)": 70.5, "step": 39000, "token_acc": 0.9529780564263323, "train_speed(iter/s)": 1.454077 }, { "epoch": 1.6708795681418962, "eval_loss": 2.7255446910858154, "eval_runtime": 13.6809, "eval_samples_per_second": 7.309, "eval_steps_per_second": 7.309, "eval_token_acc": 0.4157441574415744, "step": 39000 }, { "epoch": 1.6710937834711452, "grad_norm": 0.18652161955833435, "learning_rate": 7.487945491173012e-05, "loss": 0.4250608444213867, "memory(GiB)": 70.5, "step": 39005, "token_acc": 0.5471014492753623, "train_speed(iter/s)": 1.453282 }, { "epoch": 1.671307998800394, "grad_norm": 1.8253540992736816, "learning_rate": 7.487361719462883e-05, "loss": 0.38498940467834475, "memory(GiB)": 70.5, "step": 39010, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.453288 }, { "epoch": 1.671522214129643, "grad_norm": 7.587583541870117, "learning_rate": 7.486777902691864e-05, "loss": 0.46563873291015623, "memory(GiB)": 70.5, "step": 39015, "token_acc": 0.8849315068493151, "train_speed(iter/s)": 1.453302 }, { "epoch": 1.6717364294588921, "grad_norm": 2.021646499633789, "learning_rate": 7.48619404087053e-05, "loss": 0.30240283012390134, "memory(GiB)": 70.5, "step": 39020, "token_acc": 0.9227642276422764, "train_speed(iter/s)": 1.45331 }, { "epoch": 1.671950644788141, "grad_norm": 4.101499557495117, "learning_rate": 7.485610134009458e-05, "loss": 0.5715813636779785, "memory(GiB)": 70.5, "step": 39025, "token_acc": 0.8687258687258688, "train_speed(iter/s)": 1.453329 }, { "epoch": 1.67216486011739, "grad_norm": 1.5441930294036865, "learning_rate": 7.485026182119225e-05, "loss": 0.3116016864776611, "memory(GiB)": 70.5, "step": 39030, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.453321 }, { "epoch": 1.672379075446639, "grad_norm": 2.8982908725738525, "learning_rate": 7.484442185210414e-05, "loss": 0.281200647354126, "memory(GiB)": 70.5, "step": 39035, "token_acc": 0.9518900343642611, "train_speed(iter/s)": 1.453309 }, { "epoch": 1.6725932907758878, "grad_norm": 2.913484811782837, "learning_rate": 7.483858143293602e-05, "loss": 0.5732311725616455, "memory(GiB)": 70.5, "step": 39040, "token_acc": 0.8876712328767123, "train_speed(iter/s)": 1.453306 }, { "epoch": 1.6728075061051368, "grad_norm": 2.2026970386505127, "learning_rate": 7.483274056379368e-05, "loss": 0.4163971900939941, "memory(GiB)": 70.5, "step": 39045, "token_acc": 0.9180887372013652, "train_speed(iter/s)": 1.453317 }, { "epoch": 1.6730217214343859, "grad_norm": 6.1174516677856445, "learning_rate": 7.482689924478297e-05, "loss": 0.3163867235183716, "memory(GiB)": 70.5, "step": 39050, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.453317 }, { "epoch": 1.6732359367636347, "grad_norm": 4.481912612915039, "learning_rate": 7.482105747600968e-05, "loss": 0.4177393913269043, "memory(GiB)": 70.5, "step": 39055, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.45333 }, { "epoch": 1.6734501520928837, "grad_norm": 4.711408615112305, "learning_rate": 7.481521525757966e-05, "loss": 0.4736626148223877, "memory(GiB)": 70.5, "step": 39060, "token_acc": 0.9006211180124224, "train_speed(iter/s)": 1.453352 }, { "epoch": 1.6736643674221328, "grad_norm": 4.550480842590332, "learning_rate": 7.480937258959872e-05, "loss": 0.3201404333114624, "memory(GiB)": 70.5, "step": 39065, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.453352 }, { "epoch": 1.6738785827513816, "grad_norm": 2.9967703819274902, "learning_rate": 7.480352947217274e-05, "loss": 0.27299392223358154, "memory(GiB)": 70.5, "step": 39070, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.453353 }, { "epoch": 1.6740927980806306, "grad_norm": 2.473414659500122, "learning_rate": 7.479768590540755e-05, "loss": 0.578348970413208, "memory(GiB)": 70.5, "step": 39075, "token_acc": 0.8825503355704698, "train_speed(iter/s)": 1.453374 }, { "epoch": 1.6743070134098796, "grad_norm": 1.2777820825576782, "learning_rate": 7.479184188940901e-05, "loss": 0.1531907558441162, "memory(GiB)": 70.5, "step": 39080, "token_acc": 0.9652777777777778, "train_speed(iter/s)": 1.453379 }, { "epoch": 1.6745212287391285, "grad_norm": 1.6519969701766968, "learning_rate": 7.4785997424283e-05, "loss": 0.5900639533996582, "memory(GiB)": 70.5, "step": 39085, "token_acc": 0.8316151202749141, "train_speed(iter/s)": 1.453389 }, { "epoch": 1.6747354440683775, "grad_norm": 1.1631139516830444, "learning_rate": 7.47801525101354e-05, "loss": 0.3736544370651245, "memory(GiB)": 70.5, "step": 39090, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.453387 }, { "epoch": 1.6749496593976265, "grad_norm": 3.04616641998291, "learning_rate": 7.477430714707208e-05, "loss": 0.5009597778320313, "memory(GiB)": 70.5, "step": 39095, "token_acc": 0.9010238907849829, "train_speed(iter/s)": 1.45339 }, { "epoch": 1.6751638747268753, "grad_norm": 5.149415969848633, "learning_rate": 7.476846133519896e-05, "loss": 0.5274594783782959, "memory(GiB)": 70.5, "step": 39100, "token_acc": 0.9080882352941176, "train_speed(iter/s)": 1.453414 }, { "epoch": 1.6753780900561244, "grad_norm": 2.807962656021118, "learning_rate": 7.476261507462194e-05, "loss": 0.3642343282699585, "memory(GiB)": 70.5, "step": 39105, "token_acc": 0.9236947791164659, "train_speed(iter/s)": 1.453412 }, { "epoch": 1.6755923053853734, "grad_norm": 5.102998733520508, "learning_rate": 7.47567683654469e-05, "loss": 0.31098480224609376, "memory(GiB)": 70.5, "step": 39110, "token_acc": 0.9357429718875502, "train_speed(iter/s)": 1.453429 }, { "epoch": 1.6758065207146222, "grad_norm": 2.0566043853759766, "learning_rate": 7.475092120777978e-05, "loss": 0.38287081718444826, "memory(GiB)": 70.5, "step": 39115, "token_acc": 0.9316239316239316, "train_speed(iter/s)": 1.453441 }, { "epoch": 1.6760207360438713, "grad_norm": 3.6876673698425293, "learning_rate": 7.47450736017265e-05, "loss": 0.4461972236633301, "memory(GiB)": 70.5, "step": 39120, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.453439 }, { "epoch": 1.6762349513731203, "grad_norm": 0.7607577443122864, "learning_rate": 7.4739225547393e-05, "loss": 0.2663120746612549, "memory(GiB)": 70.5, "step": 39125, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.453446 }, { "epoch": 1.676449166702369, "grad_norm": 2.4050633907318115, "learning_rate": 7.473337704488523e-05, "loss": 0.4499367237091064, "memory(GiB)": 70.5, "step": 39130, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.453427 }, { "epoch": 1.6766633820316181, "grad_norm": 2.2471706867218018, "learning_rate": 7.472752809430913e-05, "loss": 0.6266918182373047, "memory(GiB)": 70.5, "step": 39135, "token_acc": 0.8765432098765432, "train_speed(iter/s)": 1.453422 }, { "epoch": 1.6768775973608672, "grad_norm": 2.6011922359466553, "learning_rate": 7.472167869577066e-05, "loss": 0.4197732925415039, "memory(GiB)": 70.5, "step": 39140, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.453426 }, { "epoch": 1.677091812690116, "grad_norm": 2.811494827270508, "learning_rate": 7.471582884937579e-05, "loss": 0.44982032775878905, "memory(GiB)": 70.5, "step": 39145, "token_acc": 0.9054441260744985, "train_speed(iter/s)": 1.453466 }, { "epoch": 1.677306028019365, "grad_norm": 4.256427764892578, "learning_rate": 7.470997855523049e-05, "loss": 0.4517363548278809, "memory(GiB)": 70.5, "step": 39150, "token_acc": 0.9050632911392406, "train_speed(iter/s)": 1.453467 }, { "epoch": 1.677520243348614, "grad_norm": 3.073030948638916, "learning_rate": 7.470412781344075e-05, "loss": 0.36068058013916016, "memory(GiB)": 70.5, "step": 39155, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.453481 }, { "epoch": 1.6777344586778629, "grad_norm": 1.2666654586791992, "learning_rate": 7.469827662411257e-05, "loss": 0.3710102796554565, "memory(GiB)": 70.5, "step": 39160, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.453492 }, { "epoch": 1.677948674007112, "grad_norm": 5.577808856964111, "learning_rate": 7.469242498735193e-05, "loss": 0.4367549419403076, "memory(GiB)": 70.5, "step": 39165, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.453491 }, { "epoch": 1.678162889336361, "grad_norm": 7.909582138061523, "learning_rate": 7.468657290326486e-05, "loss": 0.5910032272338868, "memory(GiB)": 70.5, "step": 39170, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.453503 }, { "epoch": 1.6783771046656097, "grad_norm": 1.100648045539856, "learning_rate": 7.468072037195735e-05, "loss": 0.33699545860290525, "memory(GiB)": 70.5, "step": 39175, "token_acc": 0.9136690647482014, "train_speed(iter/s)": 1.4535 }, { "epoch": 1.6785913199948588, "grad_norm": 4.490677356719971, "learning_rate": 7.467486739353545e-05, "loss": 0.4786252021789551, "memory(GiB)": 70.5, "step": 39180, "token_acc": 0.9122137404580153, "train_speed(iter/s)": 1.453504 }, { "epoch": 1.6788055353241078, "grad_norm": 5.275033950805664, "learning_rate": 7.466901396810517e-05, "loss": 0.2973208427429199, "memory(GiB)": 70.5, "step": 39185, "token_acc": 0.9405204460966543, "train_speed(iter/s)": 1.453492 }, { "epoch": 1.6790197506533566, "grad_norm": 7.088995933532715, "learning_rate": 7.466316009577258e-05, "loss": 0.5301689147949219, "memory(GiB)": 70.5, "step": 39190, "token_acc": 0.8917910447761194, "train_speed(iter/s)": 1.453485 }, { "epoch": 1.6792339659826059, "grad_norm": 3.319906711578369, "learning_rate": 7.465730577664368e-05, "loss": 0.4136033058166504, "memory(GiB)": 70.5, "step": 39195, "token_acc": 0.8926380368098159, "train_speed(iter/s)": 1.453467 }, { "epoch": 1.6794481813118547, "grad_norm": 1.5990879535675049, "learning_rate": 7.465145101082458e-05, "loss": 0.4620302677154541, "memory(GiB)": 70.5, "step": 39200, "token_acc": 0.8990825688073395, "train_speed(iter/s)": 1.453466 }, { "epoch": 1.6796623966411035, "grad_norm": 3.863457441329956, "learning_rate": 7.464559579842132e-05, "loss": 0.3371325969696045, "memory(GiB)": 70.5, "step": 39205, "token_acc": 0.9335443037974683, "train_speed(iter/s)": 1.453482 }, { "epoch": 1.6798766119703528, "grad_norm": 4.019735336303711, "learning_rate": 7.463974013953995e-05, "loss": 0.5343682765960693, "memory(GiB)": 70.5, "step": 39210, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.45348 }, { "epoch": 1.6800908272996016, "grad_norm": 3.2902777194976807, "learning_rate": 7.463388403428659e-05, "loss": 0.5971166610717773, "memory(GiB)": 70.5, "step": 39215, "token_acc": 0.8970588235294118, "train_speed(iter/s)": 1.453499 }, { "epoch": 1.6803050426288504, "grad_norm": 4.3085618019104, "learning_rate": 7.46280274827673e-05, "loss": 0.4954718589782715, "memory(GiB)": 70.5, "step": 39220, "token_acc": 0.9073359073359073, "train_speed(iter/s)": 1.453498 }, { "epoch": 1.6805192579580996, "grad_norm": 4.862689971923828, "learning_rate": 7.46221704850882e-05, "loss": 0.2961777448654175, "memory(GiB)": 70.5, "step": 39225, "token_acc": 0.9250814332247557, "train_speed(iter/s)": 1.453495 }, { "epoch": 1.6807334732873485, "grad_norm": 7.628137588500977, "learning_rate": 7.461631304135538e-05, "loss": 0.6630476951599121, "memory(GiB)": 70.5, "step": 39230, "token_acc": 0.875, "train_speed(iter/s)": 1.453507 }, { "epoch": 1.6809476886165973, "grad_norm": 1.9827187061309814, "learning_rate": 7.461045515167497e-05, "loss": 0.6108921051025391, "memory(GiB)": 70.5, "step": 39235, "token_acc": 0.8805970149253731, "train_speed(iter/s)": 1.453531 }, { "epoch": 1.6811619039458465, "grad_norm": 1.0328272581100464, "learning_rate": 7.460459681615305e-05, "loss": 0.35951335430145265, "memory(GiB)": 70.5, "step": 39240, "token_acc": 0.9061371841155235, "train_speed(iter/s)": 1.453551 }, { "epoch": 1.6813761192750953, "grad_norm": 4.144232749938965, "learning_rate": 7.45987380348958e-05, "loss": 0.386875057220459, "memory(GiB)": 70.5, "step": 39245, "token_acc": 0.9135802469135802, "train_speed(iter/s)": 1.453575 }, { "epoch": 1.6815903346043442, "grad_norm": 3.2281792163848877, "learning_rate": 7.459287880800933e-05, "loss": 0.5085355281829834, "memory(GiB)": 70.5, "step": 39250, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.453605 }, { "epoch": 1.6818045499335934, "grad_norm": 1.9063239097595215, "learning_rate": 7.458701913559978e-05, "loss": 0.36411266326904296, "memory(GiB)": 70.5, "step": 39255, "token_acc": 0.9217171717171717, "train_speed(iter/s)": 1.453602 }, { "epoch": 1.6820187652628422, "grad_norm": 3.2494068145751953, "learning_rate": 7.458115901777334e-05, "loss": 0.35272836685180664, "memory(GiB)": 70.5, "step": 39260, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.453612 }, { "epoch": 1.682232980592091, "grad_norm": 2.0854570865631104, "learning_rate": 7.457529845463612e-05, "loss": 0.3392197132110596, "memory(GiB)": 70.5, "step": 39265, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.453626 }, { "epoch": 1.6824471959213403, "grad_norm": 4.085208415985107, "learning_rate": 7.456943744629433e-05, "loss": 0.41049861907958984, "memory(GiB)": 70.5, "step": 39270, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.453631 }, { "epoch": 1.682661411250589, "grad_norm": 3.076906442642212, "learning_rate": 7.456357599285413e-05, "loss": 0.4479008674621582, "memory(GiB)": 70.5, "step": 39275, "token_acc": 0.9034090909090909, "train_speed(iter/s)": 1.453642 }, { "epoch": 1.682875626579838, "grad_norm": 3.33467960357666, "learning_rate": 7.455771409442171e-05, "loss": 0.5907634735107422, "memory(GiB)": 70.5, "step": 39280, "token_acc": 0.8868613138686131, "train_speed(iter/s)": 1.453652 }, { "epoch": 1.6830898419090872, "grad_norm": 0.07366061955690384, "learning_rate": 7.455185175110325e-05, "loss": 0.35550885200500487, "memory(GiB)": 70.5, "step": 39285, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.45366 }, { "epoch": 1.683304057238336, "grad_norm": 3.096719980239868, "learning_rate": 7.454598896300498e-05, "loss": 0.1438979148864746, "memory(GiB)": 70.5, "step": 39290, "token_acc": 0.9700996677740864, "train_speed(iter/s)": 1.453677 }, { "epoch": 1.6835182725675848, "grad_norm": 4.366148948669434, "learning_rate": 7.454012573023308e-05, "loss": 0.5419706344604492, "memory(GiB)": 70.5, "step": 39295, "token_acc": 0.8860294117647058, "train_speed(iter/s)": 1.453684 }, { "epoch": 1.683732487896834, "grad_norm": 3.1833105087280273, "learning_rate": 7.453426205289379e-05, "loss": 0.5459502220153809, "memory(GiB)": 70.5, "step": 39300, "token_acc": 0.8876811594202898, "train_speed(iter/s)": 1.453712 }, { "epoch": 1.6839467032260829, "grad_norm": 1.692571997642517, "learning_rate": 7.452839793109332e-05, "loss": 0.3134181976318359, "memory(GiB)": 70.5, "step": 39305, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.45374 }, { "epoch": 1.6841609185553317, "grad_norm": 3.9829318523406982, "learning_rate": 7.452253336493791e-05, "loss": 0.30823550224304197, "memory(GiB)": 70.5, "step": 39310, "token_acc": 0.9383116883116883, "train_speed(iter/s)": 1.453744 }, { "epoch": 1.684375133884581, "grad_norm": 4.201863765716553, "learning_rate": 7.451666835453382e-05, "loss": 0.4910627841949463, "memory(GiB)": 70.5, "step": 39315, "token_acc": 0.9039145907473309, "train_speed(iter/s)": 1.453737 }, { "epoch": 1.6845893492138297, "grad_norm": 4.115212917327881, "learning_rate": 7.451080289998729e-05, "loss": 0.6057473182678222, "memory(GiB)": 70.5, "step": 39320, "token_acc": 0.8933333333333333, "train_speed(iter/s)": 1.453736 }, { "epoch": 1.6848035645430786, "grad_norm": 4.390686511993408, "learning_rate": 7.450493700140454e-05, "loss": 0.4732190132141113, "memory(GiB)": 70.5, "step": 39325, "token_acc": 0.9042904290429042, "train_speed(iter/s)": 1.453748 }, { "epoch": 1.6850177798723278, "grad_norm": 3.7310924530029297, "learning_rate": 7.44990706588919e-05, "loss": 0.32591967582702636, "memory(GiB)": 70.5, "step": 39330, "token_acc": 0.9184952978056427, "train_speed(iter/s)": 1.453775 }, { "epoch": 1.6852319952015766, "grad_norm": 0.26358890533447266, "learning_rate": 7.44932038725556e-05, "loss": 0.42298226356506347, "memory(GiB)": 70.5, "step": 39335, "token_acc": 0.9105960264900662, "train_speed(iter/s)": 1.453783 }, { "epoch": 1.6854462105308254, "grad_norm": 0.08083267509937286, "learning_rate": 7.448733664250191e-05, "loss": 0.2973008394241333, "memory(GiB)": 70.5, "step": 39340, "token_acc": 0.9348534201954397, "train_speed(iter/s)": 1.453791 }, { "epoch": 1.6856604258600747, "grad_norm": 0.7274655699729919, "learning_rate": 7.448146896883719e-05, "loss": 0.1812673807144165, "memory(GiB)": 70.5, "step": 39345, "token_acc": 0.9606986899563319, "train_speed(iter/s)": 1.453791 }, { "epoch": 1.6858746411893235, "grad_norm": 2.2656431198120117, "learning_rate": 7.447560085166768e-05, "loss": 0.493457555770874, "memory(GiB)": 70.5, "step": 39350, "token_acc": 0.8925925925925926, "train_speed(iter/s)": 1.453796 }, { "epoch": 1.6860888565185723, "grad_norm": 4.0561652183532715, "learning_rate": 7.446973229109969e-05, "loss": 0.7436046123504638, "memory(GiB)": 70.5, "step": 39355, "token_acc": 0.8565217391304348, "train_speed(iter/s)": 1.453808 }, { "epoch": 1.6863030718478216, "grad_norm": 5.254225730895996, "learning_rate": 7.446386328723955e-05, "loss": 0.38333725929260254, "memory(GiB)": 70.5, "step": 39360, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.453802 }, { "epoch": 1.6865172871770704, "grad_norm": 2.7551751136779785, "learning_rate": 7.445799384019358e-05, "loss": 0.3114474773406982, "memory(GiB)": 70.5, "step": 39365, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.453804 }, { "epoch": 1.6867315025063192, "grad_norm": 2.132268190383911, "learning_rate": 7.44521239500681e-05, "loss": 0.2530024290084839, "memory(GiB)": 70.5, "step": 39370, "token_acc": 0.9416342412451362, "train_speed(iter/s)": 1.453811 }, { "epoch": 1.6869457178355685, "grad_norm": 2.8480935096740723, "learning_rate": 7.444625361696948e-05, "loss": 0.4387578010559082, "memory(GiB)": 70.5, "step": 39375, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.453806 }, { "epoch": 1.6871599331648173, "grad_norm": 1.5193346738815308, "learning_rate": 7.444038284100401e-05, "loss": 0.3233656406402588, "memory(GiB)": 70.5, "step": 39380, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.453813 }, { "epoch": 1.687374148494066, "grad_norm": 1.696535587310791, "learning_rate": 7.44345116222781e-05, "loss": 0.2985325813293457, "memory(GiB)": 70.5, "step": 39385, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.453825 }, { "epoch": 1.6875883638233153, "grad_norm": 2.9630465507507324, "learning_rate": 7.442863996089809e-05, "loss": 0.27475039958953856, "memory(GiB)": 70.5, "step": 39390, "token_acc": 0.9540983606557377, "train_speed(iter/s)": 1.453832 }, { "epoch": 1.6878025791525642, "grad_norm": 0.8186569213867188, "learning_rate": 7.442276785697035e-05, "loss": 0.28530886173248293, "memory(GiB)": 70.5, "step": 39395, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.453835 }, { "epoch": 1.688016794481813, "grad_norm": 2.3991196155548096, "learning_rate": 7.441689531060125e-05, "loss": 0.4431631565093994, "memory(GiB)": 70.5, "step": 39400, "token_acc": 0.897119341563786, "train_speed(iter/s)": 1.453849 }, { "epoch": 1.6882310098110622, "grad_norm": 2.882767915725708, "learning_rate": 7.441102232189721e-05, "loss": 0.41167435646057127, "memory(GiB)": 70.5, "step": 39405, "token_acc": 0.9147286821705426, "train_speed(iter/s)": 1.453856 }, { "epoch": 1.688445225140311, "grad_norm": 3.3137733936309814, "learning_rate": 7.440514889096457e-05, "loss": 0.3560910224914551, "memory(GiB)": 70.5, "step": 39410, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.453865 }, { "epoch": 1.6886594404695598, "grad_norm": 4.3923749923706055, "learning_rate": 7.439927501790978e-05, "loss": 0.6875655174255371, "memory(GiB)": 70.5, "step": 39415, "token_acc": 0.8346153846153846, "train_speed(iter/s)": 1.453859 }, { "epoch": 1.688873655798809, "grad_norm": 2.249577522277832, "learning_rate": 7.439340070283923e-05, "loss": 0.29154417514801023, "memory(GiB)": 70.5, "step": 39420, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.453867 }, { "epoch": 1.689087871128058, "grad_norm": 0.641918957233429, "learning_rate": 7.438752594585935e-05, "loss": 0.2288520574569702, "memory(GiB)": 70.5, "step": 39425, "token_acc": 0.9295154185022027, "train_speed(iter/s)": 1.453864 }, { "epoch": 1.6893020864573067, "grad_norm": 2.885976791381836, "learning_rate": 7.438165074707655e-05, "loss": 0.33098981380462644, "memory(GiB)": 70.5, "step": 39430, "token_acc": 0.933993399339934, "train_speed(iter/s)": 1.453877 }, { "epoch": 1.689516301786556, "grad_norm": 4.120304584503174, "learning_rate": 7.437577510659729e-05, "loss": 0.5590179920196533, "memory(GiB)": 70.5, "step": 39435, "token_acc": 0.8773946360153256, "train_speed(iter/s)": 1.453894 }, { "epoch": 1.6897305171158048, "grad_norm": 6.189676761627197, "learning_rate": 7.436989902452798e-05, "loss": 0.45163555145263673, "memory(GiB)": 70.5, "step": 39440, "token_acc": 0.8964285714285715, "train_speed(iter/s)": 1.45391 }, { "epoch": 1.6899447324450536, "grad_norm": 1.2433191537857056, "learning_rate": 7.43640225009751e-05, "loss": 0.461668062210083, "memory(GiB)": 70.5, "step": 39445, "token_acc": 0.8940677966101694, "train_speed(iter/s)": 1.453929 }, { "epoch": 1.6901589477743029, "grad_norm": 2.3357009887695312, "learning_rate": 7.43581455360451e-05, "loss": 0.31682634353637695, "memory(GiB)": 70.5, "step": 39450, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.453929 }, { "epoch": 1.6903731631035517, "grad_norm": 2.1324923038482666, "learning_rate": 7.435226812984443e-05, "loss": 0.39417898654937744, "memory(GiB)": 70.5, "step": 39455, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.453931 }, { "epoch": 1.6905873784328005, "grad_norm": 6.436469078063965, "learning_rate": 7.434639028247959e-05, "loss": 0.6245346546173096, "memory(GiB)": 70.5, "step": 39460, "token_acc": 0.8766233766233766, "train_speed(iter/s)": 1.45395 }, { "epoch": 1.6908015937620497, "grad_norm": 4.043915748596191, "learning_rate": 7.434051199405705e-05, "loss": 0.426737642288208, "memory(GiB)": 70.5, "step": 39465, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.453961 }, { "epoch": 1.6910158090912986, "grad_norm": 4.237301826477051, "learning_rate": 7.43346332646833e-05, "loss": 0.4572641849517822, "memory(GiB)": 70.5, "step": 39470, "token_acc": 0.9088050314465409, "train_speed(iter/s)": 1.453964 }, { "epoch": 1.6912300244205474, "grad_norm": 3.2054364681243896, "learning_rate": 7.432875409446483e-05, "loss": 0.4971658706665039, "memory(GiB)": 70.5, "step": 39475, "token_acc": 0.8827160493827161, "train_speed(iter/s)": 1.453968 }, { "epoch": 1.6914442397497966, "grad_norm": 0.9760488271713257, "learning_rate": 7.432287448350819e-05, "loss": 0.20381858348846435, "memory(GiB)": 70.5, "step": 39480, "token_acc": 0.948220064724919, "train_speed(iter/s)": 1.453974 }, { "epoch": 1.6916584550790454, "grad_norm": 1.940956473350525, "learning_rate": 7.431699443191982e-05, "loss": 0.31040225028991697, "memory(GiB)": 70.5, "step": 39485, "token_acc": 0.9283489096573209, "train_speed(iter/s)": 1.453977 }, { "epoch": 1.6918726704082943, "grad_norm": 3.9587314128875732, "learning_rate": 7.431111393980631e-05, "loss": 0.4365867614746094, "memory(GiB)": 70.5, "step": 39490, "token_acc": 0.9262536873156342, "train_speed(iter/s)": 1.453971 }, { "epoch": 1.6920868857375435, "grad_norm": 0.16576997935771942, "learning_rate": 7.430523300727416e-05, "loss": 0.5477306365966796, "memory(GiB)": 70.5, "step": 39495, "token_acc": 0.8893280632411067, "train_speed(iter/s)": 1.453992 }, { "epoch": 1.6923011010667923, "grad_norm": 3.175525188446045, "learning_rate": 7.429935163442991e-05, "loss": 0.1810485005378723, "memory(GiB)": 70.5, "step": 39500, "token_acc": 0.9621993127147767, "train_speed(iter/s)": 1.454007 }, { "epoch": 1.6923011010667923, "eval_loss": 2.4277913570404053, "eval_runtime": 13.348, "eval_samples_per_second": 7.492, "eval_steps_per_second": 7.492, "eval_token_acc": 0.4665757162346521, "step": 39500 }, { "epoch": 1.6925153163960411, "grad_norm": 6.847656726837158, "learning_rate": 7.429346982138013e-05, "loss": 0.6738387107849121, "memory(GiB)": 70.5, "step": 39505, "token_acc": 0.5905292479108635, "train_speed(iter/s)": 1.453256 }, { "epoch": 1.6927295317252904, "grad_norm": 3.5379340648651123, "learning_rate": 7.428758756823134e-05, "loss": 0.528270673751831, "memory(GiB)": 70.5, "step": 39510, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.453259 }, { "epoch": 1.6929437470545392, "grad_norm": 0.06027791276574135, "learning_rate": 7.42817048750901e-05, "loss": 0.19927074909210205, "memory(GiB)": 70.5, "step": 39515, "token_acc": 0.9547169811320755, "train_speed(iter/s)": 1.453275 }, { "epoch": 1.693157962383788, "grad_norm": 1.699671745300293, "learning_rate": 7.427582174206303e-05, "loss": 0.48091773986816405, "memory(GiB)": 70.5, "step": 39520, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.453274 }, { "epoch": 1.6933721777130373, "grad_norm": 3.7019312381744385, "learning_rate": 7.426993816925665e-05, "loss": 0.2232785701751709, "memory(GiB)": 70.5, "step": 39525, "token_acc": 0.9352941176470588, "train_speed(iter/s)": 1.453276 }, { "epoch": 1.693586393042286, "grad_norm": 3.2981677055358887, "learning_rate": 7.426405415677758e-05, "loss": 0.6087026596069336, "memory(GiB)": 70.5, "step": 39530, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.453288 }, { "epoch": 1.6938006083715351, "grad_norm": 1.7985786199569702, "learning_rate": 7.425816970473241e-05, "loss": 0.37131471633911134, "memory(GiB)": 70.5, "step": 39535, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.453288 }, { "epoch": 1.6940148237007842, "grad_norm": 2.9293789863586426, "learning_rate": 7.425228481322775e-05, "loss": 0.518099594116211, "memory(GiB)": 70.5, "step": 39540, "token_acc": 0.8981132075471698, "train_speed(iter/s)": 1.453288 }, { "epoch": 1.694229039030033, "grad_norm": 1.6029047966003418, "learning_rate": 7.424639948237019e-05, "loss": 0.47744436264038087, "memory(GiB)": 70.5, "step": 39545, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.453279 }, { "epoch": 1.694443254359282, "grad_norm": 3.4541280269622803, "learning_rate": 7.424051371226636e-05, "loss": 0.4991215705871582, "memory(GiB)": 70.5, "step": 39550, "token_acc": 0.9008746355685131, "train_speed(iter/s)": 1.453278 }, { "epoch": 1.694657469688531, "grad_norm": 2.0251386165618896, "learning_rate": 7.423462750302289e-05, "loss": 0.4405148983001709, "memory(GiB)": 70.5, "step": 39555, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.453306 }, { "epoch": 1.6948716850177798, "grad_norm": 4.645370006561279, "learning_rate": 7.42287408547464e-05, "loss": 0.4061575412750244, "memory(GiB)": 70.5, "step": 39560, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.453338 }, { "epoch": 1.6950859003470289, "grad_norm": 5.8109636306762695, "learning_rate": 7.422285376754354e-05, "loss": 0.3941459655761719, "memory(GiB)": 70.5, "step": 39565, "token_acc": 0.9147982062780269, "train_speed(iter/s)": 1.453336 }, { "epoch": 1.695300115676278, "grad_norm": 3.1488802433013916, "learning_rate": 7.421696624152098e-05, "loss": 0.15615925788879395, "memory(GiB)": 70.5, "step": 39570, "token_acc": 0.9681274900398407, "train_speed(iter/s)": 1.453344 }, { "epoch": 1.6955143310055267, "grad_norm": 10.813016891479492, "learning_rate": 7.421107827678533e-05, "loss": 0.7244872093200684, "memory(GiB)": 70.5, "step": 39575, "token_acc": 0.8436363636363636, "train_speed(iter/s)": 1.45334 }, { "epoch": 1.6957285463347758, "grad_norm": 6.663282871246338, "learning_rate": 7.420518987344331e-05, "loss": 0.48666629791259763, "memory(GiB)": 70.5, "step": 39580, "token_acc": 0.8945454545454545, "train_speed(iter/s)": 1.453338 }, { "epoch": 1.6959427616640248, "grad_norm": 3.5082290172576904, "learning_rate": 7.419930103160155e-05, "loss": 0.4343743324279785, "memory(GiB)": 70.5, "step": 39585, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.453342 }, { "epoch": 1.6961569769932736, "grad_norm": 4.364416122436523, "learning_rate": 7.419341175136677e-05, "loss": 0.3516599893569946, "memory(GiB)": 70.5, "step": 39590, "token_acc": 0.915129151291513, "train_speed(iter/s)": 1.453339 }, { "epoch": 1.6963711923225226, "grad_norm": 4.954138278961182, "learning_rate": 7.418752203284563e-05, "loss": 0.43683204650878904, "memory(GiB)": 70.5, "step": 39595, "token_acc": 0.9159159159159159, "train_speed(iter/s)": 1.45335 }, { "epoch": 1.6965854076517717, "grad_norm": 7.349950313568115, "learning_rate": 7.418163187614485e-05, "loss": 0.26599276065826416, "memory(GiB)": 70.5, "step": 39600, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.453351 }, { "epoch": 1.6967996229810205, "grad_norm": 2.046761989593506, "learning_rate": 7.41757412813711e-05, "loss": 0.36175649166107177, "memory(GiB)": 70.5, "step": 39605, "token_acc": 0.9006622516556292, "train_speed(iter/s)": 1.453355 }, { "epoch": 1.6970138383102695, "grad_norm": 4.17209005355835, "learning_rate": 7.416985024863115e-05, "loss": 0.4988722324371338, "memory(GiB)": 70.5, "step": 39610, "token_acc": 0.8828125, "train_speed(iter/s)": 1.453351 }, { "epoch": 1.6972280536395186, "grad_norm": 4.518965721130371, "learning_rate": 7.416395877803166e-05, "loss": 0.3137868165969849, "memory(GiB)": 70.5, "step": 39615, "token_acc": 0.9442379182156134, "train_speed(iter/s)": 1.453358 }, { "epoch": 1.6974422689687674, "grad_norm": 2.072848320007324, "learning_rate": 7.415806686967941e-05, "loss": 0.3316534996032715, "memory(GiB)": 70.5, "step": 39620, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.453358 }, { "epoch": 1.6976564842980164, "grad_norm": 2.0106027126312256, "learning_rate": 7.41521745236811e-05, "loss": 0.5722187042236329, "memory(GiB)": 70.5, "step": 39625, "token_acc": 0.8745980707395499, "train_speed(iter/s)": 1.453371 }, { "epoch": 1.6978706996272654, "grad_norm": 5.31198787689209, "learning_rate": 7.414628174014351e-05, "loss": 0.38184027671813964, "memory(GiB)": 70.5, "step": 39630, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.453392 }, { "epoch": 1.6980849149565143, "grad_norm": 1.0502800941467285, "learning_rate": 7.414038851917335e-05, "loss": 0.2318826675415039, "memory(GiB)": 70.5, "step": 39635, "token_acc": 0.9377431906614786, "train_speed(iter/s)": 1.453385 }, { "epoch": 1.6982991302857633, "grad_norm": 3.5061192512512207, "learning_rate": 7.413449486087742e-05, "loss": 0.2777836322784424, "memory(GiB)": 70.5, "step": 39640, "token_acc": 0.9276595744680851, "train_speed(iter/s)": 1.453439 }, { "epoch": 1.6985133456150123, "grad_norm": 10.078221321105957, "learning_rate": 7.412860076536247e-05, "loss": 0.3857893943786621, "memory(GiB)": 70.5, "step": 39645, "token_acc": 0.9214876033057852, "train_speed(iter/s)": 1.45345 }, { "epoch": 1.6987275609442611, "grad_norm": 1.1167449951171875, "learning_rate": 7.41227062327353e-05, "loss": 0.3161031246185303, "memory(GiB)": 70.5, "step": 39650, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.453436 }, { "epoch": 1.6989417762735102, "grad_norm": 4.202400207519531, "learning_rate": 7.411681126310268e-05, "loss": 0.6033103466033936, "memory(GiB)": 70.5, "step": 39655, "token_acc": 0.8630573248407644, "train_speed(iter/s)": 1.453426 }, { "epoch": 1.6991559916027592, "grad_norm": 3.0862536430358887, "learning_rate": 7.411091585657139e-05, "loss": 0.4504354476928711, "memory(GiB)": 70.5, "step": 39660, "token_acc": 0.8914473684210527, "train_speed(iter/s)": 1.453423 }, { "epoch": 1.699370206932008, "grad_norm": 2.1973953247070312, "learning_rate": 7.410502001324824e-05, "loss": 0.43562917709350585, "memory(GiB)": 70.5, "step": 39665, "token_acc": 0.9141104294478528, "train_speed(iter/s)": 1.453428 }, { "epoch": 1.699584422261257, "grad_norm": 4.871042251586914, "learning_rate": 7.409912373324003e-05, "loss": 0.37212185859680175, "memory(GiB)": 70.5, "step": 39670, "token_acc": 0.9164086687306502, "train_speed(iter/s)": 1.453457 }, { "epoch": 1.699798637590506, "grad_norm": 3.4923312664031982, "learning_rate": 7.409322701665359e-05, "loss": 0.5081354141235351, "memory(GiB)": 70.5, "step": 39675, "token_acc": 0.8768115942028986, "train_speed(iter/s)": 1.453465 }, { "epoch": 1.700012852919755, "grad_norm": 1.3322924375534058, "learning_rate": 7.408732986359576e-05, "loss": 0.22161543369293213, "memory(GiB)": 70.5, "step": 39680, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.453463 }, { "epoch": 1.700227068249004, "grad_norm": 2.2100770473480225, "learning_rate": 7.408143227417334e-05, "loss": 0.34739949703216555, "memory(GiB)": 70.5, "step": 39685, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.453473 }, { "epoch": 1.700441283578253, "grad_norm": 2.2044146060943604, "learning_rate": 7.407553424849318e-05, "loss": 0.41053457260131837, "memory(GiB)": 70.5, "step": 39690, "token_acc": 0.9190283400809717, "train_speed(iter/s)": 1.45347 }, { "epoch": 1.7006554989075018, "grad_norm": 5.500365734100342, "learning_rate": 7.406963578666214e-05, "loss": 0.43144936561584474, "memory(GiB)": 70.5, "step": 39695, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.453477 }, { "epoch": 1.7008697142367508, "grad_norm": 5.909147262573242, "learning_rate": 7.406373688878706e-05, "loss": 0.3650266170501709, "memory(GiB)": 70.5, "step": 39700, "token_acc": 0.9312977099236641, "train_speed(iter/s)": 1.453481 }, { "epoch": 1.7010839295659999, "grad_norm": 5.929091930389404, "learning_rate": 7.405783755497482e-05, "loss": 0.8873443603515625, "memory(GiB)": 70.5, "step": 39705, "token_acc": 0.8304093567251462, "train_speed(iter/s)": 1.453499 }, { "epoch": 1.7012981448952487, "grad_norm": 2.3058664798736572, "learning_rate": 7.405193778533229e-05, "loss": 0.33358030319213866, "memory(GiB)": 70.5, "step": 39710, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.453495 }, { "epoch": 1.7015123602244977, "grad_norm": 2.8444371223449707, "learning_rate": 7.404603757996634e-05, "loss": 0.3866913080215454, "memory(GiB)": 70.5, "step": 39715, "token_acc": 0.9110320284697508, "train_speed(iter/s)": 1.453493 }, { "epoch": 1.7017265755537467, "grad_norm": 4.6373677253723145, "learning_rate": 7.404013693898385e-05, "loss": 0.7098280906677246, "memory(GiB)": 70.5, "step": 39720, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.453511 }, { "epoch": 1.7019407908829955, "grad_norm": 4.468612194061279, "learning_rate": 7.403423586249174e-05, "loss": 0.4826189041137695, "memory(GiB)": 70.5, "step": 39725, "token_acc": 0.9013605442176871, "train_speed(iter/s)": 1.453502 }, { "epoch": 1.7021550062122446, "grad_norm": 6.562770366668701, "learning_rate": 7.402833435059691e-05, "loss": 0.4107752799987793, "memory(GiB)": 70.5, "step": 39730, "token_acc": 0.9063670411985019, "train_speed(iter/s)": 1.453501 }, { "epoch": 1.7023692215414936, "grad_norm": 2.0124118328094482, "learning_rate": 7.402243240340624e-05, "loss": 0.41216888427734377, "memory(GiB)": 70.5, "step": 39735, "token_acc": 0.9108527131782945, "train_speed(iter/s)": 1.453528 }, { "epoch": 1.7025834368707424, "grad_norm": 3.5274059772491455, "learning_rate": 7.401653002102669e-05, "loss": 0.17669813632965087, "memory(GiB)": 70.5, "step": 39740, "token_acc": 0.959731543624161, "train_speed(iter/s)": 1.453522 }, { "epoch": 1.7027976521999915, "grad_norm": 1.7960559129714966, "learning_rate": 7.401062720356516e-05, "loss": 0.5387727737426757, "memory(GiB)": 70.5, "step": 39745, "token_acc": 0.8895027624309392, "train_speed(iter/s)": 1.453525 }, { "epoch": 1.7030118675292405, "grad_norm": 2.642179250717163, "learning_rate": 7.400472395112861e-05, "loss": 0.48280348777771, "memory(GiB)": 70.5, "step": 39750, "token_acc": 0.8895522388059701, "train_speed(iter/s)": 1.453541 }, { "epoch": 1.7032260828584893, "grad_norm": 1.4964817762374878, "learning_rate": 7.399882026382395e-05, "loss": 0.5389759540557861, "memory(GiB)": 70.5, "step": 39755, "token_acc": 0.8872832369942196, "train_speed(iter/s)": 1.453544 }, { "epoch": 1.7034402981877383, "grad_norm": 1.653286099433899, "learning_rate": 7.399291614175815e-05, "loss": 0.29520258903503416, "memory(GiB)": 70.5, "step": 39760, "token_acc": 0.9426751592356688, "train_speed(iter/s)": 1.453547 }, { "epoch": 1.7036545135169874, "grad_norm": 5.932728290557861, "learning_rate": 7.398701158503818e-05, "loss": 0.3504958152770996, "memory(GiB)": 70.5, "step": 39765, "token_acc": 0.9216300940438872, "train_speed(iter/s)": 1.453572 }, { "epoch": 1.7038687288462362, "grad_norm": 4.66813325881958, "learning_rate": 7.398110659377098e-05, "loss": 0.6130581378936768, "memory(GiB)": 70.5, "step": 39770, "token_acc": 0.8793650793650793, "train_speed(iter/s)": 1.453577 }, { "epoch": 1.7040829441754852, "grad_norm": 3.807218551635742, "learning_rate": 7.397520116806354e-05, "loss": 0.44211502075195314, "memory(GiB)": 70.5, "step": 39775, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.453574 }, { "epoch": 1.7042971595047343, "grad_norm": 7.327968597412109, "learning_rate": 7.396929530802286e-05, "loss": 0.2837367534637451, "memory(GiB)": 70.5, "step": 39780, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.453582 }, { "epoch": 1.704511374833983, "grad_norm": 1.8006559610366821, "learning_rate": 7.396338901375588e-05, "loss": 0.606471586227417, "memory(GiB)": 70.5, "step": 39785, "token_acc": 0.8688524590163934, "train_speed(iter/s)": 1.453582 }, { "epoch": 1.704725590163232, "grad_norm": 1.4814585447311401, "learning_rate": 7.395748228536964e-05, "loss": 0.22670974731445312, "memory(GiB)": 70.5, "step": 39790, "token_acc": 0.9560117302052786, "train_speed(iter/s)": 1.453582 }, { "epoch": 1.7049398054924811, "grad_norm": 0.11664406955242157, "learning_rate": 7.395157512297114e-05, "loss": 0.3097347021102905, "memory(GiB)": 70.5, "step": 39795, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.453604 }, { "epoch": 1.70515402082173, "grad_norm": 6.697225093841553, "learning_rate": 7.394566752666739e-05, "loss": 0.42646331787109376, "memory(GiB)": 70.5, "step": 39800, "token_acc": 0.9003690036900369, "train_speed(iter/s)": 1.453607 }, { "epoch": 1.705368236150979, "grad_norm": 2.87565279006958, "learning_rate": 7.39397594965654e-05, "loss": 0.5181509017944336, "memory(GiB)": 70.5, "step": 39805, "token_acc": 0.9120521172638436, "train_speed(iter/s)": 1.453612 }, { "epoch": 1.705582451480228, "grad_norm": 0.18971221148967743, "learning_rate": 7.393385103277222e-05, "loss": 0.16731547117233275, "memory(GiB)": 70.5, "step": 39810, "token_acc": 0.9553903345724907, "train_speed(iter/s)": 1.453624 }, { "epoch": 1.7057966668094768, "grad_norm": 2.24284029006958, "learning_rate": 7.392794213539486e-05, "loss": 0.31624279022216795, "memory(GiB)": 70.5, "step": 39815, "token_acc": 0.9271255060728745, "train_speed(iter/s)": 1.453638 }, { "epoch": 1.7060108821387259, "grad_norm": 2.3621318340301514, "learning_rate": 7.392203280454039e-05, "loss": 0.5615414619445801, "memory(GiB)": 70.5, "step": 39820, "token_acc": 0.8697068403908795, "train_speed(iter/s)": 1.453634 }, { "epoch": 1.706225097467975, "grad_norm": 2.8982651233673096, "learning_rate": 7.391612304031585e-05, "loss": 0.39543824195861815, "memory(GiB)": 70.5, "step": 39825, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.453642 }, { "epoch": 1.7064393127972237, "grad_norm": 4.82472038269043, "learning_rate": 7.391021284282833e-05, "loss": 0.6706676483154297, "memory(GiB)": 70.5, "step": 39830, "token_acc": 0.8513513513513513, "train_speed(iter/s)": 1.453636 }, { "epoch": 1.7066535281264728, "grad_norm": 4.396647930145264, "learning_rate": 7.390430221218485e-05, "loss": 0.2966156482696533, "memory(GiB)": 70.5, "step": 39835, "token_acc": 0.9380530973451328, "train_speed(iter/s)": 1.453645 }, { "epoch": 1.7068677434557218, "grad_norm": 4.492462635040283, "learning_rate": 7.389839114849253e-05, "loss": 0.3112152099609375, "memory(GiB)": 70.5, "step": 39840, "token_acc": 0.925, "train_speed(iter/s)": 1.453644 }, { "epoch": 1.7070819587849706, "grad_norm": 4.217214584350586, "learning_rate": 7.389247965185841e-05, "loss": 0.4428996562957764, "memory(GiB)": 70.5, "step": 39845, "token_acc": 0.9113149847094801, "train_speed(iter/s)": 1.45364 }, { "epoch": 1.7072961741142196, "grad_norm": 3.912961721420288, "learning_rate": 7.388656772238964e-05, "loss": 0.5370938777923584, "memory(GiB)": 70.5, "step": 39850, "token_acc": 0.889937106918239, "train_speed(iter/s)": 1.45364 }, { "epoch": 1.7075103894434687, "grad_norm": 3.5605461597442627, "learning_rate": 7.388065536019327e-05, "loss": 0.7068113327026367, "memory(GiB)": 70.5, "step": 39855, "token_acc": 0.8503184713375797, "train_speed(iter/s)": 1.453623 }, { "epoch": 1.7077246047727175, "grad_norm": 1.6122169494628906, "learning_rate": 7.387474256537642e-05, "loss": 0.3290520668029785, "memory(GiB)": 70.5, "step": 39860, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.453624 }, { "epoch": 1.7079388201019665, "grad_norm": 3.3263192176818848, "learning_rate": 7.386882933804621e-05, "loss": 0.5449754238128662, "memory(GiB)": 70.5, "step": 39865, "token_acc": 0.8970588235294118, "train_speed(iter/s)": 1.453619 }, { "epoch": 1.7081530354312155, "grad_norm": 1.114397644996643, "learning_rate": 7.386291567830978e-05, "loss": 0.1982872724533081, "memory(GiB)": 70.5, "step": 39870, "token_acc": 0.9605263157894737, "train_speed(iter/s)": 1.453636 }, { "epoch": 1.7083672507604644, "grad_norm": 2.9174795150756836, "learning_rate": 7.385700158627424e-05, "loss": 0.36777381896972655, "memory(GiB)": 70.5, "step": 39875, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.453632 }, { "epoch": 1.7085814660897134, "grad_norm": 2.2404842376708984, "learning_rate": 7.385108706204675e-05, "loss": 0.578748369216919, "memory(GiB)": 70.5, "step": 39880, "token_acc": 0.8802588996763754, "train_speed(iter/s)": 1.453636 }, { "epoch": 1.7087956814189624, "grad_norm": 5.900612831115723, "learning_rate": 7.384517210573444e-05, "loss": 0.6765610694885253, "memory(GiB)": 70.5, "step": 39885, "token_acc": 0.8633333333333333, "train_speed(iter/s)": 1.45365 }, { "epoch": 1.7090098967482112, "grad_norm": 5.01853084564209, "learning_rate": 7.383925671744446e-05, "loss": 0.33933405876159667, "memory(GiB)": 70.5, "step": 39890, "token_acc": 0.9203980099502488, "train_speed(iter/s)": 1.453681 }, { "epoch": 1.7092241120774603, "grad_norm": 4.088042259216309, "learning_rate": 7.383334089728398e-05, "loss": 0.6002077102661133, "memory(GiB)": 70.5, "step": 39895, "token_acc": 0.8620689655172413, "train_speed(iter/s)": 1.453687 }, { "epoch": 1.7094383274067093, "grad_norm": 1.980559229850769, "learning_rate": 7.382742464536017e-05, "loss": 0.36275572776794435, "memory(GiB)": 70.5, "step": 39900, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.453731 }, { "epoch": 1.7096525427359581, "grad_norm": 2.0722663402557373, "learning_rate": 7.382150796178022e-05, "loss": 0.2503673791885376, "memory(GiB)": 70.5, "step": 39905, "token_acc": 0.9305135951661632, "train_speed(iter/s)": 1.45374 }, { "epoch": 1.7098667580652072, "grad_norm": 2.238386392593384, "learning_rate": 7.38155908466513e-05, "loss": 0.2872618198394775, "memory(GiB)": 70.5, "step": 39910, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.453759 }, { "epoch": 1.7100809733944562, "grad_norm": 2.973771810531616, "learning_rate": 7.380967330008063e-05, "loss": 0.17315187454223632, "memory(GiB)": 70.5, "step": 39915, "token_acc": 0.9530201342281879, "train_speed(iter/s)": 1.453745 }, { "epoch": 1.710295188723705, "grad_norm": 2.130073308944702, "learning_rate": 7.380375532217537e-05, "loss": 0.3906387805938721, "memory(GiB)": 70.5, "step": 39920, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.453743 }, { "epoch": 1.710509404052954, "grad_norm": 7.174804210662842, "learning_rate": 7.379783691304276e-05, "loss": 0.5717191696166992, "memory(GiB)": 70.5, "step": 39925, "token_acc": 0.8593155893536122, "train_speed(iter/s)": 1.45373 }, { "epoch": 1.710723619382203, "grad_norm": 2.804737091064453, "learning_rate": 7.379191807279002e-05, "loss": 0.3589658498764038, "memory(GiB)": 70.5, "step": 39930, "token_acc": 0.91701244813278, "train_speed(iter/s)": 1.453743 }, { "epoch": 1.7109378347114519, "grad_norm": 1.4279381036758423, "learning_rate": 7.378599880152434e-05, "loss": 0.44177865982055664, "memory(GiB)": 70.5, "step": 39935, "token_acc": 0.9046242774566474, "train_speed(iter/s)": 1.453745 }, { "epoch": 1.711152050040701, "grad_norm": 5.142737865447998, "learning_rate": 7.3780079099353e-05, "loss": 0.5519931793212891, "memory(GiB)": 70.5, "step": 39940, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.453738 }, { "epoch": 1.71136626536995, "grad_norm": 1.8219811916351318, "learning_rate": 7.377415896638322e-05, "loss": 0.24969334602355958, "memory(GiB)": 70.5, "step": 39945, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.453733 }, { "epoch": 1.7115804806991988, "grad_norm": 4.7654619216918945, "learning_rate": 7.376823840272224e-05, "loss": 0.7671055793762207, "memory(GiB)": 70.5, "step": 39950, "token_acc": 0.8305555555555556, "train_speed(iter/s)": 1.453728 }, { "epoch": 1.7117946960284478, "grad_norm": 1.3879125118255615, "learning_rate": 7.376231740847734e-05, "loss": 0.1832043170928955, "memory(GiB)": 70.5, "step": 39955, "token_acc": 0.9541284403669725, "train_speed(iter/s)": 1.453742 }, { "epoch": 1.7120089113576968, "grad_norm": 1.6608667373657227, "learning_rate": 7.375639598375574e-05, "loss": 0.3524176120758057, "memory(GiB)": 70.5, "step": 39960, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.45374 }, { "epoch": 1.7122231266869457, "grad_norm": 6.806647300720215, "learning_rate": 7.375047412866476e-05, "loss": 0.3468447208404541, "memory(GiB)": 70.5, "step": 39965, "token_acc": 0.9377431906614786, "train_speed(iter/s)": 1.453738 }, { "epoch": 1.7124373420161947, "grad_norm": 3.251248836517334, "learning_rate": 7.374455184331167e-05, "loss": 0.4272147178649902, "memory(GiB)": 70.5, "step": 39970, "token_acc": 0.9008620689655172, "train_speed(iter/s)": 1.453735 }, { "epoch": 1.7126515573454437, "grad_norm": 4.939101696014404, "learning_rate": 7.373862912780373e-05, "loss": 0.29285976886749265, "memory(GiB)": 70.5, "step": 39975, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.453747 }, { "epoch": 1.7128657726746925, "grad_norm": 3.0331406593322754, "learning_rate": 7.373270598224827e-05, "loss": 0.4797659873962402, "memory(GiB)": 70.5, "step": 39980, "token_acc": 0.9015873015873016, "train_speed(iter/s)": 1.453746 }, { "epoch": 1.7130799880039416, "grad_norm": 4.906414985656738, "learning_rate": 7.372678240675256e-05, "loss": 0.15132291316986085, "memory(GiB)": 70.5, "step": 39985, "token_acc": 0.9691119691119691, "train_speed(iter/s)": 1.453764 }, { "epoch": 1.7132942033331906, "grad_norm": 3.2881481647491455, "learning_rate": 7.372085840142394e-05, "loss": 0.34013614654541013, "memory(GiB)": 70.5, "step": 39990, "token_acc": 0.9415204678362573, "train_speed(iter/s)": 1.453766 }, { "epoch": 1.7135084186624394, "grad_norm": 1.8002043962478638, "learning_rate": 7.371493396636973e-05, "loss": 0.24905917644500733, "memory(GiB)": 70.5, "step": 39995, "token_acc": 0.952247191011236, "train_speed(iter/s)": 1.453769 }, { "epoch": 1.7137226339916884, "grad_norm": 3.570923089981079, "learning_rate": 7.370900910169723e-05, "loss": 0.2509594917297363, "memory(GiB)": 70.5, "step": 40000, "token_acc": 0.9345238095238095, "train_speed(iter/s)": 1.453774 }, { "epoch": 1.7137226339916884, "eval_loss": 2.294095039367676, "eval_runtime": 13.0536, "eval_samples_per_second": 7.661, "eval_steps_per_second": 7.661, "eval_token_acc": 0.4421965317919075, "step": 40000 }, { "epoch": 1.7139368493209375, "grad_norm": 6.318020820617676, "learning_rate": 7.370308380751378e-05, "loss": 0.7445897102355957, "memory(GiB)": 70.5, "step": 40005, "token_acc": 0.5689320388349515, "train_speed(iter/s)": 1.453041 }, { "epoch": 1.7141510646501863, "grad_norm": 1.8453130722045898, "learning_rate": 7.369715808392675e-05, "loss": 0.18822838068008424, "memory(GiB)": 70.5, "step": 40010, "token_acc": 0.9566787003610109, "train_speed(iter/s)": 1.453052 }, { "epoch": 1.7143652799794353, "grad_norm": 7.220891952514648, "learning_rate": 7.369123193104347e-05, "loss": 0.6752798557281494, "memory(GiB)": 70.5, "step": 40015, "token_acc": 0.8847457627118644, "train_speed(iter/s)": 1.453052 }, { "epoch": 1.7145794953086844, "grad_norm": 3.4730148315429688, "learning_rate": 7.368530534897128e-05, "loss": 0.6098117828369141, "memory(GiB)": 70.5, "step": 40020, "token_acc": 0.8625954198473282, "train_speed(iter/s)": 1.45305 }, { "epoch": 1.7147937106379332, "grad_norm": 0.39719921350479126, "learning_rate": 7.36793783378176e-05, "loss": 0.24991161823272706, "memory(GiB)": 70.5, "step": 40025, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.453058 }, { "epoch": 1.7150079259671822, "grad_norm": 3.5211613178253174, "learning_rate": 7.367345089768976e-05, "loss": 0.25710110664367675, "memory(GiB)": 70.5, "step": 40030, "token_acc": 0.9580645161290322, "train_speed(iter/s)": 1.453058 }, { "epoch": 1.7152221412964312, "grad_norm": 2.2936787605285645, "learning_rate": 7.366752302869513e-05, "loss": 0.4603695869445801, "memory(GiB)": 70.5, "step": 40035, "token_acc": 0.8761904761904762, "train_speed(iter/s)": 1.453051 }, { "epoch": 1.71543635662568, "grad_norm": 2.1816041469573975, "learning_rate": 7.366159473094112e-05, "loss": 0.36108386516571045, "memory(GiB)": 70.5, "step": 40040, "token_acc": 0.9400749063670412, "train_speed(iter/s)": 1.453076 }, { "epoch": 1.715650571954929, "grad_norm": 3.6904640197753906, "learning_rate": 7.365566600453513e-05, "loss": 0.3421593427658081, "memory(GiB)": 70.5, "step": 40045, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.45308 }, { "epoch": 1.7158647872841781, "grad_norm": 5.137673377990723, "learning_rate": 7.364973684958456e-05, "loss": 0.5656716346740722, "memory(GiB)": 70.5, "step": 40050, "token_acc": 0.8782287822878229, "train_speed(iter/s)": 1.453065 }, { "epoch": 1.716079002613427, "grad_norm": 1.327527642250061, "learning_rate": 7.364380726619681e-05, "loss": 0.3794901371002197, "memory(GiB)": 70.5, "step": 40055, "token_acc": 0.9112426035502958, "train_speed(iter/s)": 1.453085 }, { "epoch": 1.716293217942676, "grad_norm": 2.97070574760437, "learning_rate": 7.363787725447935e-05, "loss": 0.42485966682434084, "memory(GiB)": 70.5, "step": 40060, "token_acc": 0.9177215189873418, "train_speed(iter/s)": 1.453084 }, { "epoch": 1.716507433271925, "grad_norm": 2.3106179237365723, "learning_rate": 7.363194681453953e-05, "loss": 0.39740848541259766, "memory(GiB)": 70.5, "step": 40065, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.453092 }, { "epoch": 1.7167216486011738, "grad_norm": 0.5247345566749573, "learning_rate": 7.362601594648484e-05, "loss": 0.5045602798461915, "memory(GiB)": 70.5, "step": 40070, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.453109 }, { "epoch": 1.7169358639304229, "grad_norm": 4.344967842102051, "learning_rate": 7.36200846504227e-05, "loss": 0.4804414749145508, "memory(GiB)": 70.5, "step": 40075, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.453111 }, { "epoch": 1.717150079259672, "grad_norm": 4.591266632080078, "learning_rate": 7.361415292646057e-05, "loss": 0.3370544195175171, "memory(GiB)": 70.5, "step": 40080, "token_acc": 0.9314516129032258, "train_speed(iter/s)": 1.453112 }, { "epoch": 1.7173642945889207, "grad_norm": 3.957625150680542, "learning_rate": 7.360822077470592e-05, "loss": 0.598933219909668, "memory(GiB)": 70.5, "step": 40085, "token_acc": 0.8786127167630058, "train_speed(iter/s)": 1.453133 }, { "epoch": 1.7175785099181697, "grad_norm": 3.1562070846557617, "learning_rate": 7.360228819526621e-05, "loss": 0.3233060836791992, "memory(GiB)": 70.5, "step": 40090, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.453139 }, { "epoch": 1.7177927252474188, "grad_norm": 2.9212331771850586, "learning_rate": 7.35963551882489e-05, "loss": 0.3382901906967163, "memory(GiB)": 70.5, "step": 40095, "token_acc": 0.9409448818897638, "train_speed(iter/s)": 1.453139 }, { "epoch": 1.7180069405766676, "grad_norm": 0.7904892563819885, "learning_rate": 7.359042175376148e-05, "loss": 0.29409942626953123, "memory(GiB)": 70.5, "step": 40100, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.453136 }, { "epoch": 1.7182211559059166, "grad_norm": 4.559720516204834, "learning_rate": 7.358448789191142e-05, "loss": 0.5519734859466553, "memory(GiB)": 70.5, "step": 40105, "token_acc": 0.8790849673202614, "train_speed(iter/s)": 1.453129 }, { "epoch": 1.7184353712351657, "grad_norm": 4.2336602210998535, "learning_rate": 7.357855360280626e-05, "loss": 0.20005688667297364, "memory(GiB)": 70.5, "step": 40110, "token_acc": 0.950530035335689, "train_speed(iter/s)": 1.453126 }, { "epoch": 1.7186495865644145, "grad_norm": 0.7095297574996948, "learning_rate": 7.357261888655349e-05, "loss": 0.3711925745010376, "memory(GiB)": 70.5, "step": 40115, "token_acc": 0.9298245614035088, "train_speed(iter/s)": 1.453115 }, { "epoch": 1.7188638018936635, "grad_norm": 2.1976704597473145, "learning_rate": 7.35666837432606e-05, "loss": 0.29005942344665525, "memory(GiB)": 70.5, "step": 40120, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.453127 }, { "epoch": 1.7190780172229125, "grad_norm": 3.3036117553710938, "learning_rate": 7.356074817303513e-05, "loss": 0.39294216632843015, "memory(GiB)": 70.5, "step": 40125, "token_acc": 0.9272030651340997, "train_speed(iter/s)": 1.453137 }, { "epoch": 1.7192922325521613, "grad_norm": 1.949184536933899, "learning_rate": 7.355481217598461e-05, "loss": 0.1665647506713867, "memory(GiB)": 70.5, "step": 40130, "token_acc": 0.9668508287292817, "train_speed(iter/s)": 1.453148 }, { "epoch": 1.7195064478814104, "grad_norm": 4.6507673263549805, "learning_rate": 7.354887575221656e-05, "loss": 0.37160024642944334, "memory(GiB)": 70.5, "step": 40135, "token_acc": 0.9345794392523364, "train_speed(iter/s)": 1.45315 }, { "epoch": 1.7197206632106594, "grad_norm": 4.823300361633301, "learning_rate": 7.354293890183854e-05, "loss": 0.409254264831543, "memory(GiB)": 70.5, "step": 40140, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.453155 }, { "epoch": 1.7199348785399082, "grad_norm": 1.444100260734558, "learning_rate": 7.353700162495811e-05, "loss": 0.3086298704147339, "memory(GiB)": 70.5, "step": 40145, "token_acc": 0.936, "train_speed(iter/s)": 1.453153 }, { "epoch": 1.7201490938691573, "grad_norm": 5.9218668937683105, "learning_rate": 7.35310639216828e-05, "loss": 0.5112557888031006, "memory(GiB)": 70.5, "step": 40150, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.45317 }, { "epoch": 1.7203633091984063, "grad_norm": 2.5478289127349854, "learning_rate": 7.35251257921202e-05, "loss": 0.1282513380050659, "memory(GiB)": 70.5, "step": 40155, "token_acc": 0.9752650176678446, "train_speed(iter/s)": 1.453166 }, { "epoch": 1.7205775245276551, "grad_norm": 4.938320636749268, "learning_rate": 7.351918723637788e-05, "loss": 0.5451516151428223, "memory(GiB)": 70.5, "step": 40160, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.453192 }, { "epoch": 1.7207917398569041, "grad_norm": 6.41141414642334, "learning_rate": 7.35132482545634e-05, "loss": 0.39808359146118166, "memory(GiB)": 70.5, "step": 40165, "token_acc": 0.9141104294478528, "train_speed(iter/s)": 1.453184 }, { "epoch": 1.7210059551861532, "grad_norm": 3.418522357940674, "learning_rate": 7.350730884678441e-05, "loss": 0.6259201526641845, "memory(GiB)": 70.5, "step": 40170, "token_acc": 0.8791946308724832, "train_speed(iter/s)": 1.453194 }, { "epoch": 1.721220170515402, "grad_norm": 4.051334381103516, "learning_rate": 7.350136901314845e-05, "loss": 0.26037890911102296, "memory(GiB)": 70.5, "step": 40175, "token_acc": 0.9498327759197325, "train_speed(iter/s)": 1.453203 }, { "epoch": 1.721434385844651, "grad_norm": 4.875328063964844, "learning_rate": 7.349542875376312e-05, "loss": 0.9502505302429199, "memory(GiB)": 70.5, "step": 40180, "token_acc": 0.8156028368794326, "train_speed(iter/s)": 1.453227 }, { "epoch": 1.7216486011739, "grad_norm": 5.61622428894043, "learning_rate": 7.348948806873609e-05, "loss": 0.44144444465637206, "memory(GiB)": 70.5, "step": 40185, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.453249 }, { "epoch": 1.7218628165031489, "grad_norm": 4.245009899139404, "learning_rate": 7.348354695817493e-05, "loss": 0.16372921466827392, "memory(GiB)": 70.5, "step": 40190, "token_acc": 0.9621212121212122, "train_speed(iter/s)": 1.453247 }, { "epoch": 1.722077031832398, "grad_norm": 3.1303248405456543, "learning_rate": 7.347760542218729e-05, "loss": 0.39597363471984864, "memory(GiB)": 70.5, "step": 40195, "token_acc": 0.9176829268292683, "train_speed(iter/s)": 1.453249 }, { "epoch": 1.722291247161647, "grad_norm": 4.359979629516602, "learning_rate": 7.347166346088081e-05, "loss": 0.879051399230957, "memory(GiB)": 70.5, "step": 40200, "token_acc": 0.8074074074074075, "train_speed(iter/s)": 1.453307 }, { "epoch": 1.7225054624908958, "grad_norm": 0.4837651252746582, "learning_rate": 7.346572107436313e-05, "loss": 0.4395142078399658, "memory(GiB)": 70.5, "step": 40205, "token_acc": 0.9053030303030303, "train_speed(iter/s)": 1.453315 }, { "epoch": 1.7227196778201448, "grad_norm": 2.898780584335327, "learning_rate": 7.345977826274189e-05, "loss": 0.7224379062652588, "memory(GiB)": 70.5, "step": 40210, "token_acc": 0.8878205128205128, "train_speed(iter/s)": 1.453331 }, { "epoch": 1.7229338931493938, "grad_norm": 2.8262717723846436, "learning_rate": 7.345383502612477e-05, "loss": 0.16394083499908446, "memory(GiB)": 70.5, "step": 40215, "token_acc": 0.9492753623188406, "train_speed(iter/s)": 1.45332 }, { "epoch": 1.7231481084786426, "grad_norm": 4.460483551025391, "learning_rate": 7.344789136461942e-05, "loss": 0.4113733768463135, "memory(GiB)": 70.5, "step": 40220, "token_acc": 0.9036544850498339, "train_speed(iter/s)": 1.453318 }, { "epoch": 1.7233623238078917, "grad_norm": 6.099575996398926, "learning_rate": 7.344194727833354e-05, "loss": 0.17746372222900392, "memory(GiB)": 70.5, "step": 40225, "token_acc": 0.9607142857142857, "train_speed(iter/s)": 1.453315 }, { "epoch": 1.7235765391371407, "grad_norm": 2.972721576690674, "learning_rate": 7.343600276737477e-05, "loss": 0.49957799911499023, "memory(GiB)": 70.5, "step": 40230, "token_acc": 0.9071207430340558, "train_speed(iter/s)": 1.453318 }, { "epoch": 1.7237907544663895, "grad_norm": 1.6001838445663452, "learning_rate": 7.343005783185085e-05, "loss": 0.23803446292877198, "memory(GiB)": 70.5, "step": 40235, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.453318 }, { "epoch": 1.7240049697956386, "grad_norm": 5.1008710861206055, "learning_rate": 7.342411247186942e-05, "loss": 0.46133909225463865, "memory(GiB)": 70.5, "step": 40240, "token_acc": 0.8951310861423221, "train_speed(iter/s)": 1.453327 }, { "epoch": 1.7242191851248876, "grad_norm": 5.470424652099609, "learning_rate": 7.341816668753825e-05, "loss": 0.4276398181915283, "memory(GiB)": 70.5, "step": 40245, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.453339 }, { "epoch": 1.7244334004541364, "grad_norm": 2.0748584270477295, "learning_rate": 7.341222047896501e-05, "loss": 0.5194760799407959, "memory(GiB)": 70.5, "step": 40250, "token_acc": 0.8794788273615635, "train_speed(iter/s)": 1.453366 }, { "epoch": 1.7246476157833854, "grad_norm": 1.9538187980651855, "learning_rate": 7.340627384625743e-05, "loss": 0.43518266677856443, "memory(GiB)": 70.5, "step": 40255, "token_acc": 0.9120521172638436, "train_speed(iter/s)": 1.453371 }, { "epoch": 1.7248618311126345, "grad_norm": 6.657334327697754, "learning_rate": 7.340032678952325e-05, "loss": 0.5976600170135498, "memory(GiB)": 70.5, "step": 40260, "token_acc": 0.8990825688073395, "train_speed(iter/s)": 1.453382 }, { "epoch": 1.7250760464418833, "grad_norm": 2.3538126945495605, "learning_rate": 7.339437930887018e-05, "loss": 0.45953822135925293, "memory(GiB)": 70.5, "step": 40265, "token_acc": 0.905982905982906, "train_speed(iter/s)": 1.453405 }, { "epoch": 1.7252902617711323, "grad_norm": 3.4689881801605225, "learning_rate": 7.338843140440601e-05, "loss": 0.2318148136138916, "memory(GiB)": 70.5, "step": 40270, "token_acc": 0.9535714285714286, "train_speed(iter/s)": 1.453413 }, { "epoch": 1.7255044771003814, "grad_norm": 5.176538467407227, "learning_rate": 7.338248307623844e-05, "loss": 0.43714194297790526, "memory(GiB)": 70.5, "step": 40275, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.453403 }, { "epoch": 1.7257186924296302, "grad_norm": 2.829555034637451, "learning_rate": 7.337653432447527e-05, "loss": 0.32476885318756105, "memory(GiB)": 70.5, "step": 40280, "token_acc": 0.9300291545189504, "train_speed(iter/s)": 1.453404 }, { "epoch": 1.7259329077588792, "grad_norm": 8.71096420288086, "learning_rate": 7.337058514922425e-05, "loss": 0.3032036542892456, "memory(GiB)": 70.5, "step": 40285, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.453397 }, { "epoch": 1.7261471230881282, "grad_norm": 1.4137799739837646, "learning_rate": 7.336463555059316e-05, "loss": 0.3239565849304199, "memory(GiB)": 70.5, "step": 40290, "token_acc": 0.9301587301587302, "train_speed(iter/s)": 1.453393 }, { "epoch": 1.726361338417377, "grad_norm": 3.086643934249878, "learning_rate": 7.335868552868979e-05, "loss": 0.22927119731903076, "memory(GiB)": 70.5, "step": 40295, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.453402 }, { "epoch": 1.726575553746626, "grad_norm": 5.989239692687988, "learning_rate": 7.335273508362189e-05, "loss": 0.2359870433807373, "memory(GiB)": 70.5, "step": 40300, "token_acc": 0.952, "train_speed(iter/s)": 1.453404 }, { "epoch": 1.7267897690758751, "grad_norm": 1.2615348100662231, "learning_rate": 7.334678421549731e-05, "loss": 0.5597031116485596, "memory(GiB)": 70.5, "step": 40305, "token_acc": 0.8707482993197279, "train_speed(iter/s)": 1.453429 }, { "epoch": 1.727003984405124, "grad_norm": 2.650519371032715, "learning_rate": 7.334083292442382e-05, "loss": 0.16431413888931273, "memory(GiB)": 70.5, "step": 40310, "token_acc": 0.9611307420494699, "train_speed(iter/s)": 1.453429 }, { "epoch": 1.727218199734373, "grad_norm": 2.8968324661254883, "learning_rate": 7.333488121050925e-05, "loss": 0.47839603424072263, "memory(GiB)": 70.5, "step": 40315, "token_acc": 0.9050632911392406, "train_speed(iter/s)": 1.453445 }, { "epoch": 1.727432415063622, "grad_norm": 5.528570175170898, "learning_rate": 7.332892907386142e-05, "loss": 0.4668393135070801, "memory(GiB)": 70.5, "step": 40320, "token_acc": 0.9105058365758755, "train_speed(iter/s)": 1.453439 }, { "epoch": 1.7276466303928708, "grad_norm": 3.461344003677368, "learning_rate": 7.332297651458815e-05, "loss": 0.3131478548049927, "memory(GiB)": 70.5, "step": 40325, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.453452 }, { "epoch": 1.7278608457221198, "grad_norm": 1.7028754949569702, "learning_rate": 7.33170235327973e-05, "loss": 0.13053436279296876, "memory(GiB)": 70.5, "step": 40330, "token_acc": 0.9787234042553191, "train_speed(iter/s)": 1.453465 }, { "epoch": 1.7280750610513689, "grad_norm": 9.164621353149414, "learning_rate": 7.331107012859667e-05, "loss": 0.5375349521636963, "memory(GiB)": 70.5, "step": 40335, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.453458 }, { "epoch": 1.7282892763806177, "grad_norm": 2.2134335041046143, "learning_rate": 7.330511630209414e-05, "loss": 0.2543428897857666, "memory(GiB)": 70.5, "step": 40340, "token_acc": 0.9504950495049505, "train_speed(iter/s)": 1.453476 }, { "epoch": 1.7285034917098667, "grad_norm": 2.8980443477630615, "learning_rate": 7.329916205339757e-05, "loss": 0.3455937385559082, "memory(GiB)": 70.5, "step": 40345, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.453475 }, { "epoch": 1.7287177070391158, "grad_norm": 2.9655704498291016, "learning_rate": 7.329320738261484e-05, "loss": 0.36616714000701905, "memory(GiB)": 70.5, "step": 40350, "token_acc": 0.921875, "train_speed(iter/s)": 1.453468 }, { "epoch": 1.7289319223683646, "grad_norm": 6.28439998626709, "learning_rate": 7.32872522898538e-05, "loss": 0.30898633003234866, "memory(GiB)": 70.5, "step": 40355, "token_acc": 0.9442231075697212, "train_speed(iter/s)": 1.453492 }, { "epoch": 1.7291461376976136, "grad_norm": 4.036338806152344, "learning_rate": 7.328129677522234e-05, "loss": 0.3040364503860474, "memory(GiB)": 70.5, "step": 40360, "token_acc": 0.9319148936170213, "train_speed(iter/s)": 1.453506 }, { "epoch": 1.7293603530268626, "grad_norm": 4.1497883796691895, "learning_rate": 7.327534083882832e-05, "loss": 0.6322391510009766, "memory(GiB)": 70.5, "step": 40365, "token_acc": 0.8776223776223776, "train_speed(iter/s)": 1.453499 }, { "epoch": 1.7295745683561115, "grad_norm": 12.136799812316895, "learning_rate": 7.32693844807797e-05, "loss": 0.2776843786239624, "memory(GiB)": 70.5, "step": 40370, "token_acc": 0.9338842975206612, "train_speed(iter/s)": 1.453493 }, { "epoch": 1.7297887836853605, "grad_norm": 1.4480987787246704, "learning_rate": 7.326342770118434e-05, "loss": 0.3408064603805542, "memory(GiB)": 70.5, "step": 40375, "token_acc": 0.937007874015748, "train_speed(iter/s)": 1.453492 }, { "epoch": 1.7300029990146095, "grad_norm": 0.5587754249572754, "learning_rate": 7.325747050015017e-05, "loss": 0.25330343246459963, "memory(GiB)": 70.5, "step": 40380, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.453496 }, { "epoch": 1.7302172143438583, "grad_norm": 5.598374843597412, "learning_rate": 7.325151287778509e-05, "loss": 0.48800172805786135, "memory(GiB)": 70.5, "step": 40385, "token_acc": 0.8916666666666667, "train_speed(iter/s)": 1.453489 }, { "epoch": 1.7304314296731074, "grad_norm": 0.8687006831169128, "learning_rate": 7.324555483419707e-05, "loss": 0.3292912244796753, "memory(GiB)": 70.5, "step": 40390, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.453492 }, { "epoch": 1.7306456450023564, "grad_norm": 2.4780118465423584, "learning_rate": 7.323959636949398e-05, "loss": 0.3728486061096191, "memory(GiB)": 70.5, "step": 40395, "token_acc": 0.9201277955271565, "train_speed(iter/s)": 1.453528 }, { "epoch": 1.7308598603316052, "grad_norm": 4.887020587921143, "learning_rate": 7.323363748378384e-05, "loss": 0.6591782569885254, "memory(GiB)": 70.5, "step": 40400, "token_acc": 0.8697318007662835, "train_speed(iter/s)": 1.453527 }, { "epoch": 1.7310740756608543, "grad_norm": 6.556215763092041, "learning_rate": 7.322767817717454e-05, "loss": 0.30317959785461424, "memory(GiB)": 70.5, "step": 40405, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.453544 }, { "epoch": 1.7312882909901033, "grad_norm": 2.807382106781006, "learning_rate": 7.322171844977407e-05, "loss": 0.25700578689575193, "memory(GiB)": 70.5, "step": 40410, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.453542 }, { "epoch": 1.731502506319352, "grad_norm": 4.6388702392578125, "learning_rate": 7.321575830169037e-05, "loss": 0.3501457929611206, "memory(GiB)": 70.5, "step": 40415, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.453567 }, { "epoch": 1.7317167216486011, "grad_norm": 3.61609148979187, "learning_rate": 7.320979773303144e-05, "loss": 0.3680567264556885, "memory(GiB)": 70.5, "step": 40420, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.453563 }, { "epoch": 1.7319309369778502, "grad_norm": 4.854184150695801, "learning_rate": 7.320383674390525e-05, "loss": 0.2546211242675781, "memory(GiB)": 70.5, "step": 40425, "token_acc": 0.9341317365269461, "train_speed(iter/s)": 1.453564 }, { "epoch": 1.732145152307099, "grad_norm": 1.7105958461761475, "learning_rate": 7.319787533441981e-05, "loss": 0.22917232513427735, "memory(GiB)": 70.5, "step": 40430, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.453575 }, { "epoch": 1.732359367636348, "grad_norm": 2.8602757453918457, "learning_rate": 7.319191350468308e-05, "loss": 0.5305458068847656, "memory(GiB)": 70.5, "step": 40435, "token_acc": 0.8987854251012146, "train_speed(iter/s)": 1.453581 }, { "epoch": 1.732573582965597, "grad_norm": 3.025681257247925, "learning_rate": 7.318595125480308e-05, "loss": 0.33367042541503905, "memory(GiB)": 70.5, "step": 40440, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.453579 }, { "epoch": 1.7327877982948459, "grad_norm": 1.4105027914047241, "learning_rate": 7.317998858488783e-05, "loss": 0.4479053020477295, "memory(GiB)": 70.5, "step": 40445, "token_acc": 0.9058823529411765, "train_speed(iter/s)": 1.453562 }, { "epoch": 1.733002013624095, "grad_norm": 6.473011493682861, "learning_rate": 7.317402549504533e-05, "loss": 0.6790610313415527, "memory(GiB)": 70.5, "step": 40450, "token_acc": 0.8204419889502762, "train_speed(iter/s)": 1.453552 }, { "epoch": 1.733216228953344, "grad_norm": 4.027694225311279, "learning_rate": 7.316806198538362e-05, "loss": 0.24476776123046876, "memory(GiB)": 70.5, "step": 40455, "token_acc": 0.9543973941368078, "train_speed(iter/s)": 1.453553 }, { "epoch": 1.7334304442825927, "grad_norm": 8.912589073181152, "learning_rate": 7.316209805601074e-05, "loss": 0.5841786384582519, "memory(GiB)": 70.5, "step": 40460, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.453552 }, { "epoch": 1.7336446596118418, "grad_norm": 3.0817153453826904, "learning_rate": 7.315613370703473e-05, "loss": 0.4031072616577148, "memory(GiB)": 70.5, "step": 40465, "token_acc": 0.9122257053291536, "train_speed(iter/s)": 1.45357 }, { "epoch": 1.7338588749410908, "grad_norm": 3.7713842391967773, "learning_rate": 7.315016893856363e-05, "loss": 0.22853851318359375, "memory(GiB)": 70.5, "step": 40470, "token_acc": 0.9341692789968652, "train_speed(iter/s)": 1.453565 }, { "epoch": 1.7340730902703396, "grad_norm": 1.3660703897476196, "learning_rate": 7.31442037507055e-05, "loss": 0.2651691436767578, "memory(GiB)": 70.5, "step": 40475, "token_acc": 0.9383116883116883, "train_speed(iter/s)": 1.45357 }, { "epoch": 1.7342873055995887, "grad_norm": 3.8537590503692627, "learning_rate": 7.313823814356841e-05, "loss": 0.3478795051574707, "memory(GiB)": 70.5, "step": 40480, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.453567 }, { "epoch": 1.7345015209288377, "grad_norm": 1.9890464544296265, "learning_rate": 7.313227211726044e-05, "loss": 0.3220844745635986, "memory(GiB)": 70.5, "step": 40485, "token_acc": 0.9346938775510204, "train_speed(iter/s)": 1.453566 }, { "epoch": 1.7347157362580865, "grad_norm": 1.706761121749878, "learning_rate": 7.312630567188965e-05, "loss": 0.41495494842529296, "memory(GiB)": 70.5, "step": 40490, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.453574 }, { "epoch": 1.7349299515873355, "grad_norm": 1.9631283283233643, "learning_rate": 7.312033880756415e-05, "loss": 0.3844139575958252, "memory(GiB)": 70.5, "step": 40495, "token_acc": 0.9172932330827067, "train_speed(iter/s)": 1.453576 }, { "epoch": 1.7351441669165846, "grad_norm": 3.704627513885498, "learning_rate": 7.311437152439201e-05, "loss": 0.5987714290618896, "memory(GiB)": 70.5, "step": 40500, "token_acc": 0.8659420289855072, "train_speed(iter/s)": 1.453599 }, { "epoch": 1.7351441669165846, "eval_loss": 2.33716082572937, "eval_runtime": 13.5146, "eval_samples_per_second": 7.399, "eval_steps_per_second": 7.399, "eval_token_acc": 0.4471428571428571, "step": 40500 }, { "epoch": 1.7353583822458334, "grad_norm": 0.07064919173717499, "learning_rate": 7.310840382248136e-05, "loss": 0.26027989387512207, "memory(GiB)": 70.5, "step": 40505, "token_acc": 0.5966135458167331, "train_speed(iter/s)": 1.452866 }, { "epoch": 1.7355725975750824, "grad_norm": 2.661994457244873, "learning_rate": 7.310243570194029e-05, "loss": 0.5104427814483643, "memory(GiB)": 70.5, "step": 40510, "token_acc": 0.8831168831168831, "train_speed(iter/s)": 1.452905 }, { "epoch": 1.7357868129043315, "grad_norm": 10.782179832458496, "learning_rate": 7.309646716287692e-05, "loss": 0.5881637573242188, "memory(GiB)": 70.5, "step": 40515, "token_acc": 0.8856209150326797, "train_speed(iter/s)": 1.452902 }, { "epoch": 1.7360010282335803, "grad_norm": 1.6604477167129517, "learning_rate": 7.30904982053994e-05, "loss": 0.36104109287261965, "memory(GiB)": 70.5, "step": 40520, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.452898 }, { "epoch": 1.7362152435628293, "grad_norm": 8.41551399230957, "learning_rate": 7.308452882961582e-05, "loss": 0.36159591674804686, "memory(GiB)": 70.5, "step": 40525, "token_acc": 0.9288702928870293, "train_speed(iter/s)": 1.45292 }, { "epoch": 1.7364294588920783, "grad_norm": 4.889471054077148, "learning_rate": 7.307855903563435e-05, "loss": 0.3300872802734375, "memory(GiB)": 70.5, "step": 40530, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.452918 }, { "epoch": 1.7366436742213271, "grad_norm": 4.843908786773682, "learning_rate": 7.307258882356314e-05, "loss": 0.40898666381835935, "memory(GiB)": 70.5, "step": 40535, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.452917 }, { "epoch": 1.7368578895505762, "grad_norm": 4.538218021392822, "learning_rate": 7.306661819351033e-05, "loss": 0.4643255233764648, "memory(GiB)": 70.5, "step": 40540, "token_acc": 0.8954703832752613, "train_speed(iter/s)": 1.452915 }, { "epoch": 1.7370721048798252, "grad_norm": 1.6358081102371216, "learning_rate": 7.306064714558412e-05, "loss": 0.15880756378173827, "memory(GiB)": 70.5, "step": 40545, "token_acc": 0.9562043795620438, "train_speed(iter/s)": 1.452918 }, { "epoch": 1.737286320209074, "grad_norm": 1.8680380582809448, "learning_rate": 7.305467567989263e-05, "loss": 0.2928775310516357, "memory(GiB)": 70.5, "step": 40550, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.452921 }, { "epoch": 1.7375005355383233, "grad_norm": 3.192169427871704, "learning_rate": 7.304870379654407e-05, "loss": 0.564995002746582, "memory(GiB)": 70.5, "step": 40555, "token_acc": 0.8753993610223643, "train_speed(iter/s)": 1.452944 }, { "epoch": 1.737714750867572, "grad_norm": 5.397185325622559, "learning_rate": 7.304273149564662e-05, "loss": 0.43779377937316893, "memory(GiB)": 70.5, "step": 40560, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.452951 }, { "epoch": 1.737928966196821, "grad_norm": 0.1232997328042984, "learning_rate": 7.303675877730848e-05, "loss": 0.17093877792358397, "memory(GiB)": 70.5, "step": 40565, "token_acc": 0.9769452449567724, "train_speed(iter/s)": 1.452952 }, { "epoch": 1.7381431815260702, "grad_norm": 5.2372589111328125, "learning_rate": 7.303078564163783e-05, "loss": 0.41280155181884765, "memory(GiB)": 70.5, "step": 40570, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.452968 }, { "epoch": 1.738357396855319, "grad_norm": 6.204864501953125, "learning_rate": 7.30248120887429e-05, "loss": 0.4576852321624756, "memory(GiB)": 70.5, "step": 40575, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.452978 }, { "epoch": 1.7385716121845678, "grad_norm": 6.345576763153076, "learning_rate": 7.301883811873191e-05, "loss": 0.881580924987793, "memory(GiB)": 70.5, "step": 40580, "token_acc": 0.8249158249158249, "train_speed(iter/s)": 1.452981 }, { "epoch": 1.738785827513817, "grad_norm": 3.455521821975708, "learning_rate": 7.301286373171305e-05, "loss": 0.2755561351776123, "memory(GiB)": 70.5, "step": 40585, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.452969 }, { "epoch": 1.7390000428430659, "grad_norm": 5.835824489593506, "learning_rate": 7.30068889277946e-05, "loss": 0.3859124183654785, "memory(GiB)": 70.5, "step": 40590, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.452975 }, { "epoch": 1.7392142581723147, "grad_norm": 4.032382965087891, "learning_rate": 7.300091370708476e-05, "loss": 0.46831231117248534, "memory(GiB)": 70.5, "step": 40595, "token_acc": 0.8779661016949153, "train_speed(iter/s)": 1.452983 }, { "epoch": 1.739428473501564, "grad_norm": 2.330599784851074, "learning_rate": 7.29949380696918e-05, "loss": 0.4776790142059326, "memory(GiB)": 70.5, "step": 40600, "token_acc": 0.9020771513353115, "train_speed(iter/s)": 1.452986 }, { "epoch": 1.7396426888308127, "grad_norm": 7.4780497550964355, "learning_rate": 7.298896201572396e-05, "loss": 0.4204780101776123, "memory(GiB)": 70.5, "step": 40605, "token_acc": 0.9034749034749034, "train_speed(iter/s)": 1.452998 }, { "epoch": 1.7398569041600616, "grad_norm": 6.385276794433594, "learning_rate": 7.298298554528952e-05, "loss": 0.24161434173583984, "memory(GiB)": 70.5, "step": 40610, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.452996 }, { "epoch": 1.7400711194893108, "grad_norm": 7.364758014678955, "learning_rate": 7.297700865849672e-05, "loss": 0.4819455146789551, "memory(GiB)": 70.5, "step": 40615, "token_acc": 0.8913738019169329, "train_speed(iter/s)": 1.452993 }, { "epoch": 1.7402853348185596, "grad_norm": 4.143052101135254, "learning_rate": 7.297103135545387e-05, "loss": 0.262390661239624, "memory(GiB)": 70.5, "step": 40620, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.452996 }, { "epoch": 1.7404995501478084, "grad_norm": 2.0601418018341064, "learning_rate": 7.296505363626922e-05, "loss": 0.21121621131896973, "memory(GiB)": 70.5, "step": 40625, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.452988 }, { "epoch": 1.7407137654770577, "grad_norm": 0.2994835078716278, "learning_rate": 7.29590755010511e-05, "loss": 0.4806956768035889, "memory(GiB)": 70.5, "step": 40630, "token_acc": 0.907185628742515, "train_speed(iter/s)": 1.452986 }, { "epoch": 1.7409279808063065, "grad_norm": 4.082507610321045, "learning_rate": 7.295309694990777e-05, "loss": 0.30594072341918943, "memory(GiB)": 70.5, "step": 40635, "token_acc": 0.9343629343629344, "train_speed(iter/s)": 1.45298 }, { "epoch": 1.7411421961355553, "grad_norm": 0.960518479347229, "learning_rate": 7.294711798294758e-05, "loss": 0.34610891342163086, "memory(GiB)": 70.5, "step": 40640, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.452977 }, { "epoch": 1.7413564114648046, "grad_norm": 4.852276802062988, "learning_rate": 7.29411386002788e-05, "loss": 0.40949373245239257, "memory(GiB)": 70.5, "step": 40645, "token_acc": 0.9224489795918367, "train_speed(iter/s)": 1.452988 }, { "epoch": 1.7415706267940534, "grad_norm": 3.09836483001709, "learning_rate": 7.293515880200978e-05, "loss": 0.5801135063171386, "memory(GiB)": 70.5, "step": 40650, "token_acc": 0.869281045751634, "train_speed(iter/s)": 1.452983 }, { "epoch": 1.7417848421233022, "grad_norm": 3.351820945739746, "learning_rate": 7.292917858824885e-05, "loss": 0.31097328662872314, "memory(GiB)": 70.5, "step": 40655, "token_acc": 0.9323843416370107, "train_speed(iter/s)": 1.452994 }, { "epoch": 1.7419990574525515, "grad_norm": 3.29121994972229, "learning_rate": 7.292319795910434e-05, "loss": 0.3393620729446411, "memory(GiB)": 70.5, "step": 40660, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.452991 }, { "epoch": 1.7422132727818003, "grad_norm": 4.875582218170166, "learning_rate": 7.29172169146846e-05, "loss": 0.3006559133529663, "memory(GiB)": 70.5, "step": 40665, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.453008 }, { "epoch": 1.742427488111049, "grad_norm": 3.0932137966156006, "learning_rate": 7.291123545509796e-05, "loss": 0.26564242839813235, "memory(GiB)": 70.5, "step": 40670, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.453013 }, { "epoch": 1.7426417034402983, "grad_norm": 3.8461363315582275, "learning_rate": 7.290525358045279e-05, "loss": 0.5034307479858399, "memory(GiB)": 70.5, "step": 40675, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.453011 }, { "epoch": 1.7428559187695472, "grad_norm": 2.4869813919067383, "learning_rate": 7.289927129085749e-05, "loss": 0.42221660614013673, "memory(GiB)": 70.5, "step": 40680, "token_acc": 0.908745247148289, "train_speed(iter/s)": 1.453027 }, { "epoch": 1.743070134098796, "grad_norm": 4.8197150230407715, "learning_rate": 7.28932885864204e-05, "loss": 0.2974228382110596, "memory(GiB)": 70.5, "step": 40685, "token_acc": 0.928125, "train_speed(iter/s)": 1.453037 }, { "epoch": 1.7432843494280452, "grad_norm": 0.1921384334564209, "learning_rate": 7.28873054672499e-05, "loss": 0.43068294525146483, "memory(GiB)": 70.5, "step": 40690, "token_acc": 0.9198717948717948, "train_speed(iter/s)": 1.453038 }, { "epoch": 1.743498564757294, "grad_norm": 3.450421094894409, "learning_rate": 7.288132193345443e-05, "loss": 0.28315718173980714, "memory(GiB)": 70.5, "step": 40695, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.453048 }, { "epoch": 1.7437127800865428, "grad_norm": 3.3086929321289062, "learning_rate": 7.287533798514231e-05, "loss": 0.40976772308349607, "memory(GiB)": 70.5, "step": 40700, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.453055 }, { "epoch": 1.743926995415792, "grad_norm": 5.562314510345459, "learning_rate": 7.2869353622422e-05, "loss": 0.33507046699523924, "memory(GiB)": 70.5, "step": 40705, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.453081 }, { "epoch": 1.744141210745041, "grad_norm": 4.698585510253906, "learning_rate": 7.286336884540189e-05, "loss": 0.3638323783874512, "memory(GiB)": 70.5, "step": 40710, "token_acc": 0.9461279461279462, "train_speed(iter/s)": 1.453086 }, { "epoch": 1.7443554260742897, "grad_norm": 0.099254310131073, "learning_rate": 7.28573836541904e-05, "loss": 0.24344930648803711, "memory(GiB)": 70.5, "step": 40715, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.453087 }, { "epoch": 1.744569641403539, "grad_norm": 1.6918498277664185, "learning_rate": 7.285139804889598e-05, "loss": 0.2743523359298706, "memory(GiB)": 70.5, "step": 40720, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.453086 }, { "epoch": 1.7447838567327878, "grad_norm": 5.153543949127197, "learning_rate": 7.284541202962704e-05, "loss": 0.6083207607269288, "memory(GiB)": 70.5, "step": 40725, "token_acc": 0.8582089552238806, "train_speed(iter/s)": 1.453086 }, { "epoch": 1.7449980720620366, "grad_norm": 0.4476706087589264, "learning_rate": 7.283942559649202e-05, "loss": 0.7240954399108886, "memory(GiB)": 70.5, "step": 40730, "token_acc": 0.832258064516129, "train_speed(iter/s)": 1.453085 }, { "epoch": 1.7452122873912859, "grad_norm": 2.9159669876098633, "learning_rate": 7.283343874959941e-05, "loss": 0.42104291915893555, "memory(GiB)": 70.5, "step": 40735, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.453082 }, { "epoch": 1.7454265027205347, "grad_norm": 1.4918889999389648, "learning_rate": 7.282745148905759e-05, "loss": 0.22633006572723388, "memory(GiB)": 70.5, "step": 40740, "token_acc": 0.9571984435797666, "train_speed(iter/s)": 1.453092 }, { "epoch": 1.7456407180497835, "grad_norm": 9.360455513000488, "learning_rate": 7.28214638149751e-05, "loss": 0.8529887199401855, "memory(GiB)": 70.5, "step": 40745, "token_acc": 0.8152492668621701, "train_speed(iter/s)": 1.453086 }, { "epoch": 1.7458549333790327, "grad_norm": 4.623678684234619, "learning_rate": 7.28154757274604e-05, "loss": 0.3026554107666016, "memory(GiB)": 70.5, "step": 40750, "token_acc": 0.9288702928870293, "train_speed(iter/s)": 1.453091 }, { "epoch": 1.7460691487082816, "grad_norm": 2.1090574264526367, "learning_rate": 7.280948722662194e-05, "loss": 0.4811709880828857, "memory(GiB)": 70.5, "step": 40755, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.453109 }, { "epoch": 1.7462833640375304, "grad_norm": 3.807640314102173, "learning_rate": 7.280349831256821e-05, "loss": 0.4716053009033203, "memory(GiB)": 70.5, "step": 40760, "token_acc": 0.9161073825503355, "train_speed(iter/s)": 1.453111 }, { "epoch": 1.7464975793667796, "grad_norm": 2.35546612739563, "learning_rate": 7.279750898540774e-05, "loss": 0.43348302841186526, "memory(GiB)": 70.5, "step": 40765, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.453112 }, { "epoch": 1.7467117946960284, "grad_norm": 4.922080039978027, "learning_rate": 7.279151924524899e-05, "loss": 0.4353238582611084, "memory(GiB)": 70.5, "step": 40770, "token_acc": 0.9042553191489362, "train_speed(iter/s)": 1.453113 }, { "epoch": 1.7469260100252773, "grad_norm": 3.4945731163024902, "learning_rate": 7.278552909220049e-05, "loss": 0.25947568416595457, "memory(GiB)": 70.5, "step": 40775, "token_acc": 0.9420731707317073, "train_speed(iter/s)": 1.453112 }, { "epoch": 1.7471402253545265, "grad_norm": 5.045832633972168, "learning_rate": 7.277953852637076e-05, "loss": 0.41231064796447753, "memory(GiB)": 70.5, "step": 40780, "token_acc": 0.9226361031518625, "train_speed(iter/s)": 1.453116 }, { "epoch": 1.7473544406837753, "grad_norm": 4.837140083312988, "learning_rate": 7.277354754786832e-05, "loss": 0.415160608291626, "memory(GiB)": 70.5, "step": 40785, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.453118 }, { "epoch": 1.7475686560130241, "grad_norm": 2.2664923667907715, "learning_rate": 7.27675561568017e-05, "loss": 0.35213661193847656, "memory(GiB)": 70.5, "step": 40790, "token_acc": 0.9124579124579124, "train_speed(iter/s)": 1.453106 }, { "epoch": 1.7477828713422734, "grad_norm": 1.8868213891983032, "learning_rate": 7.276156435327946e-05, "loss": 0.27673959732055664, "memory(GiB)": 70.5, "step": 40795, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.453118 }, { "epoch": 1.7479970866715222, "grad_norm": 2.552107572555542, "learning_rate": 7.27555721374101e-05, "loss": 0.6800847530364991, "memory(GiB)": 70.5, "step": 40800, "token_acc": 0.8576779026217228, "train_speed(iter/s)": 1.453123 }, { "epoch": 1.748211302000771, "grad_norm": 6.529347896575928, "learning_rate": 7.274957950930223e-05, "loss": 0.38958823680877686, "memory(GiB)": 70.5, "step": 40805, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.453137 }, { "epoch": 1.7484255173300203, "grad_norm": 2.7900595664978027, "learning_rate": 7.274358646906438e-05, "loss": 0.28809499740600586, "memory(GiB)": 70.5, "step": 40810, "token_acc": 0.9455782312925171, "train_speed(iter/s)": 1.453133 }, { "epoch": 1.748639732659269, "grad_norm": 1.6108241081237793, "learning_rate": 7.273759301680511e-05, "loss": 0.3899667501449585, "memory(GiB)": 70.5, "step": 40815, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.453155 }, { "epoch": 1.748853947988518, "grad_norm": 0.10849565267562866, "learning_rate": 7.273159915263303e-05, "loss": 0.29034316539764404, "memory(GiB)": 70.5, "step": 40820, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.453158 }, { "epoch": 1.7490681633177672, "grad_norm": 2.147810220718384, "learning_rate": 7.27256048766567e-05, "loss": 0.42934865951538087, "memory(GiB)": 70.5, "step": 40825, "token_acc": 0.9036144578313253, "train_speed(iter/s)": 1.453154 }, { "epoch": 1.749282378647016, "grad_norm": 3.0041043758392334, "learning_rate": 7.271961018898473e-05, "loss": 0.4525859832763672, "memory(GiB)": 70.5, "step": 40830, "token_acc": 0.9116719242902208, "train_speed(iter/s)": 1.453135 }, { "epoch": 1.7494965939762648, "grad_norm": 2.9597883224487305, "learning_rate": 7.27136150897257e-05, "loss": 0.42850165367126464, "memory(GiB)": 70.5, "step": 40835, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.453149 }, { "epoch": 1.749710809305514, "grad_norm": 2.5466487407684326, "learning_rate": 7.270761957898823e-05, "loss": 0.28571457862854005, "memory(GiB)": 70.5, "step": 40840, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.453152 }, { "epoch": 1.7499250246347628, "grad_norm": 1.4057806730270386, "learning_rate": 7.270162365688092e-05, "loss": 0.269247841835022, "memory(GiB)": 70.5, "step": 40845, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.45315 }, { "epoch": 1.7501392399640117, "grad_norm": 1.6270804405212402, "learning_rate": 7.269562732351241e-05, "loss": 0.23981738090515137, "memory(GiB)": 70.5, "step": 40850, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.45316 }, { "epoch": 1.750353455293261, "grad_norm": 1.473235845565796, "learning_rate": 7.268963057899132e-05, "loss": 0.26083710193634035, "memory(GiB)": 70.5, "step": 40855, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.453164 }, { "epoch": 1.7505676706225097, "grad_norm": 4.992797374725342, "learning_rate": 7.268363342342628e-05, "loss": 0.43131556510925295, "memory(GiB)": 70.5, "step": 40860, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.453164 }, { "epoch": 1.7507818859517585, "grad_norm": 2.9769480228424072, "learning_rate": 7.267763585692595e-05, "loss": 0.5087237358093262, "memory(GiB)": 70.5, "step": 40865, "token_acc": 0.8793650793650793, "train_speed(iter/s)": 1.453178 }, { "epoch": 1.7509961012810078, "grad_norm": 2.3940391540527344, "learning_rate": 7.267163787959898e-05, "loss": 0.30297625064849854, "memory(GiB)": 70.5, "step": 40870, "token_acc": 0.9423728813559322, "train_speed(iter/s)": 1.453182 }, { "epoch": 1.7512103166102566, "grad_norm": 4.439852714538574, "learning_rate": 7.2665639491554e-05, "loss": 0.39026408195495604, "memory(GiB)": 70.5, "step": 40875, "token_acc": 0.9337748344370861, "train_speed(iter/s)": 1.453192 }, { "epoch": 1.7514245319395054, "grad_norm": 4.124843120574951, "learning_rate": 7.265964069289972e-05, "loss": 0.3926650047302246, "memory(GiB)": 70.5, "step": 40880, "token_acc": 0.908745247148289, "train_speed(iter/s)": 1.453209 }, { "epoch": 1.7516387472687547, "grad_norm": 4.541670322418213, "learning_rate": 7.265364148374478e-05, "loss": 0.35043118000030515, "memory(GiB)": 70.5, "step": 40885, "token_acc": 0.9375, "train_speed(iter/s)": 1.453206 }, { "epoch": 1.7518529625980035, "grad_norm": 4.933161735534668, "learning_rate": 7.264764186419788e-05, "loss": 0.5608706951141358, "memory(GiB)": 70.5, "step": 40890, "token_acc": 0.8745762711864407, "train_speed(iter/s)": 1.453226 }, { "epoch": 1.7520671779272525, "grad_norm": 6.352331638336182, "learning_rate": 7.26416418343677e-05, "loss": 0.5528499126434326, "memory(GiB)": 70.5, "step": 40895, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.453232 }, { "epoch": 1.7522813932565016, "grad_norm": 2.2126429080963135, "learning_rate": 7.263564139436294e-05, "loss": 0.3685596942901611, "memory(GiB)": 70.5, "step": 40900, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.45324 }, { "epoch": 1.7524956085857504, "grad_norm": 2.434467077255249, "learning_rate": 7.26296405442923e-05, "loss": 0.4165768623352051, "memory(GiB)": 70.5, "step": 40905, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.453248 }, { "epoch": 1.7527098239149994, "grad_norm": 3.7179665565490723, "learning_rate": 7.26236392842645e-05, "loss": 0.4770782470703125, "memory(GiB)": 70.5, "step": 40910, "token_acc": 0.9105960264900662, "train_speed(iter/s)": 1.453256 }, { "epoch": 1.7529240392442484, "grad_norm": 4.1798481941223145, "learning_rate": 7.261763761438824e-05, "loss": 0.28473820686340334, "memory(GiB)": 70.5, "step": 40915, "token_acc": 0.9453376205787781, "train_speed(iter/s)": 1.453259 }, { "epoch": 1.7531382545734973, "grad_norm": 5.098208427429199, "learning_rate": 7.261163553477226e-05, "loss": 0.5231746673583985, "memory(GiB)": 70.5, "step": 40920, "token_acc": 0.8704318936877077, "train_speed(iter/s)": 1.453263 }, { "epoch": 1.7533524699027463, "grad_norm": 2.399217367172241, "learning_rate": 7.26056330455253e-05, "loss": 0.3960615634918213, "memory(GiB)": 70.5, "step": 40925, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.453277 }, { "epoch": 1.7535666852319953, "grad_norm": 1.1981902122497559, "learning_rate": 7.259963014675608e-05, "loss": 0.3803962230682373, "memory(GiB)": 70.5, "step": 40930, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.45329 }, { "epoch": 1.7537809005612441, "grad_norm": 3.081566095352173, "learning_rate": 7.259362683857336e-05, "loss": 0.26850271224975586, "memory(GiB)": 70.5, "step": 40935, "token_acc": 0.94, "train_speed(iter/s)": 1.453288 }, { "epoch": 1.7539951158904932, "grad_norm": 2.4820210933685303, "learning_rate": 7.258762312108591e-05, "loss": 0.39460906982421873, "memory(GiB)": 70.5, "step": 40940, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.453295 }, { "epoch": 1.7542093312197422, "grad_norm": 5.445715427398682, "learning_rate": 7.258161899440246e-05, "loss": 0.45780577659606936, "memory(GiB)": 70.5, "step": 40945, "token_acc": 0.89568345323741, "train_speed(iter/s)": 1.453299 }, { "epoch": 1.754423546548991, "grad_norm": 2.9772329330444336, "learning_rate": 7.257561445863182e-05, "loss": 0.3631184816360474, "memory(GiB)": 70.5, "step": 40950, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.453307 }, { "epoch": 1.75463776187824, "grad_norm": 11.598746299743652, "learning_rate": 7.256960951388274e-05, "loss": 0.5940025329589844, "memory(GiB)": 70.5, "step": 40955, "token_acc": 0.8727272727272727, "train_speed(iter/s)": 1.45334 }, { "epoch": 1.754851977207489, "grad_norm": 0.49305713176727295, "learning_rate": 7.2563604160264e-05, "loss": 0.2509725332260132, "memory(GiB)": 70.5, "step": 40960, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.45335 }, { "epoch": 1.755066192536738, "grad_norm": 3.2446513175964355, "learning_rate": 7.25575983978844e-05, "loss": 0.4974111557006836, "memory(GiB)": 70.5, "step": 40965, "token_acc": 0.8858024691358025, "train_speed(iter/s)": 1.453348 }, { "epoch": 1.755280407865987, "grad_norm": 4.666252613067627, "learning_rate": 7.255159222685277e-05, "loss": 0.5157023429870605, "memory(GiB)": 70.5, "step": 40970, "token_acc": 0.887719298245614, "train_speed(iter/s)": 1.453358 }, { "epoch": 1.755494623195236, "grad_norm": 0.7634987235069275, "learning_rate": 7.254558564727786e-05, "loss": 0.1651094913482666, "memory(GiB)": 70.5, "step": 40975, "token_acc": 0.9607843137254902, "train_speed(iter/s)": 1.453359 }, { "epoch": 1.7557088385244848, "grad_norm": 5.184378623962402, "learning_rate": 7.253957865926854e-05, "loss": 0.4339774131774902, "memory(GiB)": 70.5, "step": 40980, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.453365 }, { "epoch": 1.7559230538537338, "grad_norm": 2.9403975009918213, "learning_rate": 7.253357126293361e-05, "loss": 0.42200322151184083, "memory(GiB)": 70.5, "step": 40985, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.45337 }, { "epoch": 1.7561372691829829, "grad_norm": 1.2198090553283691, "learning_rate": 7.252756345838187e-05, "loss": 0.31290631294250487, "memory(GiB)": 70.5, "step": 40990, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.453384 }, { "epoch": 1.7563514845122317, "grad_norm": 5.100419044494629, "learning_rate": 7.252155524572222e-05, "loss": 0.4370081424713135, "memory(GiB)": 70.5, "step": 40995, "token_acc": 0.8996138996138996, "train_speed(iter/s)": 1.453397 }, { "epoch": 1.7565656998414807, "grad_norm": 6.65383768081665, "learning_rate": 7.251554662506346e-05, "loss": 0.5046916484832764, "memory(GiB)": 70.5, "step": 41000, "token_acc": 0.901060070671378, "train_speed(iter/s)": 1.453408 }, { "epoch": 1.7565656998414807, "eval_loss": 2.603668451309204, "eval_runtime": 13.3284, "eval_samples_per_second": 7.503, "eval_steps_per_second": 7.503, "eval_token_acc": 0.4312977099236641, "step": 41000 }, { "epoch": 1.7567799151707297, "grad_norm": 3.31339955329895, "learning_rate": 7.250953759651443e-05, "loss": 0.3109133720397949, "memory(GiB)": 70.5, "step": 41005, "token_acc": 0.572744014732965, "train_speed(iter/s)": 1.452661 }, { "epoch": 1.7569941304999785, "grad_norm": 2.4698262214660645, "learning_rate": 7.250352816018402e-05, "loss": 0.35180437564849854, "memory(GiB)": 70.5, "step": 41010, "token_acc": 0.9300291545189504, "train_speed(iter/s)": 1.452661 }, { "epoch": 1.7572083458292276, "grad_norm": 3.29483962059021, "learning_rate": 7.24975183161811e-05, "loss": 0.35736660957336425, "memory(GiB)": 70.5, "step": 41015, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.45267 }, { "epoch": 1.7574225611584766, "grad_norm": 1.4938533306121826, "learning_rate": 7.249150806461453e-05, "loss": 0.30640997886657717, "memory(GiB)": 70.5, "step": 41020, "token_acc": 0.9397590361445783, "train_speed(iter/s)": 1.452679 }, { "epoch": 1.7576367764877254, "grad_norm": 6.070269584655762, "learning_rate": 7.248549740559319e-05, "loss": 0.31871671676635743, "memory(GiB)": 70.5, "step": 41025, "token_acc": 0.9390681003584229, "train_speed(iter/s)": 1.452691 }, { "epoch": 1.7578509918169745, "grad_norm": 3.522376775741577, "learning_rate": 7.247948633922597e-05, "loss": 0.1601046919822693, "memory(GiB)": 70.5, "step": 41030, "token_acc": 0.96, "train_speed(iter/s)": 1.452701 }, { "epoch": 1.7580652071462235, "grad_norm": 2.3011045455932617, "learning_rate": 7.247347486562177e-05, "loss": 0.3048996925354004, "memory(GiB)": 70.5, "step": 41035, "token_acc": 0.9446254071661238, "train_speed(iter/s)": 1.452704 }, { "epoch": 1.7582794224754723, "grad_norm": 6.582056045532227, "learning_rate": 7.246746298488949e-05, "loss": 0.29910922050476074, "memory(GiB)": 70.5, "step": 41040, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.452698 }, { "epoch": 1.7584936378047213, "grad_norm": 2.395578384399414, "learning_rate": 7.246145069713804e-05, "loss": 0.6105207920074462, "memory(GiB)": 70.5, "step": 41045, "token_acc": 0.8828337874659401, "train_speed(iter/s)": 1.452712 }, { "epoch": 1.7587078531339704, "grad_norm": 2.3796844482421875, "learning_rate": 7.245543800247634e-05, "loss": 0.2757678747177124, "memory(GiB)": 70.5, "step": 41050, "token_acc": 0.95, "train_speed(iter/s)": 1.452717 }, { "epoch": 1.7589220684632192, "grad_norm": 0.48376592993736267, "learning_rate": 7.244942490101332e-05, "loss": 0.22827811241149903, "memory(GiB)": 70.5, "step": 41055, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.452725 }, { "epoch": 1.7591362837924682, "grad_norm": 3.3915927410125732, "learning_rate": 7.24434113928579e-05, "loss": 0.2759871959686279, "memory(GiB)": 70.5, "step": 41060, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.452729 }, { "epoch": 1.7593504991217173, "grad_norm": 2.938312292098999, "learning_rate": 7.243739747811903e-05, "loss": 0.3097441911697388, "memory(GiB)": 70.5, "step": 41065, "token_acc": 0.93125, "train_speed(iter/s)": 1.45274 }, { "epoch": 1.759564714450966, "grad_norm": 2.621328830718994, "learning_rate": 7.243138315690567e-05, "loss": 0.6350876808166503, "memory(GiB)": 70.5, "step": 41070, "token_acc": 0.8920454545454546, "train_speed(iter/s)": 1.452756 }, { "epoch": 1.759778929780215, "grad_norm": 0.6046125292778015, "learning_rate": 7.242536842932675e-05, "loss": 0.37547600269317627, "memory(GiB)": 70.5, "step": 41075, "token_acc": 0.9119496855345912, "train_speed(iter/s)": 1.452761 }, { "epoch": 1.7599931451094641, "grad_norm": 3.093473434448242, "learning_rate": 7.241935329549125e-05, "loss": 0.5883667469024658, "memory(GiB)": 70.5, "step": 41080, "token_acc": 0.873015873015873, "train_speed(iter/s)": 1.452768 }, { "epoch": 1.760207360438713, "grad_norm": 1.5587327480316162, "learning_rate": 7.241333775550813e-05, "loss": 0.3041940212249756, "memory(GiB)": 70.5, "step": 41085, "token_acc": 0.9375, "train_speed(iter/s)": 1.452768 }, { "epoch": 1.760421575767962, "grad_norm": 5.337957859039307, "learning_rate": 7.240732180948637e-05, "loss": 0.4924555778503418, "memory(GiB)": 70.5, "step": 41090, "token_acc": 0.8892128279883382, "train_speed(iter/s)": 1.452761 }, { "epoch": 1.760635791097211, "grad_norm": 1.9050015211105347, "learning_rate": 7.240130545753496e-05, "loss": 0.16079612970352172, "memory(GiB)": 70.5, "step": 41095, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.452761 }, { "epoch": 1.7608500064264598, "grad_norm": 8.75805950164795, "learning_rate": 7.239528869976288e-05, "loss": 0.31352553367614744, "memory(GiB)": 70.5, "step": 41100, "token_acc": 0.9157088122605364, "train_speed(iter/s)": 1.452783 }, { "epoch": 1.7610642217557089, "grad_norm": 3.6870477199554443, "learning_rate": 7.238927153627914e-05, "loss": 0.3976861238479614, "memory(GiB)": 70.5, "step": 41105, "token_acc": 0.912751677852349, "train_speed(iter/s)": 1.452792 }, { "epoch": 1.761278437084958, "grad_norm": 0.8606336116790771, "learning_rate": 7.238325396719275e-05, "loss": 0.18741439580917357, "memory(GiB)": 70.5, "step": 41110, "token_acc": 0.9598393574297188, "train_speed(iter/s)": 1.452787 }, { "epoch": 1.7614926524142067, "grad_norm": 2.733031988143921, "learning_rate": 7.23772359926127e-05, "loss": 0.2287053346633911, "memory(GiB)": 70.5, "step": 41115, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.452787 }, { "epoch": 1.7617068677434558, "grad_norm": 4.862912178039551, "learning_rate": 7.237121761264805e-05, "loss": 0.31458892822265627, "memory(GiB)": 70.5, "step": 41120, "token_acc": 0.9462025316455697, "train_speed(iter/s)": 1.45279 }, { "epoch": 1.7619210830727048, "grad_norm": 3.1095755100250244, "learning_rate": 7.23651988274078e-05, "loss": 0.40685153007507324, "memory(GiB)": 70.5, "step": 41125, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.452805 }, { "epoch": 1.7621352984019536, "grad_norm": 3.329308271408081, "learning_rate": 7.235917963700098e-05, "loss": 0.3863210916519165, "memory(GiB)": 70.5, "step": 41130, "token_acc": 0.9144542772861357, "train_speed(iter/s)": 1.4528 }, { "epoch": 1.7623495137312026, "grad_norm": 2.0961430072784424, "learning_rate": 7.235316004153667e-05, "loss": 0.28809328079223634, "memory(GiB)": 70.5, "step": 41135, "token_acc": 0.951417004048583, "train_speed(iter/s)": 1.45282 }, { "epoch": 1.7625637290604517, "grad_norm": 3.8429512977600098, "learning_rate": 7.234714004112388e-05, "loss": 0.46563191413879396, "memory(GiB)": 70.5, "step": 41140, "token_acc": 0.9116022099447514, "train_speed(iter/s)": 1.452855 }, { "epoch": 1.7627779443897005, "grad_norm": 3.0166778564453125, "learning_rate": 7.234111963587169e-05, "loss": 0.631031847000122, "memory(GiB)": 70.5, "step": 41145, "token_acc": 0.8745387453874539, "train_speed(iter/s)": 1.452869 }, { "epoch": 1.7629921597189495, "grad_norm": 3.727104663848877, "learning_rate": 7.233509882588914e-05, "loss": 0.49788408279418944, "memory(GiB)": 70.5, "step": 41150, "token_acc": 0.9154078549848943, "train_speed(iter/s)": 1.452861 }, { "epoch": 1.7632063750481985, "grad_norm": 2.56402587890625, "learning_rate": 7.232907761128535e-05, "loss": 0.37325661182403563, "memory(GiB)": 70.5, "step": 41155, "token_acc": 0.9352941176470588, "train_speed(iter/s)": 1.452853 }, { "epoch": 1.7634205903774474, "grad_norm": 5.494442462921143, "learning_rate": 7.232305599216938e-05, "loss": 0.3952138900756836, "memory(GiB)": 70.5, "step": 41160, "token_acc": 0.925, "train_speed(iter/s)": 1.452852 }, { "epoch": 1.7636348057066964, "grad_norm": 3.6450657844543457, "learning_rate": 7.231703396865029e-05, "loss": 0.490583610534668, "memory(GiB)": 70.5, "step": 41165, "token_acc": 0.8854166666666666, "train_speed(iter/s)": 1.452855 }, { "epoch": 1.7638490210359454, "grad_norm": 2.6768405437469482, "learning_rate": 7.23110115408372e-05, "loss": 0.5055151462554932, "memory(GiB)": 70.5, "step": 41170, "token_acc": 0.8927536231884058, "train_speed(iter/s)": 1.452856 }, { "epoch": 1.7640632363651942, "grad_norm": 0.3030024766921997, "learning_rate": 7.23049887088392e-05, "loss": 0.2724770069122314, "memory(GiB)": 70.5, "step": 41175, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.452844 }, { "epoch": 1.7642774516944433, "grad_norm": 7.130879878997803, "learning_rate": 7.22989654727654e-05, "loss": 0.5667361259460449, "memory(GiB)": 70.5, "step": 41180, "token_acc": 0.8835616438356164, "train_speed(iter/s)": 1.452817 }, { "epoch": 1.7644916670236923, "grad_norm": 1.6673266887664795, "learning_rate": 7.229294183272495e-05, "loss": 0.3254960775375366, "memory(GiB)": 70.5, "step": 41185, "token_acc": 0.9263456090651558, "train_speed(iter/s)": 1.452835 }, { "epoch": 1.7647058823529411, "grad_norm": 7.426810264587402, "learning_rate": 7.228691778882693e-05, "loss": 0.4558091163635254, "memory(GiB)": 70.5, "step": 41190, "token_acc": 0.9099378881987578, "train_speed(iter/s)": 1.452826 }, { "epoch": 1.7649200976821902, "grad_norm": 2.7840681076049805, "learning_rate": 7.228089334118047e-05, "loss": 0.5959766864776611, "memory(GiB)": 70.5, "step": 41195, "token_acc": 0.8673139158576052, "train_speed(iter/s)": 1.45281 }, { "epoch": 1.7651343130114392, "grad_norm": 1.611932635307312, "learning_rate": 7.227486848989475e-05, "loss": 0.1892843246459961, "memory(GiB)": 70.5, "step": 41200, "token_acc": 0.9561128526645768, "train_speed(iter/s)": 1.45282 }, { "epoch": 1.765348528340688, "grad_norm": 3.0349535942077637, "learning_rate": 7.226884323507887e-05, "loss": 0.2915960788726807, "memory(GiB)": 70.5, "step": 41205, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.452837 }, { "epoch": 1.765562743669937, "grad_norm": 0.9498533606529236, "learning_rate": 7.226281757684204e-05, "loss": 0.2170416831970215, "memory(GiB)": 70.5, "step": 41210, "token_acc": 0.9405204460966543, "train_speed(iter/s)": 1.452838 }, { "epoch": 1.765776958999186, "grad_norm": 3.915473461151123, "learning_rate": 7.225679151529337e-05, "loss": 0.4935890197753906, "memory(GiB)": 70.5, "step": 41215, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.452845 }, { "epoch": 1.7659911743284349, "grad_norm": 4.237518310546875, "learning_rate": 7.225076505054202e-05, "loss": 0.29362921714782714, "memory(GiB)": 70.5, "step": 41220, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.45285 }, { "epoch": 1.766205389657684, "grad_norm": 4.8768181800842285, "learning_rate": 7.224473818269721e-05, "loss": 0.6589579582214355, "memory(GiB)": 70.5, "step": 41225, "token_acc": 0.8858131487889274, "train_speed(iter/s)": 1.452853 }, { "epoch": 1.766419604986933, "grad_norm": 3.9340415000915527, "learning_rate": 7.22387109118681e-05, "loss": 0.16175860166549683, "memory(GiB)": 70.5, "step": 41230, "token_acc": 0.9607142857142857, "train_speed(iter/s)": 1.452852 }, { "epoch": 1.7666338203161818, "grad_norm": 0.2864367961883545, "learning_rate": 7.223268323816386e-05, "loss": 0.1626768946647644, "memory(GiB)": 70.5, "step": 41235, "token_acc": 0.9626168224299065, "train_speed(iter/s)": 1.452858 }, { "epoch": 1.7668480356454308, "grad_norm": 0.35499247908592224, "learning_rate": 7.222665516169373e-05, "loss": 0.3365834951400757, "memory(GiB)": 70.5, "step": 41240, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.452858 }, { "epoch": 1.7670622509746798, "grad_norm": 0.41101446747779846, "learning_rate": 7.22206266825669e-05, "loss": 0.35205085277557374, "memory(GiB)": 70.5, "step": 41245, "token_acc": 0.9455782312925171, "train_speed(iter/s)": 1.452858 }, { "epoch": 1.7672764663039287, "grad_norm": 3.7186596393585205, "learning_rate": 7.221459780089255e-05, "loss": 0.42997050285339355, "memory(GiB)": 70.5, "step": 41250, "token_acc": 0.9324894514767933, "train_speed(iter/s)": 1.452856 }, { "epoch": 1.7674906816331777, "grad_norm": 0.3590008318424225, "learning_rate": 7.220856851677994e-05, "loss": 0.2460613489151001, "memory(GiB)": 70.5, "step": 41255, "token_acc": 0.9453125, "train_speed(iter/s)": 1.452879 }, { "epoch": 1.7677048969624267, "grad_norm": 2.4097659587860107, "learning_rate": 7.220253883033827e-05, "loss": 0.18610965013504027, "memory(GiB)": 70.5, "step": 41260, "token_acc": 0.9452554744525548, "train_speed(iter/s)": 1.452873 }, { "epoch": 1.7679191122916755, "grad_norm": 1.2449893951416016, "learning_rate": 7.219650874167675e-05, "loss": 0.13523296117782593, "memory(GiB)": 70.5, "step": 41265, "token_acc": 0.973421926910299, "train_speed(iter/s)": 1.452871 }, { "epoch": 1.7681333276209246, "grad_norm": 0.33486151695251465, "learning_rate": 7.21904782509047e-05, "loss": 0.49457383155822754, "memory(GiB)": 70.5, "step": 41270, "token_acc": 0.8714859437751004, "train_speed(iter/s)": 1.452892 }, { "epoch": 1.7683475429501736, "grad_norm": 2.2274856567382812, "learning_rate": 7.218444735813132e-05, "loss": 0.6664272308349609, "memory(GiB)": 70.5, "step": 41275, "token_acc": 0.8461538461538461, "train_speed(iter/s)": 1.452911 }, { "epoch": 1.7685617582794224, "grad_norm": 2.394951820373535, "learning_rate": 7.217841606346584e-05, "loss": 0.3863740205764771, "memory(GiB)": 70.5, "step": 41280, "token_acc": 0.898989898989899, "train_speed(iter/s)": 1.452916 }, { "epoch": 1.7687759736086714, "grad_norm": 2.1818506717681885, "learning_rate": 7.217238436701756e-05, "loss": 0.26508588790893556, "memory(GiB)": 70.5, "step": 41285, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.452921 }, { "epoch": 1.7689901889379205, "grad_norm": 1.127026915550232, "learning_rate": 7.216635226889572e-05, "loss": 0.3133701324462891, "memory(GiB)": 70.5, "step": 41290, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.452934 }, { "epoch": 1.7692044042671693, "grad_norm": 0.4185769855976105, "learning_rate": 7.216031976920963e-05, "loss": 0.28543407917022706, "memory(GiB)": 70.5, "step": 41295, "token_acc": 0.9446153846153846, "train_speed(iter/s)": 1.452933 }, { "epoch": 1.7694186195964183, "grad_norm": 4.294552803039551, "learning_rate": 7.215428686806855e-05, "loss": 0.4969793796539307, "memory(GiB)": 70.5, "step": 41300, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.45296 }, { "epoch": 1.7696328349256674, "grad_norm": 3.090672016143799, "learning_rate": 7.214825356558181e-05, "loss": 0.24506075382232667, "memory(GiB)": 70.5, "step": 41305, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.452975 }, { "epoch": 1.7698470502549162, "grad_norm": 3.7169671058654785, "learning_rate": 7.214221986185865e-05, "loss": 0.4202686309814453, "memory(GiB)": 70.5, "step": 41310, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.452987 }, { "epoch": 1.7700612655841652, "grad_norm": 9.178520202636719, "learning_rate": 7.213618575700841e-05, "loss": 0.5521541118621827, "memory(GiB)": 70.5, "step": 41315, "token_acc": 0.8912280701754386, "train_speed(iter/s)": 1.452994 }, { "epoch": 1.7702754809134142, "grad_norm": 5.616382122039795, "learning_rate": 7.21301512511404e-05, "loss": 0.24986093044281005, "memory(GiB)": 70.5, "step": 41320, "token_acc": 0.9537815126050421, "train_speed(iter/s)": 1.452985 }, { "epoch": 1.770489696242663, "grad_norm": 5.162546157836914, "learning_rate": 7.212411634436396e-05, "loss": 0.21298224925994874, "memory(GiB)": 70.5, "step": 41325, "token_acc": 0.9537953795379538, "train_speed(iter/s)": 1.452986 }, { "epoch": 1.770703911571912, "grad_norm": 2.2828938961029053, "learning_rate": 7.21180810367884e-05, "loss": 0.47084383964538573, "memory(GiB)": 70.5, "step": 41330, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.452978 }, { "epoch": 1.7709181269011611, "grad_norm": 0.646457314491272, "learning_rate": 7.211204532852302e-05, "loss": 0.3310267686843872, "memory(GiB)": 70.5, "step": 41335, "token_acc": 0.9276729559748428, "train_speed(iter/s)": 1.452968 }, { "epoch": 1.77113234223041, "grad_norm": 3.956108570098877, "learning_rate": 7.21060092196772e-05, "loss": 0.4591982364654541, "memory(GiB)": 70.5, "step": 41340, "token_acc": 0.887719298245614, "train_speed(iter/s)": 1.452958 }, { "epoch": 1.771346557559659, "grad_norm": 2.772745370864868, "learning_rate": 7.209997271036031e-05, "loss": 0.44019575119018556, "memory(GiB)": 70.5, "step": 41345, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.45298 }, { "epoch": 1.771560772888908, "grad_norm": 4.233997344970703, "learning_rate": 7.209393580068167e-05, "loss": 0.5285609245300293, "memory(GiB)": 70.5, "step": 41350, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.45298 }, { "epoch": 1.7717749882181568, "grad_norm": 4.8051838874816895, "learning_rate": 7.208789849075065e-05, "loss": 0.27902750968933104, "memory(GiB)": 70.5, "step": 41355, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.453015 }, { "epoch": 1.7719892035474059, "grad_norm": 4.295307159423828, "learning_rate": 7.208186078067665e-05, "loss": 0.32774176597595217, "memory(GiB)": 70.5, "step": 41360, "token_acc": 0.932, "train_speed(iter/s)": 1.453011 }, { "epoch": 1.772203418876655, "grad_norm": 5.009275436401367, "learning_rate": 7.2075822670569e-05, "loss": 0.5556405544281006, "memory(GiB)": 70.5, "step": 41365, "token_acc": 0.8955223880597015, "train_speed(iter/s)": 1.453022 }, { "epoch": 1.7724176342059037, "grad_norm": 2.56170654296875, "learning_rate": 7.206978416053714e-05, "loss": 0.4283571243286133, "memory(GiB)": 70.5, "step": 41370, "token_acc": 0.9430604982206405, "train_speed(iter/s)": 1.453026 }, { "epoch": 1.7726318495351527, "grad_norm": 2.9119014739990234, "learning_rate": 7.206374525069041e-05, "loss": 0.7609982967376709, "memory(GiB)": 70.5, "step": 41375, "token_acc": 0.8321917808219178, "train_speed(iter/s)": 1.453032 }, { "epoch": 1.7728460648644018, "grad_norm": 0.7543812394142151, "learning_rate": 7.205770594113826e-05, "loss": 0.35111179351806643, "memory(GiB)": 70.5, "step": 41380, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.453044 }, { "epoch": 1.7730602801936506, "grad_norm": 5.138831615447998, "learning_rate": 7.205166623199007e-05, "loss": 0.375387978553772, "memory(GiB)": 70.5, "step": 41385, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.453042 }, { "epoch": 1.7732744955228996, "grad_norm": 3.6627020835876465, "learning_rate": 7.204562612335526e-05, "loss": 0.6042525768280029, "memory(GiB)": 70.5, "step": 41390, "token_acc": 0.8664122137404581, "train_speed(iter/s)": 1.453045 }, { "epoch": 1.7734887108521487, "grad_norm": 2.537858724594116, "learning_rate": 7.203958561534324e-05, "loss": 0.3185742378234863, "memory(GiB)": 70.5, "step": 41395, "token_acc": 0.926984126984127, "train_speed(iter/s)": 1.453046 }, { "epoch": 1.7737029261813975, "grad_norm": 7.243795871734619, "learning_rate": 7.203354470806348e-05, "loss": 0.8281095504760743, "memory(GiB)": 70.5, "step": 41400, "token_acc": 0.8447653429602888, "train_speed(iter/s)": 1.453067 }, { "epoch": 1.7739171415106465, "grad_norm": 2.1824707984924316, "learning_rate": 7.202750340162536e-05, "loss": 0.4002987384796143, "memory(GiB)": 70.5, "step": 41405, "token_acc": 0.9186746987951807, "train_speed(iter/s)": 1.45308 }, { "epoch": 1.7741313568398955, "grad_norm": 3.6355738639831543, "learning_rate": 7.202146169613835e-05, "loss": 0.31203112602233884, "memory(GiB)": 70.5, "step": 41410, "token_acc": 0.9478527607361963, "train_speed(iter/s)": 1.453077 }, { "epoch": 1.7743455721691443, "grad_norm": 6.529365539550781, "learning_rate": 7.201541959171191e-05, "loss": 0.45801429748535155, "memory(GiB)": 70.5, "step": 41415, "token_acc": 0.8901960784313725, "train_speed(iter/s)": 1.453082 }, { "epoch": 1.7745597874983934, "grad_norm": 2.582127332687378, "learning_rate": 7.200937708845552e-05, "loss": 0.36723182201385496, "memory(GiB)": 70.5, "step": 41420, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.453083 }, { "epoch": 1.7747740028276424, "grad_norm": 2.0308640003204346, "learning_rate": 7.200333418647859e-05, "loss": 0.37702014446258547, "memory(GiB)": 70.5, "step": 41425, "token_acc": 0.9391891891891891, "train_speed(iter/s)": 1.453081 }, { "epoch": 1.7749882181568912, "grad_norm": 3.162593364715576, "learning_rate": 7.199729088589063e-05, "loss": 0.3166424989700317, "memory(GiB)": 70.5, "step": 41430, "token_acc": 0.9320754716981132, "train_speed(iter/s)": 1.453074 }, { "epoch": 1.7752024334861403, "grad_norm": 5.350160121917725, "learning_rate": 7.199124718680111e-05, "loss": 0.4613516330718994, "memory(GiB)": 70.5, "step": 41435, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.453075 }, { "epoch": 1.7754166488153893, "grad_norm": 10.224472999572754, "learning_rate": 7.198520308931951e-05, "loss": 0.23155601024627687, "memory(GiB)": 70.5, "step": 41440, "token_acc": 0.9398496240601504, "train_speed(iter/s)": 1.453089 }, { "epoch": 1.775630864144638, "grad_norm": 3.448495626449585, "learning_rate": 7.197915859355534e-05, "loss": 0.2562771558761597, "memory(GiB)": 70.5, "step": 41445, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.453097 }, { "epoch": 1.7758450794738871, "grad_norm": 5.086706161499023, "learning_rate": 7.197311369961809e-05, "loss": 0.4031463623046875, "memory(GiB)": 70.5, "step": 41450, "token_acc": 0.9177631578947368, "train_speed(iter/s)": 1.453096 }, { "epoch": 1.7760592948031362, "grad_norm": 9.589238166809082, "learning_rate": 7.196706840761727e-05, "loss": 0.48363347053527833, "memory(GiB)": 70.5, "step": 41455, "token_acc": 0.8993055555555556, "train_speed(iter/s)": 1.4531 }, { "epoch": 1.776273510132385, "grad_norm": 3.3067073822021484, "learning_rate": 7.196102271766241e-05, "loss": 0.3596835136413574, "memory(GiB)": 70.5, "step": 41460, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.453087 }, { "epoch": 1.776487725461634, "grad_norm": 2.564406633377075, "learning_rate": 7.1954976629863e-05, "loss": 0.21811344623565673, "memory(GiB)": 70.5, "step": 41465, "token_acc": 0.9516129032258065, "train_speed(iter/s)": 1.453093 }, { "epoch": 1.776701940790883, "grad_norm": 4.551553726196289, "learning_rate": 7.194893014432862e-05, "loss": 0.41861696243286134, "memory(GiB)": 70.5, "step": 41470, "token_acc": 0.9114583333333334, "train_speed(iter/s)": 1.453101 }, { "epoch": 1.7769161561201319, "grad_norm": 4.2792253494262695, "learning_rate": 7.194288326116876e-05, "loss": 0.5820718765258789, "memory(GiB)": 70.5, "step": 41475, "token_acc": 0.890728476821192, "train_speed(iter/s)": 1.453126 }, { "epoch": 1.777130371449381, "grad_norm": 1.1184718608856201, "learning_rate": 7.1936835980493e-05, "loss": 0.3248513460159302, "memory(GiB)": 70.5, "step": 41480, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.453122 }, { "epoch": 1.77734458677863, "grad_norm": 5.430880546569824, "learning_rate": 7.193078830241087e-05, "loss": 0.40962491035461424, "memory(GiB)": 70.5, "step": 41485, "token_acc": 0.9190283400809717, "train_speed(iter/s)": 1.453132 }, { "epoch": 1.7775588021078788, "grad_norm": 1.150283932685852, "learning_rate": 7.192474022703194e-05, "loss": 0.3690898656845093, "memory(GiB)": 70.5, "step": 41490, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.453125 }, { "epoch": 1.7777730174371278, "grad_norm": 8.743621826171875, "learning_rate": 7.191869175446576e-05, "loss": 0.49982495307922364, "memory(GiB)": 70.5, "step": 41495, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.453126 }, { "epoch": 1.7779872327663768, "grad_norm": 7.160375118255615, "learning_rate": 7.191264288482194e-05, "loss": 0.7466540336608887, "memory(GiB)": 70.5, "step": 41500, "token_acc": 0.8456973293768546, "train_speed(iter/s)": 1.453128 }, { "epoch": 1.7779872327663768, "eval_loss": 2.6776578426361084, "eval_runtime": 13.2735, "eval_samples_per_second": 7.534, "eval_steps_per_second": 7.534, "eval_token_acc": 0.44919786096256686, "step": 41500 }, { "epoch": 1.7782014480956256, "grad_norm": 3.6390750408172607, "learning_rate": 7.190659361821003e-05, "loss": 0.4067963123321533, "memory(GiB)": 70.5, "step": 41505, "token_acc": 0.5819750719079578, "train_speed(iter/s)": 1.452427 }, { "epoch": 1.7784156634248747, "grad_norm": 3.490949869155884, "learning_rate": 7.190054395473962e-05, "loss": 0.5561655044555665, "memory(GiB)": 70.5, "step": 41510, "token_acc": 0.8832116788321168, "train_speed(iter/s)": 1.452438 }, { "epoch": 1.7786298787541237, "grad_norm": 4.318096160888672, "learning_rate": 7.189449389452032e-05, "loss": 0.20571882724761964, "memory(GiB)": 70.5, "step": 41515, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.452453 }, { "epoch": 1.7788440940833725, "grad_norm": 3.8933684825897217, "learning_rate": 7.188844343766173e-05, "loss": 0.462283992767334, "memory(GiB)": 70.5, "step": 41520, "token_acc": 0.915625, "train_speed(iter/s)": 1.452453 }, { "epoch": 1.7790583094126216, "grad_norm": 5.40664005279541, "learning_rate": 7.188239258427343e-05, "loss": 0.5017581939697265, "memory(GiB)": 70.5, "step": 41525, "token_acc": 0.8787878787878788, "train_speed(iter/s)": 1.452444 }, { "epoch": 1.7792725247418706, "grad_norm": 1.5541529655456543, "learning_rate": 7.187634133446507e-05, "loss": 0.4353282928466797, "memory(GiB)": 70.5, "step": 41530, "token_acc": 0.8866279069767442, "train_speed(iter/s)": 1.452467 }, { "epoch": 1.7794867400711194, "grad_norm": 2.190575361251831, "learning_rate": 7.18702896883463e-05, "loss": 0.32464070320129396, "memory(GiB)": 70.5, "step": 41535, "token_acc": 0.9363057324840764, "train_speed(iter/s)": 1.45247 }, { "epoch": 1.7797009554003684, "grad_norm": 2.414029836654663, "learning_rate": 7.186423764602668e-05, "loss": 0.1896358847618103, "memory(GiB)": 70.5, "step": 41540, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.452472 }, { "epoch": 1.7799151707296175, "grad_norm": 1.1820380687713623, "learning_rate": 7.185818520761589e-05, "loss": 0.3909574031829834, "memory(GiB)": 70.5, "step": 41545, "token_acc": 0.9213483146067416, "train_speed(iter/s)": 1.452485 }, { "epoch": 1.7801293860588663, "grad_norm": 5.020418167114258, "learning_rate": 7.185213237322358e-05, "loss": 0.5392949104309082, "memory(GiB)": 70.5, "step": 41550, "token_acc": 0.8961937716262975, "train_speed(iter/s)": 1.452492 }, { "epoch": 1.7803436013881153, "grad_norm": 2.4443938732147217, "learning_rate": 7.184607914295937e-05, "loss": 0.2896125316619873, "memory(GiB)": 70.5, "step": 41555, "token_acc": 0.9407407407407408, "train_speed(iter/s)": 1.452497 }, { "epoch": 1.7805578167173644, "grad_norm": 5.058330535888672, "learning_rate": 7.184002551693296e-05, "loss": 0.3719912052154541, "memory(GiB)": 70.5, "step": 41560, "token_acc": 0.9128630705394191, "train_speed(iter/s)": 1.452494 }, { "epoch": 1.7807720320466132, "grad_norm": 3.879171848297119, "learning_rate": 7.183397149525401e-05, "loss": 0.4213155746459961, "memory(GiB)": 70.5, "step": 41565, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.452509 }, { "epoch": 1.7809862473758622, "grad_norm": 4.1328253746032715, "learning_rate": 7.182791707803216e-05, "loss": 0.3800839900970459, "memory(GiB)": 70.5, "step": 41570, "token_acc": 0.9221902017291066, "train_speed(iter/s)": 1.45252 }, { "epoch": 1.7812004627051112, "grad_norm": 2.478199005126953, "learning_rate": 7.182186226537714e-05, "loss": 0.46788859367370605, "memory(GiB)": 70.5, "step": 41575, "token_acc": 0.8992248062015504, "train_speed(iter/s)": 1.452529 }, { "epoch": 1.78141467803436, "grad_norm": 3.183281660079956, "learning_rate": 7.18158070573986e-05, "loss": 0.3930633068084717, "memory(GiB)": 70.5, "step": 41580, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.452532 }, { "epoch": 1.781628893363609, "grad_norm": 2.9466147422790527, "learning_rate": 7.180975145420625e-05, "loss": 0.4732706546783447, "memory(GiB)": 70.5, "step": 41585, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.45254 }, { "epoch": 1.7818431086928581, "grad_norm": 2.240945816040039, "learning_rate": 7.180369545590981e-05, "loss": 0.26071193218231203, "memory(GiB)": 70.5, "step": 41590, "token_acc": 0.9251968503937008, "train_speed(iter/s)": 1.452558 }, { "epoch": 1.782057324022107, "grad_norm": 3.285740613937378, "learning_rate": 7.179763906261896e-05, "loss": 0.37091004848480225, "memory(GiB)": 70.5, "step": 41595, "token_acc": 0.9133858267716536, "train_speed(iter/s)": 1.45256 }, { "epoch": 1.782271539351356, "grad_norm": 4.988839626312256, "learning_rate": 7.179158227444343e-05, "loss": 0.5772286415100097, "memory(GiB)": 70.5, "step": 41600, "token_acc": 0.853125, "train_speed(iter/s)": 1.452553 }, { "epoch": 1.782485754680605, "grad_norm": 5.906022548675537, "learning_rate": 7.178552509149294e-05, "loss": 0.24827198982238768, "memory(GiB)": 70.5, "step": 41605, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.452572 }, { "epoch": 1.7826999700098538, "grad_norm": 2.589186429977417, "learning_rate": 7.177946751387723e-05, "loss": 0.38198180198669435, "memory(GiB)": 70.5, "step": 41610, "token_acc": 0.8993506493506493, "train_speed(iter/s)": 1.452574 }, { "epoch": 1.7829141853391028, "grad_norm": 2.611095428466797, "learning_rate": 7.177340954170604e-05, "loss": 0.2936186552047729, "memory(GiB)": 70.5, "step": 41615, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.452582 }, { "epoch": 1.7831284006683519, "grad_norm": 1.2993764877319336, "learning_rate": 7.176735117508911e-05, "loss": 0.4656956672668457, "memory(GiB)": 70.5, "step": 41620, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.45258 }, { "epoch": 1.7833426159976007, "grad_norm": 3.5509111881256104, "learning_rate": 7.176129241413619e-05, "loss": 0.5044883728027344, "memory(GiB)": 70.5, "step": 41625, "token_acc": 0.8961937716262975, "train_speed(iter/s)": 1.452592 }, { "epoch": 1.7835568313268497, "grad_norm": 3.1214377880096436, "learning_rate": 7.175523325895705e-05, "loss": 0.33735659122467043, "memory(GiB)": 70.5, "step": 41630, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.452586 }, { "epoch": 1.7837710466560988, "grad_norm": 1.707295536994934, "learning_rate": 7.174917370966145e-05, "loss": 0.14683088064193725, "memory(GiB)": 70.5, "step": 41635, "token_acc": 0.9713261648745519, "train_speed(iter/s)": 1.452588 }, { "epoch": 1.7839852619853476, "grad_norm": 6.989225387573242, "learning_rate": 7.174311376635916e-05, "loss": 0.5271139144897461, "memory(GiB)": 70.5, "step": 41640, "token_acc": 0.8956228956228957, "train_speed(iter/s)": 1.45258 }, { "epoch": 1.7841994773145966, "grad_norm": 4.759068489074707, "learning_rate": 7.173705342915998e-05, "loss": 0.2794996976852417, "memory(GiB)": 70.5, "step": 41645, "token_acc": 0.9260869565217391, "train_speed(iter/s)": 1.452581 }, { "epoch": 1.7844136926438456, "grad_norm": 4.040140151977539, "learning_rate": 7.173099269817368e-05, "loss": 0.5090903282165528, "memory(GiB)": 70.5, "step": 41650, "token_acc": 0.9037037037037037, "train_speed(iter/s)": 1.452595 }, { "epoch": 1.7846279079730945, "grad_norm": 2.400664806365967, "learning_rate": 7.172493157351006e-05, "loss": 0.32577242851257326, "memory(GiB)": 70.5, "step": 41655, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.452604 }, { "epoch": 1.7848421233023435, "grad_norm": 2.9351751804351807, "learning_rate": 7.171887005527893e-05, "loss": 0.42700862884521484, "memory(GiB)": 70.5, "step": 41660, "token_acc": 0.9129129129129129, "train_speed(iter/s)": 1.452614 }, { "epoch": 1.7850563386315925, "grad_norm": 2.4743094444274902, "learning_rate": 7.17128081435901e-05, "loss": 0.3455036163330078, "memory(GiB)": 70.5, "step": 41665, "token_acc": 0.9415584415584416, "train_speed(iter/s)": 1.452613 }, { "epoch": 1.7852705539608413, "grad_norm": 3.853506326675415, "learning_rate": 7.170674583855335e-05, "loss": 0.5326683044433593, "memory(GiB)": 70.5, "step": 41670, "token_acc": 0.8873239436619719, "train_speed(iter/s)": 1.45263 }, { "epoch": 1.7854847692900904, "grad_norm": 3.5439486503601074, "learning_rate": 7.170068314027855e-05, "loss": 0.3239924669265747, "memory(GiB)": 70.5, "step": 41675, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.452629 }, { "epoch": 1.7856989846193394, "grad_norm": 5.2498555183410645, "learning_rate": 7.169462004887555e-05, "loss": 0.6093085765838623, "memory(GiB)": 70.5, "step": 41680, "token_acc": 0.8724137931034482, "train_speed(iter/s)": 1.452626 }, { "epoch": 1.7859131999485882, "grad_norm": 1.7252758741378784, "learning_rate": 7.168855656445412e-05, "loss": 0.357589054107666, "memory(GiB)": 70.5, "step": 41685, "token_acc": 0.9105691056910569, "train_speed(iter/s)": 1.452653 }, { "epoch": 1.7861274152778372, "grad_norm": 0.07961058616638184, "learning_rate": 7.168249268712415e-05, "loss": 0.15448522567749023, "memory(GiB)": 70.5, "step": 41690, "token_acc": 0.9669117647058824, "train_speed(iter/s)": 1.452657 }, { "epoch": 1.7863416306070863, "grad_norm": 1.257407784461975, "learning_rate": 7.16764284169955e-05, "loss": 0.42643041610717775, "memory(GiB)": 70.5, "step": 41695, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.452672 }, { "epoch": 1.786555845936335, "grad_norm": 1.402803659439087, "learning_rate": 7.167036375417801e-05, "loss": 0.4138291835784912, "memory(GiB)": 70.5, "step": 41700, "token_acc": 0.9123867069486404, "train_speed(iter/s)": 1.452671 }, { "epoch": 1.7867700612655841, "grad_norm": 3.4554617404937744, "learning_rate": 7.166429869878154e-05, "loss": 0.5070513725280762, "memory(GiB)": 70.5, "step": 41705, "token_acc": 0.9065155807365439, "train_speed(iter/s)": 1.452702 }, { "epoch": 1.7869842765948332, "grad_norm": 2.4076223373413086, "learning_rate": 7.165823325091599e-05, "loss": 0.4426224708557129, "memory(GiB)": 70.5, "step": 41710, "token_acc": 0.9363636363636364, "train_speed(iter/s)": 1.452715 }, { "epoch": 1.787198491924082, "grad_norm": 4.969460487365723, "learning_rate": 7.165216741069122e-05, "loss": 0.33766679763793944, "memory(GiB)": 70.5, "step": 41715, "token_acc": 0.9407114624505929, "train_speed(iter/s)": 1.452714 }, { "epoch": 1.787412707253331, "grad_norm": 5.144708633422852, "learning_rate": 7.164610117821713e-05, "loss": 0.20754942893981934, "memory(GiB)": 70.5, "step": 41720, "token_acc": 0.9589552238805971, "train_speed(iter/s)": 1.452719 }, { "epoch": 1.78762692258258, "grad_norm": 4.356907844543457, "learning_rate": 7.16400345536036e-05, "loss": 0.5454702377319336, "memory(GiB)": 70.5, "step": 41725, "token_acc": 0.8872727272727273, "train_speed(iter/s)": 1.452718 }, { "epoch": 1.7878411379118289, "grad_norm": 1.6419849395751953, "learning_rate": 7.163396753696057e-05, "loss": 0.19260450601577758, "memory(GiB)": 70.5, "step": 41730, "token_acc": 0.9646302250803859, "train_speed(iter/s)": 1.452716 }, { "epoch": 1.788055353241078, "grad_norm": 8.095528602600098, "learning_rate": 7.162790012839791e-05, "loss": 0.6324310779571534, "memory(GiB)": 70.5, "step": 41735, "token_acc": 0.8697183098591549, "train_speed(iter/s)": 1.45273 }, { "epoch": 1.788269568570327, "grad_norm": 3.6482903957366943, "learning_rate": 7.162183232802554e-05, "loss": 0.38828697204589846, "memory(GiB)": 70.5, "step": 41740, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.452718 }, { "epoch": 1.7884837838995757, "grad_norm": 2.1134817600250244, "learning_rate": 7.161576413595339e-05, "loss": 0.351927661895752, "memory(GiB)": 70.5, "step": 41745, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.452718 }, { "epoch": 1.7886979992288248, "grad_norm": 3.061795949935913, "learning_rate": 7.160969555229142e-05, "loss": 0.27180984020233157, "memory(GiB)": 70.5, "step": 41750, "token_acc": 0.9438943894389439, "train_speed(iter/s)": 1.452731 }, { "epoch": 1.7889122145580738, "grad_norm": 2.6900794506073, "learning_rate": 7.160362657714953e-05, "loss": 0.4605147361755371, "memory(GiB)": 70.5, "step": 41755, "token_acc": 0.884375, "train_speed(iter/s)": 1.452742 }, { "epoch": 1.7891264298873226, "grad_norm": 6.8613762855529785, "learning_rate": 7.159755721063768e-05, "loss": 0.628001070022583, "memory(GiB)": 70.5, "step": 41760, "token_acc": 0.8650519031141869, "train_speed(iter/s)": 1.452724 }, { "epoch": 1.7893406452165717, "grad_norm": 4.955134391784668, "learning_rate": 7.159148745286582e-05, "loss": 0.45148553848266604, "memory(GiB)": 70.5, "step": 41765, "token_acc": 0.9058823529411765, "train_speed(iter/s)": 1.452726 }, { "epoch": 1.7895548605458207, "grad_norm": 2.6849594116210938, "learning_rate": 7.158541730394391e-05, "loss": 0.23024826049804686, "memory(GiB)": 70.5, "step": 41770, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.452714 }, { "epoch": 1.7897690758750695, "grad_norm": 3.193829298019409, "learning_rate": 7.157934676398192e-05, "loss": 0.5030865669250488, "memory(GiB)": 70.5, "step": 41775, "token_acc": 0.8916666666666667, "train_speed(iter/s)": 1.452713 }, { "epoch": 1.7899832912043185, "grad_norm": 4.285099983215332, "learning_rate": 7.157327583308981e-05, "loss": 0.2993576765060425, "memory(GiB)": 70.5, "step": 41780, "token_acc": 0.9450980392156862, "train_speed(iter/s)": 1.452704 }, { "epoch": 1.7901975065335676, "grad_norm": 2.075598955154419, "learning_rate": 7.156720451137759e-05, "loss": 0.3701626777648926, "memory(GiB)": 70.5, "step": 41785, "token_acc": 0.9320754716981132, "train_speed(iter/s)": 1.452707 }, { "epoch": 1.7904117218628164, "grad_norm": 4.710295677185059, "learning_rate": 7.156113279895522e-05, "loss": 0.38420305252075193, "memory(GiB)": 70.5, "step": 41790, "token_acc": 0.9209486166007905, "train_speed(iter/s)": 1.452737 }, { "epoch": 1.7906259371920654, "grad_norm": 1.8919997215270996, "learning_rate": 7.15550606959327e-05, "loss": 0.2748755693435669, "memory(GiB)": 70.5, "step": 41795, "token_acc": 0.9446254071661238, "train_speed(iter/s)": 1.452738 }, { "epoch": 1.7908401525213145, "grad_norm": 1.980763554573059, "learning_rate": 7.154898820242003e-05, "loss": 0.2558917045593262, "memory(GiB)": 70.5, "step": 41800, "token_acc": 0.9488054607508533, "train_speed(iter/s)": 1.452743 }, { "epoch": 1.7910543678505633, "grad_norm": 4.460105895996094, "learning_rate": 7.154291531852723e-05, "loss": 0.33253729343414307, "memory(GiB)": 70.5, "step": 41805, "token_acc": 0.9045801526717557, "train_speed(iter/s)": 1.452749 }, { "epoch": 1.7912685831798123, "grad_norm": 3.794426918029785, "learning_rate": 7.153684204436433e-05, "loss": 0.39499309062957766, "memory(GiB)": 70.5, "step": 41810, "token_acc": 0.9023569023569024, "train_speed(iter/s)": 1.452769 }, { "epoch": 1.7914827985090613, "grad_norm": 3.365703582763672, "learning_rate": 7.153076838004129e-05, "loss": 0.3562678337097168, "memory(GiB)": 70.5, "step": 41815, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.45278 }, { "epoch": 1.7916970138383101, "grad_norm": 8.906434059143066, "learning_rate": 7.152469432566822e-05, "loss": 0.5658807754516602, "memory(GiB)": 70.5, "step": 41820, "token_acc": 0.8924418604651163, "train_speed(iter/s)": 1.452783 }, { "epoch": 1.7919112291675592, "grad_norm": 4.518669605255127, "learning_rate": 7.151861988135511e-05, "loss": 0.6101918697357178, "memory(GiB)": 70.5, "step": 41825, "token_acc": 0.8828828828828829, "train_speed(iter/s)": 1.452787 }, { "epoch": 1.7921254444968082, "grad_norm": 4.2581281661987305, "learning_rate": 7.151254504721201e-05, "loss": 0.5990650177001953, "memory(GiB)": 70.5, "step": 41830, "token_acc": 0.87248322147651, "train_speed(iter/s)": 1.452796 }, { "epoch": 1.792339659826057, "grad_norm": 1.5400680303573608, "learning_rate": 7.150646982334897e-05, "loss": 0.3497078657150269, "memory(GiB)": 70.5, "step": 41835, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.452806 }, { "epoch": 1.792553875155306, "grad_norm": 2.2721991539001465, "learning_rate": 7.150039420987606e-05, "loss": 0.41789731979370115, "memory(GiB)": 70.5, "step": 41840, "token_acc": 0.9121212121212121, "train_speed(iter/s)": 1.452839 }, { "epoch": 1.792768090484555, "grad_norm": 2.981149196624756, "learning_rate": 7.149431820690335e-05, "loss": 0.39328575134277344, "memory(GiB)": 70.5, "step": 41845, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.452851 }, { "epoch": 1.792982305813804, "grad_norm": 4.964023113250732, "learning_rate": 7.14882418145409e-05, "loss": 0.35712347030639646, "memory(GiB)": 70.5, "step": 41850, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.452864 }, { "epoch": 1.793196521143053, "grad_norm": 2.8138160705566406, "learning_rate": 7.148216503289878e-05, "loss": 0.42259368896484373, "memory(GiB)": 70.5, "step": 41855, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.45288 }, { "epoch": 1.793410736472302, "grad_norm": 2.4861464500427246, "learning_rate": 7.147608786208709e-05, "loss": 0.3145188093185425, "memory(GiB)": 70.5, "step": 41860, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.452895 }, { "epoch": 1.7936249518015508, "grad_norm": 3.7854998111724854, "learning_rate": 7.147001030221594e-05, "loss": 0.15920451879501343, "memory(GiB)": 70.5, "step": 41865, "token_acc": 0.9622641509433962, "train_speed(iter/s)": 1.452902 }, { "epoch": 1.7938391671307998, "grad_norm": 5.665269374847412, "learning_rate": 7.146393235339539e-05, "loss": 0.6304004192352295, "memory(GiB)": 70.5, "step": 41870, "token_acc": 0.8726114649681529, "train_speed(iter/s)": 1.452891 }, { "epoch": 1.7940533824600489, "grad_norm": 3.0816917419433594, "learning_rate": 7.145785401573559e-05, "loss": 0.3627488136291504, "memory(GiB)": 70.5, "step": 41875, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.452885 }, { "epoch": 1.7942675977892977, "grad_norm": 2.736788511276245, "learning_rate": 7.145177528934663e-05, "loss": 0.24887516498565673, "memory(GiB)": 70.5, "step": 41880, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.452896 }, { "epoch": 1.7944818131185467, "grad_norm": 4.89387845993042, "learning_rate": 7.144569617433863e-05, "loss": 0.45538835525512694, "memory(GiB)": 70.5, "step": 41885, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.452916 }, { "epoch": 1.7946960284477957, "grad_norm": 0.2610904276371002, "learning_rate": 7.143961667082173e-05, "loss": 0.326352858543396, "memory(GiB)": 70.5, "step": 41890, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.452916 }, { "epoch": 1.7949102437770446, "grad_norm": 3.17948842048645, "learning_rate": 7.143353677890607e-05, "loss": 0.35785095691680907, "memory(GiB)": 70.5, "step": 41895, "token_acc": 0.9299363057324841, "train_speed(iter/s)": 1.452922 }, { "epoch": 1.7951244591062938, "grad_norm": 3.1745495796203613, "learning_rate": 7.142745649870177e-05, "loss": 0.42084426879882814, "memory(GiB)": 70.5, "step": 41900, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.452929 }, { "epoch": 1.7953386744355426, "grad_norm": 1.0376875400543213, "learning_rate": 7.142137583031901e-05, "loss": 0.30988345146179197, "memory(GiB)": 70.5, "step": 41905, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.452949 }, { "epoch": 1.7955528897647914, "grad_norm": 5.0907511711120605, "learning_rate": 7.141529477386792e-05, "loss": 0.40573558807373045, "memory(GiB)": 70.5, "step": 41910, "token_acc": 0.8991228070175439, "train_speed(iter/s)": 1.452951 }, { "epoch": 1.7957671050940407, "grad_norm": 2.453829050064087, "learning_rate": 7.140921332945868e-05, "loss": 0.297459077835083, "memory(GiB)": 70.5, "step": 41915, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.452981 }, { "epoch": 1.7959813204232895, "grad_norm": 2.6994574069976807, "learning_rate": 7.140313149720145e-05, "loss": 0.5889752864837646, "memory(GiB)": 70.5, "step": 41920, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.452992 }, { "epoch": 1.7961955357525383, "grad_norm": 4.96846342086792, "learning_rate": 7.139704927720644e-05, "loss": 0.3171527862548828, "memory(GiB)": 70.5, "step": 41925, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.45302 }, { "epoch": 1.7964097510817876, "grad_norm": 2.8398513793945312, "learning_rate": 7.139096666958378e-05, "loss": 0.32324748039245604, "memory(GiB)": 70.5, "step": 41930, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.453021 }, { "epoch": 1.7966239664110364, "grad_norm": 4.390688419342041, "learning_rate": 7.13848836744437e-05, "loss": 0.5740853309631347, "memory(GiB)": 70.5, "step": 41935, "token_acc": 0.8892988929889298, "train_speed(iter/s)": 1.453034 }, { "epoch": 1.7968381817402852, "grad_norm": 5.333402633666992, "learning_rate": 7.137880029189641e-05, "loss": 0.41851134300231935, "memory(GiB)": 70.5, "step": 41940, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.453034 }, { "epoch": 1.7970523970695345, "grad_norm": 3.6801247596740723, "learning_rate": 7.137271652205208e-05, "loss": 0.6245378494262696, "memory(GiB)": 70.5, "step": 41945, "token_acc": 0.8746081504702194, "train_speed(iter/s)": 1.453043 }, { "epoch": 1.7972666123987833, "grad_norm": 0.21482005715370178, "learning_rate": 7.136663236502095e-05, "loss": 0.366714334487915, "memory(GiB)": 70.5, "step": 41950, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.453054 }, { "epoch": 1.797480827728032, "grad_norm": 1.8663514852523804, "learning_rate": 7.136054782091322e-05, "loss": 0.43352975845336916, "memory(GiB)": 70.5, "step": 41955, "token_acc": 0.900398406374502, "train_speed(iter/s)": 1.453053 }, { "epoch": 1.7976950430572813, "grad_norm": 1.8394652605056763, "learning_rate": 7.135446288983914e-05, "loss": 0.13456997871398926, "memory(GiB)": 70.5, "step": 41960, "token_acc": 0.9682539682539683, "train_speed(iter/s)": 1.453046 }, { "epoch": 1.7979092583865302, "grad_norm": 2.1299643516540527, "learning_rate": 7.134837757190891e-05, "loss": 0.39633593559265134, "memory(GiB)": 70.5, "step": 41965, "token_acc": 0.9129129129129129, "train_speed(iter/s)": 1.453031 }, { "epoch": 1.798123473715779, "grad_norm": 2.3081653118133545, "learning_rate": 7.134229186723282e-05, "loss": 0.19673519134521483, "memory(GiB)": 70.5, "step": 41970, "token_acc": 0.9595588235294118, "train_speed(iter/s)": 1.453021 }, { "epoch": 1.7983376890450282, "grad_norm": 2.4200539588928223, "learning_rate": 7.133620577592108e-05, "loss": 0.31533222198486327, "memory(GiB)": 70.5, "step": 41975, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.453012 }, { "epoch": 1.798551904374277, "grad_norm": 3.0603978633880615, "learning_rate": 7.133011929808398e-05, "loss": 0.19516438245773315, "memory(GiB)": 70.5, "step": 41980, "token_acc": 0.9535714285714286, "train_speed(iter/s)": 1.453024 }, { "epoch": 1.7987661197035258, "grad_norm": 1.0239856243133545, "learning_rate": 7.132403243383173e-05, "loss": 0.18945075273513795, "memory(GiB)": 70.5, "step": 41985, "token_acc": 0.9647887323943662, "train_speed(iter/s)": 1.453034 }, { "epoch": 1.798980335032775, "grad_norm": 3.358395576477051, "learning_rate": 7.131794518327463e-05, "loss": 0.25840535163879397, "memory(GiB)": 70.5, "step": 41990, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.453026 }, { "epoch": 1.799194550362024, "grad_norm": 1.8083140850067139, "learning_rate": 7.131185754652299e-05, "loss": 0.227734375, "memory(GiB)": 70.5, "step": 41995, "token_acc": 0.9484126984126984, "train_speed(iter/s)": 1.453025 }, { "epoch": 1.7994087656912727, "grad_norm": 2.173149824142456, "learning_rate": 7.130576952368701e-05, "loss": 0.21593520641326905, "memory(GiB)": 70.5, "step": 42000, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.453015 }, { "epoch": 1.7994087656912727, "eval_loss": 2.709421396255493, "eval_runtime": 13.7083, "eval_samples_per_second": 7.295, "eval_steps_per_second": 7.295, "eval_token_acc": 0.4126984126984127, "step": 42000 }, { "epoch": 1.799622981020522, "grad_norm": 2.1826319694519043, "learning_rate": 7.129968111487706e-05, "loss": 0.3111643075942993, "memory(GiB)": 70.5, "step": 42005, "token_acc": 0.5580985915492958, "train_speed(iter/s)": 1.452278 }, { "epoch": 1.7998371963497708, "grad_norm": 4.5063796043396, "learning_rate": 7.12935923202034e-05, "loss": 0.8829240798950195, "memory(GiB)": 70.5, "step": 42010, "token_acc": 0.7986798679867987, "train_speed(iter/s)": 1.452279 }, { "epoch": 1.8000514116790196, "grad_norm": 5.52262020111084, "learning_rate": 7.128750313977633e-05, "loss": 0.4713892936706543, "memory(GiB)": 70.5, "step": 42015, "token_acc": 0.9169675090252708, "train_speed(iter/s)": 1.452299 }, { "epoch": 1.8002656270082689, "grad_norm": 3.360215425491333, "learning_rate": 7.128141357370618e-05, "loss": 0.2589277744293213, "memory(GiB)": 70.5, "step": 42020, "token_acc": 0.9353932584269663, "train_speed(iter/s)": 1.452305 }, { "epoch": 1.8004798423375177, "grad_norm": 3.2414164543151855, "learning_rate": 7.127532362210327e-05, "loss": 0.2671512126922607, "memory(GiB)": 70.5, "step": 42025, "token_acc": 0.9427480916030534, "train_speed(iter/s)": 1.452303 }, { "epoch": 1.8006940576667665, "grad_norm": 4.351088047027588, "learning_rate": 7.12692332850779e-05, "loss": 0.8586606025695801, "memory(GiB)": 70.5, "step": 42030, "token_acc": 0.8673740053050398, "train_speed(iter/s)": 1.45231 }, { "epoch": 1.8009082729960157, "grad_norm": 3.6430435180664062, "learning_rate": 7.126314256274042e-05, "loss": 0.6846446990966797, "memory(GiB)": 70.5, "step": 42035, "token_acc": 0.8464566929133859, "train_speed(iter/s)": 1.452318 }, { "epoch": 1.8011224883252646, "grad_norm": 1.781693458557129, "learning_rate": 7.125705145520117e-05, "loss": 0.5172185897827148, "memory(GiB)": 70.5, "step": 42040, "token_acc": 0.8786764705882353, "train_speed(iter/s)": 1.452328 }, { "epoch": 1.8013367036545134, "grad_norm": 0.23058313131332397, "learning_rate": 7.125095996257048e-05, "loss": 0.26462841033935547, "memory(GiB)": 70.5, "step": 42045, "token_acc": 0.9375, "train_speed(iter/s)": 1.452323 }, { "epoch": 1.8015509189837626, "grad_norm": 2.8295786380767822, "learning_rate": 7.124486808495873e-05, "loss": 0.4166316032409668, "memory(GiB)": 70.5, "step": 42050, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.452325 }, { "epoch": 1.8017651343130114, "grad_norm": 5.929265975952148, "learning_rate": 7.123877582247626e-05, "loss": 0.6116162300109863, "memory(GiB)": 70.5, "step": 42055, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.452323 }, { "epoch": 1.8019793496422603, "grad_norm": 1.7793787717819214, "learning_rate": 7.123268317523344e-05, "loss": 0.30252666473388673, "memory(GiB)": 70.5, "step": 42060, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.452312 }, { "epoch": 1.8021935649715095, "grad_norm": 5.879805088043213, "learning_rate": 7.122659014334065e-05, "loss": 0.423005199432373, "memory(GiB)": 70.5, "step": 42065, "token_acc": 0.9040247678018576, "train_speed(iter/s)": 1.452337 }, { "epoch": 1.8024077803007583, "grad_norm": 6.4032063484191895, "learning_rate": 7.122049672690828e-05, "loss": 0.5568048000335694, "memory(GiB)": 70.5, "step": 42070, "token_acc": 0.8963210702341137, "train_speed(iter/s)": 1.452347 }, { "epoch": 1.8026219956300071, "grad_norm": 2.982900619506836, "learning_rate": 7.121440292604667e-05, "loss": 0.2628678798675537, "memory(GiB)": 70.5, "step": 42075, "token_acc": 0.9488054607508533, "train_speed(iter/s)": 1.452349 }, { "epoch": 1.8028362109592564, "grad_norm": 2.5495381355285645, "learning_rate": 7.120830874086627e-05, "loss": 0.30063481330871583, "memory(GiB)": 70.5, "step": 42080, "token_acc": 0.9314516129032258, "train_speed(iter/s)": 1.452354 }, { "epoch": 1.8030504262885052, "grad_norm": 3.907656669616699, "learning_rate": 7.120221417147747e-05, "loss": 0.3681309938430786, "memory(GiB)": 70.5, "step": 42085, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.452352 }, { "epoch": 1.803264641617754, "grad_norm": 2.7240686416625977, "learning_rate": 7.119611921799065e-05, "loss": 0.22327961921691894, "memory(GiB)": 70.5, "step": 42090, "token_acc": 0.941358024691358, "train_speed(iter/s)": 1.452356 }, { "epoch": 1.8034788569470033, "grad_norm": 1.3352652788162231, "learning_rate": 7.119002388051627e-05, "loss": 0.318222713470459, "memory(GiB)": 70.5, "step": 42095, "token_acc": 0.9407894736842105, "train_speed(iter/s)": 1.452375 }, { "epoch": 1.803693072276252, "grad_norm": 2.9939873218536377, "learning_rate": 7.118392815916472e-05, "loss": 0.20417180061340331, "memory(GiB)": 70.5, "step": 42100, "token_acc": 0.9494584837545126, "train_speed(iter/s)": 1.452363 }, { "epoch": 1.803907287605501, "grad_norm": 3.5503742694854736, "learning_rate": 7.117783205404644e-05, "loss": 0.5174329280853271, "memory(GiB)": 70.5, "step": 42105, "token_acc": 0.9026845637583892, "train_speed(iter/s)": 1.452393 }, { "epoch": 1.8041215029347502, "grad_norm": 4.12925386428833, "learning_rate": 7.117173556527187e-05, "loss": 0.3832792520523071, "memory(GiB)": 70.5, "step": 42110, "token_acc": 0.9172185430463576, "train_speed(iter/s)": 1.452395 }, { "epoch": 1.804335718263999, "grad_norm": 3.1932175159454346, "learning_rate": 7.116563869295145e-05, "loss": 0.3819037199020386, "memory(GiB)": 70.5, "step": 42115, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.452403 }, { "epoch": 1.8045499335932478, "grad_norm": 1.5898545980453491, "learning_rate": 7.11595414371956e-05, "loss": 0.20027048587799073, "memory(GiB)": 70.5, "step": 42120, "token_acc": 0.9556213017751479, "train_speed(iter/s)": 1.452394 }, { "epoch": 1.804764148922497, "grad_norm": 4.166143894195557, "learning_rate": 7.115344379811485e-05, "loss": 0.29294095039367674, "memory(GiB)": 70.5, "step": 42125, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.452393 }, { "epoch": 1.8049783642517458, "grad_norm": 1.7356730699539185, "learning_rate": 7.11473457758196e-05, "loss": 0.3470311641693115, "memory(GiB)": 70.5, "step": 42130, "token_acc": 0.8973509933774835, "train_speed(iter/s)": 1.452383 }, { "epoch": 1.8051925795809947, "grad_norm": 3.5033891201019287, "learning_rate": 7.114124737042035e-05, "loss": 0.5549688816070557, "memory(GiB)": 70.5, "step": 42135, "token_acc": 0.8923611111111112, "train_speed(iter/s)": 1.452387 }, { "epoch": 1.805406794910244, "grad_norm": 0.6526398062705994, "learning_rate": 7.113514858202758e-05, "loss": 0.3809556722640991, "memory(GiB)": 70.5, "step": 42140, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.452394 }, { "epoch": 1.8056210102394927, "grad_norm": 2.4092133045196533, "learning_rate": 7.112904941075175e-05, "loss": 0.5492062568664551, "memory(GiB)": 70.5, "step": 42145, "token_acc": 0.8706896551724138, "train_speed(iter/s)": 1.452401 }, { "epoch": 1.8058352255687415, "grad_norm": 2.669748544692993, "learning_rate": 7.112294985670337e-05, "loss": 0.3030374765396118, "memory(GiB)": 70.5, "step": 42150, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.452403 }, { "epoch": 1.8060494408979908, "grad_norm": 0.5268579721450806, "learning_rate": 7.111684991999295e-05, "loss": 0.290040111541748, "memory(GiB)": 70.5, "step": 42155, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.4524 }, { "epoch": 1.8062636562272396, "grad_norm": 2.7226617336273193, "learning_rate": 7.111074960073098e-05, "loss": 0.4143199920654297, "memory(GiB)": 70.5, "step": 42160, "token_acc": 0.8768115942028986, "train_speed(iter/s)": 1.452394 }, { "epoch": 1.8064778715564884, "grad_norm": 5.478462219238281, "learning_rate": 7.110464889902796e-05, "loss": 0.5359173774719238, "memory(GiB)": 70.5, "step": 42165, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.4524 }, { "epoch": 1.8066920868857377, "grad_norm": 2.614807605743408, "learning_rate": 7.109854781499446e-05, "loss": 0.44916467666625975, "memory(GiB)": 70.5, "step": 42170, "token_acc": 0.9051724137931034, "train_speed(iter/s)": 1.452415 }, { "epoch": 1.8069063022149865, "grad_norm": 3.172013998031616, "learning_rate": 7.109244634874096e-05, "loss": 0.46584339141845704, "memory(GiB)": 70.5, "step": 42175, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.452416 }, { "epoch": 1.8071205175442353, "grad_norm": 3.069183588027954, "learning_rate": 7.108634450037802e-05, "loss": 0.3977344512939453, "memory(GiB)": 70.5, "step": 42180, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.45242 }, { "epoch": 1.8073347328734846, "grad_norm": 2.205812454223633, "learning_rate": 7.108024227001615e-05, "loss": 0.35320773124694826, "memory(GiB)": 70.5, "step": 42185, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.452448 }, { "epoch": 1.8075489482027334, "grad_norm": 2.679020643234253, "learning_rate": 7.10741396577659e-05, "loss": 0.2957797050476074, "memory(GiB)": 70.5, "step": 42190, "token_acc": 0.9395973154362416, "train_speed(iter/s)": 1.45245 }, { "epoch": 1.8077631635319822, "grad_norm": 2.9036808013916016, "learning_rate": 7.106803666373787e-05, "loss": 0.49407367706298827, "memory(GiB)": 70.5, "step": 42195, "token_acc": 0.9035714285714286, "train_speed(iter/s)": 1.452452 }, { "epoch": 1.8079773788612314, "grad_norm": 2.932349920272827, "learning_rate": 7.10619332880426e-05, "loss": 0.27759289741516113, "memory(GiB)": 70.5, "step": 42200, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.452446 }, { "epoch": 1.8081915941904803, "grad_norm": 0.058626689016819, "learning_rate": 7.105582953079063e-05, "loss": 0.463405704498291, "memory(GiB)": 70.5, "step": 42205, "token_acc": 0.9122137404580153, "train_speed(iter/s)": 1.452441 }, { "epoch": 1.808405809519729, "grad_norm": 0.01821625977754593, "learning_rate": 7.104972539209256e-05, "loss": 0.18241850137710572, "memory(GiB)": 70.5, "step": 42210, "token_acc": 0.9591194968553459, "train_speed(iter/s)": 1.452438 }, { "epoch": 1.8086200248489783, "grad_norm": 3.2004926204681396, "learning_rate": 7.104362087205898e-05, "loss": 0.29882068634033204, "memory(GiB)": 70.5, "step": 42215, "token_acc": 0.9369369369369369, "train_speed(iter/s)": 1.452437 }, { "epoch": 1.8088342401782271, "grad_norm": 4.402462482452393, "learning_rate": 7.103751597080045e-05, "loss": 0.43909664154052735, "memory(GiB)": 70.5, "step": 42220, "token_acc": 0.9007352941176471, "train_speed(iter/s)": 1.452442 }, { "epoch": 1.809048455507476, "grad_norm": 0.7546576261520386, "learning_rate": 7.103141068842759e-05, "loss": 0.13279091119766234, "memory(GiB)": 70.5, "step": 42225, "token_acc": 0.9755244755244755, "train_speed(iter/s)": 1.452442 }, { "epoch": 1.8092626708367252, "grad_norm": 2.77581787109375, "learning_rate": 7.1025305025051e-05, "loss": 0.21057751178741455, "memory(GiB)": 70.5, "step": 42230, "token_acc": 0.9490909090909091, "train_speed(iter/s)": 1.45245 }, { "epoch": 1.809476886165974, "grad_norm": 3.3327431678771973, "learning_rate": 7.101919898078128e-05, "loss": 0.6211094856262207, "memory(GiB)": 70.5, "step": 42235, "token_acc": 0.8660130718954249, "train_speed(iter/s)": 1.45246 }, { "epoch": 1.809691101495223, "grad_norm": 4.172296047210693, "learning_rate": 7.101309255572905e-05, "loss": 0.3082176446914673, "memory(GiB)": 70.5, "step": 42240, "token_acc": 0.9403409090909091, "train_speed(iter/s)": 1.452476 }, { "epoch": 1.809905316824472, "grad_norm": 3.0685300827026367, "learning_rate": 7.100698575000496e-05, "loss": 0.37961182594299314, "memory(GiB)": 70.5, "step": 42245, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.45248 }, { "epoch": 1.810119532153721, "grad_norm": 3.7708911895751953, "learning_rate": 7.10008785637196e-05, "loss": 0.6803537368774414, "memory(GiB)": 70.5, "step": 42250, "token_acc": 0.8493589743589743, "train_speed(iter/s)": 1.452473 }, { "epoch": 1.81033374748297, "grad_norm": 6.248483180999756, "learning_rate": 7.099477099698361e-05, "loss": 0.5071362495422364, "memory(GiB)": 70.5, "step": 42255, "token_acc": 0.8941605839416058, "train_speed(iter/s)": 1.452476 }, { "epoch": 1.810547962812219, "grad_norm": 2.193427085876465, "learning_rate": 7.098866304990767e-05, "loss": 0.2722414493560791, "memory(GiB)": 70.5, "step": 42260, "token_acc": 0.9288537549407114, "train_speed(iter/s)": 1.452482 }, { "epoch": 1.8107621781414678, "grad_norm": 1.227735996246338, "learning_rate": 7.09825547226024e-05, "loss": 0.20759820938110352, "memory(GiB)": 70.5, "step": 42265, "token_acc": 0.9528619528619529, "train_speed(iter/s)": 1.452481 }, { "epoch": 1.8109763934707168, "grad_norm": 5.455148220062256, "learning_rate": 7.097644601517848e-05, "loss": 0.4202564239501953, "memory(GiB)": 70.5, "step": 42270, "token_acc": 0.9022556390977443, "train_speed(iter/s)": 1.452489 }, { "epoch": 1.8111906087999659, "grad_norm": 3.845583438873291, "learning_rate": 7.097033692774653e-05, "loss": 0.3647459030151367, "memory(GiB)": 70.5, "step": 42275, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.452491 }, { "epoch": 1.8114048241292147, "grad_norm": 3.4427649974823, "learning_rate": 7.09642274604173e-05, "loss": 0.4606820583343506, "memory(GiB)": 70.5, "step": 42280, "token_acc": 0.906832298136646, "train_speed(iter/s)": 1.452494 }, { "epoch": 1.8116190394584637, "grad_norm": 2.940770149230957, "learning_rate": 7.09581176133014e-05, "loss": 0.3147725582122803, "memory(GiB)": 70.5, "step": 42285, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.452504 }, { "epoch": 1.8118332547877127, "grad_norm": 3.8532373905181885, "learning_rate": 7.095200738650953e-05, "loss": 0.37131805419921876, "memory(GiB)": 70.5, "step": 42290, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.452491 }, { "epoch": 1.8120474701169615, "grad_norm": 6.948325157165527, "learning_rate": 7.094589678015242e-05, "loss": 0.33642170429229734, "memory(GiB)": 70.5, "step": 42295, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.452493 }, { "epoch": 1.8122616854462106, "grad_norm": 3.8843586444854736, "learning_rate": 7.093978579434072e-05, "loss": 0.4418639183044434, "memory(GiB)": 70.5, "step": 42300, "token_acc": 0.9135802469135802, "train_speed(iter/s)": 1.452495 }, { "epoch": 1.8124759007754596, "grad_norm": 3.197378158569336, "learning_rate": 7.093367442918515e-05, "loss": 0.2782602310180664, "memory(GiB)": 70.5, "step": 42305, "token_acc": 0.924187725631769, "train_speed(iter/s)": 1.452503 }, { "epoch": 1.8126901161047084, "grad_norm": 2.6837127208709717, "learning_rate": 7.092756268479646e-05, "loss": 0.3206287145614624, "memory(GiB)": 70.5, "step": 42310, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.452505 }, { "epoch": 1.8129043314339575, "grad_norm": 0.9694409966468811, "learning_rate": 7.092145056128532e-05, "loss": 0.16878424882888793, "memory(GiB)": 70.5, "step": 42315, "token_acc": 0.9645390070921985, "train_speed(iter/s)": 1.452508 }, { "epoch": 1.8131185467632065, "grad_norm": 1.9273813962936401, "learning_rate": 7.091533805876247e-05, "loss": 0.34309566020965576, "memory(GiB)": 70.5, "step": 42320, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.452504 }, { "epoch": 1.8133327620924553, "grad_norm": 3.144162893295288, "learning_rate": 7.090922517733867e-05, "loss": 0.4297780990600586, "memory(GiB)": 70.5, "step": 42325, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.452498 }, { "epoch": 1.8135469774217043, "grad_norm": 2.5421316623687744, "learning_rate": 7.090311191712464e-05, "loss": 0.38542191982269286, "memory(GiB)": 70.5, "step": 42330, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.452507 }, { "epoch": 1.8137611927509534, "grad_norm": 2.1265807151794434, "learning_rate": 7.08969982782311e-05, "loss": 0.17894819974899293, "memory(GiB)": 70.5, "step": 42335, "token_acc": 0.9525547445255474, "train_speed(iter/s)": 1.452502 }, { "epoch": 1.8139754080802022, "grad_norm": 4.4751739501953125, "learning_rate": 7.089088426076885e-05, "loss": 0.5594378471374511, "memory(GiB)": 70.5, "step": 42340, "token_acc": 0.8774193548387097, "train_speed(iter/s)": 1.452503 }, { "epoch": 1.8141896234094512, "grad_norm": 18.869426727294922, "learning_rate": 7.088476986484865e-05, "loss": 0.39328436851501464, "memory(GiB)": 70.5, "step": 42345, "token_acc": 0.89568345323741, "train_speed(iter/s)": 1.452521 }, { "epoch": 1.8144038387387003, "grad_norm": 2.695847511291504, "learning_rate": 7.087865509058124e-05, "loss": 0.30458357334136965, "memory(GiB)": 70.5, "step": 42350, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.452518 }, { "epoch": 1.814618054067949, "grad_norm": 1.98494291305542, "learning_rate": 7.087253993807741e-05, "loss": 0.29515173435211184, "memory(GiB)": 70.5, "step": 42355, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.452515 }, { "epoch": 1.814832269397198, "grad_norm": 2.6460719108581543, "learning_rate": 7.086642440744794e-05, "loss": 0.39136788845062254, "memory(GiB)": 70.5, "step": 42360, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.452525 }, { "epoch": 1.8150464847264471, "grad_norm": 2.663468360900879, "learning_rate": 7.08603084988036e-05, "loss": 0.3996147632598877, "memory(GiB)": 70.5, "step": 42365, "token_acc": 0.9277566539923955, "train_speed(iter/s)": 1.452525 }, { "epoch": 1.815260700055696, "grad_norm": 2.643157958984375, "learning_rate": 7.085419221225522e-05, "loss": 0.48468265533447263, "memory(GiB)": 70.5, "step": 42370, "token_acc": 0.9181818181818182, "train_speed(iter/s)": 1.452529 }, { "epoch": 1.815474915384945, "grad_norm": 2.645188808441162, "learning_rate": 7.084807554791358e-05, "loss": 0.6819581031799317, "memory(GiB)": 70.5, "step": 42375, "token_acc": 0.8882175226586103, "train_speed(iter/s)": 1.452568 }, { "epoch": 1.815689130714194, "grad_norm": 7.454162120819092, "learning_rate": 7.08419585058895e-05, "loss": 0.30049381256103513, "memory(GiB)": 70.5, "step": 42380, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.452591 }, { "epoch": 1.8159033460434428, "grad_norm": 2.0261175632476807, "learning_rate": 7.083584108629379e-05, "loss": 0.5482788562774659, "memory(GiB)": 70.5, "step": 42385, "token_acc": 0.8871595330739299, "train_speed(iter/s)": 1.452602 }, { "epoch": 1.8161175613726919, "grad_norm": 5.208925724029541, "learning_rate": 7.082972328923726e-05, "loss": 0.6029400825500488, "memory(GiB)": 70.5, "step": 42390, "token_acc": 0.8673469387755102, "train_speed(iter/s)": 1.452604 }, { "epoch": 1.816331776701941, "grad_norm": 9.171188354492188, "learning_rate": 7.082360511483077e-05, "loss": 0.49738402366638185, "memory(GiB)": 70.5, "step": 42395, "token_acc": 0.8895705521472392, "train_speed(iter/s)": 1.452621 }, { "epoch": 1.8165459920311897, "grad_norm": 2.4890575408935547, "learning_rate": 7.081748656318514e-05, "loss": 0.12147742509841919, "memory(GiB)": 70.5, "step": 42400, "token_acc": 0.9612403100775194, "train_speed(iter/s)": 1.452621 }, { "epoch": 1.8167602073604388, "grad_norm": 3.120859384536743, "learning_rate": 7.081136763441119e-05, "loss": 0.3972614049911499, "memory(GiB)": 70.5, "step": 42405, "token_acc": 0.8938906752411575, "train_speed(iter/s)": 1.452611 }, { "epoch": 1.8169744226896878, "grad_norm": 3.899550676345825, "learning_rate": 7.080524832861982e-05, "loss": 0.45512542724609373, "memory(GiB)": 70.5, "step": 42410, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.452616 }, { "epoch": 1.8171886380189366, "grad_norm": 0.3878963589668274, "learning_rate": 7.079912864592185e-05, "loss": 0.39560756683349607, "memory(GiB)": 70.5, "step": 42415, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.452604 }, { "epoch": 1.8174028533481856, "grad_norm": 6.961048603057861, "learning_rate": 7.079300858642814e-05, "loss": 0.4498088836669922, "memory(GiB)": 70.5, "step": 42420, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.452616 }, { "epoch": 1.8176170686774347, "grad_norm": 3.776576519012451, "learning_rate": 7.078688815024959e-05, "loss": 0.47626509666442873, "memory(GiB)": 70.5, "step": 42425, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.452636 }, { "epoch": 1.8178312840066835, "grad_norm": 6.330057144165039, "learning_rate": 7.078076733749706e-05, "loss": 0.7195763111114502, "memory(GiB)": 70.5, "step": 42430, "token_acc": 0.8643067846607669, "train_speed(iter/s)": 1.45265 }, { "epoch": 1.8180454993359325, "grad_norm": 2.9207701683044434, "learning_rate": 7.077464614828142e-05, "loss": 0.5300133705139161, "memory(GiB)": 70.5, "step": 42435, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.452663 }, { "epoch": 1.8182597146651815, "grad_norm": 3.0599265098571777, "learning_rate": 7.07685245827136e-05, "loss": 0.4404360294342041, "memory(GiB)": 70.5, "step": 42440, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.452672 }, { "epoch": 1.8184739299944304, "grad_norm": 2.1618499755859375, "learning_rate": 7.076240264090446e-05, "loss": 0.32312617301940916, "memory(GiB)": 70.5, "step": 42445, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.452683 }, { "epoch": 1.8186881453236794, "grad_norm": 5.8692755699157715, "learning_rate": 7.075628032296491e-05, "loss": 0.3496082782745361, "memory(GiB)": 70.5, "step": 42450, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.452698 }, { "epoch": 1.8189023606529284, "grad_norm": 4.6050639152526855, "learning_rate": 7.07501576290059e-05, "loss": 0.34130191802978516, "memory(GiB)": 70.5, "step": 42455, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.452691 }, { "epoch": 1.8191165759821772, "grad_norm": 3.2310307025909424, "learning_rate": 7.074403455913829e-05, "loss": 0.47139930725097656, "memory(GiB)": 70.5, "step": 42460, "token_acc": 0.9030470914127424, "train_speed(iter/s)": 1.452699 }, { "epoch": 1.8193307913114263, "grad_norm": 14.109204292297363, "learning_rate": 7.073791111347305e-05, "loss": 0.459945011138916, "memory(GiB)": 70.5, "step": 42465, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.452721 }, { "epoch": 1.8195450066406753, "grad_norm": 3.7010347843170166, "learning_rate": 7.07317872921211e-05, "loss": 0.395068359375, "memory(GiB)": 70.5, "step": 42470, "token_acc": 0.8977272727272727, "train_speed(iter/s)": 1.452718 }, { "epoch": 1.8197592219699241, "grad_norm": 2.446230888366699, "learning_rate": 7.072566309519338e-05, "loss": 0.19378007650375367, "memory(GiB)": 70.5, "step": 42475, "token_acc": 0.948936170212766, "train_speed(iter/s)": 1.452725 }, { "epoch": 1.8199734372991732, "grad_norm": 2.7153282165527344, "learning_rate": 7.071953852280081e-05, "loss": 0.19645072221755983, "memory(GiB)": 70.5, "step": 42480, "token_acc": 0.9628252788104089, "train_speed(iter/s)": 1.45274 }, { "epoch": 1.8201876526284222, "grad_norm": 4.69265604019165, "learning_rate": 7.071341357505437e-05, "loss": 0.4528674125671387, "memory(GiB)": 70.5, "step": 42485, "token_acc": 0.8996282527881041, "train_speed(iter/s)": 1.452742 }, { "epoch": 1.820401867957671, "grad_norm": 0.6030561923980713, "learning_rate": 7.070728825206501e-05, "loss": 0.3079576015472412, "memory(GiB)": 70.5, "step": 42490, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.452742 }, { "epoch": 1.82061608328692, "grad_norm": 2.840162515640259, "learning_rate": 7.070116255394371e-05, "loss": 0.5368343830108643, "memory(GiB)": 70.5, "step": 42495, "token_acc": 0.8898809523809523, "train_speed(iter/s)": 1.452746 }, { "epoch": 1.820830298616169, "grad_norm": 4.303774833679199, "learning_rate": 7.069503648080143e-05, "loss": 0.277114725112915, "memory(GiB)": 70.5, "step": 42500, "token_acc": 0.9240924092409241, "train_speed(iter/s)": 1.452742 }, { "epoch": 1.820830298616169, "eval_loss": 2.5561394691467285, "eval_runtime": 13.4668, "eval_samples_per_second": 7.426, "eval_steps_per_second": 7.426, "eval_token_acc": 0.4337194337194337, "step": 42500 }, { "epoch": 1.8210445139454179, "grad_norm": 3.501460552215576, "learning_rate": 7.068891003274915e-05, "loss": 0.21883187294006348, "memory(GiB)": 70.5, "step": 42505, "token_acc": 0.5658914728682171, "train_speed(iter/s)": 1.452033 }, { "epoch": 1.821258729274667, "grad_norm": 1.4700156450271606, "learning_rate": 7.068278320989785e-05, "loss": 0.19363255500793458, "memory(GiB)": 70.5, "step": 42510, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.452041 }, { "epoch": 1.821472944603916, "grad_norm": 5.665672779083252, "learning_rate": 7.067665601235853e-05, "loss": 0.3859058380126953, "memory(GiB)": 70.5, "step": 42515, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.452035 }, { "epoch": 1.8216871599331648, "grad_norm": 4.800632476806641, "learning_rate": 7.067052844024219e-05, "loss": 0.2698251485824585, "memory(GiB)": 70.5, "step": 42520, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.452024 }, { "epoch": 1.8219013752624138, "grad_norm": 4.062705039978027, "learning_rate": 7.066440049365983e-05, "loss": 0.527443790435791, "memory(GiB)": 70.5, "step": 42525, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.452018 }, { "epoch": 1.8221155905916628, "grad_norm": 2.816516399383545, "learning_rate": 7.06582721727225e-05, "loss": 0.3765134334564209, "memory(GiB)": 70.5, "step": 42530, "token_acc": 0.9153225806451613, "train_speed(iter/s)": 1.452023 }, { "epoch": 1.8223298059209116, "grad_norm": 3.577683448791504, "learning_rate": 7.065214347754115e-05, "loss": 0.32967305183410645, "memory(GiB)": 70.5, "step": 42535, "token_acc": 0.9249146757679181, "train_speed(iter/s)": 1.452018 }, { "epoch": 1.8225440212501607, "grad_norm": 5.057351589202881, "learning_rate": 7.064601440822688e-05, "loss": 0.09779543876647949, "memory(GiB)": 70.5, "step": 42540, "token_acc": 0.9759450171821306, "train_speed(iter/s)": 1.452008 }, { "epoch": 1.8227582365794097, "grad_norm": 2.1147522926330566, "learning_rate": 7.063988496489067e-05, "loss": 0.591908597946167, "memory(GiB)": 70.5, "step": 42545, "token_acc": 0.8859649122807017, "train_speed(iter/s)": 1.452033 }, { "epoch": 1.8229724519086585, "grad_norm": 5.1204938888549805, "learning_rate": 7.06337551476436e-05, "loss": 0.4653663635253906, "memory(GiB)": 70.5, "step": 42550, "token_acc": 0.8934169278996865, "train_speed(iter/s)": 1.452043 }, { "epoch": 1.8231866672379076, "grad_norm": 4.464017868041992, "learning_rate": 7.062762495659669e-05, "loss": 0.1456344723701477, "memory(GiB)": 70.5, "step": 42555, "token_acc": 0.9663299663299664, "train_speed(iter/s)": 1.452041 }, { "epoch": 1.8234008825671566, "grad_norm": 0.43220576643943787, "learning_rate": 7.062149439186101e-05, "loss": 0.30295934677124026, "memory(GiB)": 70.5, "step": 42560, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.452044 }, { "epoch": 1.8236150978964054, "grad_norm": 2.4799208641052246, "learning_rate": 7.06153634535476e-05, "loss": 0.24780383110046386, "memory(GiB)": 70.5, "step": 42565, "token_acc": 0.9446254071661238, "train_speed(iter/s)": 1.452037 }, { "epoch": 1.8238293132256544, "grad_norm": 2.082418203353882, "learning_rate": 7.060923214176756e-05, "loss": 0.24977493286132812, "memory(GiB)": 70.5, "step": 42570, "token_acc": 0.9399293286219081, "train_speed(iter/s)": 1.452039 }, { "epoch": 1.8240435285549035, "grad_norm": 3.3669331073760986, "learning_rate": 7.060310045663193e-05, "loss": 0.504559850692749, "memory(GiB)": 70.5, "step": 42575, "token_acc": 0.8964285714285715, "train_speed(iter/s)": 1.452071 }, { "epoch": 1.8242577438841523, "grad_norm": 1.3270509243011475, "learning_rate": 7.059696839825182e-05, "loss": 0.2889647245407104, "memory(GiB)": 70.5, "step": 42580, "token_acc": 0.9475806451612904, "train_speed(iter/s)": 1.452079 }, { "epoch": 1.8244719592134013, "grad_norm": 3.4456491470336914, "learning_rate": 7.05908359667383e-05, "loss": 0.5256254196166992, "memory(GiB)": 70.5, "step": 42585, "token_acc": 0.8837920489296636, "train_speed(iter/s)": 1.452082 }, { "epoch": 1.8246861745426504, "grad_norm": 5.40407133102417, "learning_rate": 7.058470316220248e-05, "loss": 0.4330785274505615, "memory(GiB)": 70.5, "step": 42590, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.452093 }, { "epoch": 1.8249003898718992, "grad_norm": 2.2225611209869385, "learning_rate": 7.057856998475543e-05, "loss": 0.39080142974853516, "memory(GiB)": 70.5, "step": 42595, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.452087 }, { "epoch": 1.8251146052011482, "grad_norm": 0.19693033397197723, "learning_rate": 7.05724364345083e-05, "loss": 0.4129174709320068, "memory(GiB)": 70.5, "step": 42600, "token_acc": 0.9262820512820513, "train_speed(iter/s)": 1.45209 }, { "epoch": 1.8253288205303972, "grad_norm": 3.9716742038726807, "learning_rate": 7.056630251157219e-05, "loss": 0.2712222099304199, "memory(GiB)": 70.5, "step": 42605, "token_acc": 0.941358024691358, "train_speed(iter/s)": 1.452088 }, { "epoch": 1.825543035859646, "grad_norm": 3.7778191566467285, "learning_rate": 7.05601682160582e-05, "loss": 0.14079880714416504, "memory(GiB)": 70.5, "step": 42610, "token_acc": 0.9806949806949807, "train_speed(iter/s)": 1.452103 }, { "epoch": 1.825757251188895, "grad_norm": 0.5728937983512878, "learning_rate": 7.055403354807749e-05, "loss": 0.3030005216598511, "memory(GiB)": 70.5, "step": 42615, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.452093 }, { "epoch": 1.8259714665181441, "grad_norm": 2.2669339179992676, "learning_rate": 7.054789850774118e-05, "loss": 0.29581642150878906, "memory(GiB)": 70.5, "step": 42620, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.452106 }, { "epoch": 1.826185681847393, "grad_norm": 5.552096366882324, "learning_rate": 7.054176309516041e-05, "loss": 0.9240379333496094, "memory(GiB)": 70.5, "step": 42625, "token_acc": 0.8376068376068376, "train_speed(iter/s)": 1.452115 }, { "epoch": 1.826399897176642, "grad_norm": 6.165074825286865, "learning_rate": 7.053562731044632e-05, "loss": 0.35542032718658445, "memory(GiB)": 70.5, "step": 42630, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.452156 }, { "epoch": 1.826614112505891, "grad_norm": 1.0480998754501343, "learning_rate": 7.05294911537101e-05, "loss": 0.32143187522888184, "memory(GiB)": 70.5, "step": 42635, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.45219 }, { "epoch": 1.8268283278351398, "grad_norm": 4.206269264221191, "learning_rate": 7.052335462506285e-05, "loss": 0.4819530487060547, "memory(GiB)": 70.5, "step": 42640, "token_acc": 0.8963414634146342, "train_speed(iter/s)": 1.452179 }, { "epoch": 1.8270425431643889, "grad_norm": 3.270240545272827, "learning_rate": 7.051721772461582e-05, "loss": 0.4516347885131836, "memory(GiB)": 70.5, "step": 42645, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.452175 }, { "epoch": 1.827256758493638, "grad_norm": 5.379746913909912, "learning_rate": 7.051108045248014e-05, "loss": 0.19847395420074462, "memory(GiB)": 70.5, "step": 42650, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.452174 }, { "epoch": 1.8274709738228867, "grad_norm": 5.589035987854004, "learning_rate": 7.050494280876697e-05, "loss": 0.793037223815918, "memory(GiB)": 70.5, "step": 42655, "token_acc": 0.8384146341463414, "train_speed(iter/s)": 1.452169 }, { "epoch": 1.8276851891521357, "grad_norm": 4.776634693145752, "learning_rate": 7.049880479358754e-05, "loss": 0.4490814208984375, "memory(GiB)": 70.5, "step": 42660, "token_acc": 0.8975265017667845, "train_speed(iter/s)": 1.452186 }, { "epoch": 1.8278994044813848, "grad_norm": 5.228368759155273, "learning_rate": 7.049266640705304e-05, "loss": 0.4671604156494141, "memory(GiB)": 70.5, "step": 42665, "token_acc": 0.9100719424460432, "train_speed(iter/s)": 1.452183 }, { "epoch": 1.8281136198106336, "grad_norm": 8.377272605895996, "learning_rate": 7.048652764927466e-05, "loss": 0.34523801803588866, "memory(GiB)": 70.5, "step": 42670, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.452181 }, { "epoch": 1.8283278351398826, "grad_norm": 4.249256134033203, "learning_rate": 7.04803885203636e-05, "loss": 0.3373737812042236, "memory(GiB)": 70.5, "step": 42675, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.452181 }, { "epoch": 1.8285420504691317, "grad_norm": 2.650662899017334, "learning_rate": 7.047424902043111e-05, "loss": 0.3512780427932739, "memory(GiB)": 70.5, "step": 42680, "token_acc": 0.9114391143911439, "train_speed(iter/s)": 1.452183 }, { "epoch": 1.8287562657983805, "grad_norm": 4.093042850494385, "learning_rate": 7.046810914958839e-05, "loss": 0.317500376701355, "memory(GiB)": 70.5, "step": 42685, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.452184 }, { "epoch": 1.8289704811276295, "grad_norm": 1.25660240650177, "learning_rate": 7.046196890794666e-05, "loss": 0.5863771438598633, "memory(GiB)": 70.5, "step": 42690, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.452186 }, { "epoch": 1.8291846964568785, "grad_norm": 4.7878336906433105, "learning_rate": 7.045582829561718e-05, "loss": 0.3611738681793213, "memory(GiB)": 70.5, "step": 42695, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.452185 }, { "epoch": 1.8293989117861273, "grad_norm": 2.535947322845459, "learning_rate": 7.044968731271118e-05, "loss": 0.4894257545471191, "memory(GiB)": 70.5, "step": 42700, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.452185 }, { "epoch": 1.8296131271153764, "grad_norm": 3.859196901321411, "learning_rate": 7.04435459593399e-05, "loss": 0.5130582809448242, "memory(GiB)": 70.5, "step": 42705, "token_acc": 0.8976897689768977, "train_speed(iter/s)": 1.45219 }, { "epoch": 1.8298273424446254, "grad_norm": 3.051485061645508, "learning_rate": 7.04374042356146e-05, "loss": 0.32922213077545165, "memory(GiB)": 70.5, "step": 42710, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.452187 }, { "epoch": 1.8300415577738742, "grad_norm": 4.161774158477783, "learning_rate": 7.043126214164657e-05, "loss": 0.5018106460571289, "memory(GiB)": 70.5, "step": 42715, "token_acc": 0.8766666666666667, "train_speed(iter/s)": 1.452208 }, { "epoch": 1.8302557731031233, "grad_norm": 1.9345529079437256, "learning_rate": 7.042511967754707e-05, "loss": 0.3667482852935791, "memory(GiB)": 70.5, "step": 42720, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.452211 }, { "epoch": 1.8304699884323723, "grad_norm": 3.0200681686401367, "learning_rate": 7.041897684342734e-05, "loss": 0.3110856533050537, "memory(GiB)": 70.5, "step": 42725, "token_acc": 0.924924924924925, "train_speed(iter/s)": 1.452203 }, { "epoch": 1.830684203761621, "grad_norm": 4.637816905975342, "learning_rate": 7.04128336393987e-05, "loss": 0.20720069408416747, "memory(GiB)": 70.5, "step": 42730, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.452207 }, { "epoch": 1.8308984190908701, "grad_norm": 7.448917388916016, "learning_rate": 7.040669006557245e-05, "loss": 0.5092107772827148, "memory(GiB)": 70.5, "step": 42735, "token_acc": 0.9113149847094801, "train_speed(iter/s)": 1.452217 }, { "epoch": 1.8311126344201192, "grad_norm": 3.5731375217437744, "learning_rate": 7.040054612205982e-05, "loss": 0.31483845710754393, "memory(GiB)": 70.5, "step": 42740, "token_acc": 0.9496644295302014, "train_speed(iter/s)": 1.452221 }, { "epoch": 1.831326849749368, "grad_norm": 2.481987237930298, "learning_rate": 7.03944018089722e-05, "loss": 0.23277430534362792, "memory(GiB)": 70.5, "step": 42745, "token_acc": 0.940809968847352, "train_speed(iter/s)": 1.452222 }, { "epoch": 1.831541065078617, "grad_norm": 2.130659818649292, "learning_rate": 7.038825712642085e-05, "loss": 0.5124423980712891, "memory(GiB)": 70.5, "step": 42750, "token_acc": 0.9037656903765691, "train_speed(iter/s)": 1.452233 }, { "epoch": 1.831755280407866, "grad_norm": 3.6477346420288086, "learning_rate": 7.03821120745171e-05, "loss": 0.3556635618209839, "memory(GiB)": 70.5, "step": 42755, "token_acc": 0.9236947791164659, "train_speed(iter/s)": 1.452223 }, { "epoch": 1.8319694957371149, "grad_norm": 1.5227099657058716, "learning_rate": 7.037596665337226e-05, "loss": 0.41545805931091306, "memory(GiB)": 70.5, "step": 42760, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.452226 }, { "epoch": 1.832183711066364, "grad_norm": 2.5857083797454834, "learning_rate": 7.036982086309769e-05, "loss": 0.3780375957489014, "memory(GiB)": 70.5, "step": 42765, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.452233 }, { "epoch": 1.832397926395613, "grad_norm": 6.2473225593566895, "learning_rate": 7.036367470380467e-05, "loss": 0.22647955417633056, "memory(GiB)": 70.5, "step": 42770, "token_acc": 0.946058091286307, "train_speed(iter/s)": 1.452237 }, { "epoch": 1.8326121417248618, "grad_norm": 4.223305702209473, "learning_rate": 7.03575281756046e-05, "loss": 0.3459761142730713, "memory(GiB)": 70.5, "step": 42775, "token_acc": 0.9326599326599326, "train_speed(iter/s)": 1.452236 }, { "epoch": 1.8328263570541108, "grad_norm": 4.1624531745910645, "learning_rate": 7.03513812786088e-05, "loss": 0.4928587913513184, "memory(GiB)": 70.5, "step": 42780, "token_acc": 0.9115853658536586, "train_speed(iter/s)": 1.452236 }, { "epoch": 1.8330405723833598, "grad_norm": 2.470017671585083, "learning_rate": 7.034523401292866e-05, "loss": 0.2901711702346802, "memory(GiB)": 70.5, "step": 42785, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.452251 }, { "epoch": 1.8332547877126086, "grad_norm": 1.4304897785186768, "learning_rate": 7.03390863786755e-05, "loss": 0.5824021816253662, "memory(GiB)": 70.5, "step": 42790, "token_acc": 0.8509933774834437, "train_speed(iter/s)": 1.452261 }, { "epoch": 1.8334690030418577, "grad_norm": 3.2496635913848877, "learning_rate": 7.033293837596073e-05, "loss": 0.23654301166534425, "memory(GiB)": 70.5, "step": 42795, "token_acc": 0.9571984435797666, "train_speed(iter/s)": 1.452285 }, { "epoch": 1.8336832183711067, "grad_norm": 3.9505650997161865, "learning_rate": 7.032679000489568e-05, "loss": 0.4777482032775879, "memory(GiB)": 70.5, "step": 42800, "token_acc": 0.8954423592493298, "train_speed(iter/s)": 1.452298 }, { "epoch": 1.8338974337003555, "grad_norm": 4.160984516143799, "learning_rate": 7.032064126559179e-05, "loss": 0.3130008697509766, "memory(GiB)": 70.5, "step": 42805, "token_acc": 0.930635838150289, "train_speed(iter/s)": 1.452303 }, { "epoch": 1.8341116490296046, "grad_norm": 4.672788143157959, "learning_rate": 7.031449215816041e-05, "loss": 0.3550281047821045, "memory(GiB)": 70.5, "step": 42810, "token_acc": 0.9169960474308301, "train_speed(iter/s)": 1.452301 }, { "epoch": 1.8343258643588536, "grad_norm": 2.550250768661499, "learning_rate": 7.030834268271294e-05, "loss": 0.46448574066162107, "memory(GiB)": 70.5, "step": 42815, "token_acc": 0.899390243902439, "train_speed(iter/s)": 1.45229 }, { "epoch": 1.8345400796881024, "grad_norm": 1.913807988166809, "learning_rate": 7.030219283936079e-05, "loss": 0.33357198238372804, "memory(GiB)": 70.5, "step": 42820, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.452287 }, { "epoch": 1.8347542950173514, "grad_norm": 7.18216609954834, "learning_rate": 7.029604262821539e-05, "loss": 0.2829891204833984, "memory(GiB)": 70.5, "step": 42825, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.452288 }, { "epoch": 1.8349685103466005, "grad_norm": 4.143774032592773, "learning_rate": 7.028989204938812e-05, "loss": 0.153847074508667, "memory(GiB)": 70.5, "step": 42830, "token_acc": 0.9630872483221476, "train_speed(iter/s)": 1.452312 }, { "epoch": 1.8351827256758493, "grad_norm": 3.9531195163726807, "learning_rate": 7.028374110299044e-05, "loss": 0.30200240612030027, "memory(GiB)": 70.5, "step": 42835, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.452316 }, { "epoch": 1.8353969410050983, "grad_norm": 10.92807674407959, "learning_rate": 7.027758978913374e-05, "loss": 0.5135953426361084, "memory(GiB)": 70.5, "step": 42840, "token_acc": 0.8590078328981723, "train_speed(iter/s)": 1.452309 }, { "epoch": 1.8356111563343473, "grad_norm": 3.8236083984375, "learning_rate": 7.02714381079295e-05, "loss": 0.38279540538787843, "memory(GiB)": 70.5, "step": 42845, "token_acc": 0.9066265060240963, "train_speed(iter/s)": 1.452332 }, { "epoch": 1.8358253716635962, "grad_norm": 2.426063299179077, "learning_rate": 7.026528605948913e-05, "loss": 0.3726980686187744, "memory(GiB)": 70.5, "step": 42850, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.45233 }, { "epoch": 1.8360395869928452, "grad_norm": 0.7303560972213745, "learning_rate": 7.025913364392409e-05, "loss": 0.21955695152282714, "memory(GiB)": 70.5, "step": 42855, "token_acc": 0.9540229885057471, "train_speed(iter/s)": 1.452319 }, { "epoch": 1.8362538023220942, "grad_norm": 2.9374139308929443, "learning_rate": 7.025298086134586e-05, "loss": 0.5311499118804932, "memory(GiB)": 70.5, "step": 42860, "token_acc": 0.8776758409785933, "train_speed(iter/s)": 1.452319 }, { "epoch": 1.836468017651343, "grad_norm": 3.6985416412353516, "learning_rate": 7.024682771186586e-05, "loss": 0.4897608757019043, "memory(GiB)": 70.5, "step": 42865, "token_acc": 0.9071428571428571, "train_speed(iter/s)": 1.452322 }, { "epoch": 1.836682232980592, "grad_norm": 4.381135940551758, "learning_rate": 7.02406741955956e-05, "loss": 0.3346710681915283, "memory(GiB)": 70.5, "step": 42870, "token_acc": 0.935672514619883, "train_speed(iter/s)": 1.452324 }, { "epoch": 1.8368964483098411, "grad_norm": 2.901416540145874, "learning_rate": 7.023452031264653e-05, "loss": 0.6165480613708496, "memory(GiB)": 70.5, "step": 42875, "token_acc": 0.8618618618618619, "train_speed(iter/s)": 1.452322 }, { "epoch": 1.83711066363909, "grad_norm": 3.4433188438415527, "learning_rate": 7.022836606313015e-05, "loss": 0.29783172607421876, "memory(GiB)": 70.5, "step": 42880, "token_acc": 0.9320754716981132, "train_speed(iter/s)": 1.452319 }, { "epoch": 1.837324878968339, "grad_norm": 1.460149884223938, "learning_rate": 7.022221144715793e-05, "loss": 0.27271504402160646, "memory(GiB)": 70.5, "step": 42885, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.452317 }, { "epoch": 1.837539094297588, "grad_norm": 5.211626052856445, "learning_rate": 7.021605646484137e-05, "loss": 0.3638000965118408, "memory(GiB)": 70.5, "step": 42890, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.452309 }, { "epoch": 1.8377533096268368, "grad_norm": 3.7342655658721924, "learning_rate": 7.020990111629202e-05, "loss": 0.457997465133667, "memory(GiB)": 70.5, "step": 42895, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.452304 }, { "epoch": 1.8379675249560858, "grad_norm": 4.213478088378906, "learning_rate": 7.020374540162132e-05, "loss": 0.3493010997772217, "memory(GiB)": 70.5, "step": 42900, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.452312 }, { "epoch": 1.8381817402853349, "grad_norm": 2.5736560821533203, "learning_rate": 7.019758932094083e-05, "loss": 0.17698792219161988, "memory(GiB)": 70.5, "step": 42905, "token_acc": 0.9586776859504132, "train_speed(iter/s)": 1.452311 }, { "epoch": 1.8383959556145837, "grad_norm": 4.139040470123291, "learning_rate": 7.019143287436207e-05, "loss": 0.538484525680542, "memory(GiB)": 70.5, "step": 42910, "token_acc": 0.8858131487889274, "train_speed(iter/s)": 1.452305 }, { "epoch": 1.8386101709438327, "grad_norm": 5.037586688995361, "learning_rate": 7.018527606199657e-05, "loss": 0.4728094577789307, "memory(GiB)": 70.5, "step": 42915, "token_acc": 0.888030888030888, "train_speed(iter/s)": 1.4523 }, { "epoch": 1.8388243862730818, "grad_norm": 2.8497843742370605, "learning_rate": 7.017911888395584e-05, "loss": 0.5388402938842773, "memory(GiB)": 70.5, "step": 42920, "token_acc": 0.8926380368098159, "train_speed(iter/s)": 1.452291 }, { "epoch": 1.8390386016023306, "grad_norm": 4.098728656768799, "learning_rate": 7.017296134035145e-05, "loss": 0.6061457633972168, "memory(GiB)": 70.5, "step": 42925, "token_acc": 0.8759124087591241, "train_speed(iter/s)": 1.452288 }, { "epoch": 1.8392528169315796, "grad_norm": 4.067751407623291, "learning_rate": 7.016680343129493e-05, "loss": 0.2674825668334961, "memory(GiB)": 70.5, "step": 42930, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.452302 }, { "epoch": 1.8394670322608286, "grad_norm": 4.372717380523682, "learning_rate": 7.016064515689787e-05, "loss": 0.5949650764465332, "memory(GiB)": 70.5, "step": 42935, "token_acc": 0.8705501618122977, "train_speed(iter/s)": 1.452308 }, { "epoch": 1.8396812475900775, "grad_norm": 2.3629939556121826, "learning_rate": 7.015448651727179e-05, "loss": 0.6535268783569336, "memory(GiB)": 70.5, "step": 42940, "token_acc": 0.8509933774834437, "train_speed(iter/s)": 1.452324 }, { "epoch": 1.8398954629193265, "grad_norm": 4.147362232208252, "learning_rate": 7.01483275125283e-05, "loss": 0.8477432250976562, "memory(GiB)": 70.5, "step": 42945, "token_acc": 0.8216783216783217, "train_speed(iter/s)": 1.452358 }, { "epoch": 1.8401096782485755, "grad_norm": 5.091159820556641, "learning_rate": 7.014216814277894e-05, "loss": 0.6038545608520508, "memory(GiB)": 70.5, "step": 42950, "token_acc": 0.8480565371024735, "train_speed(iter/s)": 1.452382 }, { "epoch": 1.8403238935778243, "grad_norm": 2.940087080001831, "learning_rate": 7.01360084081353e-05, "loss": 0.3455674409866333, "memory(GiB)": 70.5, "step": 42955, "token_acc": 0.9250814332247557, "train_speed(iter/s)": 1.45239 }, { "epoch": 1.8405381089070734, "grad_norm": 3.4878830909729004, "learning_rate": 7.0129848308709e-05, "loss": 0.18656045198440552, "memory(GiB)": 70.5, "step": 42960, "token_acc": 0.9501779359430605, "train_speed(iter/s)": 1.452409 }, { "epoch": 1.8407523242363224, "grad_norm": 1.8921613693237305, "learning_rate": 7.012368784461161e-05, "loss": 0.14174253940582277, "memory(GiB)": 70.5, "step": 42965, "token_acc": 0.956953642384106, "train_speed(iter/s)": 1.452412 }, { "epoch": 1.8409665395655712, "grad_norm": 4.53183126449585, "learning_rate": 7.011752701595471e-05, "loss": 0.345490837097168, "memory(GiB)": 70.5, "step": 42970, "token_acc": 0.9051094890510949, "train_speed(iter/s)": 1.452417 }, { "epoch": 1.8411807548948202, "grad_norm": 3.2722105979919434, "learning_rate": 7.011136582284998e-05, "loss": 0.5528882026672364, "memory(GiB)": 70.5, "step": 42975, "token_acc": 0.8885542168674698, "train_speed(iter/s)": 1.452418 }, { "epoch": 1.8413949702240693, "grad_norm": 2.68408465385437, "learning_rate": 7.010520426540897e-05, "loss": 0.27409336566925047, "memory(GiB)": 70.5, "step": 42980, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.45242 }, { "epoch": 1.841609185553318, "grad_norm": 7.348254680633545, "learning_rate": 7.009904234374332e-05, "loss": 0.6339303970336914, "memory(GiB)": 70.5, "step": 42985, "token_acc": 0.8692579505300353, "train_speed(iter/s)": 1.452421 }, { "epoch": 1.8418234008825671, "grad_norm": 3.016690969467163, "learning_rate": 7.009288005796469e-05, "loss": 0.40348329544067385, "memory(GiB)": 70.5, "step": 42990, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.452424 }, { "epoch": 1.8420376162118162, "grad_norm": 5.397497653961182, "learning_rate": 7.008671740818466e-05, "loss": 0.36996462345123293, "memory(GiB)": 70.5, "step": 42995, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.45243 }, { "epoch": 1.842251831541065, "grad_norm": 2.3398256301879883, "learning_rate": 7.008055439451491e-05, "loss": 0.23854057788848876, "memory(GiB)": 70.5, "step": 43000, "token_acc": 0.9429530201342282, "train_speed(iter/s)": 1.45243 }, { "epoch": 1.842251831541065, "eval_loss": 2.5321078300476074, "eval_runtime": 13.1372, "eval_samples_per_second": 7.612, "eval_steps_per_second": 7.612, "eval_token_acc": 0.44369063772048845, "step": 43000 }, { "epoch": 1.842466046870314, "grad_norm": 1.7719447612762451, "learning_rate": 7.007439101706708e-05, "loss": 0.46820878982543945, "memory(GiB)": 70.5, "step": 43005, "token_acc": 0.5689149560117303, "train_speed(iter/s)": 1.451733 }, { "epoch": 1.842680262199563, "grad_norm": 2.272819995880127, "learning_rate": 7.006822727595284e-05, "loss": 0.5328145980834961, "memory(GiB)": 70.5, "step": 43010, "token_acc": 0.8828125, "train_speed(iter/s)": 1.451733 }, { "epoch": 1.8428944775288119, "grad_norm": 1.9697456359863281, "learning_rate": 7.00620631712838e-05, "loss": 0.23933892250061034, "memory(GiB)": 70.5, "step": 43015, "token_acc": 0.9382352941176471, "train_speed(iter/s)": 1.451738 }, { "epoch": 1.843108692858061, "grad_norm": 2.941415309906006, "learning_rate": 7.00558987031717e-05, "loss": 0.40178236961364744, "memory(GiB)": 70.5, "step": 43020, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.451752 }, { "epoch": 1.84332290818731, "grad_norm": 4.101125240325928, "learning_rate": 7.004973387172818e-05, "loss": 0.43100662231445314, "memory(GiB)": 70.5, "step": 43025, "token_acc": 0.8778877887788779, "train_speed(iter/s)": 1.451756 }, { "epoch": 1.8435371235165587, "grad_norm": 0.6884315013885498, "learning_rate": 7.00435686770649e-05, "loss": 0.4832772254943848, "memory(GiB)": 70.5, "step": 43030, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.451768 }, { "epoch": 1.8437513388458078, "grad_norm": 1.664048433303833, "learning_rate": 7.003740311929358e-05, "loss": 0.16953375339508056, "memory(GiB)": 70.5, "step": 43035, "token_acc": 0.9640522875816994, "train_speed(iter/s)": 1.451772 }, { "epoch": 1.8439655541750568, "grad_norm": 1.621167778968811, "learning_rate": 7.003123719852592e-05, "loss": 0.3063133478164673, "memory(GiB)": 70.5, "step": 43040, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.451789 }, { "epoch": 1.8441797695043056, "grad_norm": 4.1443095207214355, "learning_rate": 7.00250709148736e-05, "loss": 0.32558958530426024, "memory(GiB)": 70.5, "step": 43045, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.451784 }, { "epoch": 1.8443939848335547, "grad_norm": 5.252964019775391, "learning_rate": 7.001890426844833e-05, "loss": 0.8616040229797364, "memory(GiB)": 70.5, "step": 43050, "token_acc": 0.8215384615384616, "train_speed(iter/s)": 1.451795 }, { "epoch": 1.8446082001628037, "grad_norm": 2.5389957427978516, "learning_rate": 7.001273725936184e-05, "loss": 0.5911857604980468, "memory(GiB)": 70.5, "step": 43055, "token_acc": 0.8997050147492626, "train_speed(iter/s)": 1.451792 }, { "epoch": 1.8448224154920525, "grad_norm": 2.198652505874634, "learning_rate": 7.000656988772583e-05, "loss": 0.20365595817565918, "memory(GiB)": 70.5, "step": 43060, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.451797 }, { "epoch": 1.8450366308213015, "grad_norm": 8.202596664428711, "learning_rate": 7.000040215365205e-05, "loss": 0.3729071617126465, "memory(GiB)": 70.5, "step": 43065, "token_acc": 0.9224489795918367, "train_speed(iter/s)": 1.451791 }, { "epoch": 1.8452508461505506, "grad_norm": 3.133600950241089, "learning_rate": 6.999423405725221e-05, "loss": 0.4747908592224121, "memory(GiB)": 70.5, "step": 43070, "token_acc": 0.889261744966443, "train_speed(iter/s)": 1.451781 }, { "epoch": 1.8454650614797994, "grad_norm": 2.3391730785369873, "learning_rate": 6.998806559863806e-05, "loss": 0.32243316173553466, "memory(GiB)": 70.5, "step": 43075, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.451786 }, { "epoch": 1.8456792768090484, "grad_norm": 0.43941423296928406, "learning_rate": 6.998189677792138e-05, "loss": 0.26106510162353513, "memory(GiB)": 70.5, "step": 43080, "token_acc": 0.9483282674772037, "train_speed(iter/s)": 1.45181 }, { "epoch": 1.8458934921382975, "grad_norm": 2.2247822284698486, "learning_rate": 6.997572759521386e-05, "loss": 0.43874435424804686, "memory(GiB)": 70.5, "step": 43085, "token_acc": 0.9141104294478528, "train_speed(iter/s)": 1.451828 }, { "epoch": 1.8461077074675463, "grad_norm": 3.0237865447998047, "learning_rate": 6.996955805062732e-05, "loss": 0.4500910758972168, "memory(GiB)": 70.5, "step": 43090, "token_acc": 0.9044117647058824, "train_speed(iter/s)": 1.45183 }, { "epoch": 1.8463219227967953, "grad_norm": 2.5088894367218018, "learning_rate": 6.99633881442735e-05, "loss": 0.3569206237792969, "memory(GiB)": 70.5, "step": 43095, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.45183 }, { "epoch": 1.8465361381260443, "grad_norm": 2.6046085357666016, "learning_rate": 6.995721787626417e-05, "loss": 0.2734581232070923, "memory(GiB)": 70.5, "step": 43100, "token_acc": 0.950381679389313, "train_speed(iter/s)": 1.451823 }, { "epoch": 1.8467503534552931, "grad_norm": 3.038236379623413, "learning_rate": 6.995104724671112e-05, "loss": 0.3654165506362915, "memory(GiB)": 70.5, "step": 43105, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.451827 }, { "epoch": 1.8469645687845422, "grad_norm": 2.452018976211548, "learning_rate": 6.994487625572613e-05, "loss": 0.3357478141784668, "memory(GiB)": 70.5, "step": 43110, "token_acc": 0.9036144578313253, "train_speed(iter/s)": 1.451863 }, { "epoch": 1.8471787841137912, "grad_norm": 1.5720901489257812, "learning_rate": 6.993870490342099e-05, "loss": 0.33622863292694094, "memory(GiB)": 70.5, "step": 43115, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.45187 }, { "epoch": 1.84739299944304, "grad_norm": 1.1599267721176147, "learning_rate": 6.993253318990753e-05, "loss": 0.19651631116867066, "memory(GiB)": 70.5, "step": 43120, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.451866 }, { "epoch": 1.847607214772289, "grad_norm": 3.432929754257202, "learning_rate": 6.992636111529752e-05, "loss": 0.5959908485412597, "memory(GiB)": 70.5, "step": 43125, "token_acc": 0.8724137931034482, "train_speed(iter/s)": 1.451873 }, { "epoch": 1.847821430101538, "grad_norm": 3.7462170124053955, "learning_rate": 6.992018867970278e-05, "loss": 0.48769383430480956, "memory(GiB)": 70.5, "step": 43130, "token_acc": 0.9049586776859504, "train_speed(iter/s)": 1.451868 }, { "epoch": 1.848035645430787, "grad_norm": 2.7982900142669678, "learning_rate": 6.991401588323514e-05, "loss": 0.4872868061065674, "memory(GiB)": 70.5, "step": 43135, "token_acc": 0.887459807073955, "train_speed(iter/s)": 1.45188 }, { "epoch": 1.848249860760036, "grad_norm": 1.526465892791748, "learning_rate": 6.990784272600643e-05, "loss": 0.499970006942749, "memory(GiB)": 70.5, "step": 43140, "token_acc": 0.8996282527881041, "train_speed(iter/s)": 1.451867 }, { "epoch": 1.848464076089285, "grad_norm": 1.250643014907837, "learning_rate": 6.990166920812847e-05, "loss": 0.32599468231201173, "memory(GiB)": 70.5, "step": 43145, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.451877 }, { "epoch": 1.8486782914185338, "grad_norm": 4.9753193855285645, "learning_rate": 6.989549532971309e-05, "loss": 0.32602341175079347, "memory(GiB)": 70.5, "step": 43150, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.451894 }, { "epoch": 1.8488925067477828, "grad_norm": 3.5867185592651367, "learning_rate": 6.988932109087216e-05, "loss": 0.5012857913970947, "memory(GiB)": 70.5, "step": 43155, "token_acc": 0.8884615384615384, "train_speed(iter/s)": 1.451911 }, { "epoch": 1.8491067220770319, "grad_norm": 4.611388683319092, "learning_rate": 6.988314649171751e-05, "loss": 0.4457257270812988, "memory(GiB)": 70.5, "step": 43160, "token_acc": 0.9114285714285715, "train_speed(iter/s)": 1.451929 }, { "epoch": 1.8493209374062807, "grad_norm": 2.6022884845733643, "learning_rate": 6.987697153236102e-05, "loss": 0.43273372650146485, "memory(GiB)": 70.5, "step": 43165, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.451929 }, { "epoch": 1.8495351527355297, "grad_norm": 3.5008883476257324, "learning_rate": 6.987079621291455e-05, "loss": 0.4505763530731201, "memory(GiB)": 70.5, "step": 43170, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.451953 }, { "epoch": 1.8497493680647787, "grad_norm": 2.9184577465057373, "learning_rate": 6.986462053348996e-05, "loss": 0.36132230758666994, "memory(GiB)": 70.5, "step": 43175, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.451973 }, { "epoch": 1.8499635833940276, "grad_norm": 1.4097775220870972, "learning_rate": 6.985844449419913e-05, "loss": 0.5152536869049072, "memory(GiB)": 70.5, "step": 43180, "token_acc": 0.8831521739130435, "train_speed(iter/s)": 1.451973 }, { "epoch": 1.8501777987232766, "grad_norm": 5.996248245239258, "learning_rate": 6.985226809515395e-05, "loss": 0.5750139713287353, "memory(GiB)": 70.5, "step": 43185, "token_acc": 0.86875, "train_speed(iter/s)": 1.451976 }, { "epoch": 1.8503920140525256, "grad_norm": 3.3981871604919434, "learning_rate": 6.98460913364663e-05, "loss": 0.15177100896835327, "memory(GiB)": 70.5, "step": 43190, "token_acc": 0.9662921348314607, "train_speed(iter/s)": 1.451982 }, { "epoch": 1.8506062293817744, "grad_norm": 0.42295652627944946, "learning_rate": 6.983991421824811e-05, "loss": 0.39751324653625486, "memory(GiB)": 70.5, "step": 43195, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.451982 }, { "epoch": 1.8508204447110235, "grad_norm": 3.4271392822265625, "learning_rate": 6.983373674061126e-05, "loss": 0.4564621925354004, "memory(GiB)": 70.5, "step": 43200, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.451992 }, { "epoch": 1.8510346600402725, "grad_norm": 6.369333267211914, "learning_rate": 6.982755890366766e-05, "loss": 0.2853787183761597, "memory(GiB)": 70.5, "step": 43205, "token_acc": 0.9380165289256198, "train_speed(iter/s)": 1.45201 }, { "epoch": 1.8512488753695213, "grad_norm": 6.249757289886475, "learning_rate": 6.982138070752923e-05, "loss": 0.6390312194824219, "memory(GiB)": 70.5, "step": 43210, "token_acc": 0.8662207357859532, "train_speed(iter/s)": 1.452006 }, { "epoch": 1.8514630906987704, "grad_norm": 4.198172569274902, "learning_rate": 6.981520215230788e-05, "loss": 0.5384607315063477, "memory(GiB)": 70.5, "step": 43215, "token_acc": 0.8931750741839762, "train_speed(iter/s)": 1.452025 }, { "epoch": 1.8516773060280194, "grad_norm": 3.53922176361084, "learning_rate": 6.980902323811557e-05, "loss": 0.44115939140319826, "memory(GiB)": 70.5, "step": 43220, "token_acc": 0.879245283018868, "train_speed(iter/s)": 1.452022 }, { "epoch": 1.8518915213572682, "grad_norm": 4.4631452560424805, "learning_rate": 6.980284396506421e-05, "loss": 0.601005744934082, "memory(GiB)": 70.5, "step": 43225, "token_acc": 0.8827361563517915, "train_speed(iter/s)": 1.452015 }, { "epoch": 1.8521057366865172, "grad_norm": 2.8383147716522217, "learning_rate": 6.979666433326577e-05, "loss": 0.22453293800354004, "memory(GiB)": 70.5, "step": 43230, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.452031 }, { "epoch": 1.8523199520157663, "grad_norm": 4.014501571655273, "learning_rate": 6.979048434283218e-05, "loss": 0.33372044563293457, "memory(GiB)": 70.5, "step": 43235, "token_acc": 0.9342857142857143, "train_speed(iter/s)": 1.452041 }, { "epoch": 1.852534167345015, "grad_norm": 2.42706298828125, "learning_rate": 6.978430399387541e-05, "loss": 0.5163267135620118, "memory(GiB)": 70.5, "step": 43240, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.452054 }, { "epoch": 1.8527483826742641, "grad_norm": 3.0776987075805664, "learning_rate": 6.97781232865074e-05, "loss": 0.3994445323944092, "memory(GiB)": 70.5, "step": 43245, "token_acc": 0.901685393258427, "train_speed(iter/s)": 1.452045 }, { "epoch": 1.8529625980035132, "grad_norm": 0.7160181999206543, "learning_rate": 6.977194222084013e-05, "loss": 0.21154017448425294, "memory(GiB)": 70.5, "step": 43250, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.452053 }, { "epoch": 1.853176813332762, "grad_norm": 1.3541107177734375, "learning_rate": 6.97657607969856e-05, "loss": 0.28443112373352053, "memory(GiB)": 70.5, "step": 43255, "token_acc": 0.953125, "train_speed(iter/s)": 1.452055 }, { "epoch": 1.8533910286620112, "grad_norm": 2.7590508460998535, "learning_rate": 6.975957901505574e-05, "loss": 0.39308733940124513, "memory(GiB)": 70.5, "step": 43260, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.452065 }, { "epoch": 1.85360524399126, "grad_norm": 3.000147581100464, "learning_rate": 6.97533968751626e-05, "loss": 0.2844958543777466, "memory(GiB)": 70.5, "step": 43265, "token_acc": 0.9299610894941635, "train_speed(iter/s)": 1.452066 }, { "epoch": 1.8538194593205088, "grad_norm": 3.3251333236694336, "learning_rate": 6.974721437741813e-05, "loss": 0.39196953773498533, "memory(GiB)": 70.5, "step": 43270, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.452065 }, { "epoch": 1.854033674649758, "grad_norm": 4.016794681549072, "learning_rate": 6.974103152193434e-05, "loss": 0.6108639717102051, "memory(GiB)": 70.5, "step": 43275, "token_acc": 0.8645418326693227, "train_speed(iter/s)": 1.452097 }, { "epoch": 1.854247889979007, "grad_norm": 1.7773137092590332, "learning_rate": 6.973484830882326e-05, "loss": 0.20001142024993895, "memory(GiB)": 70.5, "step": 43280, "token_acc": 0.9456066945606695, "train_speed(iter/s)": 1.452109 }, { "epoch": 1.8544621053082557, "grad_norm": 2.58038067817688, "learning_rate": 6.97286647381969e-05, "loss": 0.3353499174118042, "memory(GiB)": 70.5, "step": 43285, "token_acc": 0.9305019305019305, "train_speed(iter/s)": 1.452115 }, { "epoch": 1.854676320637505, "grad_norm": 4.191169261932373, "learning_rate": 6.972248081016724e-05, "loss": 0.5535025596618652, "memory(GiB)": 70.5, "step": 43290, "token_acc": 0.8671875, "train_speed(iter/s)": 1.452112 }, { "epoch": 1.8548905359667538, "grad_norm": 2.499180793762207, "learning_rate": 6.971629652484635e-05, "loss": 0.3293895244598389, "memory(GiB)": 70.5, "step": 43295, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.45211 }, { "epoch": 1.8551047512960026, "grad_norm": 4.024322032928467, "learning_rate": 6.971011188234628e-05, "loss": 0.5951536655426025, "memory(GiB)": 70.5, "step": 43300, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.452129 }, { "epoch": 1.8553189666252519, "grad_norm": 4.957983493804932, "learning_rate": 6.9703926882779e-05, "loss": 0.3531531095504761, "memory(GiB)": 70.5, "step": 43305, "token_acc": 0.9227941176470589, "train_speed(iter/s)": 1.452129 }, { "epoch": 1.8555331819545007, "grad_norm": 1.604386568069458, "learning_rate": 6.969774152625664e-05, "loss": 0.3316755771636963, "memory(GiB)": 70.5, "step": 43310, "token_acc": 0.9158249158249159, "train_speed(iter/s)": 1.452157 }, { "epoch": 1.8557473972837495, "grad_norm": 2.459630250930786, "learning_rate": 6.969155581289119e-05, "loss": 0.18193318843841552, "memory(GiB)": 70.5, "step": 43315, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.452151 }, { "epoch": 1.8559616126129987, "grad_norm": 3.275463819503784, "learning_rate": 6.968536974279475e-05, "loss": 0.5173603057861328, "memory(GiB)": 70.5, "step": 43320, "token_acc": 0.8695652173913043, "train_speed(iter/s)": 1.452148 }, { "epoch": 1.8561758279422476, "grad_norm": 0.6903193593025208, "learning_rate": 6.967918331607937e-05, "loss": 0.3392236948013306, "memory(GiB)": 70.5, "step": 43325, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.452149 }, { "epoch": 1.8563900432714964, "grad_norm": 3.139601707458496, "learning_rate": 6.967299653285711e-05, "loss": 0.6634061336517334, "memory(GiB)": 70.5, "step": 43330, "token_acc": 0.8641114982578397, "train_speed(iter/s)": 1.452166 }, { "epoch": 1.8566042586007456, "grad_norm": 1.0848972797393799, "learning_rate": 6.966680939324006e-05, "loss": 0.27934608459472654, "memory(GiB)": 70.5, "step": 43335, "token_acc": 0.9313725490196079, "train_speed(iter/s)": 1.452171 }, { "epoch": 1.8568184739299944, "grad_norm": 5.303282737731934, "learning_rate": 6.966062189734033e-05, "loss": 0.20262598991394043, "memory(GiB)": 70.5, "step": 43340, "token_acc": 0.9406392694063926, "train_speed(iter/s)": 1.452188 }, { "epoch": 1.8570326892592433, "grad_norm": 3.1308023929595947, "learning_rate": 6.965443404526998e-05, "loss": 0.26539173126220705, "memory(GiB)": 70.5, "step": 43345, "token_acc": 0.9475409836065574, "train_speed(iter/s)": 1.452187 }, { "epoch": 1.8572469045884925, "grad_norm": 2.7730841636657715, "learning_rate": 6.964824583714111e-05, "loss": 0.34964728355407715, "memory(GiB)": 70.5, "step": 43350, "token_acc": 0.9209726443768997, "train_speed(iter/s)": 1.452191 }, { "epoch": 1.8574611199177413, "grad_norm": 3.9684691429138184, "learning_rate": 6.964205727306586e-05, "loss": 0.2767913818359375, "memory(GiB)": 70.5, "step": 43355, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.452203 }, { "epoch": 1.8576753352469901, "grad_norm": 1.2740353345870972, "learning_rate": 6.963586835315629e-05, "loss": 0.47815465927124023, "memory(GiB)": 70.5, "step": 43360, "token_acc": 0.8984615384615384, "train_speed(iter/s)": 1.452194 }, { "epoch": 1.8578895505762394, "grad_norm": 5.525235176086426, "learning_rate": 6.962967907752455e-05, "loss": 0.6539225578308105, "memory(GiB)": 70.5, "step": 43365, "token_acc": 0.8515625, "train_speed(iter/s)": 1.452209 }, { "epoch": 1.8581037659054882, "grad_norm": 3.5335328578948975, "learning_rate": 6.962348944628276e-05, "loss": 0.6153782367706299, "memory(GiB)": 70.5, "step": 43370, "token_acc": 0.8308823529411765, "train_speed(iter/s)": 1.45224 }, { "epoch": 1.858317981234737, "grad_norm": 5.783607006072998, "learning_rate": 6.961729945954307e-05, "loss": 0.36401140689849854, "memory(GiB)": 70.5, "step": 43375, "token_acc": 0.9328358208955224, "train_speed(iter/s)": 1.452244 }, { "epoch": 1.8585321965639863, "grad_norm": 8.841778755187988, "learning_rate": 6.961110911741757e-05, "loss": 0.4235715389251709, "memory(GiB)": 70.5, "step": 43380, "token_acc": 0.9121338912133892, "train_speed(iter/s)": 1.452243 }, { "epoch": 1.858746411893235, "grad_norm": 1.8918207883834839, "learning_rate": 6.960491842001846e-05, "loss": 0.38731701374053956, "memory(GiB)": 70.5, "step": 43385, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.452267 }, { "epoch": 1.858960627222484, "grad_norm": 3.1148719787597656, "learning_rate": 6.959872736745784e-05, "loss": 0.5517853736877442, "memory(GiB)": 70.5, "step": 43390, "token_acc": 0.875, "train_speed(iter/s)": 1.452297 }, { "epoch": 1.8591748425517332, "grad_norm": 1.6455812454223633, "learning_rate": 6.95925359598479e-05, "loss": 0.197260844707489, "memory(GiB)": 70.5, "step": 43395, "token_acc": 0.9626666666666667, "train_speed(iter/s)": 1.452311 }, { "epoch": 1.859389057880982, "grad_norm": 2.9055323600769043, "learning_rate": 6.95863441973008e-05, "loss": 0.3576691150665283, "memory(GiB)": 70.5, "step": 43400, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.452322 }, { "epoch": 1.8596032732102308, "grad_norm": 0.6349327564239502, "learning_rate": 6.958015207992867e-05, "loss": 0.32574336528778075, "memory(GiB)": 70.5, "step": 43405, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.452342 }, { "epoch": 1.85981748853948, "grad_norm": 1.679184913635254, "learning_rate": 6.957395960784374e-05, "loss": 0.19928351640701295, "memory(GiB)": 70.5, "step": 43410, "token_acc": 0.9535864978902954, "train_speed(iter/s)": 1.45235 }, { "epoch": 1.8600317038687288, "grad_norm": 2.667829751968384, "learning_rate": 6.956776678115817e-05, "loss": 0.5308109283447265, "memory(GiB)": 70.5, "step": 43415, "token_acc": 0.8901734104046243, "train_speed(iter/s)": 1.452376 }, { "epoch": 1.8602459191979777, "grad_norm": 5.793257236480713, "learning_rate": 6.956157359998414e-05, "loss": 0.5274762153625489, "memory(GiB)": 70.5, "step": 43420, "token_acc": 0.8808664259927798, "train_speed(iter/s)": 1.452375 }, { "epoch": 1.860460134527227, "grad_norm": 5.278283596038818, "learning_rate": 6.955538006443386e-05, "loss": 0.8253595352172851, "memory(GiB)": 70.5, "step": 43425, "token_acc": 0.835820895522388, "train_speed(iter/s)": 1.452386 }, { "epoch": 1.8606743498564757, "grad_norm": 3.419365406036377, "learning_rate": 6.954918617461952e-05, "loss": 0.31203255653381345, "memory(GiB)": 70.5, "step": 43430, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.452393 }, { "epoch": 1.8608885651857245, "grad_norm": 2.6483054161071777, "learning_rate": 6.954299193065334e-05, "loss": 0.50248122215271, "memory(GiB)": 70.5, "step": 43435, "token_acc": 0.90234375, "train_speed(iter/s)": 1.452388 }, { "epoch": 1.8611027805149738, "grad_norm": 4.296971797943115, "learning_rate": 6.953679733264753e-05, "loss": 0.5801695823669434, "memory(GiB)": 70.5, "step": 43440, "token_acc": 0.8801652892561983, "train_speed(iter/s)": 1.452387 }, { "epoch": 1.8613169958442226, "grad_norm": 4.218311309814453, "learning_rate": 6.95306023807143e-05, "loss": 0.39118242263793945, "memory(GiB)": 70.5, "step": 43445, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.452394 }, { "epoch": 1.8615312111734714, "grad_norm": 1.9020854234695435, "learning_rate": 6.952440707496589e-05, "loss": 0.1624547004699707, "memory(GiB)": 70.5, "step": 43450, "token_acc": 0.9576547231270358, "train_speed(iter/s)": 1.452393 }, { "epoch": 1.8617454265027207, "grad_norm": 1.8589808940887451, "learning_rate": 6.951821141551455e-05, "loss": 0.30275650024414064, "memory(GiB)": 70.5, "step": 43455, "token_acc": 0.9498432601880877, "train_speed(iter/s)": 1.452395 }, { "epoch": 1.8619596418319695, "grad_norm": 1.9545197486877441, "learning_rate": 6.951201540247249e-05, "loss": 0.30533580780029296, "memory(GiB)": 70.5, "step": 43460, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.452393 }, { "epoch": 1.8621738571612183, "grad_norm": 4.043076515197754, "learning_rate": 6.950581903595194e-05, "loss": 0.3035188913345337, "memory(GiB)": 70.5, "step": 43465, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.45239 }, { "epoch": 1.8623880724904676, "grad_norm": 3.341001272201538, "learning_rate": 6.949962231606522e-05, "loss": 0.2193384885787964, "memory(GiB)": 70.5, "step": 43470, "token_acc": 0.9532374100719424, "train_speed(iter/s)": 1.452389 }, { "epoch": 1.8626022878197164, "grad_norm": 5.355679035186768, "learning_rate": 6.949342524292453e-05, "loss": 0.3853058099746704, "memory(GiB)": 70.5, "step": 43475, "token_acc": 0.91796875, "train_speed(iter/s)": 1.452401 }, { "epoch": 1.8628165031489652, "grad_norm": 3.324726104736328, "learning_rate": 6.948722781664215e-05, "loss": 0.29797921180725095, "memory(GiB)": 70.5, "step": 43480, "token_acc": 0.9409448818897638, "train_speed(iter/s)": 1.452404 }, { "epoch": 1.8630307184782144, "grad_norm": 2.976569175720215, "learning_rate": 6.948103003733036e-05, "loss": 0.3039536476135254, "memory(GiB)": 70.5, "step": 43485, "token_acc": 0.9312977099236641, "train_speed(iter/s)": 1.452404 }, { "epoch": 1.8632449338074633, "grad_norm": 3.80293869972229, "learning_rate": 6.947483190510144e-05, "loss": 0.5334471702575684, "memory(GiB)": 70.5, "step": 43490, "token_acc": 0.8686567164179104, "train_speed(iter/s)": 1.452421 }, { "epoch": 1.863459149136712, "grad_norm": 2.4530014991760254, "learning_rate": 6.946863342006768e-05, "loss": 0.3007020950317383, "memory(GiB)": 70.5, "step": 43495, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.452412 }, { "epoch": 1.8636733644659613, "grad_norm": 3.9361064434051514, "learning_rate": 6.946243458234135e-05, "loss": 0.5257722854614257, "memory(GiB)": 70.5, "step": 43500, "token_acc": 0.9067164179104478, "train_speed(iter/s)": 1.452422 }, { "epoch": 1.8636733644659613, "eval_loss": 2.5297725200653076, "eval_runtime": 13.5233, "eval_samples_per_second": 7.395, "eval_steps_per_second": 7.395, "eval_token_acc": 0.41851368970013036, "step": 43500 }, { "epoch": 1.8638875797952101, "grad_norm": 2.91582989692688, "learning_rate": 6.945623539203477e-05, "loss": 0.5600770950317383, "memory(GiB)": 70.5, "step": 43505, "token_acc": 0.5507380073800738, "train_speed(iter/s)": 1.451714 }, { "epoch": 1.864101795124459, "grad_norm": 2.5948116779327393, "learning_rate": 6.945003584926021e-05, "loss": 0.23923563957214355, "memory(GiB)": 70.5, "step": 43510, "token_acc": 0.9489795918367347, "train_speed(iter/s)": 1.451718 }, { "epoch": 1.8643160104537082, "grad_norm": 2.566500425338745, "learning_rate": 6.944383595413003e-05, "loss": 0.35009849071502686, "memory(GiB)": 70.5, "step": 43515, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.451726 }, { "epoch": 1.864530225782957, "grad_norm": 5.331145286560059, "learning_rate": 6.943763570675652e-05, "loss": 0.11240289211273194, "memory(GiB)": 70.5, "step": 43520, "token_acc": 0.9709090909090909, "train_speed(iter/s)": 1.451734 }, { "epoch": 1.8647444411122058, "grad_norm": 3.562998056411743, "learning_rate": 6.943143510725199e-05, "loss": 0.5163163661956787, "memory(GiB)": 70.5, "step": 43525, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.451729 }, { "epoch": 1.864958656441455, "grad_norm": 4.2298688888549805, "learning_rate": 6.942523415572882e-05, "loss": 0.2262502908706665, "memory(GiB)": 70.5, "step": 43530, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.451736 }, { "epoch": 1.865172871770704, "grad_norm": 4.163058280944824, "learning_rate": 6.941903285229927e-05, "loss": 0.23749728202819825, "memory(GiB)": 70.5, "step": 43535, "token_acc": 0.9462809917355371, "train_speed(iter/s)": 1.451729 }, { "epoch": 1.8653870870999527, "grad_norm": 4.0319366455078125, "learning_rate": 6.941283119707575e-05, "loss": 0.2893799066543579, "memory(GiB)": 70.5, "step": 43540, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.451716 }, { "epoch": 1.865601302429202, "grad_norm": 1.924960970878601, "learning_rate": 6.940662919017057e-05, "loss": 0.18458034992218017, "memory(GiB)": 70.5, "step": 43545, "token_acc": 0.9552715654952076, "train_speed(iter/s)": 1.451718 }, { "epoch": 1.8658155177584508, "grad_norm": 3.044846773147583, "learning_rate": 6.940042683169609e-05, "loss": 0.38361122608184817, "memory(GiB)": 70.5, "step": 43550, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.451732 }, { "epoch": 1.8660297330876996, "grad_norm": 5.232573986053467, "learning_rate": 6.939422412176468e-05, "loss": 0.2459540843963623, "memory(GiB)": 70.5, "step": 43555, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.451734 }, { "epoch": 1.8662439484169489, "grad_norm": 2.9669113159179688, "learning_rate": 6.938802106048873e-05, "loss": 0.1748797059059143, "memory(GiB)": 70.5, "step": 43560, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.451736 }, { "epoch": 1.8664581637461977, "grad_norm": 6.697007656097412, "learning_rate": 6.938181764798058e-05, "loss": 0.5726563930511475, "memory(GiB)": 70.5, "step": 43565, "token_acc": 0.8850931677018633, "train_speed(iter/s)": 1.451768 }, { "epoch": 1.8666723790754465, "grad_norm": 5.490982532501221, "learning_rate": 6.93756138843526e-05, "loss": 0.4686111927032471, "memory(GiB)": 70.5, "step": 43570, "token_acc": 0.901840490797546, "train_speed(iter/s)": 1.451755 }, { "epoch": 1.8668865944046957, "grad_norm": 6.880911350250244, "learning_rate": 6.936940976971724e-05, "loss": 0.39052581787109375, "memory(GiB)": 70.5, "step": 43575, "token_acc": 0.92578125, "train_speed(iter/s)": 1.451755 }, { "epoch": 1.8671008097339445, "grad_norm": 3.1940581798553467, "learning_rate": 6.936320530418681e-05, "loss": 0.5358729362487793, "memory(GiB)": 70.5, "step": 43580, "token_acc": 0.89501312335958, "train_speed(iter/s)": 1.451756 }, { "epoch": 1.8673150250631934, "grad_norm": 2.9318244457244873, "learning_rate": 6.935700048787379e-05, "loss": 0.605244255065918, "memory(GiB)": 70.5, "step": 43585, "token_acc": 0.8463949843260188, "train_speed(iter/s)": 1.451747 }, { "epoch": 1.8675292403924426, "grad_norm": 4.106739044189453, "learning_rate": 6.935079532089052e-05, "loss": 0.40794572830200193, "memory(GiB)": 70.5, "step": 43590, "token_acc": 0.9104938271604939, "train_speed(iter/s)": 1.451761 }, { "epoch": 1.8677434557216914, "grad_norm": 2.6636195182800293, "learning_rate": 6.934458980334944e-05, "loss": 0.3609644412994385, "memory(GiB)": 70.5, "step": 43595, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.451765 }, { "epoch": 1.8679576710509405, "grad_norm": 3.9499430656433105, "learning_rate": 6.933838393536299e-05, "loss": 0.39779510498046877, "memory(GiB)": 70.5, "step": 43600, "token_acc": 0.910958904109589, "train_speed(iter/s)": 1.451776 }, { "epoch": 1.8681718863801895, "grad_norm": 5.516026496887207, "learning_rate": 6.933217771704356e-05, "loss": 0.46988439559936523, "memory(GiB)": 70.5, "step": 43605, "token_acc": 0.9037037037037037, "train_speed(iter/s)": 1.451769 }, { "epoch": 1.8683861017094383, "grad_norm": 1.2714909315109253, "learning_rate": 6.932597114850359e-05, "loss": 0.414174747467041, "memory(GiB)": 70.5, "step": 43610, "token_acc": 0.9142011834319527, "train_speed(iter/s)": 1.451762 }, { "epoch": 1.8686003170386873, "grad_norm": 0.7864627838134766, "learning_rate": 6.931976422985554e-05, "loss": 0.1680317997932434, "memory(GiB)": 70.5, "step": 43615, "token_acc": 0.9696969696969697, "train_speed(iter/s)": 1.451766 }, { "epoch": 1.8688145323679364, "grad_norm": 4.907686233520508, "learning_rate": 6.931355696121183e-05, "loss": 0.4362056732177734, "memory(GiB)": 70.5, "step": 43620, "token_acc": 0.8961538461538462, "train_speed(iter/s)": 1.451776 }, { "epoch": 1.8690287476971852, "grad_norm": 1.3573800325393677, "learning_rate": 6.930734934268491e-05, "loss": 0.3328873157501221, "memory(GiB)": 70.5, "step": 43625, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.451766 }, { "epoch": 1.8692429630264342, "grad_norm": 1.6737076044082642, "learning_rate": 6.930114137438725e-05, "loss": 0.11362202167510986, "memory(GiB)": 70.5, "step": 43630, "token_acc": 0.9704918032786886, "train_speed(iter/s)": 1.451771 }, { "epoch": 1.8694571783556833, "grad_norm": 4.6457085609436035, "learning_rate": 6.92949330564313e-05, "loss": 0.48972234725952146, "memory(GiB)": 70.5, "step": 43635, "token_acc": 0.9139344262295082, "train_speed(iter/s)": 1.451783 }, { "epoch": 1.869671393684932, "grad_norm": 2.002485513687134, "learning_rate": 6.928872438892956e-05, "loss": 0.3690300464630127, "memory(GiB)": 70.5, "step": 43640, "token_acc": 0.9112426035502958, "train_speed(iter/s)": 1.451794 }, { "epoch": 1.869885609014181, "grad_norm": 3.3380696773529053, "learning_rate": 6.928251537199446e-05, "loss": 0.3631249189376831, "memory(GiB)": 70.5, "step": 43645, "token_acc": 0.91875, "train_speed(iter/s)": 1.451814 }, { "epoch": 1.8700998243434301, "grad_norm": 3.6107845306396484, "learning_rate": 6.927630600573851e-05, "loss": 0.28763561248779296, "memory(GiB)": 70.5, "step": 43650, "token_acc": 0.9195710455764075, "train_speed(iter/s)": 1.451821 }, { "epoch": 1.870314039672679, "grad_norm": 4.368166446685791, "learning_rate": 6.927009629027421e-05, "loss": 0.4955615997314453, "memory(GiB)": 70.5, "step": 43655, "token_acc": 0.8785714285714286, "train_speed(iter/s)": 1.451825 }, { "epoch": 1.870528255001928, "grad_norm": 2.2440783977508545, "learning_rate": 6.926388622571403e-05, "loss": 0.3652777910232544, "memory(GiB)": 70.5, "step": 43660, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.451854 }, { "epoch": 1.870742470331177, "grad_norm": 4.051190376281738, "learning_rate": 6.925767581217046e-05, "loss": 0.5334230899810791, "memory(GiB)": 70.5, "step": 43665, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.451858 }, { "epoch": 1.8709566856604258, "grad_norm": 3.8120505809783936, "learning_rate": 6.925146504975606e-05, "loss": 0.4552948474884033, "memory(GiB)": 70.5, "step": 43670, "token_acc": 0.9228395061728395, "train_speed(iter/s)": 1.451861 }, { "epoch": 1.8711709009896749, "grad_norm": 1.837588906288147, "learning_rate": 6.92452539385833e-05, "loss": 0.18216547966003419, "memory(GiB)": 70.5, "step": 43675, "token_acc": 0.9630996309963099, "train_speed(iter/s)": 1.451866 }, { "epoch": 1.871385116318924, "grad_norm": 6.9782867431640625, "learning_rate": 6.92390424787647e-05, "loss": 0.3976099967956543, "memory(GiB)": 70.5, "step": 43680, "token_acc": 0.9163346613545816, "train_speed(iter/s)": 1.45187 }, { "epoch": 1.8715993316481727, "grad_norm": 0.607567548751831, "learning_rate": 6.923283067041282e-05, "loss": 0.4096812725067139, "memory(GiB)": 70.5, "step": 43685, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.451871 }, { "epoch": 1.8718135469774217, "grad_norm": 4.661214828491211, "learning_rate": 6.922661851364016e-05, "loss": 0.571504020690918, "memory(GiB)": 70.5, "step": 43690, "token_acc": 0.8713235294117647, "train_speed(iter/s)": 1.451876 }, { "epoch": 1.8720277623066708, "grad_norm": 6.790126800537109, "learning_rate": 6.922040600855926e-05, "loss": 0.4191570281982422, "memory(GiB)": 70.5, "step": 43695, "token_acc": 0.9085545722713865, "train_speed(iter/s)": 1.451888 }, { "epoch": 1.8722419776359196, "grad_norm": 3.5171055793762207, "learning_rate": 6.921419315528268e-05, "loss": 0.6940335273742676, "memory(GiB)": 70.5, "step": 43700, "token_acc": 0.8862068965517241, "train_speed(iter/s)": 1.451908 }, { "epoch": 1.8724561929651686, "grad_norm": 3.2929253578186035, "learning_rate": 6.920797995392297e-05, "loss": 0.37142791748046877, "memory(GiB)": 70.5, "step": 43705, "token_acc": 0.907258064516129, "train_speed(iter/s)": 1.451921 }, { "epoch": 1.8726704082944177, "grad_norm": 0.7995607256889343, "learning_rate": 6.920176640459268e-05, "loss": 0.22956335544586182, "memory(GiB)": 70.5, "step": 43710, "token_acc": 0.9587301587301588, "train_speed(iter/s)": 1.451922 }, { "epoch": 1.8728846236236665, "grad_norm": 2.6886494159698486, "learning_rate": 6.919555250740439e-05, "loss": 0.25861871242523193, "memory(GiB)": 70.5, "step": 43715, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.451941 }, { "epoch": 1.8730988389529155, "grad_norm": 4.51811408996582, "learning_rate": 6.918933826247065e-05, "loss": 0.5712627410888672, "memory(GiB)": 70.5, "step": 43720, "token_acc": 0.8737201365187713, "train_speed(iter/s)": 1.451943 }, { "epoch": 1.8733130542821645, "grad_norm": 2.784900188446045, "learning_rate": 6.918312366990405e-05, "loss": 0.3275540828704834, "memory(GiB)": 70.5, "step": 43725, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.451942 }, { "epoch": 1.8735272696114134, "grad_norm": 1.0185794830322266, "learning_rate": 6.917690872981717e-05, "loss": 0.2383432626724243, "memory(GiB)": 70.5, "step": 43730, "token_acc": 0.9516129032258065, "train_speed(iter/s)": 1.451936 }, { "epoch": 1.8737414849406624, "grad_norm": 1.2368760108947754, "learning_rate": 6.917069344232258e-05, "loss": 0.5655033111572265, "memory(GiB)": 70.5, "step": 43735, "token_acc": 0.8854489164086687, "train_speed(iter/s)": 1.451941 }, { "epoch": 1.8739557002699114, "grad_norm": 2.4612839221954346, "learning_rate": 6.916447780753291e-05, "loss": 0.6245272636413575, "memory(GiB)": 70.5, "step": 43740, "token_acc": 0.8790849673202614, "train_speed(iter/s)": 1.451922 }, { "epoch": 1.8741699155991602, "grad_norm": 4.765115261077881, "learning_rate": 6.915826182556075e-05, "loss": 0.5671334266662598, "memory(GiB)": 70.5, "step": 43745, "token_acc": 0.8617021276595744, "train_speed(iter/s)": 1.451935 }, { "epoch": 1.8743841309284093, "grad_norm": 3.691317081451416, "learning_rate": 6.915204549651871e-05, "loss": 0.37847816944122314, "memory(GiB)": 70.5, "step": 43750, "token_acc": 0.9120521172638436, "train_speed(iter/s)": 1.451949 }, { "epoch": 1.8745983462576583, "grad_norm": 1.4589388370513916, "learning_rate": 6.914582882051938e-05, "loss": 0.3943014144897461, "memory(GiB)": 70.5, "step": 43755, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.451957 }, { "epoch": 1.8748125615869071, "grad_norm": 2.5885281562805176, "learning_rate": 6.913961179767543e-05, "loss": 0.3186378002166748, "memory(GiB)": 70.5, "step": 43760, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.451961 }, { "epoch": 1.8750267769161562, "grad_norm": 3.2826387882232666, "learning_rate": 6.913339442809942e-05, "loss": 0.1459587812423706, "memory(GiB)": 70.5, "step": 43765, "token_acc": 0.9715447154471545, "train_speed(iter/s)": 1.45197 }, { "epoch": 1.8752409922454052, "grad_norm": 2.1074841022491455, "learning_rate": 6.912717671190407e-05, "loss": 0.380908465385437, "memory(GiB)": 70.5, "step": 43770, "token_acc": 0.9214876033057852, "train_speed(iter/s)": 1.45198 }, { "epoch": 1.875455207574654, "grad_norm": 2.7885305881500244, "learning_rate": 6.912095864920193e-05, "loss": 0.40111751556396485, "memory(GiB)": 70.5, "step": 43775, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.451987 }, { "epoch": 1.875669422903903, "grad_norm": 2.986259698867798, "learning_rate": 6.911474024010569e-05, "loss": 0.2073798656463623, "memory(GiB)": 70.5, "step": 43780, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.451975 }, { "epoch": 1.875883638233152, "grad_norm": 4.369964599609375, "learning_rate": 6.910852148472802e-05, "loss": 0.40186347961425783, "memory(GiB)": 70.5, "step": 43785, "token_acc": 0.9478260869565217, "train_speed(iter/s)": 1.451976 }, { "epoch": 1.8760978535624009, "grad_norm": 4.502390384674072, "learning_rate": 6.910230238318157e-05, "loss": 0.688658618927002, "memory(GiB)": 70.5, "step": 43790, "token_acc": 0.860248447204969, "train_speed(iter/s)": 1.45198 }, { "epoch": 1.87631206889165, "grad_norm": 3.4084348678588867, "learning_rate": 6.909608293557895e-05, "loss": 0.3445389747619629, "memory(GiB)": 70.5, "step": 43795, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.451975 }, { "epoch": 1.876526284220899, "grad_norm": 3.804459571838379, "learning_rate": 6.908986314203289e-05, "loss": 0.41765737533569336, "memory(GiB)": 70.5, "step": 43800, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.452016 }, { "epoch": 1.8767404995501478, "grad_norm": 4.789471626281738, "learning_rate": 6.908364300265607e-05, "loss": 0.31494252681732177, "memory(GiB)": 70.5, "step": 43805, "token_acc": 0.9476923076923077, "train_speed(iter/s)": 1.452013 }, { "epoch": 1.8769547148793968, "grad_norm": 3.699329137802124, "learning_rate": 6.907742251756114e-05, "loss": 0.47412691116333006, "memory(GiB)": 70.5, "step": 43810, "token_acc": 0.889967637540453, "train_speed(iter/s)": 1.452009 }, { "epoch": 1.8771689302086458, "grad_norm": 5.099518299102783, "learning_rate": 6.907120168686079e-05, "loss": 0.34996833801269533, "memory(GiB)": 70.5, "step": 43815, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.452014 }, { "epoch": 1.8773831455378946, "grad_norm": 3.135625123977661, "learning_rate": 6.906498051066776e-05, "loss": 0.5338631629943847, "memory(GiB)": 70.5, "step": 43820, "token_acc": 0.8604651162790697, "train_speed(iter/s)": 1.452029 }, { "epoch": 1.8775973608671437, "grad_norm": 4.437517166137695, "learning_rate": 6.90587589890947e-05, "loss": 0.38397789001464844, "memory(GiB)": 70.5, "step": 43825, "token_acc": 0.9201520912547528, "train_speed(iter/s)": 1.452025 }, { "epoch": 1.8778115761963927, "grad_norm": 6.888981342315674, "learning_rate": 6.905253712225436e-05, "loss": 0.6115746021270752, "memory(GiB)": 70.5, "step": 43830, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.452009 }, { "epoch": 1.8780257915256415, "grad_norm": 5.435399055480957, "learning_rate": 6.904631491025945e-05, "loss": 0.7232629299163819, "memory(GiB)": 70.5, "step": 43835, "token_acc": 0.8770226537216829, "train_speed(iter/s)": 1.45201 }, { "epoch": 1.8782400068548906, "grad_norm": 3.8509840965270996, "learning_rate": 6.904009235322265e-05, "loss": 0.36627674102783203, "memory(GiB)": 70.5, "step": 43840, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.452014 }, { "epoch": 1.8784542221841396, "grad_norm": 4.315464973449707, "learning_rate": 6.903386945125673e-05, "loss": 0.2674837112426758, "memory(GiB)": 70.5, "step": 43845, "token_acc": 0.9272030651340997, "train_speed(iter/s)": 1.452018 }, { "epoch": 1.8786684375133884, "grad_norm": 4.2721967697143555, "learning_rate": 6.90276462044744e-05, "loss": 0.5246981143951416, "memory(GiB)": 70.5, "step": 43850, "token_acc": 0.8918032786885246, "train_speed(iter/s)": 1.452021 }, { "epoch": 1.8788826528426374, "grad_norm": 2.5621654987335205, "learning_rate": 6.90214226129884e-05, "loss": 0.26271440982818606, "memory(GiB)": 70.5, "step": 43855, "token_acc": 0.9375, "train_speed(iter/s)": 1.452043 }, { "epoch": 1.8790968681718865, "grad_norm": 4.514749526977539, "learning_rate": 6.901519867691151e-05, "loss": 0.34380981922149656, "memory(GiB)": 70.5, "step": 43860, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.452045 }, { "epoch": 1.8793110835011353, "grad_norm": 5.509635925292969, "learning_rate": 6.900897439635646e-05, "loss": 0.49297657012939455, "memory(GiB)": 70.5, "step": 43865, "token_acc": 0.9026217228464419, "train_speed(iter/s)": 1.452039 }, { "epoch": 1.8795252988303843, "grad_norm": 2.41359806060791, "learning_rate": 6.900274977143599e-05, "loss": 0.2993456840515137, "memory(GiB)": 70.5, "step": 43870, "token_acc": 0.951310861423221, "train_speed(iter/s)": 1.452038 }, { "epoch": 1.8797395141596334, "grad_norm": 4.263243198394775, "learning_rate": 6.89965248022629e-05, "loss": 0.41717844009399413, "memory(GiB)": 70.5, "step": 43875, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.452035 }, { "epoch": 1.8799537294888822, "grad_norm": 6.43966007232666, "learning_rate": 6.899029948894993e-05, "loss": 0.41710672378540037, "memory(GiB)": 70.5, "step": 43880, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.452022 }, { "epoch": 1.8801679448181312, "grad_norm": 1.7233517169952393, "learning_rate": 6.898407383160985e-05, "loss": 0.13784841299057007, "memory(GiB)": 70.5, "step": 43885, "token_acc": 0.9688715953307393, "train_speed(iter/s)": 1.452034 }, { "epoch": 1.8803821601473802, "grad_norm": 0.6021106243133545, "learning_rate": 6.89778478303555e-05, "loss": 0.34660112857818604, "memory(GiB)": 70.5, "step": 43890, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.452044 }, { "epoch": 1.880596375476629, "grad_norm": 5.795792579650879, "learning_rate": 6.897162148529963e-05, "loss": 0.1967138409614563, "memory(GiB)": 70.5, "step": 43895, "token_acc": 0.9525316455696202, "train_speed(iter/s)": 1.452049 }, { "epoch": 1.880810590805878, "grad_norm": 3.6946377754211426, "learning_rate": 6.896539479655501e-05, "loss": 0.31217260360717775, "memory(GiB)": 70.5, "step": 43900, "token_acc": 0.9312714776632303, "train_speed(iter/s)": 1.452053 }, { "epoch": 1.8810248061351271, "grad_norm": 8.589975357055664, "learning_rate": 6.89591677642345e-05, "loss": 0.13215548992156984, "memory(GiB)": 70.5, "step": 43905, "token_acc": 0.956081081081081, "train_speed(iter/s)": 1.45205 }, { "epoch": 1.881239021464376, "grad_norm": 3.442708969116211, "learning_rate": 6.895294038845087e-05, "loss": 0.3213359355926514, "memory(GiB)": 70.5, "step": 43910, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.452046 }, { "epoch": 1.881453236793625, "grad_norm": 4.345897674560547, "learning_rate": 6.894671266931697e-05, "loss": 0.5924683570861816, "memory(GiB)": 70.5, "step": 43915, "token_acc": 0.8795986622073578, "train_speed(iter/s)": 1.452069 }, { "epoch": 1.881667452122874, "grad_norm": 3.764136552810669, "learning_rate": 6.894048460694557e-05, "loss": 0.37220778465271, "memory(GiB)": 70.5, "step": 43920, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.452083 }, { "epoch": 1.8818816674521228, "grad_norm": 1.2671687602996826, "learning_rate": 6.893425620144952e-05, "loss": 0.2511159658432007, "memory(GiB)": 70.5, "step": 43925, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.452092 }, { "epoch": 1.8820958827813719, "grad_norm": 5.484409332275391, "learning_rate": 6.892802745294165e-05, "loss": 0.49232850074768064, "memory(GiB)": 70.5, "step": 43930, "token_acc": 0.8931034482758621, "train_speed(iter/s)": 1.452087 }, { "epoch": 1.8823100981106209, "grad_norm": 4.624021053314209, "learning_rate": 6.892179836153483e-05, "loss": 0.32375755310058596, "memory(GiB)": 70.5, "step": 43935, "token_acc": 0.9398148148148148, "train_speed(iter/s)": 1.452097 }, { "epoch": 1.8825243134398697, "grad_norm": 3.2565009593963623, "learning_rate": 6.891556892734188e-05, "loss": 0.4908305168151855, "memory(GiB)": 70.5, "step": 43940, "token_acc": 0.8853503184713376, "train_speed(iter/s)": 1.452108 }, { "epoch": 1.8827385287691187, "grad_norm": 2.2670021057128906, "learning_rate": 6.890933915047565e-05, "loss": 0.3772355318069458, "memory(GiB)": 70.5, "step": 43945, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.452127 }, { "epoch": 1.8829527440983678, "grad_norm": 3.0539817810058594, "learning_rate": 6.8903109031049e-05, "loss": 0.28450164794921873, "memory(GiB)": 70.5, "step": 43950, "token_acc": 0.9580152671755725, "train_speed(iter/s)": 1.452149 }, { "epoch": 1.8831669594276166, "grad_norm": 0.46458086371421814, "learning_rate": 6.88968785691748e-05, "loss": 0.4856846332550049, "memory(GiB)": 70.5, "step": 43955, "token_acc": 0.9125, "train_speed(iter/s)": 1.452167 }, { "epoch": 1.8833811747568656, "grad_norm": 4.718154430389404, "learning_rate": 6.88906477649659e-05, "loss": 0.18923487663269042, "memory(GiB)": 70.5, "step": 43960, "token_acc": 0.9578059071729957, "train_speed(iter/s)": 1.452185 }, { "epoch": 1.8835953900861147, "grad_norm": 3.542008399963379, "learning_rate": 6.888441661853523e-05, "loss": 0.29094505310058594, "memory(GiB)": 70.5, "step": 43965, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.452186 }, { "epoch": 1.8838096054153635, "grad_norm": 4.561777114868164, "learning_rate": 6.887818512999562e-05, "loss": 0.3439945220947266, "memory(GiB)": 70.5, "step": 43970, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.452204 }, { "epoch": 1.8840238207446125, "grad_norm": 3.7214713096618652, "learning_rate": 6.887195329945997e-05, "loss": 0.5882221698760987, "memory(GiB)": 70.5, "step": 43975, "token_acc": 0.875, "train_speed(iter/s)": 1.45219 }, { "epoch": 1.8842380360738615, "grad_norm": 3.5611579418182373, "learning_rate": 6.88657211270412e-05, "loss": 0.2988587379455566, "memory(GiB)": 70.5, "step": 43980, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.452194 }, { "epoch": 1.8844522514031103, "grad_norm": 1.3918755054473877, "learning_rate": 6.885948861285219e-05, "loss": 0.546180009841919, "memory(GiB)": 70.5, "step": 43985, "token_acc": 0.8841059602649006, "train_speed(iter/s)": 1.452193 }, { "epoch": 1.8846664667323594, "grad_norm": 2.1768805980682373, "learning_rate": 6.885325575700584e-05, "loss": 0.45027918815612794, "memory(GiB)": 70.5, "step": 43990, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.45219 }, { "epoch": 1.8848806820616084, "grad_norm": 2.882347583770752, "learning_rate": 6.884702255961508e-05, "loss": 0.2484117031097412, "memory(GiB)": 70.5, "step": 43995, "token_acc": 0.9490445859872612, "train_speed(iter/s)": 1.452186 }, { "epoch": 1.8850948973908572, "grad_norm": 3.006289482116699, "learning_rate": 6.884078902079283e-05, "loss": 0.4511958122253418, "memory(GiB)": 70.5, "step": 44000, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.452198 }, { "epoch": 1.8850948973908572, "eval_loss": 2.8615024089813232, "eval_runtime": 13.1109, "eval_samples_per_second": 7.627, "eval_steps_per_second": 7.627, "eval_token_acc": 0.3867334167709637, "step": 44000 }, { "epoch": 1.8853091127201063, "grad_norm": 1.6032472848892212, "learning_rate": 6.8834555140652e-05, "loss": 0.497043514251709, "memory(GiB)": 70.5, "step": 44005, "token_acc": 0.5427312775330396, "train_speed(iter/s)": 1.451544 }, { "epoch": 1.8855233280493553, "grad_norm": 1.82854425907135, "learning_rate": 6.882832091930556e-05, "loss": 0.27115740776062014, "memory(GiB)": 70.5, "step": 44010, "token_acc": 0.9298245614035088, "train_speed(iter/s)": 1.451537 }, { "epoch": 1.885737543378604, "grad_norm": 4.572357654571533, "learning_rate": 6.882208635686641e-05, "loss": 0.34126715660095214, "memory(GiB)": 70.5, "step": 44015, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.451532 }, { "epoch": 1.8859517587078531, "grad_norm": 4.236844539642334, "learning_rate": 6.881585145344751e-05, "loss": 0.3459657669067383, "memory(GiB)": 70.5, "step": 44020, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.451535 }, { "epoch": 1.8861659740371022, "grad_norm": 5.416170120239258, "learning_rate": 6.88096162091618e-05, "loss": 0.6968661308288574, "memory(GiB)": 70.5, "step": 44025, "token_acc": 0.8483606557377049, "train_speed(iter/s)": 1.451546 }, { "epoch": 1.886380189366351, "grad_norm": 2.595121383666992, "learning_rate": 6.880338062412228e-05, "loss": 0.24949798583984376, "memory(GiB)": 70.5, "step": 44030, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.451569 }, { "epoch": 1.8865944046956, "grad_norm": 4.1152238845825195, "learning_rate": 6.879714469844185e-05, "loss": 0.3568146228790283, "memory(GiB)": 70.5, "step": 44035, "token_acc": 0.911504424778761, "train_speed(iter/s)": 1.45158 }, { "epoch": 1.886808620024849, "grad_norm": 2.703397274017334, "learning_rate": 6.879090843223353e-05, "loss": 0.15218840837478637, "memory(GiB)": 70.5, "step": 44040, "token_acc": 0.9726027397260274, "train_speed(iter/s)": 1.451581 }, { "epoch": 1.8870228353540979, "grad_norm": 4.204517841339111, "learning_rate": 6.878467182561026e-05, "loss": 0.3085540771484375, "memory(GiB)": 70.5, "step": 44045, "token_acc": 0.9214501510574018, "train_speed(iter/s)": 1.451601 }, { "epoch": 1.887237050683347, "grad_norm": 1.4741573333740234, "learning_rate": 6.877843487868505e-05, "loss": 0.44513988494873047, "memory(GiB)": 70.5, "step": 44050, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.451626 }, { "epoch": 1.887451266012596, "grad_norm": 6.0394978523254395, "learning_rate": 6.877219759157087e-05, "loss": 0.5401659965515136, "memory(GiB)": 70.5, "step": 44055, "token_acc": 0.8781362007168458, "train_speed(iter/s)": 1.451654 }, { "epoch": 1.8876654813418448, "grad_norm": 2.39343523979187, "learning_rate": 6.876595996438072e-05, "loss": 0.4530781269073486, "memory(GiB)": 70.5, "step": 44060, "token_acc": 0.9213483146067416, "train_speed(iter/s)": 1.451661 }, { "epoch": 1.8878796966710938, "grad_norm": 3.527745246887207, "learning_rate": 6.875972199722761e-05, "loss": 0.21060025691986084, "memory(GiB)": 70.5, "step": 44065, "token_acc": 0.9536423841059603, "train_speed(iter/s)": 1.451671 }, { "epoch": 1.8880939120003428, "grad_norm": 1.5553686618804932, "learning_rate": 6.875348369022452e-05, "loss": 0.2701220750808716, "memory(GiB)": 70.5, "step": 44070, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.451667 }, { "epoch": 1.8883081273295916, "grad_norm": 3.1740105152130127, "learning_rate": 6.87472450434845e-05, "loss": 0.22222728729248048, "memory(GiB)": 70.5, "step": 44075, "token_acc": 0.9471544715447154, "train_speed(iter/s)": 1.451686 }, { "epoch": 1.8885223426588407, "grad_norm": 0.5832892656326294, "learning_rate": 6.874100605712054e-05, "loss": 0.4832926273345947, "memory(GiB)": 70.5, "step": 44080, "token_acc": 0.8982456140350877, "train_speed(iter/s)": 1.451691 }, { "epoch": 1.8887365579880897, "grad_norm": 4.483786582946777, "learning_rate": 6.873476673124567e-05, "loss": 0.5378880977630616, "memory(GiB)": 70.5, "step": 44085, "token_acc": 0.9052287581699346, "train_speed(iter/s)": 1.451679 }, { "epoch": 1.8889507733173385, "grad_norm": 3.7394790649414062, "learning_rate": 6.872852706597295e-05, "loss": 0.6499654293060303, "memory(GiB)": 70.5, "step": 44090, "token_acc": 0.8652694610778443, "train_speed(iter/s)": 1.451695 }, { "epoch": 1.8891649886465876, "grad_norm": 3.664334774017334, "learning_rate": 6.872228706141537e-05, "loss": 0.5232954025268555, "memory(GiB)": 70.5, "step": 44095, "token_acc": 0.8935483870967742, "train_speed(iter/s)": 1.451692 }, { "epoch": 1.8893792039758366, "grad_norm": 0.6878902316093445, "learning_rate": 6.8716046717686e-05, "loss": 0.24449706077575684, "memory(GiB)": 70.5, "step": 44100, "token_acc": 0.9508670520231214, "train_speed(iter/s)": 1.451696 }, { "epoch": 1.8895934193050854, "grad_norm": 3.6471340656280518, "learning_rate": 6.87098060348979e-05, "loss": 0.48811817169189453, "memory(GiB)": 70.5, "step": 44105, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.451701 }, { "epoch": 1.8898076346343344, "grad_norm": 3.629727840423584, "learning_rate": 6.87035650131641e-05, "loss": 0.5044245719909668, "memory(GiB)": 70.5, "step": 44110, "token_acc": 0.8952702702702703, "train_speed(iter/s)": 1.451724 }, { "epoch": 1.8900218499635835, "grad_norm": 4.368208408355713, "learning_rate": 6.869732365259767e-05, "loss": 0.5059621810913086, "memory(GiB)": 70.5, "step": 44115, "token_acc": 0.8962264150943396, "train_speed(iter/s)": 1.451735 }, { "epoch": 1.8902360652928323, "grad_norm": 1.9658479690551758, "learning_rate": 6.869108195331169e-05, "loss": 0.15659680366516113, "memory(GiB)": 70.5, "step": 44120, "token_acc": 0.9691119691119691, "train_speed(iter/s)": 1.451747 }, { "epoch": 1.8904502806220813, "grad_norm": 2.3284716606140137, "learning_rate": 6.868483991541923e-05, "loss": 0.3405050277709961, "memory(GiB)": 70.5, "step": 44125, "token_acc": 0.9328358208955224, "train_speed(iter/s)": 1.451769 }, { "epoch": 1.8906644959513303, "grad_norm": 3.8079886436462402, "learning_rate": 6.867859753903336e-05, "loss": 0.3218986034393311, "memory(GiB)": 70.5, "step": 44130, "token_acc": 0.9295774647887324, "train_speed(iter/s)": 1.451764 }, { "epoch": 1.8908787112805792, "grad_norm": 5.348166465759277, "learning_rate": 6.867235482426719e-05, "loss": 0.29342260360717776, "memory(GiB)": 70.5, "step": 44135, "token_acc": 0.9312714776632303, "train_speed(iter/s)": 1.451778 }, { "epoch": 1.8910929266098282, "grad_norm": 2.353856086730957, "learning_rate": 6.866611177123377e-05, "loss": 0.390685510635376, "memory(GiB)": 70.5, "step": 44140, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.45177 }, { "epoch": 1.8913071419390772, "grad_norm": 5.419622421264648, "learning_rate": 6.865986838004624e-05, "loss": 0.5060473442077636, "memory(GiB)": 70.5, "step": 44145, "token_acc": 0.9006849315068494, "train_speed(iter/s)": 1.451764 }, { "epoch": 1.891521357268326, "grad_norm": 3.6098506450653076, "learning_rate": 6.865362465081768e-05, "loss": 0.3900465965270996, "memory(GiB)": 70.5, "step": 44150, "token_acc": 0.933993399339934, "train_speed(iter/s)": 1.451772 }, { "epoch": 1.891735572597575, "grad_norm": 4.271976470947266, "learning_rate": 6.864738058366122e-05, "loss": 0.45163421630859374, "memory(GiB)": 70.5, "step": 44155, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.451788 }, { "epoch": 1.8919497879268241, "grad_norm": 4.294041633605957, "learning_rate": 6.864113617868997e-05, "loss": 0.5374879837036133, "memory(GiB)": 70.5, "step": 44160, "token_acc": 0.8798701298701299, "train_speed(iter/s)": 1.451796 }, { "epoch": 1.892164003256073, "grad_norm": 4.404813766479492, "learning_rate": 6.863489143601705e-05, "loss": 0.7512392044067383, "memory(GiB)": 70.5, "step": 44165, "token_acc": 0.8333333333333334, "train_speed(iter/s)": 1.451796 }, { "epoch": 1.892378218585322, "grad_norm": 3.3711838722229004, "learning_rate": 6.862864635575559e-05, "loss": 0.4308065414428711, "memory(GiB)": 70.5, "step": 44170, "token_acc": 0.9030837004405287, "train_speed(iter/s)": 1.451804 }, { "epoch": 1.892592433914571, "grad_norm": 1.732492208480835, "learning_rate": 6.862240093801873e-05, "loss": 0.3121363639831543, "memory(GiB)": 70.5, "step": 44175, "token_acc": 0.9205020920502092, "train_speed(iter/s)": 1.451815 }, { "epoch": 1.8928066492438198, "grad_norm": 1.508434772491455, "learning_rate": 6.861615518291962e-05, "loss": 0.576385498046875, "memory(GiB)": 70.5, "step": 44180, "token_acc": 0.8633333333333333, "train_speed(iter/s)": 1.451819 }, { "epoch": 1.8930208645730688, "grad_norm": 3.0405635833740234, "learning_rate": 6.860990909057137e-05, "loss": 0.34144761562347414, "memory(GiB)": 70.5, "step": 44185, "token_acc": 0.9383886255924171, "train_speed(iter/s)": 1.451831 }, { "epoch": 1.8932350799023179, "grad_norm": 1.72967529296875, "learning_rate": 6.860366266108716e-05, "loss": 0.43835062980651857, "memory(GiB)": 70.5, "step": 44190, "token_acc": 0.8880866425992779, "train_speed(iter/s)": 1.451827 }, { "epoch": 1.8934492952315667, "grad_norm": 1.9980590343475342, "learning_rate": 6.859741589458015e-05, "loss": 0.23943667411804198, "memory(GiB)": 70.5, "step": 44195, "token_acc": 0.964516129032258, "train_speed(iter/s)": 1.451825 }, { "epoch": 1.8936635105608157, "grad_norm": 3.110569477081299, "learning_rate": 6.859116879116352e-05, "loss": 0.5568036079406739, "memory(GiB)": 70.5, "step": 44200, "token_acc": 0.8881118881118881, "train_speed(iter/s)": 1.45183 }, { "epoch": 1.8938777258900648, "grad_norm": 3.419764518737793, "learning_rate": 6.858492135095043e-05, "loss": 0.5683652400970459, "memory(GiB)": 70.5, "step": 44205, "token_acc": 0.8876811594202898, "train_speed(iter/s)": 1.451818 }, { "epoch": 1.8940919412193136, "grad_norm": 1.2126426696777344, "learning_rate": 6.857867357405404e-05, "loss": 0.32800140380859377, "memory(GiB)": 70.5, "step": 44210, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.451827 }, { "epoch": 1.8943061565485626, "grad_norm": 4.289023399353027, "learning_rate": 6.857242546058756e-05, "loss": 0.31035528182983396, "memory(GiB)": 70.5, "step": 44215, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.451831 }, { "epoch": 1.8945203718778116, "grad_norm": 3.5484402179718018, "learning_rate": 6.856617701066416e-05, "loss": 0.604001808166504, "memory(GiB)": 70.5, "step": 44220, "token_acc": 0.8706624605678234, "train_speed(iter/s)": 1.451832 }, { "epoch": 1.8947345872070604, "grad_norm": 2.9973666667938232, "learning_rate": 6.855992822439705e-05, "loss": 0.44740004539489747, "memory(GiB)": 70.5, "step": 44225, "token_acc": 0.9258160237388724, "train_speed(iter/s)": 1.451828 }, { "epoch": 1.8949488025363095, "grad_norm": 2.6471071243286133, "learning_rate": 6.855367910189942e-05, "loss": 0.3041806221008301, "memory(GiB)": 70.5, "step": 44230, "token_acc": 0.946031746031746, "train_speed(iter/s)": 1.451845 }, { "epoch": 1.8951630178655585, "grad_norm": 2.420254707336426, "learning_rate": 6.854742964328451e-05, "loss": 0.4937274932861328, "memory(GiB)": 70.5, "step": 44235, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.451856 }, { "epoch": 1.8953772331948073, "grad_norm": 2.4420359134674072, "learning_rate": 6.854117984866552e-05, "loss": 0.33153090476989744, "memory(GiB)": 70.5, "step": 44240, "token_acc": 0.924924924924925, "train_speed(iter/s)": 1.451855 }, { "epoch": 1.8955914485240564, "grad_norm": 1.836103081703186, "learning_rate": 6.853492971815564e-05, "loss": 0.5368210792541503, "memory(GiB)": 70.5, "step": 44245, "token_acc": 0.8867924528301887, "train_speed(iter/s)": 1.451853 }, { "epoch": 1.8958056638533054, "grad_norm": 2.899444818496704, "learning_rate": 6.852867925186814e-05, "loss": 0.31542172431945803, "memory(GiB)": 70.5, "step": 44250, "token_acc": 0.9313725490196079, "train_speed(iter/s)": 1.451859 }, { "epoch": 1.8960198791825542, "grad_norm": 2.4813947677612305, "learning_rate": 6.85224284499162e-05, "loss": 0.43293299674987795, "memory(GiB)": 70.5, "step": 44255, "token_acc": 0.9293286219081273, "train_speed(iter/s)": 1.451874 }, { "epoch": 1.8962340945118032, "grad_norm": 2.863929271697998, "learning_rate": 6.851617731241312e-05, "loss": 0.2400416135787964, "memory(GiB)": 70.5, "step": 44260, "token_acc": 0.9274924471299094, "train_speed(iter/s)": 1.45187 }, { "epoch": 1.8964483098410523, "grad_norm": 2.3226959705352783, "learning_rate": 6.850992583947212e-05, "loss": 0.5524667739868164, "memory(GiB)": 70.5, "step": 44265, "token_acc": 0.8801261829652997, "train_speed(iter/s)": 1.451887 }, { "epoch": 1.896662525170301, "grad_norm": 3.680569887161255, "learning_rate": 6.850367403120645e-05, "loss": 0.6719714641571045, "memory(GiB)": 70.5, "step": 44270, "token_acc": 0.827485380116959, "train_speed(iter/s)": 1.451903 }, { "epoch": 1.8968767404995501, "grad_norm": 1.5980206727981567, "learning_rate": 6.849742188772936e-05, "loss": 0.3482077598571777, "memory(GiB)": 70.5, "step": 44275, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.451907 }, { "epoch": 1.8970909558287992, "grad_norm": 13.683908462524414, "learning_rate": 6.849116940915412e-05, "loss": 0.6192901134490967, "memory(GiB)": 70.5, "step": 44280, "token_acc": 0.8621908127208481, "train_speed(iter/s)": 1.451926 }, { "epoch": 1.897305171158048, "grad_norm": 0.2640795111656189, "learning_rate": 6.8484916595594e-05, "loss": 0.22737329006195067, "memory(GiB)": 70.5, "step": 44285, "token_acc": 0.9589905362776026, "train_speed(iter/s)": 1.451928 }, { "epoch": 1.897519386487297, "grad_norm": 3.439664125442505, "learning_rate": 6.847866344716228e-05, "loss": 0.32313036918640137, "memory(GiB)": 70.5, "step": 44290, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.451937 }, { "epoch": 1.897733601816546, "grad_norm": 2.837158203125, "learning_rate": 6.847240996397224e-05, "loss": 0.5325938701629639, "memory(GiB)": 70.5, "step": 44295, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.451943 }, { "epoch": 1.8979478171457949, "grad_norm": 2.894155740737915, "learning_rate": 6.846615614613716e-05, "loss": 0.28525938987731936, "memory(GiB)": 70.5, "step": 44300, "token_acc": 0.9448818897637795, "train_speed(iter/s)": 1.451949 }, { "epoch": 1.898162032475044, "grad_norm": 4.955081462860107, "learning_rate": 6.845990199377033e-05, "loss": 0.43184943199157716, "memory(GiB)": 70.5, "step": 44305, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.451967 }, { "epoch": 1.898376247804293, "grad_norm": 2.1190285682678223, "learning_rate": 6.845364750698507e-05, "loss": 0.2798203468322754, "memory(GiB)": 70.5, "step": 44310, "token_acc": 0.9315589353612167, "train_speed(iter/s)": 1.451969 }, { "epoch": 1.8985904631335417, "grad_norm": 2.1753287315368652, "learning_rate": 6.844739268589467e-05, "loss": 0.24802632331848146, "memory(GiB)": 70.5, "step": 44315, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.451983 }, { "epoch": 1.8988046784627908, "grad_norm": 4.97253942489624, "learning_rate": 6.844113753061246e-05, "loss": 0.4750679492950439, "memory(GiB)": 70.5, "step": 44320, "token_acc": 0.8973384030418251, "train_speed(iter/s)": 1.452 }, { "epoch": 1.8990188937920398, "grad_norm": 3.3163816928863525, "learning_rate": 6.843488204125174e-05, "loss": 0.3261264801025391, "memory(GiB)": 70.5, "step": 44325, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.45201 }, { "epoch": 1.8992331091212886, "grad_norm": 3.158409595489502, "learning_rate": 6.842862621792581e-05, "loss": 0.27178144454956055, "memory(GiB)": 70.5, "step": 44330, "token_acc": 0.9446640316205533, "train_speed(iter/s)": 1.452006 }, { "epoch": 1.8994473244505377, "grad_norm": 4.138465404510498, "learning_rate": 6.842237006074805e-05, "loss": 0.2139575242996216, "memory(GiB)": 70.5, "step": 44335, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.45202 }, { "epoch": 1.8996615397797867, "grad_norm": 5.348426818847656, "learning_rate": 6.841611356983179e-05, "loss": 0.4781841278076172, "memory(GiB)": 70.5, "step": 44340, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.452017 }, { "epoch": 1.8998757551090355, "grad_norm": 1.944390892982483, "learning_rate": 6.840985674529033e-05, "loss": 0.3761838436126709, "memory(GiB)": 70.5, "step": 44345, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.45203 }, { "epoch": 1.9000899704382845, "grad_norm": 3.3927595615386963, "learning_rate": 6.840359958723705e-05, "loss": 0.4540530204772949, "memory(GiB)": 70.5, "step": 44350, "token_acc": 0.897887323943662, "train_speed(iter/s)": 1.452029 }, { "epoch": 1.9003041857675336, "grad_norm": 2.361480474472046, "learning_rate": 6.839734209578532e-05, "loss": 0.4030906677246094, "memory(GiB)": 70.5, "step": 44355, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.452046 }, { "epoch": 1.9005184010967824, "grad_norm": 1.2843596935272217, "learning_rate": 6.839108427104846e-05, "loss": 0.5795387268066406, "memory(GiB)": 70.5, "step": 44360, "token_acc": 0.8844984802431611, "train_speed(iter/s)": 1.452055 }, { "epoch": 1.9007326164260314, "grad_norm": 2.719017744064331, "learning_rate": 6.838482611313985e-05, "loss": 0.3662297487258911, "memory(GiB)": 70.5, "step": 44365, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.452092 }, { "epoch": 1.9009468317552805, "grad_norm": 5.338452339172363, "learning_rate": 6.83785676221729e-05, "loss": 0.46343555450439455, "memory(GiB)": 70.5, "step": 44370, "token_acc": 0.8835341365461847, "train_speed(iter/s)": 1.452096 }, { "epoch": 1.9011610470845293, "grad_norm": 1.7520983219146729, "learning_rate": 6.837230879826093e-05, "loss": 0.34244256019592284, "memory(GiB)": 70.5, "step": 44375, "token_acc": 0.9131944444444444, "train_speed(iter/s)": 1.452091 }, { "epoch": 1.9013752624137783, "grad_norm": 5.071771144866943, "learning_rate": 6.836604964151737e-05, "loss": 0.14941849708557128, "memory(GiB)": 70.5, "step": 44380, "token_acc": 0.9628252788104089, "train_speed(iter/s)": 1.452095 }, { "epoch": 1.9015894777430273, "grad_norm": 3.1570558547973633, "learning_rate": 6.835979015205558e-05, "loss": 0.5399243354797363, "memory(GiB)": 70.5, "step": 44385, "token_acc": 0.889967637540453, "train_speed(iter/s)": 1.452109 }, { "epoch": 1.9018036930722761, "grad_norm": 3.7467923164367676, "learning_rate": 6.835353032998896e-05, "loss": 0.4188541412353516, "memory(GiB)": 70.5, "step": 44390, "token_acc": 0.9094827586206896, "train_speed(iter/s)": 1.45211 }, { "epoch": 1.9020179084015252, "grad_norm": 2.56479549407959, "learning_rate": 6.834727017543094e-05, "loss": 0.20518174171447753, "memory(GiB)": 70.5, "step": 44395, "token_acc": 0.9451612903225807, "train_speed(iter/s)": 1.452117 }, { "epoch": 1.9022321237307742, "grad_norm": 2.44376277923584, "learning_rate": 6.83410096884949e-05, "loss": 0.5725867748260498, "memory(GiB)": 70.5, "step": 44400, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.452116 }, { "epoch": 1.902446339060023, "grad_norm": 3.202169418334961, "learning_rate": 6.833474886929425e-05, "loss": 0.41407151222229005, "memory(GiB)": 70.5, "step": 44405, "token_acc": 0.890625, "train_speed(iter/s)": 1.452113 }, { "epoch": 1.902660554389272, "grad_norm": 4.853146076202393, "learning_rate": 6.832848771794245e-05, "loss": 0.5559415817260742, "memory(GiB)": 70.5, "step": 44410, "token_acc": 0.875, "train_speed(iter/s)": 1.452127 }, { "epoch": 1.902874769718521, "grad_norm": 2.6637561321258545, "learning_rate": 6.83222262345529e-05, "loss": 0.23867831230163575, "memory(GiB)": 70.5, "step": 44415, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.452127 }, { "epoch": 1.90308898504777, "grad_norm": 3.9379231929779053, "learning_rate": 6.831596441923902e-05, "loss": 0.4007572174072266, "memory(GiB)": 70.5, "step": 44420, "token_acc": 0.9176029962546817, "train_speed(iter/s)": 1.452126 }, { "epoch": 1.903303200377019, "grad_norm": 0.8938109278678894, "learning_rate": 6.830970227211427e-05, "loss": 0.25772571563720703, "memory(GiB)": 70.5, "step": 44425, "token_acc": 0.9518072289156626, "train_speed(iter/s)": 1.452138 }, { "epoch": 1.903517415706268, "grad_norm": 0.9110605120658875, "learning_rate": 6.830343979329208e-05, "loss": 0.5181807994842529, "memory(GiB)": 70.5, "step": 44430, "token_acc": 0.9011627906976745, "train_speed(iter/s)": 1.452137 }, { "epoch": 1.9037316310355168, "grad_norm": 2.8570735454559326, "learning_rate": 6.829717698288591e-05, "loss": 0.4942580223083496, "memory(GiB)": 70.5, "step": 44435, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.452134 }, { "epoch": 1.9039458463647658, "grad_norm": 6.720126628875732, "learning_rate": 6.829091384100922e-05, "loss": 0.6942769527435303, "memory(GiB)": 70.5, "step": 44440, "token_acc": 0.8637992831541219, "train_speed(iter/s)": 1.452115 }, { "epoch": 1.9041600616940149, "grad_norm": 1.6890099048614502, "learning_rate": 6.828465036777548e-05, "loss": 0.268379807472229, "memory(GiB)": 70.5, "step": 44445, "token_acc": 0.925, "train_speed(iter/s)": 1.452114 }, { "epoch": 1.9043742770232637, "grad_norm": 3.550039768218994, "learning_rate": 6.827838656329812e-05, "loss": 0.616408109664917, "memory(GiB)": 70.5, "step": 44450, "token_acc": 0.8981132075471698, "train_speed(iter/s)": 1.452125 }, { "epoch": 1.9045884923525127, "grad_norm": 4.37041711807251, "learning_rate": 6.827212242769065e-05, "loss": 0.5975085735321045, "memory(GiB)": 70.5, "step": 44455, "token_acc": 0.8941605839416058, "train_speed(iter/s)": 1.452132 }, { "epoch": 1.9048027076817617, "grad_norm": 1.4349822998046875, "learning_rate": 6.826585796106654e-05, "loss": 0.10295835733413697, "memory(GiB)": 70.5, "step": 44460, "token_acc": 0.9810606060606061, "train_speed(iter/s)": 1.452142 }, { "epoch": 1.9050169230110106, "grad_norm": 0.2485978603363037, "learning_rate": 6.82595931635393e-05, "loss": 0.29340457916259766, "memory(GiB)": 70.5, "step": 44465, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.452148 }, { "epoch": 1.9052311383402596, "grad_norm": 4.6900506019592285, "learning_rate": 6.825332803522238e-05, "loss": 0.34782469272613525, "memory(GiB)": 70.5, "step": 44470, "token_acc": 0.9064516129032258, "train_speed(iter/s)": 1.45214 }, { "epoch": 1.9054453536695086, "grad_norm": 5.069394111633301, "learning_rate": 6.82470625762293e-05, "loss": 0.4193582057952881, "memory(GiB)": 70.5, "step": 44475, "token_acc": 0.908256880733945, "train_speed(iter/s)": 1.452134 }, { "epoch": 1.9056595689987574, "grad_norm": 1.9076106548309326, "learning_rate": 6.824079678667357e-05, "loss": 0.4275655269622803, "memory(GiB)": 70.5, "step": 44480, "token_acc": 0.9037656903765691, "train_speed(iter/s)": 1.452135 }, { "epoch": 1.9058737843280065, "grad_norm": 8.589553833007812, "learning_rate": 6.82345306666687e-05, "loss": 0.2677839517593384, "memory(GiB)": 70.5, "step": 44485, "token_acc": 0.9462025316455697, "train_speed(iter/s)": 1.45215 }, { "epoch": 1.9060879996572555, "grad_norm": 4.134900093078613, "learning_rate": 6.822826421632819e-05, "loss": 0.3383822441101074, "memory(GiB)": 70.5, "step": 44490, "token_acc": 0.9209486166007905, "train_speed(iter/s)": 1.452175 }, { "epoch": 1.9063022149865043, "grad_norm": 0.4846310615539551, "learning_rate": 6.822199743576558e-05, "loss": 0.15794326066970826, "memory(GiB)": 70.5, "step": 44495, "token_acc": 0.962457337883959, "train_speed(iter/s)": 1.452173 }, { "epoch": 1.9065164303157534, "grad_norm": 5.031870365142822, "learning_rate": 6.82157303250944e-05, "loss": 0.4281972885131836, "memory(GiB)": 70.5, "step": 44500, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.452171 }, { "epoch": 1.9065164303157534, "eval_loss": 2.5822436809539795, "eval_runtime": 13.3671, "eval_samples_per_second": 7.481, "eval_steps_per_second": 7.481, "eval_token_acc": 0.4588938714499253, "step": 44500 }, { "epoch": 1.9067306456450024, "grad_norm": 2.8330776691436768, "learning_rate": 6.820946288442816e-05, "loss": 0.45418405532836914, "memory(GiB)": 70.5, "step": 44505, "token_acc": 0.6030456852791878, "train_speed(iter/s)": 1.451492 }, { "epoch": 1.9069448609742512, "grad_norm": 3.3618788719177246, "learning_rate": 6.820319511388043e-05, "loss": 0.20016241073608398, "memory(GiB)": 70.5, "step": 44510, "token_acc": 0.9603658536585366, "train_speed(iter/s)": 1.451502 }, { "epoch": 1.9071590763035002, "grad_norm": 1.468738079071045, "learning_rate": 6.819692701356474e-05, "loss": 0.37876389026641843, "memory(GiB)": 70.5, "step": 44515, "token_acc": 0.9309309309309309, "train_speed(iter/s)": 1.451502 }, { "epoch": 1.9073732916327493, "grad_norm": 2.7426233291625977, "learning_rate": 6.819065858359464e-05, "loss": 0.4178001403808594, "memory(GiB)": 70.5, "step": 44520, "token_acc": 0.890625, "train_speed(iter/s)": 1.451515 }, { "epoch": 1.907587506961998, "grad_norm": 4.271426677703857, "learning_rate": 6.818438982408371e-05, "loss": 0.5631073474884033, "memory(GiB)": 70.5, "step": 44525, "token_acc": 0.9113149847094801, "train_speed(iter/s)": 1.45152 }, { "epoch": 1.9078017222912471, "grad_norm": 3.7489676475524902, "learning_rate": 6.81781207351455e-05, "loss": 0.4144099712371826, "memory(GiB)": 70.5, "step": 44530, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.45152 }, { "epoch": 1.9080159376204961, "grad_norm": 0.09867183119058609, "learning_rate": 6.817185131689356e-05, "loss": 0.264039421081543, "memory(GiB)": 70.5, "step": 44535, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.451534 }, { "epoch": 1.908230152949745, "grad_norm": 2.7261924743652344, "learning_rate": 6.816558156944151e-05, "loss": 0.34284775257110595, "memory(GiB)": 70.5, "step": 44540, "token_acc": 0.9244712990936556, "train_speed(iter/s)": 1.451522 }, { "epoch": 1.908444368278994, "grad_norm": 5.746874809265137, "learning_rate": 6.81593114929029e-05, "loss": 0.3505500078201294, "memory(GiB)": 70.5, "step": 44545, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.451527 }, { "epoch": 1.908658583608243, "grad_norm": 2.1750881671905518, "learning_rate": 6.815304108739133e-05, "loss": 0.3730180740356445, "memory(GiB)": 70.5, "step": 44550, "token_acc": 0.90625, "train_speed(iter/s)": 1.451533 }, { "epoch": 1.9088727989374918, "grad_norm": 1.2986565828323364, "learning_rate": 6.814677035302038e-05, "loss": 0.311111307144165, "memory(GiB)": 70.5, "step": 44555, "token_acc": 0.9446640316205533, "train_speed(iter/s)": 1.451531 }, { "epoch": 1.9090870142667409, "grad_norm": 2.953409433364868, "learning_rate": 6.814049928990369e-05, "loss": 0.28985164165496824, "memory(GiB)": 70.5, "step": 44560, "token_acc": 0.9369369369369369, "train_speed(iter/s)": 1.45153 }, { "epoch": 1.90930122959599, "grad_norm": 3.698091506958008, "learning_rate": 6.81342278981548e-05, "loss": 0.2631662607192993, "memory(GiB)": 70.5, "step": 44565, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.451531 }, { "epoch": 1.9095154449252387, "grad_norm": 3.179194927215576, "learning_rate": 6.812795617788739e-05, "loss": 0.5168797492980957, "memory(GiB)": 70.5, "step": 44570, "token_acc": 0.8900343642611683, "train_speed(iter/s)": 1.451533 }, { "epoch": 1.9097296602544878, "grad_norm": 3.4087271690368652, "learning_rate": 6.812168412921504e-05, "loss": 0.5069704055786133, "memory(GiB)": 70.5, "step": 44575, "token_acc": 0.8557377049180328, "train_speed(iter/s)": 1.451538 }, { "epoch": 1.9099438755837368, "grad_norm": 6.239006996154785, "learning_rate": 6.811541175225138e-05, "loss": 0.8961823463439942, "memory(GiB)": 70.5, "step": 44580, "token_acc": 0.8214285714285714, "train_speed(iter/s)": 1.451566 }, { "epoch": 1.9101580909129856, "grad_norm": 2.1917641162872314, "learning_rate": 6.810913904711004e-05, "loss": 0.31635465621948244, "memory(GiB)": 70.5, "step": 44585, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.451567 }, { "epoch": 1.9103723062422346, "grad_norm": 4.616771697998047, "learning_rate": 6.810286601390466e-05, "loss": 0.6139188766479492, "memory(GiB)": 70.5, "step": 44590, "token_acc": 0.8716216216216216, "train_speed(iter/s)": 1.451572 }, { "epoch": 1.9105865215714837, "grad_norm": 7.062067985534668, "learning_rate": 6.809659265274888e-05, "loss": 0.5636542320251465, "memory(GiB)": 70.5, "step": 44595, "token_acc": 0.889273356401384, "train_speed(iter/s)": 1.451623 }, { "epoch": 1.9108007369007325, "grad_norm": 2.031219244003296, "learning_rate": 6.809031896375636e-05, "loss": 0.36254119873046875, "memory(GiB)": 70.5, "step": 44600, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.451639 }, { "epoch": 1.9110149522299815, "grad_norm": 3.517127752304077, "learning_rate": 6.808404494704073e-05, "loss": 0.5082364559173584, "memory(GiB)": 70.5, "step": 44605, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.451636 }, { "epoch": 1.9112291675592306, "grad_norm": 0.6588791012763977, "learning_rate": 6.807777060271566e-05, "loss": 0.32592434883117677, "memory(GiB)": 70.5, "step": 44610, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.45164 }, { "epoch": 1.9114433828884794, "grad_norm": 5.07497501373291, "learning_rate": 6.807149593089484e-05, "loss": 0.4133310317993164, "memory(GiB)": 70.5, "step": 44615, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.451651 }, { "epoch": 1.9116575982177286, "grad_norm": 5.534651756286621, "learning_rate": 6.806522093169189e-05, "loss": 0.6708320617675781, "memory(GiB)": 70.5, "step": 44620, "token_acc": 0.8698630136986302, "train_speed(iter/s)": 1.451671 }, { "epoch": 1.9118718135469774, "grad_norm": 5.740650177001953, "learning_rate": 6.805894560522051e-05, "loss": 0.7326728820800781, "memory(GiB)": 70.5, "step": 44625, "token_acc": 0.8426573426573427, "train_speed(iter/s)": 1.451671 }, { "epoch": 1.9120860288762263, "grad_norm": 0.3559148907661438, "learning_rate": 6.805266995159442e-05, "loss": 0.35750460624694824, "memory(GiB)": 70.5, "step": 44630, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.451673 }, { "epoch": 1.9123002442054755, "grad_norm": 1.895262360572815, "learning_rate": 6.804639397092726e-05, "loss": 0.4345544338226318, "memory(GiB)": 70.5, "step": 44635, "token_acc": 0.8885542168674698, "train_speed(iter/s)": 1.451665 }, { "epoch": 1.9125144595347243, "grad_norm": 3.1951687335968018, "learning_rate": 6.804011766333276e-05, "loss": 0.40873026847839355, "memory(GiB)": 70.5, "step": 44640, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.451673 }, { "epoch": 1.9127286748639731, "grad_norm": 3.6093099117279053, "learning_rate": 6.80338410289246e-05, "loss": 0.2542152166366577, "memory(GiB)": 70.5, "step": 44645, "token_acc": 0.9396825396825397, "train_speed(iter/s)": 1.451666 }, { "epoch": 1.9129428901932224, "grad_norm": 4.126529216766357, "learning_rate": 6.802756406781648e-05, "loss": 0.40642237663269043, "memory(GiB)": 70.5, "step": 44650, "token_acc": 0.9110320284697508, "train_speed(iter/s)": 1.451698 }, { "epoch": 1.9131571055224712, "grad_norm": 3.695253372192383, "learning_rate": 6.802128678012214e-05, "loss": 0.40708270072937014, "memory(GiB)": 70.5, "step": 44655, "token_acc": 0.9022988505747126, "train_speed(iter/s)": 1.451687 }, { "epoch": 1.91337132085172, "grad_norm": 2.20450758934021, "learning_rate": 6.801500916595527e-05, "loss": 0.4151634216308594, "memory(GiB)": 70.5, "step": 44660, "token_acc": 0.9048991354466859, "train_speed(iter/s)": 1.451692 }, { "epoch": 1.9135855361809693, "grad_norm": 3.1131157875061035, "learning_rate": 6.800873122542962e-05, "loss": 0.538954782485962, "memory(GiB)": 70.5, "step": 44665, "token_acc": 0.8624161073825504, "train_speed(iter/s)": 1.451691 }, { "epoch": 1.913799751510218, "grad_norm": 3.145223617553711, "learning_rate": 6.80024529586589e-05, "loss": 0.6051069736480713, "memory(GiB)": 70.5, "step": 44670, "token_acc": 0.8601036269430051, "train_speed(iter/s)": 1.451688 }, { "epoch": 1.914013966839467, "grad_norm": 4.7823967933654785, "learning_rate": 6.799617436575688e-05, "loss": 0.6686517715454101, "memory(GiB)": 70.5, "step": 44675, "token_acc": 0.8591065292096219, "train_speed(iter/s)": 1.451693 }, { "epoch": 1.9142281821687162, "grad_norm": 3.052830696105957, "learning_rate": 6.798989544683725e-05, "loss": 0.2906618595123291, "memory(GiB)": 70.5, "step": 44680, "token_acc": 0.9407114624505929, "train_speed(iter/s)": 1.451692 }, { "epoch": 1.914442397497965, "grad_norm": 2.6229031085968018, "learning_rate": 6.798361620201382e-05, "loss": 0.6502031803131103, "memory(GiB)": 70.5, "step": 44685, "token_acc": 0.8710691823899371, "train_speed(iter/s)": 1.451696 }, { "epoch": 1.9146566128272138, "grad_norm": 5.083703517913818, "learning_rate": 6.797733663140028e-05, "loss": 0.2944627285003662, "memory(GiB)": 70.5, "step": 44690, "token_acc": 0.9346153846153846, "train_speed(iter/s)": 1.451698 }, { "epoch": 1.914870828156463, "grad_norm": 5.08853816986084, "learning_rate": 6.797105673511044e-05, "loss": 0.43045845031738283, "memory(GiB)": 70.5, "step": 44695, "token_acc": 0.9145569620253164, "train_speed(iter/s)": 1.451746 }, { "epoch": 1.9150850434857118, "grad_norm": 4.0116777420043945, "learning_rate": 6.796477651325805e-05, "loss": 0.46267046928405764, "memory(GiB)": 70.5, "step": 44700, "token_acc": 0.885593220338983, "train_speed(iter/s)": 1.451765 }, { "epoch": 1.9152992588149607, "grad_norm": 2.6358730792999268, "learning_rate": 6.795849596595686e-05, "loss": 0.2872361183166504, "memory(GiB)": 70.5, "step": 44705, "token_acc": 0.9280575539568345, "train_speed(iter/s)": 1.45178 }, { "epoch": 1.91551347414421, "grad_norm": 1.3974169492721558, "learning_rate": 6.79522150933207e-05, "loss": 0.4041621685028076, "memory(GiB)": 70.5, "step": 44710, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.451812 }, { "epoch": 1.9157276894734587, "grad_norm": 9.347654342651367, "learning_rate": 6.794593389546331e-05, "loss": 0.3436591386795044, "memory(GiB)": 70.5, "step": 44715, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.451811 }, { "epoch": 1.9159419048027075, "grad_norm": 4.108773708343506, "learning_rate": 6.793965237249848e-05, "loss": 0.3402069330215454, "memory(GiB)": 70.5, "step": 44720, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.451818 }, { "epoch": 1.9161561201319568, "grad_norm": 2.555652379989624, "learning_rate": 6.793337052454005e-05, "loss": 0.4560938835144043, "memory(GiB)": 70.5, "step": 44725, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.451824 }, { "epoch": 1.9163703354612056, "grad_norm": 3.066060781478882, "learning_rate": 6.792708835170177e-05, "loss": 0.43596491813659666, "memory(GiB)": 70.5, "step": 44730, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.451819 }, { "epoch": 1.9165845507904544, "grad_norm": 2.485666513442993, "learning_rate": 6.792080585409746e-05, "loss": 0.4254495143890381, "memory(GiB)": 70.5, "step": 44735, "token_acc": 0.9007633587786259, "train_speed(iter/s)": 1.451814 }, { "epoch": 1.9167987661197037, "grad_norm": 4.297685623168945, "learning_rate": 6.791452303184094e-05, "loss": 0.2446979522705078, "memory(GiB)": 70.5, "step": 44740, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.451814 }, { "epoch": 1.9170129814489525, "grad_norm": 0.8299254775047302, "learning_rate": 6.790823988504606e-05, "loss": 0.587213659286499, "memory(GiB)": 70.5, "step": 44745, "token_acc": 0.8798701298701299, "train_speed(iter/s)": 1.451822 }, { "epoch": 1.9172271967782013, "grad_norm": 0.40856438875198364, "learning_rate": 6.790195641382658e-05, "loss": 0.6289827823638916, "memory(GiB)": 70.5, "step": 44750, "token_acc": 0.8461538461538461, "train_speed(iter/s)": 1.451825 }, { "epoch": 1.9174414121074506, "grad_norm": 1.237499713897705, "learning_rate": 6.789567261829639e-05, "loss": 0.35813260078430176, "memory(GiB)": 70.5, "step": 44755, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.451827 }, { "epoch": 1.9176556274366994, "grad_norm": 2.046530246734619, "learning_rate": 6.788938849856929e-05, "loss": 0.36262993812561034, "memory(GiB)": 70.5, "step": 44760, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.451826 }, { "epoch": 1.9178698427659482, "grad_norm": 5.258194446563721, "learning_rate": 6.788310405475915e-05, "loss": 0.5224085807800293, "memory(GiB)": 70.5, "step": 44765, "token_acc": 0.889273356401384, "train_speed(iter/s)": 1.451818 }, { "epoch": 1.9180840580951974, "grad_norm": 4.028496742248535, "learning_rate": 6.78768192869798e-05, "loss": 0.3537308692932129, "memory(GiB)": 70.5, "step": 44770, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.451824 }, { "epoch": 1.9182982734244463, "grad_norm": 4.3320112228393555, "learning_rate": 6.787053419534508e-05, "loss": 0.37033178806304934, "memory(GiB)": 70.5, "step": 44775, "token_acc": 0.9256965944272446, "train_speed(iter/s)": 1.451828 }, { "epoch": 1.918512488753695, "grad_norm": 5.4527201652526855, "learning_rate": 6.786424877996887e-05, "loss": 0.23885741233825683, "memory(GiB)": 70.5, "step": 44780, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.451872 }, { "epoch": 1.9187267040829443, "grad_norm": 0.3533739745616913, "learning_rate": 6.785796304096506e-05, "loss": 0.25617494583129885, "memory(GiB)": 70.5, "step": 44785, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.451871 }, { "epoch": 1.9189409194121931, "grad_norm": 3.6569886207580566, "learning_rate": 6.78516769784475e-05, "loss": 0.3669154644012451, "memory(GiB)": 70.5, "step": 44790, "token_acc": 0.9151515151515152, "train_speed(iter/s)": 1.451892 }, { "epoch": 1.919155134741442, "grad_norm": 1.5739600658416748, "learning_rate": 6.784539059253004e-05, "loss": 0.3693782567977905, "memory(GiB)": 70.5, "step": 44795, "token_acc": 0.9182389937106918, "train_speed(iter/s)": 1.451892 }, { "epoch": 1.9193693500706912, "grad_norm": 2.4321324825286865, "learning_rate": 6.783910388332661e-05, "loss": 0.6033660888671875, "memory(GiB)": 70.5, "step": 44800, "token_acc": 0.8686131386861314, "train_speed(iter/s)": 1.45191 }, { "epoch": 1.91958356539994, "grad_norm": 5.712518215179443, "learning_rate": 6.783281685095105e-05, "loss": 0.3844718933105469, "memory(GiB)": 70.5, "step": 44805, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.451908 }, { "epoch": 1.9197977807291888, "grad_norm": 5.416888236999512, "learning_rate": 6.78265294955173e-05, "loss": 0.202445912361145, "memory(GiB)": 70.5, "step": 44810, "token_acc": 0.9510204081632653, "train_speed(iter/s)": 1.451902 }, { "epoch": 1.920011996058438, "grad_norm": 2.9387216567993164, "learning_rate": 6.782024181713925e-05, "loss": 0.46495256423950193, "memory(GiB)": 70.5, "step": 44815, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.451902 }, { "epoch": 1.920226211387687, "grad_norm": 4.205740928649902, "learning_rate": 6.781395381593082e-05, "loss": 0.35728232860565184, "memory(GiB)": 70.5, "step": 44820, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.451893 }, { "epoch": 1.9204404267169357, "grad_norm": 3.4224395751953125, "learning_rate": 6.780766549200587e-05, "loss": 0.38591291904449465, "memory(GiB)": 70.5, "step": 44825, "token_acc": 0.9433198380566802, "train_speed(iter/s)": 1.451901 }, { "epoch": 1.920654642046185, "grad_norm": 6.905014514923096, "learning_rate": 6.780137684547838e-05, "loss": 0.6474920749664307, "memory(GiB)": 70.5, "step": 44830, "token_acc": 0.8618421052631579, "train_speed(iter/s)": 1.451905 }, { "epoch": 1.9208688573754338, "grad_norm": 1.4017428159713745, "learning_rate": 6.779508787646222e-05, "loss": 0.44762482643127444, "memory(GiB)": 70.5, "step": 44835, "token_acc": 0.9073033707865169, "train_speed(iter/s)": 1.45191 }, { "epoch": 1.9210830727046826, "grad_norm": 4.464878082275391, "learning_rate": 6.778879858507138e-05, "loss": 0.28850584030151366, "memory(GiB)": 70.5, "step": 44840, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.451916 }, { "epoch": 1.9212972880339318, "grad_norm": 2.2045094966888428, "learning_rate": 6.778250897141976e-05, "loss": 0.5159276008605957, "memory(GiB)": 70.5, "step": 44845, "token_acc": 0.8885245901639345, "train_speed(iter/s)": 1.451924 }, { "epoch": 1.9215115033631807, "grad_norm": 2.668731689453125, "learning_rate": 6.777621903562132e-05, "loss": 0.14865370988845825, "memory(GiB)": 70.5, "step": 44850, "token_acc": 0.9577922077922078, "train_speed(iter/s)": 1.451921 }, { "epoch": 1.9217257186924295, "grad_norm": 4.012466907501221, "learning_rate": 6.776992877778997e-05, "loss": 0.379517650604248, "memory(GiB)": 70.5, "step": 44855, "token_acc": 0.9392857142857143, "train_speed(iter/s)": 1.45196 }, { "epoch": 1.9219399340216787, "grad_norm": 3.180234909057617, "learning_rate": 6.77636381980397e-05, "loss": 0.24796345233917236, "memory(GiB)": 70.5, "step": 44860, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.451969 }, { "epoch": 1.9221541493509275, "grad_norm": 2.9542150497436523, "learning_rate": 6.775734729648447e-05, "loss": 0.5861196517944336, "memory(GiB)": 70.5, "step": 44865, "token_acc": 0.8825622775800712, "train_speed(iter/s)": 1.451979 }, { "epoch": 1.9223683646801764, "grad_norm": 2.2186601161956787, "learning_rate": 6.775105607323822e-05, "loss": 0.36651489734649656, "memory(GiB)": 70.5, "step": 44870, "token_acc": 0.9088145896656535, "train_speed(iter/s)": 1.45197 }, { "epoch": 1.9225825800094256, "grad_norm": 1.9636552333831787, "learning_rate": 6.774476452841496e-05, "loss": 0.31376669406890867, "memory(GiB)": 70.5, "step": 44875, "token_acc": 0.9353846153846154, "train_speed(iter/s)": 1.451973 }, { "epoch": 1.9227967953386744, "grad_norm": 1.2472196817398071, "learning_rate": 6.773847266212863e-05, "loss": 0.5491735935211182, "memory(GiB)": 70.5, "step": 44880, "token_acc": 0.89, "train_speed(iter/s)": 1.451976 }, { "epoch": 1.9230110106679232, "grad_norm": 3.642439126968384, "learning_rate": 6.773218047449323e-05, "loss": 0.47922821044921876, "memory(GiB)": 70.5, "step": 44885, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.451972 }, { "epoch": 1.9232252259971725, "grad_norm": 1.882948637008667, "learning_rate": 6.772588796562274e-05, "loss": 0.26516835689544677, "memory(GiB)": 70.5, "step": 44890, "token_acc": 0.9380165289256198, "train_speed(iter/s)": 1.451973 }, { "epoch": 1.9234394413264213, "grad_norm": 2.371131658554077, "learning_rate": 6.771959513563116e-05, "loss": 0.27867281436920166, "memory(GiB)": 70.5, "step": 44895, "token_acc": 0.9553264604810997, "train_speed(iter/s)": 1.451973 }, { "epoch": 1.9236536566556701, "grad_norm": 3.9412150382995605, "learning_rate": 6.77133019846325e-05, "loss": 0.6115674495697021, "memory(GiB)": 70.5, "step": 44900, "token_acc": 0.8631178707224335, "train_speed(iter/s)": 1.451985 }, { "epoch": 1.9238678719849194, "grad_norm": 3.589470863342285, "learning_rate": 6.770700851274077e-05, "loss": 0.3471419334411621, "memory(GiB)": 70.5, "step": 44905, "token_acc": 0.9329268292682927, "train_speed(iter/s)": 1.451993 }, { "epoch": 1.9240820873141682, "grad_norm": 4.5901689529418945, "learning_rate": 6.770071472006994e-05, "loss": 0.39868388175964353, "memory(GiB)": 70.5, "step": 44910, "token_acc": 0.9120521172638436, "train_speed(iter/s)": 1.451997 }, { "epoch": 1.924296302643417, "grad_norm": 4.991305828094482, "learning_rate": 6.769442060673409e-05, "loss": 0.596130084991455, "memory(GiB)": 70.5, "step": 44915, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.451989 }, { "epoch": 1.9245105179726663, "grad_norm": 2.1333351135253906, "learning_rate": 6.768812617284719e-05, "loss": 0.46387171745300293, "memory(GiB)": 70.5, "step": 44920, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.45199 }, { "epoch": 1.924724733301915, "grad_norm": 2.7027719020843506, "learning_rate": 6.76818314185233e-05, "loss": 0.5194323539733887, "memory(GiB)": 70.5, "step": 44925, "token_acc": 0.8805031446540881, "train_speed(iter/s)": 1.451993 }, { "epoch": 1.9249389486311639, "grad_norm": 1.8518084287643433, "learning_rate": 6.767553634387645e-05, "loss": 0.5917342662811279, "memory(GiB)": 70.5, "step": 44930, "token_acc": 0.9161290322580645, "train_speed(iter/s)": 1.452015 }, { "epoch": 1.9251531639604131, "grad_norm": 3.692744493484497, "learning_rate": 6.766924094902067e-05, "loss": 0.27416989803314207, "memory(GiB)": 70.5, "step": 44935, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.45202 }, { "epoch": 1.925367379289662, "grad_norm": 1.7386587858200073, "learning_rate": 6.766294523407002e-05, "loss": 0.24869842529296876, "memory(GiB)": 70.5, "step": 44940, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.452043 }, { "epoch": 1.9255815946189108, "grad_norm": 2.362440586090088, "learning_rate": 6.765664919913856e-05, "loss": 0.20621953010559083, "memory(GiB)": 70.5, "step": 44945, "token_acc": 0.9548387096774194, "train_speed(iter/s)": 1.452044 }, { "epoch": 1.92579580994816, "grad_norm": 0.713252067565918, "learning_rate": 6.765035284434031e-05, "loss": 0.4588137626647949, "memory(GiB)": 70.5, "step": 44950, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.452054 }, { "epoch": 1.9260100252774088, "grad_norm": 1.6493099927902222, "learning_rate": 6.764405616978938e-05, "loss": 0.2653883695602417, "memory(GiB)": 70.5, "step": 44955, "token_acc": 0.9481481481481482, "train_speed(iter/s)": 1.452061 }, { "epoch": 1.9262242406066579, "grad_norm": 2.2456204891204834, "learning_rate": 6.763775917559982e-05, "loss": 0.3508100748062134, "memory(GiB)": 70.5, "step": 44960, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.452061 }, { "epoch": 1.926438455935907, "grad_norm": 3.9899301528930664, "learning_rate": 6.76314618618857e-05, "loss": 0.5295789241790771, "memory(GiB)": 70.5, "step": 44965, "token_acc": 0.8809523809523809, "train_speed(iter/s)": 1.452084 }, { "epoch": 1.9266526712651557, "grad_norm": 4.035673141479492, "learning_rate": 6.76251642287611e-05, "loss": 0.5195623874664307, "memory(GiB)": 70.5, "step": 44970, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.452087 }, { "epoch": 1.9268668865944047, "grad_norm": 3.911750555038452, "learning_rate": 6.761886627634013e-05, "loss": 0.6685268402099609, "memory(GiB)": 70.5, "step": 44975, "token_acc": 0.8731884057971014, "train_speed(iter/s)": 1.452088 }, { "epoch": 1.9270811019236538, "grad_norm": 2.2591755390167236, "learning_rate": 6.761256800473686e-05, "loss": 0.254088306427002, "memory(GiB)": 70.5, "step": 44980, "token_acc": 0.95, "train_speed(iter/s)": 1.452079 }, { "epoch": 1.9272953172529026, "grad_norm": 6.309294700622559, "learning_rate": 6.760626941406541e-05, "loss": 0.6502872467041015, "memory(GiB)": 70.5, "step": 44985, "token_acc": 0.8597785977859779, "train_speed(iter/s)": 1.452095 }, { "epoch": 1.9275095325821516, "grad_norm": 2.1979734897613525, "learning_rate": 6.759997050443987e-05, "loss": 0.6932840824127198, "memory(GiB)": 70.5, "step": 44990, "token_acc": 0.8688524590163934, "train_speed(iter/s)": 1.452084 }, { "epoch": 1.9277237479114007, "grad_norm": 3.4773552417755127, "learning_rate": 6.759367127597436e-05, "loss": 0.45500454902648924, "memory(GiB)": 70.5, "step": 44995, "token_acc": 0.8938356164383562, "train_speed(iter/s)": 1.452082 }, { "epoch": 1.9279379632406495, "grad_norm": 2.8473947048187256, "learning_rate": 6.758737172878298e-05, "loss": 0.37399075031280515, "memory(GiB)": 70.5, "step": 45000, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.452087 }, { "epoch": 1.9279379632406495, "eval_loss": 2.58504581451416, "eval_runtime": 13.6995, "eval_samples_per_second": 7.3, "eval_steps_per_second": 7.3, "eval_token_acc": 0.4470868014268728, "step": 45000 }, { "epoch": 1.9281521785698985, "grad_norm": 8.531996726989746, "learning_rate": 6.758107186297987e-05, "loss": 0.4881740093231201, "memory(GiB)": 70.5, "step": 45005, "token_acc": 0.5594280607685433, "train_speed(iter/s)": 1.451425 }, { "epoch": 1.9283663938991475, "grad_norm": 2.397306442260742, "learning_rate": 6.757477167867914e-05, "loss": 0.3146442174911499, "memory(GiB)": 70.5, "step": 45010, "token_acc": 0.9375, "train_speed(iter/s)": 1.451449 }, { "epoch": 1.9285806092283964, "grad_norm": 0.5518183708190918, "learning_rate": 6.756847117599495e-05, "loss": 0.6006715297698975, "memory(GiB)": 70.5, "step": 45015, "token_acc": 0.8738461538461538, "train_speed(iter/s)": 1.451489 }, { "epoch": 1.9287948245576454, "grad_norm": 0.6488230228424072, "learning_rate": 6.756217035504142e-05, "loss": 0.15174100399017335, "memory(GiB)": 70.5, "step": 45020, "token_acc": 0.9561128526645768, "train_speed(iter/s)": 1.451506 }, { "epoch": 1.9290090398868944, "grad_norm": 1.010709524154663, "learning_rate": 6.755586921593268e-05, "loss": 0.2661752700805664, "memory(GiB)": 70.5, "step": 45025, "token_acc": 0.9469387755102041, "train_speed(iter/s)": 1.45151 }, { "epoch": 1.9292232552161432, "grad_norm": 1.5823372602462769, "learning_rate": 6.754956775878294e-05, "loss": 0.19199846982955932, "memory(GiB)": 70.5, "step": 45030, "token_acc": 0.95, "train_speed(iter/s)": 1.451515 }, { "epoch": 1.9294374705453923, "grad_norm": 7.112021446228027, "learning_rate": 6.754326598370628e-05, "loss": 0.44910688400268556, "memory(GiB)": 70.5, "step": 45035, "token_acc": 0.92, "train_speed(iter/s)": 1.451513 }, { "epoch": 1.9296516858746413, "grad_norm": 0.41226834058761597, "learning_rate": 6.75369638908169e-05, "loss": 0.653288459777832, "memory(GiB)": 70.5, "step": 45040, "token_acc": 0.8537414965986394, "train_speed(iter/s)": 1.451521 }, { "epoch": 1.9298659012038901, "grad_norm": 1.7886251211166382, "learning_rate": 6.7530661480229e-05, "loss": 0.31144840717315675, "memory(GiB)": 70.5, "step": 45045, "token_acc": 0.9343832020997376, "train_speed(iter/s)": 1.451517 }, { "epoch": 1.9300801165331392, "grad_norm": 2.0928609371185303, "learning_rate": 6.75243587520567e-05, "loss": 0.29195261001586914, "memory(GiB)": 70.5, "step": 45050, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.451515 }, { "epoch": 1.9302943318623882, "grad_norm": 3.692504644393921, "learning_rate": 6.751805570641421e-05, "loss": 0.5076125144958497, "memory(GiB)": 70.5, "step": 45055, "token_acc": 0.875, "train_speed(iter/s)": 1.451516 }, { "epoch": 1.930508547191637, "grad_norm": 3.622889518737793, "learning_rate": 6.75117523434157e-05, "loss": 0.47061381340026853, "memory(GiB)": 70.5, "step": 45060, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.451536 }, { "epoch": 1.930722762520886, "grad_norm": 2.7207469940185547, "learning_rate": 6.750544866317536e-05, "loss": 0.23364782333374023, "memory(GiB)": 70.5, "step": 45065, "token_acc": 0.9588607594936709, "train_speed(iter/s)": 1.451548 }, { "epoch": 1.930936977850135, "grad_norm": 2.539652109146118, "learning_rate": 6.749914466580741e-05, "loss": 0.40989270210266116, "memory(GiB)": 70.5, "step": 45070, "token_acc": 0.9054054054054054, "train_speed(iter/s)": 1.451552 }, { "epoch": 1.9311511931793839, "grad_norm": 3.8527233600616455, "learning_rate": 6.749284035142604e-05, "loss": 0.352811861038208, "memory(GiB)": 70.5, "step": 45075, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.451566 }, { "epoch": 1.931365408508633, "grad_norm": 4.568136215209961, "learning_rate": 6.748653572014546e-05, "loss": 0.22986383438110353, "memory(GiB)": 70.5, "step": 45080, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.45156 }, { "epoch": 1.931579623837882, "grad_norm": 2.307692050933838, "learning_rate": 6.748023077207988e-05, "loss": 0.3056380271911621, "memory(GiB)": 70.5, "step": 45085, "token_acc": 0.927797833935018, "train_speed(iter/s)": 1.451556 }, { "epoch": 1.9317938391671308, "grad_norm": 2.487539768218994, "learning_rate": 6.747392550734352e-05, "loss": 0.25531854629516604, "memory(GiB)": 70.5, "step": 45090, "token_acc": 0.937888198757764, "train_speed(iter/s)": 1.451553 }, { "epoch": 1.9320080544963798, "grad_norm": 4.116090297698975, "learning_rate": 6.74676199260506e-05, "loss": 0.3743924140930176, "memory(GiB)": 70.5, "step": 45095, "token_acc": 0.9415584415584416, "train_speed(iter/s)": 1.451557 }, { "epoch": 1.9322222698256288, "grad_norm": 1.2131890058517456, "learning_rate": 6.746131402831537e-05, "loss": 0.17088727951049804, "memory(GiB)": 70.5, "step": 45100, "token_acc": 0.9606557377049181, "train_speed(iter/s)": 1.451567 }, { "epoch": 1.9324364851548776, "grad_norm": 4.109292984008789, "learning_rate": 6.745500781425205e-05, "loss": 0.41065387725830077, "memory(GiB)": 70.5, "step": 45105, "token_acc": 0.9163346613545816, "train_speed(iter/s)": 1.45156 }, { "epoch": 1.9326507004841267, "grad_norm": 2.7234723567962646, "learning_rate": 6.74487012839749e-05, "loss": 0.581898546218872, "memory(GiB)": 70.5, "step": 45110, "token_acc": 0.8929889298892989, "train_speed(iter/s)": 1.451561 }, { "epoch": 1.9328649158133757, "grad_norm": 5.624485492706299, "learning_rate": 6.744239443759813e-05, "loss": 0.5561336040496826, "memory(GiB)": 70.5, "step": 45115, "token_acc": 0.9029850746268657, "train_speed(iter/s)": 1.451579 }, { "epoch": 1.9330791311426245, "grad_norm": 2.2993688583374023, "learning_rate": 6.743608727523605e-05, "loss": 0.24250175952911376, "memory(GiB)": 70.5, "step": 45120, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.451577 }, { "epoch": 1.9332933464718736, "grad_norm": 0.8650975227355957, "learning_rate": 6.742977979700287e-05, "loss": 0.32702600955963135, "memory(GiB)": 70.5, "step": 45125, "token_acc": 0.9297520661157025, "train_speed(iter/s)": 1.45159 }, { "epoch": 1.9335075618011226, "grad_norm": 4.448582172393799, "learning_rate": 6.74234720030129e-05, "loss": 0.4003459930419922, "memory(GiB)": 70.5, "step": 45130, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.451584 }, { "epoch": 1.9337217771303714, "grad_norm": 1.5738221406936646, "learning_rate": 6.741716389338039e-05, "loss": 0.21466028690338135, "memory(GiB)": 70.5, "step": 45135, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.451589 }, { "epoch": 1.9339359924596204, "grad_norm": 3.6689183712005615, "learning_rate": 6.74108554682196e-05, "loss": 0.30264036655426024, "memory(GiB)": 70.5, "step": 45140, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.451605 }, { "epoch": 1.9341502077888695, "grad_norm": 4.208964824676514, "learning_rate": 6.740454672764484e-05, "loss": 0.6159896850585938, "memory(GiB)": 70.5, "step": 45145, "token_acc": 0.8801498127340824, "train_speed(iter/s)": 1.451637 }, { "epoch": 1.9343644231181183, "grad_norm": 2.9617676734924316, "learning_rate": 6.739823767177039e-05, "loss": 0.3256326198577881, "memory(GiB)": 70.5, "step": 45150, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.451641 }, { "epoch": 1.9345786384473673, "grad_norm": 0.7559248805046082, "learning_rate": 6.739192830071054e-05, "loss": 0.5589967727661133, "memory(GiB)": 70.5, "step": 45155, "token_acc": 0.889261744966443, "train_speed(iter/s)": 1.451643 }, { "epoch": 1.9347928537766164, "grad_norm": 2.783951759338379, "learning_rate": 6.73856186145796e-05, "loss": 0.22110388278961182, "memory(GiB)": 70.5, "step": 45160, "token_acc": 0.963265306122449, "train_speed(iter/s)": 1.451651 }, { "epoch": 1.9350070691058652, "grad_norm": 2.823251724243164, "learning_rate": 6.737930861349186e-05, "loss": 0.3259900093078613, "memory(GiB)": 70.5, "step": 45165, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.451652 }, { "epoch": 1.9352212844351142, "grad_norm": 2.5518641471862793, "learning_rate": 6.737299829756165e-05, "loss": 0.7795463085174561, "memory(GiB)": 70.5, "step": 45170, "token_acc": 0.8467741935483871, "train_speed(iter/s)": 1.45165 }, { "epoch": 1.9354354997643632, "grad_norm": 4.144002437591553, "learning_rate": 6.736668766690328e-05, "loss": 0.6250722885131836, "memory(GiB)": 70.5, "step": 45175, "token_acc": 0.8680555555555556, "train_speed(iter/s)": 1.451664 }, { "epoch": 1.935649715093612, "grad_norm": 5.86954402923584, "learning_rate": 6.736037672163107e-05, "loss": 0.4875943660736084, "memory(GiB)": 70.5, "step": 45180, "token_acc": 0.8633333333333333, "train_speed(iter/s)": 1.451683 }, { "epoch": 1.935863930422861, "grad_norm": 2.957733631134033, "learning_rate": 6.735406546185934e-05, "loss": 0.33964715003967283, "memory(GiB)": 70.5, "step": 45185, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.45168 }, { "epoch": 1.9360781457521101, "grad_norm": 4.267940044403076, "learning_rate": 6.734775388770245e-05, "loss": 0.35851540565490725, "memory(GiB)": 70.5, "step": 45190, "token_acc": 0.9148148148148149, "train_speed(iter/s)": 1.451684 }, { "epoch": 1.936292361081359, "grad_norm": 0.9028540849685669, "learning_rate": 6.734144199927472e-05, "loss": 0.26942174434661864, "memory(GiB)": 70.5, "step": 45195, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.451689 }, { "epoch": 1.936506576410608, "grad_norm": 3.614943742752075, "learning_rate": 6.73351297966905e-05, "loss": 0.4041290760040283, "memory(GiB)": 70.5, "step": 45200, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.451682 }, { "epoch": 1.936720791739857, "grad_norm": 1.882200002670288, "learning_rate": 6.732881728006414e-05, "loss": 0.42886695861816404, "memory(GiB)": 70.5, "step": 45205, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.451699 }, { "epoch": 1.9369350070691058, "grad_norm": 3.9106593132019043, "learning_rate": 6.732250444951e-05, "loss": 0.3522289276123047, "memory(GiB)": 70.5, "step": 45210, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.451712 }, { "epoch": 1.9371492223983549, "grad_norm": 3.130099296569824, "learning_rate": 6.731619130514243e-05, "loss": 0.6233046054840088, "memory(GiB)": 70.5, "step": 45215, "token_acc": 0.8829787234042553, "train_speed(iter/s)": 1.45173 }, { "epoch": 1.9373634377276039, "grad_norm": 1.4244327545166016, "learning_rate": 6.730987784707584e-05, "loss": 0.2064577579498291, "memory(GiB)": 70.5, "step": 45220, "token_acc": 0.9513888888888888, "train_speed(iter/s)": 1.451736 }, { "epoch": 1.9375776530568527, "grad_norm": 1.9566102027893066, "learning_rate": 6.730356407542456e-05, "loss": 0.5169362545013427, "memory(GiB)": 70.5, "step": 45225, "token_acc": 0.887459807073955, "train_speed(iter/s)": 1.451747 }, { "epoch": 1.9377918683861017, "grad_norm": 6.205230712890625, "learning_rate": 6.729724999030297e-05, "loss": 0.37291603088378905, "memory(GiB)": 70.5, "step": 45230, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.451744 }, { "epoch": 1.9380060837153508, "grad_norm": 5.568604946136475, "learning_rate": 6.729093559182549e-05, "loss": 0.6486070632934571, "memory(GiB)": 70.5, "step": 45235, "token_acc": 0.8776371308016878, "train_speed(iter/s)": 1.451732 }, { "epoch": 1.9382202990445996, "grad_norm": 3.1603662967681885, "learning_rate": 6.728462088010647e-05, "loss": 0.42101221084594725, "memory(GiB)": 70.5, "step": 45240, "token_acc": 0.9058441558441559, "train_speed(iter/s)": 1.45173 }, { "epoch": 1.9384345143738486, "grad_norm": 6.493555068969727, "learning_rate": 6.727830585526035e-05, "loss": 0.5338586807250977, "memory(GiB)": 70.5, "step": 45245, "token_acc": 0.8986928104575164, "train_speed(iter/s)": 1.451728 }, { "epoch": 1.9386487297030977, "grad_norm": 3.5365731716156006, "learning_rate": 6.727199051740151e-05, "loss": 0.5666374683380127, "memory(GiB)": 70.5, "step": 45250, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.451753 }, { "epoch": 1.9388629450323465, "grad_norm": 3.699120283126831, "learning_rate": 6.726567486664434e-05, "loss": 0.33085706233978274, "memory(GiB)": 70.5, "step": 45255, "token_acc": 0.933933933933934, "train_speed(iter/s)": 1.451753 }, { "epoch": 1.9390771603615955, "grad_norm": 2.0212972164154053, "learning_rate": 6.725935890310328e-05, "loss": 0.3810264110565186, "memory(GiB)": 70.5, "step": 45260, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.451762 }, { "epoch": 1.9392913756908445, "grad_norm": 2.162177562713623, "learning_rate": 6.725304262689275e-05, "loss": 0.3037224531173706, "memory(GiB)": 70.5, "step": 45265, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.451763 }, { "epoch": 1.9395055910200933, "grad_norm": 5.221333026885986, "learning_rate": 6.724672603812717e-05, "loss": 0.42095260620117186, "memory(GiB)": 70.5, "step": 45270, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.451776 }, { "epoch": 1.9397198063493424, "grad_norm": 0.5565624237060547, "learning_rate": 6.724040913692098e-05, "loss": 0.24775445461273193, "memory(GiB)": 70.5, "step": 45275, "token_acc": 0.953405017921147, "train_speed(iter/s)": 1.451787 }, { "epoch": 1.9399340216785914, "grad_norm": 2.825206756591797, "learning_rate": 6.723409192338859e-05, "loss": 0.3672954559326172, "memory(GiB)": 70.5, "step": 45280, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.451785 }, { "epoch": 1.9401482370078402, "grad_norm": 4.329905986785889, "learning_rate": 6.722777439764446e-05, "loss": 0.34661839008331297, "memory(GiB)": 70.5, "step": 45285, "token_acc": 0.925, "train_speed(iter/s)": 1.451788 }, { "epoch": 1.9403624523370893, "grad_norm": 10.58515453338623, "learning_rate": 6.722145655980304e-05, "loss": 0.2783386468887329, "memory(GiB)": 70.5, "step": 45290, "token_acc": 0.9240924092409241, "train_speed(iter/s)": 1.451794 }, { "epoch": 1.9405766676663383, "grad_norm": 3.569965362548828, "learning_rate": 6.721513840997878e-05, "loss": 0.3969924211502075, "memory(GiB)": 70.5, "step": 45295, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.451794 }, { "epoch": 1.940790882995587, "grad_norm": 3.681288003921509, "learning_rate": 6.720881994828612e-05, "loss": 0.509169054031372, "memory(GiB)": 70.5, "step": 45300, "token_acc": 0.8866396761133604, "train_speed(iter/s)": 1.451805 }, { "epoch": 1.9410050983248361, "grad_norm": 2.98722505569458, "learning_rate": 6.720250117483956e-05, "loss": 0.5183483123779297, "memory(GiB)": 70.5, "step": 45305, "token_acc": 0.8873720136518771, "train_speed(iter/s)": 1.451795 }, { "epoch": 1.9412193136540852, "grad_norm": 5.668407440185547, "learning_rate": 6.719618208975355e-05, "loss": 0.5148998260498047, "memory(GiB)": 70.5, "step": 45310, "token_acc": 0.9015873015873016, "train_speed(iter/s)": 1.451796 }, { "epoch": 1.941433528983334, "grad_norm": 2.071547269821167, "learning_rate": 6.718986269314257e-05, "loss": 0.3433209419250488, "memory(GiB)": 70.5, "step": 45315, "token_acc": 0.9227941176470589, "train_speed(iter/s)": 1.451803 }, { "epoch": 1.941647744312583, "grad_norm": 2.3032753467559814, "learning_rate": 6.718354298512111e-05, "loss": 0.3925936698913574, "memory(GiB)": 70.5, "step": 45320, "token_acc": 0.9400749063670412, "train_speed(iter/s)": 1.451817 }, { "epoch": 1.941861959641832, "grad_norm": 1.4310706853866577, "learning_rate": 6.717722296580364e-05, "loss": 0.22864339351654053, "memory(GiB)": 70.5, "step": 45325, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.451823 }, { "epoch": 1.9420761749710809, "grad_norm": 4.577850341796875, "learning_rate": 6.717090263530466e-05, "loss": 0.5222259521484375, "memory(GiB)": 70.5, "step": 45330, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.45182 }, { "epoch": 1.94229039030033, "grad_norm": 2.3721232414245605, "learning_rate": 6.716458199373868e-05, "loss": 0.39238853454589845, "memory(GiB)": 70.5, "step": 45335, "token_acc": 0.8892508143322475, "train_speed(iter/s)": 1.451842 }, { "epoch": 1.942504605629579, "grad_norm": 3.1107490062713623, "learning_rate": 6.71582610412202e-05, "loss": 0.3132979393005371, "memory(GiB)": 70.5, "step": 45340, "token_acc": 0.9342105263157895, "train_speed(iter/s)": 1.451851 }, { "epoch": 1.9427188209588278, "grad_norm": 3.267244577407837, "learning_rate": 6.71519397778637e-05, "loss": 0.42044887542724607, "memory(GiB)": 70.5, "step": 45345, "token_acc": 0.9083665338645418, "train_speed(iter/s)": 1.451861 }, { "epoch": 1.9429330362880768, "grad_norm": 0.24007348716259003, "learning_rate": 6.714561820378375e-05, "loss": 0.3088704586029053, "memory(GiB)": 70.5, "step": 45350, "token_acc": 0.9581589958158996, "train_speed(iter/s)": 1.451865 }, { "epoch": 1.9431472516173258, "grad_norm": 4.681672096252441, "learning_rate": 6.713929631909483e-05, "loss": 0.3498310327529907, "memory(GiB)": 70.5, "step": 45355, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.451869 }, { "epoch": 1.9433614669465746, "grad_norm": 5.731127738952637, "learning_rate": 6.713297412391148e-05, "loss": 0.4983759880065918, "memory(GiB)": 70.5, "step": 45360, "token_acc": 0.8844765342960289, "train_speed(iter/s)": 1.451882 }, { "epoch": 1.9435756822758237, "grad_norm": 4.680256366729736, "learning_rate": 6.712665161834822e-05, "loss": 0.4205026149749756, "memory(GiB)": 70.5, "step": 45365, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.451911 }, { "epoch": 1.9437898976050727, "grad_norm": 2.2142794132232666, "learning_rate": 6.712032880251962e-05, "loss": 0.5960860252380371, "memory(GiB)": 70.5, "step": 45370, "token_acc": 0.9, "train_speed(iter/s)": 1.451922 }, { "epoch": 1.9440041129343215, "grad_norm": 1.6411354541778564, "learning_rate": 6.711400567654019e-05, "loss": 0.24636800289154054, "memory(GiB)": 70.5, "step": 45375, "token_acc": 0.9416909620991254, "train_speed(iter/s)": 1.45193 }, { "epoch": 1.9442183282635705, "grad_norm": 1.9471813440322876, "learning_rate": 6.710768224052451e-05, "loss": 0.3286593914031982, "memory(GiB)": 70.5, "step": 45380, "token_acc": 0.9480968858131488, "train_speed(iter/s)": 1.451922 }, { "epoch": 1.9444325435928196, "grad_norm": 3.4015767574310303, "learning_rate": 6.710135849458709e-05, "loss": 0.29661853313446046, "memory(GiB)": 70.5, "step": 45385, "token_acc": 0.9361022364217252, "train_speed(iter/s)": 1.45192 }, { "epoch": 1.9446467589220684, "grad_norm": 2.7002968788146973, "learning_rate": 6.709503443884255e-05, "loss": 0.5779501438140869, "memory(GiB)": 70.5, "step": 45390, "token_acc": 0.899641577060932, "train_speed(iter/s)": 1.45194 }, { "epoch": 1.9448609742513174, "grad_norm": 1.7622663974761963, "learning_rate": 6.708871007340541e-05, "loss": 0.2433779239654541, "memory(GiB)": 70.5, "step": 45395, "token_acc": 0.926605504587156, "train_speed(iter/s)": 1.451935 }, { "epoch": 1.9450751895805665, "grad_norm": 3.9232192039489746, "learning_rate": 6.708238539839025e-05, "loss": 0.6993198871612549, "memory(GiB)": 70.5, "step": 45400, "token_acc": 0.8739837398373984, "train_speed(iter/s)": 1.451954 }, { "epoch": 1.9452894049098153, "grad_norm": 8.363224983215332, "learning_rate": 6.707606041391165e-05, "loss": 0.45844459533691406, "memory(GiB)": 70.5, "step": 45405, "token_acc": 0.8937728937728938, "train_speed(iter/s)": 1.451967 }, { "epoch": 1.9455036202390643, "grad_norm": 2.2159972190856934, "learning_rate": 6.706973512008421e-05, "loss": 0.3295897006988525, "memory(GiB)": 70.5, "step": 45410, "token_acc": 0.9240506329113924, "train_speed(iter/s)": 1.451986 }, { "epoch": 1.9457178355683133, "grad_norm": 6.14854097366333, "learning_rate": 6.70634095170225e-05, "loss": 0.43574962615966795, "memory(GiB)": 70.5, "step": 45415, "token_acc": 0.9214876033057852, "train_speed(iter/s)": 1.451983 }, { "epoch": 1.9459320508975622, "grad_norm": 3.1121699810028076, "learning_rate": 6.705708360484111e-05, "loss": 0.3184756517410278, "memory(GiB)": 70.5, "step": 45420, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.451977 }, { "epoch": 1.9461462662268112, "grad_norm": 12.031450271606445, "learning_rate": 6.705075738365468e-05, "loss": 0.2891284465789795, "memory(GiB)": 70.5, "step": 45425, "token_acc": 0.946843853820598, "train_speed(iter/s)": 1.451972 }, { "epoch": 1.9463604815560602, "grad_norm": 3.698270082473755, "learning_rate": 6.704443085357777e-05, "loss": 0.2842875957489014, "memory(GiB)": 70.5, "step": 45430, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.451982 }, { "epoch": 1.946574696885309, "grad_norm": 6.184753894805908, "learning_rate": 6.7038104014725e-05, "loss": 0.3130188465118408, "memory(GiB)": 70.5, "step": 45435, "token_acc": 0.9414225941422594, "train_speed(iter/s)": 1.451991 }, { "epoch": 1.946788912214558, "grad_norm": 3.0905404090881348, "learning_rate": 6.703177686721098e-05, "loss": 0.438897705078125, "memory(GiB)": 70.5, "step": 45440, "token_acc": 0.9195046439628483, "train_speed(iter/s)": 1.452002 }, { "epoch": 1.947003127543807, "grad_norm": 2.667320966720581, "learning_rate": 6.702544941115036e-05, "loss": 0.33676486015319823, "memory(GiB)": 70.5, "step": 45445, "token_acc": 0.9470198675496688, "train_speed(iter/s)": 1.452004 }, { "epoch": 1.947217342873056, "grad_norm": 2.2138543128967285, "learning_rate": 6.701912164665776e-05, "loss": 0.5606766223907471, "memory(GiB)": 70.5, "step": 45450, "token_acc": 0.8759398496240601, "train_speed(iter/s)": 1.452005 }, { "epoch": 1.947431558202305, "grad_norm": 1.3388304710388184, "learning_rate": 6.70127935738478e-05, "loss": 0.39470930099487306, "memory(GiB)": 70.5, "step": 45455, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.452005 }, { "epoch": 1.947645773531554, "grad_norm": 1.1418355703353882, "learning_rate": 6.700646519283513e-05, "loss": 0.19720749855041503, "memory(GiB)": 70.5, "step": 45460, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.452011 }, { "epoch": 1.9478599888608028, "grad_norm": 1.2554430961608887, "learning_rate": 6.700013650373438e-05, "loss": 0.31516208648681643, "memory(GiB)": 70.5, "step": 45465, "token_acc": 0.9085173501577287, "train_speed(iter/s)": 1.452005 }, { "epoch": 1.9480742041900518, "grad_norm": 2.6695430278778076, "learning_rate": 6.699380750666023e-05, "loss": 0.3350430965423584, "memory(GiB)": 70.5, "step": 45470, "token_acc": 0.933852140077821, "train_speed(iter/s)": 1.452009 }, { "epoch": 1.9482884195193009, "grad_norm": 0.7755566239356995, "learning_rate": 6.698747820172728e-05, "loss": 0.3913798570632935, "memory(GiB)": 70.5, "step": 45475, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.452024 }, { "epoch": 1.9485026348485497, "grad_norm": 3.7485904693603516, "learning_rate": 6.698114858905024e-05, "loss": 0.2254849672317505, "memory(GiB)": 70.5, "step": 45480, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.452016 }, { "epoch": 1.9487168501777987, "grad_norm": 2.8903656005859375, "learning_rate": 6.697481866874377e-05, "loss": 0.37684292793273927, "memory(GiB)": 70.5, "step": 45485, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.452029 }, { "epoch": 1.9489310655070478, "grad_norm": 3.5611822605133057, "learning_rate": 6.696848844092254e-05, "loss": 0.3155937433242798, "memory(GiB)": 70.5, "step": 45490, "token_acc": 0.9318885448916409, "train_speed(iter/s)": 1.452037 }, { "epoch": 1.9491452808362966, "grad_norm": 4.1729960441589355, "learning_rate": 6.696215790570122e-05, "loss": 0.3996626853942871, "memory(GiB)": 70.5, "step": 45495, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.452051 }, { "epoch": 1.9493594961655456, "grad_norm": 3.4291858673095703, "learning_rate": 6.69558270631945e-05, "loss": 0.3574655771255493, "memory(GiB)": 70.5, "step": 45500, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.452041 }, { "epoch": 1.9493594961655456, "eval_loss": 2.461495876312256, "eval_runtime": 13.9872, "eval_samples_per_second": 7.149, "eval_steps_per_second": 7.149, "eval_token_acc": 0.46551724137931033, "step": 45500 }, { "epoch": 1.9495737114947946, "grad_norm": 1.4339064359664917, "learning_rate": 6.694949591351706e-05, "loss": 0.24376363754272462, "memory(GiB)": 70.5, "step": 45505, "token_acc": 0.5943852855759922, "train_speed(iter/s)": 1.451373 }, { "epoch": 1.9497879268240434, "grad_norm": 3.793579339981079, "learning_rate": 6.694316445678363e-05, "loss": 0.22704267501831055, "memory(GiB)": 70.5, "step": 45510, "token_acc": 0.9578544061302682, "train_speed(iter/s)": 1.451374 }, { "epoch": 1.9500021421532925, "grad_norm": 3.6343114376068115, "learning_rate": 6.693683269310884e-05, "loss": 0.348048996925354, "memory(GiB)": 70.5, "step": 45515, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.451368 }, { "epoch": 1.9502163574825415, "grad_norm": 3.080909252166748, "learning_rate": 6.693050062260747e-05, "loss": 0.48906478881835935, "memory(GiB)": 70.5, "step": 45520, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.451366 }, { "epoch": 1.9504305728117903, "grad_norm": 4.848405838012695, "learning_rate": 6.692416824539418e-05, "loss": 0.34385249614715574, "memory(GiB)": 70.5, "step": 45525, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.45138 }, { "epoch": 1.9506447881410394, "grad_norm": 8.16375732421875, "learning_rate": 6.69178355615837e-05, "loss": 0.3943598985671997, "memory(GiB)": 70.5, "step": 45530, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.451422 }, { "epoch": 1.9508590034702884, "grad_norm": 3.05499005317688, "learning_rate": 6.691150257129077e-05, "loss": 0.5812463760375977, "memory(GiB)": 70.5, "step": 45535, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.451426 }, { "epoch": 1.9510732187995372, "grad_norm": 4.708807945251465, "learning_rate": 6.690516927463012e-05, "loss": 0.43180408477783205, "memory(GiB)": 70.5, "step": 45540, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.451443 }, { "epoch": 1.9512874341287862, "grad_norm": 11.820655822753906, "learning_rate": 6.689883567171646e-05, "loss": 0.6001622200012207, "memory(GiB)": 70.5, "step": 45545, "token_acc": 0.8550185873605948, "train_speed(iter/s)": 1.451463 }, { "epoch": 1.9515016494580353, "grad_norm": 3.0742766857147217, "learning_rate": 6.689250176266452e-05, "loss": 0.2992063283920288, "memory(GiB)": 70.5, "step": 45550, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.451464 }, { "epoch": 1.951715864787284, "grad_norm": 0.18404535949230194, "learning_rate": 6.688616754758908e-05, "loss": 0.2822566986083984, "memory(GiB)": 70.5, "step": 45555, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.451476 }, { "epoch": 1.9519300801165331, "grad_norm": 2.5942273139953613, "learning_rate": 6.687983302660487e-05, "loss": 0.220491361618042, "memory(GiB)": 70.5, "step": 45560, "token_acc": 0.9476744186046512, "train_speed(iter/s)": 1.451479 }, { "epoch": 1.9521442954457822, "grad_norm": 4.00173282623291, "learning_rate": 6.687349819982667e-05, "loss": 0.6753308296203613, "memory(GiB)": 70.5, "step": 45565, "token_acc": 0.8515151515151516, "train_speed(iter/s)": 1.45148 }, { "epoch": 1.952358510775031, "grad_norm": 1.291894555091858, "learning_rate": 6.686716306736921e-05, "loss": 0.43312649726867675, "memory(GiB)": 70.5, "step": 45570, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.45148 }, { "epoch": 1.95257272610428, "grad_norm": 2.9749674797058105, "learning_rate": 6.686082762934725e-05, "loss": 0.31299073696136476, "memory(GiB)": 70.5, "step": 45575, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.451495 }, { "epoch": 1.952786941433529, "grad_norm": 4.928858280181885, "learning_rate": 6.68544918858756e-05, "loss": 0.5175841331481934, "memory(GiB)": 70.5, "step": 45580, "token_acc": 0.8694158075601375, "train_speed(iter/s)": 1.451511 }, { "epoch": 1.9530011567627779, "grad_norm": 1.6920076608657837, "learning_rate": 6.684815583706902e-05, "loss": 0.12678853273391724, "memory(GiB)": 70.5, "step": 45585, "token_acc": 0.9700854700854701, "train_speed(iter/s)": 1.451504 }, { "epoch": 1.953215372092027, "grad_norm": 2.2498018741607666, "learning_rate": 6.684181948304228e-05, "loss": 0.21493816375732422, "memory(GiB)": 70.5, "step": 45590, "token_acc": 0.9465875370919882, "train_speed(iter/s)": 1.451514 }, { "epoch": 1.953429587421276, "grad_norm": 0.49199602007865906, "learning_rate": 6.683548282391019e-05, "loss": 0.2130990982055664, "memory(GiB)": 70.5, "step": 45595, "token_acc": 0.963076923076923, "train_speed(iter/s)": 1.451514 }, { "epoch": 1.9536438027505247, "grad_norm": 0.8551783561706543, "learning_rate": 6.682914585978754e-05, "loss": 0.5770214080810547, "memory(GiB)": 70.5, "step": 45600, "token_acc": 0.8923766816143498, "train_speed(iter/s)": 1.451535 }, { "epoch": 1.9538580180797738, "grad_norm": 0.2222834825515747, "learning_rate": 6.682280859078911e-05, "loss": 0.6972807884216309, "memory(GiB)": 70.5, "step": 45605, "token_acc": 0.8645833333333334, "train_speed(iter/s)": 1.451546 }, { "epoch": 1.9540722334090228, "grad_norm": 3.024155855178833, "learning_rate": 6.681647101702973e-05, "loss": 0.31195409297943116, "memory(GiB)": 70.5, "step": 45610, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.451546 }, { "epoch": 1.9542864487382716, "grad_norm": 2.026484966278076, "learning_rate": 6.681013313862421e-05, "loss": 0.3188143730163574, "memory(GiB)": 70.5, "step": 45615, "token_acc": 0.9288389513108615, "train_speed(iter/s)": 1.451538 }, { "epoch": 1.9545006640675207, "grad_norm": 3.910100221633911, "learning_rate": 6.680379495568735e-05, "loss": 0.3633927345275879, "memory(GiB)": 70.5, "step": 45620, "token_acc": 0.9297520661157025, "train_speed(iter/s)": 1.45154 }, { "epoch": 1.9547148793967697, "grad_norm": 3.3132550716400146, "learning_rate": 6.679745646833397e-05, "loss": 0.45610551834106444, "memory(GiB)": 70.5, "step": 45625, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.451542 }, { "epoch": 1.9549290947260185, "grad_norm": 3.0916318893432617, "learning_rate": 6.679111767667894e-05, "loss": 0.4024197578430176, "memory(GiB)": 70.5, "step": 45630, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.451543 }, { "epoch": 1.9551433100552675, "grad_norm": 3.3756866455078125, "learning_rate": 6.678477858083703e-05, "loss": 0.5029620170593262, "memory(GiB)": 70.5, "step": 45635, "token_acc": 0.875, "train_speed(iter/s)": 1.451545 }, { "epoch": 1.9553575253845166, "grad_norm": 2.475494623184204, "learning_rate": 6.677843918092312e-05, "loss": 0.35249345302581786, "memory(GiB)": 70.5, "step": 45640, "token_acc": 0.9427480916030534, "train_speed(iter/s)": 1.451565 }, { "epoch": 1.9555717407137654, "grad_norm": 4.109879493713379, "learning_rate": 6.677209947705205e-05, "loss": 0.3237507104873657, "memory(GiB)": 70.5, "step": 45645, "token_acc": 0.946360153256705, "train_speed(iter/s)": 1.451568 }, { "epoch": 1.9557859560430144, "grad_norm": 1.6195130348205566, "learning_rate": 6.676575946933867e-05, "loss": 0.2068406581878662, "memory(GiB)": 70.5, "step": 45650, "token_acc": 0.9598393574297188, "train_speed(iter/s)": 1.451565 }, { "epoch": 1.9560001713722635, "grad_norm": 3.90722393989563, "learning_rate": 6.675941915789783e-05, "loss": 0.34683985710144044, "memory(GiB)": 70.5, "step": 45655, "token_acc": 0.8793103448275862, "train_speed(iter/s)": 1.451575 }, { "epoch": 1.9562143867015123, "grad_norm": 9.068275451660156, "learning_rate": 6.675307854284438e-05, "loss": 0.4110887050628662, "memory(GiB)": 70.5, "step": 45660, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.45157 }, { "epoch": 1.9564286020307613, "grad_norm": 1.727604627609253, "learning_rate": 6.67467376242932e-05, "loss": 0.22727608680725098, "memory(GiB)": 70.5, "step": 45665, "token_acc": 0.9595141700404858, "train_speed(iter/s)": 1.451558 }, { "epoch": 1.9566428173600103, "grad_norm": 6.8562211990356445, "learning_rate": 6.674039640235917e-05, "loss": 0.5079395771026611, "memory(GiB)": 70.5, "step": 45670, "token_acc": 0.9020771513353115, "train_speed(iter/s)": 1.451575 }, { "epoch": 1.9568570326892591, "grad_norm": 2.737666130065918, "learning_rate": 6.673405487715713e-05, "loss": 0.41080517768859864, "memory(GiB)": 70.5, "step": 45675, "token_acc": 0.9221789883268483, "train_speed(iter/s)": 1.451577 }, { "epoch": 1.9570712480185082, "grad_norm": 1.6437915563583374, "learning_rate": 6.6727713048802e-05, "loss": 0.5270620822906494, "memory(GiB)": 70.5, "step": 45680, "token_acc": 0.8835616438356164, "train_speed(iter/s)": 1.451601 }, { "epoch": 1.9572854633477572, "grad_norm": 1.5522645711898804, "learning_rate": 6.672137091740867e-05, "loss": 0.2977179288864136, "memory(GiB)": 70.5, "step": 45685, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.451636 }, { "epoch": 1.957499678677006, "grad_norm": 5.251926898956299, "learning_rate": 6.671502848309198e-05, "loss": 0.46374974250793455, "memory(GiB)": 70.5, "step": 45690, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.451661 }, { "epoch": 1.957713894006255, "grad_norm": 5.072287559509277, "learning_rate": 6.670868574596691e-05, "loss": 0.29173433780670166, "memory(GiB)": 70.5, "step": 45695, "token_acc": 0.932806324110672, "train_speed(iter/s)": 1.45166 }, { "epoch": 1.957928109335504, "grad_norm": 3.6558425426483154, "learning_rate": 6.670234270614832e-05, "loss": 0.3732417583465576, "memory(GiB)": 70.5, "step": 45700, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.451651 }, { "epoch": 1.958142324664753, "grad_norm": 3.6292524337768555, "learning_rate": 6.66959993637511e-05, "loss": 0.21321563720703124, "memory(GiB)": 70.5, "step": 45705, "token_acc": 0.9548872180451128, "train_speed(iter/s)": 1.451672 }, { "epoch": 1.958356539994002, "grad_norm": 1.8277637958526611, "learning_rate": 6.668965571889021e-05, "loss": 0.22272706031799316, "memory(GiB)": 70.5, "step": 45710, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.451666 }, { "epoch": 1.958570755323251, "grad_norm": 5.84249210357666, "learning_rate": 6.668331177168055e-05, "loss": 0.5070577144622803, "memory(GiB)": 70.5, "step": 45715, "token_acc": 0.8851851851851852, "train_speed(iter/s)": 1.451667 }, { "epoch": 1.9587849706524998, "grad_norm": 3.3885343074798584, "learning_rate": 6.667696752223703e-05, "loss": 0.28270366191864016, "memory(GiB)": 70.5, "step": 45720, "token_acc": 0.9397590361445783, "train_speed(iter/s)": 1.451673 }, { "epoch": 1.9589991859817488, "grad_norm": 3.228820562362671, "learning_rate": 6.667062297067462e-05, "loss": 0.46433076858520506, "memory(GiB)": 70.5, "step": 45725, "token_acc": 0.9009584664536742, "train_speed(iter/s)": 1.451685 }, { "epoch": 1.9592134013109979, "grad_norm": 0.08638256043195724, "learning_rate": 6.666427811710824e-05, "loss": 0.6300058841705323, "memory(GiB)": 70.5, "step": 45730, "token_acc": 0.8833333333333333, "train_speed(iter/s)": 1.451707 }, { "epoch": 1.9594276166402467, "grad_norm": 5.289611339569092, "learning_rate": 6.66579329616528e-05, "loss": 0.5301832675933837, "memory(GiB)": 70.5, "step": 45735, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.451738 }, { "epoch": 1.9596418319694957, "grad_norm": 7.877406597137451, "learning_rate": 6.66515875044233e-05, "loss": 0.36732707023620603, "memory(GiB)": 70.5, "step": 45740, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.451741 }, { "epoch": 1.9598560472987447, "grad_norm": 1.2845335006713867, "learning_rate": 6.664524174553467e-05, "loss": 0.37143356800079347, "memory(GiB)": 70.5, "step": 45745, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.451754 }, { "epoch": 1.9600702626279936, "grad_norm": 2.0356650352478027, "learning_rate": 6.663889568510186e-05, "loss": 0.2738390684127808, "memory(GiB)": 70.5, "step": 45750, "token_acc": 0.9386973180076629, "train_speed(iter/s)": 1.45179 }, { "epoch": 1.9602844779572426, "grad_norm": 0.40649619698524475, "learning_rate": 6.663254932323986e-05, "loss": 0.5274063587188721, "memory(GiB)": 70.5, "step": 45755, "token_acc": 0.8807947019867549, "train_speed(iter/s)": 1.451806 }, { "epoch": 1.9604986932864916, "grad_norm": 3.4884157180786133, "learning_rate": 6.662620266006361e-05, "loss": 0.43277945518493655, "memory(GiB)": 70.5, "step": 45760, "token_acc": 0.9129129129129129, "train_speed(iter/s)": 1.451815 }, { "epoch": 1.9607129086157404, "grad_norm": 1.4214472770690918, "learning_rate": 6.661985569568811e-05, "loss": 0.34862728118896485, "memory(GiB)": 70.5, "step": 45765, "token_acc": 0.938375350140056, "train_speed(iter/s)": 1.451812 }, { "epoch": 1.9609271239449895, "grad_norm": 3.514798641204834, "learning_rate": 6.661350843022833e-05, "loss": 0.5301523208618164, "memory(GiB)": 70.5, "step": 45770, "token_acc": 0.8862068965517241, "train_speed(iter/s)": 1.451844 }, { "epoch": 1.9611413392742385, "grad_norm": 1.8152719736099243, "learning_rate": 6.660716086379927e-05, "loss": 0.24348206520080568, "memory(GiB)": 70.5, "step": 45775, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.451833 }, { "epoch": 1.9613555546034873, "grad_norm": 4.378045558929443, "learning_rate": 6.660081299651591e-05, "loss": 0.5558921337127686, "memory(GiB)": 70.5, "step": 45780, "token_acc": 0.8761904761904762, "train_speed(iter/s)": 1.45185 }, { "epoch": 1.9615697699327364, "grad_norm": 2.366605520248413, "learning_rate": 6.659446482849325e-05, "loss": 0.3117208003997803, "memory(GiB)": 70.5, "step": 45785, "token_acc": 0.9375, "train_speed(iter/s)": 1.451838 }, { "epoch": 1.9617839852619854, "grad_norm": 4.773201942443848, "learning_rate": 6.658811635984629e-05, "loss": 0.3832448720932007, "memory(GiB)": 70.5, "step": 45790, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.451835 }, { "epoch": 1.9619982005912342, "grad_norm": 3.6083974838256836, "learning_rate": 6.658176759069004e-05, "loss": 0.442883825302124, "memory(GiB)": 70.5, "step": 45795, "token_acc": 0.8933823529411765, "train_speed(iter/s)": 1.451844 }, { "epoch": 1.9622124159204832, "grad_norm": 4.979638576507568, "learning_rate": 6.657541852113953e-05, "loss": 0.3515910625457764, "memory(GiB)": 70.5, "step": 45800, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.451836 }, { "epoch": 1.9624266312497323, "grad_norm": 6.419509410858154, "learning_rate": 6.656906915130975e-05, "loss": 0.40229454040527346, "memory(GiB)": 70.5, "step": 45805, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.451852 }, { "epoch": 1.962640846578981, "grad_norm": 0.2233155518770218, "learning_rate": 6.656271948131573e-05, "loss": 0.30745174884796145, "memory(GiB)": 70.5, "step": 45810, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.451857 }, { "epoch": 1.9628550619082301, "grad_norm": 2.3602404594421387, "learning_rate": 6.655636951127254e-05, "loss": 0.28049089908599856, "memory(GiB)": 70.5, "step": 45815, "token_acc": 0.9343629343629344, "train_speed(iter/s)": 1.451859 }, { "epoch": 1.9630692772374791, "grad_norm": 1.0910837650299072, "learning_rate": 6.655001924129515e-05, "loss": 0.3440136671066284, "memory(GiB)": 70.5, "step": 45820, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.45186 }, { "epoch": 1.963283492566728, "grad_norm": 3.142814874649048, "learning_rate": 6.654366867149865e-05, "loss": 0.43267502784729006, "memory(GiB)": 70.5, "step": 45825, "token_acc": 0.9009584664536742, "train_speed(iter/s)": 1.451867 }, { "epoch": 1.963497707895977, "grad_norm": 4.200631141662598, "learning_rate": 6.653731780199807e-05, "loss": 0.4408390045166016, "memory(GiB)": 70.5, "step": 45830, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.451874 }, { "epoch": 1.963711923225226, "grad_norm": 1.6949756145477295, "learning_rate": 6.653096663290847e-05, "loss": 0.2722727537155151, "memory(GiB)": 70.5, "step": 45835, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.451872 }, { "epoch": 1.9639261385544748, "grad_norm": 3.9358861446380615, "learning_rate": 6.652461516434492e-05, "loss": 0.522749948501587, "memory(GiB)": 70.5, "step": 45840, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.451874 }, { "epoch": 1.9641403538837239, "grad_norm": 2.494837522506714, "learning_rate": 6.651826339642243e-05, "loss": 0.5103562355041504, "memory(GiB)": 70.5, "step": 45845, "token_acc": 0.8795620437956204, "train_speed(iter/s)": 1.451874 }, { "epoch": 1.964354569212973, "grad_norm": 4.192636013031006, "learning_rate": 6.651191132925612e-05, "loss": 0.19972175359725952, "memory(GiB)": 70.5, "step": 45850, "token_acc": 0.9631336405529954, "train_speed(iter/s)": 1.451885 }, { "epoch": 1.9645687845422217, "grad_norm": 1.2645243406295776, "learning_rate": 6.650555896296104e-05, "loss": 0.5117164134979248, "memory(GiB)": 70.5, "step": 45855, "token_acc": 0.8900343642611683, "train_speed(iter/s)": 1.451906 }, { "epoch": 1.9647829998714708, "grad_norm": 2.548030138015747, "learning_rate": 6.649920629765229e-05, "loss": 0.22238707542419434, "memory(GiB)": 70.5, "step": 45860, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.451912 }, { "epoch": 1.9649972152007198, "grad_norm": 3.9133105278015137, "learning_rate": 6.649285333344492e-05, "loss": 0.5177770614624023, "memory(GiB)": 70.5, "step": 45865, "token_acc": 0.8991935483870968, "train_speed(iter/s)": 1.451913 }, { "epoch": 1.9652114305299686, "grad_norm": 4.2154059410095215, "learning_rate": 6.648650007045407e-05, "loss": 0.43278074264526367, "memory(GiB)": 70.5, "step": 45870, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.451917 }, { "epoch": 1.9654256458592176, "grad_norm": 4.186916828155518, "learning_rate": 6.648014650879479e-05, "loss": 0.39331774711608886, "memory(GiB)": 70.5, "step": 45875, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.451938 }, { "epoch": 1.9656398611884667, "grad_norm": 5.34207820892334, "learning_rate": 6.64737926485822e-05, "loss": 0.4884763717651367, "memory(GiB)": 70.5, "step": 45880, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.451947 }, { "epoch": 1.9658540765177155, "grad_norm": 5.090673446655273, "learning_rate": 6.646743848993139e-05, "loss": 0.3983408689498901, "memory(GiB)": 70.5, "step": 45885, "token_acc": 0.9140127388535032, "train_speed(iter/s)": 1.451948 }, { "epoch": 1.9660682918469645, "grad_norm": 0.09216241538524628, "learning_rate": 6.64610840329575e-05, "loss": 0.17084946632385253, "memory(GiB)": 70.5, "step": 45890, "token_acc": 0.956081081081081, "train_speed(iter/s)": 1.451967 }, { "epoch": 1.9662825071762136, "grad_norm": 6.008804798126221, "learning_rate": 6.64547292777756e-05, "loss": 0.2209157705307007, "memory(GiB)": 70.5, "step": 45895, "token_acc": 0.9497716894977168, "train_speed(iter/s)": 1.451964 }, { "epoch": 1.9664967225054624, "grad_norm": 4.143402099609375, "learning_rate": 6.644837422450087e-05, "loss": 0.2665811061859131, "memory(GiB)": 70.5, "step": 45900, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.451968 }, { "epoch": 1.9667109378347114, "grad_norm": 3.266660451889038, "learning_rate": 6.644201887324841e-05, "loss": 0.6753283977508545, "memory(GiB)": 70.5, "step": 45905, "token_acc": 0.8566775244299675, "train_speed(iter/s)": 1.451963 }, { "epoch": 1.9669251531639604, "grad_norm": 2.3650951385498047, "learning_rate": 6.643566322413336e-05, "loss": 0.44538125991821287, "memory(GiB)": 70.5, "step": 45910, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.451964 }, { "epoch": 1.9671393684932093, "grad_norm": 4.3517045974731445, "learning_rate": 6.642930727727085e-05, "loss": 0.34755916595458985, "memory(GiB)": 70.5, "step": 45915, "token_acc": 0.9183006535947712, "train_speed(iter/s)": 1.451967 }, { "epoch": 1.9673535838224583, "grad_norm": 7.659725189208984, "learning_rate": 6.642295103277602e-05, "loss": 0.4132052421569824, "memory(GiB)": 70.5, "step": 45920, "token_acc": 0.906158357771261, "train_speed(iter/s)": 1.451964 }, { "epoch": 1.9675677991517073, "grad_norm": 4.894577980041504, "learning_rate": 6.641659449076401e-05, "loss": 0.5867265224456787, "memory(GiB)": 70.5, "step": 45925, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.451983 }, { "epoch": 1.9677820144809561, "grad_norm": 3.6215994358062744, "learning_rate": 6.641023765135001e-05, "loss": 0.5232521533966065, "memory(GiB)": 70.5, "step": 45930, "token_acc": 0.890728476821192, "train_speed(iter/s)": 1.451995 }, { "epoch": 1.9679962298102052, "grad_norm": 3.8248064517974854, "learning_rate": 6.640388051464917e-05, "loss": 0.41985383033752444, "memory(GiB)": 70.5, "step": 45935, "token_acc": 0.9161490683229814, "train_speed(iter/s)": 1.452003 }, { "epoch": 1.9682104451394542, "grad_norm": 0.36894097924232483, "learning_rate": 6.639752308077662e-05, "loss": 0.3084916353225708, "memory(GiB)": 70.5, "step": 45940, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.452009 }, { "epoch": 1.968424660468703, "grad_norm": 2.847372531890869, "learning_rate": 6.639116534984758e-05, "loss": 0.47753682136535647, "memory(GiB)": 70.5, "step": 45945, "token_acc": 0.8695652173913043, "train_speed(iter/s)": 1.452024 }, { "epoch": 1.968638875797952, "grad_norm": 2.8408849239349365, "learning_rate": 6.63848073219772e-05, "loss": 0.34610486030578613, "memory(GiB)": 70.5, "step": 45950, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.452025 }, { "epoch": 1.968853091127201, "grad_norm": 5.100044250488281, "learning_rate": 6.637844899728065e-05, "loss": 0.6507465362548828, "memory(GiB)": 70.5, "step": 45955, "token_acc": 0.8466666666666667, "train_speed(iter/s)": 1.452033 }, { "epoch": 1.96906730645645, "grad_norm": 0.4308326840400696, "learning_rate": 6.637209037587315e-05, "loss": 0.5233535766601562, "memory(GiB)": 70.5, "step": 45960, "token_acc": 0.8964285714285715, "train_speed(iter/s)": 1.452032 }, { "epoch": 1.969281521785699, "grad_norm": 2.483266830444336, "learning_rate": 6.636573145786987e-05, "loss": 0.4350834846496582, "memory(GiB)": 70.5, "step": 45965, "token_acc": 0.8954248366013072, "train_speed(iter/s)": 1.452029 }, { "epoch": 1.969495737114948, "grad_norm": 2.5400688648223877, "learning_rate": 6.635937224338601e-05, "loss": 0.6838459014892578, "memory(GiB)": 70.5, "step": 45970, "token_acc": 0.8552188552188552, "train_speed(iter/s)": 1.452031 }, { "epoch": 1.9697099524441968, "grad_norm": 4.282803535461426, "learning_rate": 6.635301273253678e-05, "loss": 0.7741165161132812, "memory(GiB)": 70.5, "step": 45975, "token_acc": 0.8535825545171339, "train_speed(iter/s)": 1.452048 }, { "epoch": 1.969924167773446, "grad_norm": 1.3523943424224854, "learning_rate": 6.634665292543737e-05, "loss": 0.3252854585647583, "memory(GiB)": 70.5, "step": 45980, "token_acc": 0.933993399339934, "train_speed(iter/s)": 1.452053 }, { "epoch": 1.9701383831026948, "grad_norm": 4.03738260269165, "learning_rate": 6.634029282220303e-05, "loss": 0.31678147315979005, "memory(GiB)": 70.5, "step": 45985, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.452066 }, { "epoch": 1.9703525984319437, "grad_norm": 7.75169563293457, "learning_rate": 6.633393242294894e-05, "loss": 0.6460024356842041, "memory(GiB)": 70.5, "step": 45990, "token_acc": 0.8613861386138614, "train_speed(iter/s)": 1.4521 }, { "epoch": 1.970566813761193, "grad_norm": 6.479301452636719, "learning_rate": 6.632757172779034e-05, "loss": 0.43602991104125977, "memory(GiB)": 70.5, "step": 45995, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.452098 }, { "epoch": 1.9707810290904417, "grad_norm": 0.5155187249183655, "learning_rate": 6.632121073684246e-05, "loss": 0.407031774520874, "memory(GiB)": 70.5, "step": 46000, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.452115 }, { "epoch": 1.9707810290904417, "eval_loss": 2.3190441131591797, "eval_runtime": 14.2497, "eval_samples_per_second": 7.018, "eval_steps_per_second": 7.018, "eval_token_acc": 0.4498186215235792, "step": 46000 }, { "epoch": 1.9709952444196905, "grad_norm": 4.2860026359558105, "learning_rate": 6.631484945022055e-05, "loss": 0.40054750442504883, "memory(GiB)": 70.5, "step": 46005, "token_acc": 0.56353591160221, "train_speed(iter/s)": 1.451415 }, { "epoch": 1.9712094597489398, "grad_norm": 2.994234085083008, "learning_rate": 6.630848786803983e-05, "loss": 0.5646192550659179, "memory(GiB)": 70.5, "step": 46010, "token_acc": 0.8681672025723473, "train_speed(iter/s)": 1.451411 }, { "epoch": 1.9714236750781886, "grad_norm": 4.021666049957275, "learning_rate": 6.630212599041556e-05, "loss": 0.5239994525909424, "memory(GiB)": 70.5, "step": 46015, "token_acc": 0.8945578231292517, "train_speed(iter/s)": 1.451427 }, { "epoch": 1.9716378904074374, "grad_norm": 4.510948657989502, "learning_rate": 6.629576381746299e-05, "loss": 0.6414666652679444, "memory(GiB)": 70.5, "step": 46020, "token_acc": 0.8680981595092024, "train_speed(iter/s)": 1.451428 }, { "epoch": 1.9718521057366867, "grad_norm": 3.978959083557129, "learning_rate": 6.628940134929736e-05, "loss": 0.41211299896240233, "memory(GiB)": 70.5, "step": 46025, "token_acc": 0.9, "train_speed(iter/s)": 1.451427 }, { "epoch": 1.9720663210659355, "grad_norm": 0.5697994828224182, "learning_rate": 6.628303858603395e-05, "loss": 0.5182897567749023, "memory(GiB)": 70.5, "step": 46030, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.451426 }, { "epoch": 1.9722805363951843, "grad_norm": 2.9070394039154053, "learning_rate": 6.627667552778802e-05, "loss": 0.3879220962524414, "memory(GiB)": 70.5, "step": 46035, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.451435 }, { "epoch": 1.9724947517244336, "grad_norm": 2.103632926940918, "learning_rate": 6.627031217467484e-05, "loss": 0.5640363693237305, "memory(GiB)": 70.5, "step": 46040, "token_acc": 0.8791946308724832, "train_speed(iter/s)": 1.451459 }, { "epoch": 1.9727089670536824, "grad_norm": 3.59536075592041, "learning_rate": 6.626394852680968e-05, "loss": 0.3246379137039185, "memory(GiB)": 70.5, "step": 46045, "token_acc": 0.916030534351145, "train_speed(iter/s)": 1.451464 }, { "epoch": 1.9729231823829312, "grad_norm": 3.1143598556518555, "learning_rate": 6.625758458430785e-05, "loss": 0.5442780494689942, "memory(GiB)": 70.5, "step": 46050, "token_acc": 0.8802816901408451, "train_speed(iter/s)": 1.451476 }, { "epoch": 1.9731373977121804, "grad_norm": 2.9375319480895996, "learning_rate": 6.625122034728463e-05, "loss": 0.7214717388153076, "memory(GiB)": 70.5, "step": 46055, "token_acc": 0.8482142857142857, "train_speed(iter/s)": 1.451487 }, { "epoch": 1.9733516130414293, "grad_norm": 1.588377594947815, "learning_rate": 6.624485581585531e-05, "loss": 0.29381797313690183, "memory(GiB)": 70.5, "step": 46060, "token_acc": 0.9171428571428571, "train_speed(iter/s)": 1.451478 }, { "epoch": 1.973565828370678, "grad_norm": 3.992398977279663, "learning_rate": 6.623849099013517e-05, "loss": 0.7107771873474121, "memory(GiB)": 70.5, "step": 46065, "token_acc": 0.8557993730407524, "train_speed(iter/s)": 1.451475 }, { "epoch": 1.9737800436999273, "grad_norm": 3.390859842300415, "learning_rate": 6.623212587023955e-05, "loss": 0.4847549915313721, "memory(GiB)": 70.5, "step": 46070, "token_acc": 0.8843283582089553, "train_speed(iter/s)": 1.451487 }, { "epoch": 1.9739942590291761, "grad_norm": 6.128483772277832, "learning_rate": 6.622576045628374e-05, "loss": 0.2754200458526611, "memory(GiB)": 70.5, "step": 46075, "token_acc": 0.9471698113207547, "train_speed(iter/s)": 1.451488 }, { "epoch": 1.974208474358425, "grad_norm": 5.416232109069824, "learning_rate": 6.621939474838304e-05, "loss": 0.6260978698730468, "memory(GiB)": 70.5, "step": 46080, "token_acc": 0.8754325259515571, "train_speed(iter/s)": 1.451496 }, { "epoch": 1.9744226896876742, "grad_norm": 3.581355571746826, "learning_rate": 6.621302874665282e-05, "loss": 0.6167004585266114, "memory(GiB)": 70.5, "step": 46085, "token_acc": 0.86084142394822, "train_speed(iter/s)": 1.451499 }, { "epoch": 1.974636905016923, "grad_norm": 0.8004480004310608, "learning_rate": 6.620666245120836e-05, "loss": 0.16374752521514893, "memory(GiB)": 70.5, "step": 46090, "token_acc": 0.9693251533742331, "train_speed(iter/s)": 1.451505 }, { "epoch": 1.9748511203461718, "grad_norm": 2.779630184173584, "learning_rate": 6.620029586216499e-05, "loss": 0.34546589851379395, "memory(GiB)": 70.5, "step": 46095, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.451524 }, { "epoch": 1.975065335675421, "grad_norm": 4.7452712059021, "learning_rate": 6.61939289796381e-05, "loss": 0.4114570140838623, "memory(GiB)": 70.5, "step": 46100, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.451545 }, { "epoch": 1.97527955100467, "grad_norm": 4.521687030792236, "learning_rate": 6.618756180374295e-05, "loss": 0.43131537437438966, "memory(GiB)": 70.5, "step": 46105, "token_acc": 0.9177215189873418, "train_speed(iter/s)": 1.451561 }, { "epoch": 1.9754937663339187, "grad_norm": 1.508616328239441, "learning_rate": 6.618119433459496e-05, "loss": 0.10511970520019531, "memory(GiB)": 70.5, "step": 46110, "token_acc": 0.9764705882352941, "train_speed(iter/s)": 1.451568 }, { "epoch": 1.975707981663168, "grad_norm": 5.062739849090576, "learning_rate": 6.617482657230945e-05, "loss": 0.32379651069641113, "memory(GiB)": 70.5, "step": 46115, "token_acc": 0.9317507418397626, "train_speed(iter/s)": 1.451572 }, { "epoch": 1.9759221969924168, "grad_norm": 3.4052891731262207, "learning_rate": 6.61684585170018e-05, "loss": 0.28129193782806394, "memory(GiB)": 70.5, "step": 46120, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.451585 }, { "epoch": 1.9761364123216656, "grad_norm": 0.5443854331970215, "learning_rate": 6.616209016878733e-05, "loss": 0.5785437107086182, "memory(GiB)": 70.5, "step": 46125, "token_acc": 0.887719298245614, "train_speed(iter/s)": 1.451613 }, { "epoch": 1.9763506276509148, "grad_norm": 6.342444896697998, "learning_rate": 6.615572152778147e-05, "loss": 0.4468189239501953, "memory(GiB)": 70.5, "step": 46130, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.451611 }, { "epoch": 1.9765648429801637, "grad_norm": 1.0896060466766357, "learning_rate": 6.614935259409954e-05, "loss": 0.1676504373550415, "memory(GiB)": 70.5, "step": 46135, "token_acc": 0.9652777777777778, "train_speed(iter/s)": 1.451619 }, { "epoch": 1.9767790583094125, "grad_norm": 0.9435102343559265, "learning_rate": 6.614298336785692e-05, "loss": 0.17770780324935914, "memory(GiB)": 70.5, "step": 46140, "token_acc": 0.9601328903654485, "train_speed(iter/s)": 1.451618 }, { "epoch": 1.9769932736386617, "grad_norm": 1.802048921585083, "learning_rate": 6.613661384916902e-05, "loss": 0.4724776268005371, "memory(GiB)": 70.5, "step": 46145, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.451618 }, { "epoch": 1.9772074889679105, "grad_norm": 5.3741044998168945, "learning_rate": 6.613024403815125e-05, "loss": 0.54361891746521, "memory(GiB)": 70.5, "step": 46150, "token_acc": 0.8722741433021807, "train_speed(iter/s)": 1.451633 }, { "epoch": 1.9774217042971594, "grad_norm": 3.99656343460083, "learning_rate": 6.612387393491894e-05, "loss": 0.5505693912506103, "memory(GiB)": 70.5, "step": 46155, "token_acc": 0.8875, "train_speed(iter/s)": 1.451622 }, { "epoch": 1.9776359196264086, "grad_norm": 0.8283351063728333, "learning_rate": 6.611750353958757e-05, "loss": 0.3092385768890381, "memory(GiB)": 70.5, "step": 46160, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.451625 }, { "epoch": 1.9778501349556574, "grad_norm": 0.4761351943016052, "learning_rate": 6.611113285227247e-05, "loss": 0.17326265573501587, "memory(GiB)": 70.5, "step": 46165, "token_acc": 0.9667774086378738, "train_speed(iter/s)": 1.451624 }, { "epoch": 1.9780643502849062, "grad_norm": 2.8589274883270264, "learning_rate": 6.610476187308909e-05, "loss": 0.5593883037567139, "memory(GiB)": 70.5, "step": 46170, "token_acc": 0.877742946708464, "train_speed(iter/s)": 1.451628 }, { "epoch": 1.9782785656141555, "grad_norm": 1.06501042842865, "learning_rate": 6.609839060215287e-05, "loss": 0.41413192749023436, "memory(GiB)": 70.5, "step": 46175, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.451628 }, { "epoch": 1.9784927809434043, "grad_norm": 0.21492111682891846, "learning_rate": 6.609201903957916e-05, "loss": 0.12903913259506225, "memory(GiB)": 70.5, "step": 46180, "token_acc": 0.9632107023411371, "train_speed(iter/s)": 1.451628 }, { "epoch": 1.9787069962726531, "grad_norm": 2.301981210708618, "learning_rate": 6.608564718548344e-05, "loss": 0.3194176435470581, "memory(GiB)": 70.5, "step": 46185, "token_acc": 0.9430379746835443, "train_speed(iter/s)": 1.451624 }, { "epoch": 1.9789212116019024, "grad_norm": 2.567511796951294, "learning_rate": 6.607927503998116e-05, "loss": 0.5392478466033935, "memory(GiB)": 70.5, "step": 46190, "token_acc": 0.8692579505300353, "train_speed(iter/s)": 1.451615 }, { "epoch": 1.9791354269311512, "grad_norm": 3.1259682178497314, "learning_rate": 6.607290260318771e-05, "loss": 0.2878118991851807, "memory(GiB)": 70.5, "step": 46195, "token_acc": 0.924, "train_speed(iter/s)": 1.451615 }, { "epoch": 1.9793496422604, "grad_norm": 2.629014015197754, "learning_rate": 6.606652987521855e-05, "loss": 0.17795965671539307, "memory(GiB)": 70.5, "step": 46200, "token_acc": 0.96, "train_speed(iter/s)": 1.451614 }, { "epoch": 1.9795638575896493, "grad_norm": 1.0812666416168213, "learning_rate": 6.606015685618913e-05, "loss": 0.2625499963760376, "memory(GiB)": 70.5, "step": 46205, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.45161 }, { "epoch": 1.979778072918898, "grad_norm": 1.392549753189087, "learning_rate": 6.60537835462149e-05, "loss": 0.1989372730255127, "memory(GiB)": 70.5, "step": 46210, "token_acc": 0.9570552147239264, "train_speed(iter/s)": 1.45161 }, { "epoch": 1.9799922882481469, "grad_norm": 4.649178981781006, "learning_rate": 6.604740994541133e-05, "loss": 0.7156996726989746, "memory(GiB)": 70.5, "step": 46215, "token_acc": 0.8598484848484849, "train_speed(iter/s)": 1.451611 }, { "epoch": 1.9802065035773961, "grad_norm": 5.195955276489258, "learning_rate": 6.604103605389387e-05, "loss": 0.28867175579071047, "memory(GiB)": 70.5, "step": 46220, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.451625 }, { "epoch": 1.980420718906645, "grad_norm": 4.638171672821045, "learning_rate": 6.603466187177799e-05, "loss": 0.4302785396575928, "memory(GiB)": 70.5, "step": 46225, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.451636 }, { "epoch": 1.9806349342358938, "grad_norm": 1.6927862167358398, "learning_rate": 6.602828739917919e-05, "loss": 0.41229805946350095, "memory(GiB)": 70.5, "step": 46230, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.451636 }, { "epoch": 1.980849149565143, "grad_norm": 2.061087131500244, "learning_rate": 6.602191263621291e-05, "loss": 0.26194565296173095, "memory(GiB)": 70.5, "step": 46235, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.451641 }, { "epoch": 1.9810633648943918, "grad_norm": 4.025881767272949, "learning_rate": 6.601553758299465e-05, "loss": 0.33765835762023927, "memory(GiB)": 70.5, "step": 46240, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.451649 }, { "epoch": 1.9812775802236406, "grad_norm": 0.9659358263015747, "learning_rate": 6.60091622396399e-05, "loss": 0.16636757850646972, "memory(GiB)": 70.5, "step": 46245, "token_acc": 0.95578231292517, "train_speed(iter/s)": 1.451651 }, { "epoch": 1.98149179555289, "grad_norm": 5.939847469329834, "learning_rate": 6.600278660626418e-05, "loss": 0.3222966194152832, "memory(GiB)": 70.5, "step": 46250, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.451653 }, { "epoch": 1.9817060108821387, "grad_norm": 0.6866821050643921, "learning_rate": 6.599641068298295e-05, "loss": 0.08605865240097046, "memory(GiB)": 70.5, "step": 46255, "token_acc": 0.9690721649484536, "train_speed(iter/s)": 1.45166 }, { "epoch": 1.9819202262113875, "grad_norm": 0.5480974912643433, "learning_rate": 6.599003446991174e-05, "loss": 0.25409183502197263, "memory(GiB)": 70.5, "step": 46260, "token_acc": 0.9450980392156862, "train_speed(iter/s)": 1.451666 }, { "epoch": 1.9821344415406368, "grad_norm": 5.3887810707092285, "learning_rate": 6.598365796716606e-05, "loss": 0.24451639652252197, "memory(GiB)": 70.5, "step": 46265, "token_acc": 0.9518900343642611, "train_speed(iter/s)": 1.451673 }, { "epoch": 1.9823486568698856, "grad_norm": 3.0823020935058594, "learning_rate": 6.597728117486141e-05, "loss": 0.2866310358047485, "memory(GiB)": 70.5, "step": 46270, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.451676 }, { "epoch": 1.9825628721991344, "grad_norm": 3.1736109256744385, "learning_rate": 6.597090409311334e-05, "loss": 0.2864938020706177, "memory(GiB)": 70.5, "step": 46275, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.451678 }, { "epoch": 1.9827770875283837, "grad_norm": 3.6052908897399902, "learning_rate": 6.596452672203736e-05, "loss": 0.7528653621673584, "memory(GiB)": 70.5, "step": 46280, "token_acc": 0.8426966292134831, "train_speed(iter/s)": 1.451684 }, { "epoch": 1.9829913028576325, "grad_norm": 2.5330071449279785, "learning_rate": 6.595814906174898e-05, "loss": 0.46064534187316897, "memory(GiB)": 70.5, "step": 46285, "token_acc": 0.879746835443038, "train_speed(iter/s)": 1.45169 }, { "epoch": 1.9832055181868813, "grad_norm": 3.11877179145813, "learning_rate": 6.595177111236377e-05, "loss": 0.4346020698547363, "memory(GiB)": 70.5, "step": 46290, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.451689 }, { "epoch": 1.9834197335161305, "grad_norm": 2.7292861938476562, "learning_rate": 6.594539287399727e-05, "loss": 0.5171435832977295, "memory(GiB)": 70.5, "step": 46295, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.451713 }, { "epoch": 1.9836339488453794, "grad_norm": 4.2494730949401855, "learning_rate": 6.593901434676502e-05, "loss": 0.5712735652923584, "memory(GiB)": 70.5, "step": 46300, "token_acc": 0.8648648648648649, "train_speed(iter/s)": 1.451706 }, { "epoch": 1.9838481641746282, "grad_norm": 0.05672677606344223, "learning_rate": 6.593263553078257e-05, "loss": 0.46612725257873533, "memory(GiB)": 70.5, "step": 46305, "token_acc": 0.9093655589123867, "train_speed(iter/s)": 1.451705 }, { "epoch": 1.9840623795038774, "grad_norm": 2.643482208251953, "learning_rate": 6.592625642616546e-05, "loss": 0.2881277561187744, "memory(GiB)": 70.5, "step": 46310, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.451712 }, { "epoch": 1.9842765948331262, "grad_norm": 7.990362167358398, "learning_rate": 6.591987703302931e-05, "loss": 0.6628310203552246, "memory(GiB)": 70.5, "step": 46315, "token_acc": 0.8707482993197279, "train_speed(iter/s)": 1.451743 }, { "epoch": 1.9844908101623753, "grad_norm": 4.772984504699707, "learning_rate": 6.591349735148964e-05, "loss": 0.23194050788879395, "memory(GiB)": 70.5, "step": 46320, "token_acc": 0.9467680608365019, "train_speed(iter/s)": 1.45175 }, { "epoch": 1.9847050254916243, "grad_norm": 1.0622590780258179, "learning_rate": 6.590711738166204e-05, "loss": 0.2553361177444458, "memory(GiB)": 70.5, "step": 46325, "token_acc": 0.948905109489051, "train_speed(iter/s)": 1.451751 }, { "epoch": 1.9849192408208731, "grad_norm": 3.115152597427368, "learning_rate": 6.590073712366207e-05, "loss": 0.2695796966552734, "memory(GiB)": 70.5, "step": 46330, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.451748 }, { "epoch": 1.9851334561501222, "grad_norm": 10.258259773254395, "learning_rate": 6.589435657760535e-05, "loss": 0.8946830749511718, "memory(GiB)": 70.5, "step": 46335, "token_acc": 0.864, "train_speed(iter/s)": 1.451766 }, { "epoch": 1.9853476714793712, "grad_norm": 7.269148349761963, "learning_rate": 6.588797574360744e-05, "loss": 0.31412372589111326, "memory(GiB)": 70.5, "step": 46340, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.451769 }, { "epoch": 1.98556188680862, "grad_norm": 2.1293747425079346, "learning_rate": 6.588159462178396e-05, "loss": 0.6804367065429687, "memory(GiB)": 70.5, "step": 46345, "token_acc": 0.8545454545454545, "train_speed(iter/s)": 1.451784 }, { "epoch": 1.985776102137869, "grad_norm": 2.1867904663085938, "learning_rate": 6.587521321225048e-05, "loss": 0.3634594202041626, "memory(GiB)": 70.5, "step": 46350, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.451821 }, { "epoch": 1.985990317467118, "grad_norm": 4.912786960601807, "learning_rate": 6.586883151512263e-05, "loss": 0.2928699254989624, "memory(GiB)": 70.5, "step": 46355, "token_acc": 0.9368029739776952, "train_speed(iter/s)": 1.451822 }, { "epoch": 1.9862045327963669, "grad_norm": 2.980759620666504, "learning_rate": 6.586244953051602e-05, "loss": 0.3880244016647339, "memory(GiB)": 70.5, "step": 46360, "token_acc": 0.911504424778761, "train_speed(iter/s)": 1.451827 }, { "epoch": 1.986418748125616, "grad_norm": 1.2229403257369995, "learning_rate": 6.585606725854623e-05, "loss": 0.36127073764801027, "memory(GiB)": 70.5, "step": 46365, "token_acc": 0.9294478527607362, "train_speed(iter/s)": 1.451832 }, { "epoch": 1.986632963454865, "grad_norm": 4.038516998291016, "learning_rate": 6.58496846993289e-05, "loss": 0.5021529197692871, "memory(GiB)": 70.5, "step": 46370, "token_acc": 0.8988326848249028, "train_speed(iter/s)": 1.451825 }, { "epoch": 1.9868471787841138, "grad_norm": 0.40715882182121277, "learning_rate": 6.58433018529797e-05, "loss": 0.3317458152770996, "memory(GiB)": 70.5, "step": 46375, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.451819 }, { "epoch": 1.9870613941133628, "grad_norm": 3.855625629425049, "learning_rate": 6.583691871961422e-05, "loss": 0.11572537422180176, "memory(GiB)": 70.5, "step": 46380, "token_acc": 0.9610894941634242, "train_speed(iter/s)": 1.451816 }, { "epoch": 1.9872756094426118, "grad_norm": 3.9309799671173096, "learning_rate": 6.583053529934807e-05, "loss": 0.4514957904815674, "memory(GiB)": 70.5, "step": 46385, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.451834 }, { "epoch": 1.9874898247718606, "grad_norm": 3.4379236698150635, "learning_rate": 6.582415159229695e-05, "loss": 0.3111391305923462, "memory(GiB)": 70.5, "step": 46390, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.451843 }, { "epoch": 1.9877040401011097, "grad_norm": 2.197247266769409, "learning_rate": 6.581776759857647e-05, "loss": 0.29321904182434083, "memory(GiB)": 70.5, "step": 46395, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.45184 }, { "epoch": 1.9879182554303587, "grad_norm": 7.793574333190918, "learning_rate": 6.58113833183023e-05, "loss": 0.6435821533203125, "memory(GiB)": 70.5, "step": 46400, "token_acc": 0.8517350157728707, "train_speed(iter/s)": 1.451861 }, { "epoch": 1.9881324707596075, "grad_norm": 4.08653450012207, "learning_rate": 6.580499875159008e-05, "loss": 0.4262677192687988, "memory(GiB)": 70.5, "step": 46405, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.451865 }, { "epoch": 1.9883466860888566, "grad_norm": 6.814379692077637, "learning_rate": 6.579861389855548e-05, "loss": 0.6364480495452881, "memory(GiB)": 70.5, "step": 46410, "token_acc": 0.8729641693811075, "train_speed(iter/s)": 1.451857 }, { "epoch": 1.9885609014181056, "grad_norm": 3.622365951538086, "learning_rate": 6.579222875931417e-05, "loss": 0.49977788925170896, "memory(GiB)": 70.5, "step": 46415, "token_acc": 0.904320987654321, "train_speed(iter/s)": 1.451866 }, { "epoch": 1.9887751167473544, "grad_norm": 1.4143613576889038, "learning_rate": 6.578584333398184e-05, "loss": 0.20527472496032714, "memory(GiB)": 70.5, "step": 46420, "token_acc": 0.9689655172413794, "train_speed(iter/s)": 1.451866 }, { "epoch": 1.9889893320766034, "grad_norm": 2.6746113300323486, "learning_rate": 6.577945762267412e-05, "loss": 0.4283113956451416, "memory(GiB)": 70.5, "step": 46425, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.451869 }, { "epoch": 1.9892035474058525, "grad_norm": 4.437436580657959, "learning_rate": 6.577307162550675e-05, "loss": 0.2876727104187012, "memory(GiB)": 70.5, "step": 46430, "token_acc": 0.9413793103448276, "train_speed(iter/s)": 1.451869 }, { "epoch": 1.9894177627351013, "grad_norm": 6.86922025680542, "learning_rate": 6.576668534259536e-05, "loss": 0.6004110336303711, "memory(GiB)": 70.5, "step": 46435, "token_acc": 0.8746438746438746, "train_speed(iter/s)": 1.451873 }, { "epoch": 1.9896319780643503, "grad_norm": 2.0700736045837402, "learning_rate": 6.57602987740557e-05, "loss": 0.2621511697769165, "memory(GiB)": 70.5, "step": 46440, "token_acc": 0.9390243902439024, "train_speed(iter/s)": 1.451874 }, { "epoch": 1.9898461933935994, "grad_norm": 5.9558820724487305, "learning_rate": 6.575391192000342e-05, "loss": 0.5040513038635254, "memory(GiB)": 70.5, "step": 46445, "token_acc": 0.9042145593869731, "train_speed(iter/s)": 1.451874 }, { "epoch": 1.9900604087228482, "grad_norm": 3.372770071029663, "learning_rate": 6.574752478055428e-05, "loss": 0.2693318843841553, "memory(GiB)": 70.5, "step": 46450, "token_acc": 0.9330708661417323, "train_speed(iter/s)": 1.451867 }, { "epoch": 1.9902746240520972, "grad_norm": 3.1413025856018066, "learning_rate": 6.574113735582393e-05, "loss": 0.29860854148864746, "memory(GiB)": 70.5, "step": 46455, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.451865 }, { "epoch": 1.9904888393813462, "grad_norm": 1.5261836051940918, "learning_rate": 6.573474964592813e-05, "loss": 0.38650689125061033, "memory(GiB)": 70.5, "step": 46460, "token_acc": 0.928, "train_speed(iter/s)": 1.451871 }, { "epoch": 1.990703054710595, "grad_norm": 5.744527339935303, "learning_rate": 6.572836165098258e-05, "loss": 0.45672082901000977, "memory(GiB)": 70.5, "step": 46465, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.451871 }, { "epoch": 1.990917270039844, "grad_norm": 3.30717134475708, "learning_rate": 6.572197337110299e-05, "loss": 0.2688775062561035, "memory(GiB)": 70.5, "step": 46470, "token_acc": 0.9486486486486486, "train_speed(iter/s)": 1.451881 }, { "epoch": 1.9911314853690931, "grad_norm": 1.2170195579528809, "learning_rate": 6.57155848064051e-05, "loss": 0.09183332324028015, "memory(GiB)": 70.5, "step": 46475, "token_acc": 0.9683098591549296, "train_speed(iter/s)": 1.451883 }, { "epoch": 1.991345700698342, "grad_norm": 6.781591892242432, "learning_rate": 6.570919595700467e-05, "loss": 0.37515482902526853, "memory(GiB)": 70.5, "step": 46480, "token_acc": 0.8854961832061069, "train_speed(iter/s)": 1.45188 }, { "epoch": 1.991559916027591, "grad_norm": 5.168046474456787, "learning_rate": 6.570280682301739e-05, "loss": 0.23463726043701172, "memory(GiB)": 70.5, "step": 46485, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.451879 }, { "epoch": 1.99177413135684, "grad_norm": 4.19439172744751, "learning_rate": 6.569641740455905e-05, "loss": 0.3603007078170776, "memory(GiB)": 70.5, "step": 46490, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.451882 }, { "epoch": 1.9919883466860888, "grad_norm": 4.852924346923828, "learning_rate": 6.569002770174539e-05, "loss": 0.5684388637542724, "memory(GiB)": 70.5, "step": 46495, "token_acc": 0.8938775510204081, "train_speed(iter/s)": 1.451881 }, { "epoch": 1.9922025620153379, "grad_norm": 3.5851097106933594, "learning_rate": 6.568363771469214e-05, "loss": 0.15902050733566284, "memory(GiB)": 70.5, "step": 46500, "token_acc": 0.954983922829582, "train_speed(iter/s)": 1.451885 }, { "epoch": 1.9922025620153379, "eval_loss": 2.591038227081299, "eval_runtime": 13.5477, "eval_samples_per_second": 7.381, "eval_steps_per_second": 7.381, "eval_token_acc": 0.4092178770949721, "step": 46500 }, { "epoch": 1.9924167773445869, "grad_norm": 3.060081720352173, "learning_rate": 6.567724744351512e-05, "loss": 0.3378316402435303, "memory(GiB)": 70.5, "step": 46505, "token_acc": 0.5687984496124031, "train_speed(iter/s)": 1.451229 }, { "epoch": 1.9926309926738357, "grad_norm": 0.3416382670402527, "learning_rate": 6.567085688833001e-05, "loss": 0.19430657625198364, "memory(GiB)": 70.5, "step": 46510, "token_acc": 0.9622641509433962, "train_speed(iter/s)": 1.451231 }, { "epoch": 1.9928452080030847, "grad_norm": 2.197411060333252, "learning_rate": 6.566446604925264e-05, "loss": 0.43794808387756345, "memory(GiB)": 70.5, "step": 46515, "token_acc": 0.9183006535947712, "train_speed(iter/s)": 1.451241 }, { "epoch": 1.9930594233323338, "grad_norm": 3.9849393367767334, "learning_rate": 6.56580749263988e-05, "loss": 0.3072217464447021, "memory(GiB)": 70.5, "step": 46520, "token_acc": 0.93359375, "train_speed(iter/s)": 1.451234 }, { "epoch": 1.9932736386615826, "grad_norm": 3.477581024169922, "learning_rate": 6.565168351988422e-05, "loss": 0.4634682655334473, "memory(GiB)": 70.5, "step": 46525, "token_acc": 0.9053030303030303, "train_speed(iter/s)": 1.451234 }, { "epoch": 1.9934878539908316, "grad_norm": 3.3808748722076416, "learning_rate": 6.564529182982471e-05, "loss": 0.3516870975494385, "memory(GiB)": 70.5, "step": 46530, "token_acc": 0.9114391143911439, "train_speed(iter/s)": 1.451235 }, { "epoch": 1.9937020693200806, "grad_norm": 1.9025577306747437, "learning_rate": 6.563889985633608e-05, "loss": 0.42304286956787107, "memory(GiB)": 70.5, "step": 46535, "token_acc": 0.9322709163346613, "train_speed(iter/s)": 1.451248 }, { "epoch": 1.9939162846493295, "grad_norm": 2.8703014850616455, "learning_rate": 6.56325075995341e-05, "loss": 0.2702886343002319, "memory(GiB)": 70.5, "step": 46540, "token_acc": 0.9477124183006536, "train_speed(iter/s)": 1.451241 }, { "epoch": 1.9941304999785785, "grad_norm": 4.975704193115234, "learning_rate": 6.562611505953457e-05, "loss": 0.38912553787231446, "memory(GiB)": 70.5, "step": 46545, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.451259 }, { "epoch": 1.9943447153078275, "grad_norm": 5.906377792358398, "learning_rate": 6.561972223645331e-05, "loss": 0.4388264656066895, "memory(GiB)": 70.5, "step": 46550, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.451259 }, { "epoch": 1.9945589306370763, "grad_norm": 2.2320163249969482, "learning_rate": 6.561332913040613e-05, "loss": 0.43406524658203127, "memory(GiB)": 70.5, "step": 46555, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.451257 }, { "epoch": 1.9947731459663254, "grad_norm": 3.3208470344543457, "learning_rate": 6.560693574150885e-05, "loss": 0.41654057502746583, "memory(GiB)": 70.5, "step": 46560, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.451264 }, { "epoch": 1.9949873612955744, "grad_norm": 3.0157415866851807, "learning_rate": 6.560054206987729e-05, "loss": 0.5659527778625488, "memory(GiB)": 70.5, "step": 46565, "token_acc": 0.869281045751634, "train_speed(iter/s)": 1.451276 }, { "epoch": 1.9952015766248232, "grad_norm": 2.8832855224609375, "learning_rate": 6.559414811562728e-05, "loss": 0.3862821340560913, "memory(GiB)": 70.5, "step": 46570, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.451271 }, { "epoch": 1.9954157919540723, "grad_norm": 0.46937984228134155, "learning_rate": 6.558775387887463e-05, "loss": 0.35491487979888914, "memory(GiB)": 70.5, "step": 46575, "token_acc": 0.9385245901639344, "train_speed(iter/s)": 1.451277 }, { "epoch": 1.9956300072833213, "grad_norm": 3.634091854095459, "learning_rate": 6.558135935973521e-05, "loss": 0.6261826038360596, "memory(GiB)": 70.5, "step": 46580, "token_acc": 0.8739837398373984, "train_speed(iter/s)": 1.451285 }, { "epoch": 1.99584422261257, "grad_norm": 4.037825584411621, "learning_rate": 6.557496455832485e-05, "loss": 0.2511759281158447, "memory(GiB)": 70.5, "step": 46585, "token_acc": 0.9522058823529411, "train_speed(iter/s)": 1.45132 }, { "epoch": 1.9960584379418191, "grad_norm": 2.419969081878662, "learning_rate": 6.556856947475937e-05, "loss": 0.2882366180419922, "memory(GiB)": 70.5, "step": 46590, "token_acc": 0.934640522875817, "train_speed(iter/s)": 1.451332 }, { "epoch": 1.9962726532710682, "grad_norm": 3.360313892364502, "learning_rate": 6.55621741091547e-05, "loss": 0.5387227535247803, "memory(GiB)": 70.5, "step": 46595, "token_acc": 0.8753541076487252, "train_speed(iter/s)": 1.451323 }, { "epoch": 1.996486868600317, "grad_norm": 3.696532726287842, "learning_rate": 6.55557784616266e-05, "loss": 0.6834839344024658, "memory(GiB)": 70.5, "step": 46600, "token_acc": 0.8712574850299402, "train_speed(iter/s)": 1.451341 }, { "epoch": 1.996701083929566, "grad_norm": 4.829133987426758, "learning_rate": 6.554938253229102e-05, "loss": 0.3485468626022339, "memory(GiB)": 70.5, "step": 46605, "token_acc": 0.9233870967741935, "train_speed(iter/s)": 1.451338 }, { "epoch": 1.996915299258815, "grad_norm": 2.190255641937256, "learning_rate": 6.554298632126376e-05, "loss": 0.3254303216934204, "memory(GiB)": 70.5, "step": 46610, "token_acc": 0.9294117647058824, "train_speed(iter/s)": 1.451357 }, { "epoch": 1.9971295145880639, "grad_norm": 5.1883625984191895, "learning_rate": 6.553658982866073e-05, "loss": 0.5555108070373536, "memory(GiB)": 70.5, "step": 46615, "token_acc": 0.877742946708464, "train_speed(iter/s)": 1.451388 }, { "epoch": 1.997343729917313, "grad_norm": 2.1911263465881348, "learning_rate": 6.55301930545978e-05, "loss": 0.1602689027786255, "memory(GiB)": 70.5, "step": 46620, "token_acc": 0.9712460063897763, "train_speed(iter/s)": 1.451401 }, { "epoch": 1.997557945246562, "grad_norm": 0.7393396496772766, "learning_rate": 6.552379599919087e-05, "loss": 0.35018091201782225, "memory(GiB)": 70.5, "step": 46625, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.451411 }, { "epoch": 1.9977721605758108, "grad_norm": 4.5117034912109375, "learning_rate": 6.551739866255578e-05, "loss": 0.8790657043457031, "memory(GiB)": 70.5, "step": 46630, "token_acc": 0.840531561461794, "train_speed(iter/s)": 1.451428 }, { "epoch": 1.9979863759050598, "grad_norm": 3.8690078258514404, "learning_rate": 6.551100104480849e-05, "loss": 0.516366720199585, "memory(GiB)": 70.5, "step": 46635, "token_acc": 0.8963414634146342, "train_speed(iter/s)": 1.451433 }, { "epoch": 1.9982005912343088, "grad_norm": 7.445000648498535, "learning_rate": 6.550460314606486e-05, "loss": 0.6598622322082519, "memory(GiB)": 70.5, "step": 46640, "token_acc": 0.8372093023255814, "train_speed(iter/s)": 1.451445 }, { "epoch": 1.9984148065635576, "grad_norm": 12.345580101013184, "learning_rate": 6.549820496644078e-05, "loss": 0.38158297538757324, "memory(GiB)": 70.5, "step": 46645, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.451459 }, { "epoch": 1.9986290218928067, "grad_norm": 2.867950201034546, "learning_rate": 6.54918065060522e-05, "loss": 0.3283450841903687, "memory(GiB)": 70.5, "step": 46650, "token_acc": 0.9281437125748503, "train_speed(iter/s)": 1.451461 }, { "epoch": 1.9988432372220557, "grad_norm": 3.210010528564453, "learning_rate": 6.5485407765015e-05, "loss": 0.46695866584777834, "memory(GiB)": 70.5, "step": 46655, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.45148 }, { "epoch": 1.9990574525513045, "grad_norm": 2.534975051879883, "learning_rate": 6.54790087434451e-05, "loss": 0.53587965965271, "memory(GiB)": 70.5, "step": 46660, "token_acc": 0.8646864686468647, "train_speed(iter/s)": 1.451493 }, { "epoch": 1.9992716678805535, "grad_norm": 7.370107173919678, "learning_rate": 6.547260944145845e-05, "loss": 0.47156171798706054, "memory(GiB)": 70.5, "step": 46665, "token_acc": 0.9022556390977443, "train_speed(iter/s)": 1.451491 }, { "epoch": 1.9994858832098026, "grad_norm": 0.591545581817627, "learning_rate": 6.546620985917097e-05, "loss": 0.2581897497177124, "memory(GiB)": 70.5, "step": 46670, "token_acc": 0.9492063492063492, "train_speed(iter/s)": 1.451495 }, { "epoch": 1.9997000985390514, "grad_norm": 3.6360549926757812, "learning_rate": 6.545980999669859e-05, "loss": 0.2597395658493042, "memory(GiB)": 70.5, "step": 46675, "token_acc": 0.946031746031746, "train_speed(iter/s)": 1.451504 }, { "epoch": 1.9999143138683004, "grad_norm": 4.5583295822143555, "learning_rate": 6.545340985415726e-05, "loss": 0.4560511112213135, "memory(GiB)": 70.5, "step": 46680, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.451506 }, { "epoch": 2.0001285291975495, "grad_norm": 3.240093946456909, "learning_rate": 6.54470094316629e-05, "loss": 0.4470634460449219, "memory(GiB)": 70.5, "step": 46685, "token_acc": 0.9109311740890689, "train_speed(iter/s)": 1.451508 }, { "epoch": 2.0003427445267983, "grad_norm": 3.0506412982940674, "learning_rate": 6.544060872933146e-05, "loss": 0.6162612438201904, "memory(GiB)": 70.5, "step": 46690, "token_acc": 0.8666666666666667, "train_speed(iter/s)": 1.451508 }, { "epoch": 2.000556959856047, "grad_norm": 3.818157434463501, "learning_rate": 6.543420774727892e-05, "loss": 0.2588310956954956, "memory(GiB)": 70.5, "step": 46695, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.451505 }, { "epoch": 2.0007711751852963, "grad_norm": 1.159898042678833, "learning_rate": 6.542780648562124e-05, "loss": 0.4150959014892578, "memory(GiB)": 70.5, "step": 46700, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.451523 }, { "epoch": 2.000985390514545, "grad_norm": 2.0554709434509277, "learning_rate": 6.542140494447435e-05, "loss": 0.4314236640930176, "memory(GiB)": 70.5, "step": 46705, "token_acc": 0.9015384615384615, "train_speed(iter/s)": 1.451524 }, { "epoch": 2.001199605843794, "grad_norm": 4.868441104888916, "learning_rate": 6.541500312395427e-05, "loss": 0.5019315719604492, "memory(GiB)": 70.5, "step": 46710, "token_acc": 0.8849206349206349, "train_speed(iter/s)": 1.451526 }, { "epoch": 2.0014138211730432, "grad_norm": 1.3260124921798706, "learning_rate": 6.540860102417693e-05, "loss": 0.3678418159484863, "memory(GiB)": 70.5, "step": 46715, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.451546 }, { "epoch": 2.001628036502292, "grad_norm": 2.5912723541259766, "learning_rate": 6.540219864525835e-05, "loss": 0.2571068525314331, "memory(GiB)": 70.5, "step": 46720, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.451553 }, { "epoch": 2.001842251831541, "grad_norm": 3.7058019638061523, "learning_rate": 6.539579598731447e-05, "loss": 0.4244704723358154, "memory(GiB)": 70.5, "step": 46725, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.451554 }, { "epoch": 2.00205646716079, "grad_norm": 1.6122827529907227, "learning_rate": 6.53893930504613e-05, "loss": 0.5078072547912598, "memory(GiB)": 70.5, "step": 46730, "token_acc": 0.8920634920634921, "train_speed(iter/s)": 1.451553 }, { "epoch": 2.002270682490039, "grad_norm": 3.525681972503662, "learning_rate": 6.538298983481484e-05, "loss": 0.38283300399780273, "memory(GiB)": 70.5, "step": 46735, "token_acc": 0.8932384341637011, "train_speed(iter/s)": 1.451558 }, { "epoch": 2.0024848978192877, "grad_norm": 2.042541027069092, "learning_rate": 6.53765863404911e-05, "loss": 0.09694207906723022, "memory(GiB)": 70.5, "step": 46740, "token_acc": 0.9781021897810219, "train_speed(iter/s)": 1.45156 }, { "epoch": 2.002699113148537, "grad_norm": 2.7354395389556885, "learning_rate": 6.537018256760606e-05, "loss": 0.5716984272003174, "memory(GiB)": 70.5, "step": 46745, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.451563 }, { "epoch": 2.002913328477786, "grad_norm": 5.11538553237915, "learning_rate": 6.536377851627577e-05, "loss": 0.3466149568557739, "memory(GiB)": 70.5, "step": 46750, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.451568 }, { "epoch": 2.0031275438070346, "grad_norm": 3.084584951400757, "learning_rate": 6.53573741866162e-05, "loss": 0.4294286727905273, "memory(GiB)": 70.5, "step": 46755, "token_acc": 0.904, "train_speed(iter/s)": 1.451578 }, { "epoch": 2.003341759136284, "grad_norm": 7.769847869873047, "learning_rate": 6.53509695787434e-05, "loss": 0.5673374652862548, "memory(GiB)": 70.5, "step": 46760, "token_acc": 0.9057971014492754, "train_speed(iter/s)": 1.451585 }, { "epoch": 2.0035559744655327, "grad_norm": 2.10099196434021, "learning_rate": 6.534456469277337e-05, "loss": 0.3236544609069824, "memory(GiB)": 70.5, "step": 46765, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.451593 }, { "epoch": 2.0037701897947815, "grad_norm": 2.0918750762939453, "learning_rate": 6.533815952882216e-05, "loss": 0.48494691848754884, "memory(GiB)": 70.5, "step": 46770, "token_acc": 0.8804780876494024, "train_speed(iter/s)": 1.45159 }, { "epoch": 2.0039844051240308, "grad_norm": 4.331406593322754, "learning_rate": 6.53317540870058e-05, "loss": 0.457018518447876, "memory(GiB)": 70.5, "step": 46775, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.451603 }, { "epoch": 2.0041986204532796, "grad_norm": 4.2265119552612305, "learning_rate": 6.532534836744035e-05, "loss": 0.3235656499862671, "memory(GiB)": 70.5, "step": 46780, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.451605 }, { "epoch": 2.0044128357825284, "grad_norm": 2.6114470958709717, "learning_rate": 6.531894237024183e-05, "loss": 0.269632363319397, "memory(GiB)": 70.5, "step": 46785, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.451607 }, { "epoch": 2.0046270511117776, "grad_norm": 1.399806022644043, "learning_rate": 6.53125360955263e-05, "loss": 0.3340468883514404, "memory(GiB)": 70.5, "step": 46790, "token_acc": 0.91015625, "train_speed(iter/s)": 1.451644 }, { "epoch": 2.0048412664410264, "grad_norm": 4.107738494873047, "learning_rate": 6.530612954340981e-05, "loss": 0.4134226322174072, "memory(GiB)": 70.5, "step": 46795, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.451647 }, { "epoch": 2.0050554817702753, "grad_norm": 3.4704036712646484, "learning_rate": 6.529972271400844e-05, "loss": 0.4368389129638672, "memory(GiB)": 70.5, "step": 46800, "token_acc": 0.9171597633136095, "train_speed(iter/s)": 1.451669 }, { "epoch": 2.0052696970995245, "grad_norm": 2.0818557739257812, "learning_rate": 6.529331560743821e-05, "loss": 0.6143061637878418, "memory(GiB)": 70.5, "step": 46805, "token_acc": 0.8680981595092024, "train_speed(iter/s)": 1.451667 }, { "epoch": 2.0054839124287733, "grad_norm": 3.828904151916504, "learning_rate": 6.528690822381523e-05, "loss": 0.44026761054992675, "memory(GiB)": 70.5, "step": 46810, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.45167 }, { "epoch": 2.0056981277580226, "grad_norm": 6.581126689910889, "learning_rate": 6.528050056325558e-05, "loss": 0.12308932542800903, "memory(GiB)": 70.5, "step": 46815, "token_acc": 0.9659574468085106, "train_speed(iter/s)": 1.451681 }, { "epoch": 2.0059123430872714, "grad_norm": 2.5736300945281982, "learning_rate": 6.527409262587533e-05, "loss": 0.3964546203613281, "memory(GiB)": 70.5, "step": 46820, "token_acc": 0.9057971014492754, "train_speed(iter/s)": 1.451675 }, { "epoch": 2.00612655841652, "grad_norm": 1.5319331884384155, "learning_rate": 6.526768441179056e-05, "loss": 0.19366047382354737, "memory(GiB)": 70.5, "step": 46825, "token_acc": 0.9540983606557377, "train_speed(iter/s)": 1.451675 }, { "epoch": 2.0063407737457695, "grad_norm": 0.05087066814303398, "learning_rate": 6.526127592111737e-05, "loss": 0.28177316188812257, "memory(GiB)": 70.5, "step": 46830, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.451687 }, { "epoch": 2.0065549890750183, "grad_norm": 3.793836832046509, "learning_rate": 6.525486715397183e-05, "loss": 0.25462732315063474, "memory(GiB)": 70.5, "step": 46835, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.451688 }, { "epoch": 2.006769204404267, "grad_norm": 3.1822140216827393, "learning_rate": 6.524845811047009e-05, "loss": 0.25841498374938965, "memory(GiB)": 70.5, "step": 46840, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.451706 }, { "epoch": 2.0069834197335163, "grad_norm": 3.1367056369781494, "learning_rate": 6.52420487907282e-05, "loss": 0.08503538370132446, "memory(GiB)": 70.5, "step": 46845, "token_acc": 0.9783549783549783, "train_speed(iter/s)": 1.451708 }, { "epoch": 2.007197635062765, "grad_norm": 2.994760751724243, "learning_rate": 6.52356391948623e-05, "loss": 0.2908243417739868, "memory(GiB)": 70.5, "step": 46850, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.451699 }, { "epoch": 2.007411850392014, "grad_norm": 6.714861869812012, "learning_rate": 6.522922932298852e-05, "loss": 0.31981720924377444, "memory(GiB)": 70.5, "step": 46855, "token_acc": 0.9375, "train_speed(iter/s)": 1.451713 }, { "epoch": 2.0076260657212632, "grad_norm": 1.6056045293807983, "learning_rate": 6.522281917522295e-05, "loss": 0.20829904079437256, "memory(GiB)": 70.5, "step": 46860, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.451729 }, { "epoch": 2.007840281050512, "grad_norm": 4.865965366363525, "learning_rate": 6.521640875168175e-05, "loss": 0.45349273681640623, "memory(GiB)": 70.5, "step": 46865, "token_acc": 0.9029126213592233, "train_speed(iter/s)": 1.451752 }, { "epoch": 2.008054496379761, "grad_norm": 2.039128303527832, "learning_rate": 6.520999805248102e-05, "loss": 0.1213606357574463, "memory(GiB)": 70.5, "step": 46870, "token_acc": 0.971875, "train_speed(iter/s)": 1.451757 }, { "epoch": 2.00826871170901, "grad_norm": 4.620748043060303, "learning_rate": 6.520358707773691e-05, "loss": 0.4446725845336914, "memory(GiB)": 70.5, "step": 46875, "token_acc": 0.928, "train_speed(iter/s)": 1.45175 }, { "epoch": 2.008482927038259, "grad_norm": 3.073341131210327, "learning_rate": 6.519717582756554e-05, "loss": 0.38340053558349607, "memory(GiB)": 70.5, "step": 46880, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.451741 }, { "epoch": 2.0086971423675077, "grad_norm": 3.2163636684417725, "learning_rate": 6.519076430208308e-05, "loss": 0.34734597206115725, "memory(GiB)": 70.5, "step": 46885, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.451742 }, { "epoch": 2.008911357696757, "grad_norm": 3.9597525596618652, "learning_rate": 6.518435250140569e-05, "loss": 0.51041579246521, "memory(GiB)": 70.5, "step": 46890, "token_acc": 0.9029850746268657, "train_speed(iter/s)": 1.451741 }, { "epoch": 2.009125573026006, "grad_norm": 5.848443031311035, "learning_rate": 6.517794042564951e-05, "loss": 0.5005092620849609, "memory(GiB)": 70.5, "step": 46895, "token_acc": 0.8975409836065574, "train_speed(iter/s)": 1.451744 }, { "epoch": 2.0093397883552546, "grad_norm": 1.3993006944656372, "learning_rate": 6.51715280749307e-05, "loss": 0.6501065731048584, "memory(GiB)": 70.5, "step": 46900, "token_acc": 0.8771331058020477, "train_speed(iter/s)": 1.451746 }, { "epoch": 2.009554003684504, "grad_norm": 1.484931230545044, "learning_rate": 6.516511544936542e-05, "loss": 0.3222174167633057, "memory(GiB)": 70.5, "step": 46905, "token_acc": 0.9336283185840708, "train_speed(iter/s)": 1.451755 }, { "epoch": 2.0097682190137527, "grad_norm": 6.683872699737549, "learning_rate": 6.515870254906984e-05, "loss": 0.24622349739074706, "memory(GiB)": 70.5, "step": 46910, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.451759 }, { "epoch": 2.0099824343430015, "grad_norm": 0.05769333243370056, "learning_rate": 6.515228937416015e-05, "loss": 0.31424148082733155, "memory(GiB)": 70.5, "step": 46915, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.451754 }, { "epoch": 2.0101966496722508, "grad_norm": 3.239182710647583, "learning_rate": 6.514587592475252e-05, "loss": 0.28428149223327637, "memory(GiB)": 70.5, "step": 46920, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.451756 }, { "epoch": 2.0104108650014996, "grad_norm": 1.0646154880523682, "learning_rate": 6.513946220096316e-05, "loss": 0.4060355186462402, "memory(GiB)": 70.5, "step": 46925, "token_acc": 0.9107692307692308, "train_speed(iter/s)": 1.451779 }, { "epoch": 2.0106250803307484, "grad_norm": 7.161166191101074, "learning_rate": 6.513304820290822e-05, "loss": 0.4389499664306641, "memory(GiB)": 70.5, "step": 46930, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.451778 }, { "epoch": 2.0108392956599976, "grad_norm": 2.0002121925354004, "learning_rate": 6.512663393070392e-05, "loss": 0.22414681911468506, "memory(GiB)": 70.5, "step": 46935, "token_acc": 0.9660377358490566, "train_speed(iter/s)": 1.451777 }, { "epoch": 2.0110535109892465, "grad_norm": 0.2743092477321625, "learning_rate": 6.512021938446646e-05, "loss": 0.21558027267456054, "memory(GiB)": 70.5, "step": 46940, "token_acc": 0.9588014981273408, "train_speed(iter/s)": 1.451784 }, { "epoch": 2.0112677263184953, "grad_norm": 5.213945388793945, "learning_rate": 6.511380456431204e-05, "loss": 0.37813830375671387, "memory(GiB)": 70.5, "step": 46945, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.451801 }, { "epoch": 2.0114819416477445, "grad_norm": 3.2043333053588867, "learning_rate": 6.510738947035687e-05, "loss": 0.5016732215881348, "memory(GiB)": 70.5, "step": 46950, "token_acc": 0.8938356164383562, "train_speed(iter/s)": 1.451801 }, { "epoch": 2.0116961569769933, "grad_norm": 2.6985416412353516, "learning_rate": 6.510097410271717e-05, "loss": 0.4051307201385498, "memory(GiB)": 70.5, "step": 46955, "token_acc": 0.9153225806451613, "train_speed(iter/s)": 1.451807 }, { "epoch": 2.011910372306242, "grad_norm": 2.6075687408447266, "learning_rate": 6.509455846150916e-05, "loss": 0.46650099754333496, "memory(GiB)": 70.5, "step": 46960, "token_acc": 0.9015873015873016, "train_speed(iter/s)": 1.451824 }, { "epoch": 2.0121245876354914, "grad_norm": 2.740140914916992, "learning_rate": 6.508814254684906e-05, "loss": 0.3052715301513672, "memory(GiB)": 70.5, "step": 46965, "token_acc": 0.9375, "train_speed(iter/s)": 1.451832 }, { "epoch": 2.01233880296474, "grad_norm": 3.195817232131958, "learning_rate": 6.50817263588531e-05, "loss": 0.4444750785827637, "memory(GiB)": 70.5, "step": 46970, "token_acc": 0.9152542372881356, "train_speed(iter/s)": 1.451843 }, { "epoch": 2.012553018293989, "grad_norm": 4.774703025817871, "learning_rate": 6.507530989763754e-05, "loss": 0.6621347427368164, "memory(GiB)": 70.5, "step": 46975, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.451848 }, { "epoch": 2.0127672336232383, "grad_norm": 2.9108760356903076, "learning_rate": 6.506889316331858e-05, "loss": 0.2530683040618896, "memory(GiB)": 70.5, "step": 46980, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.451855 }, { "epoch": 2.012981448952487, "grad_norm": 3.6877927780151367, "learning_rate": 6.50624761560125e-05, "loss": 0.5045418262481689, "memory(GiB)": 70.5, "step": 46985, "token_acc": 0.8851174934725848, "train_speed(iter/s)": 1.451855 }, { "epoch": 2.013195664281736, "grad_norm": 5.7749738693237305, "learning_rate": 6.505605887583552e-05, "loss": 0.7818368911743164, "memory(GiB)": 70.5, "step": 46990, "token_acc": 0.8314176245210728, "train_speed(iter/s)": 1.451847 }, { "epoch": 2.013409879610985, "grad_norm": 2.5822103023529053, "learning_rate": 6.50496413229039e-05, "loss": 0.1538211226463318, "memory(GiB)": 70.5, "step": 46995, "token_acc": 0.9645390070921985, "train_speed(iter/s)": 1.451843 }, { "epoch": 2.013624094940234, "grad_norm": 2.948317050933838, "learning_rate": 6.504322349733393e-05, "loss": 0.3478792905807495, "memory(GiB)": 70.5, "step": 47000, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.451839 }, { "epoch": 2.013624094940234, "eval_loss": 2.660264253616333, "eval_runtime": 13.5147, "eval_samples_per_second": 7.399, "eval_steps_per_second": 7.399, "eval_token_acc": 0.43333333333333335, "step": 47000 }, { "epoch": 2.013838310269483, "grad_norm": 3.648219585418701, "learning_rate": 6.503680539924184e-05, "loss": 0.36907596588134767, "memory(GiB)": 70.5, "step": 47005, "token_acc": 0.5542857142857143, "train_speed(iter/s)": 1.451186 }, { "epoch": 2.014052525598732, "grad_norm": 3.6475589275360107, "learning_rate": 6.503038702874394e-05, "loss": 0.29269819259643554, "memory(GiB)": 70.5, "step": 47010, "token_acc": 0.9295774647887324, "train_speed(iter/s)": 1.451194 }, { "epoch": 2.014266740927981, "grad_norm": 2.0628390312194824, "learning_rate": 6.502396838595646e-05, "loss": 0.32953386306762694, "memory(GiB)": 70.5, "step": 47015, "token_acc": 0.9157303370786517, "train_speed(iter/s)": 1.451205 }, { "epoch": 2.0144809562572297, "grad_norm": 1.982171893119812, "learning_rate": 6.50175494709957e-05, "loss": 0.2854544401168823, "memory(GiB)": 70.5, "step": 47020, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.451201 }, { "epoch": 2.014695171586479, "grad_norm": 4.183255672454834, "learning_rate": 6.501113028397793e-05, "loss": 0.4259798049926758, "memory(GiB)": 70.5, "step": 47025, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.451199 }, { "epoch": 2.0149093869157277, "grad_norm": 4.082297325134277, "learning_rate": 6.500471082501948e-05, "loss": 0.563326644897461, "memory(GiB)": 70.5, "step": 47030, "token_acc": 0.8682170542635659, "train_speed(iter/s)": 1.451195 }, { "epoch": 2.0151236022449766, "grad_norm": 4.62106466293335, "learning_rate": 6.49982910942366e-05, "loss": 0.29378671646118165, "memory(GiB)": 70.5, "step": 47035, "token_acc": 0.9482071713147411, "train_speed(iter/s)": 1.451207 }, { "epoch": 2.015337817574226, "grad_norm": 3.6044228076934814, "learning_rate": 6.499187109174561e-05, "loss": 0.49365978240966796, "memory(GiB)": 70.5, "step": 47040, "token_acc": 0.9, "train_speed(iter/s)": 1.451213 }, { "epoch": 2.0155520329034746, "grad_norm": 0.0737026259303093, "learning_rate": 6.498545081766282e-05, "loss": 0.45589275360107423, "memory(GiB)": 70.5, "step": 47045, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.451217 }, { "epoch": 2.0157662482327234, "grad_norm": 2.9485421180725098, "learning_rate": 6.497903027210453e-05, "loss": 0.3436159133911133, "memory(GiB)": 70.5, "step": 47050, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.45122 }, { "epoch": 2.0159804635619727, "grad_norm": 1.7489709854125977, "learning_rate": 6.497260945518706e-05, "loss": 0.2671010494232178, "memory(GiB)": 70.5, "step": 47055, "token_acc": 0.9419354838709677, "train_speed(iter/s)": 1.451217 }, { "epoch": 2.0161946788912215, "grad_norm": 7.313312530517578, "learning_rate": 6.496618836702672e-05, "loss": 0.35755114555358886, "memory(GiB)": 70.5, "step": 47060, "token_acc": 0.9152046783625731, "train_speed(iter/s)": 1.451227 }, { "epoch": 2.0164088942204703, "grad_norm": 2.3430256843566895, "learning_rate": 6.49597670077398e-05, "loss": 0.297791051864624, "memory(GiB)": 70.5, "step": 47065, "token_acc": 0.9397590361445783, "train_speed(iter/s)": 1.451227 }, { "epoch": 2.0166231095497196, "grad_norm": 2.1767916679382324, "learning_rate": 6.495334537744271e-05, "loss": 0.2897105932235718, "memory(GiB)": 70.5, "step": 47070, "token_acc": 0.9182879377431906, "train_speed(iter/s)": 1.451229 }, { "epoch": 2.0168373248789684, "grad_norm": 3.719686269760132, "learning_rate": 6.494692347625174e-05, "loss": 0.6581297397613526, "memory(GiB)": 70.5, "step": 47075, "token_acc": 0.8545454545454545, "train_speed(iter/s)": 1.451233 }, { "epoch": 2.017051540208217, "grad_norm": 2.6945018768310547, "learning_rate": 6.494050130428323e-05, "loss": 0.3969214916229248, "memory(GiB)": 70.5, "step": 47080, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.451235 }, { "epoch": 2.0172657555374665, "grad_norm": 6.776172161102295, "learning_rate": 6.493407886165351e-05, "loss": 0.5348648071289063, "memory(GiB)": 70.5, "step": 47085, "token_acc": 0.890728476821192, "train_speed(iter/s)": 1.451239 }, { "epoch": 2.0174799708667153, "grad_norm": 4.073083400726318, "learning_rate": 6.492765614847896e-05, "loss": 0.31310880184173584, "memory(GiB)": 70.5, "step": 47090, "token_acc": 0.9466192170818505, "train_speed(iter/s)": 1.451247 }, { "epoch": 2.017694186195964, "grad_norm": 5.291154384613037, "learning_rate": 6.492123316487589e-05, "loss": 0.6809689998626709, "memory(GiB)": 70.5, "step": 47095, "token_acc": 0.8458781362007168, "train_speed(iter/s)": 1.451258 }, { "epoch": 2.0179084015252133, "grad_norm": 3.4020233154296875, "learning_rate": 6.49148099109607e-05, "loss": 0.2678339719772339, "memory(GiB)": 70.5, "step": 47100, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.451251 }, { "epoch": 2.018122616854462, "grad_norm": 1.3989211320877075, "learning_rate": 6.490838638684973e-05, "loss": 0.25531485080718996, "memory(GiB)": 70.5, "step": 47105, "token_acc": 0.9517426273458445, "train_speed(iter/s)": 1.451258 }, { "epoch": 2.018336832183711, "grad_norm": 3.8669347763061523, "learning_rate": 6.490196259265934e-05, "loss": 0.3803633451461792, "memory(GiB)": 70.5, "step": 47110, "token_acc": 0.9014598540145985, "train_speed(iter/s)": 1.451268 }, { "epoch": 2.01855104751296, "grad_norm": 4.300317287445068, "learning_rate": 6.489553852850596e-05, "loss": 0.3023999214172363, "memory(GiB)": 70.5, "step": 47115, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.451266 }, { "epoch": 2.018765262842209, "grad_norm": 1.7151836156845093, "learning_rate": 6.488911419450591e-05, "loss": 0.27112998962402346, "memory(GiB)": 70.5, "step": 47120, "token_acc": 0.9438943894389439, "train_speed(iter/s)": 1.45127 }, { "epoch": 2.018979478171458, "grad_norm": 3.8606128692626953, "learning_rate": 6.488268959077557e-05, "loss": 0.22487196922302247, "memory(GiB)": 70.5, "step": 47125, "token_acc": 0.9560439560439561, "train_speed(iter/s)": 1.451272 }, { "epoch": 2.019193693500707, "grad_norm": 3.9919300079345703, "learning_rate": 6.487626471743137e-05, "loss": 0.3523031949996948, "memory(GiB)": 70.5, "step": 47130, "token_acc": 0.928, "train_speed(iter/s)": 1.451276 }, { "epoch": 2.019407908829956, "grad_norm": 2.5557572841644287, "learning_rate": 6.486983957458967e-05, "loss": 0.20478515625, "memory(GiB)": 70.5, "step": 47135, "token_acc": 0.9453125, "train_speed(iter/s)": 1.451278 }, { "epoch": 2.0196221241592047, "grad_norm": 2.7129404544830322, "learning_rate": 6.486341416236687e-05, "loss": 0.34339170455932616, "memory(GiB)": 70.5, "step": 47140, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.451294 }, { "epoch": 2.019836339488454, "grad_norm": 5.6365580558776855, "learning_rate": 6.485698848087939e-05, "loss": 0.24019575119018555, "memory(GiB)": 70.5, "step": 47145, "token_acc": 0.953125, "train_speed(iter/s)": 1.451329 }, { "epoch": 2.020050554817703, "grad_norm": 3.64971661567688, "learning_rate": 6.485056253024362e-05, "loss": 0.3721939563751221, "memory(GiB)": 70.5, "step": 47150, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.451349 }, { "epoch": 2.0202647701469516, "grad_norm": 3.442342519760132, "learning_rate": 6.484413631057599e-05, "loss": 0.39964842796325684, "memory(GiB)": 70.5, "step": 47155, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.451364 }, { "epoch": 2.020478985476201, "grad_norm": 5.23480224609375, "learning_rate": 6.48377098219929e-05, "loss": 0.22043323516845703, "memory(GiB)": 70.5, "step": 47160, "token_acc": 0.9593220338983051, "train_speed(iter/s)": 1.451365 }, { "epoch": 2.0206932008054497, "grad_norm": 5.893966197967529, "learning_rate": 6.483128306461078e-05, "loss": 0.3933832883834839, "memory(GiB)": 70.5, "step": 47165, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.451372 }, { "epoch": 2.0209074161346985, "grad_norm": 2.5774619579315186, "learning_rate": 6.482485603854604e-05, "loss": 0.6068005084991455, "memory(GiB)": 70.5, "step": 47170, "token_acc": 0.8713450292397661, "train_speed(iter/s)": 1.451373 }, { "epoch": 2.0211216314639477, "grad_norm": 0.3474971354007721, "learning_rate": 6.481842874391514e-05, "loss": 0.3118995428085327, "memory(GiB)": 70.5, "step": 47175, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.451378 }, { "epoch": 2.0213358467931966, "grad_norm": 4.948963642120361, "learning_rate": 6.481200118083449e-05, "loss": 0.608919906616211, "memory(GiB)": 70.5, "step": 47180, "token_acc": 0.8924731182795699, "train_speed(iter/s)": 1.451385 }, { "epoch": 2.0215500621224454, "grad_norm": 0.15075275301933289, "learning_rate": 6.480557334942055e-05, "loss": 0.3352478265762329, "memory(GiB)": 70.5, "step": 47185, "token_acc": 0.9307359307359307, "train_speed(iter/s)": 1.451394 }, { "epoch": 2.0217642774516946, "grad_norm": 8.14201831817627, "learning_rate": 6.479914524978978e-05, "loss": 0.21916062831878663, "memory(GiB)": 70.5, "step": 47190, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.451397 }, { "epoch": 2.0219784927809434, "grad_norm": 2.5920989513397217, "learning_rate": 6.479271688205858e-05, "loss": 0.24250483512878418, "memory(GiB)": 70.5, "step": 47195, "token_acc": 0.9414225941422594, "train_speed(iter/s)": 1.451394 }, { "epoch": 2.0221927081101922, "grad_norm": 3.8273568153381348, "learning_rate": 6.478628824634346e-05, "loss": 0.49476017951965334, "memory(GiB)": 70.5, "step": 47200, "token_acc": 0.9078498293515358, "train_speed(iter/s)": 1.451389 }, { "epoch": 2.0224069234394415, "grad_norm": 1.9923560619354248, "learning_rate": 6.477985934276085e-05, "loss": 0.3463109254837036, "memory(GiB)": 70.5, "step": 47205, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.451382 }, { "epoch": 2.0226211387686903, "grad_norm": 0.5725401639938354, "learning_rate": 6.477343017142722e-05, "loss": 0.5295327663421631, "memory(GiB)": 70.5, "step": 47210, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.451381 }, { "epoch": 2.022835354097939, "grad_norm": 9.239989280700684, "learning_rate": 6.476700073245904e-05, "loss": 0.2611178159713745, "memory(GiB)": 70.5, "step": 47215, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.451377 }, { "epoch": 2.0230495694271884, "grad_norm": 4.581867218017578, "learning_rate": 6.47605710259728e-05, "loss": 0.29077739715576173, "memory(GiB)": 70.5, "step": 47220, "token_acc": 0.949685534591195, "train_speed(iter/s)": 1.451398 }, { "epoch": 2.023263784756437, "grad_norm": 5.484643936157227, "learning_rate": 6.475414105208497e-05, "loss": 0.642219352722168, "memory(GiB)": 70.5, "step": 47225, "token_acc": 0.8695652173913043, "train_speed(iter/s)": 1.451416 }, { "epoch": 2.023478000085686, "grad_norm": 1.5547813177108765, "learning_rate": 6.474771081091202e-05, "loss": 0.2839707374572754, "memory(GiB)": 70.5, "step": 47230, "token_acc": 0.9420731707317073, "train_speed(iter/s)": 1.451419 }, { "epoch": 2.0236922154149353, "grad_norm": 1.7063993215560913, "learning_rate": 6.474128030257047e-05, "loss": 0.6027248859405517, "memory(GiB)": 70.5, "step": 47235, "token_acc": 0.87890625, "train_speed(iter/s)": 1.451417 }, { "epoch": 2.023906430744184, "grad_norm": 3.3029356002807617, "learning_rate": 6.473484952717679e-05, "loss": 0.3993695735931396, "memory(GiB)": 70.5, "step": 47240, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.451418 }, { "epoch": 2.024120646073433, "grad_norm": 2.537686586380005, "learning_rate": 6.472841848484749e-05, "loss": 0.3960877418518066, "memory(GiB)": 70.5, "step": 47245, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.451417 }, { "epoch": 2.024334861402682, "grad_norm": 4.0113677978515625, "learning_rate": 6.472198717569907e-05, "loss": 0.34189441204071047, "memory(GiB)": 70.5, "step": 47250, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.451431 }, { "epoch": 2.024549076731931, "grad_norm": 4.1642889976501465, "learning_rate": 6.471555559984804e-05, "loss": 0.47713651657104494, "memory(GiB)": 70.5, "step": 47255, "token_acc": 0.8993506493506493, "train_speed(iter/s)": 1.451441 }, { "epoch": 2.0247632920611798, "grad_norm": 1.9434128999710083, "learning_rate": 6.470912375741093e-05, "loss": 0.4197535037994385, "memory(GiB)": 70.5, "step": 47260, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.451435 }, { "epoch": 2.024977507390429, "grad_norm": 6.219625473022461, "learning_rate": 6.470269164850424e-05, "loss": 0.3833281993865967, "memory(GiB)": 70.5, "step": 47265, "token_acc": 0.9088145896656535, "train_speed(iter/s)": 1.451438 }, { "epoch": 2.025191722719678, "grad_norm": 0.9366504549980164, "learning_rate": 6.469625927324448e-05, "loss": 0.5643921375274659, "memory(GiB)": 70.5, "step": 47270, "token_acc": 0.8674242424242424, "train_speed(iter/s)": 1.451434 }, { "epoch": 2.0254059380489267, "grad_norm": 2.091085910797119, "learning_rate": 6.468982663174822e-05, "loss": 0.22055211067199706, "memory(GiB)": 70.5, "step": 47275, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.451445 }, { "epoch": 2.025620153378176, "grad_norm": 2.6424946784973145, "learning_rate": 6.468339372413196e-05, "loss": 0.48630075454711913, "memory(GiB)": 70.5, "step": 47280, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.451468 }, { "epoch": 2.0258343687074247, "grad_norm": 3.652005910873413, "learning_rate": 6.467696055051225e-05, "loss": 0.2636903762817383, "memory(GiB)": 70.5, "step": 47285, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.451494 }, { "epoch": 2.0260485840366735, "grad_norm": 5.785483360290527, "learning_rate": 6.467052711100563e-05, "loss": 0.6169801712036133, "memory(GiB)": 70.5, "step": 47290, "token_acc": 0.8741935483870967, "train_speed(iter/s)": 1.451494 }, { "epoch": 2.026262799365923, "grad_norm": 2.1608502864837646, "learning_rate": 6.466409340572864e-05, "loss": 0.21651439666748046, "memory(GiB)": 70.5, "step": 47295, "token_acc": 0.9571428571428572, "train_speed(iter/s)": 1.451499 }, { "epoch": 2.0264770146951716, "grad_norm": 4.104649066925049, "learning_rate": 6.465765943479785e-05, "loss": 0.26978049278259275, "memory(GiB)": 70.5, "step": 47300, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.451498 }, { "epoch": 2.0266912300244204, "grad_norm": 3.2338523864746094, "learning_rate": 6.465122519832982e-05, "loss": 0.38066601753234863, "memory(GiB)": 70.5, "step": 47305, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.451499 }, { "epoch": 2.0269054453536697, "grad_norm": 1.9488838911056519, "learning_rate": 6.464479069644109e-05, "loss": 0.2885688304901123, "memory(GiB)": 70.5, "step": 47310, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.4515 }, { "epoch": 2.0271196606829185, "grad_norm": 4.697782516479492, "learning_rate": 6.463835592924825e-05, "loss": 0.33109374046325685, "memory(GiB)": 70.5, "step": 47315, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.451527 }, { "epoch": 2.0273338760121673, "grad_norm": 5.814637660980225, "learning_rate": 6.463192089686786e-05, "loss": 0.5632325649261475, "memory(GiB)": 70.5, "step": 47320, "token_acc": 0.8790322580645161, "train_speed(iter/s)": 1.451564 }, { "epoch": 2.0275480913414166, "grad_norm": 3.232966184616089, "learning_rate": 6.462548559941648e-05, "loss": 0.28245418071746825, "memory(GiB)": 70.5, "step": 47325, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.451572 }, { "epoch": 2.0277623066706654, "grad_norm": 1.476583480834961, "learning_rate": 6.461905003701072e-05, "loss": 0.3217300891876221, "memory(GiB)": 70.5, "step": 47330, "token_acc": 0.9303135888501742, "train_speed(iter/s)": 1.451586 }, { "epoch": 2.027976521999914, "grad_norm": 4.740162372589111, "learning_rate": 6.461261420976716e-05, "loss": 0.3684617280960083, "memory(GiB)": 70.5, "step": 47335, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.451598 }, { "epoch": 2.0281907373291634, "grad_norm": 0.5445603132247925, "learning_rate": 6.460617811780238e-05, "loss": 0.20175354480743407, "memory(GiB)": 70.5, "step": 47340, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.451609 }, { "epoch": 2.0284049526584123, "grad_norm": 6.008301258087158, "learning_rate": 6.459974176123299e-05, "loss": 0.3594545364379883, "memory(GiB)": 70.5, "step": 47345, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.451611 }, { "epoch": 2.028619167987661, "grad_norm": 1.2475197315216064, "learning_rate": 6.459330514017556e-05, "loss": 0.18788652420043944, "memory(GiB)": 70.5, "step": 47350, "token_acc": 0.9553264604810997, "train_speed(iter/s)": 1.451608 }, { "epoch": 2.0288333833169103, "grad_norm": 1.2478456497192383, "learning_rate": 6.458686825474674e-05, "loss": 0.3042749404907227, "memory(GiB)": 70.5, "step": 47355, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.451603 }, { "epoch": 2.029047598646159, "grad_norm": 2.7008018493652344, "learning_rate": 6.45804311050631e-05, "loss": 0.4089754581451416, "memory(GiB)": 70.5, "step": 47360, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.451621 }, { "epoch": 2.029261813975408, "grad_norm": 3.1198880672454834, "learning_rate": 6.457399369124129e-05, "loss": 0.36994802951812744, "memory(GiB)": 70.5, "step": 47365, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.45162 }, { "epoch": 2.029476029304657, "grad_norm": 2.779202461242676, "learning_rate": 6.456755601339789e-05, "loss": 0.39109833240509034, "memory(GiB)": 70.5, "step": 47370, "token_acc": 0.91875, "train_speed(iter/s)": 1.451624 }, { "epoch": 2.029690244633906, "grad_norm": 3.0201807022094727, "learning_rate": 6.456111807164958e-05, "loss": 0.3546031951904297, "memory(GiB)": 70.5, "step": 47375, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.451644 }, { "epoch": 2.029904459963155, "grad_norm": 6.2603559494018555, "learning_rate": 6.455467986611293e-05, "loss": 0.47895116806030275, "memory(GiB)": 70.5, "step": 47380, "token_acc": 0.9042904290429042, "train_speed(iter/s)": 1.451658 }, { "epoch": 2.030118675292404, "grad_norm": 1.5443087816238403, "learning_rate": 6.45482413969046e-05, "loss": 0.30798289775848386, "memory(GiB)": 70.5, "step": 47385, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.451653 }, { "epoch": 2.030332890621653, "grad_norm": 3.7345314025878906, "learning_rate": 6.454180266414125e-05, "loss": 0.16410930156707765, "memory(GiB)": 70.5, "step": 47390, "token_acc": 0.9609375, "train_speed(iter/s)": 1.451644 }, { "epoch": 2.0305471059509017, "grad_norm": 4.7918267250061035, "learning_rate": 6.453536366793949e-05, "loss": 0.3651083469390869, "memory(GiB)": 70.5, "step": 47395, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.451645 }, { "epoch": 2.030761321280151, "grad_norm": 6.61681604385376, "learning_rate": 6.452892440841597e-05, "loss": 0.47645010948181155, "memory(GiB)": 70.5, "step": 47400, "token_acc": 0.9176954732510288, "train_speed(iter/s)": 1.451665 }, { "epoch": 2.0309755366094, "grad_norm": 2.5556671619415283, "learning_rate": 6.452248488568738e-05, "loss": 0.323940372467041, "memory(GiB)": 70.5, "step": 47405, "token_acc": 0.9269662921348315, "train_speed(iter/s)": 1.451666 }, { "epoch": 2.0311897519386486, "grad_norm": 3.91011118888855, "learning_rate": 6.451604509987033e-05, "loss": 0.48607726097106935, "memory(GiB)": 70.5, "step": 47410, "token_acc": 0.8859315589353612, "train_speed(iter/s)": 1.451681 }, { "epoch": 2.031403967267898, "grad_norm": 1.8855153322219849, "learning_rate": 6.450960505108153e-05, "loss": 0.3087873935699463, "memory(GiB)": 70.5, "step": 47415, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.451671 }, { "epoch": 2.0316181825971467, "grad_norm": 2.341684579849243, "learning_rate": 6.450316473943763e-05, "loss": 0.45423474311828616, "memory(GiB)": 70.5, "step": 47420, "token_acc": 0.8875, "train_speed(iter/s)": 1.451661 }, { "epoch": 2.0318323979263955, "grad_norm": 4.503137588500977, "learning_rate": 6.449672416505526e-05, "loss": 0.22753410339355468, "memory(GiB)": 70.5, "step": 47425, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.451661 }, { "epoch": 2.0320466132556447, "grad_norm": 4.663466453552246, "learning_rate": 6.449028332805117e-05, "loss": 0.3638306617736816, "memory(GiB)": 70.5, "step": 47430, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.451669 }, { "epoch": 2.0322608285848935, "grad_norm": 5.930561542510986, "learning_rate": 6.448384222854197e-05, "loss": 0.4660353660583496, "memory(GiB)": 70.5, "step": 47435, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.451685 }, { "epoch": 2.0324750439141424, "grad_norm": 5.706390380859375, "learning_rate": 6.447740086664439e-05, "loss": 0.3546177864074707, "memory(GiB)": 70.5, "step": 47440, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.451702 }, { "epoch": 2.0326892592433916, "grad_norm": 3.2041175365448, "learning_rate": 6.447095924247512e-05, "loss": 0.4561814308166504, "memory(GiB)": 70.5, "step": 47445, "token_acc": 0.8831615120274914, "train_speed(iter/s)": 1.451701 }, { "epoch": 2.0329034745726404, "grad_norm": 6.0806193351745605, "learning_rate": 6.446451735615085e-05, "loss": 0.5855185031890869, "memory(GiB)": 70.5, "step": 47450, "token_acc": 0.890295358649789, "train_speed(iter/s)": 1.451713 }, { "epoch": 2.0331176899018892, "grad_norm": 3.1727983951568604, "learning_rate": 6.445807520778828e-05, "loss": 0.13382539749145508, "memory(GiB)": 70.5, "step": 47455, "token_acc": 0.959409594095941, "train_speed(iter/s)": 1.451713 }, { "epoch": 2.0333319052311385, "grad_norm": 4.666578769683838, "learning_rate": 6.445163279750412e-05, "loss": 0.44828367233276367, "memory(GiB)": 70.5, "step": 47460, "token_acc": 0.8994082840236687, "train_speed(iter/s)": 1.45173 }, { "epoch": 2.0335461205603873, "grad_norm": 5.31848669052124, "learning_rate": 6.444519012541507e-05, "loss": 0.6392924308776855, "memory(GiB)": 70.5, "step": 47465, "token_acc": 0.8949044585987261, "train_speed(iter/s)": 1.451753 }, { "epoch": 2.033760335889636, "grad_norm": 1.1299998760223389, "learning_rate": 6.443874719163782e-05, "loss": 0.4373302936553955, "memory(GiB)": 70.5, "step": 47470, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.451766 }, { "epoch": 2.0339745512188854, "grad_norm": 2.7930057048797607, "learning_rate": 6.443230399628916e-05, "loss": 0.33986501693725585, "memory(GiB)": 70.5, "step": 47475, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.451766 }, { "epoch": 2.034188766548134, "grad_norm": 1.2220778465270996, "learning_rate": 6.442586053948576e-05, "loss": 0.3554290056228638, "memory(GiB)": 70.5, "step": 47480, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.451769 }, { "epoch": 2.034402981877383, "grad_norm": 3.522819757461548, "learning_rate": 6.441941682134436e-05, "loss": 0.22034878730773927, "memory(GiB)": 70.5, "step": 47485, "token_acc": 0.9509433962264151, "train_speed(iter/s)": 1.451775 }, { "epoch": 2.0346171972066323, "grad_norm": 2.796985387802124, "learning_rate": 6.441297284198172e-05, "loss": 0.3558933258056641, "memory(GiB)": 70.5, "step": 47490, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.451773 }, { "epoch": 2.034831412535881, "grad_norm": 1.743396282196045, "learning_rate": 6.440652860151452e-05, "loss": 0.31530165672302246, "memory(GiB)": 70.5, "step": 47495, "token_acc": 0.9083665338645418, "train_speed(iter/s)": 1.45177 }, { "epoch": 2.03504562786513, "grad_norm": 2.1670806407928467, "learning_rate": 6.440008410005956e-05, "loss": 0.2885331869125366, "memory(GiB)": 70.5, "step": 47500, "token_acc": 0.9461279461279462, "train_speed(iter/s)": 1.45178 }, { "epoch": 2.03504562786513, "eval_loss": 2.5181310176849365, "eval_runtime": 13.3194, "eval_samples_per_second": 7.508, "eval_steps_per_second": 7.508, "eval_token_acc": 0.42909987669543775, "step": 47500 }, { "epoch": 2.035259843194379, "grad_norm": 2.826847553253174, "learning_rate": 6.439363933773358e-05, "loss": 0.44466476440429686, "memory(GiB)": 70.5, "step": 47505, "token_acc": 0.5701830863121186, "train_speed(iter/s)": 1.451141 }, { "epoch": 2.035474058523628, "grad_norm": 2.210513114929199, "learning_rate": 6.43871943146533e-05, "loss": 0.26535494327545167, "memory(GiB)": 70.5, "step": 47510, "token_acc": 0.9401993355481728, "train_speed(iter/s)": 1.451139 }, { "epoch": 2.0356882738528768, "grad_norm": 2.209998846054077, "learning_rate": 6.43807490309355e-05, "loss": 0.16166919469833374, "memory(GiB)": 70.5, "step": 47515, "token_acc": 0.9527896995708155, "train_speed(iter/s)": 1.45115 }, { "epoch": 2.035902489182126, "grad_norm": 0.9670879244804382, "learning_rate": 6.437430348669695e-05, "loss": 0.20201995372772216, "memory(GiB)": 70.5, "step": 47520, "token_acc": 0.9552845528455285, "train_speed(iter/s)": 1.451161 }, { "epoch": 2.036116704511375, "grad_norm": 5.292684555053711, "learning_rate": 6.436785768205442e-05, "loss": 0.735888671875, "memory(GiB)": 70.5, "step": 47525, "token_acc": 0.8380566801619433, "train_speed(iter/s)": 1.451182 }, { "epoch": 2.0363309198406236, "grad_norm": 4.478448390960693, "learning_rate": 6.436141161712465e-05, "loss": 0.2968492031097412, "memory(GiB)": 70.5, "step": 47530, "token_acc": 0.9339080459770115, "train_speed(iter/s)": 1.45118 }, { "epoch": 2.036545135169873, "grad_norm": 6.283172607421875, "learning_rate": 6.435496529202446e-05, "loss": 0.3554699420928955, "memory(GiB)": 70.5, "step": 47535, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.451194 }, { "epoch": 2.0367593504991217, "grad_norm": 2.2809526920318604, "learning_rate": 6.434851870687058e-05, "loss": 0.3601693153381348, "memory(GiB)": 70.5, "step": 47540, "token_acc": 0.9390243902439024, "train_speed(iter/s)": 1.451212 }, { "epoch": 2.0369735658283705, "grad_norm": 3.461440324783325, "learning_rate": 6.434207186177982e-05, "loss": 0.5012109756469727, "memory(GiB)": 70.5, "step": 47545, "token_acc": 0.9015748031496063, "train_speed(iter/s)": 1.451207 }, { "epoch": 2.03718778115762, "grad_norm": 3.0054643154144287, "learning_rate": 6.4335624756869e-05, "loss": 0.39149577617645265, "memory(GiB)": 70.5, "step": 47550, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.451205 }, { "epoch": 2.0374019964868686, "grad_norm": 0.45907869935035706, "learning_rate": 6.432917739225486e-05, "loss": 0.23729569911956788, "memory(GiB)": 70.5, "step": 47555, "token_acc": 0.9577464788732394, "train_speed(iter/s)": 1.451219 }, { "epoch": 2.0376162118161174, "grad_norm": 3.7616541385650635, "learning_rate": 6.432272976805425e-05, "loss": 0.3430230140686035, "memory(GiB)": 70.5, "step": 47560, "token_acc": 0.9173228346456693, "train_speed(iter/s)": 1.451217 }, { "epoch": 2.0378304271453667, "grad_norm": 1.7118115425109863, "learning_rate": 6.431628188438397e-05, "loss": 0.2252413272857666, "memory(GiB)": 70.5, "step": 47565, "token_acc": 0.9471698113207547, "train_speed(iter/s)": 1.451225 }, { "epoch": 2.0380446424746155, "grad_norm": 2.098641872406006, "learning_rate": 6.430983374136078e-05, "loss": 0.38119282722473147, "memory(GiB)": 70.5, "step": 47570, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.451223 }, { "epoch": 2.0382588578038643, "grad_norm": 4.140746116638184, "learning_rate": 6.430338533910155e-05, "loss": 0.3076956748962402, "memory(GiB)": 70.5, "step": 47575, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.451242 }, { "epoch": 2.0384730731331135, "grad_norm": 2.876951217651367, "learning_rate": 6.429693667772308e-05, "loss": 0.3118378400802612, "memory(GiB)": 70.5, "step": 47580, "token_acc": 0.9387096774193548, "train_speed(iter/s)": 1.451236 }, { "epoch": 2.0386872884623624, "grad_norm": 4.22162389755249, "learning_rate": 6.429048775734216e-05, "loss": 0.4438285827636719, "memory(GiB)": 70.5, "step": 47585, "token_acc": 0.9023569023569024, "train_speed(iter/s)": 1.451249 }, { "epoch": 2.038901503791611, "grad_norm": 3.0264804363250732, "learning_rate": 6.428403857807568e-05, "loss": 0.3044466972351074, "memory(GiB)": 70.5, "step": 47590, "token_acc": 0.9453924914675768, "train_speed(iter/s)": 1.451246 }, { "epoch": 2.0391157191208604, "grad_norm": 7.2021307945251465, "learning_rate": 6.427758914004044e-05, "loss": 0.5677507877349853, "memory(GiB)": 70.5, "step": 47595, "token_acc": 0.8884892086330936, "train_speed(iter/s)": 1.451245 }, { "epoch": 2.0393299344501092, "grad_norm": 17.58225440979004, "learning_rate": 6.427113944335326e-05, "loss": 0.44292635917663575, "memory(GiB)": 70.5, "step": 47600, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.451259 }, { "epoch": 2.039544149779358, "grad_norm": 2.3579773902893066, "learning_rate": 6.426468948813102e-05, "loss": 0.25006120204925536, "memory(GiB)": 70.5, "step": 47605, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.451291 }, { "epoch": 2.0397583651086073, "grad_norm": 2.56516695022583, "learning_rate": 6.425823927449055e-05, "loss": 0.29215302467346194, "memory(GiB)": 70.5, "step": 47610, "token_acc": 0.9539473684210527, "train_speed(iter/s)": 1.451286 }, { "epoch": 2.039972580437856, "grad_norm": 3.585767984390259, "learning_rate": 6.425178880254869e-05, "loss": 0.7501347541809082, "memory(GiB)": 70.5, "step": 47615, "token_acc": 0.8620689655172413, "train_speed(iter/s)": 1.451299 }, { "epoch": 2.040186795767105, "grad_norm": 5.368339538574219, "learning_rate": 6.424533807242229e-05, "loss": 0.3546141624450684, "memory(GiB)": 70.5, "step": 47620, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.4513 }, { "epoch": 2.040401011096354, "grad_norm": 5.951714992523193, "learning_rate": 6.423888708422825e-05, "loss": 0.47978978157043456, "memory(GiB)": 70.5, "step": 47625, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.451299 }, { "epoch": 2.040615226425603, "grad_norm": 1.426073431968689, "learning_rate": 6.42324358380834e-05, "loss": 0.13866318464279176, "memory(GiB)": 70.5, "step": 47630, "token_acc": 0.968421052631579, "train_speed(iter/s)": 1.451301 }, { "epoch": 2.040829441754852, "grad_norm": 2.4685301780700684, "learning_rate": 6.422598433410465e-05, "loss": 0.39597325325012206, "memory(GiB)": 70.5, "step": 47635, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.45129 }, { "epoch": 2.041043657084101, "grad_norm": 5.677618026733398, "learning_rate": 6.421953257240882e-05, "loss": 0.3252150058746338, "memory(GiB)": 70.5, "step": 47640, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.451299 }, { "epoch": 2.04125787241335, "grad_norm": 1.8210092782974243, "learning_rate": 6.421308055311284e-05, "loss": 0.19858112335205078, "memory(GiB)": 70.5, "step": 47645, "token_acc": 0.9539473684210527, "train_speed(iter/s)": 1.451297 }, { "epoch": 2.0414720877425987, "grad_norm": 3.7062745094299316, "learning_rate": 6.420662827633358e-05, "loss": 0.5079660415649414, "memory(GiB)": 70.5, "step": 47650, "token_acc": 0.8870967741935484, "train_speed(iter/s)": 1.451295 }, { "epoch": 2.041686303071848, "grad_norm": 1.6701692342758179, "learning_rate": 6.420017574218788e-05, "loss": 0.4262245655059814, "memory(GiB)": 70.5, "step": 47655, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.451308 }, { "epoch": 2.0419005184010968, "grad_norm": 2.604637861251831, "learning_rate": 6.41937229507927e-05, "loss": 0.2659695386886597, "memory(GiB)": 70.5, "step": 47660, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.451306 }, { "epoch": 2.0421147337303456, "grad_norm": 3.8146812915802, "learning_rate": 6.418726990226491e-05, "loss": 0.40534210205078125, "memory(GiB)": 70.5, "step": 47665, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.451304 }, { "epoch": 2.042328949059595, "grad_norm": 0.681645929813385, "learning_rate": 6.418081659672142e-05, "loss": 0.325938081741333, "memory(GiB)": 70.5, "step": 47670, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.451309 }, { "epoch": 2.0425431643888436, "grad_norm": 3.6689417362213135, "learning_rate": 6.417436303427914e-05, "loss": 0.5076084613800049, "memory(GiB)": 70.5, "step": 47675, "token_acc": 0.900709219858156, "train_speed(iter/s)": 1.451315 }, { "epoch": 2.0427573797180925, "grad_norm": 3.7881829738616943, "learning_rate": 6.416790921505498e-05, "loss": 0.3991727828979492, "memory(GiB)": 70.5, "step": 47680, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.451319 }, { "epoch": 2.0429715950473417, "grad_norm": 0.2154429703950882, "learning_rate": 6.416145513916584e-05, "loss": 0.1811738967895508, "memory(GiB)": 70.5, "step": 47685, "token_acc": 0.9652777777777778, "train_speed(iter/s)": 1.45132 }, { "epoch": 2.0431858103765905, "grad_norm": 2.8881444931030273, "learning_rate": 6.415500080672866e-05, "loss": 0.6699195861816406, "memory(GiB)": 70.5, "step": 47690, "token_acc": 0.8576051779935275, "train_speed(iter/s)": 1.45132 }, { "epoch": 2.0434000257058393, "grad_norm": 5.413203239440918, "learning_rate": 6.414854621786037e-05, "loss": 0.6962713718414306, "memory(GiB)": 70.5, "step": 47695, "token_acc": 0.8691275167785235, "train_speed(iter/s)": 1.451345 }, { "epoch": 2.0436142410350886, "grad_norm": 3.3217475414276123, "learning_rate": 6.414209137267788e-05, "loss": 0.22635719776153565, "memory(GiB)": 70.5, "step": 47700, "token_acc": 0.937984496124031, "train_speed(iter/s)": 1.451351 }, { "epoch": 2.0438284563643374, "grad_norm": 2.6973862648010254, "learning_rate": 6.413563627129815e-05, "loss": 0.3922914981842041, "memory(GiB)": 70.5, "step": 47705, "token_acc": 0.9054878048780488, "train_speed(iter/s)": 1.451362 }, { "epoch": 2.044042671693586, "grad_norm": 3.2773354053497314, "learning_rate": 6.41291809138381e-05, "loss": 0.3959831237792969, "memory(GiB)": 70.5, "step": 47710, "token_acc": 0.90625, "train_speed(iter/s)": 1.451371 }, { "epoch": 2.0442568870228355, "grad_norm": 5.815387725830078, "learning_rate": 6.412272530041469e-05, "loss": 0.5971796035766601, "memory(GiB)": 70.5, "step": 47715, "token_acc": 0.8807017543859649, "train_speed(iter/s)": 1.451378 }, { "epoch": 2.0444711023520843, "grad_norm": 1.8154776096343994, "learning_rate": 6.411626943114486e-05, "loss": 0.34501121044158933, "memory(GiB)": 70.5, "step": 47720, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.451383 }, { "epoch": 2.044685317681333, "grad_norm": 2.4031107425689697, "learning_rate": 6.410981330614558e-05, "loss": 0.2543515205383301, "memory(GiB)": 70.5, "step": 47725, "token_acc": 0.9498069498069498, "train_speed(iter/s)": 1.451382 }, { "epoch": 2.0448995330105824, "grad_norm": 2.069812059402466, "learning_rate": 6.410335692553376e-05, "loss": 0.36762776374816897, "memory(GiB)": 70.5, "step": 47730, "token_acc": 0.9120234604105572, "train_speed(iter/s)": 1.451383 }, { "epoch": 2.045113748339831, "grad_norm": 7.924597263336182, "learning_rate": 6.409690028942643e-05, "loss": 0.1992565870285034, "memory(GiB)": 70.5, "step": 47735, "token_acc": 0.9626865671641791, "train_speed(iter/s)": 1.451376 }, { "epoch": 2.04532796366908, "grad_norm": 4.482367992401123, "learning_rate": 6.409044339794052e-05, "loss": 0.40611972808837893, "memory(GiB)": 70.5, "step": 47740, "token_acc": 0.8996763754045307, "train_speed(iter/s)": 1.451371 }, { "epoch": 2.0455421789983292, "grad_norm": 2.282390832901001, "learning_rate": 6.408398625119301e-05, "loss": 0.3122326135635376, "memory(GiB)": 70.5, "step": 47745, "token_acc": 0.9381107491856677, "train_speed(iter/s)": 1.45137 }, { "epoch": 2.045756394327578, "grad_norm": 2.0416977405548096, "learning_rate": 6.407752884930089e-05, "loss": 0.4419189453125, "memory(GiB)": 70.5, "step": 47750, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.451368 }, { "epoch": 2.045970609656827, "grad_norm": 2.4420273303985596, "learning_rate": 6.407107119238111e-05, "loss": 0.5943962574005127, "memory(GiB)": 70.5, "step": 47755, "token_acc": 0.8726114649681529, "train_speed(iter/s)": 1.451388 }, { "epoch": 2.046184824986076, "grad_norm": 3.7436914443969727, "learning_rate": 6.40646132805507e-05, "loss": 0.4812185287475586, "memory(GiB)": 70.5, "step": 47760, "token_acc": 0.883495145631068, "train_speed(iter/s)": 1.451377 }, { "epoch": 2.046399040315325, "grad_norm": 2.9040703773498535, "learning_rate": 6.405815511392659e-05, "loss": 0.28540453910827634, "memory(GiB)": 70.5, "step": 47765, "token_acc": 0.9423868312757202, "train_speed(iter/s)": 1.451397 }, { "epoch": 2.0466132556445737, "grad_norm": 2.167358636856079, "learning_rate": 6.405169669262583e-05, "loss": 0.4102794170379639, "memory(GiB)": 70.5, "step": 47770, "token_acc": 0.8975265017667845, "train_speed(iter/s)": 1.451396 }, { "epoch": 2.046827470973823, "grad_norm": 4.024528980255127, "learning_rate": 6.40452380167654e-05, "loss": 0.28680105209350587, "memory(GiB)": 70.5, "step": 47775, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.451394 }, { "epoch": 2.047041686303072, "grad_norm": 6.72648286819458, "learning_rate": 6.403877908646232e-05, "loss": 0.4308903694152832, "memory(GiB)": 70.5, "step": 47780, "token_acc": 0.8772563176895307, "train_speed(iter/s)": 1.451382 }, { "epoch": 2.0472559016323206, "grad_norm": 6.045637607574463, "learning_rate": 6.403231990183358e-05, "loss": 0.3862269878387451, "memory(GiB)": 70.5, "step": 47785, "token_acc": 0.9101449275362319, "train_speed(iter/s)": 1.451379 }, { "epoch": 2.04747011696157, "grad_norm": 4.277620792388916, "learning_rate": 6.40258604629962e-05, "loss": 0.5478341102600097, "memory(GiB)": 70.5, "step": 47790, "token_acc": 0.8906666666666667, "train_speed(iter/s)": 1.45138 }, { "epoch": 2.0476843322908187, "grad_norm": 3.654752492904663, "learning_rate": 6.401940077006721e-05, "loss": 0.29789109230041505, "memory(GiB)": 70.5, "step": 47795, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.451378 }, { "epoch": 2.0478985476200675, "grad_norm": 1.2843971252441406, "learning_rate": 6.401294082316361e-05, "loss": 0.2832914113998413, "memory(GiB)": 70.5, "step": 47800, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.451385 }, { "epoch": 2.0481127629493168, "grad_norm": 5.6865434646606445, "learning_rate": 6.400648062240243e-05, "loss": 0.2734884977340698, "memory(GiB)": 70.5, "step": 47805, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.451396 }, { "epoch": 2.0483269782785656, "grad_norm": 4.000848770141602, "learning_rate": 6.400002016790074e-05, "loss": 0.3219779968261719, "memory(GiB)": 70.5, "step": 47810, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.451396 }, { "epoch": 2.0485411936078144, "grad_norm": 7.857330322265625, "learning_rate": 6.399355945977554e-05, "loss": 0.3403666019439697, "memory(GiB)": 70.5, "step": 47815, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.451418 }, { "epoch": 2.0487554089370636, "grad_norm": 5.017498970031738, "learning_rate": 6.39870984981439e-05, "loss": 0.30747017860412595, "memory(GiB)": 70.5, "step": 47820, "token_acc": 0.9336099585062241, "train_speed(iter/s)": 1.451432 }, { "epoch": 2.0489696242663125, "grad_norm": 5.839710712432861, "learning_rate": 6.398063728312283e-05, "loss": 0.676287841796875, "memory(GiB)": 70.5, "step": 47825, "token_acc": 0.8582677165354331, "train_speed(iter/s)": 1.451437 }, { "epoch": 2.0491838395955613, "grad_norm": 1.8029190301895142, "learning_rate": 6.39741758148294e-05, "loss": 0.503867244720459, "memory(GiB)": 70.5, "step": 47830, "token_acc": 0.889967637540453, "train_speed(iter/s)": 1.451444 }, { "epoch": 2.0493980549248105, "grad_norm": 1.3360739946365356, "learning_rate": 6.396771409338068e-05, "loss": 0.3348002672195435, "memory(GiB)": 70.5, "step": 47835, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.451459 }, { "epoch": 2.0496122702540593, "grad_norm": 1.7316187620162964, "learning_rate": 6.39612521188937e-05, "loss": 0.39394800662994384, "memory(GiB)": 70.5, "step": 47840, "token_acc": 0.9401709401709402, "train_speed(iter/s)": 1.451469 }, { "epoch": 2.049826485583308, "grad_norm": 3.1624395847320557, "learning_rate": 6.395478989148556e-05, "loss": 0.27371203899383545, "memory(GiB)": 70.5, "step": 47845, "token_acc": 0.9297297297297298, "train_speed(iter/s)": 1.451466 }, { "epoch": 2.0500407009125574, "grad_norm": 2.9025955200195312, "learning_rate": 6.39483274112733e-05, "loss": 0.5993714809417725, "memory(GiB)": 70.5, "step": 47850, "token_acc": 0.8633333333333333, "train_speed(iter/s)": 1.45149 }, { "epoch": 2.0502549162418062, "grad_norm": 5.419928550720215, "learning_rate": 6.3941864678374e-05, "loss": 0.4683380126953125, "memory(GiB)": 70.5, "step": 47855, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.451488 }, { "epoch": 2.050469131571055, "grad_norm": 4.349549770355225, "learning_rate": 6.393540169290475e-05, "loss": 0.6288832187652588, "memory(GiB)": 70.5, "step": 47860, "token_acc": 0.8979057591623036, "train_speed(iter/s)": 1.451494 }, { "epoch": 2.0506833469003043, "grad_norm": 3.9866878986358643, "learning_rate": 6.392893845498262e-05, "loss": 0.2609997749328613, "memory(GiB)": 70.5, "step": 47865, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.451496 }, { "epoch": 2.050897562229553, "grad_norm": 6.592968463897705, "learning_rate": 6.392247496472472e-05, "loss": 0.3421879768371582, "memory(GiB)": 70.5, "step": 47870, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.451501 }, { "epoch": 2.051111777558802, "grad_norm": 6.228003978729248, "learning_rate": 6.391601122224811e-05, "loss": 0.42588214874267577, "memory(GiB)": 70.5, "step": 47875, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.451508 }, { "epoch": 2.051325992888051, "grad_norm": 0.6509956121444702, "learning_rate": 6.39095472276699e-05, "loss": 0.178023362159729, "memory(GiB)": 70.5, "step": 47880, "token_acc": 0.9673590504451038, "train_speed(iter/s)": 1.451501 }, { "epoch": 2.0515402082173, "grad_norm": 2.806661367416382, "learning_rate": 6.39030829811072e-05, "loss": 0.5397136211395264, "memory(GiB)": 70.5, "step": 47885, "token_acc": 0.8881118881118881, "train_speed(iter/s)": 1.451513 }, { "epoch": 2.051754423546549, "grad_norm": 1.5170069932937622, "learning_rate": 6.38966184826771e-05, "loss": 0.13199926614761354, "memory(GiB)": 70.5, "step": 47890, "token_acc": 0.9694189602446484, "train_speed(iter/s)": 1.45151 }, { "epoch": 2.051968638875798, "grad_norm": 1.3861052989959717, "learning_rate": 6.389015373249674e-05, "loss": 0.31626479625701903, "memory(GiB)": 70.5, "step": 47895, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.451526 }, { "epoch": 2.052182854205047, "grad_norm": 1.4245842695236206, "learning_rate": 6.388368873068321e-05, "loss": 0.3117805242538452, "memory(GiB)": 70.5, "step": 47900, "token_acc": 0.9484848484848485, "train_speed(iter/s)": 1.451532 }, { "epoch": 2.0523970695342957, "grad_norm": 4.074802398681641, "learning_rate": 6.387722347735362e-05, "loss": 0.3816797733306885, "memory(GiB)": 70.5, "step": 47905, "token_acc": 0.9013333333333333, "train_speed(iter/s)": 1.451547 }, { "epoch": 2.052611284863545, "grad_norm": 3.166527509689331, "learning_rate": 6.387075797262513e-05, "loss": 0.23688156604766847, "memory(GiB)": 70.5, "step": 47910, "token_acc": 0.9469387755102041, "train_speed(iter/s)": 1.451543 }, { "epoch": 2.0528255001927938, "grad_norm": 5.1095991134643555, "learning_rate": 6.386429221661483e-05, "loss": 0.29240126609802247, "memory(GiB)": 70.5, "step": 47915, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.451542 }, { "epoch": 2.0530397155220426, "grad_norm": 4.287453651428223, "learning_rate": 6.385782620943986e-05, "loss": 0.39183506965637205, "memory(GiB)": 70.5, "step": 47920, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.451547 }, { "epoch": 2.053253930851292, "grad_norm": 1.2477099895477295, "learning_rate": 6.385135995121739e-05, "loss": 0.2773273468017578, "memory(GiB)": 70.5, "step": 47925, "token_acc": 0.9571865443425076, "train_speed(iter/s)": 1.451563 }, { "epoch": 2.0534681461805406, "grad_norm": 1.705125093460083, "learning_rate": 6.384489344206455e-05, "loss": 0.2931018590927124, "memory(GiB)": 70.5, "step": 47930, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.45157 }, { "epoch": 2.0536823615097894, "grad_norm": 5.54170560836792, "learning_rate": 6.383842668209845e-05, "loss": 0.5268583297729492, "memory(GiB)": 70.5, "step": 47935, "token_acc": 0.9173553719008265, "train_speed(iter/s)": 1.45157 }, { "epoch": 2.0538965768390387, "grad_norm": 0.375287264585495, "learning_rate": 6.383195967143629e-05, "loss": 0.3293191194534302, "memory(GiB)": 70.5, "step": 47940, "token_acc": 0.9496644295302014, "train_speed(iter/s)": 1.451572 }, { "epoch": 2.0541107921682875, "grad_norm": 0.23856593668460846, "learning_rate": 6.38254924101952e-05, "loss": 0.07564188241958618, "memory(GiB)": 70.5, "step": 47945, "token_acc": 0.9829787234042553, "train_speed(iter/s)": 1.451576 }, { "epoch": 2.0543250074975363, "grad_norm": 3.7148847579956055, "learning_rate": 6.381902489849233e-05, "loss": 0.4205787658691406, "memory(GiB)": 70.5, "step": 47950, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.451586 }, { "epoch": 2.0545392228267856, "grad_norm": 2.568650960922241, "learning_rate": 6.381255713644488e-05, "loss": 0.36500275135040283, "memory(GiB)": 70.5, "step": 47955, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.451591 }, { "epoch": 2.0547534381560344, "grad_norm": 3.8678719997406006, "learning_rate": 6.380608912416999e-05, "loss": 0.4292935848236084, "memory(GiB)": 70.5, "step": 47960, "token_acc": 0.912621359223301, "train_speed(iter/s)": 1.451592 }, { "epoch": 2.054967653485283, "grad_norm": 1.0943546295166016, "learning_rate": 6.379962086178485e-05, "loss": 0.20083014965057372, "memory(GiB)": 70.5, "step": 47965, "token_acc": 0.9635258358662614, "train_speed(iter/s)": 1.451602 }, { "epoch": 2.0551818688145325, "grad_norm": 0.9289455413818359, "learning_rate": 6.379315234940664e-05, "loss": 0.15410978794097902, "memory(GiB)": 70.5, "step": 47970, "token_acc": 0.9714285714285714, "train_speed(iter/s)": 1.451607 }, { "epoch": 2.0553960841437813, "grad_norm": 2.720078706741333, "learning_rate": 6.378668358715252e-05, "loss": 0.4129208564758301, "memory(GiB)": 70.5, "step": 47975, "token_acc": 0.9191616766467066, "train_speed(iter/s)": 1.451615 }, { "epoch": 2.05561029947303, "grad_norm": 5.856415271759033, "learning_rate": 6.378021457513971e-05, "loss": 0.4656832695007324, "memory(GiB)": 70.5, "step": 47980, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.451625 }, { "epoch": 2.0558245148022793, "grad_norm": 1.9023401737213135, "learning_rate": 6.377374531348538e-05, "loss": 0.4250325202941895, "memory(GiB)": 70.5, "step": 47985, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.451636 }, { "epoch": 2.056038730131528, "grad_norm": 4.057559490203857, "learning_rate": 6.376727580230671e-05, "loss": 0.5019278526306152, "memory(GiB)": 70.5, "step": 47990, "token_acc": 0.8910891089108911, "train_speed(iter/s)": 1.451631 }, { "epoch": 2.056252945460777, "grad_norm": 9.407608032226562, "learning_rate": 6.376080604172096e-05, "loss": 0.5921007633209229, "memory(GiB)": 70.5, "step": 47995, "token_acc": 0.8859315589353612, "train_speed(iter/s)": 1.451643 }, { "epoch": 2.0564671607900262, "grad_norm": 4.336686134338379, "learning_rate": 6.375433603184528e-05, "loss": 0.409475040435791, "memory(GiB)": 70.5, "step": 48000, "token_acc": 0.9039735099337748, "train_speed(iter/s)": 1.451662 }, { "epoch": 2.0564671607900262, "eval_loss": 2.389533519744873, "eval_runtime": 13.7295, "eval_samples_per_second": 7.284, "eval_steps_per_second": 7.284, "eval_token_acc": 0.4383033419023136, "step": 48000 }, { "epoch": 2.056681376119275, "grad_norm": 2.8811848163604736, "learning_rate": 6.37478657727969e-05, "loss": 0.358808708190918, "memory(GiB)": 70.5, "step": 48005, "token_acc": 0.5575992255566312, "train_speed(iter/s)": 1.451034 }, { "epoch": 2.056895591448524, "grad_norm": 6.313811302185059, "learning_rate": 6.374139526469304e-05, "loss": 0.5276803970336914, "memory(GiB)": 70.5, "step": 48010, "token_acc": 0.87890625, "train_speed(iter/s)": 1.451035 }, { "epoch": 2.057109806777773, "grad_norm": 4.08439826965332, "learning_rate": 6.373492450765093e-05, "loss": 0.261932373046875, "memory(GiB)": 70.5, "step": 48015, "token_acc": 0.9322493224932249, "train_speed(iter/s)": 1.451042 }, { "epoch": 2.057324022107022, "grad_norm": 3.9468321800231934, "learning_rate": 6.372845350178776e-05, "loss": 0.5644903659820557, "memory(GiB)": 70.5, "step": 48020, "token_acc": 0.8922305764411027, "train_speed(iter/s)": 1.451065 }, { "epoch": 2.0575382374362707, "grad_norm": 3.751708984375, "learning_rate": 6.372198224722077e-05, "loss": 0.5417839050292969, "memory(GiB)": 70.5, "step": 48025, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.451076 }, { "epoch": 2.05775245276552, "grad_norm": 6.307085990905762, "learning_rate": 6.37155107440672e-05, "loss": 0.2947713851928711, "memory(GiB)": 70.5, "step": 48030, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.45108 }, { "epoch": 2.057966668094769, "grad_norm": 4.896065711975098, "learning_rate": 6.370903899244429e-05, "loss": 0.3063358306884766, "memory(GiB)": 70.5, "step": 48035, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.451082 }, { "epoch": 2.0581808834240176, "grad_norm": 1.4990531206130981, "learning_rate": 6.370256699246929e-05, "loss": 0.35884206295013427, "memory(GiB)": 70.5, "step": 48040, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.451088 }, { "epoch": 2.058395098753267, "grad_norm": 4.014467716217041, "learning_rate": 6.369609474425944e-05, "loss": 0.5995320796966552, "memory(GiB)": 70.5, "step": 48045, "token_acc": 0.8817567567567568, "train_speed(iter/s)": 1.451109 }, { "epoch": 2.0586093140825157, "grad_norm": 3.0181941986083984, "learning_rate": 6.368962224793195e-05, "loss": 0.24585583209991455, "memory(GiB)": 70.5, "step": 48050, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.451114 }, { "epoch": 2.0588235294117645, "grad_norm": 7.191644668579102, "learning_rate": 6.368314950360415e-05, "loss": 0.5727811813354492, "memory(GiB)": 70.5, "step": 48055, "token_acc": 0.8856088560885609, "train_speed(iter/s)": 1.4511 }, { "epoch": 2.0590377447410138, "grad_norm": 2.185497283935547, "learning_rate": 6.367667651139324e-05, "loss": 0.22748734951019287, "memory(GiB)": 70.5, "step": 48060, "token_acc": 0.9574468085106383, "train_speed(iter/s)": 1.4511 }, { "epoch": 2.0592519600702626, "grad_norm": 4.812835216522217, "learning_rate": 6.367020327141651e-05, "loss": 0.43746423721313477, "memory(GiB)": 70.5, "step": 48065, "token_acc": 0.910394265232975, "train_speed(iter/s)": 1.451098 }, { "epoch": 2.0594661753995114, "grad_norm": 1.8288288116455078, "learning_rate": 6.366372978379122e-05, "loss": 0.30831518173217776, "memory(GiB)": 70.5, "step": 48070, "token_acc": 0.9367469879518072, "train_speed(iter/s)": 1.4511 }, { "epoch": 2.0596803907287606, "grad_norm": 3.641284465789795, "learning_rate": 6.365725604863466e-05, "loss": 0.22469968795776368, "memory(GiB)": 70.5, "step": 48075, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.451103 }, { "epoch": 2.0598946060580094, "grad_norm": 4.968398571014404, "learning_rate": 6.365078206606408e-05, "loss": 0.4356249809265137, "memory(GiB)": 70.5, "step": 48080, "token_acc": 0.9011976047904192, "train_speed(iter/s)": 1.45112 }, { "epoch": 2.0601088213872583, "grad_norm": 5.5482940673828125, "learning_rate": 6.364430783619679e-05, "loss": 0.47083053588867185, "memory(GiB)": 70.5, "step": 48085, "token_acc": 0.9058441558441559, "train_speed(iter/s)": 1.451125 }, { "epoch": 2.0603230367165075, "grad_norm": 1.9091463088989258, "learning_rate": 6.363783335915005e-05, "loss": 0.2449277877807617, "memory(GiB)": 70.5, "step": 48090, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.451147 }, { "epoch": 2.0605372520457563, "grad_norm": 3.050417423248291, "learning_rate": 6.36313586350412e-05, "loss": 0.2962085723876953, "memory(GiB)": 70.5, "step": 48095, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.451168 }, { "epoch": 2.060751467375005, "grad_norm": 3.0330426692962646, "learning_rate": 6.362488366398746e-05, "loss": 0.44993414878845217, "memory(GiB)": 70.5, "step": 48100, "token_acc": 0.8900343642611683, "train_speed(iter/s)": 1.451187 }, { "epoch": 2.0609656827042544, "grad_norm": 3.2414615154266357, "learning_rate": 6.361840844610619e-05, "loss": 0.5946744918823242, "memory(GiB)": 70.5, "step": 48105, "token_acc": 0.8845070422535212, "train_speed(iter/s)": 1.451198 }, { "epoch": 2.061179898033503, "grad_norm": 4.700540065765381, "learning_rate": 6.361193298151466e-05, "loss": 0.45325698852539065, "memory(GiB)": 70.5, "step": 48110, "token_acc": 0.8873720136518771, "train_speed(iter/s)": 1.451192 }, { "epoch": 2.061394113362752, "grad_norm": 4.388160228729248, "learning_rate": 6.360545727033023e-05, "loss": 0.3319800138473511, "memory(GiB)": 70.5, "step": 48115, "token_acc": 0.9313725490196079, "train_speed(iter/s)": 1.451197 }, { "epoch": 2.0616083286920013, "grad_norm": 0.6891873478889465, "learning_rate": 6.359898131267014e-05, "loss": 0.30564146041870116, "memory(GiB)": 70.5, "step": 48120, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.451209 }, { "epoch": 2.06182254402125, "grad_norm": 2.7167768478393555, "learning_rate": 6.359250510865177e-05, "loss": 0.643521785736084, "memory(GiB)": 70.5, "step": 48125, "token_acc": 0.8686868686868687, "train_speed(iter/s)": 1.451226 }, { "epoch": 2.062036759350499, "grad_norm": 2.4863808155059814, "learning_rate": 6.358602865839242e-05, "loss": 0.2713578701019287, "memory(GiB)": 70.5, "step": 48130, "token_acc": 0.9122137404580153, "train_speed(iter/s)": 1.451225 }, { "epoch": 2.062250974679748, "grad_norm": 6.384767055511475, "learning_rate": 6.357955196200939e-05, "loss": 0.3857272148132324, "memory(GiB)": 70.5, "step": 48135, "token_acc": 0.9383259911894273, "train_speed(iter/s)": 1.451234 }, { "epoch": 2.062465190008997, "grad_norm": 4.887454032897949, "learning_rate": 6.357307501962007e-05, "loss": 0.1505090117454529, "memory(GiB)": 70.5, "step": 48140, "token_acc": 0.967741935483871, "train_speed(iter/s)": 1.451238 }, { "epoch": 2.062679405338246, "grad_norm": 3.1159045696258545, "learning_rate": 6.356659783134174e-05, "loss": 0.41937527656555174, "memory(GiB)": 70.5, "step": 48145, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.451237 }, { "epoch": 2.062893620667495, "grad_norm": 4.348973751068115, "learning_rate": 6.356012039729177e-05, "loss": 0.3609882116317749, "memory(GiB)": 70.5, "step": 48150, "token_acc": 0.9123867069486404, "train_speed(iter/s)": 1.451234 }, { "epoch": 2.063107835996744, "grad_norm": 1.9808168411254883, "learning_rate": 6.35536427175875e-05, "loss": 0.32533926963806153, "memory(GiB)": 70.5, "step": 48155, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.451248 }, { "epoch": 2.0633220513259927, "grad_norm": 4.084928512573242, "learning_rate": 6.354716479234629e-05, "loss": 0.20751595497131348, "memory(GiB)": 70.5, "step": 48160, "token_acc": 0.9567099567099567, "train_speed(iter/s)": 1.451249 }, { "epoch": 2.063536266655242, "grad_norm": 4.4732184410095215, "learning_rate": 6.354068662168546e-05, "loss": 0.5253073692321777, "memory(GiB)": 70.5, "step": 48165, "token_acc": 0.8844884488448845, "train_speed(iter/s)": 1.451239 }, { "epoch": 2.0637504819844907, "grad_norm": 3.5772881507873535, "learning_rate": 6.353420820572242e-05, "loss": 0.3548849105834961, "memory(GiB)": 70.5, "step": 48170, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.451234 }, { "epoch": 2.0639646973137395, "grad_norm": 2.8714542388916016, "learning_rate": 6.352772954457447e-05, "loss": 0.5135534286499024, "memory(GiB)": 70.5, "step": 48175, "token_acc": 0.8807692307692307, "train_speed(iter/s)": 1.451234 }, { "epoch": 2.064178912642989, "grad_norm": 0.7874492406845093, "learning_rate": 6.3521250638359e-05, "loss": 0.22047679424285888, "memory(GiB)": 70.5, "step": 48180, "token_acc": 0.9570957095709571, "train_speed(iter/s)": 1.451226 }, { "epoch": 2.0643931279722376, "grad_norm": 3.6951744556427, "learning_rate": 6.351477148719343e-05, "loss": 0.3798687934875488, "memory(GiB)": 70.5, "step": 48185, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.451224 }, { "epoch": 2.0646073433014864, "grad_norm": 4.319723606109619, "learning_rate": 6.350829209119509e-05, "loss": 0.33077359199523926, "memory(GiB)": 70.5, "step": 48190, "token_acc": 0.9296875, "train_speed(iter/s)": 1.451223 }, { "epoch": 2.0648215586307357, "grad_norm": 3.0903186798095703, "learning_rate": 6.350181245048135e-05, "loss": 0.529024600982666, "memory(GiB)": 70.5, "step": 48195, "token_acc": 0.9003115264797508, "train_speed(iter/s)": 1.451226 }, { "epoch": 2.0650357739599845, "grad_norm": 3.8357536792755127, "learning_rate": 6.349533256516962e-05, "loss": 0.4096415042877197, "memory(GiB)": 70.5, "step": 48200, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.451228 }, { "epoch": 2.0652499892892333, "grad_norm": 7.645828723907471, "learning_rate": 6.348885243537728e-05, "loss": 0.6309397697448731, "memory(GiB)": 70.5, "step": 48205, "token_acc": 0.8905660377358491, "train_speed(iter/s)": 1.451239 }, { "epoch": 2.0654642046184826, "grad_norm": 0.6226351857185364, "learning_rate": 6.348237206122172e-05, "loss": 0.4021893501281738, "memory(GiB)": 70.5, "step": 48210, "token_acc": 0.8973384030418251, "train_speed(iter/s)": 1.451247 }, { "epoch": 2.0656784199477314, "grad_norm": 3.937929153442383, "learning_rate": 6.347589144282034e-05, "loss": 0.4293787956237793, "memory(GiB)": 70.5, "step": 48215, "token_acc": 0.90625, "train_speed(iter/s)": 1.451239 }, { "epoch": 2.06589263527698, "grad_norm": 2.4595704078674316, "learning_rate": 6.346941058029055e-05, "loss": 0.21127009391784668, "memory(GiB)": 70.5, "step": 48220, "token_acc": 0.9576271186440678, "train_speed(iter/s)": 1.451243 }, { "epoch": 2.0661068506062295, "grad_norm": 2.2167015075683594, "learning_rate": 6.346292947374975e-05, "loss": 0.2844770193099976, "memory(GiB)": 70.5, "step": 48225, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.451262 }, { "epoch": 2.0663210659354783, "grad_norm": 5.442282676696777, "learning_rate": 6.345644812331538e-05, "loss": 0.3637293100357056, "memory(GiB)": 70.5, "step": 48230, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.451261 }, { "epoch": 2.066535281264727, "grad_norm": 4.534058570861816, "learning_rate": 6.344996652910479e-05, "loss": 0.388106632232666, "memory(GiB)": 70.5, "step": 48235, "token_acc": 0.92, "train_speed(iter/s)": 1.451267 }, { "epoch": 2.0667494965939763, "grad_norm": 3.984586000442505, "learning_rate": 6.344348469123545e-05, "loss": 0.1999220848083496, "memory(GiB)": 70.5, "step": 48240, "token_acc": 0.9558823529411765, "train_speed(iter/s)": 1.451259 }, { "epoch": 2.066963711923225, "grad_norm": 2.230614423751831, "learning_rate": 6.343700260982479e-05, "loss": 0.3024479150772095, "memory(GiB)": 70.5, "step": 48245, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.451275 }, { "epoch": 2.0671779272524744, "grad_norm": 2.8498497009277344, "learning_rate": 6.343052028499019e-05, "loss": 0.2045196771621704, "memory(GiB)": 70.5, "step": 48250, "token_acc": 0.9541284403669725, "train_speed(iter/s)": 1.451297 }, { "epoch": 2.067392142581723, "grad_norm": 3.2706964015960693, "learning_rate": 6.342403771684916e-05, "loss": 0.4225929737091064, "memory(GiB)": 70.5, "step": 48255, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.451296 }, { "epoch": 2.067606357910972, "grad_norm": 2.346573829650879, "learning_rate": 6.341755490551906e-05, "loss": 0.5500638008117675, "memory(GiB)": 70.5, "step": 48260, "token_acc": 0.8856209150326797, "train_speed(iter/s)": 1.451314 }, { "epoch": 2.0678205732402213, "grad_norm": 0.8814651370048523, "learning_rate": 6.341107185111737e-05, "loss": 0.2615276575088501, "memory(GiB)": 70.5, "step": 48265, "token_acc": 0.9345238095238095, "train_speed(iter/s)": 1.451308 }, { "epoch": 2.06803478856947, "grad_norm": 4.866836071014404, "learning_rate": 6.340458855376156e-05, "loss": 0.48769302368164064, "memory(GiB)": 70.5, "step": 48270, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.451308 }, { "epoch": 2.068249003898719, "grad_norm": 3.5726635456085205, "learning_rate": 6.339810501356903e-05, "loss": 0.4059539794921875, "memory(GiB)": 70.5, "step": 48275, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.451304 }, { "epoch": 2.068463219227968, "grad_norm": 2.8569376468658447, "learning_rate": 6.339162123065727e-05, "loss": 0.5747577667236328, "memory(GiB)": 70.5, "step": 48280, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.451306 }, { "epoch": 2.068677434557217, "grad_norm": 3.8270390033721924, "learning_rate": 6.338513720514371e-05, "loss": 0.37429807186126707, "memory(GiB)": 70.5, "step": 48285, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.451329 }, { "epoch": 2.068891649886466, "grad_norm": 4.506779670715332, "learning_rate": 6.337865293714584e-05, "loss": 0.26418678760528563, "memory(GiB)": 70.5, "step": 48290, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.451323 }, { "epoch": 2.069105865215715, "grad_norm": 1.515657901763916, "learning_rate": 6.337216842678114e-05, "loss": 0.18222551345825194, "memory(GiB)": 70.5, "step": 48295, "token_acc": 0.9645669291338582, "train_speed(iter/s)": 1.451323 }, { "epoch": 2.069320080544964, "grad_norm": 1.0335134267807007, "learning_rate": 6.336568367416706e-05, "loss": 0.17600982189178466, "memory(GiB)": 70.5, "step": 48300, "token_acc": 0.9578059071729957, "train_speed(iter/s)": 1.45132 }, { "epoch": 2.0695342958742127, "grad_norm": 0.9222844839096069, "learning_rate": 6.335919867942107e-05, "loss": 0.27444992065429685, "memory(GiB)": 70.5, "step": 48305, "token_acc": 0.9404145077720207, "train_speed(iter/s)": 1.451331 }, { "epoch": 2.069748511203462, "grad_norm": 3.0724093914031982, "learning_rate": 6.335271344266066e-05, "loss": 0.2522777795791626, "memory(GiB)": 70.5, "step": 48310, "token_acc": 0.9435215946843853, "train_speed(iter/s)": 1.45134 }, { "epoch": 2.0699627265327107, "grad_norm": 5.264583110809326, "learning_rate": 6.334622796400333e-05, "loss": 0.4976529598236084, "memory(GiB)": 70.5, "step": 48315, "token_acc": 0.8830188679245283, "train_speed(iter/s)": 1.451348 }, { "epoch": 2.0701769418619596, "grad_norm": 1.609141230583191, "learning_rate": 6.333974224356656e-05, "loss": 0.4840717315673828, "memory(GiB)": 70.5, "step": 48320, "token_acc": 0.9035714285714286, "train_speed(iter/s)": 1.451347 }, { "epoch": 2.070391157191209, "grad_norm": 2.7679245471954346, "learning_rate": 6.333325628146785e-05, "loss": 0.3503784894943237, "memory(GiB)": 70.5, "step": 48325, "token_acc": 0.9330543933054394, "train_speed(iter/s)": 1.45135 }, { "epoch": 2.0706053725204576, "grad_norm": 5.344973564147949, "learning_rate": 6.332677007782468e-05, "loss": 0.4707513332366943, "memory(GiB)": 70.5, "step": 48330, "token_acc": 0.8867313915857605, "train_speed(iter/s)": 1.451348 }, { "epoch": 2.0708195878497064, "grad_norm": 9.293176651000977, "learning_rate": 6.33202836327546e-05, "loss": 0.5121758460998536, "memory(GiB)": 70.5, "step": 48335, "token_acc": 0.8814229249011858, "train_speed(iter/s)": 1.45134 }, { "epoch": 2.0710338031789557, "grad_norm": 3.3109278678894043, "learning_rate": 6.331379694637504e-05, "loss": 0.3999786376953125, "memory(GiB)": 70.5, "step": 48340, "token_acc": 0.9201520912547528, "train_speed(iter/s)": 1.45134 }, { "epoch": 2.0712480185082045, "grad_norm": 7.352919578552246, "learning_rate": 6.33073100188036e-05, "loss": 0.3011223077774048, "memory(GiB)": 70.5, "step": 48345, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.451342 }, { "epoch": 2.0714622338374533, "grad_norm": 1.7683584690093994, "learning_rate": 6.330082285015773e-05, "loss": 0.5738058090209961, "memory(GiB)": 70.5, "step": 48350, "token_acc": 0.8733333333333333, "train_speed(iter/s)": 1.451335 }, { "epoch": 2.0716764491667026, "grad_norm": 3.946901321411133, "learning_rate": 6.329433544055499e-05, "loss": 0.1914306402206421, "memory(GiB)": 70.5, "step": 48355, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.451335 }, { "epoch": 2.0718906644959514, "grad_norm": 2.774897813796997, "learning_rate": 6.328784779011288e-05, "loss": 0.31199333667755125, "memory(GiB)": 70.5, "step": 48360, "token_acc": 0.9430604982206405, "train_speed(iter/s)": 1.451341 }, { "epoch": 2.0721048798252, "grad_norm": 3.4309096336364746, "learning_rate": 6.328135989894896e-05, "loss": 0.42864227294921875, "memory(GiB)": 70.5, "step": 48365, "token_acc": 0.903114186851211, "train_speed(iter/s)": 1.45134 }, { "epoch": 2.0723190951544495, "grad_norm": 0.4437292218208313, "learning_rate": 6.327487176718075e-05, "loss": 0.40665764808654786, "memory(GiB)": 70.5, "step": 48370, "token_acc": 0.9105960264900662, "train_speed(iter/s)": 1.451342 }, { "epoch": 2.0725333104836983, "grad_norm": 0.3609107434749603, "learning_rate": 6.326838339492577e-05, "loss": 0.28383355140686034, "memory(GiB)": 70.5, "step": 48375, "token_acc": 0.9369369369369369, "train_speed(iter/s)": 1.451341 }, { "epoch": 2.072747525812947, "grad_norm": 3.6881914138793945, "learning_rate": 6.326189478230157e-05, "loss": 0.44507341384887694, "memory(GiB)": 70.5, "step": 48380, "token_acc": 0.8971061093247589, "train_speed(iter/s)": 1.451351 }, { "epoch": 2.0729617411421963, "grad_norm": 0.9459409117698669, "learning_rate": 6.325540592942571e-05, "loss": 0.2676565408706665, "memory(GiB)": 70.5, "step": 48385, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.451351 }, { "epoch": 2.073175956471445, "grad_norm": 6.145468235015869, "learning_rate": 6.324891683641576e-05, "loss": 0.40180068016052245, "memory(GiB)": 70.5, "step": 48390, "token_acc": 0.9128787878787878, "train_speed(iter/s)": 1.451364 }, { "epoch": 2.073390171800694, "grad_norm": 3.153555154800415, "learning_rate": 6.324242750338922e-05, "loss": 0.44175381660461427, "memory(GiB)": 70.5, "step": 48395, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.451363 }, { "epoch": 2.073604387129943, "grad_norm": 5.040831089019775, "learning_rate": 6.323593793046371e-05, "loss": 0.4428744316101074, "memory(GiB)": 70.5, "step": 48400, "token_acc": 0.9294871794871795, "train_speed(iter/s)": 1.451355 }, { "epoch": 2.073818602459192, "grad_norm": 4.1550798416137695, "learning_rate": 6.322944811775676e-05, "loss": 0.40236778259277345, "memory(GiB)": 70.5, "step": 48405, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.451354 }, { "epoch": 2.074032817788441, "grad_norm": 3.0844340324401855, "learning_rate": 6.322295806538594e-05, "loss": 0.3048505067825317, "memory(GiB)": 70.5, "step": 48410, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.451361 }, { "epoch": 2.07424703311769, "grad_norm": 2.3861794471740723, "learning_rate": 6.321646777346883e-05, "loss": 0.19413909912109376, "memory(GiB)": 70.5, "step": 48415, "token_acc": 0.9471947194719472, "train_speed(iter/s)": 1.451364 }, { "epoch": 2.074461248446939, "grad_norm": 4.15631103515625, "learning_rate": 6.320997724212304e-05, "loss": 0.22923941612243653, "memory(GiB)": 70.5, "step": 48420, "token_acc": 0.9514563106796117, "train_speed(iter/s)": 1.45136 }, { "epoch": 2.0746754637761877, "grad_norm": 3.860234022140503, "learning_rate": 6.320348647146608e-05, "loss": 0.22573680877685548, "memory(GiB)": 70.5, "step": 48425, "token_acc": 0.9372822299651568, "train_speed(iter/s)": 1.451358 }, { "epoch": 2.074889679105437, "grad_norm": 6.875819683074951, "learning_rate": 6.319699546161559e-05, "loss": 0.35115323066711424, "memory(GiB)": 70.5, "step": 48430, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.451357 }, { "epoch": 2.075103894434686, "grad_norm": 1.4829641580581665, "learning_rate": 6.319050421268915e-05, "loss": 0.22831578254699708, "memory(GiB)": 70.5, "step": 48435, "token_acc": 0.9539295392953929, "train_speed(iter/s)": 1.451385 }, { "epoch": 2.0753181097639346, "grad_norm": 5.403709888458252, "learning_rate": 6.318401272480435e-05, "loss": 0.36785035133361815, "memory(GiB)": 70.5, "step": 48440, "token_acc": 0.8951612903225806, "train_speed(iter/s)": 1.451385 }, { "epoch": 2.075532325093184, "grad_norm": 4.110930442810059, "learning_rate": 6.31775209980788e-05, "loss": 0.472745418548584, "memory(GiB)": 70.5, "step": 48445, "token_acc": 0.9064516129032258, "train_speed(iter/s)": 1.451396 }, { "epoch": 2.0757465404224327, "grad_norm": 2.5683846473693848, "learning_rate": 6.317102903263008e-05, "loss": 0.4315854549407959, "memory(GiB)": 70.5, "step": 48450, "token_acc": 0.9325153374233128, "train_speed(iter/s)": 1.451402 }, { "epoch": 2.0759607557516815, "grad_norm": 5.404969692230225, "learning_rate": 6.316453682857582e-05, "loss": 0.2669505596160889, "memory(GiB)": 70.5, "step": 48455, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.451404 }, { "epoch": 2.0761749710809307, "grad_norm": 0.2349676638841629, "learning_rate": 6.315804438603363e-05, "loss": 0.27611379623413085, "memory(GiB)": 70.5, "step": 48460, "token_acc": 0.9375, "train_speed(iter/s)": 1.451415 }, { "epoch": 2.0763891864101796, "grad_norm": 1.854982614517212, "learning_rate": 6.315155170512111e-05, "loss": 0.2904746294021606, "memory(GiB)": 70.5, "step": 48465, "token_acc": 0.9355828220858896, "train_speed(iter/s)": 1.451411 }, { "epoch": 2.0766034017394284, "grad_norm": 4.194306373596191, "learning_rate": 6.31450587859559e-05, "loss": 0.45969905853271487, "memory(GiB)": 70.5, "step": 48470, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.451412 }, { "epoch": 2.0768176170686776, "grad_norm": 1.7136905193328857, "learning_rate": 6.313856562865564e-05, "loss": 0.3416821718215942, "memory(GiB)": 70.5, "step": 48475, "token_acc": 0.9301587301587302, "train_speed(iter/s)": 1.451411 }, { "epoch": 2.0770318323979264, "grad_norm": 2.77394437789917, "learning_rate": 6.313207223333793e-05, "loss": 0.23768210411071777, "memory(GiB)": 70.5, "step": 48480, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.451426 }, { "epoch": 2.0772460477271752, "grad_norm": 2.6992650032043457, "learning_rate": 6.312557860012039e-05, "loss": 0.254965615272522, "memory(GiB)": 70.5, "step": 48485, "token_acc": 0.9675324675324676, "train_speed(iter/s)": 1.451432 }, { "epoch": 2.0774602630564245, "grad_norm": 2.92889404296875, "learning_rate": 6.31190847291207e-05, "loss": 0.36970202922821044, "memory(GiB)": 70.5, "step": 48490, "token_acc": 0.9123505976095617, "train_speed(iter/s)": 1.451428 }, { "epoch": 2.0776744783856733, "grad_norm": 1.9874992370605469, "learning_rate": 6.311259062045647e-05, "loss": 0.3791363716125488, "memory(GiB)": 70.5, "step": 48495, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.451432 }, { "epoch": 2.077888693714922, "grad_norm": 2.997889280319214, "learning_rate": 6.310609627424537e-05, "loss": 0.3341566801071167, "memory(GiB)": 70.5, "step": 48500, "token_acc": 0.926530612244898, "train_speed(iter/s)": 1.451434 }, { "epoch": 2.077888693714922, "eval_loss": 2.3897643089294434, "eval_runtime": 14.151, "eval_samples_per_second": 7.067, "eval_steps_per_second": 7.067, "eval_token_acc": 0.4704, "step": 48500 }, { "epoch": 2.0781029090441714, "grad_norm": 5.15714693069458, "learning_rate": 6.309960169060504e-05, "loss": 0.2890408277511597, "memory(GiB)": 70.5, "step": 48505, "token_acc": 0.5986238532110092, "train_speed(iter/s)": 1.450771 }, { "epoch": 2.07831712437342, "grad_norm": 3.099945545196533, "learning_rate": 6.309310686965313e-05, "loss": 0.40518579483032224, "memory(GiB)": 70.5, "step": 48510, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.450761 }, { "epoch": 2.078531339702669, "grad_norm": 4.387178897857666, "learning_rate": 6.308661181150733e-05, "loss": 0.34815068244934083, "memory(GiB)": 70.5, "step": 48515, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.450763 }, { "epoch": 2.0787455550319183, "grad_norm": 3.3985838890075684, "learning_rate": 6.308011651628525e-05, "loss": 0.5290556907653808, "memory(GiB)": 70.5, "step": 48520, "token_acc": 0.9003215434083601, "train_speed(iter/s)": 1.450764 }, { "epoch": 2.078959770361167, "grad_norm": 1.965965747833252, "learning_rate": 6.30736209841046e-05, "loss": 0.3601155996322632, "memory(GiB)": 70.5, "step": 48525, "token_acc": 0.9112627986348123, "train_speed(iter/s)": 1.450773 }, { "epoch": 2.079173985690416, "grad_norm": 3.240102529525757, "learning_rate": 6.306712521508306e-05, "loss": 0.39252729415893556, "memory(GiB)": 70.5, "step": 48530, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.450779 }, { "epoch": 2.079388201019665, "grad_norm": 3.5627031326293945, "learning_rate": 6.306062920933829e-05, "loss": 0.48215112686157224, "memory(GiB)": 70.5, "step": 48535, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.450775 }, { "epoch": 2.079602416348914, "grad_norm": 4.158056735992432, "learning_rate": 6.305413296698795e-05, "loss": 0.34019103050231936, "memory(GiB)": 70.5, "step": 48540, "token_acc": 0.9330708661417323, "train_speed(iter/s)": 1.450789 }, { "epoch": 2.0798166316781628, "grad_norm": 3.374767780303955, "learning_rate": 6.304763648814974e-05, "loss": 0.2567232370376587, "memory(GiB)": 70.5, "step": 48545, "token_acc": 0.9313984168865436, "train_speed(iter/s)": 1.450791 }, { "epoch": 2.080030847007412, "grad_norm": 2.484605073928833, "learning_rate": 6.304113977294137e-05, "loss": 0.3007645130157471, "memory(GiB)": 70.5, "step": 48550, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.450792 }, { "epoch": 2.080245062336661, "grad_norm": 9.165810585021973, "learning_rate": 6.30346428214805e-05, "loss": 0.4940351963043213, "memory(GiB)": 70.5, "step": 48555, "token_acc": 0.9204545454545454, "train_speed(iter/s)": 1.450803 }, { "epoch": 2.0804592776659097, "grad_norm": 0.6089269518852234, "learning_rate": 6.302814563388487e-05, "loss": 0.2175212621688843, "memory(GiB)": 70.5, "step": 48560, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.450819 }, { "epoch": 2.080673492995159, "grad_norm": 6.179553985595703, "learning_rate": 6.302164821027217e-05, "loss": 0.3328144073486328, "memory(GiB)": 70.5, "step": 48565, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.450821 }, { "epoch": 2.0808877083244077, "grad_norm": 1.3606855869293213, "learning_rate": 6.301515055076007e-05, "loss": 0.39638509750366213, "memory(GiB)": 70.5, "step": 48570, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.450828 }, { "epoch": 2.0811019236536565, "grad_norm": 4.70668888092041, "learning_rate": 6.300865265546632e-05, "loss": 0.21503567695617676, "memory(GiB)": 70.5, "step": 48575, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.450827 }, { "epoch": 2.081316138982906, "grad_norm": 3.5074479579925537, "learning_rate": 6.300215452450862e-05, "loss": 0.26340522766113283, "memory(GiB)": 70.5, "step": 48580, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.450838 }, { "epoch": 2.0815303543121546, "grad_norm": 3.9492857456207275, "learning_rate": 6.29956561580047e-05, "loss": 0.8618377685546875, "memory(GiB)": 70.5, "step": 48585, "token_acc": 0.836676217765043, "train_speed(iter/s)": 1.450852 }, { "epoch": 2.0817445696414034, "grad_norm": 2.6277756690979004, "learning_rate": 6.298915755607228e-05, "loss": 0.28576154708862306, "memory(GiB)": 70.5, "step": 48590, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.450864 }, { "epoch": 2.0819587849706527, "grad_norm": 5.630936622619629, "learning_rate": 6.298265871882908e-05, "loss": 0.5818511962890625, "memory(GiB)": 70.5, "step": 48595, "token_acc": 0.8666666666666667, "train_speed(iter/s)": 1.450861 }, { "epoch": 2.0821730002999015, "grad_norm": 1.1109917163848877, "learning_rate": 6.297615964639283e-05, "loss": 0.5757677555084229, "memory(GiB)": 70.5, "step": 48600, "token_acc": 0.8892307692307693, "train_speed(iter/s)": 1.450855 }, { "epoch": 2.0823872156291503, "grad_norm": 0.30146318674087524, "learning_rate": 6.29696603388813e-05, "loss": 0.7059422492980957, "memory(GiB)": 70.5, "step": 48605, "token_acc": 0.8566775244299675, "train_speed(iter/s)": 1.450852 }, { "epoch": 2.0826014309583996, "grad_norm": 8.6114501953125, "learning_rate": 6.296316079641218e-05, "loss": 0.37674365043640134, "memory(GiB)": 70.5, "step": 48610, "token_acc": 0.9198717948717948, "train_speed(iter/s)": 1.450848 }, { "epoch": 2.0828156462876484, "grad_norm": 2.0797200202941895, "learning_rate": 6.295666101910325e-05, "loss": 0.3650477409362793, "memory(GiB)": 70.5, "step": 48615, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.450845 }, { "epoch": 2.083029861616897, "grad_norm": 3.6763556003570557, "learning_rate": 6.295016100707226e-05, "loss": 0.3471231937408447, "memory(GiB)": 70.5, "step": 48620, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.450844 }, { "epoch": 2.0832440769461464, "grad_norm": 4.333076477050781, "learning_rate": 6.294366076043695e-05, "loss": 0.7515505790710449, "memory(GiB)": 70.5, "step": 48625, "token_acc": 0.8120567375886525, "train_speed(iter/s)": 1.450845 }, { "epoch": 2.0834582922753953, "grad_norm": 1.15593683719635, "learning_rate": 6.293716027931507e-05, "loss": 0.5116789817810059, "memory(GiB)": 70.5, "step": 48630, "token_acc": 0.9053627760252366, "train_speed(iter/s)": 1.450859 }, { "epoch": 2.083672507604644, "grad_norm": 6.188859939575195, "learning_rate": 6.293065956382442e-05, "loss": 0.5216177940368653, "memory(GiB)": 70.5, "step": 48635, "token_acc": 0.899641577060932, "train_speed(iter/s)": 1.450853 }, { "epoch": 2.0838867229338933, "grad_norm": 4.061227798461914, "learning_rate": 6.292415861408273e-05, "loss": 0.5690081119537354, "memory(GiB)": 70.5, "step": 48640, "token_acc": 0.876984126984127, "train_speed(iter/s)": 1.450856 }, { "epoch": 2.084100938263142, "grad_norm": 2.9685475826263428, "learning_rate": 6.291765743020779e-05, "loss": 0.2780740737915039, "memory(GiB)": 70.5, "step": 48645, "token_acc": 0.9337349397590361, "train_speed(iter/s)": 1.450866 }, { "epoch": 2.084315153592391, "grad_norm": 1.4230252504348755, "learning_rate": 6.291115601231736e-05, "loss": 0.1783167839050293, "memory(GiB)": 70.5, "step": 48650, "token_acc": 0.9590163934426229, "train_speed(iter/s)": 1.450863 }, { "epoch": 2.08452936892164, "grad_norm": 2.289489984512329, "learning_rate": 6.290465436052921e-05, "loss": 0.275003981590271, "memory(GiB)": 70.5, "step": 48655, "token_acc": 0.946875, "train_speed(iter/s)": 1.450873 }, { "epoch": 2.084743584250889, "grad_norm": 1.2914769649505615, "learning_rate": 6.289815247496117e-05, "loss": 0.19538272619247438, "memory(GiB)": 70.5, "step": 48660, "token_acc": 0.9672727272727273, "train_speed(iter/s)": 1.450872 }, { "epoch": 2.084957799580138, "grad_norm": 5.049959659576416, "learning_rate": 6.289165035573098e-05, "loss": 0.6093783378601074, "memory(GiB)": 70.5, "step": 48665, "token_acc": 0.8963210702341137, "train_speed(iter/s)": 1.45087 }, { "epoch": 2.085172014909387, "grad_norm": 1.037330150604248, "learning_rate": 6.288514800295647e-05, "loss": 0.43264317512512207, "memory(GiB)": 70.5, "step": 48670, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.450872 }, { "epoch": 2.085386230238636, "grad_norm": 3.580899715423584, "learning_rate": 6.287864541675542e-05, "loss": 0.2615530014038086, "memory(GiB)": 70.5, "step": 48675, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.450895 }, { "epoch": 2.0856004455678847, "grad_norm": 0.1935407966375351, "learning_rate": 6.287214259724559e-05, "loss": 0.1984245777130127, "memory(GiB)": 70.5, "step": 48680, "token_acc": 0.9455782312925171, "train_speed(iter/s)": 1.450916 }, { "epoch": 2.085814660897134, "grad_norm": 5.137082099914551, "learning_rate": 6.286563954454485e-05, "loss": 0.18660953044891357, "memory(GiB)": 70.5, "step": 48685, "token_acc": 0.9689655172413794, "train_speed(iter/s)": 1.450915 }, { "epoch": 2.0860288762263828, "grad_norm": 2.337155818939209, "learning_rate": 6.285913625877097e-05, "loss": 0.42983036041259765, "memory(GiB)": 70.5, "step": 48690, "token_acc": 0.8964285714285715, "train_speed(iter/s)": 1.450917 }, { "epoch": 2.0862430915556316, "grad_norm": 1.5980885028839111, "learning_rate": 6.285263274004179e-05, "loss": 0.34491715431213377, "memory(GiB)": 70.5, "step": 48695, "token_acc": 0.916083916083916, "train_speed(iter/s)": 1.450919 }, { "epoch": 2.086457306884881, "grad_norm": 4.148499965667725, "learning_rate": 6.284612898847508e-05, "loss": 0.43911104202270507, "memory(GiB)": 70.5, "step": 48700, "token_acc": 0.9028571428571428, "train_speed(iter/s)": 1.450929 }, { "epoch": 2.0866715222141297, "grad_norm": 1.7769840955734253, "learning_rate": 6.283962500418872e-05, "loss": 0.4108292579650879, "memory(GiB)": 70.5, "step": 48705, "token_acc": 0.92, "train_speed(iter/s)": 1.450933 }, { "epoch": 2.0868857375433785, "grad_norm": 7.415637493133545, "learning_rate": 6.28331207873005e-05, "loss": 0.32784745693206785, "memory(GiB)": 70.5, "step": 48710, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.450926 }, { "epoch": 2.0870999528726277, "grad_norm": 2.4195077419281006, "learning_rate": 6.282661633792826e-05, "loss": 0.4561657428741455, "memory(GiB)": 70.5, "step": 48715, "token_acc": 0.916030534351145, "train_speed(iter/s)": 1.450941 }, { "epoch": 2.0873141682018765, "grad_norm": 3.400618314743042, "learning_rate": 6.282011165618984e-05, "loss": 0.1919680953025818, "memory(GiB)": 70.5, "step": 48720, "token_acc": 0.9551724137931035, "train_speed(iter/s)": 1.450938 }, { "epoch": 2.0875283835311254, "grad_norm": 2.795142889022827, "learning_rate": 6.281360674220305e-05, "loss": 0.337675666809082, "memory(GiB)": 70.5, "step": 48725, "token_acc": 0.9262820512820513, "train_speed(iter/s)": 1.450953 }, { "epoch": 2.0877425988603746, "grad_norm": 7.879018306732178, "learning_rate": 6.280710159608578e-05, "loss": 0.5332611083984375, "memory(GiB)": 70.5, "step": 48730, "token_acc": 0.888135593220339, "train_speed(iter/s)": 1.450965 }, { "epoch": 2.0879568141896234, "grad_norm": 4.594374656677246, "learning_rate": 6.280059621795583e-05, "loss": 0.27802438735961915, "memory(GiB)": 70.5, "step": 48735, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.45096 }, { "epoch": 2.0881710295188722, "grad_norm": 2.8630759716033936, "learning_rate": 6.279409060793109e-05, "loss": 0.3421756744384766, "memory(GiB)": 70.5, "step": 48740, "token_acc": 0.9363957597173145, "train_speed(iter/s)": 1.450959 }, { "epoch": 2.0883852448481215, "grad_norm": 5.997331619262695, "learning_rate": 6.278758476612937e-05, "loss": 0.4696558952331543, "memory(GiB)": 70.5, "step": 48745, "token_acc": 0.9040247678018576, "train_speed(iter/s)": 1.450943 }, { "epoch": 2.0885994601773703, "grad_norm": 3.1972477436065674, "learning_rate": 6.278107869266859e-05, "loss": 0.32395186424255373, "memory(GiB)": 70.5, "step": 48750, "token_acc": 0.9289940828402367, "train_speed(iter/s)": 1.450947 }, { "epoch": 2.088813675506619, "grad_norm": 5.403107166290283, "learning_rate": 6.277457238766655e-05, "loss": 0.45914878845214846, "memory(GiB)": 70.5, "step": 48755, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.450955 }, { "epoch": 2.0890278908358684, "grad_norm": 2.466822385787964, "learning_rate": 6.276806585124116e-05, "loss": 0.47505550384521483, "memory(GiB)": 70.5, "step": 48760, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.450961 }, { "epoch": 2.089242106165117, "grad_norm": 2.0634052753448486, "learning_rate": 6.276155908351027e-05, "loss": 0.4240112781524658, "memory(GiB)": 70.5, "step": 48765, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.450962 }, { "epoch": 2.089456321494366, "grad_norm": 2.0088281631469727, "learning_rate": 6.275505208459178e-05, "loss": 0.48057923316955564, "memory(GiB)": 70.5, "step": 48770, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.450963 }, { "epoch": 2.0896705368236153, "grad_norm": 5.785497188568115, "learning_rate": 6.274854485460355e-05, "loss": 0.4321177005767822, "memory(GiB)": 70.5, "step": 48775, "token_acc": 0.9097744360902256, "train_speed(iter/s)": 1.450965 }, { "epoch": 2.089884752152864, "grad_norm": 2.06437611579895, "learning_rate": 6.274203739366347e-05, "loss": 0.4153263568878174, "memory(GiB)": 70.5, "step": 48780, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.450968 }, { "epoch": 2.090098967482113, "grad_norm": 1.0311211347579956, "learning_rate": 6.273552970188942e-05, "loss": 0.18832343816757202, "memory(GiB)": 70.5, "step": 48785, "token_acc": 0.9530791788856305, "train_speed(iter/s)": 1.450977 }, { "epoch": 2.090313182811362, "grad_norm": 4.477016448974609, "learning_rate": 6.272902177939933e-05, "loss": 0.5020005226135253, "memory(GiB)": 70.5, "step": 48790, "token_acc": 0.8905775075987842, "train_speed(iter/s)": 1.450974 }, { "epoch": 2.090527398140611, "grad_norm": 4.804495334625244, "learning_rate": 6.272251362631107e-05, "loss": 0.38275647163391113, "memory(GiB)": 70.5, "step": 48795, "token_acc": 0.9213483146067416, "train_speed(iter/s)": 1.450984 }, { "epoch": 2.0907416134698598, "grad_norm": 2.9836015701293945, "learning_rate": 6.27160052427425e-05, "loss": 0.2642987728118896, "memory(GiB)": 70.5, "step": 48800, "token_acc": 0.9205020920502092, "train_speed(iter/s)": 1.450984 }, { "epoch": 2.090955828799109, "grad_norm": 3.385003089904785, "learning_rate": 6.27094966288116e-05, "loss": 0.22817730903625488, "memory(GiB)": 70.5, "step": 48805, "token_acc": 0.9480968858131488, "train_speed(iter/s)": 1.450987 }, { "epoch": 2.091170044128358, "grad_norm": 3.1508185863494873, "learning_rate": 6.270298778463624e-05, "loss": 0.35045804977416994, "memory(GiB)": 70.5, "step": 48810, "token_acc": 0.9313432835820895, "train_speed(iter/s)": 1.450988 }, { "epoch": 2.0913842594576066, "grad_norm": 2.196338415145874, "learning_rate": 6.269647871033432e-05, "loss": 0.4168813705444336, "memory(GiB)": 70.5, "step": 48815, "token_acc": 0.9226006191950464, "train_speed(iter/s)": 1.450993 }, { "epoch": 2.091598474786856, "grad_norm": 4.945016384124756, "learning_rate": 6.26899694060238e-05, "loss": 0.3370008945465088, "memory(GiB)": 70.5, "step": 48820, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.450992 }, { "epoch": 2.0918126901161047, "grad_norm": 2.3158817291259766, "learning_rate": 6.26834598718226e-05, "loss": 0.33293101787567136, "memory(GiB)": 70.5, "step": 48825, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.450991 }, { "epoch": 2.0920269054453535, "grad_norm": 4.108760833740234, "learning_rate": 6.267695010784859e-05, "loss": 0.5591896057128907, "memory(GiB)": 70.5, "step": 48830, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.450995 }, { "epoch": 2.092241120774603, "grad_norm": 3.8878707885742188, "learning_rate": 6.267044011421975e-05, "loss": 0.2610881805419922, "memory(GiB)": 70.5, "step": 48835, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.451004 }, { "epoch": 2.0924553361038516, "grad_norm": 1.5194402933120728, "learning_rate": 6.2663929891054e-05, "loss": 0.4816396236419678, "memory(GiB)": 70.5, "step": 48840, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.451011 }, { "epoch": 2.0926695514331004, "grad_norm": 6.371880531311035, "learning_rate": 6.265741943846926e-05, "loss": 0.5816590309143066, "memory(GiB)": 70.5, "step": 48845, "token_acc": 0.8915343915343915, "train_speed(iter/s)": 1.451014 }, { "epoch": 2.0928837667623497, "grad_norm": 3.1127243041992188, "learning_rate": 6.265090875658353e-05, "loss": 0.5447323799133301, "memory(GiB)": 70.5, "step": 48850, "token_acc": 0.8962536023054755, "train_speed(iter/s)": 1.451012 }, { "epoch": 2.0930979820915985, "grad_norm": 2.6646816730499268, "learning_rate": 6.264439784551472e-05, "loss": 0.46080570220947265, "memory(GiB)": 70.5, "step": 48855, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.451007 }, { "epoch": 2.0933121974208473, "grad_norm": 1.7190806865692139, "learning_rate": 6.263788670538075e-05, "loss": 0.22560627460479737, "memory(GiB)": 70.5, "step": 48860, "token_acc": 0.9659863945578231, "train_speed(iter/s)": 1.451008 }, { "epoch": 2.0935264127500965, "grad_norm": 3.5743324756622314, "learning_rate": 6.263137533629963e-05, "loss": 0.22698256969451905, "memory(GiB)": 70.5, "step": 48865, "token_acc": 0.9631901840490797, "train_speed(iter/s)": 1.451017 }, { "epoch": 2.0937406280793454, "grad_norm": 1.930320382118225, "learning_rate": 6.262486373838929e-05, "loss": 0.6803817749023438, "memory(GiB)": 70.5, "step": 48870, "token_acc": 0.8600583090379009, "train_speed(iter/s)": 1.451029 }, { "epoch": 2.093954843408594, "grad_norm": 1.5556195974349976, "learning_rate": 6.261835191176769e-05, "loss": 0.3802447557449341, "memory(GiB)": 70.5, "step": 48875, "token_acc": 0.9066147859922179, "train_speed(iter/s)": 1.451038 }, { "epoch": 2.0941690587378434, "grad_norm": 0.19306300580501556, "learning_rate": 6.261183985655281e-05, "loss": 0.22728865146636962, "memory(GiB)": 70.5, "step": 48880, "token_acc": 0.952191235059761, "train_speed(iter/s)": 1.451045 }, { "epoch": 2.0943832740670922, "grad_norm": 3.053591251373291, "learning_rate": 6.260532757286264e-05, "loss": 0.34217195510864257, "memory(GiB)": 70.5, "step": 48885, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.451044 }, { "epoch": 2.094597489396341, "grad_norm": 3.559790849685669, "learning_rate": 6.259881506081512e-05, "loss": 0.5891656875610352, "memory(GiB)": 70.5, "step": 48890, "token_acc": 0.86875, "train_speed(iter/s)": 1.451069 }, { "epoch": 2.0948117047255903, "grad_norm": 3.6704745292663574, "learning_rate": 6.259230232052826e-05, "loss": 0.5680595397949219, "memory(GiB)": 70.5, "step": 48895, "token_acc": 0.8545454545454545, "train_speed(iter/s)": 1.451061 }, { "epoch": 2.095025920054839, "grad_norm": 3.121000051498413, "learning_rate": 6.258578935212e-05, "loss": 0.25533106327056887, "memory(GiB)": 70.5, "step": 48900, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.45106 }, { "epoch": 2.095240135384088, "grad_norm": 1.6977643966674805, "learning_rate": 6.257927615570839e-05, "loss": 0.27080526351928713, "memory(GiB)": 70.5, "step": 48905, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.451067 }, { "epoch": 2.095454350713337, "grad_norm": 9.019216537475586, "learning_rate": 6.257276273141139e-05, "loss": 0.3344238758087158, "memory(GiB)": 70.5, "step": 48910, "token_acc": 0.9516728624535316, "train_speed(iter/s)": 1.451063 }, { "epoch": 2.095668566042586, "grad_norm": 2.7934629917144775, "learning_rate": 6.256624907934699e-05, "loss": 0.5543661117553711, "memory(GiB)": 70.5, "step": 48915, "token_acc": 0.8996539792387543, "train_speed(iter/s)": 1.451067 }, { "epoch": 2.095882781371835, "grad_norm": 0.6299164295196533, "learning_rate": 6.25597351996332e-05, "loss": 0.47747135162353516, "memory(GiB)": 70.5, "step": 48920, "token_acc": 0.8977272727272727, "train_speed(iter/s)": 1.451055 }, { "epoch": 2.096096996701084, "grad_norm": 2.3269848823547363, "learning_rate": 6.255322109238803e-05, "loss": 0.1856738805770874, "memory(GiB)": 70.5, "step": 48925, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.45105 }, { "epoch": 2.096311212030333, "grad_norm": 5.0594892501831055, "learning_rate": 6.254670675772947e-05, "loss": 0.4291266441345215, "memory(GiB)": 70.5, "step": 48930, "token_acc": 0.910958904109589, "train_speed(iter/s)": 1.451042 }, { "epoch": 2.0965254273595817, "grad_norm": 4.3346452713012695, "learning_rate": 6.254019219577556e-05, "loss": 0.4048892021179199, "memory(GiB)": 70.5, "step": 48935, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.451046 }, { "epoch": 2.096739642688831, "grad_norm": 0.31731462478637695, "learning_rate": 6.253367740664431e-05, "loss": 0.4844684600830078, "memory(GiB)": 70.5, "step": 48940, "token_acc": 0.8825301204819277, "train_speed(iter/s)": 1.451056 }, { "epoch": 2.0969538580180798, "grad_norm": 4.787808895111084, "learning_rate": 6.252716239045372e-05, "loss": 0.782163667678833, "memory(GiB)": 70.5, "step": 48945, "token_acc": 0.8435754189944135, "train_speed(iter/s)": 1.45108 }, { "epoch": 2.0971680733473286, "grad_norm": 3.0487823486328125, "learning_rate": 6.252064714732185e-05, "loss": 0.34403438568115235, "memory(GiB)": 70.5, "step": 48950, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.451077 }, { "epoch": 2.097382288676578, "grad_norm": 3.385474920272827, "learning_rate": 6.25141316773667e-05, "loss": 0.30891273021697996, "memory(GiB)": 70.5, "step": 48955, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.451079 }, { "epoch": 2.0975965040058266, "grad_norm": 2.5737500190734863, "learning_rate": 6.250761598070632e-05, "loss": 0.35570459365844725, "memory(GiB)": 70.5, "step": 48960, "token_acc": 0.925531914893617, "train_speed(iter/s)": 1.451077 }, { "epoch": 2.0978107193350755, "grad_norm": 6.589915752410889, "learning_rate": 6.250110005745874e-05, "loss": 0.28542590141296387, "memory(GiB)": 70.5, "step": 48965, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.451077 }, { "epoch": 2.0980249346643247, "grad_norm": 3.9060862064361572, "learning_rate": 6.249458390774201e-05, "loss": 0.36191093921661377, "memory(GiB)": 70.5, "step": 48970, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.451075 }, { "epoch": 2.0982391499935735, "grad_norm": 1.8830088376998901, "learning_rate": 6.248806753167417e-05, "loss": 0.3813464641571045, "memory(GiB)": 70.5, "step": 48975, "token_acc": 0.9240121580547113, "train_speed(iter/s)": 1.451076 }, { "epoch": 2.0984533653228223, "grad_norm": 3.903520345687866, "learning_rate": 6.248155092937326e-05, "loss": 0.4449927806854248, "memory(GiB)": 70.5, "step": 48980, "token_acc": 0.9088145896656535, "train_speed(iter/s)": 1.451091 }, { "epoch": 2.0986675806520716, "grad_norm": 3.577705144882202, "learning_rate": 6.247503410095737e-05, "loss": 0.4419257164001465, "memory(GiB)": 70.5, "step": 48985, "token_acc": 0.8881118881118881, "train_speed(iter/s)": 1.451109 }, { "epoch": 2.0988817959813204, "grad_norm": 1.3505948781967163, "learning_rate": 6.246851704654451e-05, "loss": 0.44902777671813965, "memory(GiB)": 70.5, "step": 48990, "token_acc": 0.9005376344086021, "train_speed(iter/s)": 1.451122 }, { "epoch": 2.099096011310569, "grad_norm": 6.309841156005859, "learning_rate": 6.246199976625277e-05, "loss": 0.4425994873046875, "memory(GiB)": 70.5, "step": 48995, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.45114 }, { "epoch": 2.0993102266398185, "grad_norm": 1.8237158060073853, "learning_rate": 6.245548226020024e-05, "loss": 0.3071705102920532, "memory(GiB)": 70.5, "step": 49000, "token_acc": 0.9427480916030534, "train_speed(iter/s)": 1.451144 }, { "epoch": 2.0993102266398185, "eval_loss": 2.4354279041290283, "eval_runtime": 13.7787, "eval_samples_per_second": 7.258, "eval_steps_per_second": 7.258, "eval_token_acc": 0.46321525885558584, "step": 49000 }, { "epoch": 2.0995244419690673, "grad_norm": 3.0524396896362305, "learning_rate": 6.244896452850494e-05, "loss": 0.23472244739532472, "memory(GiB)": 70.5, "step": 49005, "token_acc": 0.6085714285714285, "train_speed(iter/s)": 1.450515 }, { "epoch": 2.099738657298316, "grad_norm": 2.637918472290039, "learning_rate": 6.244244657128498e-05, "loss": 0.2538157939910889, "memory(GiB)": 70.5, "step": 49010, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.450542 }, { "epoch": 2.0999528726275654, "grad_norm": 3.1463634967803955, "learning_rate": 6.243592838865842e-05, "loss": 0.2791051149368286, "memory(GiB)": 70.5, "step": 49015, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.450561 }, { "epoch": 2.100167087956814, "grad_norm": 3.4188790321350098, "learning_rate": 6.242940998074336e-05, "loss": 0.5734274864196778, "memory(GiB)": 70.5, "step": 49020, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.450562 }, { "epoch": 2.100381303286063, "grad_norm": 1.8595079183578491, "learning_rate": 6.242289134765788e-05, "loss": 0.13034957647323608, "memory(GiB)": 70.5, "step": 49025, "token_acc": 0.9761904761904762, "train_speed(iter/s)": 1.450563 }, { "epoch": 2.1005955186153122, "grad_norm": 2.3614296913146973, "learning_rate": 6.241637248952006e-05, "loss": 0.2629213571548462, "memory(GiB)": 70.5, "step": 49030, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.450558 }, { "epoch": 2.100809733944561, "grad_norm": 6.112635612487793, "learning_rate": 6.2409853406448e-05, "loss": 0.7544786930084229, "memory(GiB)": 70.5, "step": 49035, "token_acc": 0.839142091152815, "train_speed(iter/s)": 1.450574 }, { "epoch": 2.10102394927381, "grad_norm": 5.785514831542969, "learning_rate": 6.240333409855983e-05, "loss": 0.20059309005737305, "memory(GiB)": 70.5, "step": 49040, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.450583 }, { "epoch": 2.101238164603059, "grad_norm": 5.450356483459473, "learning_rate": 6.239681456597361e-05, "loss": 0.49997835159301757, "memory(GiB)": 70.5, "step": 49045, "token_acc": 0.9052287581699346, "train_speed(iter/s)": 1.450591 }, { "epoch": 2.101452379932308, "grad_norm": 6.176684379577637, "learning_rate": 6.239029480880747e-05, "loss": 0.32757866382598877, "memory(GiB)": 70.5, "step": 49050, "token_acc": 0.9233576642335767, "train_speed(iter/s)": 1.450589 }, { "epoch": 2.1016665952615567, "grad_norm": 2.219521999359131, "learning_rate": 6.238377482717951e-05, "loss": 0.1732121229171753, "memory(GiB)": 70.5, "step": 49055, "token_acc": 0.9612403100775194, "train_speed(iter/s)": 1.450593 }, { "epoch": 2.101880810590806, "grad_norm": 2.8577823638916016, "learning_rate": 6.237725462120784e-05, "loss": 0.3631949186325073, "memory(GiB)": 70.5, "step": 49060, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.450596 }, { "epoch": 2.102095025920055, "grad_norm": 2.8905727863311768, "learning_rate": 6.237073419101061e-05, "loss": 0.5338173389434815, "memory(GiB)": 70.5, "step": 49065, "token_acc": 0.8972602739726028, "train_speed(iter/s)": 1.450606 }, { "epoch": 2.1023092412493036, "grad_norm": 6.025857925415039, "learning_rate": 6.236421353670592e-05, "loss": 0.5129109859466553, "memory(GiB)": 70.5, "step": 49070, "token_acc": 0.9042145593869731, "train_speed(iter/s)": 1.4506 }, { "epoch": 2.102523456578553, "grad_norm": 7.165661334991455, "learning_rate": 6.235769265841191e-05, "loss": 0.5030503273010254, "memory(GiB)": 70.5, "step": 49075, "token_acc": 0.8870431893687708, "train_speed(iter/s)": 1.450615 }, { "epoch": 2.1027376719078017, "grad_norm": 2.7811691761016846, "learning_rate": 6.235117155624671e-05, "loss": 0.48076434135437013, "memory(GiB)": 70.5, "step": 49080, "token_acc": 0.8945578231292517, "train_speed(iter/s)": 1.450639 }, { "epoch": 2.1029518872370505, "grad_norm": 6.528554916381836, "learning_rate": 6.234465023032844e-05, "loss": 0.5637404441833496, "memory(GiB)": 70.5, "step": 49085, "token_acc": 0.8549382716049383, "train_speed(iter/s)": 1.450655 }, { "epoch": 2.1031661025662998, "grad_norm": 2.5267324447631836, "learning_rate": 6.233812868077525e-05, "loss": 0.2975089311599731, "memory(GiB)": 70.5, "step": 49090, "token_acc": 0.9345238095238095, "train_speed(iter/s)": 1.45066 }, { "epoch": 2.1033803178955486, "grad_norm": 1.9254422187805176, "learning_rate": 6.233160690770528e-05, "loss": 0.2635096549987793, "memory(GiB)": 70.5, "step": 49095, "token_acc": 0.9433333333333334, "train_speed(iter/s)": 1.450667 }, { "epoch": 2.1035945332247974, "grad_norm": 1.4183447360992432, "learning_rate": 6.23250849112367e-05, "loss": 0.15317966938018798, "memory(GiB)": 70.5, "step": 49100, "token_acc": 0.9605734767025089, "train_speed(iter/s)": 1.450667 }, { "epoch": 2.1038087485540466, "grad_norm": 1.9185805320739746, "learning_rate": 6.231856269148762e-05, "loss": 0.26027276515960696, "memory(GiB)": 70.5, "step": 49105, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.450668 }, { "epoch": 2.1040229638832955, "grad_norm": 4.741833209991455, "learning_rate": 6.231204024857624e-05, "loss": 0.4668295383453369, "memory(GiB)": 70.5, "step": 49110, "token_acc": 0.908675799086758, "train_speed(iter/s)": 1.450698 }, { "epoch": 2.1042371792125443, "grad_norm": 4.055414199829102, "learning_rate": 6.23055175826207e-05, "loss": 0.35059025287628176, "memory(GiB)": 70.5, "step": 49115, "token_acc": 0.9305019305019305, "train_speed(iter/s)": 1.450696 }, { "epoch": 2.1044513945417935, "grad_norm": 2.1454503536224365, "learning_rate": 6.229899469373917e-05, "loss": 0.3525214195251465, "memory(GiB)": 70.5, "step": 49120, "token_acc": 0.9244186046511628, "train_speed(iter/s)": 1.45069 }, { "epoch": 2.1046656098710423, "grad_norm": 3.1106154918670654, "learning_rate": 6.229247158204981e-05, "loss": 0.21239638328552246, "memory(GiB)": 70.5, "step": 49125, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.450686 }, { "epoch": 2.104879825200291, "grad_norm": 3.6827263832092285, "learning_rate": 6.228594824767078e-05, "loss": 0.33902225494384763, "memory(GiB)": 70.5, "step": 49130, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.450677 }, { "epoch": 2.1050940405295404, "grad_norm": 3.2343292236328125, "learning_rate": 6.227942469072027e-05, "loss": 0.3091240644454956, "memory(GiB)": 70.5, "step": 49135, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.450681 }, { "epoch": 2.1053082558587892, "grad_norm": 5.716817378997803, "learning_rate": 6.227290091131648e-05, "loss": 0.23414213657379152, "memory(GiB)": 70.5, "step": 49140, "token_acc": 0.9702127659574468, "train_speed(iter/s)": 1.450687 }, { "epoch": 2.105522471188038, "grad_norm": 9.096633911132812, "learning_rate": 6.22663769095776e-05, "loss": 0.46207103729248045, "memory(GiB)": 70.5, "step": 49145, "token_acc": 0.8861209964412812, "train_speed(iter/s)": 1.450678 }, { "epoch": 2.1057366865172873, "grad_norm": 4.203575134277344, "learning_rate": 6.225985268562175e-05, "loss": 0.603237533569336, "memory(GiB)": 70.5, "step": 49150, "token_acc": 0.8538205980066446, "train_speed(iter/s)": 1.450669 }, { "epoch": 2.105950901846536, "grad_norm": 5.721044063568115, "learning_rate": 6.22533282395672e-05, "loss": 0.46817588806152344, "memory(GiB)": 70.5, "step": 49155, "token_acc": 0.8958904109589041, "train_speed(iter/s)": 1.450668 }, { "epoch": 2.106165117175785, "grad_norm": 0.38815775513648987, "learning_rate": 6.22468035715321e-05, "loss": 0.24956235885620118, "memory(GiB)": 70.5, "step": 49160, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.450669 }, { "epoch": 2.106379332505034, "grad_norm": 4.362469673156738, "learning_rate": 6.224027868163467e-05, "loss": 0.32048027515411376, "memory(GiB)": 70.5, "step": 49165, "token_acc": 0.9465408805031447, "train_speed(iter/s)": 1.450676 }, { "epoch": 2.106593547834283, "grad_norm": 1.40098237991333, "learning_rate": 6.223375356999311e-05, "loss": 0.3968137502670288, "memory(GiB)": 70.5, "step": 49170, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.450685 }, { "epoch": 2.106807763163532, "grad_norm": 3.003443956375122, "learning_rate": 6.222722823672562e-05, "loss": 0.2882606029510498, "memory(GiB)": 70.5, "step": 49175, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.450687 }, { "epoch": 2.107021978492781, "grad_norm": 3.11020565032959, "learning_rate": 6.222070268195041e-05, "loss": 0.3897082805633545, "memory(GiB)": 70.5, "step": 49180, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.450714 }, { "epoch": 2.10723619382203, "grad_norm": 2.74399471282959, "learning_rate": 6.221417690578574e-05, "loss": 0.3054746389389038, "memory(GiB)": 70.5, "step": 49185, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.450726 }, { "epoch": 2.1074504091512787, "grad_norm": 4.276601791381836, "learning_rate": 6.220765090834977e-05, "loss": 0.5457033634185791, "memory(GiB)": 70.5, "step": 49190, "token_acc": 0.8973384030418251, "train_speed(iter/s)": 1.450766 }, { "epoch": 2.107664624480528, "grad_norm": 1.8728678226470947, "learning_rate": 6.220112468976076e-05, "loss": 0.4744720935821533, "memory(GiB)": 70.5, "step": 49195, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.450771 }, { "epoch": 2.1078788398097767, "grad_norm": 4.888040542602539, "learning_rate": 6.219459825013694e-05, "loss": 0.24380898475646973, "memory(GiB)": 70.5, "step": 49200, "token_acc": 0.9494584837545126, "train_speed(iter/s)": 1.450771 }, { "epoch": 2.1080930551390256, "grad_norm": 1.5281093120574951, "learning_rate": 6.218807158959652e-05, "loss": 0.3939269304275513, "memory(GiB)": 70.5, "step": 49205, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.450775 }, { "epoch": 2.108307270468275, "grad_norm": 2.4364027976989746, "learning_rate": 6.218154470825775e-05, "loss": 0.38640332221984863, "memory(GiB)": 70.5, "step": 49210, "token_acc": 0.9401993355481728, "train_speed(iter/s)": 1.450769 }, { "epoch": 2.1085214857975236, "grad_norm": 3.762741804122925, "learning_rate": 6.217501760623889e-05, "loss": 0.2219792604446411, "memory(GiB)": 70.5, "step": 49215, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.450768 }, { "epoch": 2.1087357011267724, "grad_norm": 5.926715850830078, "learning_rate": 6.216849028365815e-05, "loss": 0.4676963806152344, "memory(GiB)": 70.5, "step": 49220, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.450769 }, { "epoch": 2.1089499164560217, "grad_norm": 1.4025437831878662, "learning_rate": 6.216196274063379e-05, "loss": 0.2683104038238525, "memory(GiB)": 70.5, "step": 49225, "token_acc": 0.9544072948328267, "train_speed(iter/s)": 1.450769 }, { "epoch": 2.1091641317852705, "grad_norm": 3.8019704818725586, "learning_rate": 6.215543497728407e-05, "loss": 0.14097288846969605, "memory(GiB)": 70.5, "step": 49230, "token_acc": 0.9637096774193549, "train_speed(iter/s)": 1.450771 }, { "epoch": 2.1093783471145193, "grad_norm": 0.6982013583183289, "learning_rate": 6.214890699372724e-05, "loss": 0.3265953540802002, "memory(GiB)": 70.5, "step": 49235, "token_acc": 0.9155313351498637, "train_speed(iter/s)": 1.450776 }, { "epoch": 2.1095925624437686, "grad_norm": 3.796754837036133, "learning_rate": 6.214237879008157e-05, "loss": 0.31077089309692385, "memory(GiB)": 70.5, "step": 49240, "token_acc": 0.942652329749104, "train_speed(iter/s)": 1.45077 }, { "epoch": 2.1098067777730174, "grad_norm": 5.054220676422119, "learning_rate": 6.213585036646531e-05, "loss": 0.44664630889892576, "memory(GiB)": 70.5, "step": 49245, "token_acc": 0.8896321070234113, "train_speed(iter/s)": 1.450769 }, { "epoch": 2.110020993102266, "grad_norm": 4.4333672523498535, "learning_rate": 6.212932172299674e-05, "loss": 0.583078670501709, "memory(GiB)": 70.5, "step": 49250, "token_acc": 0.8953068592057761, "train_speed(iter/s)": 1.450769 }, { "epoch": 2.1102352084315155, "grad_norm": 3.325958728790283, "learning_rate": 6.212279285979412e-05, "loss": 0.26731574535369873, "memory(GiB)": 70.5, "step": 49255, "token_acc": 0.9571428571428572, "train_speed(iter/s)": 1.450791 }, { "epoch": 2.1104494237607643, "grad_norm": 2.0511932373046875, "learning_rate": 6.211626377697575e-05, "loss": 0.28163230419158936, "memory(GiB)": 70.5, "step": 49260, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.45079 }, { "epoch": 2.110663639090013, "grad_norm": 0.025639796629548073, "learning_rate": 6.210973447465988e-05, "loss": 0.26322696208953855, "memory(GiB)": 70.5, "step": 49265, "token_acc": 0.9184952978056427, "train_speed(iter/s)": 1.450784 }, { "epoch": 2.1108778544192623, "grad_norm": 0.5256868600845337, "learning_rate": 6.210320495296484e-05, "loss": 0.36933577060699463, "memory(GiB)": 70.5, "step": 49270, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.45079 }, { "epoch": 2.111092069748511, "grad_norm": 5.947211265563965, "learning_rate": 6.209667521200886e-05, "loss": 0.4145852565765381, "memory(GiB)": 70.5, "step": 49275, "token_acc": 0.92, "train_speed(iter/s)": 1.450785 }, { "epoch": 2.11130628507776, "grad_norm": 4.671642780303955, "learning_rate": 6.209014525191025e-05, "loss": 0.36086416244506836, "memory(GiB)": 70.5, "step": 49280, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.450794 }, { "epoch": 2.1115205004070092, "grad_norm": 2.805104970932007, "learning_rate": 6.208361507278735e-05, "loss": 0.12375988960266113, "memory(GiB)": 70.5, "step": 49285, "token_acc": 0.9695817490494296, "train_speed(iter/s)": 1.450797 }, { "epoch": 2.111734715736258, "grad_norm": 0.48893988132476807, "learning_rate": 6.207708467475842e-05, "loss": 0.13885644674301148, "memory(GiB)": 70.5, "step": 49290, "token_acc": 0.9611650485436893, "train_speed(iter/s)": 1.450801 }, { "epoch": 2.111948931065507, "grad_norm": 3.574618101119995, "learning_rate": 6.207055405794176e-05, "loss": 0.2778846263885498, "memory(GiB)": 70.5, "step": 49295, "token_acc": 0.9375, "train_speed(iter/s)": 1.450802 }, { "epoch": 2.112163146394756, "grad_norm": 4.680306911468506, "learning_rate": 6.20640232224557e-05, "loss": 0.2740614891052246, "memory(GiB)": 70.5, "step": 49300, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.450803 }, { "epoch": 2.112377361724005, "grad_norm": 4.060408592224121, "learning_rate": 6.205749216841855e-05, "loss": 0.44332351684570315, "memory(GiB)": 70.5, "step": 49305, "token_acc": 0.9, "train_speed(iter/s)": 1.450806 }, { "epoch": 2.1125915770532537, "grad_norm": 5.2356977462768555, "learning_rate": 6.20509608959486e-05, "loss": 0.5358894348144532, "memory(GiB)": 70.5, "step": 49310, "token_acc": 0.8729508196721312, "train_speed(iter/s)": 1.450819 }, { "epoch": 2.112805792382503, "grad_norm": 3.1814210414886475, "learning_rate": 6.20444294051642e-05, "loss": 0.3024327754974365, "memory(GiB)": 70.5, "step": 49315, "token_acc": 0.9356913183279743, "train_speed(iter/s)": 1.450819 }, { "epoch": 2.113020007711752, "grad_norm": 7.450423240661621, "learning_rate": 6.203789769618365e-05, "loss": 0.5361366271972656, "memory(GiB)": 70.5, "step": 49320, "token_acc": 0.8700361010830325, "train_speed(iter/s)": 1.450803 }, { "epoch": 2.1132342230410006, "grad_norm": 3.1397321224212646, "learning_rate": 6.203136576912529e-05, "loss": 0.16687397956848143, "memory(GiB)": 70.5, "step": 49325, "token_acc": 0.9602649006622517, "train_speed(iter/s)": 1.45082 }, { "epoch": 2.11344843837025, "grad_norm": 5.941906452178955, "learning_rate": 6.202483362410748e-05, "loss": 0.4995596885681152, "memory(GiB)": 70.5, "step": 49330, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.450823 }, { "epoch": 2.1136626536994987, "grad_norm": 4.761825084686279, "learning_rate": 6.20183012612485e-05, "loss": 0.23645172119140626, "memory(GiB)": 70.5, "step": 49335, "token_acc": 0.9540636042402827, "train_speed(iter/s)": 1.450818 }, { "epoch": 2.1138768690287475, "grad_norm": 5.811190128326416, "learning_rate": 6.201176868066674e-05, "loss": 0.49507961273193357, "memory(GiB)": 70.5, "step": 49340, "token_acc": 0.9013605442176871, "train_speed(iter/s)": 1.450818 }, { "epoch": 2.1140910843579968, "grad_norm": 11.99102783203125, "learning_rate": 6.20052358824805e-05, "loss": 0.3558556318283081, "memory(GiB)": 70.5, "step": 49345, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.450825 }, { "epoch": 2.1143052996872456, "grad_norm": 3.771183967590332, "learning_rate": 6.199870286680817e-05, "loss": 0.2924570322036743, "memory(GiB)": 70.5, "step": 49350, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.450821 }, { "epoch": 2.1145195150164944, "grad_norm": 4.508339881896973, "learning_rate": 6.199216963376806e-05, "loss": 0.3020374536514282, "memory(GiB)": 70.5, "step": 49355, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.450823 }, { "epoch": 2.1147337303457436, "grad_norm": 1.5904834270477295, "learning_rate": 6.198563618347857e-05, "loss": 0.09429430961608887, "memory(GiB)": 70.5, "step": 49360, "token_acc": 0.9841897233201581, "train_speed(iter/s)": 1.450826 }, { "epoch": 2.1149479456749924, "grad_norm": 7.241287708282471, "learning_rate": 6.197910251605803e-05, "loss": 0.3829049825668335, "memory(GiB)": 70.5, "step": 49365, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.450824 }, { "epoch": 2.1151621610042413, "grad_norm": 1.6868982315063477, "learning_rate": 6.19725686316248e-05, "loss": 0.12404441833496094, "memory(GiB)": 70.5, "step": 49370, "token_acc": 0.9525691699604744, "train_speed(iter/s)": 1.450827 }, { "epoch": 2.1153763763334905, "grad_norm": 4.050381660461426, "learning_rate": 6.196603453029728e-05, "loss": 0.5886729717254638, "memory(GiB)": 70.5, "step": 49375, "token_acc": 0.884272997032641, "train_speed(iter/s)": 1.450836 }, { "epoch": 2.1155905916627393, "grad_norm": 0.4246578514575958, "learning_rate": 6.19595002121938e-05, "loss": 0.29735569953918456, "memory(GiB)": 70.5, "step": 49380, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.45084 }, { "epoch": 2.115804806991988, "grad_norm": 3.877185106277466, "learning_rate": 6.195296567743277e-05, "loss": 0.36621870994567873, "memory(GiB)": 70.5, "step": 49385, "token_acc": 0.8853754940711462, "train_speed(iter/s)": 1.450848 }, { "epoch": 2.1160190223212374, "grad_norm": 2.360340118408203, "learning_rate": 6.194643092613254e-05, "loss": 0.3847385883331299, "memory(GiB)": 70.5, "step": 49390, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.450863 }, { "epoch": 2.116233237650486, "grad_norm": 2.963496446609497, "learning_rate": 6.193989595841151e-05, "loss": 0.26360931396484377, "memory(GiB)": 70.5, "step": 49395, "token_acc": 0.95, "train_speed(iter/s)": 1.450866 }, { "epoch": 2.116447452979735, "grad_norm": 2.149461030960083, "learning_rate": 6.193336077438807e-05, "loss": 0.6646921157836914, "memory(GiB)": 70.5, "step": 49400, "token_acc": 0.8645418326693227, "train_speed(iter/s)": 1.450857 }, { "epoch": 2.1166616683089843, "grad_norm": 3.254155397415161, "learning_rate": 6.192682537418061e-05, "loss": 0.35159454345703123, "memory(GiB)": 70.5, "step": 49405, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.450854 }, { "epoch": 2.116875883638233, "grad_norm": 2.709575891494751, "learning_rate": 6.19202897579075e-05, "loss": 0.5172066211700439, "memory(GiB)": 70.5, "step": 49410, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.45086 }, { "epoch": 2.117090098967482, "grad_norm": 2.927258014678955, "learning_rate": 6.191375392568718e-05, "loss": 0.4656976699829102, "memory(GiB)": 70.5, "step": 49415, "token_acc": 0.889261744966443, "train_speed(iter/s)": 1.450863 }, { "epoch": 2.117304314296731, "grad_norm": 2.9905354976654053, "learning_rate": 6.190721787763801e-05, "loss": 0.4352055549621582, "memory(GiB)": 70.5, "step": 49420, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.450859 }, { "epoch": 2.11751852962598, "grad_norm": 2.0383543968200684, "learning_rate": 6.190068161387844e-05, "loss": 0.46997575759887694, "memory(GiB)": 70.5, "step": 49425, "token_acc": 0.9073569482288828, "train_speed(iter/s)": 1.450852 }, { "epoch": 2.117732744955229, "grad_norm": 1.755405306816101, "learning_rate": 6.189414513452685e-05, "loss": 0.1070871353149414, "memory(GiB)": 70.5, "step": 49430, "token_acc": 0.971830985915493, "train_speed(iter/s)": 1.450854 }, { "epoch": 2.117946960284478, "grad_norm": 4.641988754272461, "learning_rate": 6.188760843970166e-05, "loss": 0.5504646301269531, "memory(GiB)": 70.5, "step": 49435, "token_acc": 0.8746177370030581, "train_speed(iter/s)": 1.450869 }, { "epoch": 2.118161175613727, "grad_norm": 5.426955223083496, "learning_rate": 6.188107152952129e-05, "loss": 0.5973721027374268, "memory(GiB)": 70.5, "step": 49440, "token_acc": 0.9026845637583892, "train_speed(iter/s)": 1.450863 }, { "epoch": 2.1183753909429757, "grad_norm": 0.3072928190231323, "learning_rate": 6.187453440410418e-05, "loss": 0.419842004776001, "memory(GiB)": 70.5, "step": 49445, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.450876 }, { "epoch": 2.118589606272225, "grad_norm": 3.182579755783081, "learning_rate": 6.186799706356872e-05, "loss": 0.30778188705444337, "memory(GiB)": 70.5, "step": 49450, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.450872 }, { "epoch": 2.1188038216014737, "grad_norm": 5.289299964904785, "learning_rate": 6.186145950803337e-05, "loss": 0.518027400970459, "memory(GiB)": 70.5, "step": 49455, "token_acc": 0.8934707903780069, "train_speed(iter/s)": 1.450876 }, { "epoch": 2.1190180369307225, "grad_norm": 2.9008688926696777, "learning_rate": 6.185492173761655e-05, "loss": 0.431547737121582, "memory(GiB)": 70.5, "step": 49460, "token_acc": 0.9054545454545454, "train_speed(iter/s)": 1.45089 }, { "epoch": 2.119232252259972, "grad_norm": 2.488769292831421, "learning_rate": 6.184838375243671e-05, "loss": 0.20255160331726074, "memory(GiB)": 70.5, "step": 49465, "token_acc": 0.9554794520547946, "train_speed(iter/s)": 1.450887 }, { "epoch": 2.1194464675892206, "grad_norm": 2.8858227729797363, "learning_rate": 6.184184555261227e-05, "loss": 0.3390321731567383, "memory(GiB)": 70.5, "step": 49470, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.450877 }, { "epoch": 2.11966068291847, "grad_norm": 2.563291549682617, "learning_rate": 6.18353071382617e-05, "loss": 0.2687685966491699, "memory(GiB)": 70.5, "step": 49475, "token_acc": 0.9360902255639098, "train_speed(iter/s)": 1.450883 }, { "epoch": 2.1198748982477187, "grad_norm": 3.460632085800171, "learning_rate": 6.182876850950344e-05, "loss": 0.6237831592559815, "memory(GiB)": 70.5, "step": 49480, "token_acc": 0.8772563176895307, "train_speed(iter/s)": 1.450884 }, { "epoch": 2.1200891135769675, "grad_norm": 4.3401618003845215, "learning_rate": 6.182222966645593e-05, "loss": 0.3708014011383057, "memory(GiB)": 70.5, "step": 49485, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.450888 }, { "epoch": 2.1203033289062168, "grad_norm": 3.2084524631500244, "learning_rate": 6.181569060923765e-05, "loss": 0.5494359970092774, "memory(GiB)": 70.5, "step": 49490, "token_acc": 0.8904593639575972, "train_speed(iter/s)": 1.450905 }, { "epoch": 2.1205175442354656, "grad_norm": 1.8226737976074219, "learning_rate": 6.180915133796705e-05, "loss": 0.32672724723815916, "memory(GiB)": 70.5, "step": 49495, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.450907 }, { "epoch": 2.1207317595647144, "grad_norm": 2.3241984844207764, "learning_rate": 6.180261185276259e-05, "loss": 0.34798550605773926, "memory(GiB)": 70.5, "step": 49500, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.450915 }, { "epoch": 2.1207317595647144, "eval_loss": 2.2960877418518066, "eval_runtime": 13.0357, "eval_samples_per_second": 7.671, "eval_steps_per_second": 7.671, "eval_token_acc": 0.4722662440570523, "step": 49500 }, { "epoch": 2.1209459748939636, "grad_norm": 4.325350284576416, "learning_rate": 6.179607215374274e-05, "loss": 0.40251574516296384, "memory(GiB)": 70.5, "step": 49505, "token_acc": 0.6266804550155118, "train_speed(iter/s)": 1.450312 }, { "epoch": 2.1211601902232124, "grad_norm": 1.1057504415512085, "learning_rate": 6.178953224102599e-05, "loss": 0.30783216953277587, "memory(GiB)": 70.5, "step": 49510, "token_acc": 0.946969696969697, "train_speed(iter/s)": 1.450335 }, { "epoch": 2.1213744055524613, "grad_norm": 3.5966548919677734, "learning_rate": 6.178299211473081e-05, "loss": 0.34077014923095705, "memory(GiB)": 70.5, "step": 49515, "token_acc": 0.9337748344370861, "train_speed(iter/s)": 1.450338 }, { "epoch": 2.1215886208817105, "grad_norm": 3.884377956390381, "learning_rate": 6.177645177497566e-05, "loss": 0.4154163360595703, "memory(GiB)": 70.5, "step": 49520, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.450336 }, { "epoch": 2.1218028362109593, "grad_norm": 2.702943801879883, "learning_rate": 6.176991122187904e-05, "loss": 0.4773528575897217, "memory(GiB)": 70.5, "step": 49525, "token_acc": 0.8951048951048951, "train_speed(iter/s)": 1.450333 }, { "epoch": 2.122017051540208, "grad_norm": 4.920912265777588, "learning_rate": 6.176337045555944e-05, "loss": 0.4744349479675293, "memory(GiB)": 70.5, "step": 49530, "token_acc": 0.9006410256410257, "train_speed(iter/s)": 1.450333 }, { "epoch": 2.1222312668694574, "grad_norm": 1.2741467952728271, "learning_rate": 6.175682947613534e-05, "loss": 0.11225640773773193, "memory(GiB)": 70.5, "step": 49535, "token_acc": 0.9638157894736842, "train_speed(iter/s)": 1.450339 }, { "epoch": 2.122445482198706, "grad_norm": 4.54136323928833, "learning_rate": 6.175028828372527e-05, "loss": 0.33525924682617186, "memory(GiB)": 70.5, "step": 49540, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.45035 }, { "epoch": 2.122659697527955, "grad_norm": 3.0000429153442383, "learning_rate": 6.174374687844769e-05, "loss": 0.37850222587585447, "memory(GiB)": 70.5, "step": 49545, "token_acc": 0.9045801526717557, "train_speed(iter/s)": 1.45037 }, { "epoch": 2.1228739128572043, "grad_norm": 3.583761692047119, "learning_rate": 6.173720526042112e-05, "loss": 0.3380183219909668, "memory(GiB)": 70.5, "step": 49550, "token_acc": 0.9145569620253164, "train_speed(iter/s)": 1.450366 }, { "epoch": 2.123088128186453, "grad_norm": 7.940977096557617, "learning_rate": 6.173066342976405e-05, "loss": 0.2847935438156128, "memory(GiB)": 70.5, "step": 49555, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.450379 }, { "epoch": 2.123302343515702, "grad_norm": 3.903062105178833, "learning_rate": 6.172412138659504e-05, "loss": 0.42363834381103516, "memory(GiB)": 70.5, "step": 49560, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.450374 }, { "epoch": 2.123516558844951, "grad_norm": 2.340092658996582, "learning_rate": 6.171757913103255e-05, "loss": 0.6717251300811767, "memory(GiB)": 70.5, "step": 49565, "token_acc": 0.8818681318681318, "train_speed(iter/s)": 1.450381 }, { "epoch": 2.1237307741742, "grad_norm": 2.489624500274658, "learning_rate": 6.171103666319514e-05, "loss": 0.41077728271484376, "memory(GiB)": 70.5, "step": 49570, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.4504 }, { "epoch": 2.123944989503449, "grad_norm": 4.428746223449707, "learning_rate": 6.17044939832013e-05, "loss": 0.1687699317932129, "memory(GiB)": 70.5, "step": 49575, "token_acc": 0.9781021897810219, "train_speed(iter/s)": 1.4504 }, { "epoch": 2.124159204832698, "grad_norm": 4.744144916534424, "learning_rate": 6.169795109116957e-05, "loss": 0.6304924964904786, "memory(GiB)": 70.5, "step": 49580, "token_acc": 0.8489932885906041, "train_speed(iter/s)": 1.450401 }, { "epoch": 2.124373420161947, "grad_norm": 1.1366310119628906, "learning_rate": 6.169140798721847e-05, "loss": 0.2977325916290283, "memory(GiB)": 70.5, "step": 49585, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.450417 }, { "epoch": 2.1245876354911957, "grad_norm": 3.717878818511963, "learning_rate": 6.168486467146658e-05, "loss": 0.4422417640686035, "memory(GiB)": 70.5, "step": 49590, "token_acc": 0.9122340425531915, "train_speed(iter/s)": 1.450402 }, { "epoch": 2.124801850820445, "grad_norm": 2.0847246646881104, "learning_rate": 6.167832114403238e-05, "loss": 0.43256196975708006, "memory(GiB)": 70.5, "step": 49595, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.450401 }, { "epoch": 2.1250160661496937, "grad_norm": 3.5232222080230713, "learning_rate": 6.167177740503444e-05, "loss": 0.5076728343963623, "memory(GiB)": 70.5, "step": 49600, "token_acc": 0.889967637540453, "train_speed(iter/s)": 1.450402 }, { "epoch": 2.1252302814789426, "grad_norm": 1.484273910522461, "learning_rate": 6.166523345459132e-05, "loss": 0.2881158828735352, "memory(GiB)": 70.5, "step": 49605, "token_acc": 0.9413680781758957, "train_speed(iter/s)": 1.450403 }, { "epoch": 2.125444496808192, "grad_norm": 8.430697441101074, "learning_rate": 6.165868929282155e-05, "loss": 0.40240015983581545, "memory(GiB)": 70.5, "step": 49610, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.450404 }, { "epoch": 2.1256587121374406, "grad_norm": 2.2571117877960205, "learning_rate": 6.165214491984367e-05, "loss": 0.31809914112091064, "memory(GiB)": 70.5, "step": 49615, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.45042 }, { "epoch": 2.1258729274666894, "grad_norm": 0.2484825998544693, "learning_rate": 6.164560033577626e-05, "loss": 0.18771156072616577, "memory(GiB)": 70.5, "step": 49620, "token_acc": 0.9442724458204335, "train_speed(iter/s)": 1.450422 }, { "epoch": 2.1260871427959387, "grad_norm": 0.23814237117767334, "learning_rate": 6.163905554073787e-05, "loss": 0.35030326843261717, "memory(GiB)": 70.5, "step": 49625, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.450429 }, { "epoch": 2.1263013581251875, "grad_norm": 2.684218645095825, "learning_rate": 6.16325105348471e-05, "loss": 0.4248077392578125, "memory(GiB)": 70.5, "step": 49630, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.450442 }, { "epoch": 2.1265155734544363, "grad_norm": 5.748961448669434, "learning_rate": 6.162596531822247e-05, "loss": 0.49936494827270506, "memory(GiB)": 70.5, "step": 49635, "token_acc": 0.8819672131147541, "train_speed(iter/s)": 1.450453 }, { "epoch": 2.1267297887836856, "grad_norm": 3.2087249755859375, "learning_rate": 6.161941989098256e-05, "loss": 0.31483774185180663, "memory(GiB)": 70.5, "step": 49640, "token_acc": 0.924187725631769, "train_speed(iter/s)": 1.450455 }, { "epoch": 2.1269440041129344, "grad_norm": 1.9438050985336304, "learning_rate": 6.161287425324597e-05, "loss": 0.3236798524856567, "memory(GiB)": 70.5, "step": 49645, "token_acc": 0.9456869009584664, "train_speed(iter/s)": 1.450461 }, { "epoch": 2.127158219442183, "grad_norm": 5.951128005981445, "learning_rate": 6.160632840513127e-05, "loss": 0.157762348651886, "memory(GiB)": 70.5, "step": 49650, "token_acc": 0.9581881533101045, "train_speed(iter/s)": 1.450462 }, { "epoch": 2.1273724347714325, "grad_norm": 8.297576904296875, "learning_rate": 6.159978234675704e-05, "loss": 0.44900999069213865, "memory(GiB)": 70.5, "step": 49655, "token_acc": 0.9065743944636678, "train_speed(iter/s)": 1.450467 }, { "epoch": 2.1275866501006813, "grad_norm": 0.8307307362556458, "learning_rate": 6.159323607824188e-05, "loss": 0.40829854011535643, "memory(GiB)": 70.5, "step": 49660, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.450462 }, { "epoch": 2.12780086542993, "grad_norm": 4.473964214324951, "learning_rate": 6.158668959970437e-05, "loss": 0.5321093559265136, "memory(GiB)": 70.5, "step": 49665, "token_acc": 0.8905109489051095, "train_speed(iter/s)": 1.450482 }, { "epoch": 2.1280150807591793, "grad_norm": 2.947810173034668, "learning_rate": 6.158014291126311e-05, "loss": 0.18451964855194092, "memory(GiB)": 70.5, "step": 49670, "token_acc": 0.9552238805970149, "train_speed(iter/s)": 1.45048 }, { "epoch": 2.128229296088428, "grad_norm": 6.500638961791992, "learning_rate": 6.15735960130367e-05, "loss": 0.294842791557312, "memory(GiB)": 70.5, "step": 49675, "token_acc": 0.9387096774193548, "train_speed(iter/s)": 1.450478 }, { "epoch": 2.128443511417677, "grad_norm": 3.4711692333221436, "learning_rate": 6.156704890514372e-05, "loss": 0.33528857231140136, "memory(GiB)": 70.5, "step": 49680, "token_acc": 0.9049079754601227, "train_speed(iter/s)": 1.450479 }, { "epoch": 2.128657726746926, "grad_norm": 8.764448165893555, "learning_rate": 6.156050158770282e-05, "loss": 0.3215871095657349, "memory(GiB)": 70.5, "step": 49685, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.450479 }, { "epoch": 2.128871942076175, "grad_norm": 3.8302125930786133, "learning_rate": 6.155395406083257e-05, "loss": 0.4725905418395996, "memory(GiB)": 70.5, "step": 49690, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.450488 }, { "epoch": 2.129086157405424, "grad_norm": 3.259497880935669, "learning_rate": 6.154740632465162e-05, "loss": 0.3306746482849121, "memory(GiB)": 70.5, "step": 49695, "token_acc": 0.9485530546623794, "train_speed(iter/s)": 1.450491 }, { "epoch": 2.129300372734673, "grad_norm": 3.6458215713500977, "learning_rate": 6.154085837927857e-05, "loss": 0.3604025363922119, "memory(GiB)": 70.5, "step": 49700, "token_acc": 0.9042145593869731, "train_speed(iter/s)": 1.450506 }, { "epoch": 2.129514588063922, "grad_norm": 2.294743061065674, "learning_rate": 6.153431022483205e-05, "loss": 0.12956466674804687, "memory(GiB)": 70.5, "step": 49705, "token_acc": 0.9742647058823529, "train_speed(iter/s)": 1.450527 }, { "epoch": 2.1297288033931707, "grad_norm": 3.667027711868286, "learning_rate": 6.152776186143067e-05, "loss": 0.4394217491149902, "memory(GiB)": 70.5, "step": 49710, "token_acc": 0.9153005464480874, "train_speed(iter/s)": 1.450526 }, { "epoch": 2.12994301872242, "grad_norm": 3.109819173812866, "learning_rate": 6.152121328919307e-05, "loss": 0.41442456245422366, "memory(GiB)": 70.5, "step": 49715, "token_acc": 0.9068100358422939, "train_speed(iter/s)": 1.450535 }, { "epoch": 2.130157234051669, "grad_norm": 0.05648045986890793, "learning_rate": 6.15146645082379e-05, "loss": 0.27598488330841064, "memory(GiB)": 70.5, "step": 49720, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.450538 }, { "epoch": 2.1303714493809176, "grad_norm": 4.3352179527282715, "learning_rate": 6.150811551868377e-05, "loss": 0.35208017826080323, "memory(GiB)": 70.5, "step": 49725, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.450534 }, { "epoch": 2.130585664710167, "grad_norm": 3.8153553009033203, "learning_rate": 6.15015663206493e-05, "loss": 0.6558331489562989, "memory(GiB)": 70.5, "step": 49730, "token_acc": 0.8638132295719845, "train_speed(iter/s)": 1.450552 }, { "epoch": 2.1307998800394157, "grad_norm": 0.644271969795227, "learning_rate": 6.149501691425321e-05, "loss": 0.29440219402313234, "memory(GiB)": 70.5, "step": 49735, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.450544 }, { "epoch": 2.1310140953686645, "grad_norm": 2.776294469833374, "learning_rate": 6.148846729961409e-05, "loss": 0.38455400466918943, "memory(GiB)": 70.5, "step": 49740, "token_acc": 0.9293286219081273, "train_speed(iter/s)": 1.450547 }, { "epoch": 2.1312283106979137, "grad_norm": 2.728818893432617, "learning_rate": 6.148191747685061e-05, "loss": 0.34362890720367434, "memory(GiB)": 70.5, "step": 49745, "token_acc": 0.9177631578947368, "train_speed(iter/s)": 1.450548 }, { "epoch": 2.1314425260271626, "grad_norm": 1.6108945608139038, "learning_rate": 6.147536744608143e-05, "loss": 0.4213698387145996, "memory(GiB)": 70.5, "step": 49750, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.450548 }, { "epoch": 2.1316567413564114, "grad_norm": 6.285868167877197, "learning_rate": 6.146881720742519e-05, "loss": 0.6033853530883789, "memory(GiB)": 70.5, "step": 49755, "token_acc": 0.8854961832061069, "train_speed(iter/s)": 1.450549 }, { "epoch": 2.1318709566856606, "grad_norm": 3.173476457595825, "learning_rate": 6.146226676100058e-05, "loss": 0.28673782348632815, "memory(GiB)": 70.5, "step": 49760, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.450551 }, { "epoch": 2.1320851720149094, "grad_norm": 0.7446104884147644, "learning_rate": 6.145571610692624e-05, "loss": 0.30190110206604004, "memory(GiB)": 70.5, "step": 49765, "token_acc": 0.9442231075697212, "train_speed(iter/s)": 1.450551 }, { "epoch": 2.1322993873441582, "grad_norm": 5.083645820617676, "learning_rate": 6.144916524532086e-05, "loss": 0.34647550582885744, "memory(GiB)": 70.5, "step": 49770, "token_acc": 0.9349593495934959, "train_speed(iter/s)": 1.450552 }, { "epoch": 2.1325136026734075, "grad_norm": 2.013232707977295, "learning_rate": 6.144261417630313e-05, "loss": 0.5436834812164306, "memory(GiB)": 70.5, "step": 49775, "token_acc": 0.8870431893687708, "train_speed(iter/s)": 1.450567 }, { "epoch": 2.1327278180026563, "grad_norm": 2.9616942405700684, "learning_rate": 6.143606289999169e-05, "loss": 0.3237110137939453, "memory(GiB)": 70.5, "step": 49780, "token_acc": 0.9386973180076629, "train_speed(iter/s)": 1.450571 }, { "epoch": 2.132942033331905, "grad_norm": 2.7611846923828125, "learning_rate": 6.142951141650527e-05, "loss": 0.26019716262817383, "memory(GiB)": 70.5, "step": 49785, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.450567 }, { "epoch": 2.1331562486611544, "grad_norm": 3.6258842945098877, "learning_rate": 6.14229597259625e-05, "loss": 0.30288896560668943, "memory(GiB)": 70.5, "step": 49790, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.450577 }, { "epoch": 2.133370463990403, "grad_norm": 4.684276580810547, "learning_rate": 6.141640782848211e-05, "loss": 0.42064828872680665, "memory(GiB)": 70.5, "step": 49795, "token_acc": 0.9233870967741935, "train_speed(iter/s)": 1.450591 }, { "epoch": 2.133584679319652, "grad_norm": 7.094675540924072, "learning_rate": 6.140985572418276e-05, "loss": 0.39852118492126465, "memory(GiB)": 70.5, "step": 49800, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.450602 }, { "epoch": 2.1337988946489013, "grad_norm": 2.075152635574341, "learning_rate": 6.14033034131832e-05, "loss": 0.3075856685638428, "memory(GiB)": 70.5, "step": 49805, "token_acc": 0.9218106995884774, "train_speed(iter/s)": 1.450623 }, { "epoch": 2.13401310997815, "grad_norm": 4.444478511810303, "learning_rate": 6.13967508956021e-05, "loss": 0.4699970245361328, "memory(GiB)": 70.5, "step": 49810, "token_acc": 0.8986301369863013, "train_speed(iter/s)": 1.450626 }, { "epoch": 2.134227325307399, "grad_norm": 7.987437725067139, "learning_rate": 6.139019817155815e-05, "loss": 0.30823872089385984, "memory(GiB)": 70.5, "step": 49815, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.450632 }, { "epoch": 2.134441540636648, "grad_norm": 3.769148349761963, "learning_rate": 6.13836452411701e-05, "loss": 0.3444085121154785, "memory(GiB)": 70.5, "step": 49820, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.450637 }, { "epoch": 2.134655755965897, "grad_norm": 2.198981285095215, "learning_rate": 6.137709210455661e-05, "loss": 0.4709984302520752, "memory(GiB)": 70.5, "step": 49825, "token_acc": 0.903125, "train_speed(iter/s)": 1.450639 }, { "epoch": 2.1348699712951458, "grad_norm": 5.435634613037109, "learning_rate": 6.137053876183644e-05, "loss": 0.2413943290710449, "memory(GiB)": 70.5, "step": 49830, "token_acc": 0.94140625, "train_speed(iter/s)": 1.450643 }, { "epoch": 2.135084186624395, "grad_norm": 0.1659930944442749, "learning_rate": 6.136398521312829e-05, "loss": 0.2517061710357666, "memory(GiB)": 70.5, "step": 49835, "token_acc": 0.9456869009584664, "train_speed(iter/s)": 1.450649 }, { "epoch": 2.135298401953644, "grad_norm": 2.132169485092163, "learning_rate": 6.135743145855088e-05, "loss": 0.2360771417617798, "memory(GiB)": 70.5, "step": 49840, "token_acc": 0.9528985507246377, "train_speed(iter/s)": 1.450649 }, { "epoch": 2.1355126172828927, "grad_norm": 8.427508354187012, "learning_rate": 6.135087749822296e-05, "loss": 0.3232978343963623, "memory(GiB)": 70.5, "step": 49845, "token_acc": 0.934984520123839, "train_speed(iter/s)": 1.450648 }, { "epoch": 2.135726832612142, "grad_norm": 4.757858753204346, "learning_rate": 6.134432333226324e-05, "loss": 0.2235802173614502, "memory(GiB)": 70.5, "step": 49850, "token_acc": 0.958904109589041, "train_speed(iter/s)": 1.450651 }, { "epoch": 2.1359410479413907, "grad_norm": 1.272644281387329, "learning_rate": 6.133776896079045e-05, "loss": 0.4459801197052002, "memory(GiB)": 70.5, "step": 49855, "token_acc": 0.8819444444444444, "train_speed(iter/s)": 1.450646 }, { "epoch": 2.1361552632706395, "grad_norm": 3.683065414428711, "learning_rate": 6.133121438392336e-05, "loss": 0.2855182647705078, "memory(GiB)": 70.5, "step": 49860, "token_acc": 0.9490909090909091, "train_speed(iter/s)": 1.45065 }, { "epoch": 2.136369478599889, "grad_norm": 2.3154795169830322, "learning_rate": 6.132465960178069e-05, "loss": 0.3163213014602661, "memory(GiB)": 70.5, "step": 49865, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.45065 }, { "epoch": 2.1365836939291376, "grad_norm": 4.096757888793945, "learning_rate": 6.131810461448118e-05, "loss": 0.43628778457641604, "memory(GiB)": 70.5, "step": 49870, "token_acc": 0.9233870967741935, "train_speed(iter/s)": 1.450658 }, { "epoch": 2.1367979092583864, "grad_norm": 5.417862415313721, "learning_rate": 6.131154942214356e-05, "loss": 0.3959050178527832, "memory(GiB)": 70.5, "step": 49875, "token_acc": 0.9111969111969112, "train_speed(iter/s)": 1.450662 }, { "epoch": 2.1370121245876357, "grad_norm": 4.724027633666992, "learning_rate": 6.130499402488665e-05, "loss": 0.5276658058166503, "memory(GiB)": 70.5, "step": 49880, "token_acc": 0.8896551724137931, "train_speed(iter/s)": 1.450669 }, { "epoch": 2.1372263399168845, "grad_norm": 2.1320345401763916, "learning_rate": 6.129843842282915e-05, "loss": 0.15985093116760254, "memory(GiB)": 70.5, "step": 49885, "token_acc": 0.9804878048780488, "train_speed(iter/s)": 1.450673 }, { "epoch": 2.1374405552461333, "grad_norm": 2.6555964946746826, "learning_rate": 6.129188261608985e-05, "loss": 0.456959867477417, "memory(GiB)": 70.5, "step": 49890, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.450676 }, { "epoch": 2.1376547705753826, "grad_norm": 6.316608428955078, "learning_rate": 6.12853266047875e-05, "loss": 0.624326753616333, "memory(GiB)": 70.5, "step": 49895, "token_acc": 0.8746081504702194, "train_speed(iter/s)": 1.450686 }, { "epoch": 2.1378689859046314, "grad_norm": 3.6266541481018066, "learning_rate": 6.127877038904087e-05, "loss": 0.564473819732666, "memory(GiB)": 70.5, "step": 49900, "token_acc": 0.8721311475409836, "train_speed(iter/s)": 1.450679 }, { "epoch": 2.13808320123388, "grad_norm": 5.089602947235107, "learning_rate": 6.127221396896876e-05, "loss": 0.30765795707702637, "memory(GiB)": 70.5, "step": 49905, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.450668 }, { "epoch": 2.1382974165631294, "grad_norm": 4.840006351470947, "learning_rate": 6.126565734468987e-05, "loss": 0.6921665668487549, "memory(GiB)": 70.5, "step": 49910, "token_acc": 0.8591065292096219, "train_speed(iter/s)": 1.450669 }, { "epoch": 2.1385116318923783, "grad_norm": 2.265814781188965, "learning_rate": 6.125910051632305e-05, "loss": 0.4193385124206543, "memory(GiB)": 70.5, "step": 49915, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.450674 }, { "epoch": 2.138725847221627, "grad_norm": 2.9113152027130127, "learning_rate": 6.125254348398708e-05, "loss": 0.3833383798599243, "memory(GiB)": 70.5, "step": 49920, "token_acc": 0.9095744680851063, "train_speed(iter/s)": 1.450671 }, { "epoch": 2.1389400625508763, "grad_norm": 2.6722753047943115, "learning_rate": 6.124598624780071e-05, "loss": 0.25833561420440676, "memory(GiB)": 70.5, "step": 49925, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.45067 }, { "epoch": 2.139154277880125, "grad_norm": 3.146204948425293, "learning_rate": 6.123942880788276e-05, "loss": 0.296728515625, "memory(GiB)": 70.5, "step": 49930, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.450679 }, { "epoch": 2.139368493209374, "grad_norm": 3.2731542587280273, "learning_rate": 6.123287116435201e-05, "loss": 0.43311634063720705, "memory(GiB)": 70.5, "step": 49935, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.45069 }, { "epoch": 2.139582708538623, "grad_norm": 3.5629069805145264, "learning_rate": 6.122631331732726e-05, "loss": 0.3536886692047119, "memory(GiB)": 70.5, "step": 49940, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.450707 }, { "epoch": 2.139796923867872, "grad_norm": 2.4860095977783203, "learning_rate": 6.121975526692731e-05, "loss": 0.48342461585998536, "memory(GiB)": 70.5, "step": 49945, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.450705 }, { "epoch": 2.140011139197121, "grad_norm": 4.695987224578857, "learning_rate": 6.121319701327097e-05, "loss": 0.4782230377197266, "memory(GiB)": 70.5, "step": 49950, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.450715 }, { "epoch": 2.14022535452637, "grad_norm": 4.5059638023376465, "learning_rate": 6.120663855647706e-05, "loss": 0.28949048519134524, "memory(GiB)": 70.5, "step": 49955, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.450716 }, { "epoch": 2.140439569855619, "grad_norm": 1.1755454540252686, "learning_rate": 6.120007989666437e-05, "loss": 0.29462504386901855, "memory(GiB)": 70.5, "step": 49960, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.450727 }, { "epoch": 2.1406537851848677, "grad_norm": 3.8625333309173584, "learning_rate": 6.119352103395172e-05, "loss": 0.40517454147338866, "memory(GiB)": 70.5, "step": 49965, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.450722 }, { "epoch": 2.140868000514117, "grad_norm": 2.570239782333374, "learning_rate": 6.118696196845793e-05, "loss": 0.37235336303710936, "memory(GiB)": 70.5, "step": 49970, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.450718 }, { "epoch": 2.1410822158433658, "grad_norm": 4.393589973449707, "learning_rate": 6.118040270030185e-05, "loss": 0.5374437808990479, "memory(GiB)": 70.5, "step": 49975, "token_acc": 0.8916666666666667, "train_speed(iter/s)": 1.45074 }, { "epoch": 2.1412964311726146, "grad_norm": 2.950833797454834, "learning_rate": 6.117384322960228e-05, "loss": 0.36583311557769777, "memory(GiB)": 70.5, "step": 49980, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.450739 }, { "epoch": 2.141510646501864, "grad_norm": 1.6629612445831299, "learning_rate": 6.116728355647805e-05, "loss": 0.17534021139144898, "memory(GiB)": 70.5, "step": 49985, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.450744 }, { "epoch": 2.1417248618311127, "grad_norm": 1.0725805759429932, "learning_rate": 6.1160723681048e-05, "loss": 0.6880894660949707, "memory(GiB)": 70.5, "step": 49990, "token_acc": 0.875, "train_speed(iter/s)": 1.450749 }, { "epoch": 2.1419390771603615, "grad_norm": 4.480474472045898, "learning_rate": 6.115416360343099e-05, "loss": 0.3329556941986084, "memory(GiB)": 70.5, "step": 49995, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.450757 }, { "epoch": 2.1421532924896107, "grad_norm": 2.101500988006592, "learning_rate": 6.114760332374582e-05, "loss": 0.26920945644378663, "memory(GiB)": 70.5, "step": 50000, "token_acc": 0.9467680608365019, "train_speed(iter/s)": 1.450756 }, { "epoch": 2.1421532924896107, "eval_loss": 2.3717589378356934, "eval_runtime": 13.7066, "eval_samples_per_second": 7.296, "eval_steps_per_second": 7.296, "eval_token_acc": 0.4648876404494382, "step": 50000 }, { "epoch": 2.1423675078188595, "grad_norm": 2.5574636459350586, "learning_rate": 6.114104284211139e-05, "loss": 0.2740926265716553, "memory(GiB)": 70.5, "step": 50005, "token_acc": 0.6145733461169702, "train_speed(iter/s)": 1.450146 }, { "epoch": 2.1425817231481084, "grad_norm": 7.230663776397705, "learning_rate": 6.11344821586465e-05, "loss": 0.45656557083129884, "memory(GiB)": 70.5, "step": 50010, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.450147 }, { "epoch": 2.1427959384773576, "grad_norm": 7.3729634284973145, "learning_rate": 6.112792127347001e-05, "loss": 0.5310036182403565, "memory(GiB)": 70.5, "step": 50015, "token_acc": 0.8897338403041825, "train_speed(iter/s)": 1.450156 }, { "epoch": 2.1430101538066064, "grad_norm": 2.9889378547668457, "learning_rate": 6.112136018670079e-05, "loss": 0.2736427068710327, "memory(GiB)": 70.5, "step": 50020, "token_acc": 0.9363636363636364, "train_speed(iter/s)": 1.45015 }, { "epoch": 2.1432243691358552, "grad_norm": 8.200289726257324, "learning_rate": 6.111479889845772e-05, "loss": 0.6138160705566407, "memory(GiB)": 70.5, "step": 50025, "token_acc": 0.8582995951417004, "train_speed(iter/s)": 1.450156 }, { "epoch": 2.1434385844651045, "grad_norm": 2.813225746154785, "learning_rate": 6.110823740885962e-05, "loss": 0.36423320770263673, "memory(GiB)": 70.5, "step": 50030, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.450156 }, { "epoch": 2.1436527997943533, "grad_norm": 1.3578354120254517, "learning_rate": 6.110167571802538e-05, "loss": 0.2281052827835083, "memory(GiB)": 70.5, "step": 50035, "token_acc": 0.9413333333333334, "train_speed(iter/s)": 1.450168 }, { "epoch": 2.143867015123602, "grad_norm": 0.8087402582168579, "learning_rate": 6.109511382607388e-05, "loss": 0.2443930149078369, "memory(GiB)": 70.5, "step": 50040, "token_acc": 0.9482071713147411, "train_speed(iter/s)": 1.450172 }, { "epoch": 2.1440812304528514, "grad_norm": 5.215057373046875, "learning_rate": 6.108855173312397e-05, "loss": 0.34290294647216796, "memory(GiB)": 70.5, "step": 50045, "token_acc": 0.9276595744680851, "train_speed(iter/s)": 1.45019 }, { "epoch": 2.1442954457821, "grad_norm": 5.006034851074219, "learning_rate": 6.108198943929457e-05, "loss": 0.45378837585449217, "memory(GiB)": 70.5, "step": 50050, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.450191 }, { "epoch": 2.144509661111349, "grad_norm": 3.1352295875549316, "learning_rate": 6.107542694470452e-05, "loss": 0.19923439025878906, "memory(GiB)": 70.5, "step": 50055, "token_acc": 0.9568627450980393, "train_speed(iter/s)": 1.450198 }, { "epoch": 2.1447238764405983, "grad_norm": 2.1636769771575928, "learning_rate": 6.10688642494727e-05, "loss": 0.19793252944946288, "memory(GiB)": 70.5, "step": 50060, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.450214 }, { "epoch": 2.144938091769847, "grad_norm": 1.344519853591919, "learning_rate": 6.106230135371804e-05, "loss": 0.3176865816116333, "memory(GiB)": 70.5, "step": 50065, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.450209 }, { "epoch": 2.145152307099096, "grad_norm": 11.115313529968262, "learning_rate": 6.105573825755942e-05, "loss": 0.5455578804016114, "memory(GiB)": 70.5, "step": 50070, "token_acc": 0.890282131661442, "train_speed(iter/s)": 1.450201 }, { "epoch": 2.145366522428345, "grad_norm": 2.1019678115844727, "learning_rate": 6.104917496111574e-05, "loss": 0.2782400608062744, "memory(GiB)": 70.5, "step": 50075, "token_acc": 0.932258064516129, "train_speed(iter/s)": 1.450215 }, { "epoch": 2.145580737757594, "grad_norm": 2.2925798892974854, "learning_rate": 6.104261146450588e-05, "loss": 0.2624505519866943, "memory(GiB)": 70.5, "step": 50080, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.450218 }, { "epoch": 2.1457949530868428, "grad_norm": 2.5921385288238525, "learning_rate": 6.103604776784872e-05, "loss": 0.3656193733215332, "memory(GiB)": 70.5, "step": 50085, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.450217 }, { "epoch": 2.146009168416092, "grad_norm": 2.3794260025024414, "learning_rate": 6.102948387126325e-05, "loss": 0.3263404846191406, "memory(GiB)": 70.5, "step": 50090, "token_acc": 0.9245901639344263, "train_speed(iter/s)": 1.45021 }, { "epoch": 2.146223383745341, "grad_norm": 2.627988576889038, "learning_rate": 6.10229197748683e-05, "loss": 0.345902419090271, "memory(GiB)": 70.5, "step": 50095, "token_acc": 0.9105058365758755, "train_speed(iter/s)": 1.450215 }, { "epoch": 2.1464375990745896, "grad_norm": 1.015257477760315, "learning_rate": 6.101635547878285e-05, "loss": 0.24346086978912354, "memory(GiB)": 70.5, "step": 50100, "token_acc": 0.956386292834891, "train_speed(iter/s)": 1.450215 }, { "epoch": 2.146651814403839, "grad_norm": 1.776109218597412, "learning_rate": 6.100979098312576e-05, "loss": 0.26396956443786623, "memory(GiB)": 70.5, "step": 50105, "token_acc": 0.9476584022038568, "train_speed(iter/s)": 1.450223 }, { "epoch": 2.1468660297330877, "grad_norm": 4.139390468597412, "learning_rate": 6.100322628801599e-05, "loss": 0.37558326721191404, "memory(GiB)": 70.5, "step": 50110, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.45022 }, { "epoch": 2.1470802450623365, "grad_norm": 2.3416898250579834, "learning_rate": 6.0996661393572454e-05, "loss": 0.35666346549987793, "memory(GiB)": 70.5, "step": 50115, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.450225 }, { "epoch": 2.147294460391586, "grad_norm": 1.1910723447799683, "learning_rate": 6.099009629991408e-05, "loss": 0.14797109365463257, "memory(GiB)": 70.5, "step": 50120, "token_acc": 0.9550561797752809, "train_speed(iter/s)": 1.450228 }, { "epoch": 2.1475086757208346, "grad_norm": 5.049123764038086, "learning_rate": 6.098353100715981e-05, "loss": 0.6330349445343018, "memory(GiB)": 70.5, "step": 50125, "token_acc": 0.8598901098901099, "train_speed(iter/s)": 1.450245 }, { "epoch": 2.1477228910500834, "grad_norm": 1.7075624465942383, "learning_rate": 6.0976965515428554e-05, "loss": 0.4505485534667969, "memory(GiB)": 70.5, "step": 50130, "token_acc": 0.9053497942386831, "train_speed(iter/s)": 1.450261 }, { "epoch": 2.1479371063793327, "grad_norm": 7.914571285247803, "learning_rate": 6.097039982483927e-05, "loss": 0.2504589080810547, "memory(GiB)": 70.5, "step": 50135, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.450264 }, { "epoch": 2.1481513217085815, "grad_norm": 1.3338418006896973, "learning_rate": 6.0963833935510916e-05, "loss": 0.36746559143066404, "memory(GiB)": 70.5, "step": 50140, "token_acc": 0.9331210191082803, "train_speed(iter/s)": 1.450274 }, { "epoch": 2.1483655370378303, "grad_norm": 3.403475046157837, "learning_rate": 6.0957267847562414e-05, "loss": 0.25697941780090333, "memory(GiB)": 70.5, "step": 50145, "token_acc": 0.940809968847352, "train_speed(iter/s)": 1.450278 }, { "epoch": 2.1485797523670795, "grad_norm": 0.4617578983306885, "learning_rate": 6.095070156111274e-05, "loss": 0.347818660736084, "memory(GiB)": 70.5, "step": 50150, "token_acc": 0.901685393258427, "train_speed(iter/s)": 1.450279 }, { "epoch": 2.1487939676963284, "grad_norm": 3.2261416912078857, "learning_rate": 6.094413507628084e-05, "loss": 0.46534123420715334, "memory(GiB)": 70.5, "step": 50155, "token_acc": 0.8986486486486487, "train_speed(iter/s)": 1.450271 }, { "epoch": 2.149008183025577, "grad_norm": 2.675490140914917, "learning_rate": 6.093756839318565e-05, "loss": 0.2719964265823364, "memory(GiB)": 70.5, "step": 50160, "token_acc": 0.9571428571428572, "train_speed(iter/s)": 1.450295 }, { "epoch": 2.1492223983548264, "grad_norm": 3.8755810260772705, "learning_rate": 6.093100151194615e-05, "loss": 0.268280029296875, "memory(GiB)": 70.5, "step": 50165, "token_acc": 0.9429657794676806, "train_speed(iter/s)": 1.450295 }, { "epoch": 2.1494366136840752, "grad_norm": 4.4576334953308105, "learning_rate": 6.09244344326813e-05, "loss": 0.3412520408630371, "memory(GiB)": 70.5, "step": 50170, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.4503 }, { "epoch": 2.149650829013324, "grad_norm": 3.696401596069336, "learning_rate": 6.091786715551008e-05, "loss": 0.630208683013916, "memory(GiB)": 70.5, "step": 50175, "token_acc": 0.8541666666666666, "train_speed(iter/s)": 1.450289 }, { "epoch": 2.1498650443425733, "grad_norm": 4.5849504470825195, "learning_rate": 6.091129968055146e-05, "loss": 0.35932435989379885, "memory(GiB)": 70.5, "step": 50180, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.450283 }, { "epoch": 2.150079259671822, "grad_norm": 2.096135377883911, "learning_rate": 6.09047320079244e-05, "loss": 0.43087615966796877, "memory(GiB)": 70.5, "step": 50185, "token_acc": 0.9051094890510949, "train_speed(iter/s)": 1.450278 }, { "epoch": 2.150293475001071, "grad_norm": 3.0071117877960205, "learning_rate": 6.0898164137747893e-05, "loss": 0.4678149700164795, "memory(GiB)": 70.5, "step": 50190, "token_acc": 0.8861209964412812, "train_speed(iter/s)": 1.450276 }, { "epoch": 2.15050769033032, "grad_norm": 7.0533270835876465, "learning_rate": 6.089159607014092e-05, "loss": 0.42994346618652346, "memory(GiB)": 70.5, "step": 50195, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.45026 }, { "epoch": 2.150721905659569, "grad_norm": 3.9689533710479736, "learning_rate": 6.0885027805222484e-05, "loss": 0.19094167947769164, "memory(GiB)": 70.5, "step": 50200, "token_acc": 0.9611307420494699, "train_speed(iter/s)": 1.45026 }, { "epoch": 2.150936120988818, "grad_norm": 4.158168792724609, "learning_rate": 6.0878459343111517e-05, "loss": 0.3780506134033203, "memory(GiB)": 70.5, "step": 50205, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.450268 }, { "epoch": 2.151150336318067, "grad_norm": 2.5306191444396973, "learning_rate": 6.087189068392709e-05, "loss": 0.3125194311141968, "memory(GiB)": 70.5, "step": 50210, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.45027 }, { "epoch": 2.151364551647316, "grad_norm": 5.0024094581604, "learning_rate": 6.0865321827788154e-05, "loss": 0.5212277412414551, "memory(GiB)": 70.5, "step": 50215, "token_acc": 0.8966789667896679, "train_speed(iter/s)": 1.450264 }, { "epoch": 2.1515787669765647, "grad_norm": 5.486734390258789, "learning_rate": 6.085875277481372e-05, "loss": 0.5100666046142578, "memory(GiB)": 70.5, "step": 50220, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.450268 }, { "epoch": 2.151792982305814, "grad_norm": 4.444644451141357, "learning_rate": 6.08521835251228e-05, "loss": 0.48678112030029297, "memory(GiB)": 70.5, "step": 50225, "token_acc": 0.8954248366013072, "train_speed(iter/s)": 1.450264 }, { "epoch": 2.1520071976350628, "grad_norm": 1.1633846759796143, "learning_rate": 6.084561407883438e-05, "loss": 0.10787702798843384, "memory(GiB)": 70.5, "step": 50230, "token_acc": 0.9705882352941176, "train_speed(iter/s)": 1.45026 }, { "epoch": 2.1522214129643116, "grad_norm": 2.1764180660247803, "learning_rate": 6.083904443606751e-05, "loss": 0.06146761178970337, "memory(GiB)": 70.5, "step": 50235, "token_acc": 0.9882352941176471, "train_speed(iter/s)": 1.450265 }, { "epoch": 2.152435628293561, "grad_norm": 4.698911666870117, "learning_rate": 6.083247459694117e-05, "loss": 0.4610863208770752, "memory(GiB)": 70.5, "step": 50240, "token_acc": 0.8986928104575164, "train_speed(iter/s)": 1.450279 }, { "epoch": 2.1526498436228096, "grad_norm": 5.063790798187256, "learning_rate": 6.0825904561574374e-05, "loss": 0.29933457374572753, "memory(GiB)": 70.5, "step": 50245, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.450281 }, { "epoch": 2.1528640589520585, "grad_norm": 2.9078400135040283, "learning_rate": 6.081933433008617e-05, "loss": 0.23710918426513672, "memory(GiB)": 70.5, "step": 50250, "token_acc": 0.9563758389261745, "train_speed(iter/s)": 1.450285 }, { "epoch": 2.1530782742813077, "grad_norm": 1.5007801055908203, "learning_rate": 6.081276390259559e-05, "loss": 0.3692553281784058, "memory(GiB)": 70.5, "step": 50255, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.450287 }, { "epoch": 2.1532924896105565, "grad_norm": 2.1289985179901123, "learning_rate": 6.0806193279221634e-05, "loss": 0.3482968330383301, "memory(GiB)": 70.5, "step": 50260, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.450289 }, { "epoch": 2.1535067049398053, "grad_norm": 4.648497581481934, "learning_rate": 6.079962246008336e-05, "loss": 0.34279890060424806, "memory(GiB)": 70.5, "step": 50265, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.450287 }, { "epoch": 2.1537209202690546, "grad_norm": 4.447849273681641, "learning_rate": 6.07930514452998e-05, "loss": 0.27801928520202634, "memory(GiB)": 70.5, "step": 50270, "token_acc": 0.9421487603305785, "train_speed(iter/s)": 1.450294 }, { "epoch": 2.1539351355983034, "grad_norm": 5.763927459716797, "learning_rate": 6.0786480234989976e-05, "loss": 0.6751011371612549, "memory(GiB)": 70.5, "step": 50275, "token_acc": 0.8870292887029289, "train_speed(iter/s)": 1.45029 }, { "epoch": 2.154149350927552, "grad_norm": 2.758915424346924, "learning_rate": 6.0779908829272936e-05, "loss": 0.3597527265548706, "memory(GiB)": 70.5, "step": 50280, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.450289 }, { "epoch": 2.1543635662568015, "grad_norm": 3.9133641719818115, "learning_rate": 6.077333722826775e-05, "loss": 0.2569149971008301, "memory(GiB)": 70.5, "step": 50285, "token_acc": 0.9388646288209607, "train_speed(iter/s)": 1.450288 }, { "epoch": 2.1545777815860503, "grad_norm": 3.071854829788208, "learning_rate": 6.076676543209344e-05, "loss": 0.285294771194458, "memory(GiB)": 70.5, "step": 50290, "token_acc": 0.948170731707317, "train_speed(iter/s)": 1.450315 }, { "epoch": 2.154791996915299, "grad_norm": 6.542153358459473, "learning_rate": 6.07601934408691e-05, "loss": 0.5218152046203614, "memory(GiB)": 70.5, "step": 50295, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.450311 }, { "epoch": 2.1550062122445484, "grad_norm": 9.179766654968262, "learning_rate": 6.075362125471374e-05, "loss": 0.3385780334472656, "memory(GiB)": 70.5, "step": 50300, "token_acc": 0.9446640316205533, "train_speed(iter/s)": 1.450335 }, { "epoch": 2.155220427573797, "grad_norm": 2.9221243858337402, "learning_rate": 6.0747048873746446e-05, "loss": 0.31282360553741456, "memory(GiB)": 70.5, "step": 50305, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.45034 }, { "epoch": 2.155434642903046, "grad_norm": 5.462530136108398, "learning_rate": 6.074047629808629e-05, "loss": 0.5600555419921875, "memory(GiB)": 70.5, "step": 50310, "token_acc": 0.9060402684563759, "train_speed(iter/s)": 1.450341 }, { "epoch": 2.1556488582322952, "grad_norm": 1.5522381067276, "learning_rate": 6.073390352785232e-05, "loss": 0.6052928924560547, "memory(GiB)": 70.5, "step": 50315, "token_acc": 0.8528528528528528, "train_speed(iter/s)": 1.450357 }, { "epoch": 2.155863073561544, "grad_norm": 2.8583483695983887, "learning_rate": 6.0727330563163624e-05, "loss": 0.5615878105163574, "memory(GiB)": 70.5, "step": 50320, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.450359 }, { "epoch": 2.156077288890793, "grad_norm": 5.391454219818115, "learning_rate": 6.072075740413926e-05, "loss": 0.5705677509307862, "memory(GiB)": 70.5, "step": 50325, "token_acc": 0.8851963746223565, "train_speed(iter/s)": 1.450356 }, { "epoch": 2.156291504220042, "grad_norm": 2.405059576034546, "learning_rate": 6.071418405089834e-05, "loss": 0.38620116710662844, "memory(GiB)": 70.5, "step": 50330, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.45037 }, { "epoch": 2.156505719549291, "grad_norm": 1.9872689247131348, "learning_rate": 6.070761050355991e-05, "loss": 0.38354015350341797, "memory(GiB)": 70.5, "step": 50335, "token_acc": 0.9274193548387096, "train_speed(iter/s)": 1.45037 }, { "epoch": 2.1567199348785397, "grad_norm": 3.0163612365722656, "learning_rate": 6.070103676224308e-05, "loss": 0.4165000915527344, "memory(GiB)": 70.5, "step": 50340, "token_acc": 0.9026217228464419, "train_speed(iter/s)": 1.450359 }, { "epoch": 2.156934150207789, "grad_norm": 6.0543389320373535, "learning_rate": 6.069446282706692e-05, "loss": 0.2592761993408203, "memory(GiB)": 70.5, "step": 50345, "token_acc": 0.9486301369863014, "train_speed(iter/s)": 1.450358 }, { "epoch": 2.157148365537038, "grad_norm": 2.4032280445098877, "learning_rate": 6.068788869815054e-05, "loss": 0.27058985233306887, "memory(GiB)": 70.5, "step": 50350, "token_acc": 0.9448818897637795, "train_speed(iter/s)": 1.450369 }, { "epoch": 2.1573625808662866, "grad_norm": 1.7153714895248413, "learning_rate": 6.068131437561303e-05, "loss": 0.23852040767669677, "memory(GiB)": 70.5, "step": 50355, "token_acc": 0.9516728624535316, "train_speed(iter/s)": 1.450369 }, { "epoch": 2.157576796195536, "grad_norm": 2.694429636001587, "learning_rate": 6.067473985957349e-05, "loss": 0.2512845516204834, "memory(GiB)": 70.5, "step": 50360, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.45037 }, { "epoch": 2.1577910115247847, "grad_norm": 1.5465333461761475, "learning_rate": 6.066816515015101e-05, "loss": 0.2531964063644409, "memory(GiB)": 70.5, "step": 50365, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.450366 }, { "epoch": 2.1580052268540335, "grad_norm": 5.618694305419922, "learning_rate": 6.0661590247464736e-05, "loss": 0.37811832427978515, "memory(GiB)": 70.5, "step": 50370, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.450367 }, { "epoch": 2.1582194421832828, "grad_norm": 2.5729730129241943, "learning_rate": 6.065501515163374e-05, "loss": 0.4047966957092285, "memory(GiB)": 70.5, "step": 50375, "token_acc": 0.9010695187165776, "train_speed(iter/s)": 1.450369 }, { "epoch": 2.1584336575125316, "grad_norm": 6.683036804199219, "learning_rate": 6.064843986277715e-05, "loss": 0.6440689563751221, "memory(GiB)": 70.5, "step": 50380, "token_acc": 0.9045801526717557, "train_speed(iter/s)": 1.450384 }, { "epoch": 2.1586478728417804, "grad_norm": 4.114692687988281, "learning_rate": 6.064186438101409e-05, "loss": 0.5234070301055909, "memory(GiB)": 70.5, "step": 50385, "token_acc": 0.8736462093862816, "train_speed(iter/s)": 1.450393 }, { "epoch": 2.1588620881710296, "grad_norm": 0.1409483551979065, "learning_rate": 6.063528870646367e-05, "loss": 0.41842103004455566, "memory(GiB)": 70.5, "step": 50390, "token_acc": 0.9157088122605364, "train_speed(iter/s)": 1.450395 }, { "epoch": 2.1590763035002785, "grad_norm": 2.3140628337860107, "learning_rate": 6.0628712839245005e-05, "loss": 0.26376821994781496, "memory(GiB)": 70.5, "step": 50395, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.450399 }, { "epoch": 2.1592905188295273, "grad_norm": 3.5447325706481934, "learning_rate": 6.0622136779477254e-05, "loss": 0.3272446393966675, "memory(GiB)": 70.5, "step": 50400, "token_acc": 0.9226006191950464, "train_speed(iter/s)": 1.450403 }, { "epoch": 2.1595047341587765, "grad_norm": 3.569477081298828, "learning_rate": 6.0615560527279514e-05, "loss": 0.3573725938796997, "memory(GiB)": 70.5, "step": 50405, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.450423 }, { "epoch": 2.1597189494880253, "grad_norm": 1.5329039096832275, "learning_rate": 6.060898408277096e-05, "loss": 0.1684481382369995, "memory(GiB)": 70.5, "step": 50410, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.450425 }, { "epoch": 2.159933164817274, "grad_norm": 4.514549255371094, "learning_rate": 6.06024074460707e-05, "loss": 0.3763023138046265, "memory(GiB)": 70.5, "step": 50415, "token_acc": 0.924901185770751, "train_speed(iter/s)": 1.450428 }, { "epoch": 2.1601473801465234, "grad_norm": 4.131153106689453, "learning_rate": 6.059583061729787e-05, "loss": 0.3034748792648315, "memory(GiB)": 70.5, "step": 50420, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.450427 }, { "epoch": 2.1603615954757722, "grad_norm": 3.693256139755249, "learning_rate": 6.058925359657164e-05, "loss": 0.5924712181091308, "memory(GiB)": 70.5, "step": 50425, "token_acc": 0.8792569659442725, "train_speed(iter/s)": 1.450423 }, { "epoch": 2.160575810805021, "grad_norm": 3.022909164428711, "learning_rate": 6.058267638401114e-05, "loss": 0.20727732181549072, "memory(GiB)": 70.5, "step": 50430, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.450437 }, { "epoch": 2.1607900261342703, "grad_norm": 4.513272285461426, "learning_rate": 6.057609897973552e-05, "loss": 0.5032817840576171, "memory(GiB)": 70.5, "step": 50435, "token_acc": 0.8720238095238095, "train_speed(iter/s)": 1.450456 }, { "epoch": 2.161004241463519, "grad_norm": 0.8097392320632935, "learning_rate": 6.056952138386397e-05, "loss": 0.33360862731933594, "memory(GiB)": 70.5, "step": 50440, "token_acc": 0.9240121580547113, "train_speed(iter/s)": 1.450464 }, { "epoch": 2.161218456792768, "grad_norm": 3.026848316192627, "learning_rate": 6.056294359651562e-05, "loss": 0.2413865566253662, "memory(GiB)": 70.5, "step": 50445, "token_acc": 0.953168044077135, "train_speed(iter/s)": 1.450461 }, { "epoch": 2.161432672122017, "grad_norm": 1.9898422956466675, "learning_rate": 6.0556365617809615e-05, "loss": 0.49372687339782717, "memory(GiB)": 70.5, "step": 50450, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.45046 }, { "epoch": 2.161646887451266, "grad_norm": 2.8418028354644775, "learning_rate": 6.0549787447865166e-05, "loss": 0.36460859775543214, "memory(GiB)": 70.5, "step": 50455, "token_acc": 0.9277566539923955, "train_speed(iter/s)": 1.450462 }, { "epoch": 2.161861102780515, "grad_norm": 6.64315128326416, "learning_rate": 6.0543209086801434e-05, "loss": 0.6724228382110595, "memory(GiB)": 70.5, "step": 50460, "token_acc": 0.8876811594202898, "train_speed(iter/s)": 1.450463 }, { "epoch": 2.162075318109764, "grad_norm": 2.3482744693756104, "learning_rate": 6.053663053473754e-05, "loss": 0.4040745735168457, "memory(GiB)": 70.5, "step": 50465, "token_acc": 0.9099099099099099, "train_speed(iter/s)": 1.450467 }, { "epoch": 2.162289533439013, "grad_norm": 4.998032093048096, "learning_rate": 6.053005179179273e-05, "loss": 0.6280900955200195, "memory(GiB)": 70.5, "step": 50470, "token_acc": 0.8860294117647058, "train_speed(iter/s)": 1.450467 }, { "epoch": 2.1625037487682617, "grad_norm": 1.6790562868118286, "learning_rate": 6.052347285808615e-05, "loss": 0.43697094917297363, "memory(GiB)": 70.5, "step": 50475, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.450474 }, { "epoch": 2.162717964097511, "grad_norm": 2.350163459777832, "learning_rate": 6.051689373373698e-05, "loss": 0.23639962673187256, "memory(GiB)": 70.5, "step": 50480, "token_acc": 0.9633699633699634, "train_speed(iter/s)": 1.45048 }, { "epoch": 2.1629321794267597, "grad_norm": 2.100372076034546, "learning_rate": 6.0510314418864413e-05, "loss": 0.15725407600402833, "memory(GiB)": 70.5, "step": 50485, "token_acc": 0.955719557195572, "train_speed(iter/s)": 1.450484 }, { "epoch": 2.1631463947560086, "grad_norm": 3.1015419960021973, "learning_rate": 6.050373491358764e-05, "loss": 0.31547520160675047, "memory(GiB)": 70.5, "step": 50490, "token_acc": 0.9288537549407114, "train_speed(iter/s)": 1.450501 }, { "epoch": 2.163360610085258, "grad_norm": 3.2656090259552, "learning_rate": 6.049715521802587e-05, "loss": 0.399636435508728, "memory(GiB)": 70.5, "step": 50495, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.450495 }, { "epoch": 2.1635748254145066, "grad_norm": 8.556272506713867, "learning_rate": 6.0490575332298274e-05, "loss": 0.391419506072998, "memory(GiB)": 70.5, "step": 50500, "token_acc": 0.94375, "train_speed(iter/s)": 1.45051 }, { "epoch": 2.1635748254145066, "eval_loss": 2.5923960208892822, "eval_runtime": 13.495, "eval_samples_per_second": 7.41, "eval_steps_per_second": 7.41, "eval_token_acc": 0.4452054794520548, "step": 50500 }, { "epoch": 2.1637890407437554, "grad_norm": 12.670923233032227, "learning_rate": 6.048399525652406e-05, "loss": 0.34789199829101564, "memory(GiB)": 70.5, "step": 50505, "token_acc": 0.5797819623389494, "train_speed(iter/s)": 1.449923 }, { "epoch": 2.1640032560730047, "grad_norm": 5.175434112548828, "learning_rate": 6.0477414990822444e-05, "loss": 0.30921671390533445, "memory(GiB)": 70.5, "step": 50510, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.449914 }, { "epoch": 2.1642174714022535, "grad_norm": 5.639260292053223, "learning_rate": 6.0470834535312636e-05, "loss": 0.564429759979248, "memory(GiB)": 70.5, "step": 50515, "token_acc": 0.8770764119601329, "train_speed(iter/s)": 1.449909 }, { "epoch": 2.1644316867315023, "grad_norm": 3.4619226455688477, "learning_rate": 6.046425389011382e-05, "loss": 0.28351678848266604, "memory(GiB)": 70.5, "step": 50520, "token_acc": 0.941908713692946, "train_speed(iter/s)": 1.449915 }, { "epoch": 2.1646459020607516, "grad_norm": 3.3895816802978516, "learning_rate": 6.045767305534524e-05, "loss": 0.3387812614440918, "memory(GiB)": 70.5, "step": 50525, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.449918 }, { "epoch": 2.1648601173900004, "grad_norm": 1.755401849746704, "learning_rate": 6.045109203112611e-05, "loss": 0.3025153398513794, "memory(GiB)": 70.5, "step": 50530, "token_acc": 0.943217665615142, "train_speed(iter/s)": 1.44993 }, { "epoch": 2.165074332719249, "grad_norm": 1.273904800415039, "learning_rate": 6.044451081757563e-05, "loss": 0.4991893291473389, "memory(GiB)": 70.5, "step": 50535, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.449948 }, { "epoch": 2.1652885480484985, "grad_norm": 3.3584203720092773, "learning_rate": 6.043792941481303e-05, "loss": 0.4423415184020996, "memory(GiB)": 70.5, "step": 50540, "token_acc": 0.8914728682170543, "train_speed(iter/s)": 1.44995 }, { "epoch": 2.1655027633777473, "grad_norm": 7.488539218902588, "learning_rate": 6.0431347822957574e-05, "loss": 0.5717312812805175, "memory(GiB)": 70.5, "step": 50545, "token_acc": 0.884, "train_speed(iter/s)": 1.44997 }, { "epoch": 2.165716978706996, "grad_norm": 3.210742712020874, "learning_rate": 6.042476604212844e-05, "loss": 0.2585152149200439, "memory(GiB)": 70.5, "step": 50550, "token_acc": 0.9261538461538461, "train_speed(iter/s)": 1.449967 }, { "epoch": 2.1659311940362453, "grad_norm": 3.596564769744873, "learning_rate": 6.041818407244492e-05, "loss": 0.48058762550354006, "memory(GiB)": 70.5, "step": 50555, "token_acc": 0.9025974025974026, "train_speed(iter/s)": 1.449983 }, { "epoch": 2.166145409365494, "grad_norm": 3.660250425338745, "learning_rate": 6.0411601914026205e-05, "loss": 0.3062582492828369, "memory(GiB)": 70.5, "step": 50560, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.449986 }, { "epoch": 2.166359624694743, "grad_norm": 4.263165473937988, "learning_rate": 6.040501956699155e-05, "loss": 0.2811107873916626, "memory(GiB)": 70.5, "step": 50565, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.449985 }, { "epoch": 2.1665738400239922, "grad_norm": 4.620199680328369, "learning_rate": 6.039843703146022e-05, "loss": 0.426584529876709, "memory(GiB)": 70.5, "step": 50570, "token_acc": 0.9162011173184358, "train_speed(iter/s)": 1.449988 }, { "epoch": 2.166788055353241, "grad_norm": 1.232830286026001, "learning_rate": 6.039185430755143e-05, "loss": 0.4644639015197754, "memory(GiB)": 70.5, "step": 50575, "token_acc": 0.8902439024390244, "train_speed(iter/s)": 1.450015 }, { "epoch": 2.16700227068249, "grad_norm": 3.5579488277435303, "learning_rate": 6.038527139538445e-05, "loss": 0.26367716789245604, "memory(GiB)": 70.5, "step": 50580, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.450021 }, { "epoch": 2.167216486011739, "grad_norm": 0.9253590703010559, "learning_rate": 6.0378688295078556e-05, "loss": 0.3698856353759766, "memory(GiB)": 70.5, "step": 50585, "token_acc": 0.904, "train_speed(iter/s)": 1.45002 }, { "epoch": 2.167430701340988, "grad_norm": 4.819793701171875, "learning_rate": 6.037210500675298e-05, "loss": 0.3007026672363281, "memory(GiB)": 70.5, "step": 50590, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.45003 }, { "epoch": 2.1676449166702367, "grad_norm": 3.6802706718444824, "learning_rate": 6.036552153052698e-05, "loss": 0.43890671730041503, "memory(GiB)": 70.5, "step": 50595, "token_acc": 0.9108910891089109, "train_speed(iter/s)": 1.450043 }, { "epoch": 2.167859131999486, "grad_norm": 3.3731322288513184, "learning_rate": 6.035893786651985e-05, "loss": 0.33712828159332275, "memory(GiB)": 70.5, "step": 50600, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.450049 }, { "epoch": 2.168073347328735, "grad_norm": 2.9495229721069336, "learning_rate": 6.035235401485084e-05, "loss": 0.23694026470184326, "memory(GiB)": 70.5, "step": 50605, "token_acc": 0.9484978540772532, "train_speed(iter/s)": 1.450068 }, { "epoch": 2.1682875626579836, "grad_norm": 2.4004592895507812, "learning_rate": 6.034576997563921e-05, "loss": 0.5282141208648682, "memory(GiB)": 70.5, "step": 50610, "token_acc": 0.8708609271523179, "train_speed(iter/s)": 1.450066 }, { "epoch": 2.168501777987233, "grad_norm": 4.100773811340332, "learning_rate": 6.0339185749004265e-05, "loss": 0.3914056301116943, "memory(GiB)": 70.5, "step": 50615, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.450086 }, { "epoch": 2.1687159933164817, "grad_norm": 2.4297852516174316, "learning_rate": 6.033260133506528e-05, "loss": 0.3595496654510498, "memory(GiB)": 70.5, "step": 50620, "token_acc": 0.919732441471572, "train_speed(iter/s)": 1.45009 }, { "epoch": 2.1689302086457305, "grad_norm": 1.092568278312683, "learning_rate": 6.03260167339415e-05, "loss": 0.3191986083984375, "memory(GiB)": 70.5, "step": 50625, "token_acc": 0.9288389513108615, "train_speed(iter/s)": 1.4501 }, { "epoch": 2.1691444239749798, "grad_norm": 0.6710596084594727, "learning_rate": 6.031943194575227e-05, "loss": 0.5258928775787354, "memory(GiB)": 70.5, "step": 50630, "token_acc": 0.90234375, "train_speed(iter/s)": 1.450107 }, { "epoch": 2.1693586393042286, "grad_norm": 2.9561519622802734, "learning_rate": 6.031284697061683e-05, "loss": 0.6219576835632324, "memory(GiB)": 70.5, "step": 50635, "token_acc": 0.8626865671641791, "train_speed(iter/s)": 1.450121 }, { "epoch": 2.1695728546334774, "grad_norm": 3.5043416023254395, "learning_rate": 6.030626180865451e-05, "loss": 0.2751473426818848, "memory(GiB)": 70.5, "step": 50640, "token_acc": 0.9378378378378378, "train_speed(iter/s)": 1.450114 }, { "epoch": 2.1697870699627266, "grad_norm": 2.746171712875366, "learning_rate": 6.029967645998459e-05, "loss": 0.40851850509643556, "memory(GiB)": 70.5, "step": 50645, "token_acc": 0.9347181008902077, "train_speed(iter/s)": 1.450113 }, { "epoch": 2.1700012852919754, "grad_norm": 3.3916001319885254, "learning_rate": 6.0293090924726346e-05, "loss": 0.23308019638061522, "memory(GiB)": 70.5, "step": 50650, "token_acc": 0.9463722397476341, "train_speed(iter/s)": 1.450119 }, { "epoch": 2.1702155006212243, "grad_norm": 1.0808597803115845, "learning_rate": 6.028650520299912e-05, "loss": 0.28803603649139403, "memory(GiB)": 70.5, "step": 50655, "token_acc": 0.9295302013422819, "train_speed(iter/s)": 1.45013 }, { "epoch": 2.1704297159504735, "grad_norm": 5.271799564361572, "learning_rate": 6.0279919294922206e-05, "loss": 0.22374839782714845, "memory(GiB)": 70.5, "step": 50660, "token_acc": 0.9618320610687023, "train_speed(iter/s)": 1.450125 }, { "epoch": 2.1706439312797223, "grad_norm": 3.335956573486328, "learning_rate": 6.02733332006149e-05, "loss": 0.5024633407592773, "memory(GiB)": 70.5, "step": 50665, "token_acc": 0.9015873015873016, "train_speed(iter/s)": 1.45015 }, { "epoch": 2.170858146608971, "grad_norm": 1.7045818567276, "learning_rate": 6.026674692019654e-05, "loss": 0.42264270782470703, "memory(GiB)": 70.5, "step": 50670, "token_acc": 0.9235474006116208, "train_speed(iter/s)": 1.450149 }, { "epoch": 2.1710723619382204, "grad_norm": 5.937655925750732, "learning_rate": 6.0260160453786416e-05, "loss": 0.5328501701354981, "memory(GiB)": 70.5, "step": 50675, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.450148 }, { "epoch": 2.171286577267469, "grad_norm": 3.854870080947876, "learning_rate": 6.025357380150387e-05, "loss": 0.3493131399154663, "memory(GiB)": 70.5, "step": 50680, "token_acc": 0.9501557632398754, "train_speed(iter/s)": 1.450144 }, { "epoch": 2.171500792596718, "grad_norm": 3.636934995651245, "learning_rate": 6.02469869634682e-05, "loss": 0.3863845348358154, "memory(GiB)": 70.5, "step": 50685, "token_acc": 0.9049429657794676, "train_speed(iter/s)": 1.450159 }, { "epoch": 2.1717150079259673, "grad_norm": 1.8298152685165405, "learning_rate": 6.0240399939798766e-05, "loss": 0.17030801773071289, "memory(GiB)": 70.5, "step": 50690, "token_acc": 0.9707792207792207, "train_speed(iter/s)": 1.450161 }, { "epoch": 2.171929223255216, "grad_norm": 5.10529899597168, "learning_rate": 6.023381273061487e-05, "loss": 0.9092180252075195, "memory(GiB)": 70.5, "step": 50695, "token_acc": 0.8012048192771084, "train_speed(iter/s)": 1.450145 }, { "epoch": 2.172143438584465, "grad_norm": 4.012739181518555, "learning_rate": 6.0227225336035866e-05, "loss": 0.48504176139831545, "memory(GiB)": 70.5, "step": 50700, "token_acc": 0.900990099009901, "train_speed(iter/s)": 1.450136 }, { "epoch": 2.172357653913714, "grad_norm": 8.615055084228516, "learning_rate": 6.022063775618107e-05, "loss": 0.5546581268310546, "memory(GiB)": 70.5, "step": 50705, "token_acc": 0.8837209302325582, "train_speed(iter/s)": 1.450156 }, { "epoch": 2.172571869242963, "grad_norm": 2.8993635177612305, "learning_rate": 6.0214049991169844e-05, "loss": 0.296434211730957, "memory(GiB)": 70.5, "step": 50710, "token_acc": 0.95, "train_speed(iter/s)": 1.450155 }, { "epoch": 2.172786084572212, "grad_norm": 0.8645787835121155, "learning_rate": 6.0207462041121524e-05, "loss": 0.5426628112792968, "memory(GiB)": 70.5, "step": 50715, "token_acc": 0.8672086720867209, "train_speed(iter/s)": 1.450167 }, { "epoch": 2.173000299901461, "grad_norm": 3.5879170894622803, "learning_rate": 6.0200873906155455e-05, "loss": 0.21097178459167482, "memory(GiB)": 70.5, "step": 50720, "token_acc": 0.9451612903225807, "train_speed(iter/s)": 1.450168 }, { "epoch": 2.17321451523071, "grad_norm": 1.6429110765457153, "learning_rate": 6.0194285586390955e-05, "loss": 0.3435809135437012, "memory(GiB)": 70.5, "step": 50725, "token_acc": 0.9182879377431906, "train_speed(iter/s)": 1.450172 }, { "epoch": 2.1734287305599587, "grad_norm": 4.476863861083984, "learning_rate": 6.0187697081947434e-05, "loss": 0.6583369255065918, "memory(GiB)": 70.5, "step": 50730, "token_acc": 0.8811188811188811, "train_speed(iter/s)": 1.450169 }, { "epoch": 2.173642945889208, "grad_norm": 1.9101914167404175, "learning_rate": 6.0181108392944216e-05, "loss": 0.3402487993240356, "memory(GiB)": 70.5, "step": 50735, "token_acc": 0.935672514619883, "train_speed(iter/s)": 1.450179 }, { "epoch": 2.1738571612184567, "grad_norm": 1.5978426933288574, "learning_rate": 6.017451951950067e-05, "loss": 0.36893749237060547, "memory(GiB)": 70.5, "step": 50740, "token_acc": 0.9123505976095617, "train_speed(iter/s)": 1.450215 }, { "epoch": 2.1740713765477055, "grad_norm": 2.0848536491394043, "learning_rate": 6.0167930461736165e-05, "loss": 0.39859638214111326, "memory(GiB)": 70.5, "step": 50745, "token_acc": 0.9137254901960784, "train_speed(iter/s)": 1.450218 }, { "epoch": 2.174285591876955, "grad_norm": 1.8936057090759277, "learning_rate": 6.016134121977006e-05, "loss": 0.31246128082275393, "memory(GiB)": 70.5, "step": 50750, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.450219 }, { "epoch": 2.1744998072062036, "grad_norm": 4.131645202636719, "learning_rate": 6.015475179372173e-05, "loss": 0.3909461498260498, "memory(GiB)": 70.5, "step": 50755, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.450221 }, { "epoch": 2.1747140225354524, "grad_norm": 2.9218990802764893, "learning_rate": 6.0148162183710534e-05, "loss": 0.3356438159942627, "memory(GiB)": 70.5, "step": 50760, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.450226 }, { "epoch": 2.1749282378647017, "grad_norm": 5.598459720611572, "learning_rate": 6.014157238985587e-05, "loss": 0.3893770694732666, "memory(GiB)": 70.5, "step": 50765, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.450221 }, { "epoch": 2.1751424531939505, "grad_norm": 4.1312150955200195, "learning_rate": 6.0134982412277095e-05, "loss": 0.3023512363433838, "memory(GiB)": 70.5, "step": 50770, "token_acc": 0.9490909090909091, "train_speed(iter/s)": 1.450219 }, { "epoch": 2.1753566685231993, "grad_norm": 4.15233850479126, "learning_rate": 6.0128392251093624e-05, "loss": 0.3392078399658203, "memory(GiB)": 70.5, "step": 50775, "token_acc": 0.9128787878787878, "train_speed(iter/s)": 1.450221 }, { "epoch": 2.1755708838524486, "grad_norm": 3.2555618286132812, "learning_rate": 6.012180190642481e-05, "loss": 0.5061532497406006, "memory(GiB)": 70.5, "step": 50780, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.450228 }, { "epoch": 2.1757850991816974, "grad_norm": 4.4852614402771, "learning_rate": 6.011521137839007e-05, "loss": 0.40663585662841795, "memory(GiB)": 70.5, "step": 50785, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.450227 }, { "epoch": 2.175999314510946, "grad_norm": 2.6591453552246094, "learning_rate": 6.0108620667108794e-05, "loss": 0.5149047374725342, "memory(GiB)": 70.5, "step": 50790, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.450248 }, { "epoch": 2.1762135298401954, "grad_norm": 4.918720245361328, "learning_rate": 6.010202977270035e-05, "loss": 0.3292094707489014, "memory(GiB)": 70.5, "step": 50795, "token_acc": 0.9465648854961832, "train_speed(iter/s)": 1.450251 }, { "epoch": 2.1764277451694443, "grad_norm": 2.3173131942749023, "learning_rate": 6.009543869528417e-05, "loss": 0.2565911054611206, "memory(GiB)": 70.5, "step": 50800, "token_acc": 0.9487951807228916, "train_speed(iter/s)": 1.450253 }, { "epoch": 2.176641960498693, "grad_norm": 4.2892022132873535, "learning_rate": 6.008884743497966e-05, "loss": 0.23151252269744874, "memory(GiB)": 70.5, "step": 50805, "token_acc": 0.9633699633699634, "train_speed(iter/s)": 1.450262 }, { "epoch": 2.1768561758279423, "grad_norm": 2.5561747550964355, "learning_rate": 6.00822559919062e-05, "loss": 0.39697904586791993, "memory(GiB)": 70.5, "step": 50810, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.450263 }, { "epoch": 2.177070391157191, "grad_norm": 4.587259769439697, "learning_rate": 6.007566436618321e-05, "loss": 0.5739389419555664, "memory(GiB)": 70.5, "step": 50815, "token_acc": 0.9022556390977443, "train_speed(iter/s)": 1.450257 }, { "epoch": 2.17728460648644, "grad_norm": 3.2156083583831787, "learning_rate": 6.006907255793013e-05, "loss": 0.3968796968460083, "memory(GiB)": 70.5, "step": 50820, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.450265 }, { "epoch": 2.177498821815689, "grad_norm": 2.8929522037506104, "learning_rate": 6.006248056726634e-05, "loss": 0.30507328510284426, "memory(GiB)": 70.5, "step": 50825, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.450283 }, { "epoch": 2.177713037144938, "grad_norm": 5.098251819610596, "learning_rate": 6.005588839431129e-05, "loss": 0.7591546058654786, "memory(GiB)": 70.5, "step": 50830, "token_acc": 0.8408408408408409, "train_speed(iter/s)": 1.450299 }, { "epoch": 2.177927252474187, "grad_norm": 1.2483460903167725, "learning_rate": 6.0049296039184364e-05, "loss": 0.5337582111358643, "memory(GiB)": 70.5, "step": 50835, "token_acc": 0.9041916167664671, "train_speed(iter/s)": 1.4503 }, { "epoch": 2.178141467803436, "grad_norm": 4.255964756011963, "learning_rate": 6.0042703502005015e-05, "loss": 0.5335560798645019, "memory(GiB)": 70.5, "step": 50840, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.450303 }, { "epoch": 2.178355683132685, "grad_norm": 3.8531720638275146, "learning_rate": 6.00361107828927e-05, "loss": 0.48204526901245115, "memory(GiB)": 70.5, "step": 50845, "token_acc": 0.8723404255319149, "train_speed(iter/s)": 1.450304 }, { "epoch": 2.1785698984619337, "grad_norm": 3.491903781890869, "learning_rate": 6.00295178819668e-05, "loss": 0.39439728260040285, "memory(GiB)": 70.5, "step": 50850, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.450303 }, { "epoch": 2.178784113791183, "grad_norm": 5.094521522521973, "learning_rate": 6.002292479934678e-05, "loss": 0.7063254833221435, "memory(GiB)": 70.5, "step": 50855, "token_acc": 0.863481228668942, "train_speed(iter/s)": 1.450298 }, { "epoch": 2.178998329120432, "grad_norm": 2.5922508239746094, "learning_rate": 6.0016331535152084e-05, "loss": 0.5594659328460694, "memory(GiB)": 70.5, "step": 50860, "token_acc": 0.8896103896103896, "train_speed(iter/s)": 1.450316 }, { "epoch": 2.1792125444496806, "grad_norm": 1.0435105562210083, "learning_rate": 6.000973808950214e-05, "loss": 0.14698625802993776, "memory(GiB)": 70.5, "step": 50865, "token_acc": 0.9686411149825784, "train_speed(iter/s)": 1.450324 }, { "epoch": 2.17942675977893, "grad_norm": 6.542809963226318, "learning_rate": 6.000314446251638e-05, "loss": 0.340877628326416, "memory(GiB)": 70.5, "step": 50870, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.450329 }, { "epoch": 2.1796409751081787, "grad_norm": 2.5310957431793213, "learning_rate": 5.99965506543143e-05, "loss": 0.6822574615478516, "memory(GiB)": 70.5, "step": 50875, "token_acc": 0.8483870967741935, "train_speed(iter/s)": 1.450343 }, { "epoch": 2.1798551904374275, "grad_norm": 4.222930908203125, "learning_rate": 5.9989956665015324e-05, "loss": 0.309874963760376, "memory(GiB)": 70.5, "step": 50880, "token_acc": 0.9438943894389439, "train_speed(iter/s)": 1.450343 }, { "epoch": 2.1800694057666767, "grad_norm": 4.666376113891602, "learning_rate": 5.99833624947389e-05, "loss": 0.4180905342102051, "memory(GiB)": 70.5, "step": 50885, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.450342 }, { "epoch": 2.1802836210959256, "grad_norm": 3.2566232681274414, "learning_rate": 5.997676814360451e-05, "loss": 0.4702473163604736, "memory(GiB)": 70.5, "step": 50890, "token_acc": 0.9105960264900662, "train_speed(iter/s)": 1.450355 }, { "epoch": 2.1804978364251744, "grad_norm": 4.381570816040039, "learning_rate": 5.9970173611731616e-05, "loss": 0.36413044929504396, "memory(GiB)": 70.5, "step": 50895, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.45035 }, { "epoch": 2.1807120517544236, "grad_norm": 2.350660800933838, "learning_rate": 5.996357889923965e-05, "loss": 0.2897210121154785, "memory(GiB)": 70.5, "step": 50900, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.450348 }, { "epoch": 2.1809262670836724, "grad_norm": 2.6837234497070312, "learning_rate": 5.995698400624813e-05, "loss": 0.43457584381103515, "memory(GiB)": 70.5, "step": 50905, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.450352 }, { "epoch": 2.1811404824129212, "grad_norm": 2.5613183975219727, "learning_rate": 5.995038893287648e-05, "loss": 0.38179285526275636, "memory(GiB)": 70.5, "step": 50910, "token_acc": 0.8941605839416058, "train_speed(iter/s)": 1.450359 }, { "epoch": 2.1813546977421705, "grad_norm": 3.6755473613739014, "learning_rate": 5.994379367924421e-05, "loss": 0.5074479103088378, "memory(GiB)": 70.5, "step": 50915, "token_acc": 0.8732394366197183, "train_speed(iter/s)": 1.450366 }, { "epoch": 2.1815689130714193, "grad_norm": 3.511270523071289, "learning_rate": 5.993719824547079e-05, "loss": 0.4831888198852539, "memory(GiB)": 70.5, "step": 50920, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.450365 }, { "epoch": 2.181783128400668, "grad_norm": 3.8288726806640625, "learning_rate": 5.9930602631675705e-05, "loss": 0.4659147262573242, "memory(GiB)": 70.5, "step": 50925, "token_acc": 0.8859315589353612, "train_speed(iter/s)": 1.450378 }, { "epoch": 2.1819973437299174, "grad_norm": 3.3015096187591553, "learning_rate": 5.992400683797843e-05, "loss": 0.3359243154525757, "memory(GiB)": 70.5, "step": 50930, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.450378 }, { "epoch": 2.182211559059166, "grad_norm": 4.322972297668457, "learning_rate": 5.991741086449848e-05, "loss": 0.31626391410827637, "memory(GiB)": 70.5, "step": 50935, "token_acc": 0.9307692307692308, "train_speed(iter/s)": 1.450368 }, { "epoch": 2.182425774388415, "grad_norm": 1.029123306274414, "learning_rate": 5.991081471135531e-05, "loss": 0.2956243515014648, "memory(GiB)": 70.5, "step": 50940, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.450373 }, { "epoch": 2.1826399897176643, "grad_norm": 4.294251918792725, "learning_rate": 5.990421837866843e-05, "loss": 0.42691750526428224, "memory(GiB)": 70.5, "step": 50945, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.450383 }, { "epoch": 2.182854205046913, "grad_norm": 5.9248552322387695, "learning_rate": 5.989762186655736e-05, "loss": 0.19426398277282714, "memory(GiB)": 70.5, "step": 50950, "token_acc": 0.9566787003610109, "train_speed(iter/s)": 1.450389 }, { "epoch": 2.183068420376162, "grad_norm": 6.775415897369385, "learning_rate": 5.989102517514158e-05, "loss": 0.6069326400756836, "memory(GiB)": 70.5, "step": 50955, "token_acc": 0.8938906752411575, "train_speed(iter/s)": 1.450398 }, { "epoch": 2.183282635705411, "grad_norm": 4.81499719619751, "learning_rate": 5.9884428304540595e-05, "loss": 0.48601346015930175, "memory(GiB)": 70.5, "step": 50960, "token_acc": 0.9176954732510288, "train_speed(iter/s)": 1.450431 }, { "epoch": 2.18349685103466, "grad_norm": 3.1970746517181396, "learning_rate": 5.987783125487394e-05, "loss": 0.37641403675079343, "memory(GiB)": 70.5, "step": 50965, "token_acc": 0.9251968503937008, "train_speed(iter/s)": 1.450429 }, { "epoch": 2.1837110663639088, "grad_norm": 0.7025327086448669, "learning_rate": 5.987123402626108e-05, "loss": 0.2537921190261841, "memory(GiB)": 70.5, "step": 50970, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.45044 }, { "epoch": 2.183925281693158, "grad_norm": 1.469771146774292, "learning_rate": 5.986463661882157e-05, "loss": 0.4959570407867432, "memory(GiB)": 70.5, "step": 50975, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.450438 }, { "epoch": 2.184139497022407, "grad_norm": 1.8534303903579712, "learning_rate": 5.985803903267491e-05, "loss": 0.27616114616394044, "memory(GiB)": 70.5, "step": 50980, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.450442 }, { "epoch": 2.1843537123516557, "grad_norm": 2.6703264713287354, "learning_rate": 5.985144126794061e-05, "loss": 0.49498748779296875, "memory(GiB)": 70.5, "step": 50985, "token_acc": 0.8608058608058609, "train_speed(iter/s)": 1.450444 }, { "epoch": 2.184567927680905, "grad_norm": 1.6884695291519165, "learning_rate": 5.984484332473823e-05, "loss": 0.14628790616989135, "memory(GiB)": 70.5, "step": 50990, "token_acc": 0.9724409448818898, "train_speed(iter/s)": 1.45045 }, { "epoch": 2.1847821430101537, "grad_norm": 0.38976579904556274, "learning_rate": 5.983824520318728e-05, "loss": 0.22680156230926513, "memory(GiB)": 70.5, "step": 50995, "token_acc": 0.9426751592356688, "train_speed(iter/s)": 1.450459 }, { "epoch": 2.1849963583394025, "grad_norm": 1.8618870973587036, "learning_rate": 5.983164690340727e-05, "loss": 0.22542252540588378, "memory(GiB)": 70.5, "step": 51000, "token_acc": 0.9616858237547893, "train_speed(iter/s)": 1.450472 }, { "epoch": 2.1849963583394025, "eval_loss": 2.5033349990844727, "eval_runtime": 14.3484, "eval_samples_per_second": 6.969, "eval_steps_per_second": 6.969, "eval_token_acc": 0.44054054054054054, "step": 51000 }, { "epoch": 2.185210573668652, "grad_norm": 3.5495004653930664, "learning_rate": 5.982504842551777e-05, "loss": 0.6544827938079834, "memory(GiB)": 70.5, "step": 51005, "token_acc": 0.5666982024597919, "train_speed(iter/s)": 1.449854 }, { "epoch": 2.1854247889979006, "grad_norm": 0.6601597666740417, "learning_rate": 5.981844976963831e-05, "loss": 0.3984886646270752, "memory(GiB)": 70.5, "step": 51010, "token_acc": 0.9145569620253164, "train_speed(iter/s)": 1.449857 }, { "epoch": 2.1856390043271494, "grad_norm": 3.863405227661133, "learning_rate": 5.981185093588839e-05, "loss": 0.33829736709594727, "memory(GiB)": 70.5, "step": 51015, "token_acc": 0.9288135593220339, "train_speed(iter/s)": 1.449857 }, { "epoch": 2.1858532196563987, "grad_norm": 2.1617472171783447, "learning_rate": 5.980525192438761e-05, "loss": 0.3811631202697754, "memory(GiB)": 70.5, "step": 51020, "token_acc": 0.9316770186335404, "train_speed(iter/s)": 1.44986 }, { "epoch": 2.1860674349856475, "grad_norm": 15.906596183776855, "learning_rate": 5.979865273525549e-05, "loss": 0.46365838050842284, "memory(GiB)": 70.5, "step": 51025, "token_acc": 0.9079497907949791, "train_speed(iter/s)": 1.449866 }, { "epoch": 2.1862816503148967, "grad_norm": 3.158656120300293, "learning_rate": 5.9792053368611565e-05, "loss": 0.4353679656982422, "memory(GiB)": 70.5, "step": 51030, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.449873 }, { "epoch": 2.1864958656441456, "grad_norm": 4.493058681488037, "learning_rate": 5.978545382457543e-05, "loss": 0.3438121795654297, "memory(GiB)": 70.5, "step": 51035, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.449871 }, { "epoch": 2.1867100809733944, "grad_norm": 3.145226240158081, "learning_rate": 5.977885410326661e-05, "loss": 0.6803734302520752, "memory(GiB)": 70.5, "step": 51040, "token_acc": 0.8753993610223643, "train_speed(iter/s)": 1.449885 }, { "epoch": 2.1869242963026436, "grad_norm": 3.1874423027038574, "learning_rate": 5.977225420480468e-05, "loss": 0.46413888931274416, "memory(GiB)": 70.5, "step": 51045, "token_acc": 0.9129129129129129, "train_speed(iter/s)": 1.449886 }, { "epoch": 2.1871385116318924, "grad_norm": 2.6805074214935303, "learning_rate": 5.97656541293092e-05, "loss": 0.3700529098510742, "memory(GiB)": 70.5, "step": 51050, "token_acc": 0.900709219858156, "train_speed(iter/s)": 1.449907 }, { "epoch": 2.1873527269611412, "grad_norm": 3.552578926086426, "learning_rate": 5.975905387689973e-05, "loss": 0.8501643180847168, "memory(GiB)": 70.5, "step": 51055, "token_acc": 0.8412698412698413, "train_speed(iter/s)": 1.449895 }, { "epoch": 2.1875669422903905, "grad_norm": 0.1346604973077774, "learning_rate": 5.9752453447695834e-05, "loss": 0.380810809135437, "memory(GiB)": 70.5, "step": 51060, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.449898 }, { "epoch": 2.1877811576196393, "grad_norm": 1.0206964015960693, "learning_rate": 5.974585284181712e-05, "loss": 0.46544637680053713, "memory(GiB)": 70.5, "step": 51065, "token_acc": 0.8908450704225352, "train_speed(iter/s)": 1.449898 }, { "epoch": 2.187995372948888, "grad_norm": 2.4040744304656982, "learning_rate": 5.973925205938311e-05, "loss": 0.20734648704528807, "memory(GiB)": 70.5, "step": 51070, "token_acc": 0.9548387096774194, "train_speed(iter/s)": 1.449904 }, { "epoch": 2.1882095882781374, "grad_norm": 2.513272762298584, "learning_rate": 5.973265110051344e-05, "loss": 0.3945478439331055, "memory(GiB)": 70.5, "step": 51075, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.449896 }, { "epoch": 2.188423803607386, "grad_norm": 2.537752151489258, "learning_rate": 5.9726049965327656e-05, "loss": 0.5745400428771973, "memory(GiB)": 70.5, "step": 51080, "token_acc": 0.8791208791208791, "train_speed(iter/s)": 1.449904 }, { "epoch": 2.188638018936635, "grad_norm": 3.634613037109375, "learning_rate": 5.9719448653945344e-05, "loss": 0.3844005107879639, "memory(GiB)": 70.5, "step": 51085, "token_acc": 0.93359375, "train_speed(iter/s)": 1.449909 }, { "epoch": 2.1888522342658843, "grad_norm": 3.30008602142334, "learning_rate": 5.9712847166486105e-05, "loss": 0.5045750141143799, "memory(GiB)": 70.5, "step": 51090, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.449929 }, { "epoch": 2.189066449595133, "grad_norm": 2.043272018432617, "learning_rate": 5.9706245503069534e-05, "loss": 0.23061032295227052, "memory(GiB)": 70.5, "step": 51095, "token_acc": 0.9516129032258065, "train_speed(iter/s)": 1.449928 }, { "epoch": 2.189280664924382, "grad_norm": 1.8209482431411743, "learning_rate": 5.9699643663815205e-05, "loss": 0.15258182287216188, "memory(GiB)": 70.5, "step": 51100, "token_acc": 0.9594594594594594, "train_speed(iter/s)": 1.449927 }, { "epoch": 2.189494880253631, "grad_norm": 4.67851448059082, "learning_rate": 5.969304164884275e-05, "loss": 0.42417001724243164, "memory(GiB)": 70.5, "step": 51105, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.44996 }, { "epoch": 2.18970909558288, "grad_norm": 4.721888065338135, "learning_rate": 5.968643945827176e-05, "loss": 0.307102108001709, "memory(GiB)": 70.5, "step": 51110, "token_acc": 0.9217391304347826, "train_speed(iter/s)": 1.449955 }, { "epoch": 2.1899233109121288, "grad_norm": 2.547982931137085, "learning_rate": 5.9679837092221815e-05, "loss": 0.34571263790130613, "memory(GiB)": 70.5, "step": 51115, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.449955 }, { "epoch": 2.190137526241378, "grad_norm": 3.1291699409484863, "learning_rate": 5.967323455081255e-05, "loss": 0.43664913177490233, "memory(GiB)": 70.5, "step": 51120, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.449953 }, { "epoch": 2.190351741570627, "grad_norm": 3.641458749771118, "learning_rate": 5.966663183416357e-05, "loss": 0.36188225746154784, "memory(GiB)": 70.5, "step": 51125, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.449961 }, { "epoch": 2.1905659568998757, "grad_norm": 1.7951501607894897, "learning_rate": 5.966002894239446e-05, "loss": 0.3912527561187744, "memory(GiB)": 70.5, "step": 51130, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.449967 }, { "epoch": 2.190780172229125, "grad_norm": 5.49590539932251, "learning_rate": 5.965342587562489e-05, "loss": 0.4462904930114746, "memory(GiB)": 70.5, "step": 51135, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.449981 }, { "epoch": 2.1909943875583737, "grad_norm": 3.948486566543579, "learning_rate": 5.9646822633974454e-05, "loss": 0.520871353149414, "memory(GiB)": 70.5, "step": 51140, "token_acc": 0.9020408163265307, "train_speed(iter/s)": 1.449979 }, { "epoch": 2.1912086028876225, "grad_norm": 1.158976435661316, "learning_rate": 5.964021921756277e-05, "loss": 0.23059558868408203, "memory(GiB)": 70.5, "step": 51145, "token_acc": 0.9542682926829268, "train_speed(iter/s)": 1.449982 }, { "epoch": 2.191422818216872, "grad_norm": 1.782740592956543, "learning_rate": 5.963361562650946e-05, "loss": 0.41855573654174805, "memory(GiB)": 70.5, "step": 51150, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.449984 }, { "epoch": 2.1916370335461206, "grad_norm": 3.158716917037964, "learning_rate": 5.962701186093419e-05, "loss": 0.4607736587524414, "memory(GiB)": 70.5, "step": 51155, "token_acc": 0.899390243902439, "train_speed(iter/s)": 1.449979 }, { "epoch": 2.1918512488753694, "grad_norm": 3.0478920936584473, "learning_rate": 5.962040792095656e-05, "loss": 0.37351288795471194, "memory(GiB)": 70.5, "step": 51160, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.44999 }, { "epoch": 2.1920654642046187, "grad_norm": 9.306955337524414, "learning_rate": 5.96138038066962e-05, "loss": 0.31277549266815186, "memory(GiB)": 70.5, "step": 51165, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.449986 }, { "epoch": 2.1922796795338675, "grad_norm": 0.7128827571868896, "learning_rate": 5.960719951827278e-05, "loss": 0.2718203067779541, "memory(GiB)": 70.5, "step": 51170, "token_acc": 0.9481481481481482, "train_speed(iter/s)": 1.449983 }, { "epoch": 2.1924938948631163, "grad_norm": 2.5276286602020264, "learning_rate": 5.960059505580593e-05, "loss": 0.5175028800964355, "memory(GiB)": 70.5, "step": 51175, "token_acc": 0.8867313915857605, "train_speed(iter/s)": 1.449981 }, { "epoch": 2.1927081101923656, "grad_norm": 5.026991844177246, "learning_rate": 5.9593990419415294e-05, "loss": 0.4754149913787842, "memory(GiB)": 70.5, "step": 51180, "token_acc": 0.9033333333333333, "train_speed(iter/s)": 1.44998 }, { "epoch": 2.1929223255216144, "grad_norm": 4.568122863769531, "learning_rate": 5.9587385609220516e-05, "loss": 0.7321350574493408, "memory(GiB)": 70.5, "step": 51185, "token_acc": 0.8682432432432432, "train_speed(iter/s)": 1.449984 }, { "epoch": 2.193136540850863, "grad_norm": 5.472165584564209, "learning_rate": 5.958078062534126e-05, "loss": 0.39142799377441406, "memory(GiB)": 70.5, "step": 51190, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.450008 }, { "epoch": 2.1933507561801124, "grad_norm": 4.328889846801758, "learning_rate": 5.957417546789717e-05, "loss": 0.4257911205291748, "memory(GiB)": 70.5, "step": 51195, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.450008 }, { "epoch": 2.1935649715093612, "grad_norm": 2.548952579498291, "learning_rate": 5.956757013700791e-05, "loss": 0.5193898677825928, "memory(GiB)": 70.5, "step": 51200, "token_acc": 0.9084507042253521, "train_speed(iter/s)": 1.450024 }, { "epoch": 2.19377918683861, "grad_norm": 3.2382924556732178, "learning_rate": 5.956096463279314e-05, "loss": 0.345200252532959, "memory(GiB)": 70.5, "step": 51205, "token_acc": 0.9291784702549575, "train_speed(iter/s)": 1.450045 }, { "epoch": 2.1939934021678593, "grad_norm": 1.9402272701263428, "learning_rate": 5.955435895537253e-05, "loss": 0.41257877349853517, "memory(GiB)": 70.5, "step": 51210, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.450059 }, { "epoch": 2.194207617497108, "grad_norm": 2.664508819580078, "learning_rate": 5.9547753104865746e-05, "loss": 0.584498119354248, "memory(GiB)": 70.5, "step": 51215, "token_acc": 0.8567251461988304, "train_speed(iter/s)": 1.450049 }, { "epoch": 2.194421832826357, "grad_norm": 0.17769382894039154, "learning_rate": 5.954114708139247e-05, "loss": 0.5057545661926269, "memory(GiB)": 70.5, "step": 51220, "token_acc": 0.8934707903780069, "train_speed(iter/s)": 1.450048 }, { "epoch": 2.194636048155606, "grad_norm": 4.135988712310791, "learning_rate": 5.953454088507236e-05, "loss": 0.38073699474334716, "memory(GiB)": 70.5, "step": 51225, "token_acc": 0.9123505976095617, "train_speed(iter/s)": 1.450057 }, { "epoch": 2.194850263484855, "grad_norm": 3.7053444385528564, "learning_rate": 5.952793451602507e-05, "loss": 0.42030653953552244, "memory(GiB)": 70.5, "step": 51230, "token_acc": 0.9110320284697508, "train_speed(iter/s)": 1.450064 }, { "epoch": 2.195064478814104, "grad_norm": 0.06621988117694855, "learning_rate": 5.9521327974370334e-05, "loss": 0.46125354766845705, "memory(GiB)": 70.5, "step": 51235, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.450065 }, { "epoch": 2.195278694143353, "grad_norm": 1.1664928197860718, "learning_rate": 5.95147212602278e-05, "loss": 0.38840432167053224, "memory(GiB)": 70.5, "step": 51240, "token_acc": 0.9182389937106918, "train_speed(iter/s)": 1.45006 }, { "epoch": 2.195492909472602, "grad_norm": 5.890903949737549, "learning_rate": 5.950811437371716e-05, "loss": 0.49774923324584963, "memory(GiB)": 70.5, "step": 51245, "token_acc": 0.8720238095238095, "train_speed(iter/s)": 1.450065 }, { "epoch": 2.1957071248018507, "grad_norm": 1.4045419692993164, "learning_rate": 5.950150731495813e-05, "loss": 0.3715933322906494, "memory(GiB)": 70.5, "step": 51250, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.450075 }, { "epoch": 2.1959213401311, "grad_norm": 7.51812219619751, "learning_rate": 5.949490008407037e-05, "loss": 0.1814044237136841, "memory(GiB)": 70.5, "step": 51255, "token_acc": 0.9640287769784173, "train_speed(iter/s)": 1.450087 }, { "epoch": 2.1961355554603488, "grad_norm": 5.106799602508545, "learning_rate": 5.9488292681173585e-05, "loss": 0.7546957015991211, "memory(GiB)": 70.5, "step": 51260, "token_acc": 0.8526645768025078, "train_speed(iter/s)": 1.450088 }, { "epoch": 2.1963497707895976, "grad_norm": 3.646167755126953, "learning_rate": 5.948168510638748e-05, "loss": 0.563668155670166, "memory(GiB)": 70.5, "step": 51265, "token_acc": 0.8736462093862816, "train_speed(iter/s)": 1.450108 }, { "epoch": 2.196563986118847, "grad_norm": 5.267276287078857, "learning_rate": 5.9475077359831766e-05, "loss": 0.44749135971069337, "memory(GiB)": 70.5, "step": 51270, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.450104 }, { "epoch": 2.1967782014480957, "grad_norm": 3.414707660675049, "learning_rate": 5.9468469441626116e-05, "loss": 0.43937177658081056, "memory(GiB)": 70.5, "step": 51275, "token_acc": 0.9088235294117647, "train_speed(iter/s)": 1.450103 }, { "epoch": 2.1969924167773445, "grad_norm": 2.6601991653442383, "learning_rate": 5.946186135189027e-05, "loss": 0.46574883460998534, "memory(GiB)": 70.5, "step": 51280, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.450108 }, { "epoch": 2.1972066321065937, "grad_norm": 4.106680870056152, "learning_rate": 5.945525309074393e-05, "loss": 0.09995218515396118, "memory(GiB)": 70.5, "step": 51285, "token_acc": 0.9765886287625418, "train_speed(iter/s)": 1.450112 }, { "epoch": 2.1974208474358425, "grad_norm": 2.6938745975494385, "learning_rate": 5.944864465830681e-05, "loss": 0.41347036361694334, "memory(GiB)": 70.5, "step": 51290, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.450124 }, { "epoch": 2.1976350627650914, "grad_norm": 5.8746538162231445, "learning_rate": 5.944203605469863e-05, "loss": 0.1980844497680664, "memory(GiB)": 70.5, "step": 51295, "token_acc": 0.9532374100719424, "train_speed(iter/s)": 1.450123 }, { "epoch": 2.1978492780943406, "grad_norm": 2.474396228790283, "learning_rate": 5.943542728003911e-05, "loss": 0.2958374500274658, "memory(GiB)": 70.5, "step": 51300, "token_acc": 0.945054945054945, "train_speed(iter/s)": 1.45013 }, { "epoch": 2.1980634934235894, "grad_norm": 2.4865128993988037, "learning_rate": 5.9428818334447976e-05, "loss": 0.3315902233123779, "memory(GiB)": 70.5, "step": 51305, "token_acc": 0.9223300970873787, "train_speed(iter/s)": 1.450137 }, { "epoch": 2.1982777087528382, "grad_norm": 3.3051915168762207, "learning_rate": 5.9422209218044956e-05, "loss": 0.5587521553039551, "memory(GiB)": 70.5, "step": 51310, "token_acc": 0.8940809968847352, "train_speed(iter/s)": 1.450138 }, { "epoch": 2.1984919240820875, "grad_norm": 7.414772033691406, "learning_rate": 5.941559993094976e-05, "loss": 0.3570552349090576, "memory(GiB)": 70.5, "step": 51315, "token_acc": 0.9213483146067416, "train_speed(iter/s)": 1.450138 }, { "epoch": 2.1987061394113363, "grad_norm": 1.524340271949768, "learning_rate": 5.9408990473282145e-05, "loss": 0.24772341251373292, "memory(GiB)": 70.5, "step": 51320, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.450152 }, { "epoch": 2.198920354740585, "grad_norm": 4.710116863250732, "learning_rate": 5.940238084516184e-05, "loss": 0.5519515991210937, "memory(GiB)": 70.5, "step": 51325, "token_acc": 0.8954248366013072, "train_speed(iter/s)": 1.450148 }, { "epoch": 2.1991345700698344, "grad_norm": 0.9877539277076721, "learning_rate": 5.9395771046708594e-05, "loss": 0.3144498109817505, "memory(GiB)": 70.5, "step": 51330, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.45015 }, { "epoch": 2.199348785399083, "grad_norm": 0.791766345500946, "learning_rate": 5.9389161078042143e-05, "loss": 0.2937031269073486, "memory(GiB)": 70.5, "step": 51335, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.450167 }, { "epoch": 2.199563000728332, "grad_norm": 3.9735541343688965, "learning_rate": 5.9382550939282234e-05, "loss": 0.27264227867126467, "memory(GiB)": 70.5, "step": 51340, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.450168 }, { "epoch": 2.1997772160575813, "grad_norm": 5.119778156280518, "learning_rate": 5.9375940630548597e-05, "loss": 0.6261510848999023, "memory(GiB)": 70.5, "step": 51345, "token_acc": 0.8596491228070176, "train_speed(iter/s)": 1.450179 }, { "epoch": 2.19999143138683, "grad_norm": 3.649566888809204, "learning_rate": 5.9369330151961e-05, "loss": 0.3045217990875244, "memory(GiB)": 70.5, "step": 51350, "token_acc": 0.9442622950819672, "train_speed(iter/s)": 1.450182 }, { "epoch": 2.200205646716079, "grad_norm": 12.93321704864502, "learning_rate": 5.9362719503639216e-05, "loss": 0.42865619659423826, "memory(GiB)": 70.5, "step": 51355, "token_acc": 0.9018987341772152, "train_speed(iter/s)": 1.450197 }, { "epoch": 2.200419862045328, "grad_norm": 0.7991122603416443, "learning_rate": 5.9356108685702974e-05, "loss": 0.5964747428894043, "memory(GiB)": 70.5, "step": 51360, "token_acc": 0.8794788273615635, "train_speed(iter/s)": 1.4502 }, { "epoch": 2.200634077374577, "grad_norm": 1.9408981800079346, "learning_rate": 5.934949769827205e-05, "loss": 0.4790898323059082, "memory(GiB)": 70.5, "step": 51365, "token_acc": 0.9035714285714286, "train_speed(iter/s)": 1.450213 }, { "epoch": 2.2008482927038258, "grad_norm": 2.364313840866089, "learning_rate": 5.9342886541466204e-05, "loss": 0.44541220664978026, "memory(GiB)": 70.5, "step": 51370, "token_acc": 0.9046242774566474, "train_speed(iter/s)": 1.450228 }, { "epoch": 2.201062508033075, "grad_norm": 3.2689449787139893, "learning_rate": 5.93362752154052e-05, "loss": 0.22482538223266602, "memory(GiB)": 70.5, "step": 51375, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.450234 }, { "epoch": 2.201276723362324, "grad_norm": 7.18286657333374, "learning_rate": 5.9329663720208826e-05, "loss": 0.3337057590484619, "memory(GiB)": 70.5, "step": 51380, "token_acc": 0.9372384937238494, "train_speed(iter/s)": 1.450223 }, { "epoch": 2.2014909386915726, "grad_norm": 2.29667592048645, "learning_rate": 5.932305205599683e-05, "loss": 0.42179241180419924, "memory(GiB)": 70.5, "step": 51385, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.450222 }, { "epoch": 2.201705154020822, "grad_norm": 0.06337615847587585, "learning_rate": 5.931644022288899e-05, "loss": 0.4333448886871338, "memory(GiB)": 70.5, "step": 51390, "token_acc": 0.90625, "train_speed(iter/s)": 1.450213 }, { "epoch": 2.2019193693500707, "grad_norm": 3.7145893573760986, "learning_rate": 5.9309828221005115e-05, "loss": 0.3168431282043457, "memory(GiB)": 70.5, "step": 51395, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.450228 }, { "epoch": 2.2021335846793195, "grad_norm": 0.6782521605491638, "learning_rate": 5.930321605046496e-05, "loss": 0.349801230430603, "memory(GiB)": 70.5, "step": 51400, "token_acc": 0.925, "train_speed(iter/s)": 1.450232 }, { "epoch": 2.202347800008569, "grad_norm": 3.5576584339141846, "learning_rate": 5.9296603711388324e-05, "loss": 0.383362340927124, "memory(GiB)": 70.5, "step": 51405, "token_acc": 0.9102167182662538, "train_speed(iter/s)": 1.450238 }, { "epoch": 2.2025620153378176, "grad_norm": 0.696987509727478, "learning_rate": 5.928999120389499e-05, "loss": 0.21114554405212402, "memory(GiB)": 70.5, "step": 51410, "token_acc": 0.9676375404530745, "train_speed(iter/s)": 1.450237 }, { "epoch": 2.2027762306670664, "grad_norm": 2.1492512226104736, "learning_rate": 5.928337852810475e-05, "loss": 0.2401029348373413, "memory(GiB)": 70.5, "step": 51415, "token_acc": 0.9349112426035503, "train_speed(iter/s)": 1.450242 }, { "epoch": 2.2029904459963157, "grad_norm": 3.5479490756988525, "learning_rate": 5.927676568413739e-05, "loss": 0.3101226329803467, "memory(GiB)": 70.5, "step": 51420, "token_acc": 0.9174917491749175, "train_speed(iter/s)": 1.450242 }, { "epoch": 2.2032046613255645, "grad_norm": 3.049267292022705, "learning_rate": 5.9270152672112725e-05, "loss": 0.23204026222229004, "memory(GiB)": 70.5, "step": 51425, "token_acc": 0.9303135888501742, "train_speed(iter/s)": 1.450251 }, { "epoch": 2.2034188766548133, "grad_norm": 2.936291456222534, "learning_rate": 5.9263539492150557e-05, "loss": 0.5579321384429932, "memory(GiB)": 70.5, "step": 51430, "token_acc": 0.884375, "train_speed(iter/s)": 1.45024 }, { "epoch": 2.2036330919840625, "grad_norm": 3.541712760925293, "learning_rate": 5.9256926144370663e-05, "loss": 0.4785926342010498, "memory(GiB)": 70.5, "step": 51435, "token_acc": 0.8849557522123894, "train_speed(iter/s)": 1.450237 }, { "epoch": 2.2038473073133114, "grad_norm": 1.179591178894043, "learning_rate": 5.9250312628892877e-05, "loss": 0.32626221179962156, "memory(GiB)": 70.5, "step": 51440, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.450248 }, { "epoch": 2.20406152264256, "grad_norm": 4.66850471496582, "learning_rate": 5.9243698945837014e-05, "loss": 0.3412567138671875, "memory(GiB)": 70.5, "step": 51445, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.450248 }, { "epoch": 2.2042757379718094, "grad_norm": 1.1496515274047852, "learning_rate": 5.923708509532284e-05, "loss": 0.26950273513793943, "memory(GiB)": 70.5, "step": 51450, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.450258 }, { "epoch": 2.2044899533010582, "grad_norm": 3.208296537399292, "learning_rate": 5.923047107747024e-05, "loss": 0.4769591331481934, "memory(GiB)": 70.5, "step": 51455, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.45025 }, { "epoch": 2.204704168630307, "grad_norm": 4.8395304679870605, "learning_rate": 5.9223856892398975e-05, "loss": 0.2033766031265259, "memory(GiB)": 70.5, "step": 51460, "token_acc": 0.9575971731448764, "train_speed(iter/s)": 1.450266 }, { "epoch": 2.2049183839595563, "grad_norm": 2.0114188194274902, "learning_rate": 5.921724254022889e-05, "loss": 0.16712719202041626, "memory(GiB)": 70.5, "step": 51465, "token_acc": 0.9623824451410659, "train_speed(iter/s)": 1.450264 }, { "epoch": 2.205132599288805, "grad_norm": 5.0708088874816895, "learning_rate": 5.921062802107982e-05, "loss": 0.4842782497406006, "memory(GiB)": 70.5, "step": 51470, "token_acc": 0.896, "train_speed(iter/s)": 1.45029 }, { "epoch": 2.205346814618054, "grad_norm": 5.219204902648926, "learning_rate": 5.920401333507157e-05, "loss": 0.4822529315948486, "memory(GiB)": 70.5, "step": 51475, "token_acc": 0.9, "train_speed(iter/s)": 1.450287 }, { "epoch": 2.205561029947303, "grad_norm": 3.5671870708465576, "learning_rate": 5.9197398482324e-05, "loss": 0.438889217376709, "memory(GiB)": 70.5, "step": 51480, "token_acc": 0.8900709219858156, "train_speed(iter/s)": 1.450296 }, { "epoch": 2.205775245276552, "grad_norm": 4.733739376068115, "learning_rate": 5.919078346295693e-05, "loss": 0.39969799518585203, "memory(GiB)": 70.5, "step": 51485, "token_acc": 0.9203296703296703, "train_speed(iter/s)": 1.450296 }, { "epoch": 2.205989460605801, "grad_norm": 2.8924806118011475, "learning_rate": 5.918416827709018e-05, "loss": 0.3818478584289551, "memory(GiB)": 70.5, "step": 51490, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.450297 }, { "epoch": 2.20620367593505, "grad_norm": 2.4865639209747314, "learning_rate": 5.917755292484361e-05, "loss": 0.30372593402862547, "memory(GiB)": 70.5, "step": 51495, "token_acc": 0.9127725856697819, "train_speed(iter/s)": 1.450296 }, { "epoch": 2.206417891264299, "grad_norm": 2.1032960414886475, "learning_rate": 5.917093740633707e-05, "loss": 0.33774752616882325, "memory(GiB)": 70.5, "step": 51500, "token_acc": 0.9137254901960784, "train_speed(iter/s)": 1.450297 }, { "epoch": 2.206417891264299, "eval_loss": 2.4934723377227783, "eval_runtime": 14.1383, "eval_samples_per_second": 7.073, "eval_steps_per_second": 7.073, "eval_token_acc": 0.44328552803129073, "step": 51500 }, { "epoch": 2.2066321065935477, "grad_norm": 4.093058109283447, "learning_rate": 5.916432172169038e-05, "loss": 0.4018381595611572, "memory(GiB)": 70.5, "step": 51505, "token_acc": 0.5746691871455577, "train_speed(iter/s)": 1.449693 }, { "epoch": 2.206846321922797, "grad_norm": 0.16976885497570038, "learning_rate": 5.9157705871023426e-05, "loss": 0.7254127025604248, "memory(GiB)": 70.5, "step": 51510, "token_acc": 0.8423076923076923, "train_speed(iter/s)": 1.449705 }, { "epoch": 2.2070605372520458, "grad_norm": 3.2444660663604736, "learning_rate": 5.915108985445603e-05, "loss": 0.189738130569458, "memory(GiB)": 70.5, "step": 51515, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.449717 }, { "epoch": 2.2072747525812946, "grad_norm": 4.515582084655762, "learning_rate": 5.914447367210805e-05, "loss": 0.2604857921600342, "memory(GiB)": 70.5, "step": 51520, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.44972 }, { "epoch": 2.207488967910544, "grad_norm": 6.270288944244385, "learning_rate": 5.913785732409937e-05, "loss": 0.5935777187347412, "memory(GiB)": 70.5, "step": 51525, "token_acc": 0.8552631578947368, "train_speed(iter/s)": 1.449725 }, { "epoch": 2.2077031832397926, "grad_norm": 5.082672119140625, "learning_rate": 5.913124081054981e-05, "loss": 0.2875372886657715, "memory(GiB)": 70.5, "step": 51530, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.449726 }, { "epoch": 2.2079173985690415, "grad_norm": 4.242832660675049, "learning_rate": 5.912462413157926e-05, "loss": 0.5190069675445557, "memory(GiB)": 70.5, "step": 51535, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.449728 }, { "epoch": 2.2081316138982907, "grad_norm": 4.580529689788818, "learning_rate": 5.91180072873076e-05, "loss": 0.4782750606536865, "memory(GiB)": 70.5, "step": 51540, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.449725 }, { "epoch": 2.2083458292275395, "grad_norm": 1.757004737854004, "learning_rate": 5.9111390277854675e-05, "loss": 0.18477821350097656, "memory(GiB)": 70.5, "step": 51545, "token_acc": 0.9515570934256056, "train_speed(iter/s)": 1.449731 }, { "epoch": 2.2085600445567883, "grad_norm": 0.39787065982818604, "learning_rate": 5.910477310334036e-05, "loss": 0.5085625171661377, "memory(GiB)": 70.5, "step": 51550, "token_acc": 0.8996539792387543, "train_speed(iter/s)": 1.449736 }, { "epoch": 2.2087742598860376, "grad_norm": 3.4419171810150146, "learning_rate": 5.9098155763884554e-05, "loss": 0.34318814277648924, "memory(GiB)": 70.5, "step": 51555, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.449736 }, { "epoch": 2.2089884752152864, "grad_norm": 8.027510643005371, "learning_rate": 5.909153825960711e-05, "loss": 0.6525047779083252, "memory(GiB)": 70.5, "step": 51560, "token_acc": 0.8859934853420195, "train_speed(iter/s)": 1.449731 }, { "epoch": 2.209202690544535, "grad_norm": 3.019226551055908, "learning_rate": 5.908492059062794e-05, "loss": 0.513969087600708, "memory(GiB)": 70.5, "step": 51565, "token_acc": 0.8623188405797102, "train_speed(iter/s)": 1.449726 }, { "epoch": 2.2094169058737845, "grad_norm": 2.3480610847473145, "learning_rate": 5.907830275706689e-05, "loss": 0.1021493911743164, "memory(GiB)": 70.5, "step": 51570, "token_acc": 0.9775641025641025, "train_speed(iter/s)": 1.449721 }, { "epoch": 2.2096311212030333, "grad_norm": 6.165764808654785, "learning_rate": 5.907168475904388e-05, "loss": 0.303079891204834, "memory(GiB)": 70.5, "step": 51575, "token_acc": 0.9375, "train_speed(iter/s)": 1.449742 }, { "epoch": 2.209845336532282, "grad_norm": 0.6493340134620667, "learning_rate": 5.906506659667878e-05, "loss": 0.6484148502349854, "memory(GiB)": 70.5, "step": 51580, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.449744 }, { "epoch": 2.2100595518615314, "grad_norm": 3.0141170024871826, "learning_rate": 5.905844827009151e-05, "loss": 0.30763022899627684, "memory(GiB)": 70.5, "step": 51585, "token_acc": 0.9472049689440993, "train_speed(iter/s)": 1.449741 }, { "epoch": 2.21027376719078, "grad_norm": 3.9710803031921387, "learning_rate": 5.905182977940195e-05, "loss": 0.5433608055114746, "memory(GiB)": 70.5, "step": 51590, "token_acc": 0.8916083916083916, "train_speed(iter/s)": 1.449738 }, { "epoch": 2.210487982520029, "grad_norm": 1.3597582578659058, "learning_rate": 5.904521112472999e-05, "loss": 0.36143791675567627, "memory(GiB)": 70.5, "step": 51595, "token_acc": 0.9475308641975309, "train_speed(iter/s)": 1.449726 }, { "epoch": 2.2107021978492782, "grad_norm": 1.4851993322372437, "learning_rate": 5.903859230619556e-05, "loss": 0.35996222496032715, "memory(GiB)": 70.5, "step": 51600, "token_acc": 0.9311377245508982, "train_speed(iter/s)": 1.449727 }, { "epoch": 2.210916413178527, "grad_norm": 1.2720882892608643, "learning_rate": 5.903197332391853e-05, "loss": 0.4410989761352539, "memory(GiB)": 70.5, "step": 51605, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.449725 }, { "epoch": 2.211130628507776, "grad_norm": 1.7728787660598755, "learning_rate": 5.902535417801884e-05, "loss": 0.3158538818359375, "memory(GiB)": 70.5, "step": 51610, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.449711 }, { "epoch": 2.211344843837025, "grad_norm": 4.101406574249268, "learning_rate": 5.901873486861641e-05, "loss": 0.5751792430877686, "memory(GiB)": 70.5, "step": 51615, "token_acc": 0.8498293515358362, "train_speed(iter/s)": 1.449708 }, { "epoch": 2.211559059166274, "grad_norm": 3.3314948081970215, "learning_rate": 5.901211539583111e-05, "loss": 0.42035694122314454, "memory(GiB)": 70.5, "step": 51620, "token_acc": 0.8991596638655462, "train_speed(iter/s)": 1.449712 }, { "epoch": 2.2117732744955227, "grad_norm": 3.1209747791290283, "learning_rate": 5.900549575978291e-05, "loss": 0.5162035465240479, "memory(GiB)": 70.5, "step": 51625, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.449727 }, { "epoch": 2.211987489824772, "grad_norm": 7.850344181060791, "learning_rate": 5.899887596059171e-05, "loss": 0.7029481410980225, "memory(GiB)": 70.5, "step": 51630, "token_acc": 0.8781362007168458, "train_speed(iter/s)": 1.449738 }, { "epoch": 2.212201705154021, "grad_norm": 5.772072792053223, "learning_rate": 5.899225599837741e-05, "loss": 0.45394697189331057, "memory(GiB)": 70.5, "step": 51635, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.449734 }, { "epoch": 2.2124159204832696, "grad_norm": 4.931660175323486, "learning_rate": 5.8985635873259956e-05, "loss": 0.5790389060974122, "memory(GiB)": 70.5, "step": 51640, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.44974 }, { "epoch": 2.212630135812519, "grad_norm": 5.073096752166748, "learning_rate": 5.8979015585359296e-05, "loss": 0.4463081359863281, "memory(GiB)": 70.5, "step": 51645, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.449752 }, { "epoch": 2.2128443511417677, "grad_norm": 2.5553786754608154, "learning_rate": 5.897239513479532e-05, "loss": 0.2724622249603271, "memory(GiB)": 70.5, "step": 51650, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.449759 }, { "epoch": 2.2130585664710165, "grad_norm": 0.23210838437080383, "learning_rate": 5.896577452168801e-05, "loss": 0.21689856052398682, "memory(GiB)": 70.5, "step": 51655, "token_acc": 0.9498069498069498, "train_speed(iter/s)": 1.449767 }, { "epoch": 2.2132727818002658, "grad_norm": 2.9007210731506348, "learning_rate": 5.8959153746157294e-05, "loss": 0.2384674072265625, "memory(GiB)": 70.5, "step": 51660, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.449767 }, { "epoch": 2.2134869971295146, "grad_norm": 2.412012815475464, "learning_rate": 5.895253280832308e-05, "loss": 0.213645076751709, "memory(GiB)": 70.5, "step": 51665, "token_acc": 0.9447852760736196, "train_speed(iter/s)": 1.449776 }, { "epoch": 2.2137012124587634, "grad_norm": 2.44979190826416, "learning_rate": 5.894591170830536e-05, "loss": 0.2246773958206177, "memory(GiB)": 70.5, "step": 51670, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.449781 }, { "epoch": 2.2139154277880126, "grad_norm": 2.2641701698303223, "learning_rate": 5.893929044622404e-05, "loss": 0.2315523862838745, "memory(GiB)": 70.5, "step": 51675, "token_acc": 0.948948948948949, "train_speed(iter/s)": 1.44978 }, { "epoch": 2.2141296431172615, "grad_norm": 3.960909843444824, "learning_rate": 5.8932669022199095e-05, "loss": 0.32328917980194094, "memory(GiB)": 70.5, "step": 51680, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.449782 }, { "epoch": 2.2143438584465103, "grad_norm": 5.725348949432373, "learning_rate": 5.892604743635045e-05, "loss": 0.43845090866088865, "memory(GiB)": 70.5, "step": 51685, "token_acc": 0.924, "train_speed(iter/s)": 1.44978 }, { "epoch": 2.2145580737757595, "grad_norm": 3.8541066646575928, "learning_rate": 5.891942568879811e-05, "loss": 0.3449028730392456, "memory(GiB)": 70.5, "step": 51690, "token_acc": 0.92, "train_speed(iter/s)": 1.449778 }, { "epoch": 2.2147722891050083, "grad_norm": 2.5143232345581055, "learning_rate": 5.8912803779662e-05, "loss": 0.2637375593185425, "memory(GiB)": 70.5, "step": 51695, "token_acc": 0.9323843416370107, "train_speed(iter/s)": 1.449778 }, { "epoch": 2.214986504434257, "grad_norm": 2.477506160736084, "learning_rate": 5.890618170906208e-05, "loss": 0.38475356101989744, "memory(GiB)": 70.5, "step": 51700, "token_acc": 0.9294478527607362, "train_speed(iter/s)": 1.449793 }, { "epoch": 2.2152007197635064, "grad_norm": 2.0137648582458496, "learning_rate": 5.889955947711834e-05, "loss": 0.2433213472366333, "memory(GiB)": 70.5, "step": 51705, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.449794 }, { "epoch": 2.215414935092755, "grad_norm": 2.1868538856506348, "learning_rate": 5.8892937083950704e-05, "loss": 0.5304313659667969, "memory(GiB)": 70.5, "step": 51710, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.44981 }, { "epoch": 2.215629150422004, "grad_norm": 0.9693938493728638, "learning_rate": 5.8886314529679196e-05, "loss": 0.17914758920669555, "memory(GiB)": 70.5, "step": 51715, "token_acc": 0.9567901234567902, "train_speed(iter/s)": 1.449812 }, { "epoch": 2.2158433657512533, "grad_norm": 3.1503818035125732, "learning_rate": 5.8879691814423744e-05, "loss": 0.5272690296173096, "memory(GiB)": 70.5, "step": 51720, "token_acc": 0.8982456140350877, "train_speed(iter/s)": 1.449819 }, { "epoch": 2.216057581080502, "grad_norm": 3.8430449962615967, "learning_rate": 5.8873068938304355e-05, "loss": 0.315415358543396, "memory(GiB)": 70.5, "step": 51725, "token_acc": 0.9547325102880658, "train_speed(iter/s)": 1.449831 }, { "epoch": 2.216271796409751, "grad_norm": 4.062384605407715, "learning_rate": 5.8866445901441e-05, "loss": 0.37226340770721433, "memory(GiB)": 70.5, "step": 51730, "token_acc": 0.9287671232876712, "train_speed(iter/s)": 1.449831 }, { "epoch": 2.216486011739, "grad_norm": 1.2742401361465454, "learning_rate": 5.885982270395366e-05, "loss": 0.30116612911224366, "memory(GiB)": 70.5, "step": 51735, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.44983 }, { "epoch": 2.216700227068249, "grad_norm": 4.490481376647949, "learning_rate": 5.885319934596233e-05, "loss": 0.1842116117477417, "memory(GiB)": 70.5, "step": 51740, "token_acc": 0.9646302250803859, "train_speed(iter/s)": 1.449838 }, { "epoch": 2.216914442397498, "grad_norm": 1.8207494020462036, "learning_rate": 5.884657582758698e-05, "loss": 0.2619371175765991, "memory(GiB)": 70.5, "step": 51745, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.449848 }, { "epoch": 2.217128657726747, "grad_norm": 3.8815107345581055, "learning_rate": 5.8839952148947594e-05, "loss": 0.24978852272033691, "memory(GiB)": 70.5, "step": 51750, "token_acc": 0.9419354838709677, "train_speed(iter/s)": 1.449855 }, { "epoch": 2.217342873055996, "grad_norm": 7.706562042236328, "learning_rate": 5.8833328310164215e-05, "loss": 0.48983373641967776, "memory(GiB)": 70.5, "step": 51755, "token_acc": 0.916, "train_speed(iter/s)": 1.449854 }, { "epoch": 2.2175570883852447, "grad_norm": 3.6941452026367188, "learning_rate": 5.882670431135677e-05, "loss": 0.34437811374664307, "memory(GiB)": 70.5, "step": 51760, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.449854 }, { "epoch": 2.217771303714494, "grad_norm": 2.904989719390869, "learning_rate": 5.882008015264532e-05, "loss": 0.3471388816833496, "memory(GiB)": 70.5, "step": 51765, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.449856 }, { "epoch": 2.2179855190437427, "grad_norm": 3.0537476539611816, "learning_rate": 5.8813455834149846e-05, "loss": 0.23482890129089357, "memory(GiB)": 70.5, "step": 51770, "token_acc": 0.9576923076923077, "train_speed(iter/s)": 1.449867 }, { "epoch": 2.2181997343729916, "grad_norm": 4.368670463562012, "learning_rate": 5.880683135599034e-05, "loss": 0.36423916816711427, "memory(GiB)": 70.5, "step": 51775, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.449868 }, { "epoch": 2.218413949702241, "grad_norm": 4.678079605102539, "learning_rate": 5.880020671828683e-05, "loss": 0.471403980255127, "memory(GiB)": 70.5, "step": 51780, "token_acc": 0.8817567567567568, "train_speed(iter/s)": 1.449884 }, { "epoch": 2.2186281650314896, "grad_norm": 3.7022910118103027, "learning_rate": 5.879358192115932e-05, "loss": 0.6368526935577392, "memory(GiB)": 70.5, "step": 51785, "token_acc": 0.8741721854304636, "train_speed(iter/s)": 1.449912 }, { "epoch": 2.2188423803607384, "grad_norm": 0.7959430813789368, "learning_rate": 5.8786956964727834e-05, "loss": 0.24494729042053223, "memory(GiB)": 70.5, "step": 51790, "token_acc": 0.9453376205787781, "train_speed(iter/s)": 1.449916 }, { "epoch": 2.2190565956899877, "grad_norm": 2.68110990524292, "learning_rate": 5.878033184911236e-05, "loss": 0.5054098606109619, "memory(GiB)": 70.5, "step": 51795, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.449912 }, { "epoch": 2.2192708110192365, "grad_norm": 2.9326205253601074, "learning_rate": 5.877370657443294e-05, "loss": 0.2690433025360107, "memory(GiB)": 70.5, "step": 51800, "token_acc": 0.937007874015748, "train_speed(iter/s)": 1.449916 }, { "epoch": 2.2194850263484853, "grad_norm": 4.653321743011475, "learning_rate": 5.876708114080961e-05, "loss": 0.27941868305206297, "memory(GiB)": 70.5, "step": 51805, "token_acc": 0.9549295774647887, "train_speed(iter/s)": 1.449913 }, { "epoch": 2.2196992416777346, "grad_norm": 5.816536903381348, "learning_rate": 5.876045554836237e-05, "loss": 0.21439528465270996, "memory(GiB)": 70.5, "step": 51810, "token_acc": 0.948051948051948, "train_speed(iter/s)": 1.449913 }, { "epoch": 2.2199134570069834, "grad_norm": 0.8606597185134888, "learning_rate": 5.875382979721127e-05, "loss": 0.3664278268814087, "memory(GiB)": 70.5, "step": 51815, "token_acc": 0.9300911854103343, "train_speed(iter/s)": 1.449916 }, { "epoch": 2.220127672336232, "grad_norm": 3.6307766437530518, "learning_rate": 5.874720388747632e-05, "loss": 0.39805192947387696, "memory(GiB)": 70.5, "step": 51820, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.44992 }, { "epoch": 2.2203418876654815, "grad_norm": 7.135588645935059, "learning_rate": 5.874057781927756e-05, "loss": 0.3095508813858032, "memory(GiB)": 70.5, "step": 51825, "token_acc": 0.940677966101695, "train_speed(iter/s)": 1.449911 }, { "epoch": 2.2205561029947303, "grad_norm": 4.107687950134277, "learning_rate": 5.8733951592735045e-05, "loss": 0.6605735778808594, "memory(GiB)": 70.5, "step": 51830, "token_acc": 0.8653846153846154, "train_speed(iter/s)": 1.449918 }, { "epoch": 2.220770318323979, "grad_norm": 1.9040714502334595, "learning_rate": 5.8727325207968806e-05, "loss": 0.22101821899414062, "memory(GiB)": 70.5, "step": 51835, "token_acc": 0.94921875, "train_speed(iter/s)": 1.44992 }, { "epoch": 2.2209845336532283, "grad_norm": 1.0491713285446167, "learning_rate": 5.872069866509887e-05, "loss": 0.3612942934036255, "memory(GiB)": 70.5, "step": 51840, "token_acc": 0.924901185770751, "train_speed(iter/s)": 1.449912 }, { "epoch": 2.221198748982477, "grad_norm": 3.558867931365967, "learning_rate": 5.871407196424532e-05, "loss": 0.26006624698638914, "memory(GiB)": 70.5, "step": 51845, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.449915 }, { "epoch": 2.221412964311726, "grad_norm": 6.494116306304932, "learning_rate": 5.870744510552817e-05, "loss": 0.6002049446105957, "memory(GiB)": 70.5, "step": 51850, "token_acc": 0.8884758364312267, "train_speed(iter/s)": 1.449955 }, { "epoch": 2.2216271796409752, "grad_norm": 3.6415743827819824, "learning_rate": 5.8700818089067474e-05, "loss": 0.5776042938232422, "memory(GiB)": 70.5, "step": 51855, "token_acc": 0.8931297709923665, "train_speed(iter/s)": 1.449964 }, { "epoch": 2.221841394970224, "grad_norm": 2.310748815536499, "learning_rate": 5.8694190914983317e-05, "loss": 0.34522697925567625, "memory(GiB)": 70.5, "step": 51860, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.44997 }, { "epoch": 2.222055610299473, "grad_norm": 4.885250091552734, "learning_rate": 5.868756358339572e-05, "loss": 0.5774139404296875, "memory(GiB)": 70.5, "step": 51865, "token_acc": 0.8541666666666666, "train_speed(iter/s)": 1.449988 }, { "epoch": 2.222269825628722, "grad_norm": 2.338010311126709, "learning_rate": 5.8680936094424754e-05, "loss": 0.24787886142730714, "memory(GiB)": 70.5, "step": 51870, "token_acc": 0.9453924914675768, "train_speed(iter/s)": 1.449997 }, { "epoch": 2.222484040957971, "grad_norm": 3.0369317531585693, "learning_rate": 5.8674308448190506e-05, "loss": 0.25834126472473146, "memory(GiB)": 70.5, "step": 51875, "token_acc": 0.942652329749104, "train_speed(iter/s)": 1.449993 }, { "epoch": 2.2226982562872197, "grad_norm": 0.24679502844810486, "learning_rate": 5.8667680644813005e-05, "loss": 0.21809329986572265, "memory(GiB)": 70.5, "step": 51880, "token_acc": 0.9482071713147411, "train_speed(iter/s)": 1.449997 }, { "epoch": 2.222912471616469, "grad_norm": 3.2286267280578613, "learning_rate": 5.8661052684412354e-05, "loss": 0.4511703014373779, "memory(GiB)": 70.5, "step": 51885, "token_acc": 0.8984615384615384, "train_speed(iter/s)": 1.449998 }, { "epoch": 2.223126686945718, "grad_norm": 4.690120697021484, "learning_rate": 5.86544245671086e-05, "loss": 0.6938035488128662, "memory(GiB)": 70.5, "step": 51890, "token_acc": 0.8542372881355932, "train_speed(iter/s)": 1.450002 }, { "epoch": 2.2233409022749666, "grad_norm": 1.797331690788269, "learning_rate": 5.8647796293021826e-05, "loss": 0.32368154525756837, "memory(GiB)": 70.5, "step": 51895, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.450001 }, { "epoch": 2.223555117604216, "grad_norm": 5.67279577255249, "learning_rate": 5.864116786227212e-05, "loss": 0.6454895973205567, "memory(GiB)": 70.5, "step": 51900, "token_acc": 0.8682432432432432, "train_speed(iter/s)": 1.449997 }, { "epoch": 2.2237693329334647, "grad_norm": 4.15754508972168, "learning_rate": 5.863453927497954e-05, "loss": 0.22310004234313965, "memory(GiB)": 70.5, "step": 51905, "token_acc": 0.952054794520548, "train_speed(iter/s)": 1.44999 }, { "epoch": 2.2239835482627135, "grad_norm": 5.471196174621582, "learning_rate": 5.8627910531264176e-05, "loss": 0.36933557987213134, "memory(GiB)": 70.5, "step": 51910, "token_acc": 0.92, "train_speed(iter/s)": 1.449979 }, { "epoch": 2.2241977635919628, "grad_norm": 3.3487257957458496, "learning_rate": 5.862128163124613e-05, "loss": 0.44631519317626955, "memory(GiB)": 70.5, "step": 51915, "token_acc": 0.8945686900958466, "train_speed(iter/s)": 1.449993 }, { "epoch": 2.2244119789212116, "grad_norm": 2.150068759918213, "learning_rate": 5.861465257504548e-05, "loss": 0.343535041809082, "memory(GiB)": 70.5, "step": 51920, "token_acc": 0.924812030075188, "train_speed(iter/s)": 1.449992 }, { "epoch": 2.2246261942504604, "grad_norm": 2.5648670196533203, "learning_rate": 5.8608023362782316e-05, "loss": 0.28977968692779543, "memory(GiB)": 70.5, "step": 51925, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.449997 }, { "epoch": 2.2248404095797096, "grad_norm": 2.6238832473754883, "learning_rate": 5.8601393994576734e-05, "loss": 0.49546422958374026, "memory(GiB)": 70.5, "step": 51930, "token_acc": 0.8693009118541033, "train_speed(iter/s)": 1.44999 }, { "epoch": 2.2250546249089584, "grad_norm": 4.209255218505859, "learning_rate": 5.859476447054884e-05, "loss": 0.3869462013244629, "memory(GiB)": 70.5, "step": 51935, "token_acc": 0.9294478527607362, "train_speed(iter/s)": 1.449986 }, { "epoch": 2.2252688402382073, "grad_norm": 2.630281448364258, "learning_rate": 5.8588134790818707e-05, "loss": 0.26996841430664065, "memory(GiB)": 70.5, "step": 51940, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.449981 }, { "epoch": 2.2254830555674565, "grad_norm": 2.3545429706573486, "learning_rate": 5.858150495550646e-05, "loss": 0.27882416248321534, "memory(GiB)": 70.5, "step": 51945, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.449981 }, { "epoch": 2.2256972708967053, "grad_norm": 3.340320110321045, "learning_rate": 5.857487496473221e-05, "loss": 0.391330623626709, "memory(GiB)": 70.5, "step": 51950, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.449996 }, { "epoch": 2.225911486225954, "grad_norm": 2.740168571472168, "learning_rate": 5.856824481861605e-05, "loss": 0.46643381118774413, "memory(GiB)": 70.5, "step": 51955, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.449994 }, { "epoch": 2.2261257015552034, "grad_norm": 4.273463249206543, "learning_rate": 5.85616145172781e-05, "loss": 0.17190606594085694, "memory(GiB)": 70.5, "step": 51960, "token_acc": 0.9446254071661238, "train_speed(iter/s)": 1.449993 }, { "epoch": 2.226339916884452, "grad_norm": 3.0977859497070312, "learning_rate": 5.855498406083847e-05, "loss": 0.49559311866760253, "memory(GiB)": 70.5, "step": 51965, "token_acc": 0.886986301369863, "train_speed(iter/s)": 1.450026 }, { "epoch": 2.226554132213701, "grad_norm": 1.129059076309204, "learning_rate": 5.854835344941727e-05, "loss": 0.2545957088470459, "memory(GiB)": 70.5, "step": 51970, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.450027 }, { "epoch": 2.2267683475429503, "grad_norm": 3.7239654064178467, "learning_rate": 5.854172268313465e-05, "loss": 0.48371243476867676, "memory(GiB)": 70.5, "step": 51975, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.45003 }, { "epoch": 2.226982562872199, "grad_norm": 3.681192398071289, "learning_rate": 5.8535091762110695e-05, "loss": 0.5693175792694092, "memory(GiB)": 70.5, "step": 51980, "token_acc": 0.8773584905660378, "train_speed(iter/s)": 1.450019 }, { "epoch": 2.227196778201448, "grad_norm": 1.6798123121261597, "learning_rate": 5.852846068646554e-05, "loss": 0.4093451023101807, "memory(GiB)": 70.5, "step": 51985, "token_acc": 0.8943661971830986, "train_speed(iter/s)": 1.450036 }, { "epoch": 2.227410993530697, "grad_norm": 4.795599937438965, "learning_rate": 5.8521829456319334e-05, "loss": 0.4179224014282227, "memory(GiB)": 70.5, "step": 51990, "token_acc": 0.8880597014925373, "train_speed(iter/s)": 1.450035 }, { "epoch": 2.227625208859946, "grad_norm": 5.838315010070801, "learning_rate": 5.851519807179219e-05, "loss": 0.4582851409912109, "memory(GiB)": 70.5, "step": 51995, "token_acc": 0.9053627760252366, "train_speed(iter/s)": 1.450041 }, { "epoch": 2.227839424189195, "grad_norm": 3.6778085231781006, "learning_rate": 5.850856653300424e-05, "loss": 0.25558791160583494, "memory(GiB)": 70.5, "step": 52000, "token_acc": 0.941717791411043, "train_speed(iter/s)": 1.450033 }, { "epoch": 2.227839424189195, "eval_loss": 2.4211320877075195, "eval_runtime": 14.121, "eval_samples_per_second": 7.082, "eval_steps_per_second": 7.082, "eval_token_acc": 0.44623655913978494, "step": 52000 }, { "epoch": 2.228053639518444, "grad_norm": 2.9643747806549072, "learning_rate": 5.850193484007563e-05, "loss": 0.30959696769714357, "memory(GiB)": 70.5, "step": 52005, "token_acc": 0.5878442545109211, "train_speed(iter/s)": 1.449432 }, { "epoch": 2.228267854847693, "grad_norm": 1.9665124416351318, "learning_rate": 5.849530299312649e-05, "loss": 0.17825782299041748, "memory(GiB)": 70.5, "step": 52010, "token_acc": 0.95578231292517, "train_speed(iter/s)": 1.44944 }, { "epoch": 2.2284820701769417, "grad_norm": 6.740762233734131, "learning_rate": 5.848867099227696e-05, "loss": 0.7790162086486816, "memory(GiB)": 70.5, "step": 52015, "token_acc": 0.8670694864048338, "train_speed(iter/s)": 1.449452 }, { "epoch": 2.228696285506191, "grad_norm": 1.3156815767288208, "learning_rate": 5.848203883764721e-05, "loss": 0.39795761108398436, "memory(GiB)": 70.5, "step": 52020, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.449469 }, { "epoch": 2.2289105008354397, "grad_norm": 0.976874828338623, "learning_rate": 5.8475406529357356e-05, "loss": 0.11487489938735962, "memory(GiB)": 70.5, "step": 52025, "token_acc": 0.954225352112676, "train_speed(iter/s)": 1.449475 }, { "epoch": 2.2291247161646885, "grad_norm": 0.2605103552341461, "learning_rate": 5.8468774067527575e-05, "loss": 0.2997324228286743, "memory(GiB)": 70.5, "step": 52030, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.449477 }, { "epoch": 2.229338931493938, "grad_norm": 1.858055591583252, "learning_rate": 5.8462141452277995e-05, "loss": 0.19895342588424683, "memory(GiB)": 70.5, "step": 52035, "token_acc": 0.96484375, "train_speed(iter/s)": 1.449494 }, { "epoch": 2.2295531468231866, "grad_norm": 4.758909225463867, "learning_rate": 5.845550868372879e-05, "loss": 0.3157306432723999, "memory(GiB)": 70.5, "step": 52040, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.449501 }, { "epoch": 2.2297673621524354, "grad_norm": 6.597296714782715, "learning_rate": 5.844887576200012e-05, "loss": 0.7061738967895508, "memory(GiB)": 70.5, "step": 52045, "token_acc": 0.8540925266903915, "train_speed(iter/s)": 1.449507 }, { "epoch": 2.2299815774816847, "grad_norm": 6.494606018066406, "learning_rate": 5.844224268721214e-05, "loss": 0.37275080680847167, "memory(GiB)": 70.5, "step": 52050, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.449503 }, { "epoch": 2.2301957928109335, "grad_norm": 1.7185115814208984, "learning_rate": 5.843560945948499e-05, "loss": 0.33068509101867677, "memory(GiB)": 70.5, "step": 52055, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.449518 }, { "epoch": 2.2304100081401823, "grad_norm": 2.3445537090301514, "learning_rate": 5.8428976078938877e-05, "loss": 0.283681321144104, "memory(GiB)": 70.5, "step": 52060, "token_acc": 0.9400749063670412, "train_speed(iter/s)": 1.449535 }, { "epoch": 2.2306242234694316, "grad_norm": 0.5587276220321655, "learning_rate": 5.842234254569396e-05, "loss": 0.18108830451965333, "memory(GiB)": 70.5, "step": 52065, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.44954 }, { "epoch": 2.2308384387986804, "grad_norm": 2.3909056186676025, "learning_rate": 5.84157088598704e-05, "loss": 0.366547679901123, "memory(GiB)": 70.5, "step": 52070, "token_acc": 0.9057750759878419, "train_speed(iter/s)": 1.449539 }, { "epoch": 2.231052654127929, "grad_norm": 7.275413513183594, "learning_rate": 5.840907502158839e-05, "loss": 0.6313671588897705, "memory(GiB)": 70.5, "step": 52075, "token_acc": 0.8715953307392996, "train_speed(iter/s)": 1.449537 }, { "epoch": 2.2312668694571784, "grad_norm": 0.8404770493507385, "learning_rate": 5.84024410309681e-05, "loss": 0.7675891399383545, "memory(GiB)": 70.5, "step": 52080, "token_acc": 0.8542274052478134, "train_speed(iter/s)": 1.449551 }, { "epoch": 2.2314810847864273, "grad_norm": 3.242980480194092, "learning_rate": 5.839580688812969e-05, "loss": 0.2815993309020996, "memory(GiB)": 70.5, "step": 52085, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.449549 }, { "epoch": 2.231695300115676, "grad_norm": 4.894441604614258, "learning_rate": 5.8389172593193365e-05, "loss": 0.3491365432739258, "memory(GiB)": 70.5, "step": 52090, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.449544 }, { "epoch": 2.2319095154449253, "grad_norm": 3.300809383392334, "learning_rate": 5.838253814627932e-05, "loss": 0.5104000091552734, "memory(GiB)": 70.5, "step": 52095, "token_acc": 0.9017857142857143, "train_speed(iter/s)": 1.449552 }, { "epoch": 2.232123730774174, "grad_norm": 6.306825637817383, "learning_rate": 5.8375903547507724e-05, "loss": 0.5785128116607666, "memory(GiB)": 70.5, "step": 52100, "token_acc": 0.8892307692307693, "train_speed(iter/s)": 1.449587 }, { "epoch": 2.232337946103423, "grad_norm": 5.565030574798584, "learning_rate": 5.836926879699879e-05, "loss": 0.33343145847320554, "memory(GiB)": 70.5, "step": 52105, "token_acc": 0.9264069264069265, "train_speed(iter/s)": 1.449585 }, { "epoch": 2.232552161432672, "grad_norm": 3.0183873176574707, "learning_rate": 5.83626338948727e-05, "loss": 0.4817360401153564, "memory(GiB)": 70.5, "step": 52110, "token_acc": 0.8916967509025271, "train_speed(iter/s)": 1.449589 }, { "epoch": 2.232766376761921, "grad_norm": 6.62835693359375, "learning_rate": 5.835599884124964e-05, "loss": 0.6601252079010009, "memory(GiB)": 70.5, "step": 52115, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.449592 }, { "epoch": 2.2329805920911703, "grad_norm": 3.840653657913208, "learning_rate": 5.8349363636249835e-05, "loss": 0.3552763223648071, "memory(GiB)": 70.5, "step": 52120, "token_acc": 0.9068100358422939, "train_speed(iter/s)": 1.449589 }, { "epoch": 2.233194807420419, "grad_norm": 2.5838027000427246, "learning_rate": 5.834272827999345e-05, "loss": 0.2584064960479736, "memory(GiB)": 70.5, "step": 52125, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.449594 }, { "epoch": 2.233409022749668, "grad_norm": 5.313626766204834, "learning_rate": 5.833609277260074e-05, "loss": 0.39726862907409666, "memory(GiB)": 70.5, "step": 52130, "token_acc": 0.92, "train_speed(iter/s)": 1.449622 }, { "epoch": 2.233623238078917, "grad_norm": 2.1336252689361572, "learning_rate": 5.8329457114191886e-05, "loss": 0.25207335948944093, "memory(GiB)": 70.5, "step": 52135, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.449627 }, { "epoch": 2.233837453408166, "grad_norm": 3.453535318374634, "learning_rate": 5.832282130488711e-05, "loss": 0.48475074768066406, "memory(GiB)": 70.5, "step": 52140, "token_acc": 0.9029850746268657, "train_speed(iter/s)": 1.449631 }, { "epoch": 2.234051668737415, "grad_norm": 2.134890079498291, "learning_rate": 5.8316185344806596e-05, "loss": 0.30536465644836425, "memory(GiB)": 70.5, "step": 52145, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.44964 }, { "epoch": 2.234265884066664, "grad_norm": 0.6115723252296448, "learning_rate": 5.8309549234070605e-05, "loss": 0.4223916053771973, "memory(GiB)": 70.5, "step": 52150, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.449646 }, { "epoch": 2.234480099395913, "grad_norm": 3.2320284843444824, "learning_rate": 5.8302912972799315e-05, "loss": 0.22940196990966796, "memory(GiB)": 70.5, "step": 52155, "token_acc": 0.9494584837545126, "train_speed(iter/s)": 1.449648 }, { "epoch": 2.2346943147251617, "grad_norm": 1.1332906484603882, "learning_rate": 5.8296276561112985e-05, "loss": 0.20771193504333496, "memory(GiB)": 70.5, "step": 52160, "token_acc": 0.9541666666666667, "train_speed(iter/s)": 1.449648 }, { "epoch": 2.234908530054411, "grad_norm": 2.3687613010406494, "learning_rate": 5.828963999913182e-05, "loss": 0.5951743602752686, "memory(GiB)": 70.5, "step": 52165, "token_acc": 0.8895348837209303, "train_speed(iter/s)": 1.449651 }, { "epoch": 2.2351227453836597, "grad_norm": 0.3470629155635834, "learning_rate": 5.8283003286976035e-05, "loss": 0.24321603775024414, "memory(GiB)": 70.5, "step": 52170, "token_acc": 0.9503311258278145, "train_speed(iter/s)": 1.44965 }, { "epoch": 2.2353369607129085, "grad_norm": 2.2655677795410156, "learning_rate": 5.827636642476589e-05, "loss": 0.19193859100341798, "memory(GiB)": 70.5, "step": 52175, "token_acc": 0.9522292993630573, "train_speed(iter/s)": 1.449661 }, { "epoch": 2.235551176042158, "grad_norm": 4.820706844329834, "learning_rate": 5.826972941262161e-05, "loss": 0.4374522686004639, "memory(GiB)": 70.5, "step": 52180, "token_acc": 0.8901515151515151, "train_speed(iter/s)": 1.449674 }, { "epoch": 2.2357653913714066, "grad_norm": 1.3081341981887817, "learning_rate": 5.826309225066341e-05, "loss": 0.36432180404663084, "memory(GiB)": 70.5, "step": 52185, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.449684 }, { "epoch": 2.2359796067006554, "grad_norm": 4.959054470062256, "learning_rate": 5.825645493901155e-05, "loss": 0.8540868759155273, "memory(GiB)": 70.5, "step": 52190, "token_acc": 0.8402366863905325, "train_speed(iter/s)": 1.449682 }, { "epoch": 2.2361938220299047, "grad_norm": 2.171531915664673, "learning_rate": 5.824981747778626e-05, "loss": 0.3841434955596924, "memory(GiB)": 70.5, "step": 52195, "token_acc": 0.8980891719745223, "train_speed(iter/s)": 1.44969 }, { "epoch": 2.2364080373591535, "grad_norm": 2.713148593902588, "learning_rate": 5.824317986710778e-05, "loss": 0.23450865745544433, "memory(GiB)": 70.5, "step": 52200, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.449691 }, { "epoch": 2.2366222526884023, "grad_norm": 2.9896273612976074, "learning_rate": 5.823654210709637e-05, "loss": 0.599562406539917, "memory(GiB)": 70.5, "step": 52205, "token_acc": 0.8607242339832869, "train_speed(iter/s)": 1.449703 }, { "epoch": 2.2368364680176516, "grad_norm": 3.2028040885925293, "learning_rate": 5.8229904197872284e-05, "loss": 0.44502873420715333, "memory(GiB)": 70.5, "step": 52210, "token_acc": 0.9212328767123288, "train_speed(iter/s)": 1.44969 }, { "epoch": 2.2370506833469004, "grad_norm": 3.990720510482788, "learning_rate": 5.822326613955574e-05, "loss": 0.41907339096069335, "memory(GiB)": 70.5, "step": 52215, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.449699 }, { "epoch": 2.237264898676149, "grad_norm": 1.3920363187789917, "learning_rate": 5.821662793226704e-05, "loss": 0.31293818950653074, "memory(GiB)": 70.5, "step": 52220, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.449695 }, { "epoch": 2.2374791140053985, "grad_norm": 4.155309677124023, "learning_rate": 5.820998957612641e-05, "loss": 0.38309240341186523, "memory(GiB)": 70.5, "step": 52225, "token_acc": 0.9155844155844156, "train_speed(iter/s)": 1.449699 }, { "epoch": 2.2376933293346473, "grad_norm": 0.6575645804405212, "learning_rate": 5.820335107125412e-05, "loss": 0.2978542327880859, "memory(GiB)": 70.5, "step": 52230, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.449702 }, { "epoch": 2.237907544663896, "grad_norm": 1.2977255582809448, "learning_rate": 5.819671241777043e-05, "loss": 0.29032995700836184, "memory(GiB)": 70.5, "step": 52235, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.449717 }, { "epoch": 2.2381217599931453, "grad_norm": 1.6999094486236572, "learning_rate": 5.819007361579558e-05, "loss": 0.4566225051879883, "memory(GiB)": 70.5, "step": 52240, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.449732 }, { "epoch": 2.238335975322394, "grad_norm": 0.12810122966766357, "learning_rate": 5.818343466544989e-05, "loss": 0.3187387704849243, "memory(GiB)": 70.5, "step": 52245, "token_acc": 0.9357798165137615, "train_speed(iter/s)": 1.449733 }, { "epoch": 2.238550190651643, "grad_norm": 3.4493041038513184, "learning_rate": 5.8176795566853606e-05, "loss": 0.25446171760559083, "memory(GiB)": 70.5, "step": 52250, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.449749 }, { "epoch": 2.238764405980892, "grad_norm": 6.196792125701904, "learning_rate": 5.817015632012699e-05, "loss": 0.2755770683288574, "memory(GiB)": 70.5, "step": 52255, "token_acc": 0.9323843416370107, "train_speed(iter/s)": 1.449758 }, { "epoch": 2.238978621310141, "grad_norm": 11.474093437194824, "learning_rate": 5.816351692539033e-05, "loss": 0.4658053398132324, "memory(GiB)": 70.5, "step": 52260, "token_acc": 0.8831615120274914, "train_speed(iter/s)": 1.449788 }, { "epoch": 2.23919283663939, "grad_norm": 3.4774158000946045, "learning_rate": 5.81568773827639e-05, "loss": 0.3389065027236938, "memory(GiB)": 70.5, "step": 52265, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.449794 }, { "epoch": 2.239407051968639, "grad_norm": 2.4239206314086914, "learning_rate": 5.815023769236798e-05, "loss": 0.2691312313079834, "memory(GiB)": 70.5, "step": 52270, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.449798 }, { "epoch": 2.239621267297888, "grad_norm": 3.685058355331421, "learning_rate": 5.814359785432286e-05, "loss": 0.46241350173950196, "memory(GiB)": 70.5, "step": 52275, "token_acc": 0.9147540983606557, "train_speed(iter/s)": 1.449802 }, { "epoch": 2.2398354826271367, "grad_norm": 3.6863245964050293, "learning_rate": 5.8136957868748844e-05, "loss": 0.39353864192962645, "memory(GiB)": 70.5, "step": 52280, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.449813 }, { "epoch": 2.240049697956386, "grad_norm": 4.49826192855835, "learning_rate": 5.813031773576618e-05, "loss": 0.35973904132843015, "memory(GiB)": 70.5, "step": 52285, "token_acc": 0.9085173501577287, "train_speed(iter/s)": 1.449824 }, { "epoch": 2.240263913285635, "grad_norm": 3.0505995750427246, "learning_rate": 5.81236774554952e-05, "loss": 0.3769645690917969, "memory(GiB)": 70.5, "step": 52290, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.44983 }, { "epoch": 2.2404781286148836, "grad_norm": 2.505516290664673, "learning_rate": 5.811703702805618e-05, "loss": 0.4104153156280518, "memory(GiB)": 70.5, "step": 52295, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.449859 }, { "epoch": 2.240692343944133, "grad_norm": 5.056509971618652, "learning_rate": 5.811039645356941e-05, "loss": 0.4438169956207275, "memory(GiB)": 70.5, "step": 52300, "token_acc": 0.8939393939393939, "train_speed(iter/s)": 1.449858 }, { "epoch": 2.2409065592733817, "grad_norm": 3.8106203079223633, "learning_rate": 5.810375573215521e-05, "loss": 0.5517570018768311, "memory(GiB)": 70.5, "step": 52305, "token_acc": 0.889196675900277, "train_speed(iter/s)": 1.449858 }, { "epoch": 2.2411207746026305, "grad_norm": 1.1975123882293701, "learning_rate": 5.809711486393388e-05, "loss": 0.2654069423675537, "memory(GiB)": 70.5, "step": 52310, "token_acc": 0.9365558912386707, "train_speed(iter/s)": 1.449878 }, { "epoch": 2.2413349899318797, "grad_norm": 5.38698673248291, "learning_rate": 5.8090473849025685e-05, "loss": 0.37932658195495605, "memory(GiB)": 70.5, "step": 52315, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.449883 }, { "epoch": 2.2415492052611286, "grad_norm": 2.297485589981079, "learning_rate": 5.8083832687551e-05, "loss": 0.32621941566467283, "memory(GiB)": 70.5, "step": 52320, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.449883 }, { "epoch": 2.2417634205903774, "grad_norm": 4.498394012451172, "learning_rate": 5.807719137963009e-05, "loss": 0.27379276752471926, "memory(GiB)": 70.5, "step": 52325, "token_acc": 0.9402390438247012, "train_speed(iter/s)": 1.449883 }, { "epoch": 2.2419776359196266, "grad_norm": 2.4017460346221924, "learning_rate": 5.807054992538328e-05, "loss": 0.18548257350921632, "memory(GiB)": 70.5, "step": 52330, "token_acc": 0.9585798816568047, "train_speed(iter/s)": 1.449883 }, { "epoch": 2.2421918512488754, "grad_norm": 4.3774638175964355, "learning_rate": 5.806390832493089e-05, "loss": 0.3778779745101929, "memory(GiB)": 70.5, "step": 52335, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.449912 }, { "epoch": 2.2424060665781242, "grad_norm": 2.456810712814331, "learning_rate": 5.805726657839324e-05, "loss": 0.2800804376602173, "memory(GiB)": 70.5, "step": 52340, "token_acc": 0.9532467532467532, "train_speed(iter/s)": 1.449913 }, { "epoch": 2.2426202819073735, "grad_norm": 3.5406861305236816, "learning_rate": 5.805062468589064e-05, "loss": 0.2696718454360962, "memory(GiB)": 70.5, "step": 52345, "token_acc": 0.9429657794676806, "train_speed(iter/s)": 1.449903 }, { "epoch": 2.2428344972366223, "grad_norm": 2.926668405532837, "learning_rate": 5.8043982647543426e-05, "loss": 0.3851790428161621, "memory(GiB)": 70.5, "step": 52350, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.44991 }, { "epoch": 2.243048712565871, "grad_norm": 5.329750061035156, "learning_rate": 5.803734046347192e-05, "loss": 0.35979812145233153, "memory(GiB)": 70.5, "step": 52355, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.449917 }, { "epoch": 2.2432629278951204, "grad_norm": 5.003969669342041, "learning_rate": 5.8030698133796445e-05, "loss": 0.27737994194030763, "memory(GiB)": 70.5, "step": 52360, "token_acc": 0.9392712550607287, "train_speed(iter/s)": 1.449925 }, { "epoch": 2.243477143224369, "grad_norm": 1.402706265449524, "learning_rate": 5.802405565863735e-05, "loss": 0.35401718616485595, "memory(GiB)": 70.5, "step": 52365, "token_acc": 0.9312714776632303, "train_speed(iter/s)": 1.449928 }, { "epoch": 2.243691358553618, "grad_norm": 3.6275134086608887, "learning_rate": 5.8017413038114965e-05, "loss": 0.41221070289611816, "memory(GiB)": 70.5, "step": 52370, "token_acc": 0.9144736842105263, "train_speed(iter/s)": 1.449942 }, { "epoch": 2.2439055738828673, "grad_norm": 3.1742467880249023, "learning_rate": 5.8010770272349615e-05, "loss": 0.26667261123657227, "memory(GiB)": 70.5, "step": 52375, "token_acc": 0.9372822299651568, "train_speed(iter/s)": 1.449953 }, { "epoch": 2.244119789212116, "grad_norm": 5.101417541503906, "learning_rate": 5.8004127361461644e-05, "loss": 0.538941764831543, "memory(GiB)": 70.5, "step": 52380, "token_acc": 0.8654434250764526, "train_speed(iter/s)": 1.449956 }, { "epoch": 2.244334004541365, "grad_norm": 2.237912654876709, "learning_rate": 5.799748430557139e-05, "loss": 0.6118401050567627, "memory(GiB)": 70.5, "step": 52385, "token_acc": 0.8859934853420195, "train_speed(iter/s)": 1.449952 }, { "epoch": 2.244548219870614, "grad_norm": 1.192143201828003, "learning_rate": 5.799084110479921e-05, "loss": 0.19035494327545166, "memory(GiB)": 70.5, "step": 52390, "token_acc": 0.9628571428571429, "train_speed(iter/s)": 1.449954 }, { "epoch": 2.244762435199863, "grad_norm": 1.5636405944824219, "learning_rate": 5.798419775926546e-05, "loss": 0.3797930717468262, "memory(GiB)": 70.5, "step": 52395, "token_acc": 0.9036544850498339, "train_speed(iter/s)": 1.449961 }, { "epoch": 2.2449766505291118, "grad_norm": 2.285092830657959, "learning_rate": 5.7977554269090475e-05, "loss": 0.2745864152908325, "memory(GiB)": 70.5, "step": 52400, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.44998 }, { "epoch": 2.245190865858361, "grad_norm": 1.4235252141952515, "learning_rate": 5.7970910634394594e-05, "loss": 0.21388912200927734, "memory(GiB)": 70.5, "step": 52405, "token_acc": 0.9488188976377953, "train_speed(iter/s)": 1.449997 }, { "epoch": 2.24540508118761, "grad_norm": 4.338217258453369, "learning_rate": 5.796426685529821e-05, "loss": 0.47854185104370117, "memory(GiB)": 70.5, "step": 52410, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.450007 }, { "epoch": 2.2456192965168587, "grad_norm": 5.65863037109375, "learning_rate": 5.795762293192164e-05, "loss": 0.5837369918823242, "memory(GiB)": 70.5, "step": 52415, "token_acc": 0.8877887788778878, "train_speed(iter/s)": 1.450017 }, { "epoch": 2.245833511846108, "grad_norm": 2.9844653606414795, "learning_rate": 5.7950978864385286e-05, "loss": 0.2246945858001709, "memory(GiB)": 70.5, "step": 52420, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.450031 }, { "epoch": 2.2460477271753567, "grad_norm": 3.5591013431549072, "learning_rate": 5.7944334652809485e-05, "loss": 0.31027586460113527, "memory(GiB)": 70.5, "step": 52425, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.45003 }, { "epoch": 2.2462619425046055, "grad_norm": 5.291460037231445, "learning_rate": 5.7937690297314594e-05, "loss": 0.7970645904541016, "memory(GiB)": 70.5, "step": 52430, "token_acc": 0.8587360594795539, "train_speed(iter/s)": 1.450028 }, { "epoch": 2.246476157833855, "grad_norm": 1.3253092765808105, "learning_rate": 5.793104579802102e-05, "loss": 0.32525787353515623, "memory(GiB)": 70.5, "step": 52435, "token_acc": 0.9228486646884273, "train_speed(iter/s)": 1.450034 }, { "epoch": 2.2466903731631036, "grad_norm": 2.2848453521728516, "learning_rate": 5.79244011550491e-05, "loss": 0.14725769758224488, "memory(GiB)": 70.5, "step": 52440, "token_acc": 0.9712230215827338, "train_speed(iter/s)": 1.450036 }, { "epoch": 2.2469045884923524, "grad_norm": 3.907160997390747, "learning_rate": 5.7917756368519217e-05, "loss": 0.45122594833374025, "memory(GiB)": 70.5, "step": 52445, "token_acc": 0.889763779527559, "train_speed(iter/s)": 1.450037 }, { "epoch": 2.2471188038216017, "grad_norm": 4.853945732116699, "learning_rate": 5.7911111438551754e-05, "loss": 0.6026165962219239, "memory(GiB)": 70.5, "step": 52450, "token_acc": 0.8878504672897196, "train_speed(iter/s)": 1.450038 }, { "epoch": 2.2473330191508505, "grad_norm": 3.2808356285095215, "learning_rate": 5.7904466365267097e-05, "loss": 0.4054673671722412, "memory(GiB)": 70.5, "step": 52455, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.450048 }, { "epoch": 2.2475472344800993, "grad_norm": 4.520842552185059, "learning_rate": 5.789782114878559e-05, "loss": 0.49213442802429197, "memory(GiB)": 70.5, "step": 52460, "token_acc": 0.9081632653061225, "train_speed(iter/s)": 1.450046 }, { "epoch": 2.2477614498093486, "grad_norm": 1.3325215578079224, "learning_rate": 5.789117578922767e-05, "loss": 0.38770012855529784, "memory(GiB)": 70.5, "step": 52465, "token_acc": 0.9201520912547528, "train_speed(iter/s)": 1.450075 }, { "epoch": 2.2479756651385974, "grad_norm": 0.24046337604522705, "learning_rate": 5.7884530286713687e-05, "loss": 0.10555834770202636, "memory(GiB)": 70.5, "step": 52470, "token_acc": 0.9678571428571429, "train_speed(iter/s)": 1.450082 }, { "epoch": 2.248189880467846, "grad_norm": 2.5445492267608643, "learning_rate": 5.787788464136403e-05, "loss": 0.23370919227600098, "memory(GiB)": 70.5, "step": 52475, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.450092 }, { "epoch": 2.2484040957970954, "grad_norm": 2.353748083114624, "learning_rate": 5.787123885329913e-05, "loss": 0.19180119037628174, "memory(GiB)": 70.5, "step": 52480, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.450087 }, { "epoch": 2.2486183111263442, "grad_norm": 1.2690510749816895, "learning_rate": 5.786459292263934e-05, "loss": 0.30581321716308596, "memory(GiB)": 70.5, "step": 52485, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.450087 }, { "epoch": 2.248832526455593, "grad_norm": 7.001256942749023, "learning_rate": 5.785794684950506e-05, "loss": 0.46362671852111814, "memory(GiB)": 70.5, "step": 52490, "token_acc": 0.8845070422535212, "train_speed(iter/s)": 1.450102 }, { "epoch": 2.2490467417848423, "grad_norm": 3.242931365966797, "learning_rate": 5.7851300634016724e-05, "loss": 0.30791757106781004, "memory(GiB)": 70.5, "step": 52495, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.450107 }, { "epoch": 2.249260957114091, "grad_norm": 4.177306175231934, "learning_rate": 5.784465427629469e-05, "loss": 0.3549293041229248, "memory(GiB)": 70.5, "step": 52500, "token_acc": 0.925, "train_speed(iter/s)": 1.450111 }, { "epoch": 2.249260957114091, "eval_loss": 2.4387519359588623, "eval_runtime": 13.5386, "eval_samples_per_second": 7.386, "eval_steps_per_second": 7.386, "eval_token_acc": 0.4577294685990338, "step": 52500 }, { "epoch": 2.24947517244334, "grad_norm": 5.028260707855225, "learning_rate": 5.783800777645939e-05, "loss": 0.41315221786499023, "memory(GiB)": 70.5, "step": 52505, "token_acc": 0.5681186283595923, "train_speed(iter/s)": 1.449525 }, { "epoch": 2.249689387772589, "grad_norm": 2.525174856185913, "learning_rate": 5.783136113463125e-05, "loss": 0.36772058010101316, "memory(GiB)": 70.5, "step": 52510, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.449527 }, { "epoch": 2.249903603101838, "grad_norm": 3.035057544708252, "learning_rate": 5.782471435093063e-05, "loss": 0.34537670612335203, "memory(GiB)": 70.5, "step": 52515, "token_acc": 0.9164086687306502, "train_speed(iter/s)": 1.44953 }, { "epoch": 2.250117818431087, "grad_norm": 7.354162216186523, "learning_rate": 5.7818067425477976e-05, "loss": 0.7275999069213868, "memory(GiB)": 70.5, "step": 52520, "token_acc": 0.8677966101694915, "train_speed(iter/s)": 1.449544 }, { "epoch": 2.250332033760336, "grad_norm": 3.3559162616729736, "learning_rate": 5.781142035839371e-05, "loss": 0.40292768478393554, "memory(GiB)": 70.5, "step": 52525, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.449538 }, { "epoch": 2.250546249089585, "grad_norm": 6.04686164855957, "learning_rate": 5.7804773149798216e-05, "loss": 0.45314655303955076, "memory(GiB)": 70.5, "step": 52530, "token_acc": 0.916923076923077, "train_speed(iter/s)": 1.449553 }, { "epoch": 2.2507604644188337, "grad_norm": 6.260641574859619, "learning_rate": 5.7798125799811944e-05, "loss": 0.37197935581207275, "memory(GiB)": 70.5, "step": 52535, "token_acc": 0.9206896551724137, "train_speed(iter/s)": 1.449544 }, { "epoch": 2.250974679748083, "grad_norm": 5.497158050537109, "learning_rate": 5.77914783085553e-05, "loss": 0.5177834510803223, "memory(GiB)": 70.5, "step": 52540, "token_acc": 0.8995983935742972, "train_speed(iter/s)": 1.449537 }, { "epoch": 2.2511888950773318, "grad_norm": 4.427728652954102, "learning_rate": 5.778483067614874e-05, "loss": 0.3128844976425171, "memory(GiB)": 70.5, "step": 52545, "token_acc": 0.925, "train_speed(iter/s)": 1.449547 }, { "epoch": 2.2514031104065806, "grad_norm": 4.379026412963867, "learning_rate": 5.7778182902712644e-05, "loss": 0.3440453052520752, "memory(GiB)": 70.5, "step": 52550, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.449554 }, { "epoch": 2.25161732573583, "grad_norm": 3.8940658569335938, "learning_rate": 5.777153498836748e-05, "loss": 0.4477794647216797, "memory(GiB)": 70.5, "step": 52555, "token_acc": 0.9188311688311688, "train_speed(iter/s)": 1.449554 }, { "epoch": 2.2518315410650787, "grad_norm": 4.589415550231934, "learning_rate": 5.776488693323366e-05, "loss": 0.3815009593963623, "memory(GiB)": 70.5, "step": 52560, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.449565 }, { "epoch": 2.2520457563943275, "grad_norm": 2.041130542755127, "learning_rate": 5.775823873743165e-05, "loss": 0.39951133728027344, "memory(GiB)": 70.5, "step": 52565, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.449578 }, { "epoch": 2.2522599717235767, "grad_norm": 0.3210144639015198, "learning_rate": 5.775159040108185e-05, "loss": 0.329424524307251, "memory(GiB)": 70.5, "step": 52570, "token_acc": 0.9154078549848943, "train_speed(iter/s)": 1.449565 }, { "epoch": 2.2524741870528255, "grad_norm": 3.0190553665161133, "learning_rate": 5.7744941924304716e-05, "loss": 0.2662684679031372, "memory(GiB)": 70.5, "step": 52575, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.449571 }, { "epoch": 2.2526884023820744, "grad_norm": 3.3492918014526367, "learning_rate": 5.77382933072207e-05, "loss": 0.46067514419555666, "memory(GiB)": 70.5, "step": 52580, "token_acc": 0.9076433121019108, "train_speed(iter/s)": 1.449572 }, { "epoch": 2.2529026177113236, "grad_norm": 4.5578389167785645, "learning_rate": 5.773164454995026e-05, "loss": 0.21438393592834473, "memory(GiB)": 70.5, "step": 52585, "token_acc": 0.956386292834891, "train_speed(iter/s)": 1.449572 }, { "epoch": 2.2531168330405724, "grad_norm": 3.523167133331299, "learning_rate": 5.77249956526138e-05, "loss": 0.29048476219177244, "memory(GiB)": 70.5, "step": 52590, "token_acc": 0.9276729559748428, "train_speed(iter/s)": 1.449567 }, { "epoch": 2.2533310483698212, "grad_norm": 5.43466854095459, "learning_rate": 5.7718346615331806e-05, "loss": 0.30491974353790285, "memory(GiB)": 70.5, "step": 52595, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.449569 }, { "epoch": 2.2535452636990705, "grad_norm": 3.7831838130950928, "learning_rate": 5.771169743822473e-05, "loss": 0.442803955078125, "memory(GiB)": 70.5, "step": 52600, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.449581 }, { "epoch": 2.2537594790283193, "grad_norm": 1.6243313550949097, "learning_rate": 5.770504812141301e-05, "loss": 0.3941894769668579, "memory(GiB)": 70.5, "step": 52605, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.449579 }, { "epoch": 2.253973694357568, "grad_norm": 3.417905807495117, "learning_rate": 5.7698398665017104e-05, "loss": 0.41497139930725097, "memory(GiB)": 70.5, "step": 52610, "token_acc": 0.9122257053291536, "train_speed(iter/s)": 1.449574 }, { "epoch": 2.2541879096868174, "grad_norm": 5.672442436218262, "learning_rate": 5.7691749069157505e-05, "loss": 0.4582627296447754, "memory(GiB)": 70.5, "step": 52615, "token_acc": 0.9155844155844156, "train_speed(iter/s)": 1.44957 }, { "epoch": 2.254402125016066, "grad_norm": 3.7658822536468506, "learning_rate": 5.768509933395465e-05, "loss": 0.33792340755462646, "memory(GiB)": 70.5, "step": 52620, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.44957 }, { "epoch": 2.254616340345315, "grad_norm": 4.83274507522583, "learning_rate": 5.7678449459529015e-05, "loss": 0.4129764556884766, "memory(GiB)": 70.5, "step": 52625, "token_acc": 0.8938053097345132, "train_speed(iter/s)": 1.449576 }, { "epoch": 2.2548305556745643, "grad_norm": 0.4518454968929291, "learning_rate": 5.7671799446001075e-05, "loss": 0.6419589042663574, "memory(GiB)": 70.5, "step": 52630, "token_acc": 0.8803418803418803, "train_speed(iter/s)": 1.449588 }, { "epoch": 2.255044771003813, "grad_norm": 3.7972068786621094, "learning_rate": 5.766514929349126e-05, "loss": 0.41292500495910645, "memory(GiB)": 70.5, "step": 52635, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.449579 }, { "epoch": 2.255258986333062, "grad_norm": 5.369274139404297, "learning_rate": 5.7658499002120104e-05, "loss": 0.3957796096801758, "memory(GiB)": 70.5, "step": 52640, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.449581 }, { "epoch": 2.255473201662311, "grad_norm": 0.8681516647338867, "learning_rate": 5.765184857200804e-05, "loss": 0.2444002628326416, "memory(GiB)": 70.5, "step": 52645, "token_acc": 0.948170731707317, "train_speed(iter/s)": 1.449592 }, { "epoch": 2.25568741699156, "grad_norm": 5.667004108428955, "learning_rate": 5.764519800327556e-05, "loss": 0.3891102075576782, "memory(GiB)": 70.5, "step": 52650, "token_acc": 0.9161490683229814, "train_speed(iter/s)": 1.449608 }, { "epoch": 2.2559016323208088, "grad_norm": 4.065532207489014, "learning_rate": 5.7638547296043154e-05, "loss": 0.4421842575073242, "memory(GiB)": 70.5, "step": 52655, "token_acc": 0.8996282527881041, "train_speed(iter/s)": 1.449607 }, { "epoch": 2.256115847650058, "grad_norm": 5.463143348693848, "learning_rate": 5.76318964504313e-05, "loss": 0.4165318965911865, "memory(GiB)": 70.5, "step": 52660, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.44963 }, { "epoch": 2.256330062979307, "grad_norm": 3.070361614227295, "learning_rate": 5.7625245466560474e-05, "loss": 0.28249056339263917, "memory(GiB)": 70.5, "step": 52665, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.449631 }, { "epoch": 2.2565442783085556, "grad_norm": 3.311476469039917, "learning_rate": 5.761859434455118e-05, "loss": 0.3437540054321289, "memory(GiB)": 70.5, "step": 52670, "token_acc": 0.9328358208955224, "train_speed(iter/s)": 1.449629 }, { "epoch": 2.256758493637805, "grad_norm": 4.856264114379883, "learning_rate": 5.761194308452389e-05, "loss": 0.2427760362625122, "memory(GiB)": 70.5, "step": 52675, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.449631 }, { "epoch": 2.2569727089670537, "grad_norm": 3.036844253540039, "learning_rate": 5.760529168659912e-05, "loss": 0.4215597152709961, "memory(GiB)": 70.5, "step": 52680, "token_acc": 0.916923076923077, "train_speed(iter/s)": 1.449624 }, { "epoch": 2.2571869242963025, "grad_norm": 3.8464009761810303, "learning_rate": 5.759864015089735e-05, "loss": 0.43200340270996096, "memory(GiB)": 70.5, "step": 52685, "token_acc": 0.9187279151943463, "train_speed(iter/s)": 1.449628 }, { "epoch": 2.257401139625552, "grad_norm": 5.106104373931885, "learning_rate": 5.7591988477539104e-05, "loss": 0.28839247226715087, "memory(GiB)": 70.5, "step": 52690, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.449628 }, { "epoch": 2.2576153549548006, "grad_norm": 3.8864269256591797, "learning_rate": 5.758533666664485e-05, "loss": 0.4074099063873291, "memory(GiB)": 70.5, "step": 52695, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.449635 }, { "epoch": 2.2578295702840494, "grad_norm": 6.145382404327393, "learning_rate": 5.757868471833512e-05, "loss": 0.33277325630187987, "memory(GiB)": 70.5, "step": 52700, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.449653 }, { "epoch": 2.2580437856132987, "grad_norm": 2.737121105194092, "learning_rate": 5.757203263273039e-05, "loss": 0.3038219928741455, "memory(GiB)": 70.5, "step": 52705, "token_acc": 0.9281437125748503, "train_speed(iter/s)": 1.449662 }, { "epoch": 2.2582580009425475, "grad_norm": 3.9837310314178467, "learning_rate": 5.756538040995119e-05, "loss": 0.42202110290527345, "memory(GiB)": 70.5, "step": 52710, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.44966 }, { "epoch": 2.2584722162717963, "grad_norm": 4.174581527709961, "learning_rate": 5.7558728050118036e-05, "loss": 0.21165065765380858, "memory(GiB)": 70.5, "step": 52715, "token_acc": 0.954225352112676, "train_speed(iter/s)": 1.449657 }, { "epoch": 2.2586864316010455, "grad_norm": 3.220247745513916, "learning_rate": 5.755207555335142e-05, "loss": 0.5465051651000976, "memory(GiB)": 70.5, "step": 52720, "token_acc": 0.9034267912772586, "train_speed(iter/s)": 1.449672 }, { "epoch": 2.2589006469302944, "grad_norm": 12.55324935913086, "learning_rate": 5.7545422919771874e-05, "loss": 0.28412575721740724, "memory(GiB)": 70.5, "step": 52725, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.449672 }, { "epoch": 2.259114862259543, "grad_norm": 2.7143619060516357, "learning_rate": 5.753877014949992e-05, "loss": 0.4835994720458984, "memory(GiB)": 70.5, "step": 52730, "token_acc": 0.8973607038123167, "train_speed(iter/s)": 1.449674 }, { "epoch": 2.2593290775887924, "grad_norm": 5.518698692321777, "learning_rate": 5.753211724265606e-05, "loss": 0.5149809837341308, "memory(GiB)": 70.5, "step": 52735, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.449668 }, { "epoch": 2.2595432929180412, "grad_norm": 1.3936549425125122, "learning_rate": 5.7525464199360844e-05, "loss": 0.33528695106506345, "memory(GiB)": 70.5, "step": 52740, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.449665 }, { "epoch": 2.25975750824729, "grad_norm": 6.15483283996582, "learning_rate": 5.751881101973479e-05, "loss": 0.4762080669403076, "memory(GiB)": 70.5, "step": 52745, "token_acc": 0.9027237354085603, "train_speed(iter/s)": 1.44967 }, { "epoch": 2.2599717235765393, "grad_norm": 4.226250648498535, "learning_rate": 5.75121577038984e-05, "loss": 0.3122908592224121, "memory(GiB)": 70.5, "step": 52750, "token_acc": 0.9233870967741935, "train_speed(iter/s)": 1.449666 }, { "epoch": 2.260185938905788, "grad_norm": 1.7420153617858887, "learning_rate": 5.750550425197224e-05, "loss": 0.3331716537475586, "memory(GiB)": 70.5, "step": 52755, "token_acc": 0.9258064516129032, "train_speed(iter/s)": 1.449674 }, { "epoch": 2.260400154235037, "grad_norm": 4.238653182983398, "learning_rate": 5.749885066407683e-05, "loss": 0.531134033203125, "memory(GiB)": 70.5, "step": 52760, "token_acc": 0.8984375, "train_speed(iter/s)": 1.449676 }, { "epoch": 2.260614369564286, "grad_norm": 6.468727111816406, "learning_rate": 5.74921969403327e-05, "loss": 0.5030579566955566, "memory(GiB)": 70.5, "step": 52765, "token_acc": 0.8991935483870968, "train_speed(iter/s)": 1.449672 }, { "epoch": 2.260828584893535, "grad_norm": 5.882770538330078, "learning_rate": 5.74855430808604e-05, "loss": 0.3865624666213989, "memory(GiB)": 70.5, "step": 52770, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.449697 }, { "epoch": 2.261042800222784, "grad_norm": 3.818988561630249, "learning_rate": 5.7478889085780476e-05, "loss": 0.547053050994873, "memory(GiB)": 70.5, "step": 52775, "token_acc": 0.8903654485049833, "train_speed(iter/s)": 1.449697 }, { "epoch": 2.261257015552033, "grad_norm": 5.508390426635742, "learning_rate": 5.7472234955213435e-05, "loss": 0.4715911865234375, "memory(GiB)": 70.5, "step": 52780, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.449696 }, { "epoch": 2.261471230881282, "grad_norm": 5.168531894683838, "learning_rate": 5.7465580689279864e-05, "loss": 0.6101081848144532, "memory(GiB)": 70.5, "step": 52785, "token_acc": 0.8778877887788779, "train_speed(iter/s)": 1.449698 }, { "epoch": 2.2616854462105307, "grad_norm": 4.662021636962891, "learning_rate": 5.745892628810029e-05, "loss": 0.593177080154419, "memory(GiB)": 70.5, "step": 52790, "token_acc": 0.8902439024390244, "train_speed(iter/s)": 1.449704 }, { "epoch": 2.26189966153978, "grad_norm": 4.144350051879883, "learning_rate": 5.745227175179526e-05, "loss": 0.3444117546081543, "memory(GiB)": 70.5, "step": 52795, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.449704 }, { "epoch": 2.2621138768690288, "grad_norm": 6.619505882263184, "learning_rate": 5.744561708048536e-05, "loss": 0.26885228157043456, "memory(GiB)": 70.5, "step": 52800, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.449701 }, { "epoch": 2.2623280921982776, "grad_norm": 1.651773452758789, "learning_rate": 5.743896227429111e-05, "loss": 0.21267013549804686, "memory(GiB)": 70.5, "step": 52805, "token_acc": 0.946927374301676, "train_speed(iter/s)": 1.449702 }, { "epoch": 2.262542307527527, "grad_norm": 4.422550678253174, "learning_rate": 5.743230733333307e-05, "loss": 0.4594231128692627, "memory(GiB)": 70.5, "step": 52810, "token_acc": 0.9082278481012658, "train_speed(iter/s)": 1.449695 }, { "epoch": 2.2627565228567756, "grad_norm": 1.238610863685608, "learning_rate": 5.7425652257731834e-05, "loss": 0.41801815032958983, "memory(GiB)": 70.5, "step": 52815, "token_acc": 0.9212598425196851, "train_speed(iter/s)": 1.449699 }, { "epoch": 2.2629707381860245, "grad_norm": 1.5611387491226196, "learning_rate": 5.741899704760791e-05, "loss": 0.28969900608062743, "memory(GiB)": 70.5, "step": 52820, "token_acc": 0.9465875370919882, "train_speed(iter/s)": 1.449707 }, { "epoch": 2.2631849535152737, "grad_norm": 2.763514280319214, "learning_rate": 5.741234170308193e-05, "loss": 0.2515584468841553, "memory(GiB)": 70.5, "step": 52825, "token_acc": 0.9407407407407408, "train_speed(iter/s)": 1.44971 }, { "epoch": 2.2633991688445225, "grad_norm": 4.797215461730957, "learning_rate": 5.74056862242744e-05, "loss": 0.4629405975341797, "memory(GiB)": 70.5, "step": 52830, "token_acc": 0.9157894736842105, "train_speed(iter/s)": 1.449712 }, { "epoch": 2.2636133841737713, "grad_norm": 2.837392807006836, "learning_rate": 5.7399030611305913e-05, "loss": 0.6495938301086426, "memory(GiB)": 70.5, "step": 52835, "token_acc": 0.8674351585014409, "train_speed(iter/s)": 1.449726 }, { "epoch": 2.2638275995030206, "grad_norm": 4.748310089111328, "learning_rate": 5.739237486429707e-05, "loss": 0.2842139720916748, "memory(GiB)": 70.5, "step": 52840, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.449725 }, { "epoch": 2.2640418148322694, "grad_norm": 5.662550449371338, "learning_rate": 5.738571898336842e-05, "loss": 0.21090641021728515, "memory(GiB)": 70.5, "step": 52845, "token_acc": 0.9587628865979382, "train_speed(iter/s)": 1.449726 }, { "epoch": 2.264256030161518, "grad_norm": 1.4836467504501343, "learning_rate": 5.737906296864053e-05, "loss": 0.3705893039703369, "memory(GiB)": 70.5, "step": 52850, "token_acc": 0.9246031746031746, "train_speed(iter/s)": 1.449724 }, { "epoch": 2.2644702454907675, "grad_norm": 6.0440239906311035, "learning_rate": 5.737240682023399e-05, "loss": 0.6593562126159668, "memory(GiB)": 70.5, "step": 52855, "token_acc": 0.8807017543859649, "train_speed(iter/s)": 1.44972 }, { "epoch": 2.2646844608200163, "grad_norm": 5.959343433380127, "learning_rate": 5.73657505382694e-05, "loss": 0.5569936275482178, "memory(GiB)": 70.5, "step": 52860, "token_acc": 0.8951841359773371, "train_speed(iter/s)": 1.449717 }, { "epoch": 2.264898676149265, "grad_norm": 3.1031646728515625, "learning_rate": 5.735909412286731e-05, "loss": 0.28183276653289796, "memory(GiB)": 70.5, "step": 52865, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.449712 }, { "epoch": 2.2651128914785144, "grad_norm": 1.668314814567566, "learning_rate": 5.735243757414833e-05, "loss": 0.17628538608551025, "memory(GiB)": 70.5, "step": 52870, "token_acc": 0.9575757575757575, "train_speed(iter/s)": 1.449707 }, { "epoch": 2.265327106807763, "grad_norm": 1.8515039682388306, "learning_rate": 5.734578089223306e-05, "loss": 0.24089574813842773, "memory(GiB)": 70.5, "step": 52875, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.449717 }, { "epoch": 2.265541322137012, "grad_norm": 4.844218730926514, "learning_rate": 5.7339124077242066e-05, "loss": 0.36744821071624756, "memory(GiB)": 70.5, "step": 52880, "token_acc": 0.9133858267716536, "train_speed(iter/s)": 1.449716 }, { "epoch": 2.2657555374662612, "grad_norm": 0.19748161733150482, "learning_rate": 5.7332467129295964e-05, "loss": 0.10275565385818482, "memory(GiB)": 70.5, "step": 52885, "token_acc": 0.9784172661870504, "train_speed(iter/s)": 1.449716 }, { "epoch": 2.26596975279551, "grad_norm": 5.95936918258667, "learning_rate": 5.732581004851534e-05, "loss": 0.4063377380371094, "memory(GiB)": 70.5, "step": 52890, "token_acc": 0.9020408163265307, "train_speed(iter/s)": 1.449711 }, { "epoch": 2.266183968124759, "grad_norm": 1.4996542930603027, "learning_rate": 5.731915283502079e-05, "loss": 0.22405519485473632, "memory(GiB)": 70.5, "step": 52895, "token_acc": 0.9628252788104089, "train_speed(iter/s)": 1.449733 }, { "epoch": 2.266398183454008, "grad_norm": 4.283078670501709, "learning_rate": 5.731249548893291e-05, "loss": 0.4961993217468262, "memory(GiB)": 70.5, "step": 52900, "token_acc": 0.8831615120274914, "train_speed(iter/s)": 1.449737 }, { "epoch": 2.266612398783257, "grad_norm": 1.7851955890655518, "learning_rate": 5.730583801037234e-05, "loss": 0.4868284225463867, "memory(GiB)": 70.5, "step": 52905, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.449736 }, { "epoch": 2.2668266141125057, "grad_norm": 1.9902963638305664, "learning_rate": 5.729918039945963e-05, "loss": 0.1607264518737793, "memory(GiB)": 70.5, "step": 52910, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.449735 }, { "epoch": 2.267040829441755, "grad_norm": 4.655635833740234, "learning_rate": 5.729252265631545e-05, "loss": 0.4588481426239014, "memory(GiB)": 70.5, "step": 52915, "token_acc": 0.8937728937728938, "train_speed(iter/s)": 1.449738 }, { "epoch": 2.267255044771004, "grad_norm": 3.549098491668701, "learning_rate": 5.728586478106037e-05, "loss": 0.4345695495605469, "memory(GiB)": 70.5, "step": 52920, "token_acc": 0.9065420560747663, "train_speed(iter/s)": 1.449739 }, { "epoch": 2.2674692601002526, "grad_norm": 5.16324520111084, "learning_rate": 5.727920677381501e-05, "loss": 0.3706834077835083, "memory(GiB)": 70.5, "step": 52925, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.449738 }, { "epoch": 2.267683475429502, "grad_norm": 1.3598228693008423, "learning_rate": 5.72725486347e-05, "loss": 0.5312013149261474, "memory(GiB)": 70.5, "step": 52930, "token_acc": 0.8679245283018868, "train_speed(iter/s)": 1.449741 }, { "epoch": 2.2678976907587507, "grad_norm": 3.2825918197631836, "learning_rate": 5.726589036383594e-05, "loss": 0.3489930868148804, "memory(GiB)": 70.5, "step": 52935, "token_acc": 0.9104938271604939, "train_speed(iter/s)": 1.449749 }, { "epoch": 2.2681119060879995, "grad_norm": 3.5232367515563965, "learning_rate": 5.725923196134345e-05, "loss": 0.2612337112426758, "memory(GiB)": 70.5, "step": 52940, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.449753 }, { "epoch": 2.2683261214172488, "grad_norm": 1.4787962436676025, "learning_rate": 5.725257342734318e-05, "loss": 0.3999204158782959, "memory(GiB)": 70.5, "step": 52945, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.449745 }, { "epoch": 2.2685403367464976, "grad_norm": 5.946691036224365, "learning_rate": 5.7245914761955744e-05, "loss": 0.3191575288772583, "memory(GiB)": 70.5, "step": 52950, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.449753 }, { "epoch": 2.2687545520757464, "grad_norm": 5.423698902130127, "learning_rate": 5.7239255965301755e-05, "loss": 0.5347332954406738, "memory(GiB)": 70.5, "step": 52955, "token_acc": 0.8721804511278195, "train_speed(iter/s)": 1.449767 }, { "epoch": 2.2689687674049956, "grad_norm": 3.3791604042053223, "learning_rate": 5.723259703750186e-05, "loss": 0.18113222122192382, "memory(GiB)": 70.5, "step": 52960, "token_acc": 0.9401709401709402, "train_speed(iter/s)": 1.449772 }, { "epoch": 2.2691829827342445, "grad_norm": 2.5940539836883545, "learning_rate": 5.722593797867667e-05, "loss": 0.37963266372680665, "memory(GiB)": 70.5, "step": 52965, "token_acc": 0.9171597633136095, "train_speed(iter/s)": 1.449773 }, { "epoch": 2.2693971980634933, "grad_norm": 3.122728109359741, "learning_rate": 5.721927878894685e-05, "loss": 0.2870576620101929, "memory(GiB)": 70.5, "step": 52970, "token_acc": 0.9458333333333333, "train_speed(iter/s)": 1.449795 }, { "epoch": 2.2696114133927425, "grad_norm": 0.9123919010162354, "learning_rate": 5.721261946843302e-05, "loss": 0.4531064033508301, "memory(GiB)": 70.5, "step": 52975, "token_acc": 0.8980891719745223, "train_speed(iter/s)": 1.449811 }, { "epoch": 2.2698256287219913, "grad_norm": 3.9189770221710205, "learning_rate": 5.72059600172558e-05, "loss": 0.6268417835235596, "memory(GiB)": 70.5, "step": 52980, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.449824 }, { "epoch": 2.27003984405124, "grad_norm": 0.5161292552947998, "learning_rate": 5.7199300435535884e-05, "loss": 0.4138816833496094, "memory(GiB)": 70.5, "step": 52985, "token_acc": 0.9336099585062241, "train_speed(iter/s)": 1.44984 }, { "epoch": 2.2702540593804894, "grad_norm": 3.9676201343536377, "learning_rate": 5.7192640723393874e-05, "loss": 0.29136953353881834, "memory(GiB)": 70.5, "step": 52990, "token_acc": 0.941358024691358, "train_speed(iter/s)": 1.449839 }, { "epoch": 2.270468274709738, "grad_norm": 2.9290401935577393, "learning_rate": 5.7185980880950426e-05, "loss": 0.5018796920776367, "memory(GiB)": 70.5, "step": 52995, "token_acc": 0.8929765886287625, "train_speed(iter/s)": 1.449837 }, { "epoch": 2.270682490038987, "grad_norm": 5.761385917663574, "learning_rate": 5.717932090832621e-05, "loss": 0.6170250415802002, "memory(GiB)": 70.5, "step": 53000, "token_acc": 0.8708487084870848, "train_speed(iter/s)": 1.44984 }, { "epoch": 2.270682490038987, "eval_loss": 2.3999388217926025, "eval_runtime": 13.9712, "eval_samples_per_second": 7.158, "eval_steps_per_second": 7.158, "eval_token_acc": 0.4684431977559607, "step": 53000 }, { "epoch": 2.2708967053682363, "grad_norm": 4.6669697761535645, "learning_rate": 5.7172660805641855e-05, "loss": 0.37383618354797366, "memory(GiB)": 70.5, "step": 53005, "token_acc": 0.5827263267429761, "train_speed(iter/s)": 1.449247 }, { "epoch": 2.271110920697485, "grad_norm": 3.10567045211792, "learning_rate": 5.716600057301802e-05, "loss": 0.31509671211242674, "memory(GiB)": 70.5, "step": 53010, "token_acc": 0.916923076923077, "train_speed(iter/s)": 1.449256 }, { "epoch": 2.271325136026734, "grad_norm": 3.6892271041870117, "learning_rate": 5.7159340210575355e-05, "loss": 0.5789093971252441, "memory(GiB)": 70.5, "step": 53015, "token_acc": 0.8961937716262975, "train_speed(iter/s)": 1.44926 }, { "epoch": 2.271539351355983, "grad_norm": 1.1169780492782593, "learning_rate": 5.715267971843453e-05, "loss": 0.5933608055114746, "memory(GiB)": 70.5, "step": 53020, "token_acc": 0.8392156862745098, "train_speed(iter/s)": 1.449285 }, { "epoch": 2.271753566685232, "grad_norm": 2.246265172958374, "learning_rate": 5.71460190967162e-05, "loss": 0.38626558780670167, "memory(GiB)": 70.5, "step": 53025, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.449286 }, { "epoch": 2.271967782014481, "grad_norm": 0.37796053290367126, "learning_rate": 5.713935834554104e-05, "loss": 0.20568389892578126, "memory(GiB)": 70.5, "step": 53030, "token_acc": 0.9419354838709677, "train_speed(iter/s)": 1.449285 }, { "epoch": 2.27218199734373, "grad_norm": 2.7691256999969482, "learning_rate": 5.713269746502971e-05, "loss": 0.18548436164855958, "memory(GiB)": 70.5, "step": 53035, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.449282 }, { "epoch": 2.272396212672979, "grad_norm": 5.244280815124512, "learning_rate": 5.7126036455302855e-05, "loss": 0.5616899013519288, "memory(GiB)": 70.5, "step": 53040, "token_acc": 0.8843537414965986, "train_speed(iter/s)": 1.449284 }, { "epoch": 2.2726104280022277, "grad_norm": 7.407625675201416, "learning_rate": 5.7119375316481175e-05, "loss": 0.4693611145019531, "memory(GiB)": 70.5, "step": 53045, "token_acc": 0.9121338912133892, "train_speed(iter/s)": 1.44929 }, { "epoch": 2.272824643331477, "grad_norm": 3.106170415878296, "learning_rate": 5.7112714048685324e-05, "loss": 0.36048135757446287, "memory(GiB)": 70.5, "step": 53050, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.449288 }, { "epoch": 2.2730388586607257, "grad_norm": 0.24638964235782623, "learning_rate": 5.710605265203599e-05, "loss": 0.399113392829895, "memory(GiB)": 70.5, "step": 53055, "token_acc": 0.9139072847682119, "train_speed(iter/s)": 1.44929 }, { "epoch": 2.2732530739899746, "grad_norm": 4.88716983795166, "learning_rate": 5.7099391126653855e-05, "loss": 0.5733750820159912, "memory(GiB)": 70.5, "step": 53060, "token_acc": 0.8909774436090225, "train_speed(iter/s)": 1.449306 }, { "epoch": 2.273467289319224, "grad_norm": 1.6370623111724854, "learning_rate": 5.709272947265959e-05, "loss": 0.23565452098846434, "memory(GiB)": 70.5, "step": 53065, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.44931 }, { "epoch": 2.2736815046484726, "grad_norm": 5.18342399597168, "learning_rate": 5.708606769017386e-05, "loss": 0.5440085411071778, "memory(GiB)": 70.5, "step": 53070, "token_acc": 0.8938053097345132, "train_speed(iter/s)": 1.449306 }, { "epoch": 2.2738957199777214, "grad_norm": 3.945046901702881, "learning_rate": 5.707940577931739e-05, "loss": 0.48295130729675295, "memory(GiB)": 70.5, "step": 53075, "token_acc": 0.8934169278996865, "train_speed(iter/s)": 1.449314 }, { "epoch": 2.2741099353069707, "grad_norm": 4.596092700958252, "learning_rate": 5.707274374021082e-05, "loss": 0.41078500747680663, "memory(GiB)": 70.5, "step": 53080, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.449343 }, { "epoch": 2.2743241506362195, "grad_norm": 2.855907917022705, "learning_rate": 5.706608157297488e-05, "loss": 0.27781317234039304, "memory(GiB)": 70.5, "step": 53085, "token_acc": 0.9469387755102041, "train_speed(iter/s)": 1.44936 }, { "epoch": 2.2745383659654683, "grad_norm": 4.261691570281982, "learning_rate": 5.705941927773023e-05, "loss": 0.3178340673446655, "memory(GiB)": 70.5, "step": 53090, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.449357 }, { "epoch": 2.2747525812947176, "grad_norm": 2.8425331115722656, "learning_rate": 5.70527568545976e-05, "loss": 0.5549401760101318, "memory(GiB)": 70.5, "step": 53095, "token_acc": 0.8940397350993378, "train_speed(iter/s)": 1.449367 }, { "epoch": 2.2749667966239664, "grad_norm": 5.540611267089844, "learning_rate": 5.704609430369764e-05, "loss": 0.43674049377441404, "memory(GiB)": 70.5, "step": 53100, "token_acc": 0.9187279151943463, "train_speed(iter/s)": 1.449369 }, { "epoch": 2.275181011953215, "grad_norm": 1.3810499906539917, "learning_rate": 5.703943162515109e-05, "loss": 0.38604881763458254, "memory(GiB)": 70.5, "step": 53105, "token_acc": 0.9221556886227545, "train_speed(iter/s)": 1.449383 }, { "epoch": 2.2753952272824645, "grad_norm": 5.327312469482422, "learning_rate": 5.703276881907861e-05, "loss": 0.4792295455932617, "memory(GiB)": 70.5, "step": 53110, "token_acc": 0.8984126984126984, "train_speed(iter/s)": 1.449387 }, { "epoch": 2.2756094426117133, "grad_norm": 2.284775495529175, "learning_rate": 5.7026105885600946e-05, "loss": 0.14739974737167358, "memory(GiB)": 70.5, "step": 53115, "token_acc": 0.9651162790697675, "train_speed(iter/s)": 1.44939 }, { "epoch": 2.275823657940962, "grad_norm": 6.925829887390137, "learning_rate": 5.701944282483877e-05, "loss": 0.2983029127120972, "memory(GiB)": 70.5, "step": 53120, "token_acc": 0.924924924924925, "train_speed(iter/s)": 1.449396 }, { "epoch": 2.2760378732702113, "grad_norm": 2.3106300830841064, "learning_rate": 5.7012779636912804e-05, "loss": 0.2451099395751953, "memory(GiB)": 70.5, "step": 53125, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.449398 }, { "epoch": 2.27625208859946, "grad_norm": 2.023820638656616, "learning_rate": 5.700611632194377e-05, "loss": 0.31774191856384276, "memory(GiB)": 70.5, "step": 53130, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.449404 }, { "epoch": 2.276466303928709, "grad_norm": 3.096386671066284, "learning_rate": 5.699945288005235e-05, "loss": 0.4759627342224121, "memory(GiB)": 70.5, "step": 53135, "token_acc": 0.9219219219219219, "train_speed(iter/s)": 1.449408 }, { "epoch": 2.2766805192579582, "grad_norm": 5.507349491119385, "learning_rate": 5.699278931135929e-05, "loss": 0.2775786638259888, "memory(GiB)": 70.5, "step": 53140, "token_acc": 0.9349112426035503, "train_speed(iter/s)": 1.449402 }, { "epoch": 2.276894734587207, "grad_norm": 3.1055943965911865, "learning_rate": 5.6986125615985285e-05, "loss": 0.4317826271057129, "memory(GiB)": 70.5, "step": 53145, "token_acc": 0.9087719298245615, "train_speed(iter/s)": 1.449405 }, { "epoch": 2.277108949916456, "grad_norm": 4.6006083488464355, "learning_rate": 5.6979461794051056e-05, "loss": 0.3695951461791992, "memory(GiB)": 70.5, "step": 53150, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.44941 }, { "epoch": 2.277323165245705, "grad_norm": 5.431859493255615, "learning_rate": 5.6972797845677325e-05, "loss": 0.510779857635498, "memory(GiB)": 70.5, "step": 53155, "token_acc": 0.8816793893129771, "train_speed(iter/s)": 1.44942 }, { "epoch": 2.277537380574954, "grad_norm": 2.378704786300659, "learning_rate": 5.696613377098484e-05, "loss": 0.3490100383758545, "memory(GiB)": 70.5, "step": 53160, "token_acc": 0.9260450160771704, "train_speed(iter/s)": 1.449424 }, { "epoch": 2.2777515959042027, "grad_norm": 8.274801254272461, "learning_rate": 5.695946957009429e-05, "loss": 0.571352195739746, "memory(GiB)": 70.5, "step": 53165, "token_acc": 0.8971428571428571, "train_speed(iter/s)": 1.449422 }, { "epoch": 2.277965811233452, "grad_norm": 2.288325786590576, "learning_rate": 5.6952805243126416e-05, "loss": 0.3491519451141357, "memory(GiB)": 70.5, "step": 53170, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.449416 }, { "epoch": 2.278180026562701, "grad_norm": 3.4322919845581055, "learning_rate": 5.694614079020197e-05, "loss": 0.36034741401672366, "memory(GiB)": 70.5, "step": 53175, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.449421 }, { "epoch": 2.2783942418919496, "grad_norm": 2.28483247756958, "learning_rate": 5.6939476211441664e-05, "loss": 0.45918974876403806, "memory(GiB)": 70.5, "step": 53180, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.449419 }, { "epoch": 2.278608457221199, "grad_norm": 2.395932197570801, "learning_rate": 5.6932811506966236e-05, "loss": 0.5134274005889893, "memory(GiB)": 70.5, "step": 53185, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.449423 }, { "epoch": 2.2788226725504477, "grad_norm": 3.1144320964813232, "learning_rate": 5.692614667689642e-05, "loss": 0.5457284450531006, "memory(GiB)": 70.5, "step": 53190, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.449417 }, { "epoch": 2.2790368878796965, "grad_norm": 3.15972638130188, "learning_rate": 5.6919481721352954e-05, "loss": 0.2908227205276489, "memory(GiB)": 70.5, "step": 53195, "token_acc": 0.9442379182156134, "train_speed(iter/s)": 1.449416 }, { "epoch": 2.2792511032089458, "grad_norm": 3.317673921585083, "learning_rate": 5.691281664045659e-05, "loss": 0.342284107208252, "memory(GiB)": 70.5, "step": 53200, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.449416 }, { "epoch": 2.2794653185381946, "grad_norm": 5.9461445808410645, "learning_rate": 5.690615143432807e-05, "loss": 0.49057803153991697, "memory(GiB)": 70.5, "step": 53205, "token_acc": 0.8880866425992779, "train_speed(iter/s)": 1.449425 }, { "epoch": 2.2796795338674434, "grad_norm": 2.693390369415283, "learning_rate": 5.689948610308815e-05, "loss": 0.31540038585662844, "memory(GiB)": 70.5, "step": 53210, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.449429 }, { "epoch": 2.2798937491966926, "grad_norm": 3.5278031826019287, "learning_rate": 5.689282064685755e-05, "loss": 0.1698897361755371, "memory(GiB)": 70.5, "step": 53215, "token_acc": 0.9554655870445344, "train_speed(iter/s)": 1.449432 }, { "epoch": 2.2801079645259414, "grad_norm": 5.995758056640625, "learning_rate": 5.688615506575704e-05, "loss": 0.40904674530029295, "memory(GiB)": 70.5, "step": 53220, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.449438 }, { "epoch": 2.2803221798551903, "grad_norm": 1.389531135559082, "learning_rate": 5.687948935990737e-05, "loss": 0.3067673921585083, "memory(GiB)": 70.5, "step": 53225, "token_acc": 0.9226006191950464, "train_speed(iter/s)": 1.449453 }, { "epoch": 2.2805363951844395, "grad_norm": 1.5227221250534058, "learning_rate": 5.687282352942931e-05, "loss": 0.36159625053405764, "memory(GiB)": 70.5, "step": 53230, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.449446 }, { "epoch": 2.2807506105136883, "grad_norm": 4.09245491027832, "learning_rate": 5.686615757444359e-05, "loss": 0.3823273181915283, "memory(GiB)": 70.5, "step": 53235, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.449439 }, { "epoch": 2.280964825842937, "grad_norm": 4.6702799797058105, "learning_rate": 5.685949149507099e-05, "loss": 0.42411074638366697, "memory(GiB)": 70.5, "step": 53240, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.449439 }, { "epoch": 2.2811790411721864, "grad_norm": 6.7948150634765625, "learning_rate": 5.6852825291432264e-05, "loss": 0.2802061319351196, "memory(GiB)": 70.5, "step": 53245, "token_acc": 0.9213483146067416, "train_speed(iter/s)": 1.449438 }, { "epoch": 2.281393256501435, "grad_norm": 2.3919928073883057, "learning_rate": 5.684615896364819e-05, "loss": 0.44316611289978025, "memory(GiB)": 70.5, "step": 53250, "token_acc": 0.9042904290429042, "train_speed(iter/s)": 1.449434 }, { "epoch": 2.281607471830684, "grad_norm": 3.2869982719421387, "learning_rate": 5.683949251183951e-05, "loss": 0.4902122974395752, "memory(GiB)": 70.5, "step": 53255, "token_acc": 0.8910891089108911, "train_speed(iter/s)": 1.449429 }, { "epoch": 2.2818216871599333, "grad_norm": 3.17828631401062, "learning_rate": 5.683282593612702e-05, "loss": 0.5613065719604492, "memory(GiB)": 70.5, "step": 53260, "token_acc": 0.8897338403041825, "train_speed(iter/s)": 1.449448 }, { "epoch": 2.282035902489182, "grad_norm": 3.125295639038086, "learning_rate": 5.682615923663147e-05, "loss": 0.40001716613769533, "memory(GiB)": 70.5, "step": 53265, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.449478 }, { "epoch": 2.282250117818431, "grad_norm": 3.1407928466796875, "learning_rate": 5.681949241347364e-05, "loss": 0.2714265823364258, "memory(GiB)": 70.5, "step": 53270, "token_acc": 0.9479166666666666, "train_speed(iter/s)": 1.44947 }, { "epoch": 2.28246433314768, "grad_norm": 4.161417484283447, "learning_rate": 5.681282546677431e-05, "loss": 0.32826714515686034, "memory(GiB)": 70.5, "step": 53275, "token_acc": 0.9345794392523364, "train_speed(iter/s)": 1.449468 }, { "epoch": 2.282678548476929, "grad_norm": 4.024356365203857, "learning_rate": 5.6806158396654264e-05, "loss": 0.3191695213317871, "memory(GiB)": 70.5, "step": 53280, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.449488 }, { "epoch": 2.282892763806178, "grad_norm": 4.003512859344482, "learning_rate": 5.679949120323426e-05, "loss": 0.44338092803955076, "memory(GiB)": 70.5, "step": 53285, "token_acc": 0.8867313915857605, "train_speed(iter/s)": 1.449496 }, { "epoch": 2.283106979135427, "grad_norm": 2.9352989196777344, "learning_rate": 5.6792823886635104e-05, "loss": 0.43652806282043455, "memory(GiB)": 70.5, "step": 53290, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.449504 }, { "epoch": 2.283321194464676, "grad_norm": 2.8504984378814697, "learning_rate": 5.678615644697758e-05, "loss": 0.2601557970046997, "memory(GiB)": 70.5, "step": 53295, "token_acc": 0.9541666666666667, "train_speed(iter/s)": 1.449523 }, { "epoch": 2.2835354097939247, "grad_norm": 6.868009090423584, "learning_rate": 5.6779488884382446e-05, "loss": 0.24890489578247071, "memory(GiB)": 70.5, "step": 53300, "token_acc": 0.9431438127090301, "train_speed(iter/s)": 1.449525 }, { "epoch": 2.283749625123174, "grad_norm": 2.26428484916687, "learning_rate": 5.677282119897053e-05, "loss": 0.2133488178253174, "memory(GiB)": 70.5, "step": 53305, "token_acc": 0.948051948051948, "train_speed(iter/s)": 1.449532 }, { "epoch": 2.2839638404524227, "grad_norm": 1.474932074546814, "learning_rate": 5.6766153390862585e-05, "loss": 0.3669632911682129, "memory(GiB)": 70.5, "step": 53310, "token_acc": 0.921875, "train_speed(iter/s)": 1.44953 }, { "epoch": 2.2841780557816715, "grad_norm": 5.377188682556152, "learning_rate": 5.675948546017943e-05, "loss": 0.49345989227294923, "memory(GiB)": 70.5, "step": 53315, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.449529 }, { "epoch": 2.284392271110921, "grad_norm": 0.5170162320137024, "learning_rate": 5.6752817407041855e-05, "loss": 0.2383336067199707, "memory(GiB)": 70.5, "step": 53320, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.449536 }, { "epoch": 2.2846064864401696, "grad_norm": 2.676469564437866, "learning_rate": 5.674614923157068e-05, "loss": 0.3076120615005493, "memory(GiB)": 70.5, "step": 53325, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.449555 }, { "epoch": 2.2848207017694184, "grad_norm": 1.796510100364685, "learning_rate": 5.6739480933886655e-05, "loss": 0.6351068019866943, "memory(GiB)": 70.5, "step": 53330, "token_acc": 0.863768115942029, "train_speed(iter/s)": 1.449564 }, { "epoch": 2.2850349170986677, "grad_norm": 5.7469482421875, "learning_rate": 5.6732812514110624e-05, "loss": 0.2867789030075073, "memory(GiB)": 70.5, "step": 53335, "token_acc": 0.9432624113475178, "train_speed(iter/s)": 1.44956 }, { "epoch": 2.2852491324279165, "grad_norm": 0.6336707472801208, "learning_rate": 5.672614397236337e-05, "loss": 0.2812535762786865, "memory(GiB)": 70.5, "step": 53340, "token_acc": 0.9434523809523809, "train_speed(iter/s)": 1.449558 }, { "epoch": 2.2854633477571653, "grad_norm": 4.821355819702148, "learning_rate": 5.671947530876571e-05, "loss": 0.37067503929138185, "memory(GiB)": 70.5, "step": 53345, "token_acc": 0.9413680781758957, "train_speed(iter/s)": 1.449551 }, { "epoch": 2.2856775630864146, "grad_norm": 3.747122287750244, "learning_rate": 5.671280652343846e-05, "loss": 0.31232738494873047, "memory(GiB)": 70.5, "step": 53350, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.449548 }, { "epoch": 2.2858917784156634, "grad_norm": 2.251922845840454, "learning_rate": 5.670613761650242e-05, "loss": 0.5566195487976074, "memory(GiB)": 70.5, "step": 53355, "token_acc": 0.8925081433224755, "train_speed(iter/s)": 1.449548 }, { "epoch": 2.286105993744912, "grad_norm": 2.4166882038116455, "learning_rate": 5.6699468588078394e-05, "loss": 0.3651489973068237, "memory(GiB)": 70.5, "step": 53360, "token_acc": 0.9120879120879121, "train_speed(iter/s)": 1.449543 }, { "epoch": 2.2863202090741614, "grad_norm": 2.9951207637786865, "learning_rate": 5.669279943828723e-05, "loss": 0.2467723846435547, "memory(GiB)": 70.5, "step": 53365, "token_acc": 0.9488054607508533, "train_speed(iter/s)": 1.44956 }, { "epoch": 2.2865344244034103, "grad_norm": 5.133660316467285, "learning_rate": 5.6686130167249706e-05, "loss": 0.367284369468689, "memory(GiB)": 70.5, "step": 53370, "token_acc": 0.8953068592057761, "train_speed(iter/s)": 1.449565 }, { "epoch": 2.286748639732659, "grad_norm": 0.846076250076294, "learning_rate": 5.6679460775086676e-05, "loss": 0.20354270935058594, "memory(GiB)": 70.5, "step": 53375, "token_acc": 0.9558232931726908, "train_speed(iter/s)": 1.449568 }, { "epoch": 2.2869628550619083, "grad_norm": 6.270830154418945, "learning_rate": 5.667279126191896e-05, "loss": 0.47473459243774413, "memory(GiB)": 70.5, "step": 53380, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.449573 }, { "epoch": 2.287177070391157, "grad_norm": 9.750495910644531, "learning_rate": 5.666612162786734e-05, "loss": 0.8128856658935547, "memory(GiB)": 70.5, "step": 53385, "token_acc": 0.84765625, "train_speed(iter/s)": 1.449604 }, { "epoch": 2.287391285720406, "grad_norm": 1.8908761739730835, "learning_rate": 5.665945187305268e-05, "loss": 0.3812835454940796, "memory(GiB)": 70.5, "step": 53390, "token_acc": 0.9375, "train_speed(iter/s)": 1.449612 }, { "epoch": 2.287605501049655, "grad_norm": 3.773660182952881, "learning_rate": 5.6652781997595814e-05, "loss": 0.5328668594360352, "memory(GiB)": 70.5, "step": 53395, "token_acc": 0.877742946708464, "train_speed(iter/s)": 1.449619 }, { "epoch": 2.287819716378904, "grad_norm": 2.356423854827881, "learning_rate": 5.6646112001617554e-05, "loss": 0.24006919860839843, "memory(GiB)": 70.5, "step": 53400, "token_acc": 0.9303030303030303, "train_speed(iter/s)": 1.449611 }, { "epoch": 2.288033931708153, "grad_norm": 3.5788700580596924, "learning_rate": 5.663944188523875e-05, "loss": 0.3900122165679932, "memory(GiB)": 70.5, "step": 53405, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.449612 }, { "epoch": 2.288248147037402, "grad_norm": 1.9418141841888428, "learning_rate": 5.663277164858023e-05, "loss": 0.33175044059753417, "memory(GiB)": 70.5, "step": 53410, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.449618 }, { "epoch": 2.288462362366651, "grad_norm": 8.450907707214355, "learning_rate": 5.662610129176281e-05, "loss": 0.2387521743774414, "memory(GiB)": 70.5, "step": 53415, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.449619 }, { "epoch": 2.2886765776958997, "grad_norm": 2.5113871097564697, "learning_rate": 5.661943081490736e-05, "loss": 0.39509007930755613, "memory(GiB)": 70.5, "step": 53420, "token_acc": 0.911504424778761, "train_speed(iter/s)": 1.449613 }, { "epoch": 2.288890793025149, "grad_norm": 3.9333574771881104, "learning_rate": 5.661276021813472e-05, "loss": 0.4428097724914551, "memory(GiB)": 70.5, "step": 53425, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.449631 }, { "epoch": 2.289105008354398, "grad_norm": 1.680936574935913, "learning_rate": 5.6606089501565704e-05, "loss": 0.3767989158630371, "memory(GiB)": 70.5, "step": 53430, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.449636 }, { "epoch": 2.2893192236836466, "grad_norm": 3.718277931213379, "learning_rate": 5.65994186653212e-05, "loss": 0.4427751064300537, "memory(GiB)": 70.5, "step": 53435, "token_acc": 0.9, "train_speed(iter/s)": 1.449639 }, { "epoch": 2.289533439012896, "grad_norm": 4.332479476928711, "learning_rate": 5.659274770952204e-05, "loss": 0.4826528072357178, "memory(GiB)": 70.5, "step": 53440, "token_acc": 0.9083333333333333, "train_speed(iter/s)": 1.449639 }, { "epoch": 2.2897476543421447, "grad_norm": 4.243422508239746, "learning_rate": 5.658607663428905e-05, "loss": 0.46933732032775877, "memory(GiB)": 70.5, "step": 53445, "token_acc": 0.8776223776223776, "train_speed(iter/s)": 1.449642 }, { "epoch": 2.2899618696713935, "grad_norm": 2.277090072631836, "learning_rate": 5.6579405439743114e-05, "loss": 0.23726811408996581, "memory(GiB)": 70.5, "step": 53450, "token_acc": 0.9431818181818182, "train_speed(iter/s)": 1.449645 }, { "epoch": 2.2901760850006427, "grad_norm": 0.2064744532108307, "learning_rate": 5.6572734126005076e-05, "loss": 0.23630006313323976, "memory(GiB)": 70.5, "step": 53455, "token_acc": 0.948339483394834, "train_speed(iter/s)": 1.44964 }, { "epoch": 2.2903903003298915, "grad_norm": 2.193551540374756, "learning_rate": 5.656606269319579e-05, "loss": 0.10102169513702393, "memory(GiB)": 70.5, "step": 53460, "token_acc": 0.977859778597786, "train_speed(iter/s)": 1.449641 }, { "epoch": 2.2906045156591404, "grad_norm": 3.3399651050567627, "learning_rate": 5.6559391141436126e-05, "loss": 0.5319972038269043, "memory(GiB)": 70.5, "step": 53465, "token_acc": 0.8973509933774835, "train_speed(iter/s)": 1.449657 }, { "epoch": 2.2908187309883896, "grad_norm": 10.238447189331055, "learning_rate": 5.655271947084694e-05, "loss": 0.32825703620910646, "memory(GiB)": 70.5, "step": 53470, "token_acc": 0.9495548961424333, "train_speed(iter/s)": 1.449674 }, { "epoch": 2.2910329463176384, "grad_norm": 0.8763352036476135, "learning_rate": 5.654604768154909e-05, "loss": 0.27101964950561525, "memory(GiB)": 70.5, "step": 53475, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.44967 }, { "epoch": 2.2912471616468872, "grad_norm": 3.281989097595215, "learning_rate": 5.6539375773663436e-05, "loss": 0.5150269031524658, "memory(GiB)": 70.5, "step": 53480, "token_acc": 0.8811881188118812, "train_speed(iter/s)": 1.449669 }, { "epoch": 2.2914613769761365, "grad_norm": 6.666264057159424, "learning_rate": 5.653270374731087e-05, "loss": 0.3336836338043213, "memory(GiB)": 70.5, "step": 53485, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.449675 }, { "epoch": 2.2916755923053853, "grad_norm": 5.3463969230651855, "learning_rate": 5.652603160261224e-05, "loss": 0.5435078620910645, "memory(GiB)": 70.5, "step": 53490, "token_acc": 0.8526315789473684, "train_speed(iter/s)": 1.449669 }, { "epoch": 2.291889807634634, "grad_norm": 6.436408519744873, "learning_rate": 5.651935933968843e-05, "loss": 0.33032329082489015, "memory(GiB)": 70.5, "step": 53495, "token_acc": 0.9065420560747663, "train_speed(iter/s)": 1.449664 }, { "epoch": 2.2921040229638834, "grad_norm": 1.5094674825668335, "learning_rate": 5.65126869586603e-05, "loss": 0.3972856760025024, "memory(GiB)": 70.5, "step": 53500, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.449672 }, { "epoch": 2.2921040229638834, "eval_loss": 2.4498934745788574, "eval_runtime": 13.1215, "eval_samples_per_second": 7.621, "eval_steps_per_second": 7.621, "eval_token_acc": 0.45, "step": 53500 }, { "epoch": 2.292318238293132, "grad_norm": 5.801357746124268, "learning_rate": 5.650601445964874e-05, "loss": 0.2223287582397461, "memory(GiB)": 70.5, "step": 53505, "token_acc": 0.6092066601371204, "train_speed(iter/s)": 1.44912 }, { "epoch": 2.292532453622381, "grad_norm": 2.840301275253296, "learning_rate": 5.6499341842774637e-05, "loss": 0.28338189125061036, "memory(GiB)": 70.5, "step": 53510, "token_acc": 0.940625, "train_speed(iter/s)": 1.449124 }, { "epoch": 2.2927466689516303, "grad_norm": 2.8509740829467773, "learning_rate": 5.649266910815885e-05, "loss": 0.15979973077774048, "memory(GiB)": 70.5, "step": 53515, "token_acc": 0.9636363636363636, "train_speed(iter/s)": 1.449126 }, { "epoch": 2.292960884280879, "grad_norm": 6.476071357727051, "learning_rate": 5.648599625592227e-05, "loss": 0.479799747467041, "memory(GiB)": 70.5, "step": 53520, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.449133 }, { "epoch": 2.293175099610128, "grad_norm": 0.17781279981136322, "learning_rate": 5.647932328618579e-05, "loss": 0.45985946655273435, "memory(GiB)": 70.5, "step": 53525, "token_acc": 0.9080118694362018, "train_speed(iter/s)": 1.449135 }, { "epoch": 2.293389314939377, "grad_norm": 6.220119953155518, "learning_rate": 5.647265019907027e-05, "loss": 0.4274289608001709, "memory(GiB)": 70.5, "step": 53530, "token_acc": 0.9206896551724137, "train_speed(iter/s)": 1.449138 }, { "epoch": 2.293603530268626, "grad_norm": 4.03661584854126, "learning_rate": 5.646597699469665e-05, "loss": 0.28065469264984133, "memory(GiB)": 70.5, "step": 53535, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.449146 }, { "epoch": 2.2938177455978748, "grad_norm": 4.1724395751953125, "learning_rate": 5.645930367318577e-05, "loss": 0.422794771194458, "memory(GiB)": 70.5, "step": 53540, "token_acc": 0.890625, "train_speed(iter/s)": 1.449152 }, { "epoch": 2.294031960927124, "grad_norm": 3.000312089920044, "learning_rate": 5.645263023465854e-05, "loss": 0.18635308742523193, "memory(GiB)": 70.5, "step": 53545, "token_acc": 0.9640522875816994, "train_speed(iter/s)": 1.449159 }, { "epoch": 2.294246176256373, "grad_norm": 4.875197887420654, "learning_rate": 5.644595667923589e-05, "loss": 0.5027366638183594, "memory(GiB)": 70.5, "step": 53550, "token_acc": 0.8996138996138996, "train_speed(iter/s)": 1.449174 }, { "epoch": 2.2944603915856217, "grad_norm": 4.36314582824707, "learning_rate": 5.643928300703867e-05, "loss": 0.43693079948425295, "memory(GiB)": 70.5, "step": 53555, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.449177 }, { "epoch": 2.294674606914871, "grad_norm": 3.1217849254608154, "learning_rate": 5.643260921818779e-05, "loss": 0.6142314910888672, "memory(GiB)": 70.5, "step": 53560, "token_acc": 0.8698630136986302, "train_speed(iter/s)": 1.449188 }, { "epoch": 2.2948888222441197, "grad_norm": 1.620611310005188, "learning_rate": 5.642593531280418e-05, "loss": 0.25156326293945314, "memory(GiB)": 70.5, "step": 53565, "token_acc": 0.952054794520548, "train_speed(iter/s)": 1.449206 }, { "epoch": 2.2951030375733685, "grad_norm": 5.62371826171875, "learning_rate": 5.6419261291008696e-05, "loss": 0.439546012878418, "memory(GiB)": 70.5, "step": 53570, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.449203 }, { "epoch": 2.295317252902618, "grad_norm": 2.112898111343384, "learning_rate": 5.641258715292228e-05, "loss": 0.33036787509918214, "memory(GiB)": 70.5, "step": 53575, "token_acc": 0.933852140077821, "train_speed(iter/s)": 1.449204 }, { "epoch": 2.2955314682318666, "grad_norm": 0.17571553587913513, "learning_rate": 5.640591289866585e-05, "loss": 0.2202195405960083, "memory(GiB)": 70.5, "step": 53580, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.449204 }, { "epoch": 2.2957456835611154, "grad_norm": 4.893132209777832, "learning_rate": 5.6399238528360285e-05, "loss": 0.3805395603179932, "memory(GiB)": 70.5, "step": 53585, "token_acc": 0.9133858267716536, "train_speed(iter/s)": 1.449204 }, { "epoch": 2.2959598988903647, "grad_norm": 3.3214588165283203, "learning_rate": 5.6392564042126514e-05, "loss": 0.1667158007621765, "memory(GiB)": 70.5, "step": 53590, "token_acc": 0.9718875502008032, "train_speed(iter/s)": 1.449205 }, { "epoch": 2.2961741142196135, "grad_norm": 2.1673429012298584, "learning_rate": 5.638588944008546e-05, "loss": 0.6437295913696289, "memory(GiB)": 70.5, "step": 53595, "token_acc": 0.8767123287671232, "train_speed(iter/s)": 1.449201 }, { "epoch": 2.2963883295488623, "grad_norm": 4.122439861297607, "learning_rate": 5.637921472235801e-05, "loss": 0.5037303447723389, "memory(GiB)": 70.5, "step": 53600, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.4492 }, { "epoch": 2.2966025448781116, "grad_norm": 6.5001020431518555, "learning_rate": 5.63725398890651e-05, "loss": 0.4439727306365967, "memory(GiB)": 70.5, "step": 53605, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.449208 }, { "epoch": 2.2968167602073604, "grad_norm": 6.516814231872559, "learning_rate": 5.636586494032766e-05, "loss": 0.30937848091125486, "memory(GiB)": 70.5, "step": 53610, "token_acc": 0.95, "train_speed(iter/s)": 1.449201 }, { "epoch": 2.297030975536609, "grad_norm": 3.292412042617798, "learning_rate": 5.635918987626661e-05, "loss": 0.7606129169464111, "memory(GiB)": 70.5, "step": 53615, "token_acc": 0.8367346938775511, "train_speed(iter/s)": 1.449206 }, { "epoch": 2.2972451908658584, "grad_norm": 1.9303717613220215, "learning_rate": 5.6352514697002865e-05, "loss": 0.418535852432251, "memory(GiB)": 70.5, "step": 53620, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.449206 }, { "epoch": 2.2974594061951072, "grad_norm": 0.4974150061607361, "learning_rate": 5.6345839402657364e-05, "loss": 0.4640336513519287, "memory(GiB)": 70.5, "step": 53625, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.449212 }, { "epoch": 2.297673621524356, "grad_norm": 5.9643402099609375, "learning_rate": 5.633916399335102e-05, "loss": 0.6320202827453614, "memory(GiB)": 70.5, "step": 53630, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.449215 }, { "epoch": 2.2978878368536053, "grad_norm": 5.796878814697266, "learning_rate": 5.633248846920478e-05, "loss": 0.3148148775100708, "memory(GiB)": 70.5, "step": 53635, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.449211 }, { "epoch": 2.298102052182854, "grad_norm": 2.6414058208465576, "learning_rate": 5.632581283033956e-05, "loss": 0.2180105686187744, "memory(GiB)": 70.5, "step": 53640, "token_acc": 0.9536679536679536, "train_speed(iter/s)": 1.449209 }, { "epoch": 2.298316267512103, "grad_norm": 7.705135345458984, "learning_rate": 5.631913707687632e-05, "loss": 0.4201181411743164, "memory(GiB)": 70.5, "step": 53645, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.449212 }, { "epoch": 2.298530482841352, "grad_norm": 4.373175144195557, "learning_rate": 5.6312461208935964e-05, "loss": 0.2117168664932251, "memory(GiB)": 70.5, "step": 53650, "token_acc": 0.9541666666666667, "train_speed(iter/s)": 1.449214 }, { "epoch": 2.298744698170601, "grad_norm": 8.658500671386719, "learning_rate": 5.630578522663947e-05, "loss": 0.3177162170410156, "memory(GiB)": 70.5, "step": 53655, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.449231 }, { "epoch": 2.29895891349985, "grad_norm": 3.002086877822876, "learning_rate": 5.629910913010775e-05, "loss": 0.3351463794708252, "memory(GiB)": 70.5, "step": 53660, "token_acc": 0.9295774647887324, "train_speed(iter/s)": 1.449252 }, { "epoch": 2.299173128829099, "grad_norm": 7.110208988189697, "learning_rate": 5.6292432919461777e-05, "loss": 0.32074005603790284, "memory(GiB)": 70.5, "step": 53665, "token_acc": 0.9341317365269461, "train_speed(iter/s)": 1.449263 }, { "epoch": 2.299387344158348, "grad_norm": 6.220852851867676, "learning_rate": 5.628575659482247e-05, "loss": 0.39553916454315186, "memory(GiB)": 70.5, "step": 53670, "token_acc": 0.9136490250696379, "train_speed(iter/s)": 1.449266 }, { "epoch": 2.2996015594875967, "grad_norm": 4.607945919036865, "learning_rate": 5.627908015631078e-05, "loss": 0.4637740135192871, "memory(GiB)": 70.5, "step": 53675, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.449275 }, { "epoch": 2.299815774816846, "grad_norm": 0.25310367345809937, "learning_rate": 5.6272403604047663e-05, "loss": 0.118560528755188, "memory(GiB)": 70.5, "step": 53680, "token_acc": 0.9678571428571429, "train_speed(iter/s)": 1.449276 }, { "epoch": 2.3000299901460948, "grad_norm": 2.61967396736145, "learning_rate": 5.626572693815407e-05, "loss": 0.323476505279541, "memory(GiB)": 70.5, "step": 53685, "token_acc": 0.9414225941422594, "train_speed(iter/s)": 1.449283 }, { "epoch": 2.3002442054753436, "grad_norm": 5.965197563171387, "learning_rate": 5.625905015875096e-05, "loss": 0.42984418869018554, "memory(GiB)": 70.5, "step": 53690, "token_acc": 0.900990099009901, "train_speed(iter/s)": 1.449279 }, { "epoch": 2.300458420804593, "grad_norm": 0.6055848598480225, "learning_rate": 5.625237326595929e-05, "loss": 0.07638252377510071, "memory(GiB)": 70.5, "step": 53695, "token_acc": 0.9791044776119403, "train_speed(iter/s)": 1.449285 }, { "epoch": 2.3006726361338417, "grad_norm": 5.9280171394348145, "learning_rate": 5.624569625990002e-05, "loss": 0.290201473236084, "memory(GiB)": 70.5, "step": 53700, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.449288 }, { "epoch": 2.3008868514630905, "grad_norm": 2.9993503093719482, "learning_rate": 5.623901914069407e-05, "loss": 0.2793882846832275, "memory(GiB)": 70.5, "step": 53705, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.449291 }, { "epoch": 2.3011010667923397, "grad_norm": 4.700130939483643, "learning_rate": 5.623234190846247e-05, "loss": 0.47109642028808596, "memory(GiB)": 70.5, "step": 53710, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.449293 }, { "epoch": 2.3013152821215885, "grad_norm": 3.830970048904419, "learning_rate": 5.6225664563326134e-05, "loss": 0.7237722873687744, "memory(GiB)": 70.5, "step": 53715, "token_acc": 0.8432055749128919, "train_speed(iter/s)": 1.449304 }, { "epoch": 2.3015294974508373, "grad_norm": 4.3970046043396, "learning_rate": 5.621898710540604e-05, "loss": 0.7009999752044678, "memory(GiB)": 70.5, "step": 53720, "token_acc": 0.8258064516129032, "train_speed(iter/s)": 1.44931 }, { "epoch": 2.3017437127800866, "grad_norm": 3.2931957244873047, "learning_rate": 5.621230953482317e-05, "loss": 0.32334561347961427, "memory(GiB)": 70.5, "step": 53725, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.449313 }, { "epoch": 2.3019579281093354, "grad_norm": 3.591479778289795, "learning_rate": 5.620563185169848e-05, "loss": 0.3717966079711914, "memory(GiB)": 70.5, "step": 53730, "token_acc": 0.9204892966360856, "train_speed(iter/s)": 1.449312 }, { "epoch": 2.3021721434385842, "grad_norm": 6.51910924911499, "learning_rate": 5.6198954056152954e-05, "loss": 0.292827844619751, "memory(GiB)": 70.5, "step": 53735, "token_acc": 0.9098039215686274, "train_speed(iter/s)": 1.449328 }, { "epoch": 2.3023863587678335, "grad_norm": 5.207090854644775, "learning_rate": 5.6192276148307556e-05, "loss": 0.4586700439453125, "memory(GiB)": 70.5, "step": 53740, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.449353 }, { "epoch": 2.3026005740970823, "grad_norm": 2.7928318977355957, "learning_rate": 5.618559812828327e-05, "loss": 0.28120615482330324, "memory(GiB)": 70.5, "step": 53745, "token_acc": 0.9385245901639344, "train_speed(iter/s)": 1.449358 }, { "epoch": 2.302814789426331, "grad_norm": 0.343016117811203, "learning_rate": 5.6178919996201064e-05, "loss": 0.3863049507141113, "memory(GiB)": 70.5, "step": 53750, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.44936 }, { "epoch": 2.3030290047555804, "grad_norm": 0.34627610445022583, "learning_rate": 5.617224175218193e-05, "loss": 0.19911811351776124, "memory(GiB)": 70.5, "step": 53755, "token_acc": 0.9573770491803278, "train_speed(iter/s)": 1.449361 }, { "epoch": 2.303243220084829, "grad_norm": 3.4638726711273193, "learning_rate": 5.616556339634686e-05, "loss": 0.28048694133758545, "memory(GiB)": 70.5, "step": 53760, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.449357 }, { "epoch": 2.303457435414078, "grad_norm": 0.843008279800415, "learning_rate": 5.61588849288168e-05, "loss": 0.17510461807250977, "memory(GiB)": 70.5, "step": 53765, "token_acc": 0.9652509652509652, "train_speed(iter/s)": 1.449352 }, { "epoch": 2.3036716507433272, "grad_norm": 2.9409186840057373, "learning_rate": 5.6152206349712776e-05, "loss": 0.41826720237731935, "memory(GiB)": 70.5, "step": 53770, "token_acc": 0.9124579124579124, "train_speed(iter/s)": 1.449347 }, { "epoch": 2.303885866072576, "grad_norm": 3.6663713455200195, "learning_rate": 5.614552765915575e-05, "loss": 0.3619333505630493, "memory(GiB)": 70.5, "step": 53775, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.449363 }, { "epoch": 2.304100081401825, "grad_norm": 1.0253468751907349, "learning_rate": 5.613884885726675e-05, "loss": 0.1925184726715088, "memory(GiB)": 70.5, "step": 53780, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.449361 }, { "epoch": 2.304314296731074, "grad_norm": 0.9378514885902405, "learning_rate": 5.6132169944166735e-05, "loss": 0.14609036445617676, "memory(GiB)": 70.5, "step": 53785, "token_acc": 0.9698996655518395, "train_speed(iter/s)": 1.449371 }, { "epoch": 2.304528512060323, "grad_norm": 2.8869709968566895, "learning_rate": 5.6125490919976696e-05, "loss": 0.303623104095459, "memory(GiB)": 70.5, "step": 53790, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.44938 }, { "epoch": 2.3047427273895718, "grad_norm": 4.117764949798584, "learning_rate": 5.611881178481765e-05, "loss": 0.39959309101104734, "memory(GiB)": 70.5, "step": 53795, "token_acc": 0.9004524886877828, "train_speed(iter/s)": 1.449381 }, { "epoch": 2.304956942718821, "grad_norm": 1.7091422080993652, "learning_rate": 5.611213253881059e-05, "loss": 0.2821804523468018, "memory(GiB)": 70.5, "step": 53800, "token_acc": 0.9343065693430657, "train_speed(iter/s)": 1.449383 }, { "epoch": 2.30517115804807, "grad_norm": 2.5907328128814697, "learning_rate": 5.610545318207652e-05, "loss": 0.26749367713928224, "memory(GiB)": 70.5, "step": 53805, "token_acc": 0.9418604651162791, "train_speed(iter/s)": 1.449388 }, { "epoch": 2.3053853733773186, "grad_norm": 4.497104167938232, "learning_rate": 5.609877371473643e-05, "loss": 0.4631082534790039, "memory(GiB)": 70.5, "step": 53810, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.449389 }, { "epoch": 2.305599588706568, "grad_norm": 3.301682233810425, "learning_rate": 5.6092094136911344e-05, "loss": 0.6042266368865967, "memory(GiB)": 70.5, "step": 53815, "token_acc": 0.8782894736842105, "train_speed(iter/s)": 1.449389 }, { "epoch": 2.3058138040358167, "grad_norm": 3.865720748901367, "learning_rate": 5.608541444872224e-05, "loss": 0.16694759130477904, "memory(GiB)": 70.5, "step": 53820, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.449394 }, { "epoch": 2.3060280193650655, "grad_norm": 5.021277904510498, "learning_rate": 5.607873465029017e-05, "loss": 0.1715349793434143, "memory(GiB)": 70.5, "step": 53825, "token_acc": 0.9641434262948207, "train_speed(iter/s)": 1.449412 }, { "epoch": 2.3062422346943148, "grad_norm": 5.589904308319092, "learning_rate": 5.607205474173609e-05, "loss": 0.4337559700012207, "memory(GiB)": 70.5, "step": 53830, "token_acc": 0.9018181818181819, "train_speed(iter/s)": 1.449422 }, { "epoch": 2.3064564500235636, "grad_norm": 2.3857924938201904, "learning_rate": 5.606537472318105e-05, "loss": 0.3788615226745605, "memory(GiB)": 70.5, "step": 53835, "token_acc": 0.9156118143459916, "train_speed(iter/s)": 1.449424 }, { "epoch": 2.3066706653528124, "grad_norm": 2.1560347080230713, "learning_rate": 5.605869459474608e-05, "loss": 0.5390674591064453, "memory(GiB)": 70.5, "step": 53840, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.449437 }, { "epoch": 2.3068848806820617, "grad_norm": 0.6594496965408325, "learning_rate": 5.6052014356552166e-05, "loss": 0.25436246395111084, "memory(GiB)": 70.5, "step": 53845, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.449457 }, { "epoch": 2.3070990960113105, "grad_norm": 1.6889723539352417, "learning_rate": 5.6045334008720316e-05, "loss": 0.26104116439819336, "memory(GiB)": 70.5, "step": 53850, "token_acc": 0.9432624113475178, "train_speed(iter/s)": 1.44947 }, { "epoch": 2.3073133113405593, "grad_norm": 2.3454360961914062, "learning_rate": 5.603865355137159e-05, "loss": 0.455338716506958, "memory(GiB)": 70.5, "step": 53855, "token_acc": 0.9078212290502793, "train_speed(iter/s)": 1.449483 }, { "epoch": 2.3075275266698085, "grad_norm": 2.728752613067627, "learning_rate": 5.603197298462697e-05, "loss": 0.2548649549484253, "memory(GiB)": 70.5, "step": 53860, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.449483 }, { "epoch": 2.3077417419990573, "grad_norm": 3.2312440872192383, "learning_rate": 5.60252923086075e-05, "loss": 0.2503353118896484, "memory(GiB)": 70.5, "step": 53865, "token_acc": 0.9471698113207547, "train_speed(iter/s)": 1.449482 }, { "epoch": 2.3079559573283066, "grad_norm": 1.8478693962097168, "learning_rate": 5.601861152343423e-05, "loss": 0.1925703763961792, "memory(GiB)": 70.5, "step": 53870, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.449489 }, { "epoch": 2.3081701726575554, "grad_norm": 7.490016937255859, "learning_rate": 5.601193062922816e-05, "loss": 0.3407423973083496, "memory(GiB)": 70.5, "step": 53875, "token_acc": 0.9437229437229437, "train_speed(iter/s)": 1.449491 }, { "epoch": 2.3083843879868042, "grad_norm": 3.3790435791015625, "learning_rate": 5.600524962611032e-05, "loss": 0.32505862712860106, "memory(GiB)": 70.5, "step": 53880, "token_acc": 0.9315589353612167, "train_speed(iter/s)": 1.4495 }, { "epoch": 2.3085986033160535, "grad_norm": 4.06915807723999, "learning_rate": 5.5998568514201754e-05, "loss": 0.31341025829315183, "memory(GiB)": 70.5, "step": 53885, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.449509 }, { "epoch": 2.3088128186453023, "grad_norm": 0.861772358417511, "learning_rate": 5.599188729362349e-05, "loss": 0.15059107542037964, "memory(GiB)": 70.5, "step": 53890, "token_acc": 0.956989247311828, "train_speed(iter/s)": 1.449505 }, { "epoch": 2.309027033974551, "grad_norm": 1.9860326051712036, "learning_rate": 5.598520596449657e-05, "loss": 0.33988428115844727, "memory(GiB)": 70.5, "step": 53895, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.449506 }, { "epoch": 2.3092412493038004, "grad_norm": 4.868551254272461, "learning_rate": 5.597852452694202e-05, "loss": 0.41863741874694826, "memory(GiB)": 70.5, "step": 53900, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.449508 }, { "epoch": 2.309455464633049, "grad_norm": 3.5520007610321045, "learning_rate": 5.5971842981080905e-05, "loss": 0.36502714157104493, "memory(GiB)": 70.5, "step": 53905, "token_acc": 0.9274447949526814, "train_speed(iter/s)": 1.449511 }, { "epoch": 2.309669679962298, "grad_norm": 7.9579877853393555, "learning_rate": 5.5965161327034234e-05, "loss": 0.433031702041626, "memory(GiB)": 70.5, "step": 53910, "token_acc": 0.9042553191489362, "train_speed(iter/s)": 1.449516 }, { "epoch": 2.3098838952915473, "grad_norm": 0.030878588557243347, "learning_rate": 5.595847956492308e-05, "loss": 0.3921220779418945, "memory(GiB)": 70.5, "step": 53915, "token_acc": 0.9301587301587302, "train_speed(iter/s)": 1.449519 }, { "epoch": 2.310098110620796, "grad_norm": 5.237154483795166, "learning_rate": 5.595179769486848e-05, "loss": 0.3952942371368408, "memory(GiB)": 70.5, "step": 53920, "token_acc": 0.9316546762589928, "train_speed(iter/s)": 1.449526 }, { "epoch": 2.310312325950045, "grad_norm": 3.2706494331359863, "learning_rate": 5.5945115716991484e-05, "loss": 0.2545742988586426, "memory(GiB)": 70.5, "step": 53925, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.449519 }, { "epoch": 2.310526541279294, "grad_norm": 1.8500431776046753, "learning_rate": 5.5938433631413145e-05, "loss": 0.37685275077819824, "memory(GiB)": 70.5, "step": 53930, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.44952 }, { "epoch": 2.310740756608543, "grad_norm": 7.551665306091309, "learning_rate": 5.5931751438254486e-05, "loss": 0.31137535572052, "memory(GiB)": 70.5, "step": 53935, "token_acc": 0.9575971731448764, "train_speed(iter/s)": 1.44952 }, { "epoch": 2.3109549719377918, "grad_norm": 42.9505500793457, "learning_rate": 5.59250691376366e-05, "loss": 0.46219935417175295, "memory(GiB)": 70.5, "step": 53940, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.449519 }, { "epoch": 2.311169187267041, "grad_norm": 5.364593982696533, "learning_rate": 5.5918386729680535e-05, "loss": 0.15154176950454712, "memory(GiB)": 70.5, "step": 53945, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.449517 }, { "epoch": 2.31138340259629, "grad_norm": 4.0888447761535645, "learning_rate": 5.591170421450733e-05, "loss": 0.44754953384399415, "memory(GiB)": 70.5, "step": 53950, "token_acc": 0.9075342465753424, "train_speed(iter/s)": 1.449519 }, { "epoch": 2.3115976179255386, "grad_norm": 1.5142426490783691, "learning_rate": 5.590502159223807e-05, "loss": 0.18826534748077392, "memory(GiB)": 70.5, "step": 53955, "token_acc": 0.9601328903654485, "train_speed(iter/s)": 1.449523 }, { "epoch": 2.311811833254788, "grad_norm": 3.106785535812378, "learning_rate": 5.58983388629938e-05, "loss": 0.3264858961105347, "memory(GiB)": 70.5, "step": 53960, "token_acc": 0.910958904109589, "train_speed(iter/s)": 1.449528 }, { "epoch": 2.3120260485840367, "grad_norm": 3.2899327278137207, "learning_rate": 5.589165602689559e-05, "loss": 0.4097161293029785, "memory(GiB)": 70.5, "step": 53965, "token_acc": 0.91875, "train_speed(iter/s)": 1.449522 }, { "epoch": 2.3122402639132855, "grad_norm": 4.264303684234619, "learning_rate": 5.588497308406451e-05, "loss": 0.4816551208496094, "memory(GiB)": 70.5, "step": 53970, "token_acc": 0.9029126213592233, "train_speed(iter/s)": 1.449521 }, { "epoch": 2.3124544792425348, "grad_norm": 5.246365070343018, "learning_rate": 5.5878290034621593e-05, "loss": 0.3177368640899658, "memory(GiB)": 70.5, "step": 53975, "token_acc": 0.9342857142857143, "train_speed(iter/s)": 1.449529 }, { "epoch": 2.3126686945717836, "grad_norm": 1.514316201210022, "learning_rate": 5.5871606878687956e-05, "loss": 0.29752888679504397, "memory(GiB)": 70.5, "step": 53980, "token_acc": 0.9390243902439024, "train_speed(iter/s)": 1.449529 }, { "epoch": 2.3128829099010324, "grad_norm": 3.822108030319214, "learning_rate": 5.586492361638466e-05, "loss": 0.3520516872406006, "memory(GiB)": 70.5, "step": 53985, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.449526 }, { "epoch": 2.3130971252302817, "grad_norm": 3.0148708820343018, "learning_rate": 5.585824024783277e-05, "loss": 0.2564198970794678, "memory(GiB)": 70.5, "step": 53990, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.449532 }, { "epoch": 2.3133113405595305, "grad_norm": 2.9119584560394287, "learning_rate": 5.585155677315336e-05, "loss": 0.2712787389755249, "memory(GiB)": 70.5, "step": 53995, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.449532 }, { "epoch": 2.3135255558887793, "grad_norm": 3.3794896602630615, "learning_rate": 5.5844873192467506e-05, "loss": 0.24150660037994384, "memory(GiB)": 70.5, "step": 54000, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.449543 }, { "epoch": 2.3135255558887793, "eval_loss": 2.5918097496032715, "eval_runtime": 13.4851, "eval_samples_per_second": 7.416, "eval_steps_per_second": 7.416, "eval_token_acc": 0.4373297002724796, "step": 54000 }, { "epoch": 2.3137397712180285, "grad_norm": 2.7610361576080322, "learning_rate": 5.583818950589629e-05, "loss": 0.7330290794372558, "memory(GiB)": 70.5, "step": 54005, "token_acc": 0.5722326454033771, "train_speed(iter/s)": 1.448986 }, { "epoch": 2.3139539865472774, "grad_norm": 1.332458734512329, "learning_rate": 5.583150571356079e-05, "loss": 0.3255995035171509, "memory(GiB)": 70.5, "step": 54010, "token_acc": 0.9250814332247557, "train_speed(iter/s)": 1.448987 }, { "epoch": 2.314168201876526, "grad_norm": 5.601319789886475, "learning_rate": 5.5824821815582104e-05, "loss": 0.3370448112487793, "memory(GiB)": 70.5, "step": 54015, "token_acc": 0.93359375, "train_speed(iter/s)": 1.449001 }, { "epoch": 2.3143824172057754, "grad_norm": 3.781092405319214, "learning_rate": 5.5818137812081306e-05, "loss": 0.48778977394104006, "memory(GiB)": 70.5, "step": 54020, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.449016 }, { "epoch": 2.3145966325350242, "grad_norm": 2.7286956310272217, "learning_rate": 5.581145370317948e-05, "loss": 0.42888636589050294, "memory(GiB)": 70.5, "step": 54025, "token_acc": 0.9071729957805907, "train_speed(iter/s)": 1.449019 }, { "epoch": 2.314810847864273, "grad_norm": 2.818528890609741, "learning_rate": 5.5804769488997724e-05, "loss": 0.2654047250747681, "memory(GiB)": 70.5, "step": 54030, "token_acc": 0.948051948051948, "train_speed(iter/s)": 1.449024 }, { "epoch": 2.3150250631935223, "grad_norm": 1.1314167976379395, "learning_rate": 5.579808516965711e-05, "loss": 0.24881331920623778, "memory(GiB)": 70.5, "step": 54035, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.449034 }, { "epoch": 2.315239278522771, "grad_norm": 4.439828872680664, "learning_rate": 5.579140074527877e-05, "loss": 0.4454522609710693, "memory(GiB)": 70.5, "step": 54040, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.449031 }, { "epoch": 2.31545349385202, "grad_norm": 3.4717681407928467, "learning_rate": 5.578471621598376e-05, "loss": 0.5697812080383301, "memory(GiB)": 70.5, "step": 54045, "token_acc": 0.8664383561643836, "train_speed(iter/s)": 1.449031 }, { "epoch": 2.315667709181269, "grad_norm": 2.5451061725616455, "learning_rate": 5.5778031581893175e-05, "loss": 0.5431168556213379, "memory(GiB)": 70.5, "step": 54050, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.449029 }, { "epoch": 2.315881924510518, "grad_norm": 4.578554153442383, "learning_rate": 5.5771346843128124e-05, "loss": 0.4977963924407959, "memory(GiB)": 70.5, "step": 54055, "token_acc": 0.8764044943820225, "train_speed(iter/s)": 1.44904 }, { "epoch": 2.316096139839767, "grad_norm": 1.485893964767456, "learning_rate": 5.576466199980973e-05, "loss": 0.6573754787445069, "memory(GiB)": 70.5, "step": 54060, "token_acc": 0.8844621513944223, "train_speed(iter/s)": 1.449055 }, { "epoch": 2.316310355169016, "grad_norm": 3.335533380508423, "learning_rate": 5.575797705205907e-05, "loss": 0.5168357849121094, "memory(GiB)": 70.5, "step": 54065, "token_acc": 0.8878504672897196, "train_speed(iter/s)": 1.44906 }, { "epoch": 2.316524570498265, "grad_norm": 2.343374013900757, "learning_rate": 5.5751291999997256e-05, "loss": 0.48432002067565916, "memory(GiB)": 70.5, "step": 54070, "token_acc": 0.9030303030303031, "train_speed(iter/s)": 1.449055 }, { "epoch": 2.3167387858275137, "grad_norm": 2.334937334060669, "learning_rate": 5.574460684374541e-05, "loss": 0.14840307235717773, "memory(GiB)": 70.5, "step": 54075, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.449054 }, { "epoch": 2.316953001156763, "grad_norm": 6.315237998962402, "learning_rate": 5.573792158342459e-05, "loss": 0.38669734001159667, "memory(GiB)": 70.5, "step": 54080, "token_acc": 0.9401197604790419, "train_speed(iter/s)": 1.449057 }, { "epoch": 2.3171672164860118, "grad_norm": 6.712156295776367, "learning_rate": 5.573123621915595e-05, "loss": 0.5383927345275878, "memory(GiB)": 70.5, "step": 54085, "token_acc": 0.885558583106267, "train_speed(iter/s)": 1.449053 }, { "epoch": 2.3173814318152606, "grad_norm": 3.762136697769165, "learning_rate": 5.572455075106059e-05, "loss": 0.3192451953887939, "memory(GiB)": 70.5, "step": 54090, "token_acc": 0.9367816091954023, "train_speed(iter/s)": 1.449058 }, { "epoch": 2.31759564714451, "grad_norm": 1.1561225652694702, "learning_rate": 5.5717865179259624e-05, "loss": 0.24645051956176758, "memory(GiB)": 70.5, "step": 54095, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.449062 }, { "epoch": 2.3178098624737586, "grad_norm": 3.715754747390747, "learning_rate": 5.571117950387417e-05, "loss": 0.3258678674697876, "memory(GiB)": 70.5, "step": 54100, "token_acc": 0.9283387622149837, "train_speed(iter/s)": 1.449056 }, { "epoch": 2.3180240778030075, "grad_norm": 2.510674238204956, "learning_rate": 5.570449372502534e-05, "loss": 0.3807116985321045, "memory(GiB)": 70.5, "step": 54105, "token_acc": 0.9205882352941176, "train_speed(iter/s)": 1.449055 }, { "epoch": 2.3182382931322567, "grad_norm": 3.9175047874450684, "learning_rate": 5.5697807842834245e-05, "loss": 0.4555692672729492, "memory(GiB)": 70.5, "step": 54110, "token_acc": 0.9067164179104478, "train_speed(iter/s)": 1.449047 }, { "epoch": 2.3184525084615055, "grad_norm": 4.0094218254089355, "learning_rate": 5.5691121857422034e-05, "loss": 0.37041702270507815, "memory(GiB)": 70.5, "step": 54115, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.449054 }, { "epoch": 2.3186667237907543, "grad_norm": 3.383087158203125, "learning_rate": 5.568443576890979e-05, "loss": 0.2813273906707764, "memory(GiB)": 70.5, "step": 54120, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.449066 }, { "epoch": 2.3188809391200036, "grad_norm": 4.195888042449951, "learning_rate": 5.567774957741867e-05, "loss": 0.32539892196655273, "memory(GiB)": 70.5, "step": 54125, "token_acc": 0.925, "train_speed(iter/s)": 1.44907 }, { "epoch": 2.3190951544492524, "grad_norm": 1.9411475658416748, "learning_rate": 5.56710632830698e-05, "loss": 0.16339231729507447, "memory(GiB)": 70.5, "step": 54130, "token_acc": 0.9629629629629629, "train_speed(iter/s)": 1.449075 }, { "epoch": 2.319309369778501, "grad_norm": 2.0709478855133057, "learning_rate": 5.5664376885984296e-05, "loss": 0.2719243049621582, "memory(GiB)": 70.5, "step": 54135, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.449074 }, { "epoch": 2.3195235851077505, "grad_norm": 4.233088493347168, "learning_rate": 5.565769038628328e-05, "loss": 0.5276435375213623, "memory(GiB)": 70.5, "step": 54140, "token_acc": 0.900709219858156, "train_speed(iter/s)": 1.449077 }, { "epoch": 2.3197378004369993, "grad_norm": 5.070909023284912, "learning_rate": 5.5651003784087904e-05, "loss": 0.5683378219604492, "memory(GiB)": 70.5, "step": 54145, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.449077 }, { "epoch": 2.319952015766248, "grad_norm": 3.3148036003112793, "learning_rate": 5.5644317079519296e-05, "loss": 0.4551424026489258, "memory(GiB)": 70.5, "step": 54150, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.449083 }, { "epoch": 2.3201662310954974, "grad_norm": 1.736794114112854, "learning_rate": 5.5637630272698584e-05, "loss": 0.3034923791885376, "memory(GiB)": 70.5, "step": 54155, "token_acc": 0.9317507418397626, "train_speed(iter/s)": 1.449081 }, { "epoch": 2.320380446424746, "grad_norm": 2.270966053009033, "learning_rate": 5.56309433637469e-05, "loss": 0.19863184690475463, "memory(GiB)": 70.5, "step": 54160, "token_acc": 0.9575971731448764, "train_speed(iter/s)": 1.449079 }, { "epoch": 2.320594661753995, "grad_norm": 0.6561574935913086, "learning_rate": 5.562425635278541e-05, "loss": 0.3890302896499634, "memory(GiB)": 70.5, "step": 54165, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.449076 }, { "epoch": 2.3208088770832442, "grad_norm": 2.5561726093292236, "learning_rate": 5.561756923993523e-05, "loss": 0.2778343677520752, "memory(GiB)": 70.5, "step": 54170, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.449077 }, { "epoch": 2.321023092412493, "grad_norm": 0.7962214350700378, "learning_rate": 5.561088202531752e-05, "loss": 0.3842479228973389, "memory(GiB)": 70.5, "step": 54175, "token_acc": 0.9296875, "train_speed(iter/s)": 1.449085 }, { "epoch": 2.321237307741742, "grad_norm": 4.649628639221191, "learning_rate": 5.5604194709053424e-05, "loss": 0.5672934532165528, "memory(GiB)": 70.5, "step": 54180, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.449116 }, { "epoch": 2.321451523070991, "grad_norm": 0.2887682318687439, "learning_rate": 5.5597507291264087e-05, "loss": 0.24374496936798096, "memory(GiB)": 70.5, "step": 54185, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.449116 }, { "epoch": 2.32166573840024, "grad_norm": 3.080707550048828, "learning_rate": 5.559081977207065e-05, "loss": 0.5035924911499023, "memory(GiB)": 70.5, "step": 54190, "token_acc": 0.906060606060606, "train_speed(iter/s)": 1.449121 }, { "epoch": 2.3218799537294887, "grad_norm": 3.7043676376342773, "learning_rate": 5.558413215159425e-05, "loss": 0.2914787769317627, "memory(GiB)": 70.5, "step": 54195, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.449126 }, { "epoch": 2.322094169058738, "grad_norm": 4.072885990142822, "learning_rate": 5.557744442995607e-05, "loss": 0.4025119781494141, "memory(GiB)": 70.5, "step": 54200, "token_acc": 0.8990536277602523, "train_speed(iter/s)": 1.449121 }, { "epoch": 2.322308384387987, "grad_norm": 6.1111297607421875, "learning_rate": 5.5570756607277256e-05, "loss": 0.745305871963501, "memory(GiB)": 70.5, "step": 54205, "token_acc": 0.8333333333333334, "train_speed(iter/s)": 1.449111 }, { "epoch": 2.3225225997172356, "grad_norm": 0.6214466094970703, "learning_rate": 5.556406868367895e-05, "loss": 0.24229462146759034, "memory(GiB)": 70.5, "step": 54210, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.449104 }, { "epoch": 2.322736815046485, "grad_norm": 3.3801615238189697, "learning_rate": 5.555738065928233e-05, "loss": 0.4670864105224609, "memory(GiB)": 70.5, "step": 54215, "token_acc": 0.884, "train_speed(iter/s)": 1.449106 }, { "epoch": 2.3229510303757337, "grad_norm": 0.8477426767349243, "learning_rate": 5.555069253420855e-05, "loss": 0.5601550102233886, "memory(GiB)": 70.5, "step": 54220, "token_acc": 0.8543046357615894, "train_speed(iter/s)": 1.449103 }, { "epoch": 2.3231652457049825, "grad_norm": 5.102475643157959, "learning_rate": 5.554400430857876e-05, "loss": 0.29851703643798827, "memory(GiB)": 70.5, "step": 54225, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.449115 }, { "epoch": 2.3233794610342318, "grad_norm": 4.763238906860352, "learning_rate": 5.5537315982514135e-05, "loss": 0.46166143417358396, "memory(GiB)": 70.5, "step": 54230, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.449113 }, { "epoch": 2.3235936763634806, "grad_norm": 3.0977251529693604, "learning_rate": 5.5530627556135826e-05, "loss": 0.38351123332977294, "memory(GiB)": 70.5, "step": 54235, "token_acc": 0.9011299435028248, "train_speed(iter/s)": 1.449117 }, { "epoch": 2.3238078916927294, "grad_norm": 1.0051701068878174, "learning_rate": 5.552393902956502e-05, "loss": 0.7136910438537598, "memory(GiB)": 70.5, "step": 54240, "token_acc": 0.8602150537634409, "train_speed(iter/s)": 1.449132 }, { "epoch": 2.3240221070219786, "grad_norm": 1.3971450328826904, "learning_rate": 5.551725040292288e-05, "loss": 0.28753612041473386, "memory(GiB)": 70.5, "step": 54245, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.449133 }, { "epoch": 2.3242363223512275, "grad_norm": 0.5499507188796997, "learning_rate": 5.551056167633058e-05, "loss": 0.26681180000305177, "memory(GiB)": 70.5, "step": 54250, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.449141 }, { "epoch": 2.3244505376804763, "grad_norm": 3.9726357460021973, "learning_rate": 5.550387284990927e-05, "loss": 0.4292400360107422, "memory(GiB)": 70.5, "step": 54255, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.449139 }, { "epoch": 2.3246647530097255, "grad_norm": 2.4778213500976562, "learning_rate": 5.549718392378014e-05, "loss": 0.2548077583312988, "memory(GiB)": 70.5, "step": 54260, "token_acc": 0.937007874015748, "train_speed(iter/s)": 1.449141 }, { "epoch": 2.3248789683389743, "grad_norm": 3.2474400997161865, "learning_rate": 5.549049489806437e-05, "loss": 0.3264917850494385, "memory(GiB)": 70.5, "step": 54265, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.449145 }, { "epoch": 2.325093183668223, "grad_norm": 1.2700748443603516, "learning_rate": 5.5483805772883133e-05, "loss": 0.38419513702392577, "memory(GiB)": 70.5, "step": 54270, "token_acc": 0.9208860759493671, "train_speed(iter/s)": 1.449152 }, { "epoch": 2.3253073989974724, "grad_norm": 4.850588798522949, "learning_rate": 5.547711654835761e-05, "loss": 0.5955772876739502, "memory(GiB)": 70.5, "step": 54275, "token_acc": 0.8772727272727273, "train_speed(iter/s)": 1.449164 }, { "epoch": 2.325521614326721, "grad_norm": 6.3297529220581055, "learning_rate": 5.5470427224609e-05, "loss": 0.5014420509338379, "memory(GiB)": 70.5, "step": 54280, "token_acc": 0.9075144508670521, "train_speed(iter/s)": 1.449158 }, { "epoch": 2.32573582965597, "grad_norm": 4.3065876960754395, "learning_rate": 5.5463737801758443e-05, "loss": 0.4487509250640869, "memory(GiB)": 70.5, "step": 54285, "token_acc": 0.8884758364312267, "train_speed(iter/s)": 1.449155 }, { "epoch": 2.3259500449852193, "grad_norm": 3.4933454990386963, "learning_rate": 5.545704827992717e-05, "loss": 0.3577310085296631, "memory(GiB)": 70.5, "step": 54290, "token_acc": 0.9150141643059491, "train_speed(iter/s)": 1.449151 }, { "epoch": 2.326164260314468, "grad_norm": 5.66727352142334, "learning_rate": 5.5450358659236336e-05, "loss": 0.7443987846374511, "memory(GiB)": 70.5, "step": 54295, "token_acc": 0.8524590163934426, "train_speed(iter/s)": 1.449161 }, { "epoch": 2.326378475643717, "grad_norm": 3.35476016998291, "learning_rate": 5.5443668939807156e-05, "loss": 0.48612356185913086, "memory(GiB)": 70.5, "step": 54300, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.449161 }, { "epoch": 2.326592690972966, "grad_norm": 6.2548747062683105, "learning_rate": 5.5436979121760804e-05, "loss": 0.50778226852417, "memory(GiB)": 70.5, "step": 54305, "token_acc": 0.889589905362776, "train_speed(iter/s)": 1.449157 }, { "epoch": 2.326806906302215, "grad_norm": 2.2287044525146484, "learning_rate": 5.543028920521846e-05, "loss": 0.5408354282379151, "memory(GiB)": 70.5, "step": 54310, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.449153 }, { "epoch": 2.327021121631464, "grad_norm": 2.646470069885254, "learning_rate": 5.542359919030133e-05, "loss": 0.38472325801849366, "memory(GiB)": 70.5, "step": 54315, "token_acc": 0.9088235294117647, "train_speed(iter/s)": 1.449153 }, { "epoch": 2.327235336960713, "grad_norm": 2.4435527324676514, "learning_rate": 5.541690907713063e-05, "loss": 0.32892141342163084, "memory(GiB)": 70.5, "step": 54320, "token_acc": 0.9335443037974683, "train_speed(iter/s)": 1.449153 }, { "epoch": 2.327449552289962, "grad_norm": 8.284196853637695, "learning_rate": 5.5410218865827534e-05, "loss": 0.189558744430542, "memory(GiB)": 70.5, "step": 54325, "token_acc": 0.9498207885304659, "train_speed(iter/s)": 1.449159 }, { "epoch": 2.3276637676192107, "grad_norm": 2.122724771499634, "learning_rate": 5.540352855651325e-05, "loss": 0.28056821823120115, "memory(GiB)": 70.5, "step": 54330, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.449153 }, { "epoch": 2.32787798294846, "grad_norm": 1.291551113128662, "learning_rate": 5.539683814930898e-05, "loss": 0.44289684295654297, "memory(GiB)": 70.5, "step": 54335, "token_acc": 0.9088145896656535, "train_speed(iter/s)": 1.449155 }, { "epoch": 2.3280921982777087, "grad_norm": 2.3232617378234863, "learning_rate": 5.539014764433592e-05, "loss": 0.4488636016845703, "memory(GiB)": 70.5, "step": 54340, "token_acc": 0.8996865203761756, "train_speed(iter/s)": 1.449162 }, { "epoch": 2.3283064136069576, "grad_norm": 1.9009543657302856, "learning_rate": 5.538345704171527e-05, "loss": 0.28615665435791016, "memory(GiB)": 70.5, "step": 54345, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.449157 }, { "epoch": 2.328520628936207, "grad_norm": 2.0203194618225098, "learning_rate": 5.537676634156827e-05, "loss": 0.22671544551849365, "memory(GiB)": 70.5, "step": 54350, "token_acc": 0.9492957746478873, "train_speed(iter/s)": 1.449158 }, { "epoch": 2.3287348442654556, "grad_norm": 3.7159264087677, "learning_rate": 5.537007554401608e-05, "loss": 0.507906150817871, "memory(GiB)": 70.5, "step": 54355, "token_acc": 0.9015544041450777, "train_speed(iter/s)": 1.449177 }, { "epoch": 2.3289490595947044, "grad_norm": 7.118109703063965, "learning_rate": 5.536338464917995e-05, "loss": 0.42833786010742186, "memory(GiB)": 70.5, "step": 54360, "token_acc": 0.9065040650406504, "train_speed(iter/s)": 1.449188 }, { "epoch": 2.3291632749239537, "grad_norm": 4.043100357055664, "learning_rate": 5.535669365718107e-05, "loss": 0.2873367786407471, "memory(GiB)": 70.5, "step": 54365, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.449196 }, { "epoch": 2.3293774902532025, "grad_norm": 1.8733460903167725, "learning_rate": 5.535000256814066e-05, "loss": 0.37391340732574463, "memory(GiB)": 70.5, "step": 54370, "token_acc": 0.9176829268292683, "train_speed(iter/s)": 1.449199 }, { "epoch": 2.3295917055824513, "grad_norm": 3.437798023223877, "learning_rate": 5.534331138217993e-05, "loss": 0.39759979248046873, "memory(GiB)": 70.5, "step": 54375, "token_acc": 0.908, "train_speed(iter/s)": 1.449196 }, { "epoch": 2.3298059209117006, "grad_norm": 3.517801523208618, "learning_rate": 5.53366200994201e-05, "loss": 0.3423847913742065, "memory(GiB)": 70.5, "step": 54380, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.449195 }, { "epoch": 2.3300201362409494, "grad_norm": 2.3866283893585205, "learning_rate": 5.532992871998239e-05, "loss": 0.44426631927490234, "memory(GiB)": 70.5, "step": 54385, "token_acc": 0.92, "train_speed(iter/s)": 1.449197 }, { "epoch": 2.330234351570198, "grad_norm": 0.17679674923419952, "learning_rate": 5.532323724398804e-05, "loss": 0.1542261004447937, "memory(GiB)": 70.5, "step": 54390, "token_acc": 0.9699570815450643, "train_speed(iter/s)": 1.449194 }, { "epoch": 2.3304485668994475, "grad_norm": 4.604583740234375, "learning_rate": 5.531654567155824e-05, "loss": 0.16093616485595702, "memory(GiB)": 70.5, "step": 54395, "token_acc": 0.966542750929368, "train_speed(iter/s)": 1.449205 }, { "epoch": 2.3306627822286963, "grad_norm": 2.7136995792388916, "learning_rate": 5.530985400281422e-05, "loss": 0.3458041429519653, "memory(GiB)": 70.5, "step": 54400, "token_acc": 0.9274193548387096, "train_speed(iter/s)": 1.44921 }, { "epoch": 2.330876997557945, "grad_norm": 4.146524429321289, "learning_rate": 5.530316223787723e-05, "loss": 0.344400954246521, "memory(GiB)": 70.5, "step": 54405, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.449203 }, { "epoch": 2.3310912128871943, "grad_norm": 4.256633758544922, "learning_rate": 5.529647037686847e-05, "loss": 0.32012941837310793, "memory(GiB)": 70.5, "step": 54410, "token_acc": 0.9262820512820513, "train_speed(iter/s)": 1.449208 }, { "epoch": 2.331305428216443, "grad_norm": 3.950392723083496, "learning_rate": 5.528977841990919e-05, "loss": 0.2981738805770874, "memory(GiB)": 70.5, "step": 54415, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.449217 }, { "epoch": 2.331519643545692, "grad_norm": 4.241466045379639, "learning_rate": 5.5283086367120594e-05, "loss": 0.4397602081298828, "memory(GiB)": 70.5, "step": 54420, "token_acc": 0.9195046439628483, "train_speed(iter/s)": 1.449218 }, { "epoch": 2.3317338588749412, "grad_norm": 2.4338972568511963, "learning_rate": 5.527639421862395e-05, "loss": 0.4448698997497559, "memory(GiB)": 70.5, "step": 54425, "token_acc": 0.9049295774647887, "train_speed(iter/s)": 1.449234 }, { "epoch": 2.33194807420419, "grad_norm": 4.225515842437744, "learning_rate": 5.526970197454047e-05, "loss": 0.46112523078918455, "memory(GiB)": 70.5, "step": 54430, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.449235 }, { "epoch": 2.332162289533439, "grad_norm": 3.7345261573791504, "learning_rate": 5.52630096349914e-05, "loss": 0.6366782188415527, "memory(GiB)": 70.5, "step": 54435, "token_acc": 0.8553846153846154, "train_speed(iter/s)": 1.449236 }, { "epoch": 2.332376504862688, "grad_norm": 4.115697860717773, "learning_rate": 5.525631720009796e-05, "loss": 0.41658506393432615, "memory(GiB)": 70.5, "step": 54440, "token_acc": 0.8952380952380953, "train_speed(iter/s)": 1.449237 }, { "epoch": 2.332590720191937, "grad_norm": 2.2027361392974854, "learning_rate": 5.524962466998142e-05, "loss": 0.33987979888916015, "memory(GiB)": 70.5, "step": 54445, "token_acc": 0.9022082018927445, "train_speed(iter/s)": 1.449232 }, { "epoch": 2.3328049355211857, "grad_norm": 5.103825092315674, "learning_rate": 5.5242932044762994e-05, "loss": 0.25368366241455076, "memory(GiB)": 70.5, "step": 54450, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.449224 }, { "epoch": 2.333019150850435, "grad_norm": 1.8616772890090942, "learning_rate": 5.523623932456394e-05, "loss": 0.1512528896331787, "memory(GiB)": 70.5, "step": 54455, "token_acc": 0.9624060150375939, "train_speed(iter/s)": 1.449223 }, { "epoch": 2.333233366179684, "grad_norm": 1.9309933185577393, "learning_rate": 5.522954650950549e-05, "loss": 0.3420664548873901, "memory(GiB)": 70.5, "step": 54460, "token_acc": 0.9289940828402367, "train_speed(iter/s)": 1.449225 }, { "epoch": 2.3334475815089326, "grad_norm": 3.4978270530700684, "learning_rate": 5.522285359970891e-05, "loss": 0.3365072011947632, "memory(GiB)": 70.5, "step": 54465, "token_acc": 0.9164265129682997, "train_speed(iter/s)": 1.449226 }, { "epoch": 2.333661796838182, "grad_norm": 6.93479061126709, "learning_rate": 5.521616059529543e-05, "loss": 0.6519536972045898, "memory(GiB)": 70.5, "step": 54470, "token_acc": 0.8575851393188855, "train_speed(iter/s)": 1.44922 }, { "epoch": 2.3338760121674307, "grad_norm": 4.032729148864746, "learning_rate": 5.5209467496386315e-05, "loss": 0.2671054363250732, "memory(GiB)": 70.5, "step": 54475, "token_acc": 0.9471947194719472, "train_speed(iter/s)": 1.449229 }, { "epoch": 2.3340902274966795, "grad_norm": 0.5366532802581787, "learning_rate": 5.5202774303102814e-05, "loss": 0.38132011890411377, "memory(GiB)": 70.5, "step": 54480, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.449237 }, { "epoch": 2.3343044428259287, "grad_norm": 4.5426106452941895, "learning_rate": 5.519608101556616e-05, "loss": 0.47729787826538084, "memory(GiB)": 70.5, "step": 54485, "token_acc": 0.8786885245901639, "train_speed(iter/s)": 1.449242 }, { "epoch": 2.3345186581551776, "grad_norm": 2.0265843868255615, "learning_rate": 5.518938763389764e-05, "loss": 0.3303265333175659, "memory(GiB)": 70.5, "step": 54490, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.449246 }, { "epoch": 2.3347328734844264, "grad_norm": 3.139803647994995, "learning_rate": 5.5182694158218476e-05, "loss": 0.47084755897521974, "memory(GiB)": 70.5, "step": 54495, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.449242 }, { "epoch": 2.3349470888136756, "grad_norm": 1.4078712463378906, "learning_rate": 5.5176000588649946e-05, "loss": 0.2679582118988037, "memory(GiB)": 70.5, "step": 54500, "token_acc": 0.9385026737967914, "train_speed(iter/s)": 1.449231 }, { "epoch": 2.3349470888136756, "eval_loss": 2.4766149520874023, "eval_runtime": 13.4211, "eval_samples_per_second": 7.451, "eval_steps_per_second": 7.451, "eval_token_acc": 0.44, "step": 54500 }, { "epoch": 2.3351613041429244, "grad_norm": 2.0345373153686523, "learning_rate": 5.5169306925313324e-05, "loss": 0.232285475730896, "memory(GiB)": 70.5, "step": 54505, "token_acc": 0.5782567947516402, "train_speed(iter/s)": 1.448681 }, { "epoch": 2.3353755194721733, "grad_norm": 3.467186450958252, "learning_rate": 5.5162613168329856e-05, "loss": 0.5818545341491699, "memory(GiB)": 70.5, "step": 54510, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.448679 }, { "epoch": 2.3355897348014225, "grad_norm": 4.347537517547607, "learning_rate": 5.5155919317820806e-05, "loss": 0.31451687812805174, "memory(GiB)": 70.5, "step": 54515, "token_acc": 0.9212598425196851, "train_speed(iter/s)": 1.448683 }, { "epoch": 2.3358039501306713, "grad_norm": 3.291130304336548, "learning_rate": 5.514922537390744e-05, "loss": 0.24191052913665773, "memory(GiB)": 70.5, "step": 54520, "token_acc": 0.9283489096573209, "train_speed(iter/s)": 1.448686 }, { "epoch": 2.33601816545992, "grad_norm": 3.7675275802612305, "learning_rate": 5.5142531336711034e-05, "loss": 0.43506970405578616, "memory(GiB)": 70.5, "step": 54525, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.448687 }, { "epoch": 2.3362323807891694, "grad_norm": 0.4027826189994812, "learning_rate": 5.513583720635283e-05, "loss": 0.3212038516998291, "memory(GiB)": 70.5, "step": 54530, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.448686 }, { "epoch": 2.336446596118418, "grad_norm": 3.666111946105957, "learning_rate": 5.512914298295413e-05, "loss": 0.5975586414337158, "memory(GiB)": 70.5, "step": 54535, "token_acc": 0.8702290076335878, "train_speed(iter/s)": 1.448685 }, { "epoch": 2.336660811447667, "grad_norm": 4.327959060668945, "learning_rate": 5.51224486666362e-05, "loss": 0.597440528869629, "memory(GiB)": 70.5, "step": 54540, "token_acc": 0.8721311475409836, "train_speed(iter/s)": 1.448684 }, { "epoch": 2.3368750267769163, "grad_norm": 3.365518808364868, "learning_rate": 5.511575425752029e-05, "loss": 0.7189579486846924, "memory(GiB)": 70.5, "step": 54545, "token_acc": 0.8705035971223022, "train_speed(iter/s)": 1.448675 }, { "epoch": 2.337089242106165, "grad_norm": 2.760312080383301, "learning_rate": 5.510905975572771e-05, "loss": 0.5760009765625, "memory(GiB)": 70.5, "step": 54550, "token_acc": 0.8953846153846153, "train_speed(iter/s)": 1.448673 }, { "epoch": 2.337303457435414, "grad_norm": 3.0420753955841064, "learning_rate": 5.51023651613797e-05, "loss": 0.29367125034332275, "memory(GiB)": 70.5, "step": 54555, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.448668 }, { "epoch": 2.337517672764663, "grad_norm": 1.2094660997390747, "learning_rate": 5.5095670474597574e-05, "loss": 0.4208981037139893, "memory(GiB)": 70.5, "step": 54560, "token_acc": 0.9137380191693291, "train_speed(iter/s)": 1.448675 }, { "epoch": 2.337731888093912, "grad_norm": 0.30365797877311707, "learning_rate": 5.508897569550259e-05, "loss": 0.22894582748413086, "memory(GiB)": 70.5, "step": 54565, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.448667 }, { "epoch": 2.337946103423161, "grad_norm": 6.083972930908203, "learning_rate": 5.5082280824216046e-05, "loss": 0.5062716007232666, "memory(GiB)": 70.5, "step": 54570, "token_acc": 0.8735632183908046, "train_speed(iter/s)": 1.448677 }, { "epoch": 2.33816031875241, "grad_norm": 2.986250162124634, "learning_rate": 5.5075585860859206e-05, "loss": 0.4371354579925537, "memory(GiB)": 70.5, "step": 54575, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.448687 }, { "epoch": 2.338374534081659, "grad_norm": 3.545308828353882, "learning_rate": 5.5068890805553374e-05, "loss": 0.2799223899841309, "memory(GiB)": 70.5, "step": 54580, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.448694 }, { "epoch": 2.3385887494109077, "grad_norm": 2.352712631225586, "learning_rate": 5.506219565841983e-05, "loss": 0.3813108682632446, "memory(GiB)": 70.5, "step": 54585, "token_acc": 0.9312977099236641, "train_speed(iter/s)": 1.448703 }, { "epoch": 2.338802964740157, "grad_norm": 4.952893257141113, "learning_rate": 5.5055500419579866e-05, "loss": 0.5346709251403808, "memory(GiB)": 70.5, "step": 54590, "token_acc": 0.8696969696969697, "train_speed(iter/s)": 1.448715 }, { "epoch": 2.3390171800694057, "grad_norm": 5.971453666687012, "learning_rate": 5.5048805089154765e-05, "loss": 0.34707629680633545, "memory(GiB)": 70.5, "step": 54595, "token_acc": 0.9319148936170213, "train_speed(iter/s)": 1.448709 }, { "epoch": 2.3392313953986545, "grad_norm": 4.4381232261657715, "learning_rate": 5.504210966726582e-05, "loss": 0.34178762435913085, "memory(GiB)": 70.5, "step": 54600, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.448712 }, { "epoch": 2.339445610727904, "grad_norm": 5.007004737854004, "learning_rate": 5.503541415403433e-05, "loss": 0.36654033660888674, "memory(GiB)": 70.5, "step": 54605, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.448713 }, { "epoch": 2.3396598260571526, "grad_norm": 2.511691093444824, "learning_rate": 5.502871854958159e-05, "loss": 0.28156242370605467, "memory(GiB)": 70.5, "step": 54610, "token_acc": 0.9325513196480938, "train_speed(iter/s)": 1.44871 }, { "epoch": 2.3398740413864014, "grad_norm": 1.4969135522842407, "learning_rate": 5.502202285402889e-05, "loss": 0.5587126255035401, "memory(GiB)": 70.5, "step": 54615, "token_acc": 0.8690909090909091, "train_speed(iter/s)": 1.448715 }, { "epoch": 2.3400882567156507, "grad_norm": 3.313150644302368, "learning_rate": 5.5015327067497544e-05, "loss": 0.47501540184020996, "memory(GiB)": 70.5, "step": 54620, "token_acc": 0.8802395209580839, "train_speed(iter/s)": 1.448707 }, { "epoch": 2.3403024720448995, "grad_norm": 2.4516522884368896, "learning_rate": 5.500863119010885e-05, "loss": 0.32619657516479494, "memory(GiB)": 70.5, "step": 54625, "token_acc": 0.9157608695652174, "train_speed(iter/s)": 1.448701 }, { "epoch": 2.3405166873741483, "grad_norm": 2.675204038619995, "learning_rate": 5.500193522198409e-05, "loss": 0.2507157802581787, "memory(GiB)": 70.5, "step": 54630, "token_acc": 0.9377431906614786, "train_speed(iter/s)": 1.448703 }, { "epoch": 2.3407309027033976, "grad_norm": 3.196462869644165, "learning_rate": 5.499523916324459e-05, "loss": 0.4253035545349121, "memory(GiB)": 70.5, "step": 54635, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.448705 }, { "epoch": 2.3409451180326464, "grad_norm": 6.949106216430664, "learning_rate": 5.498854301401164e-05, "loss": 0.5628348350524902, "memory(GiB)": 70.5, "step": 54640, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.448707 }, { "epoch": 2.341159333361895, "grad_norm": 2.9725534915924072, "learning_rate": 5.4981846774406555e-05, "loss": 0.44736828804016116, "memory(GiB)": 70.5, "step": 54645, "token_acc": 0.9111969111969112, "train_speed(iter/s)": 1.448708 }, { "epoch": 2.3413735486911444, "grad_norm": 3.478980779647827, "learning_rate": 5.497515044455065e-05, "loss": 0.36940999031066896, "memory(GiB)": 70.5, "step": 54650, "token_acc": 0.936026936026936, "train_speed(iter/s)": 1.448718 }, { "epoch": 2.3415877640203933, "grad_norm": 11.000109672546387, "learning_rate": 5.496845402456522e-05, "loss": 0.459564733505249, "memory(GiB)": 70.5, "step": 54655, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.44872 }, { "epoch": 2.341801979349642, "grad_norm": 5.034777641296387, "learning_rate": 5.496175751457159e-05, "loss": 0.260945463180542, "memory(GiB)": 70.5, "step": 54660, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.448712 }, { "epoch": 2.3420161946788913, "grad_norm": 3.2083706855773926, "learning_rate": 5.495506091469106e-05, "loss": 0.31927878856658937, "memory(GiB)": 70.5, "step": 54665, "token_acc": 0.9337175792507204, "train_speed(iter/s)": 1.448712 }, { "epoch": 2.34223041000814, "grad_norm": 1.4760757684707642, "learning_rate": 5.494836422504497e-05, "loss": 0.3157083988189697, "memory(GiB)": 70.5, "step": 54670, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.448721 }, { "epoch": 2.342444625337389, "grad_norm": 4.180158615112305, "learning_rate": 5.49416674457546e-05, "loss": 0.2443983554840088, "memory(GiB)": 70.5, "step": 54675, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.448726 }, { "epoch": 2.342658840666638, "grad_norm": 3.7097086906433105, "learning_rate": 5.493497057694129e-05, "loss": 0.38625168800354004, "memory(GiB)": 70.5, "step": 54680, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.448763 }, { "epoch": 2.342873055995887, "grad_norm": 3.3338510990142822, "learning_rate": 5.492827361872637e-05, "loss": 0.48162097930908204, "memory(GiB)": 70.5, "step": 54685, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.448766 }, { "epoch": 2.343087271325136, "grad_norm": 4.358829975128174, "learning_rate": 5.492157657123114e-05, "loss": 0.3793545961380005, "memory(GiB)": 70.5, "step": 54690, "token_acc": 0.9263803680981595, "train_speed(iter/s)": 1.448755 }, { "epoch": 2.343301486654385, "grad_norm": 5.751448154449463, "learning_rate": 5.4914879434576936e-05, "loss": 0.5676261425018311, "memory(GiB)": 70.5, "step": 54695, "token_acc": 0.905511811023622, "train_speed(iter/s)": 1.448765 }, { "epoch": 2.343515701983634, "grad_norm": 4.152568340301514, "learning_rate": 5.490818220888508e-05, "loss": 0.4024346828460693, "memory(GiB)": 70.5, "step": 54700, "token_acc": 0.8960674157303371, "train_speed(iter/s)": 1.448778 }, { "epoch": 2.3437299173128827, "grad_norm": 2.009751319885254, "learning_rate": 5.4901484894276914e-05, "loss": 0.47971444129943847, "memory(GiB)": 70.5, "step": 54705, "token_acc": 0.8908450704225352, "train_speed(iter/s)": 1.448776 }, { "epoch": 2.343944132642132, "grad_norm": 6.238495349884033, "learning_rate": 5.4894787490873746e-05, "loss": 0.4815852165222168, "memory(GiB)": 70.5, "step": 54710, "token_acc": 0.9086294416243654, "train_speed(iter/s)": 1.448777 }, { "epoch": 2.344158347971381, "grad_norm": 2.0660526752471924, "learning_rate": 5.48880899987969e-05, "loss": 0.49589691162109373, "memory(GiB)": 70.5, "step": 54715, "token_acc": 0.9034267912772586, "train_speed(iter/s)": 1.448791 }, { "epoch": 2.34437256330063, "grad_norm": 0.09132685512304306, "learning_rate": 5.488139241816771e-05, "loss": 0.20443391799926758, "memory(GiB)": 70.5, "step": 54720, "token_acc": 0.9578313253012049, "train_speed(iter/s)": 1.448797 }, { "epoch": 2.344586778629879, "grad_norm": 0.24600735306739807, "learning_rate": 5.487469474910754e-05, "loss": 0.20212278366088868, "memory(GiB)": 70.5, "step": 54725, "token_acc": 0.9484978540772532, "train_speed(iter/s)": 1.448794 }, { "epoch": 2.3448009939591277, "grad_norm": 6.704759120941162, "learning_rate": 5.486799699173768e-05, "loss": 0.31655848026275635, "memory(GiB)": 70.5, "step": 54730, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.448792 }, { "epoch": 2.345015209288377, "grad_norm": 4.390979290008545, "learning_rate": 5.486129914617951e-05, "loss": 0.48533382415771487, "memory(GiB)": 70.5, "step": 54735, "token_acc": 0.8856304985337243, "train_speed(iter/s)": 1.448792 }, { "epoch": 2.3452294246176257, "grad_norm": 3.0387964248657227, "learning_rate": 5.485460121255434e-05, "loss": 0.39120934009552, "memory(GiB)": 70.5, "step": 54740, "token_acc": 0.90625, "train_speed(iter/s)": 1.448805 }, { "epoch": 2.3454436399468745, "grad_norm": 0.5075785517692566, "learning_rate": 5.484790319098351e-05, "loss": 0.39440288543701174, "memory(GiB)": 70.5, "step": 54745, "token_acc": 0.9049429657794676, "train_speed(iter/s)": 1.448814 }, { "epoch": 2.345657855276124, "grad_norm": 1.6312224864959717, "learning_rate": 5.4841205081588354e-05, "loss": 0.2955231904983521, "memory(GiB)": 70.5, "step": 54750, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.448816 }, { "epoch": 2.3458720706053726, "grad_norm": 3.3077850341796875, "learning_rate": 5.483450688449024e-05, "loss": 0.19659414291381835, "memory(GiB)": 70.5, "step": 54755, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.448822 }, { "epoch": 2.3460862859346214, "grad_norm": 1.8094319105148315, "learning_rate": 5.482780859981049e-05, "loss": 0.42279973030090334, "memory(GiB)": 70.5, "step": 54760, "token_acc": 0.9171779141104295, "train_speed(iter/s)": 1.448825 }, { "epoch": 2.3463005012638707, "grad_norm": 3.3530495166778564, "learning_rate": 5.482111022767048e-05, "loss": 0.3348376274108887, "memory(GiB)": 70.5, "step": 54765, "token_acc": 0.9288389513108615, "train_speed(iter/s)": 1.448827 }, { "epoch": 2.3465147165931195, "grad_norm": 2.969212770462036, "learning_rate": 5.481441176819152e-05, "loss": 0.33481578826904296, "memory(GiB)": 70.5, "step": 54770, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.448829 }, { "epoch": 2.3467289319223683, "grad_norm": 5.199581146240234, "learning_rate": 5.480771322149497e-05, "loss": 0.6244299411773682, "memory(GiB)": 70.5, "step": 54775, "token_acc": 0.8651026392961877, "train_speed(iter/s)": 1.448834 }, { "epoch": 2.3469431472516176, "grad_norm": 4.305105686187744, "learning_rate": 5.480101458770221e-05, "loss": 0.6714318275451661, "memory(GiB)": 70.5, "step": 54780, "token_acc": 0.8551236749116607, "train_speed(iter/s)": 1.448855 }, { "epoch": 2.3471573625808664, "grad_norm": 3.7247395515441895, "learning_rate": 5.479431586693454e-05, "loss": 0.2218801736831665, "memory(GiB)": 70.5, "step": 54785, "token_acc": 0.9458333333333333, "train_speed(iter/s)": 1.448868 }, { "epoch": 2.347371577910115, "grad_norm": 3.0545804500579834, "learning_rate": 5.478761705931334e-05, "loss": 0.3074420213699341, "memory(GiB)": 70.5, "step": 54790, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.448872 }, { "epoch": 2.3475857932393644, "grad_norm": 1.4347295761108398, "learning_rate": 5.478091816495997e-05, "loss": 0.34061198234558104, "memory(GiB)": 70.5, "step": 54795, "token_acc": 0.9338842975206612, "train_speed(iter/s)": 1.448869 }, { "epoch": 2.3478000085686133, "grad_norm": 3.6501474380493164, "learning_rate": 5.4774219183995804e-05, "loss": 0.44470901489257814, "memory(GiB)": 70.5, "step": 54800, "token_acc": 0.9146005509641874, "train_speed(iter/s)": 1.448871 }, { "epoch": 2.348014223897862, "grad_norm": 8.084715843200684, "learning_rate": 5.476752011654215e-05, "loss": 0.4974452018737793, "memory(GiB)": 70.5, "step": 54805, "token_acc": 0.8968253968253969, "train_speed(iter/s)": 1.448887 }, { "epoch": 2.3482284392271113, "grad_norm": 3.607499599456787, "learning_rate": 5.4760820962720416e-05, "loss": 0.5124674320220948, "memory(GiB)": 70.5, "step": 54810, "token_acc": 0.8796296296296297, "train_speed(iter/s)": 1.448891 }, { "epoch": 2.34844265455636, "grad_norm": 3.40412974357605, "learning_rate": 5.475412172265193e-05, "loss": 0.4927422523498535, "memory(GiB)": 70.5, "step": 54815, "token_acc": 0.9097744360902256, "train_speed(iter/s)": 1.448889 }, { "epoch": 2.348656869885609, "grad_norm": 5.340081691741943, "learning_rate": 5.4747422396458085e-05, "loss": 0.47985310554504396, "memory(GiB)": 70.5, "step": 54820, "token_acc": 0.9125475285171103, "train_speed(iter/s)": 1.448896 }, { "epoch": 2.348871085214858, "grad_norm": 1.3284178972244263, "learning_rate": 5.4740722984260216e-05, "loss": 0.3664852619171143, "memory(GiB)": 70.5, "step": 54825, "token_acc": 0.9208333333333333, "train_speed(iter/s)": 1.448902 }, { "epoch": 2.349085300544107, "grad_norm": 2.0341227054595947, "learning_rate": 5.473402348617971e-05, "loss": 0.28169817924499513, "memory(GiB)": 70.5, "step": 54830, "token_acc": 0.9355828220858896, "train_speed(iter/s)": 1.448908 }, { "epoch": 2.349299515873356, "grad_norm": 4.103806972503662, "learning_rate": 5.472732390233792e-05, "loss": 0.4671617031097412, "memory(GiB)": 70.5, "step": 54835, "token_acc": 0.9029126213592233, "train_speed(iter/s)": 1.448907 }, { "epoch": 2.349513731202605, "grad_norm": 3.4400126934051514, "learning_rate": 5.472062423285623e-05, "loss": 0.4676655769348145, "memory(GiB)": 70.5, "step": 54840, "token_acc": 0.9068825910931174, "train_speed(iter/s)": 1.448898 }, { "epoch": 2.349727946531854, "grad_norm": 2.016002655029297, "learning_rate": 5.4713924477856006e-05, "loss": 0.38626389503479003, "memory(GiB)": 70.5, "step": 54845, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.448896 }, { "epoch": 2.3499421618611027, "grad_norm": 3.794734239578247, "learning_rate": 5.470722463745862e-05, "loss": 0.4361108779907227, "memory(GiB)": 70.5, "step": 54850, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.448904 }, { "epoch": 2.350156377190352, "grad_norm": 6.814071178436279, "learning_rate": 5.4700524711785436e-05, "loss": 0.5949673652648926, "memory(GiB)": 70.5, "step": 54855, "token_acc": 0.818407960199005, "train_speed(iter/s)": 1.448897 }, { "epoch": 2.350370592519601, "grad_norm": 3.8398618698120117, "learning_rate": 5.469382470095784e-05, "loss": 0.3697032928466797, "memory(GiB)": 70.5, "step": 54860, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.448915 }, { "epoch": 2.3505848078488496, "grad_norm": 2.0146751403808594, "learning_rate": 5.4687124605097196e-05, "loss": 0.6839057445526123, "memory(GiB)": 70.5, "step": 54865, "token_acc": 0.8720238095238095, "train_speed(iter/s)": 1.448919 }, { "epoch": 2.350799023178099, "grad_norm": 9.449298858642578, "learning_rate": 5.4680424424324914e-05, "loss": 0.44599413871765137, "memory(GiB)": 70.5, "step": 54870, "token_acc": 0.8929663608562691, "train_speed(iter/s)": 1.448915 }, { "epoch": 2.3510132385073477, "grad_norm": 0.7485092878341675, "learning_rate": 5.467372415876233e-05, "loss": 0.34730513095855714, "memory(GiB)": 70.5, "step": 54875, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.448919 }, { "epoch": 2.3512274538365965, "grad_norm": 3.689138412475586, "learning_rate": 5.466702380853087e-05, "loss": 0.3434727430343628, "memory(GiB)": 70.5, "step": 54880, "token_acc": 0.9271255060728745, "train_speed(iter/s)": 1.448922 }, { "epoch": 2.3514416691658457, "grad_norm": 3.2595038414001465, "learning_rate": 5.4660323373751884e-05, "loss": 0.36938698291778566, "memory(GiB)": 70.5, "step": 54885, "token_acc": 0.9066147859922179, "train_speed(iter/s)": 1.448918 }, { "epoch": 2.3516558844950946, "grad_norm": 1.0526481866836548, "learning_rate": 5.465362285454677e-05, "loss": 0.3410006523132324, "memory(GiB)": 70.5, "step": 54890, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.448924 }, { "epoch": 2.3518700998243434, "grad_norm": 0.6592255234718323, "learning_rate": 5.4646922251036924e-05, "loss": 0.217684006690979, "memory(GiB)": 70.5, "step": 54895, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.448922 }, { "epoch": 2.3520843151535926, "grad_norm": 3.4127299785614014, "learning_rate": 5.464022156334371e-05, "loss": 0.5039616584777832, "memory(GiB)": 70.5, "step": 54900, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.448925 }, { "epoch": 2.3522985304828414, "grad_norm": 3.118410110473633, "learning_rate": 5.463352079158852e-05, "loss": 0.24376914501190186, "memory(GiB)": 70.5, "step": 54905, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.44894 }, { "epoch": 2.3525127458120902, "grad_norm": 1.9958930015563965, "learning_rate": 5.462681993589277e-05, "loss": 0.18064956665039061, "memory(GiB)": 70.5, "step": 54910, "token_acc": 0.96, "train_speed(iter/s)": 1.448942 }, { "epoch": 2.3527269611413395, "grad_norm": 3.0787875652313232, "learning_rate": 5.462011899637783e-05, "loss": 0.19415065050125122, "memory(GiB)": 70.5, "step": 54915, "token_acc": 0.9570957095709571, "train_speed(iter/s)": 1.448943 }, { "epoch": 2.3529411764705883, "grad_norm": 1.4844306707382202, "learning_rate": 5.4613417973165106e-05, "loss": 0.6858043193817138, "memory(GiB)": 70.5, "step": 54920, "token_acc": 0.8507462686567164, "train_speed(iter/s)": 1.448939 }, { "epoch": 2.353155391799837, "grad_norm": 3.0115673542022705, "learning_rate": 5.4606716866375985e-05, "loss": 0.25080204010009766, "memory(GiB)": 70.5, "step": 54925, "token_acc": 0.9389067524115756, "train_speed(iter/s)": 1.448944 }, { "epoch": 2.3533696071290864, "grad_norm": 2.4867000579833984, "learning_rate": 5.460001567613188e-05, "loss": 0.4379605770111084, "memory(GiB)": 70.5, "step": 54930, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.448964 }, { "epoch": 2.353583822458335, "grad_norm": 4.0065155029296875, "learning_rate": 5.459331440255414e-05, "loss": 0.3709320306777954, "memory(GiB)": 70.5, "step": 54935, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.448969 }, { "epoch": 2.353798037787584, "grad_norm": 5.483132362365723, "learning_rate": 5.458661304576422e-05, "loss": 0.5030012130737305, "memory(GiB)": 70.5, "step": 54940, "token_acc": 0.9094488188976378, "train_speed(iter/s)": 1.448963 }, { "epoch": 2.3540122531168333, "grad_norm": 2.0243680477142334, "learning_rate": 5.457991160588351e-05, "loss": 0.32144627571105955, "memory(GiB)": 70.5, "step": 54945, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.448972 }, { "epoch": 2.354226468446082, "grad_norm": 3.2237117290496826, "learning_rate": 5.45732100830334e-05, "loss": 0.3822195529937744, "memory(GiB)": 70.5, "step": 54950, "token_acc": 0.909433962264151, "train_speed(iter/s)": 1.448982 }, { "epoch": 2.354440683775331, "grad_norm": 2.2958953380584717, "learning_rate": 5.4566508477335296e-05, "loss": 0.2880213975906372, "memory(GiB)": 70.5, "step": 54955, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.448988 }, { "epoch": 2.35465489910458, "grad_norm": 2.323312282562256, "learning_rate": 5.4559806788910606e-05, "loss": 0.3974114656448364, "memory(GiB)": 70.5, "step": 54960, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.448986 }, { "epoch": 2.354869114433829, "grad_norm": 1.5930956602096558, "learning_rate": 5.455310501788075e-05, "loss": 0.1525997996330261, "memory(GiB)": 70.5, "step": 54965, "token_acc": 0.9616858237547893, "train_speed(iter/s)": 1.448975 }, { "epoch": 2.3550833297630778, "grad_norm": 2.9893481731414795, "learning_rate": 5.454640316436712e-05, "loss": 0.5507631778717041, "memory(GiB)": 70.5, "step": 54970, "token_acc": 0.9010238907849829, "train_speed(iter/s)": 1.44898 }, { "epoch": 2.355297545092327, "grad_norm": 1.9076547622680664, "learning_rate": 5.453970122849113e-05, "loss": 0.2704340696334839, "memory(GiB)": 70.5, "step": 54975, "token_acc": 0.933920704845815, "train_speed(iter/s)": 1.448986 }, { "epoch": 2.355511760421576, "grad_norm": 11.831310272216797, "learning_rate": 5.453299921037418e-05, "loss": 0.3260432004928589, "memory(GiB)": 70.5, "step": 54980, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.448978 }, { "epoch": 2.3557259757508247, "grad_norm": 2.675819158554077, "learning_rate": 5.452629711013773e-05, "loss": 0.5603820323944092, "memory(GiB)": 70.5, "step": 54985, "token_acc": 0.878125, "train_speed(iter/s)": 1.448981 }, { "epoch": 2.355940191080074, "grad_norm": 3.4541001319885254, "learning_rate": 5.451959492790313e-05, "loss": 0.17293593883514405, "memory(GiB)": 70.5, "step": 54990, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.448986 }, { "epoch": 2.3561544064093227, "grad_norm": 3.0674033164978027, "learning_rate": 5.451289266379184e-05, "loss": 0.2837062835693359, "memory(GiB)": 70.5, "step": 54995, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.448993 }, { "epoch": 2.3563686217385715, "grad_norm": 1.508364200592041, "learning_rate": 5.4506190317925275e-05, "loss": 0.39808950424194334, "memory(GiB)": 70.5, "step": 55000, "token_acc": 0.9295302013422819, "train_speed(iter/s)": 1.449011 }, { "epoch": 2.3563686217385715, "eval_loss": 2.496994972229004, "eval_runtime": 13.6414, "eval_samples_per_second": 7.331, "eval_steps_per_second": 7.331, "eval_token_acc": 0.43898809523809523, "step": 55000 }, { "epoch": 2.356582837067821, "grad_norm": 3.945523262023926, "learning_rate": 5.4499487890424826e-05, "loss": 0.5330152034759521, "memory(GiB)": 70.5, "step": 55005, "token_acc": 0.5708245243128964, "train_speed(iter/s)": 1.448455 }, { "epoch": 2.3567970523970696, "grad_norm": 2.2528886795043945, "learning_rate": 5.4492785381411936e-05, "loss": 0.44075589179992675, "memory(GiB)": 70.5, "step": 55010, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.448467 }, { "epoch": 2.3570112677263184, "grad_norm": 4.847831726074219, "learning_rate": 5.4486082791008044e-05, "loss": 0.5679641723632812, "memory(GiB)": 70.5, "step": 55015, "token_acc": 0.8837209302325582, "train_speed(iter/s)": 1.448475 }, { "epoch": 2.3572254830555677, "grad_norm": 4.304424285888672, "learning_rate": 5.447938011933453e-05, "loss": 0.25655364990234375, "memory(GiB)": 70.5, "step": 55020, "token_acc": 0.9375, "train_speed(iter/s)": 1.448478 }, { "epoch": 2.3574396983848165, "grad_norm": 5.708155632019043, "learning_rate": 5.447267736651286e-05, "loss": 0.5851702213287353, "memory(GiB)": 70.5, "step": 55025, "token_acc": 0.875, "train_speed(iter/s)": 1.448484 }, { "epoch": 2.3576539137140653, "grad_norm": 3.800816297531128, "learning_rate": 5.446597453266444e-05, "loss": 0.48370981216430664, "memory(GiB)": 70.5, "step": 55030, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.448481 }, { "epoch": 2.3578681290433146, "grad_norm": 0.29302966594696045, "learning_rate": 5.44592716179107e-05, "loss": 0.24502294063568114, "memory(GiB)": 70.5, "step": 55035, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.44848 }, { "epoch": 2.3580823443725634, "grad_norm": 3.7983055114746094, "learning_rate": 5.445256862237308e-05, "loss": 0.44464893341064454, "memory(GiB)": 70.5, "step": 55040, "token_acc": 0.8962962962962963, "train_speed(iter/s)": 1.448473 }, { "epoch": 2.358296559701812, "grad_norm": 2.18605375289917, "learning_rate": 5.4445865546172995e-05, "loss": 0.2741574287414551, "memory(GiB)": 70.5, "step": 55045, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.44847 }, { "epoch": 2.3585107750310614, "grad_norm": 3.73864483833313, "learning_rate": 5.4439162389431876e-05, "loss": 0.43577189445495607, "memory(GiB)": 70.5, "step": 55050, "token_acc": 0.90633608815427, "train_speed(iter/s)": 1.448475 }, { "epoch": 2.3587249903603102, "grad_norm": 4.087780475616455, "learning_rate": 5.443245915227117e-05, "loss": 0.30202105045318606, "memory(GiB)": 70.5, "step": 55055, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.448483 }, { "epoch": 2.358939205689559, "grad_norm": 5.365493297576904, "learning_rate": 5.442575583481232e-05, "loss": 0.35280284881591795, "memory(GiB)": 70.5, "step": 55060, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.448501 }, { "epoch": 2.3591534210188083, "grad_norm": 1.5115262269973755, "learning_rate": 5.441905243717674e-05, "loss": 0.463959264755249, "memory(GiB)": 70.5, "step": 55065, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.448497 }, { "epoch": 2.359367636348057, "grad_norm": 2.9996426105499268, "learning_rate": 5.4412348959485894e-05, "loss": 0.4334695816040039, "memory(GiB)": 70.5, "step": 55070, "token_acc": 0.8937728937728938, "train_speed(iter/s)": 1.448486 }, { "epoch": 2.359581851677306, "grad_norm": 3.7456629276275635, "learning_rate": 5.4405645401861205e-05, "loss": 0.2805644989013672, "memory(GiB)": 70.5, "step": 55075, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.448489 }, { "epoch": 2.359796067006555, "grad_norm": 3.1305575370788574, "learning_rate": 5.439894176442409e-05, "loss": 0.35301599502563474, "memory(GiB)": 70.5, "step": 55080, "token_acc": 0.9261213720316622, "train_speed(iter/s)": 1.448491 }, { "epoch": 2.360010282335804, "grad_norm": 0.6324419379234314, "learning_rate": 5.439223804729604e-05, "loss": 0.14790232181549073, "memory(GiB)": 70.5, "step": 55085, "token_acc": 0.9611307420494699, "train_speed(iter/s)": 1.448487 }, { "epoch": 2.360224497665053, "grad_norm": 1.7106901407241821, "learning_rate": 5.438553425059848e-05, "loss": 0.3613922595977783, "memory(GiB)": 70.5, "step": 55090, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.448489 }, { "epoch": 2.360438712994302, "grad_norm": 3.5727460384368896, "learning_rate": 5.437883037445284e-05, "loss": 0.2650846004486084, "memory(GiB)": 70.5, "step": 55095, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.44848 }, { "epoch": 2.360652928323551, "grad_norm": 6.916569232940674, "learning_rate": 5.43721264189806e-05, "loss": 0.4753739833831787, "memory(GiB)": 70.5, "step": 55100, "token_acc": 0.9186046511627907, "train_speed(iter/s)": 1.448484 }, { "epoch": 2.3608671436527997, "grad_norm": 2.5760231018066406, "learning_rate": 5.4365422384303164e-05, "loss": 0.2099174976348877, "memory(GiB)": 70.5, "step": 55105, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.448491 }, { "epoch": 2.361081358982049, "grad_norm": 4.60403299331665, "learning_rate": 5.435871827054203e-05, "loss": 0.49581236839294435, "memory(GiB)": 70.5, "step": 55110, "token_acc": 0.9025157232704403, "train_speed(iter/s)": 1.448489 }, { "epoch": 2.3612955743112978, "grad_norm": 6.73586368560791, "learning_rate": 5.435201407781863e-05, "loss": 0.5708225250244141, "memory(GiB)": 70.5, "step": 55115, "token_acc": 0.8996539792387543, "train_speed(iter/s)": 1.448517 }, { "epoch": 2.3615097896405466, "grad_norm": 4.306656360626221, "learning_rate": 5.434530980625438e-05, "loss": 0.4340690612792969, "memory(GiB)": 70.5, "step": 55120, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.448528 }, { "epoch": 2.361724004969796, "grad_norm": 0.9333075284957886, "learning_rate": 5.433860545597078e-05, "loss": 0.3950053215026855, "memory(GiB)": 70.5, "step": 55125, "token_acc": 0.9044585987261147, "train_speed(iter/s)": 1.44853 }, { "epoch": 2.3619382202990447, "grad_norm": 4.33928108215332, "learning_rate": 5.433190102708928e-05, "loss": 0.27051348686218263, "memory(GiB)": 70.5, "step": 55130, "token_acc": 0.937007874015748, "train_speed(iter/s)": 1.448533 }, { "epoch": 2.3621524356282935, "grad_norm": 1.222993016242981, "learning_rate": 5.432519651973133e-05, "loss": 0.21632211208343505, "memory(GiB)": 70.5, "step": 55135, "token_acc": 0.9546925566343042, "train_speed(iter/s)": 1.448533 }, { "epoch": 2.3623666509575427, "grad_norm": 3.887842893600464, "learning_rate": 5.431849193401838e-05, "loss": 0.27980899810791016, "memory(GiB)": 70.5, "step": 55140, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.448528 }, { "epoch": 2.3625808662867915, "grad_norm": 0.5187107920646667, "learning_rate": 5.431178727007191e-05, "loss": 0.21198499202728271, "memory(GiB)": 70.5, "step": 55145, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.448532 }, { "epoch": 2.3627950816160403, "grad_norm": 3.515005111694336, "learning_rate": 5.430508252801335e-05, "loss": 0.29717793464660647, "memory(GiB)": 70.5, "step": 55150, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.448534 }, { "epoch": 2.3630092969452896, "grad_norm": 5.4751739501953125, "learning_rate": 5.42983777079642e-05, "loss": 0.48238444328308105, "memory(GiB)": 70.5, "step": 55155, "token_acc": 0.9045801526717557, "train_speed(iter/s)": 1.448529 }, { "epoch": 2.3632235122745384, "grad_norm": 2.2855305671691895, "learning_rate": 5.429167281004589e-05, "loss": 0.5223685741424561, "memory(GiB)": 70.5, "step": 55160, "token_acc": 0.8958990536277602, "train_speed(iter/s)": 1.448543 }, { "epoch": 2.3634377276037872, "grad_norm": 2.935891628265381, "learning_rate": 5.42849678343799e-05, "loss": 0.3862550973892212, "memory(GiB)": 70.5, "step": 55165, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.448541 }, { "epoch": 2.3636519429330365, "grad_norm": 2.4137966632843018, "learning_rate": 5.4278262781087706e-05, "loss": 0.15492690801620485, "memory(GiB)": 70.5, "step": 55170, "token_acc": 0.9694656488549618, "train_speed(iter/s)": 1.448543 }, { "epoch": 2.3638661582622853, "grad_norm": 3.421884059906006, "learning_rate": 5.427155765029078e-05, "loss": 0.4528526306152344, "memory(GiB)": 70.5, "step": 55175, "token_acc": 0.9, "train_speed(iter/s)": 1.448547 }, { "epoch": 2.364080373591534, "grad_norm": 2.7021644115448, "learning_rate": 5.426485244211056e-05, "loss": 0.42345123291015624, "memory(GiB)": 70.5, "step": 55180, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.448541 }, { "epoch": 2.3642945889207834, "grad_norm": 4.621204853057861, "learning_rate": 5.4258147156668545e-05, "loss": 0.6959856033325196, "memory(GiB)": 70.5, "step": 55185, "token_acc": 0.8528528528528528, "train_speed(iter/s)": 1.44853 }, { "epoch": 2.364508804250032, "grad_norm": 3.174862861633301, "learning_rate": 5.425144179408621e-05, "loss": 0.24636573791503907, "memory(GiB)": 70.5, "step": 55190, "token_acc": 0.9623824451410659, "train_speed(iter/s)": 1.44854 }, { "epoch": 2.364723019579281, "grad_norm": 1.2778573036193848, "learning_rate": 5.4244736354484994e-05, "loss": 0.2500413656234741, "memory(GiB)": 70.5, "step": 55195, "token_acc": 0.9501779359430605, "train_speed(iter/s)": 1.44854 }, { "epoch": 2.3649372349085303, "grad_norm": 2.3719308376312256, "learning_rate": 5.423803083798642e-05, "loss": 0.4668439865112305, "memory(GiB)": 70.5, "step": 55200, "token_acc": 0.8867313915857605, "train_speed(iter/s)": 1.448537 }, { "epoch": 2.365151450237779, "grad_norm": 3.2553982734680176, "learning_rate": 5.423132524471194e-05, "loss": 0.3799863576889038, "memory(GiB)": 70.5, "step": 55205, "token_acc": 0.9137254901960784, "train_speed(iter/s)": 1.448546 }, { "epoch": 2.365365665567028, "grad_norm": 2.9155383110046387, "learning_rate": 5.422461957478302e-05, "loss": 0.22402913570404054, "memory(GiB)": 70.5, "step": 55210, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.448559 }, { "epoch": 2.365579880896277, "grad_norm": 3.3568711280822754, "learning_rate": 5.4217913828321166e-05, "loss": 0.24479999542236328, "memory(GiB)": 70.5, "step": 55215, "token_acc": 0.9378698224852071, "train_speed(iter/s)": 1.448556 }, { "epoch": 2.365794096225526, "grad_norm": 9.155698776245117, "learning_rate": 5.421120800544785e-05, "loss": 0.6192069053649902, "memory(GiB)": 70.5, "step": 55220, "token_acc": 0.8659003831417624, "train_speed(iter/s)": 1.448581 }, { "epoch": 2.3660083115547748, "grad_norm": 3.0822079181671143, "learning_rate": 5.420450210628454e-05, "loss": 0.3492306709289551, "memory(GiB)": 70.5, "step": 55225, "token_acc": 0.9298245614035088, "train_speed(iter/s)": 1.448597 }, { "epoch": 2.366222526884024, "grad_norm": 2.0736465454101562, "learning_rate": 5.4197796130952724e-05, "loss": 0.15809139013290405, "memory(GiB)": 70.5, "step": 55230, "token_acc": 0.9537037037037037, "train_speed(iter/s)": 1.448598 }, { "epoch": 2.366436742213273, "grad_norm": 0.5085158944129944, "learning_rate": 5.41910900795739e-05, "loss": 0.44775638580322263, "memory(GiB)": 70.5, "step": 55235, "token_acc": 0.9169960474308301, "train_speed(iter/s)": 1.448613 }, { "epoch": 2.3666509575425216, "grad_norm": 1.8827892541885376, "learning_rate": 5.418438395226954e-05, "loss": 0.29608726501464844, "memory(GiB)": 70.5, "step": 55240, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.448625 }, { "epoch": 2.366865172871771, "grad_norm": 0.49625223875045776, "learning_rate": 5.417767774916116e-05, "loss": 0.5954806804656982, "memory(GiB)": 70.5, "step": 55245, "token_acc": 0.864406779661017, "train_speed(iter/s)": 1.448633 }, { "epoch": 2.3670793882010197, "grad_norm": 3.2143354415893555, "learning_rate": 5.417097147037021e-05, "loss": 0.4798429012298584, "memory(GiB)": 70.5, "step": 55250, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.448639 }, { "epoch": 2.3672936035302685, "grad_norm": 3.4821720123291016, "learning_rate": 5.416426511601822e-05, "loss": 0.5504418849945069, "memory(GiB)": 70.5, "step": 55255, "token_acc": 0.8964401294498382, "train_speed(iter/s)": 1.448645 }, { "epoch": 2.3675078188595178, "grad_norm": 3.031144857406616, "learning_rate": 5.415755868622665e-05, "loss": 0.6363205432891845, "memory(GiB)": 70.5, "step": 55260, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.448646 }, { "epoch": 2.3677220341887666, "grad_norm": 4.798582553863525, "learning_rate": 5.4150852181116995e-05, "loss": 0.6234495639801025, "memory(GiB)": 70.5, "step": 55265, "token_acc": 0.8732394366197183, "train_speed(iter/s)": 1.448644 }, { "epoch": 2.3679362495180154, "grad_norm": 6.880427360534668, "learning_rate": 5.414414560081076e-05, "loss": 0.6265539169311524, "memory(GiB)": 70.5, "step": 55270, "token_acc": 0.8836477987421384, "train_speed(iter/s)": 1.448644 }, { "epoch": 2.3681504648472647, "grad_norm": 3.093266725540161, "learning_rate": 5.413743894542945e-05, "loss": 0.4588153839111328, "memory(GiB)": 70.5, "step": 55275, "token_acc": 0.8884615384615384, "train_speed(iter/s)": 1.448639 }, { "epoch": 2.3683646801765135, "grad_norm": 4.126387596130371, "learning_rate": 5.413073221509454e-05, "loss": 0.44532132148742676, "memory(GiB)": 70.5, "step": 55280, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.448642 }, { "epoch": 2.3685788955057623, "grad_norm": 0.3858708441257477, "learning_rate": 5.412402540992756e-05, "loss": 0.42006778717041016, "memory(GiB)": 70.5, "step": 55285, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.448644 }, { "epoch": 2.3687931108350115, "grad_norm": 5.490840911865234, "learning_rate": 5.4117318530049976e-05, "loss": 0.36324348449707033, "memory(GiB)": 70.5, "step": 55290, "token_acc": 0.9197530864197531, "train_speed(iter/s)": 1.448663 }, { "epoch": 2.3690073261642604, "grad_norm": 6.052123546600342, "learning_rate": 5.41106115755833e-05, "loss": 0.4248487949371338, "memory(GiB)": 70.5, "step": 55295, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.44867 }, { "epoch": 2.369221541493509, "grad_norm": 3.4180164337158203, "learning_rate": 5.4103904546649063e-05, "loss": 0.4240254402160645, "memory(GiB)": 70.5, "step": 55300, "token_acc": 0.8830409356725146, "train_speed(iter/s)": 1.448667 }, { "epoch": 2.3694357568227584, "grad_norm": 3.2468059062957764, "learning_rate": 5.4097197443368716e-05, "loss": 0.3759260416030884, "memory(GiB)": 70.5, "step": 55305, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.448668 }, { "epoch": 2.3696499721520072, "grad_norm": 2.947377920150757, "learning_rate": 5.4090490265863814e-05, "loss": 0.40709357261657714, "memory(GiB)": 70.5, "step": 55310, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.448677 }, { "epoch": 2.369864187481256, "grad_norm": 3.6253888607025146, "learning_rate": 5.408378301425584e-05, "loss": 0.10892674922943116, "memory(GiB)": 70.5, "step": 55315, "token_acc": 0.9704797047970479, "train_speed(iter/s)": 1.448687 }, { "epoch": 2.3700784028105053, "grad_norm": 3.6965136528015137, "learning_rate": 5.40770756886663e-05, "loss": 0.4726593494415283, "memory(GiB)": 70.5, "step": 55320, "token_acc": 0.8878504672897196, "train_speed(iter/s)": 1.448699 }, { "epoch": 2.370292618139754, "grad_norm": 4.203708648681641, "learning_rate": 5.407036828921671e-05, "loss": 0.5767871856689453, "memory(GiB)": 70.5, "step": 55325, "token_acc": 0.8774834437086093, "train_speed(iter/s)": 1.448687 }, { "epoch": 2.370506833469003, "grad_norm": 2.844010591506958, "learning_rate": 5.406366081602859e-05, "loss": 0.31044907569885255, "memory(GiB)": 70.5, "step": 55330, "token_acc": 0.9401041666666666, "train_speed(iter/s)": 1.448685 }, { "epoch": 2.370721048798252, "grad_norm": 4.771903038024902, "learning_rate": 5.4056953269223445e-05, "loss": 0.6263085842132569, "memory(GiB)": 70.5, "step": 55335, "token_acc": 0.8644578313253012, "train_speed(iter/s)": 1.448691 }, { "epoch": 2.370935264127501, "grad_norm": 2.6001927852630615, "learning_rate": 5.405024564892277e-05, "loss": 0.6063377380371093, "memory(GiB)": 70.5, "step": 55340, "token_acc": 0.8595890410958904, "train_speed(iter/s)": 1.448688 }, { "epoch": 2.37114947945675, "grad_norm": 4.412602424621582, "learning_rate": 5.4043537955248094e-05, "loss": 0.3938585758209229, "memory(GiB)": 70.5, "step": 55345, "token_acc": 0.9054545454545454, "train_speed(iter/s)": 1.448691 }, { "epoch": 2.371363694785999, "grad_norm": 2.467076063156128, "learning_rate": 5.403683018832094e-05, "loss": 0.2379439353942871, "memory(GiB)": 70.5, "step": 55350, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.448701 }, { "epoch": 2.371577910115248, "grad_norm": 2.868135929107666, "learning_rate": 5.4030122348262824e-05, "loss": 0.46555533409118655, "memory(GiB)": 70.5, "step": 55355, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.448702 }, { "epoch": 2.3717921254444967, "grad_norm": 4.36163330078125, "learning_rate": 5.402341443519526e-05, "loss": 0.4676025867462158, "memory(GiB)": 70.5, "step": 55360, "token_acc": 0.8839590443686007, "train_speed(iter/s)": 1.448704 }, { "epoch": 2.372006340773746, "grad_norm": 4.566256999969482, "learning_rate": 5.4016706449239764e-05, "loss": 0.34829447269439695, "memory(GiB)": 70.5, "step": 55365, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.448722 }, { "epoch": 2.3722205561029948, "grad_norm": 1.5845232009887695, "learning_rate": 5.400999839051788e-05, "loss": 0.27104527950286866, "memory(GiB)": 70.5, "step": 55370, "token_acc": 0.9252336448598131, "train_speed(iter/s)": 1.448739 }, { "epoch": 2.3724347714322436, "grad_norm": 1.2160667181015015, "learning_rate": 5.40032902591511e-05, "loss": 0.3324757099151611, "memory(GiB)": 70.5, "step": 55375, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.448743 }, { "epoch": 2.372648986761493, "grad_norm": 4.299962043762207, "learning_rate": 5.399658205526096e-05, "loss": 0.3201579570770264, "memory(GiB)": 70.5, "step": 55380, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.448749 }, { "epoch": 2.3728632020907416, "grad_norm": 3.842785120010376, "learning_rate": 5.398987377896898e-05, "loss": 0.28072280883789064, "memory(GiB)": 70.5, "step": 55385, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.44876 }, { "epoch": 2.3730774174199905, "grad_norm": 2.3229212760925293, "learning_rate": 5.398316543039671e-05, "loss": 0.179887855052948, "memory(GiB)": 70.5, "step": 55390, "token_acc": 0.9576923076923077, "train_speed(iter/s)": 1.448754 }, { "epoch": 2.3732916327492397, "grad_norm": 2.8391542434692383, "learning_rate": 5.397645700966565e-05, "loss": 0.4315822601318359, "memory(GiB)": 70.5, "step": 55395, "token_acc": 0.9149797570850202, "train_speed(iter/s)": 1.448751 }, { "epoch": 2.3735058480784885, "grad_norm": 3.65451717376709, "learning_rate": 5.3969748516897354e-05, "loss": 0.3085920810699463, "memory(GiB)": 70.5, "step": 55400, "token_acc": 0.9496124031007752, "train_speed(iter/s)": 1.448757 }, { "epoch": 2.3737200634077373, "grad_norm": 3.7555980682373047, "learning_rate": 5.3963039952213336e-05, "loss": 0.30926387310028075, "memory(GiB)": 70.5, "step": 55405, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.448752 }, { "epoch": 2.3739342787369866, "grad_norm": 2.081904649734497, "learning_rate": 5.395633131573512e-05, "loss": 0.4303161144256592, "memory(GiB)": 70.5, "step": 55410, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.448751 }, { "epoch": 2.3741484940662354, "grad_norm": 2.6512467861175537, "learning_rate": 5.394962260758425e-05, "loss": 0.3677434206008911, "memory(GiB)": 70.5, "step": 55415, "token_acc": 0.9404761904761905, "train_speed(iter/s)": 1.448753 }, { "epoch": 2.374362709395484, "grad_norm": 6.695333480834961, "learning_rate": 5.394291382788228e-05, "loss": 0.40309438705444334, "memory(GiB)": 70.5, "step": 55420, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.448761 }, { "epoch": 2.3745769247247335, "grad_norm": 5.40246057510376, "learning_rate": 5.39362049767507e-05, "loss": 0.2516184329986572, "memory(GiB)": 70.5, "step": 55425, "token_acc": 0.9550173010380623, "train_speed(iter/s)": 1.448781 }, { "epoch": 2.3747911400539823, "grad_norm": 4.41666841506958, "learning_rate": 5.3929496054311104e-05, "loss": 0.5556628227233886, "memory(GiB)": 70.5, "step": 55430, "token_acc": 0.8449367088607594, "train_speed(iter/s)": 1.448772 }, { "epoch": 2.375005355383231, "grad_norm": 3.0151805877685547, "learning_rate": 5.3922787060685e-05, "loss": 0.32575235366821287, "memory(GiB)": 70.5, "step": 55435, "token_acc": 0.9289940828402367, "train_speed(iter/s)": 1.448784 }, { "epoch": 2.3752195707124804, "grad_norm": 3.8827106952667236, "learning_rate": 5.391607799599391e-05, "loss": 0.7588991641998291, "memory(GiB)": 70.5, "step": 55440, "token_acc": 0.8161993769470405, "train_speed(iter/s)": 1.448794 }, { "epoch": 2.375433786041729, "grad_norm": 1.4717357158660889, "learning_rate": 5.39093688603594e-05, "loss": 0.3492881298065186, "memory(GiB)": 70.5, "step": 55445, "token_acc": 0.9357429718875502, "train_speed(iter/s)": 1.448794 }, { "epoch": 2.375648001370978, "grad_norm": 3.427417039871216, "learning_rate": 5.390265965390301e-05, "loss": 0.2513458490371704, "memory(GiB)": 70.5, "step": 55450, "token_acc": 0.9392712550607287, "train_speed(iter/s)": 1.448792 }, { "epoch": 2.3758622167002272, "grad_norm": 6.210406303405762, "learning_rate": 5.3895950376746266e-05, "loss": 0.46922807693481444, "memory(GiB)": 70.5, "step": 55455, "token_acc": 0.8959537572254336, "train_speed(iter/s)": 1.448782 }, { "epoch": 2.376076432029476, "grad_norm": 2.6158251762390137, "learning_rate": 5.388924102901074e-05, "loss": 0.3658011436462402, "memory(GiB)": 70.5, "step": 55460, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.448794 }, { "epoch": 2.376290647358725, "grad_norm": 2.6341423988342285, "learning_rate": 5.388253161081795e-05, "loss": 0.40719308853149416, "memory(GiB)": 70.5, "step": 55465, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.448808 }, { "epoch": 2.376504862687974, "grad_norm": 2.434204339981079, "learning_rate": 5.387582212228948e-05, "loss": 0.21494896411895753, "memory(GiB)": 70.5, "step": 55470, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.448823 }, { "epoch": 2.376719078017223, "grad_norm": 2.0047688484191895, "learning_rate": 5.3869112563546844e-05, "loss": 0.3292895078659058, "memory(GiB)": 70.5, "step": 55475, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.448822 }, { "epoch": 2.3769332933464717, "grad_norm": 5.893100261688232, "learning_rate": 5.386240293471161e-05, "loss": 0.31114678382873534, "memory(GiB)": 70.5, "step": 55480, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.44883 }, { "epoch": 2.377147508675721, "grad_norm": 2.1546850204467773, "learning_rate": 5.385569323590531e-05, "loss": 0.3630533695220947, "memory(GiB)": 70.5, "step": 55485, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.44883 }, { "epoch": 2.37736172400497, "grad_norm": 2.104146957397461, "learning_rate": 5.3848983467249516e-05, "loss": 0.416066312789917, "memory(GiB)": 70.5, "step": 55490, "token_acc": 0.9159663865546218, "train_speed(iter/s)": 1.44884 }, { "epoch": 2.3775759393342186, "grad_norm": 3.73123836517334, "learning_rate": 5.384227362886578e-05, "loss": 0.21921019554138182, "memory(GiB)": 70.5, "step": 55495, "token_acc": 0.9528619528619529, "train_speed(iter/s)": 1.44884 }, { "epoch": 2.377790154663468, "grad_norm": 1.2561765909194946, "learning_rate": 5.3835563720875645e-05, "loss": 0.26819257736206054, "memory(GiB)": 70.5, "step": 55500, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.448836 }, { "epoch": 2.377790154663468, "eval_loss": 2.4212427139282227, "eval_runtime": 14.1477, "eval_samples_per_second": 7.068, "eval_steps_per_second": 7.068, "eval_token_acc": 0.4216710182767624, "step": 55500 }, { "epoch": 2.3780043699927167, "grad_norm": 5.512554168701172, "learning_rate": 5.382885374340069e-05, "loss": 0.42247943878173827, "memory(GiB)": 70.5, "step": 55505, "token_acc": 0.5560836501901141, "train_speed(iter/s)": 1.44826 }, { "epoch": 2.3782185853219655, "grad_norm": 3.647158622741699, "learning_rate": 5.382214369656244e-05, "loss": 0.6329929828643799, "memory(GiB)": 70.5, "step": 55510, "token_acc": 0.8757396449704142, "train_speed(iter/s)": 1.448265 }, { "epoch": 2.3784328006512148, "grad_norm": 4.118147850036621, "learning_rate": 5.381543358048249e-05, "loss": 0.35364508628845215, "memory(GiB)": 70.5, "step": 55515, "token_acc": 0.9308943089430894, "train_speed(iter/s)": 1.448272 }, { "epoch": 2.3786470159804636, "grad_norm": 2.5067920684814453, "learning_rate": 5.380872339528237e-05, "loss": 0.2194195032119751, "memory(GiB)": 70.5, "step": 55520, "token_acc": 0.9506172839506173, "train_speed(iter/s)": 1.448271 }, { "epoch": 2.3788612313097124, "grad_norm": 1.0267360210418701, "learning_rate": 5.380201314108365e-05, "loss": 0.22396504878997803, "memory(GiB)": 70.5, "step": 55525, "token_acc": 0.9463414634146341, "train_speed(iter/s)": 1.44828 }, { "epoch": 2.3790754466389616, "grad_norm": 2.5550026893615723, "learning_rate": 5.3795302818007895e-05, "loss": 0.4167480945587158, "memory(GiB)": 70.5, "step": 55530, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.448276 }, { "epoch": 2.3792896619682105, "grad_norm": 5.130678176879883, "learning_rate": 5.378859242617668e-05, "loss": 0.5844497680664062, "memory(GiB)": 70.5, "step": 55535, "token_acc": 0.8485915492957746, "train_speed(iter/s)": 1.448278 }, { "epoch": 2.3795038772974593, "grad_norm": 3.609571695327759, "learning_rate": 5.378188196571154e-05, "loss": 0.5490140914916992, "memory(GiB)": 70.5, "step": 55540, "token_acc": 0.8937728937728938, "train_speed(iter/s)": 1.448275 }, { "epoch": 2.3797180926267085, "grad_norm": 2.935757637023926, "learning_rate": 5.3775171436734084e-05, "loss": 0.32542009353637696, "memory(GiB)": 70.5, "step": 55545, "token_acc": 0.9372384937238494, "train_speed(iter/s)": 1.448278 }, { "epoch": 2.3799323079559573, "grad_norm": 3.4762632846832275, "learning_rate": 5.376846083936585e-05, "loss": 0.4537391662597656, "memory(GiB)": 70.5, "step": 55550, "token_acc": 0.903114186851211, "train_speed(iter/s)": 1.448281 }, { "epoch": 2.380146523285206, "grad_norm": 0.7526494860649109, "learning_rate": 5.376175017372841e-05, "loss": 0.09843238592147827, "memory(GiB)": 70.5, "step": 55555, "token_acc": 0.9602649006622517, "train_speed(iter/s)": 1.448298 }, { "epoch": 2.3803607386144554, "grad_norm": 1.3980340957641602, "learning_rate": 5.375503943994333e-05, "loss": 0.147769558429718, "memory(GiB)": 70.5, "step": 55560, "token_acc": 0.9657794676806084, "train_speed(iter/s)": 1.448299 }, { "epoch": 2.380574953943704, "grad_norm": 2.996971845626831, "learning_rate": 5.37483286381322e-05, "loss": 0.1717039465904236, "memory(GiB)": 70.5, "step": 55565, "token_acc": 0.9527896995708155, "train_speed(iter/s)": 1.448308 }, { "epoch": 2.380789169272953, "grad_norm": 2.8016679286956787, "learning_rate": 5.374161776841656e-05, "loss": 0.3512481689453125, "memory(GiB)": 70.5, "step": 55570, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.44831 }, { "epoch": 2.3810033846022023, "grad_norm": 7.469133377075195, "learning_rate": 5.373490683091802e-05, "loss": 0.35848798751831057, "memory(GiB)": 70.5, "step": 55575, "token_acc": 0.9301587301587302, "train_speed(iter/s)": 1.448307 }, { "epoch": 2.381217599931451, "grad_norm": 4.1815876960754395, "learning_rate": 5.372819582575814e-05, "loss": 0.4287982940673828, "memory(GiB)": 70.5, "step": 55580, "token_acc": 0.9176029962546817, "train_speed(iter/s)": 1.448304 }, { "epoch": 2.3814318152607, "grad_norm": 2.185800313949585, "learning_rate": 5.372148475305849e-05, "loss": 0.30771048069000245, "memory(GiB)": 70.5, "step": 55585, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.448305 }, { "epoch": 2.381646030589949, "grad_norm": 3.0318002700805664, "learning_rate": 5.371477361294066e-05, "loss": 0.499638032913208, "memory(GiB)": 70.5, "step": 55590, "token_acc": 0.9014084507042254, "train_speed(iter/s)": 1.448305 }, { "epoch": 2.381860245919198, "grad_norm": 2.0863378047943115, "learning_rate": 5.370806240552623e-05, "loss": 0.22649540901184081, "memory(GiB)": 70.5, "step": 55595, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.448314 }, { "epoch": 2.382074461248447, "grad_norm": 4.749948024749756, "learning_rate": 5.370135113093674e-05, "loss": 0.4511530876159668, "memory(GiB)": 70.5, "step": 55600, "token_acc": 0.8953488372093024, "train_speed(iter/s)": 1.448315 }, { "epoch": 2.382288676577696, "grad_norm": 2.191026210784912, "learning_rate": 5.369463978929382e-05, "loss": 0.3373162269592285, "memory(GiB)": 70.5, "step": 55605, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.448325 }, { "epoch": 2.382502891906945, "grad_norm": 4.848132610321045, "learning_rate": 5.3687928380719044e-05, "loss": 0.47725396156311034, "memory(GiB)": 70.5, "step": 55610, "token_acc": 0.9036544850498339, "train_speed(iter/s)": 1.448329 }, { "epoch": 2.3827171072361937, "grad_norm": 3.0771231651306152, "learning_rate": 5.368121690533396e-05, "loss": 0.30999300479888914, "memory(GiB)": 70.5, "step": 55615, "token_acc": 0.9378698224852071, "train_speed(iter/s)": 1.448334 }, { "epoch": 2.382931322565443, "grad_norm": 1.9778262376785278, "learning_rate": 5.3674505363260206e-05, "loss": 0.3222850561141968, "memory(GiB)": 70.5, "step": 55620, "token_acc": 0.9311740890688259, "train_speed(iter/s)": 1.448342 }, { "epoch": 2.3831455378946917, "grad_norm": 4.06095552444458, "learning_rate": 5.366779375461933e-05, "loss": 0.2530482769012451, "memory(GiB)": 70.5, "step": 55625, "token_acc": 0.9477124183006536, "train_speed(iter/s)": 1.448343 }, { "epoch": 2.3833597532239406, "grad_norm": 5.949102878570557, "learning_rate": 5.366108207953293e-05, "loss": 0.4944140911102295, "memory(GiB)": 70.5, "step": 55630, "token_acc": 0.8966666666666666, "train_speed(iter/s)": 1.448361 }, { "epoch": 2.38357396855319, "grad_norm": 7.1750359535217285, "learning_rate": 5.365437033812259e-05, "loss": 0.588493824005127, "memory(GiB)": 70.5, "step": 55635, "token_acc": 0.8907849829351536, "train_speed(iter/s)": 1.448367 }, { "epoch": 2.3837881838824386, "grad_norm": 3.303053617477417, "learning_rate": 5.3647658530509904e-05, "loss": 0.2984930515289307, "memory(GiB)": 70.5, "step": 55640, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.448372 }, { "epoch": 2.3840023992116874, "grad_norm": 3.807006597518921, "learning_rate": 5.364094665681646e-05, "loss": 0.3487492561340332, "memory(GiB)": 70.5, "step": 55645, "token_acc": 0.8991097922848664, "train_speed(iter/s)": 1.448376 }, { "epoch": 2.3842166145409367, "grad_norm": 2.844930410385132, "learning_rate": 5.363423471716386e-05, "loss": 0.4701141357421875, "memory(GiB)": 70.5, "step": 55650, "token_acc": 0.9035714285714286, "train_speed(iter/s)": 1.448392 }, { "epoch": 2.3844308298701855, "grad_norm": 0.42850908637046814, "learning_rate": 5.3627522711673675e-05, "loss": 0.367397141456604, "memory(GiB)": 70.5, "step": 55655, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.44841 }, { "epoch": 2.3846450451994343, "grad_norm": 1.5564286708831787, "learning_rate": 5.362081064046753e-05, "loss": 0.4079939365386963, "memory(GiB)": 70.5, "step": 55660, "token_acc": 0.9034267912772586, "train_speed(iter/s)": 1.448424 }, { "epoch": 2.3848592605286836, "grad_norm": 0.32447004318237305, "learning_rate": 5.3614098503667e-05, "loss": 0.35881776809692384, "memory(GiB)": 70.5, "step": 55665, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.448427 }, { "epoch": 2.3850734758579324, "grad_norm": 1.2902460098266602, "learning_rate": 5.360738630139368e-05, "loss": 0.3912456750869751, "memory(GiB)": 70.5, "step": 55670, "token_acc": 0.9155844155844156, "train_speed(iter/s)": 1.448429 }, { "epoch": 2.385287691187181, "grad_norm": 5.855057716369629, "learning_rate": 5.360067403376916e-05, "loss": 0.7395027160644532, "memory(GiB)": 70.5, "step": 55675, "token_acc": 0.853035143769968, "train_speed(iter/s)": 1.448437 }, { "epoch": 2.3855019065164305, "grad_norm": 0.28911226987838745, "learning_rate": 5.359396170091508e-05, "loss": 0.18101595640182494, "memory(GiB)": 70.5, "step": 55680, "token_acc": 0.959731543624161, "train_speed(iter/s)": 1.448433 }, { "epoch": 2.3857161218456793, "grad_norm": 4.37897253036499, "learning_rate": 5.358724930295299e-05, "loss": 0.3475008010864258, "memory(GiB)": 70.5, "step": 55685, "token_acc": 0.9191176470588235, "train_speed(iter/s)": 1.448432 }, { "epoch": 2.385930337174928, "grad_norm": 4.733485698699951, "learning_rate": 5.3580536840004524e-05, "loss": 0.25060176849365234, "memory(GiB)": 70.5, "step": 55690, "token_acc": 0.91796875, "train_speed(iter/s)": 1.44845 }, { "epoch": 2.3861445525041773, "grad_norm": 3.310045003890991, "learning_rate": 5.3573824312191276e-05, "loss": 0.5855710983276368, "memory(GiB)": 70.5, "step": 55695, "token_acc": 0.875, "train_speed(iter/s)": 1.448477 }, { "epoch": 2.386358767833426, "grad_norm": 2.732161283493042, "learning_rate": 5.356711171963484e-05, "loss": 0.2628235101699829, "memory(GiB)": 70.5, "step": 55700, "token_acc": 0.9366515837104072, "train_speed(iter/s)": 1.448498 }, { "epoch": 2.386572983162675, "grad_norm": 2.4397850036621094, "learning_rate": 5.356039906245684e-05, "loss": 0.28980958461761475, "memory(GiB)": 70.5, "step": 55705, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.448512 }, { "epoch": 2.3867871984919242, "grad_norm": 2.289607524871826, "learning_rate": 5.3553686340778875e-05, "loss": 0.5321062564849853, "memory(GiB)": 70.5, "step": 55710, "token_acc": 0.8922305764411027, "train_speed(iter/s)": 1.448517 }, { "epoch": 2.387001413821173, "grad_norm": 3.5839524269104004, "learning_rate": 5.354697355472252e-05, "loss": 0.33571295738220214, "memory(GiB)": 70.5, "step": 55715, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.448516 }, { "epoch": 2.387215629150422, "grad_norm": 3.8604419231414795, "learning_rate": 5.354026070440944e-05, "loss": 0.5751075744628906, "memory(GiB)": 70.5, "step": 55720, "token_acc": 0.8795180722891566, "train_speed(iter/s)": 1.448522 }, { "epoch": 2.387429844479671, "grad_norm": 0.1341785490512848, "learning_rate": 5.35335477899612e-05, "loss": 0.25174143314361574, "memory(GiB)": 70.5, "step": 55725, "token_acc": 0.9563758389261745, "train_speed(iter/s)": 1.448543 }, { "epoch": 2.38764405980892, "grad_norm": 0.5768601298332214, "learning_rate": 5.352683481149944e-05, "loss": 0.45335373878479, "memory(GiB)": 70.5, "step": 55730, "token_acc": 0.916030534351145, "train_speed(iter/s)": 1.448561 }, { "epoch": 2.3878582751381687, "grad_norm": 4.497368335723877, "learning_rate": 5.352012176914575e-05, "loss": 0.3115269899368286, "memory(GiB)": 70.5, "step": 55735, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.448561 }, { "epoch": 2.388072490467418, "grad_norm": 1.972151756286621, "learning_rate": 5.351340866302176e-05, "loss": 0.2950622797012329, "memory(GiB)": 70.5, "step": 55740, "token_acc": 0.9416666666666667, "train_speed(iter/s)": 1.448567 }, { "epoch": 2.388286705796667, "grad_norm": 2.273754119873047, "learning_rate": 5.350669549324907e-05, "loss": 0.5217100620269776, "memory(GiB)": 70.5, "step": 55745, "token_acc": 0.8733766233766234, "train_speed(iter/s)": 1.448567 }, { "epoch": 2.3885009211259156, "grad_norm": 3.560554027557373, "learning_rate": 5.3499982259949285e-05, "loss": 0.49999136924743653, "memory(GiB)": 70.5, "step": 55750, "token_acc": 0.8924050632911392, "train_speed(iter/s)": 1.448567 }, { "epoch": 2.388715136455165, "grad_norm": 5.235320568084717, "learning_rate": 5.349326896324407e-05, "loss": 0.47374372482299804, "memory(GiB)": 70.5, "step": 55755, "token_acc": 0.9021739130434783, "train_speed(iter/s)": 1.448567 }, { "epoch": 2.3889293517844137, "grad_norm": 1.4857321977615356, "learning_rate": 5.348655560325498e-05, "loss": 0.45328426361083984, "memory(GiB)": 70.5, "step": 55760, "token_acc": 0.88, "train_speed(iter/s)": 1.448565 }, { "epoch": 2.3891435671136625, "grad_norm": 2.591339111328125, "learning_rate": 5.347984218010369e-05, "loss": 0.4586019515991211, "memory(GiB)": 70.5, "step": 55765, "token_acc": 0.8877887788778878, "train_speed(iter/s)": 1.448559 }, { "epoch": 2.3893577824429117, "grad_norm": 3.4755699634552, "learning_rate": 5.347312869391179e-05, "loss": 0.5842247486114502, "memory(GiB)": 70.5, "step": 55770, "token_acc": 0.8745874587458746, "train_speed(iter/s)": 1.448555 }, { "epoch": 2.3895719977721606, "grad_norm": 0.6314675211906433, "learning_rate": 5.3466415144800893e-05, "loss": 0.21421704292297364, "memory(GiB)": 70.5, "step": 55775, "token_acc": 0.953125, "train_speed(iter/s)": 1.448558 }, { "epoch": 2.3897862131014094, "grad_norm": 1.1862355470657349, "learning_rate": 5.345970153289266e-05, "loss": 0.14773668050765992, "memory(GiB)": 70.5, "step": 55780, "token_acc": 0.953307392996109, "train_speed(iter/s)": 1.448576 }, { "epoch": 2.3900004284306586, "grad_norm": 2.85581111907959, "learning_rate": 5.345298785830866e-05, "loss": 0.31883835792541504, "memory(GiB)": 70.5, "step": 55785, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.448597 }, { "epoch": 2.3902146437599074, "grad_norm": 0.869495153427124, "learning_rate": 5.344627412117055e-05, "loss": 0.15649499893188476, "memory(GiB)": 70.5, "step": 55790, "token_acc": 0.9610894941634242, "train_speed(iter/s)": 1.448612 }, { "epoch": 2.3904288590891563, "grad_norm": 6.974704742431641, "learning_rate": 5.343956032159996e-05, "loss": 0.697969627380371, "memory(GiB)": 70.5, "step": 55795, "token_acc": 0.8407643312101911, "train_speed(iter/s)": 1.448612 }, { "epoch": 2.3906430744184055, "grad_norm": 2.0980236530303955, "learning_rate": 5.34328464597185e-05, "loss": 0.3625503063201904, "memory(GiB)": 70.5, "step": 55800, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.448627 }, { "epoch": 2.3908572897476543, "grad_norm": 1.3302059173583984, "learning_rate": 5.342613253564782e-05, "loss": 0.19783610105514526, "memory(GiB)": 70.5, "step": 55805, "token_acc": 0.962059620596206, "train_speed(iter/s)": 1.448629 }, { "epoch": 2.391071505076903, "grad_norm": 2.7161865234375, "learning_rate": 5.341941854950952e-05, "loss": 0.33534190654754636, "memory(GiB)": 70.5, "step": 55810, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.44863 }, { "epoch": 2.3912857204061524, "grad_norm": 1.1375740766525269, "learning_rate": 5.341270450142526e-05, "loss": 0.18480079174041747, "memory(GiB)": 70.5, "step": 55815, "token_acc": 0.954225352112676, "train_speed(iter/s)": 1.448632 }, { "epoch": 2.391499935735401, "grad_norm": 3.300143003463745, "learning_rate": 5.340599039151665e-05, "loss": 0.4699354648590088, "memory(GiB)": 70.5, "step": 55820, "token_acc": 0.9016949152542373, "train_speed(iter/s)": 1.448625 }, { "epoch": 2.39171415106465, "grad_norm": 0.4782333970069885, "learning_rate": 5.339927621990533e-05, "loss": 0.40518627166748045, "memory(GiB)": 70.5, "step": 55825, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.448648 }, { "epoch": 2.3919283663938993, "grad_norm": 2.1271920204162598, "learning_rate": 5.3392561986712916e-05, "loss": 0.3106904268264771, "memory(GiB)": 70.5, "step": 55830, "token_acc": 0.9304635761589404, "train_speed(iter/s)": 1.448668 }, { "epoch": 2.392142581723148, "grad_norm": 4.240818023681641, "learning_rate": 5.3385847692061075e-05, "loss": 0.3385958671569824, "memory(GiB)": 70.5, "step": 55835, "token_acc": 0.924812030075188, "train_speed(iter/s)": 1.44867 }, { "epoch": 2.392356797052397, "grad_norm": 5.698493957519531, "learning_rate": 5.337913333607143e-05, "loss": 0.3661458969116211, "memory(GiB)": 70.5, "step": 55840, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.448677 }, { "epoch": 2.392571012381646, "grad_norm": 5.813741683959961, "learning_rate": 5.337241891886561e-05, "loss": 0.3804053544998169, "memory(GiB)": 70.5, "step": 55845, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.448693 }, { "epoch": 2.392785227710895, "grad_norm": 4.1236114501953125, "learning_rate": 5.3365704440565255e-05, "loss": 0.4583852767944336, "memory(GiB)": 70.5, "step": 55850, "token_acc": 0.8783783783783784, "train_speed(iter/s)": 1.448704 }, { "epoch": 2.392999443040144, "grad_norm": 5.113155364990234, "learning_rate": 5.335898990129202e-05, "loss": 0.37427258491516113, "memory(GiB)": 70.5, "step": 55855, "token_acc": 0.9176954732510288, "train_speed(iter/s)": 1.44871 }, { "epoch": 2.393213658369393, "grad_norm": 10.983378410339355, "learning_rate": 5.335227530116751e-05, "loss": 0.2851008653640747, "memory(GiB)": 70.5, "step": 55860, "token_acc": 0.9397590361445783, "train_speed(iter/s)": 1.448706 }, { "epoch": 2.393427873698642, "grad_norm": 5.31074333190918, "learning_rate": 5.3345560640313395e-05, "loss": 0.701351547241211, "memory(GiB)": 70.5, "step": 55865, "token_acc": 0.8863636363636364, "train_speed(iter/s)": 1.448721 }, { "epoch": 2.3936420890278907, "grad_norm": 1.6551504135131836, "learning_rate": 5.333884591885132e-05, "loss": 0.35407583713531493, "memory(GiB)": 70.5, "step": 55870, "token_acc": 0.9221183800623053, "train_speed(iter/s)": 1.448736 }, { "epoch": 2.39385630435714, "grad_norm": 3.0028772354125977, "learning_rate": 5.333213113690291e-05, "loss": 0.45178637504577634, "memory(GiB)": 70.5, "step": 55875, "token_acc": 0.9108910891089109, "train_speed(iter/s)": 1.448734 }, { "epoch": 2.3940705196863887, "grad_norm": 2.3111965656280518, "learning_rate": 5.3325416294589826e-05, "loss": 0.5815481185913086, "memory(GiB)": 70.5, "step": 55880, "token_acc": 0.8756476683937824, "train_speed(iter/s)": 1.448735 }, { "epoch": 2.3942847350156375, "grad_norm": 3.1413955688476562, "learning_rate": 5.331870139203371e-05, "loss": 0.43957977294921874, "memory(GiB)": 70.5, "step": 55885, "token_acc": 0.901060070671378, "train_speed(iter/s)": 1.448739 }, { "epoch": 2.394498950344887, "grad_norm": 2.6638739109039307, "learning_rate": 5.33119864293562e-05, "loss": 0.5322871685028077, "memory(GiB)": 70.5, "step": 55890, "token_acc": 0.8673740053050398, "train_speed(iter/s)": 1.44874 }, { "epoch": 2.3947131656741356, "grad_norm": 2.4004907608032227, "learning_rate": 5.3305271406678936e-05, "loss": 0.45093183517456054, "memory(GiB)": 70.5, "step": 55895, "token_acc": 0.8938906752411575, "train_speed(iter/s)": 1.448745 }, { "epoch": 2.3949273810033844, "grad_norm": 3.8414435386657715, "learning_rate": 5.329855632412359e-05, "loss": 0.5760745525360107, "memory(GiB)": 70.5, "step": 55900, "token_acc": 0.879746835443038, "train_speed(iter/s)": 1.448742 }, { "epoch": 2.3951415963326337, "grad_norm": 3.9126875400543213, "learning_rate": 5.329184118181181e-05, "loss": 0.5775054931640625, "memory(GiB)": 70.5, "step": 55905, "token_acc": 0.8790849673202614, "train_speed(iter/s)": 1.448747 }, { "epoch": 2.3953558116618825, "grad_norm": 2.8137943744659424, "learning_rate": 5.3285125979865234e-05, "loss": 0.5522202491760254, "memory(GiB)": 70.5, "step": 55910, "token_acc": 0.884, "train_speed(iter/s)": 1.448755 }, { "epoch": 2.3955700269911313, "grad_norm": 3.324411630630493, "learning_rate": 5.3278410718405514e-05, "loss": 0.36068997383117674, "memory(GiB)": 70.5, "step": 55915, "token_acc": 0.9227941176470589, "train_speed(iter/s)": 1.448758 }, { "epoch": 2.3957842423203806, "grad_norm": 4.997622013092041, "learning_rate": 5.327169539755431e-05, "loss": 0.35125293731689455, "memory(GiB)": 70.5, "step": 55920, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.448756 }, { "epoch": 2.3959984576496294, "grad_norm": 3.777920722961426, "learning_rate": 5.3264980017433284e-05, "loss": 0.3812370777130127, "memory(GiB)": 70.5, "step": 55925, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.448764 }, { "epoch": 2.396212672978878, "grad_norm": 5.856020927429199, "learning_rate": 5.325826457816406e-05, "loss": 0.20560939311981202, "memory(GiB)": 70.5, "step": 55930, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.448771 }, { "epoch": 2.3964268883081274, "grad_norm": 3.5261306762695312, "learning_rate": 5.325154907986833e-05, "loss": 0.20937891006469728, "memory(GiB)": 70.5, "step": 55935, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.448785 }, { "epoch": 2.3966411036373763, "grad_norm": 5.476274490356445, "learning_rate": 5.324483352266775e-05, "loss": 0.41582584381103516, "memory(GiB)": 70.5, "step": 55940, "token_acc": 0.9311740890688259, "train_speed(iter/s)": 1.448792 }, { "epoch": 2.396855318966625, "grad_norm": 2.771097183227539, "learning_rate": 5.323811790668395e-05, "loss": 0.32744393348693845, "memory(GiB)": 70.5, "step": 55945, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.448804 }, { "epoch": 2.3970695342958743, "grad_norm": 4.485565662384033, "learning_rate": 5.323140223203862e-05, "loss": 0.3510725498199463, "memory(GiB)": 70.5, "step": 55950, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.44882 }, { "epoch": 2.397283749625123, "grad_norm": 3.3701000213623047, "learning_rate": 5.322468649885341e-05, "loss": 0.3755953311920166, "memory(GiB)": 70.5, "step": 55955, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.448819 }, { "epoch": 2.397497964954372, "grad_norm": 5.458912372589111, "learning_rate": 5.3217970707249955e-05, "loss": 0.4480452060699463, "memory(GiB)": 70.5, "step": 55960, "token_acc": 0.9111969111969112, "train_speed(iter/s)": 1.44882 }, { "epoch": 2.397712180283621, "grad_norm": 4.1462249755859375, "learning_rate": 5.3211254857349955e-05, "loss": 0.5099764347076416, "memory(GiB)": 70.5, "step": 55965, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.44882 }, { "epoch": 2.39792639561287, "grad_norm": 3.484219789505005, "learning_rate": 5.320453894927506e-05, "loss": 0.46641716957092283, "memory(GiB)": 70.5, "step": 55970, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.448819 }, { "epoch": 2.398140610942119, "grad_norm": 1.9813456535339355, "learning_rate": 5.319782298314694e-05, "loss": 0.2504000663757324, "memory(GiB)": 70.5, "step": 55975, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.448819 }, { "epoch": 2.398354826271368, "grad_norm": 4.960549354553223, "learning_rate": 5.319110695908726e-05, "loss": 0.2602841377258301, "memory(GiB)": 70.5, "step": 55980, "token_acc": 0.9506578947368421, "train_speed(iter/s)": 1.448825 }, { "epoch": 2.398569041600617, "grad_norm": 5.302353858947754, "learning_rate": 5.318439087721768e-05, "loss": 0.2430504322052002, "memory(GiB)": 70.5, "step": 55985, "token_acc": 0.9519230769230769, "train_speed(iter/s)": 1.448835 }, { "epoch": 2.3987832569298657, "grad_norm": 2.214878797531128, "learning_rate": 5.3177674737659865e-05, "loss": 0.21371288299560548, "memory(GiB)": 70.5, "step": 55990, "token_acc": 0.9511400651465798, "train_speed(iter/s)": 1.448839 }, { "epoch": 2.398997472259115, "grad_norm": 5.35909366607666, "learning_rate": 5.31709585405355e-05, "loss": 0.1850544571876526, "memory(GiB)": 70.5, "step": 55995, "token_acc": 0.9656652360515021, "train_speed(iter/s)": 1.448842 }, { "epoch": 2.399211687588364, "grad_norm": 2.695124864578247, "learning_rate": 5.316424228596625e-05, "loss": 0.2646437644958496, "memory(GiB)": 70.5, "step": 56000, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.448838 }, { "epoch": 2.399211687588364, "eval_loss": 2.4846913814544678, "eval_runtime": 13.5839, "eval_samples_per_second": 7.362, "eval_steps_per_second": 7.362, "eval_token_acc": 0.44010767160161507, "step": 56000 }, { "epoch": 2.3994259029176126, "grad_norm": 1.9825208187103271, "learning_rate": 5.315752597407376e-05, "loss": 0.36124730110168457, "memory(GiB)": 70.5, "step": 56005, "token_acc": 0.5681818181818182, "train_speed(iter/s)": 1.448294 }, { "epoch": 2.399640118246862, "grad_norm": 3.8767292499542236, "learning_rate": 5.315080960497975e-05, "loss": 0.2163395881652832, "memory(GiB)": 70.5, "step": 56010, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.448295 }, { "epoch": 2.3998543335761107, "grad_norm": 2.0202856063842773, "learning_rate": 5.3144093178805856e-05, "loss": 0.19902645349502562, "memory(GiB)": 70.5, "step": 56015, "token_acc": 0.9549295774647887, "train_speed(iter/s)": 1.448296 }, { "epoch": 2.4000685489053595, "grad_norm": 3.1377642154693604, "learning_rate": 5.313737669567377e-05, "loss": 0.2871319055557251, "memory(GiB)": 70.5, "step": 56020, "token_acc": 0.9314641744548287, "train_speed(iter/s)": 1.448301 }, { "epoch": 2.4002827642346087, "grad_norm": 3.6116483211517334, "learning_rate": 5.313066015570516e-05, "loss": 0.2796665668487549, "memory(GiB)": 70.5, "step": 56025, "token_acc": 0.9318885448916409, "train_speed(iter/s)": 1.448296 }, { "epoch": 2.4004969795638575, "grad_norm": 5.88750696182251, "learning_rate": 5.312394355902171e-05, "loss": 0.3226002216339111, "memory(GiB)": 70.5, "step": 56030, "token_acc": 0.9424778761061947, "train_speed(iter/s)": 1.448315 }, { "epoch": 2.4007111948931064, "grad_norm": 5.102730751037598, "learning_rate": 5.3117226905745075e-05, "loss": 0.3361337661743164, "memory(GiB)": 70.5, "step": 56035, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.448316 }, { "epoch": 2.4009254102223556, "grad_norm": 5.2283549308776855, "learning_rate": 5.311051019599698e-05, "loss": 0.3100406646728516, "memory(GiB)": 70.5, "step": 56040, "token_acc": 0.946875, "train_speed(iter/s)": 1.448318 }, { "epoch": 2.4011396255516044, "grad_norm": 0.3946548104286194, "learning_rate": 5.310379342989904e-05, "loss": 0.15176568031311036, "memory(GiB)": 70.5, "step": 56045, "token_acc": 0.9735849056603774, "train_speed(iter/s)": 1.448325 }, { "epoch": 2.4013538408808532, "grad_norm": 5.001772403717041, "learning_rate": 5.3097076607572984e-05, "loss": 0.41063451766967773, "memory(GiB)": 70.5, "step": 56050, "token_acc": 0.911042944785276, "train_speed(iter/s)": 1.448322 }, { "epoch": 2.4015680562101025, "grad_norm": 1.2193005084991455, "learning_rate": 5.309035972914049e-05, "loss": 0.26503193378448486, "memory(GiB)": 70.5, "step": 56055, "token_acc": 0.9363636363636364, "train_speed(iter/s)": 1.448336 }, { "epoch": 2.4017822715393513, "grad_norm": 2.221234083175659, "learning_rate": 5.308364279472322e-05, "loss": 0.6642908096313477, "memory(GiB)": 70.5, "step": 56060, "token_acc": 0.8763636363636363, "train_speed(iter/s)": 1.448334 }, { "epoch": 2.4019964868686, "grad_norm": 3.0077977180480957, "learning_rate": 5.3076925804442865e-05, "loss": 0.5386641502380372, "memory(GiB)": 70.5, "step": 56065, "token_acc": 0.8937007874015748, "train_speed(iter/s)": 1.448353 }, { "epoch": 2.4022107021978494, "grad_norm": 2.921441078186035, "learning_rate": 5.307020875842114e-05, "loss": 0.21866602897644044, "memory(GiB)": 70.5, "step": 56070, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.448361 }, { "epoch": 2.402424917527098, "grad_norm": 3.4690520763397217, "learning_rate": 5.306349165677967e-05, "loss": 0.35247230529785156, "memory(GiB)": 70.5, "step": 56075, "token_acc": 0.9420731707317073, "train_speed(iter/s)": 1.448366 }, { "epoch": 2.402639132856347, "grad_norm": 2.436572790145874, "learning_rate": 5.3056774499640184e-05, "loss": 0.25343847274780273, "memory(GiB)": 70.5, "step": 56080, "token_acc": 0.9410029498525073, "train_speed(iter/s)": 1.448365 }, { "epoch": 2.4028533481855963, "grad_norm": 1.5096888542175293, "learning_rate": 5.305005728712437e-05, "loss": 0.4352680206298828, "memory(GiB)": 70.5, "step": 56085, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.448369 }, { "epoch": 2.403067563514845, "grad_norm": 3.2785604000091553, "learning_rate": 5.304334001935389e-05, "loss": 0.2611887216567993, "memory(GiB)": 70.5, "step": 56090, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.44837 }, { "epoch": 2.403281778844094, "grad_norm": 9.06693172454834, "learning_rate": 5.3036622696450466e-05, "loss": 0.4792226791381836, "memory(GiB)": 70.5, "step": 56095, "token_acc": 0.9157088122605364, "train_speed(iter/s)": 1.44839 }, { "epoch": 2.403495994173343, "grad_norm": 4.421489715576172, "learning_rate": 5.302990531853578e-05, "loss": 0.7595810890197754, "memory(GiB)": 70.5, "step": 56100, "token_acc": 0.8547854785478548, "train_speed(iter/s)": 1.44839 }, { "epoch": 2.403710209502592, "grad_norm": 0.09554526209831238, "learning_rate": 5.302318788573149e-05, "loss": 0.5346932411193848, "memory(GiB)": 70.5, "step": 56105, "token_acc": 0.8725868725868726, "train_speed(iter/s)": 1.448402 }, { "epoch": 2.4039244248318408, "grad_norm": 4.211864471435547, "learning_rate": 5.3016470398159344e-05, "loss": 0.478441858291626, "memory(GiB)": 70.5, "step": 56110, "token_acc": 0.9007633587786259, "train_speed(iter/s)": 1.448398 }, { "epoch": 2.40413864016109, "grad_norm": 4.15187406539917, "learning_rate": 5.3009752855941e-05, "loss": 0.36103134155273436, "memory(GiB)": 70.5, "step": 56115, "token_acc": 0.9375, "train_speed(iter/s)": 1.448407 }, { "epoch": 2.404352855490339, "grad_norm": 2.13680362701416, "learning_rate": 5.300303525919813e-05, "loss": 0.3550382614135742, "memory(GiB)": 70.5, "step": 56120, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.448409 }, { "epoch": 2.4045670708195876, "grad_norm": 3.14548397064209, "learning_rate": 5.2996317608052494e-05, "loss": 0.2856159687042236, "memory(GiB)": 70.5, "step": 56125, "token_acc": 0.932806324110672, "train_speed(iter/s)": 1.448406 }, { "epoch": 2.404781286148837, "grad_norm": 0.9530471563339233, "learning_rate": 5.298959990262574e-05, "loss": 0.37140443325042727, "memory(GiB)": 70.5, "step": 56130, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.448408 }, { "epoch": 2.4049955014780857, "grad_norm": 4.441066265106201, "learning_rate": 5.298288214303958e-05, "loss": 0.2979682445526123, "memory(GiB)": 70.5, "step": 56135, "token_acc": 0.9488054607508533, "train_speed(iter/s)": 1.44842 }, { "epoch": 2.4052097168073345, "grad_norm": 0.34450763463974, "learning_rate": 5.2976164329415725e-05, "loss": 0.1878295660018921, "memory(GiB)": 70.5, "step": 56140, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.448424 }, { "epoch": 2.405423932136584, "grad_norm": 5.372649192810059, "learning_rate": 5.296944646187585e-05, "loss": 0.7006539344787598, "memory(GiB)": 70.5, "step": 56145, "token_acc": 0.8344827586206897, "train_speed(iter/s)": 1.448438 }, { "epoch": 2.4056381474658326, "grad_norm": 3.1320276260375977, "learning_rate": 5.296272854054166e-05, "loss": 0.24150094985961915, "memory(GiB)": 70.5, "step": 56150, "token_acc": 0.9385964912280702, "train_speed(iter/s)": 1.448442 }, { "epoch": 2.4058523627950814, "grad_norm": 2.7916314601898193, "learning_rate": 5.295601056553486e-05, "loss": 0.34076671600341796, "memory(GiB)": 70.5, "step": 56155, "token_acc": 0.9157088122605364, "train_speed(iter/s)": 1.448447 }, { "epoch": 2.4060665781243307, "grad_norm": 2.053184986114502, "learning_rate": 5.294929253697718e-05, "loss": 0.28090245723724366, "memory(GiB)": 70.5, "step": 56160, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.448451 }, { "epoch": 2.4062807934535795, "grad_norm": 2.3889803886413574, "learning_rate": 5.294257445499028e-05, "loss": 0.2388866901397705, "memory(GiB)": 70.5, "step": 56165, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.448467 }, { "epoch": 2.4064950087828283, "grad_norm": 4.341933727264404, "learning_rate": 5.29358563196959e-05, "loss": 0.31250050067901614, "memory(GiB)": 70.5, "step": 56170, "token_acc": 0.915625, "train_speed(iter/s)": 1.448466 }, { "epoch": 2.4067092241120775, "grad_norm": 0.20384594798088074, "learning_rate": 5.2929138131215715e-05, "loss": 0.2519153356552124, "memory(GiB)": 70.5, "step": 56175, "token_acc": 0.9326599326599326, "train_speed(iter/s)": 1.448463 }, { "epoch": 2.4069234394413264, "grad_norm": 4.9610185623168945, "learning_rate": 5.292241988967144e-05, "loss": 0.37999348640441893, "memory(GiB)": 70.5, "step": 56180, "token_acc": 0.9186046511627907, "train_speed(iter/s)": 1.448464 }, { "epoch": 2.407137654770575, "grad_norm": 1.8696271181106567, "learning_rate": 5.29157015951848e-05, "loss": 0.21249282360076904, "memory(GiB)": 70.5, "step": 56185, "token_acc": 0.9467455621301775, "train_speed(iter/s)": 1.448474 }, { "epoch": 2.4073518700998244, "grad_norm": 3.75742244720459, "learning_rate": 5.2908983247877475e-05, "loss": 0.3464756965637207, "memory(GiB)": 70.5, "step": 56190, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.44848 }, { "epoch": 2.4075660854290732, "grad_norm": 3.571605682373047, "learning_rate": 5.2902264847871196e-05, "loss": 0.5375184535980224, "memory(GiB)": 70.5, "step": 56195, "token_acc": 0.8907563025210085, "train_speed(iter/s)": 1.448486 }, { "epoch": 2.407780300758322, "grad_norm": 7.627288341522217, "learning_rate": 5.289554639528768e-05, "loss": 0.5123649597167969, "memory(GiB)": 70.5, "step": 56200, "token_acc": 0.9057750759878419, "train_speed(iter/s)": 1.448487 }, { "epoch": 2.4079945160875713, "grad_norm": 2.9474592208862305, "learning_rate": 5.2888827890248604e-05, "loss": 0.2936037540435791, "memory(GiB)": 70.5, "step": 56205, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.448489 }, { "epoch": 2.40820873141682, "grad_norm": 4.59821081161499, "learning_rate": 5.288210933287572e-05, "loss": 0.5991519451141357, "memory(GiB)": 70.5, "step": 56210, "token_acc": 0.8765822784810127, "train_speed(iter/s)": 1.448503 }, { "epoch": 2.408422946746069, "grad_norm": 2.6456167697906494, "learning_rate": 5.287539072329072e-05, "loss": 0.361010217666626, "memory(GiB)": 70.5, "step": 56215, "token_acc": 0.9190031152647975, "train_speed(iter/s)": 1.448514 }, { "epoch": 2.408637162075318, "grad_norm": 2.1825172901153564, "learning_rate": 5.286867206161531e-05, "loss": 0.2567943334579468, "memory(GiB)": 70.5, "step": 56220, "token_acc": 0.9399293286219081, "train_speed(iter/s)": 1.448522 }, { "epoch": 2.408851377404567, "grad_norm": 0.1122029647231102, "learning_rate": 5.286195334797122e-05, "loss": 0.3278908967971802, "memory(GiB)": 70.5, "step": 56225, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.448521 }, { "epoch": 2.409065592733816, "grad_norm": 7.962311744689941, "learning_rate": 5.285523458248015e-05, "loss": 0.25781261920928955, "memory(GiB)": 70.5, "step": 56230, "token_acc": 0.9539007092198581, "train_speed(iter/s)": 1.448538 }, { "epoch": 2.409279808063065, "grad_norm": 1.3966387510299683, "learning_rate": 5.284851576526383e-05, "loss": 0.37217121124267577, "memory(GiB)": 70.5, "step": 56235, "token_acc": 0.9044368600682594, "train_speed(iter/s)": 1.448544 }, { "epoch": 2.409494023392314, "grad_norm": 3.173149824142456, "learning_rate": 5.2841796896443986e-05, "loss": 0.19784202575683593, "memory(GiB)": 70.5, "step": 56240, "token_acc": 0.9645390070921985, "train_speed(iter/s)": 1.448548 }, { "epoch": 2.4097082387215627, "grad_norm": 4.4379191398620605, "learning_rate": 5.283507797614232e-05, "loss": 0.7588286399841309, "memory(GiB)": 70.5, "step": 56245, "token_acc": 0.8452830188679246, "train_speed(iter/s)": 1.448553 }, { "epoch": 2.409922454050812, "grad_norm": 0.17796368896961212, "learning_rate": 5.282835900448055e-05, "loss": 0.4204233646392822, "memory(GiB)": 70.5, "step": 56250, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.448554 }, { "epoch": 2.4101366693800608, "grad_norm": 2.2227656841278076, "learning_rate": 5.282163998158042e-05, "loss": 0.38926622867584226, "memory(GiB)": 70.5, "step": 56255, "token_acc": 0.9003831417624522, "train_speed(iter/s)": 1.448566 }, { "epoch": 2.4103508847093096, "grad_norm": 3.9824512004852295, "learning_rate": 5.281492090756364e-05, "loss": 0.23527615070343016, "memory(GiB)": 70.5, "step": 56260, "token_acc": 0.9511278195488722, "train_speed(iter/s)": 1.448573 }, { "epoch": 2.410565100038559, "grad_norm": 2.8920414447784424, "learning_rate": 5.280820178255188e-05, "loss": 0.6403561592102051, "memory(GiB)": 70.5, "step": 56265, "token_acc": 0.8565891472868217, "train_speed(iter/s)": 1.448597 }, { "epoch": 2.4107793153678077, "grad_norm": 3.120342493057251, "learning_rate": 5.2801482606666955e-05, "loss": 0.2758676052093506, "memory(GiB)": 70.5, "step": 56270, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.448603 }, { "epoch": 2.4109935306970565, "grad_norm": 4.590897560119629, "learning_rate": 5.279476338003053e-05, "loss": 0.5466650009155274, "memory(GiB)": 70.5, "step": 56275, "token_acc": 0.8691588785046729, "train_speed(iter/s)": 1.448604 }, { "epoch": 2.4112077460263057, "grad_norm": 1.8059934377670288, "learning_rate": 5.2788044102764345e-05, "loss": 0.29713680744171145, "memory(GiB)": 70.5, "step": 56280, "token_acc": 0.9420849420849421, "train_speed(iter/s)": 1.448603 }, { "epoch": 2.4114219613555545, "grad_norm": 0.12223837524652481, "learning_rate": 5.278132477499014e-05, "loss": 0.43532423973083495, "memory(GiB)": 70.5, "step": 56285, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.448607 }, { "epoch": 2.4116361766848033, "grad_norm": 3.659809112548828, "learning_rate": 5.277460539682961e-05, "loss": 0.4632091522216797, "memory(GiB)": 70.5, "step": 56290, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.448607 }, { "epoch": 2.4118503920140526, "grad_norm": 3.0942158699035645, "learning_rate": 5.276788596840451e-05, "loss": 0.456990909576416, "memory(GiB)": 70.5, "step": 56295, "token_acc": 0.9164086687306502, "train_speed(iter/s)": 1.448608 }, { "epoch": 2.4120646073433014, "grad_norm": 0.8891383409500122, "learning_rate": 5.276116648983656e-05, "loss": 0.22536242008209229, "memory(GiB)": 70.5, "step": 56300, "token_acc": 0.9540636042402827, "train_speed(iter/s)": 1.448612 }, { "epoch": 2.4122788226725502, "grad_norm": 2.8700602054595947, "learning_rate": 5.275444696124747e-05, "loss": 0.4959463119506836, "memory(GiB)": 70.5, "step": 56305, "token_acc": 0.8972602739726028, "train_speed(iter/s)": 1.448606 }, { "epoch": 2.4124930380017995, "grad_norm": 6.808250427246094, "learning_rate": 5.2747727382758993e-05, "loss": 0.34762349128723147, "memory(GiB)": 70.5, "step": 56310, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.448603 }, { "epoch": 2.4127072533310483, "grad_norm": 3.783656597137451, "learning_rate": 5.274100775449288e-05, "loss": 0.2845951557159424, "memory(GiB)": 70.5, "step": 56315, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.448604 }, { "epoch": 2.412921468660297, "grad_norm": 2.5037524700164795, "learning_rate": 5.2734288076570824e-05, "loss": 0.3973504066467285, "memory(GiB)": 70.5, "step": 56320, "token_acc": 0.9034749034749034, "train_speed(iter/s)": 1.448607 }, { "epoch": 2.4131356839895464, "grad_norm": 4.725710391998291, "learning_rate": 5.2727568349114576e-05, "loss": 0.5066546440124512, "memory(GiB)": 70.5, "step": 56325, "token_acc": 0.9021406727828746, "train_speed(iter/s)": 1.448611 }, { "epoch": 2.413349899318795, "grad_norm": 2.2745556831359863, "learning_rate": 5.2720848572245874e-05, "loss": 0.4830489635467529, "memory(GiB)": 70.5, "step": 56330, "token_acc": 0.9079754601226994, "train_speed(iter/s)": 1.448625 }, { "epoch": 2.413564114648044, "grad_norm": 2.526219129562378, "learning_rate": 5.271412874608644e-05, "loss": 0.2268143653869629, "memory(GiB)": 70.5, "step": 56335, "token_acc": 0.9541984732824428, "train_speed(iter/s)": 1.448616 }, { "epoch": 2.4137783299772932, "grad_norm": 3.7044289112091064, "learning_rate": 5.270740887075801e-05, "loss": 0.5305689334869385, "memory(GiB)": 70.5, "step": 56340, "token_acc": 0.9203821656050956, "train_speed(iter/s)": 1.448619 }, { "epoch": 2.413992545306542, "grad_norm": 1.7681632041931152, "learning_rate": 5.2700688946382345e-05, "loss": 0.4749757766723633, "memory(GiB)": 70.5, "step": 56345, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.448629 }, { "epoch": 2.414206760635791, "grad_norm": 3.7864954471588135, "learning_rate": 5.269396897308117e-05, "loss": 0.3132654905319214, "memory(GiB)": 70.5, "step": 56350, "token_acc": 0.9097744360902256, "train_speed(iter/s)": 1.448633 }, { "epoch": 2.41442097596504, "grad_norm": 1.4219261407852173, "learning_rate": 5.2687248950976206e-05, "loss": 0.3461721420288086, "memory(GiB)": 70.5, "step": 56355, "token_acc": 0.9455128205128205, "train_speed(iter/s)": 1.448632 }, { "epoch": 2.414635191294289, "grad_norm": 3.898545026779175, "learning_rate": 5.268052888018922e-05, "loss": 0.5115819931030273, "memory(GiB)": 70.5, "step": 56360, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.44865 }, { "epoch": 2.4148494066235378, "grad_norm": 1.1546623706817627, "learning_rate": 5.267380876084192e-05, "loss": 0.25722646713256836, "memory(GiB)": 70.5, "step": 56365, "token_acc": 0.9572649572649573, "train_speed(iter/s)": 1.448657 }, { "epoch": 2.415063621952787, "grad_norm": 1.8056498765945435, "learning_rate": 5.266708859305608e-05, "loss": 0.27641661167144777, "memory(GiB)": 70.5, "step": 56370, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.448659 }, { "epoch": 2.415277837282036, "grad_norm": 4.183955669403076, "learning_rate": 5.266036837695344e-05, "loss": 0.35744481086730956, "memory(GiB)": 70.5, "step": 56375, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.448662 }, { "epoch": 2.4154920526112846, "grad_norm": 5.848021507263184, "learning_rate": 5.26536481126557e-05, "loss": 0.6974071502685547, "memory(GiB)": 70.5, "step": 56380, "token_acc": 0.8594771241830066, "train_speed(iter/s)": 1.44866 }, { "epoch": 2.415706267940534, "grad_norm": 1.9702016115188599, "learning_rate": 5.264692780028465e-05, "loss": 0.24514265060424806, "memory(GiB)": 70.5, "step": 56385, "token_acc": 0.950381679389313, "train_speed(iter/s)": 1.448664 }, { "epoch": 2.4159204832697827, "grad_norm": 2.199561834335327, "learning_rate": 5.264020743996203e-05, "loss": 0.5099337577819825, "memory(GiB)": 70.5, "step": 56390, "token_acc": 0.9084507042253521, "train_speed(iter/s)": 1.448656 }, { "epoch": 2.4161346985990315, "grad_norm": 3.2395362854003906, "learning_rate": 5.263348703180956e-05, "loss": 0.34684481620788576, "memory(GiB)": 70.5, "step": 56395, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.448658 }, { "epoch": 2.4163489139282808, "grad_norm": 3.472533702850342, "learning_rate": 5.262676657594901e-05, "loss": 0.44282803535461424, "memory(GiB)": 70.5, "step": 56400, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.448661 }, { "epoch": 2.4165631292575296, "grad_norm": 10.274219512939453, "learning_rate": 5.2620046072502124e-05, "loss": 0.388257098197937, "memory(GiB)": 70.5, "step": 56405, "token_acc": 0.9108527131782945, "train_speed(iter/s)": 1.448673 }, { "epoch": 2.4167773445867784, "grad_norm": 3.2739832401275635, "learning_rate": 5.2613325521590626e-05, "loss": 0.4377938747406006, "memory(GiB)": 70.5, "step": 56410, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.448678 }, { "epoch": 2.4169915599160277, "grad_norm": 3.1359803676605225, "learning_rate": 5.26066049233363e-05, "loss": 0.39580204486846926, "memory(GiB)": 70.5, "step": 56415, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.448678 }, { "epoch": 2.4172057752452765, "grad_norm": 5.284639835357666, "learning_rate": 5.259988427786088e-05, "loss": 0.26945109367370607, "memory(GiB)": 70.5, "step": 56420, "token_acc": 0.9528619528619529, "train_speed(iter/s)": 1.448681 }, { "epoch": 2.4174199905745253, "grad_norm": 4.51198673248291, "learning_rate": 5.259316358528611e-05, "loss": 0.6620871543884277, "memory(GiB)": 70.5, "step": 56425, "token_acc": 0.8783783783783784, "train_speed(iter/s)": 1.448695 }, { "epoch": 2.4176342059037745, "grad_norm": 0.3490581810474396, "learning_rate": 5.2586442845733765e-05, "loss": 0.4030364990234375, "memory(GiB)": 70.5, "step": 56430, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.448695 }, { "epoch": 2.4178484212330233, "grad_norm": 3.934793472290039, "learning_rate": 5.257972205932558e-05, "loss": 0.354177188873291, "memory(GiB)": 70.5, "step": 56435, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.448698 }, { "epoch": 2.418062636562272, "grad_norm": 4.408319473266602, "learning_rate": 5.257300122618329e-05, "loss": 0.3419528007507324, "memory(GiB)": 70.5, "step": 56440, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.448712 }, { "epoch": 2.4182768518915214, "grad_norm": 1.876868486404419, "learning_rate": 5.2566280346428687e-05, "loss": 0.2849609851837158, "memory(GiB)": 70.5, "step": 56445, "token_acc": 0.9251497005988024, "train_speed(iter/s)": 1.44871 }, { "epoch": 2.4184910672207702, "grad_norm": 2.8259975910186768, "learning_rate": 5.25595594201835e-05, "loss": 0.2023371934890747, "memory(GiB)": 70.5, "step": 56450, "token_acc": 0.9590443686006825, "train_speed(iter/s)": 1.448713 }, { "epoch": 2.418705282550019, "grad_norm": 3.452652931213379, "learning_rate": 5.2552838447569485e-05, "loss": 0.37628746032714844, "memory(GiB)": 70.5, "step": 56455, "token_acc": 0.9201277955271565, "train_speed(iter/s)": 1.448724 }, { "epoch": 2.4189194978792683, "grad_norm": 3.1604247093200684, "learning_rate": 5.254611742870842e-05, "loss": 0.39287748336791994, "memory(GiB)": 70.5, "step": 56460, "token_acc": 0.9154518950437318, "train_speed(iter/s)": 1.448737 }, { "epoch": 2.419133713208517, "grad_norm": 4.7367119789123535, "learning_rate": 5.253939636372206e-05, "loss": 0.6469476699829102, "memory(GiB)": 70.5, "step": 56465, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.448766 }, { "epoch": 2.419347928537766, "grad_norm": 2.2128183841705322, "learning_rate": 5.253267525273213e-05, "loss": 0.3725393295288086, "memory(GiB)": 70.5, "step": 56470, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.448761 }, { "epoch": 2.419562143867015, "grad_norm": 1.5742478370666504, "learning_rate": 5.252595409586042e-05, "loss": 0.352056622505188, "memory(GiB)": 70.5, "step": 56475, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.448768 }, { "epoch": 2.419776359196264, "grad_norm": 4.352084159851074, "learning_rate": 5.251923289322868e-05, "loss": 0.3967432975769043, "memory(GiB)": 70.5, "step": 56480, "token_acc": 0.92, "train_speed(iter/s)": 1.448768 }, { "epoch": 2.419990574525513, "grad_norm": 3.279132843017578, "learning_rate": 5.251251164495868e-05, "loss": 0.4569528579711914, "memory(GiB)": 70.5, "step": 56485, "token_acc": 0.9102564102564102, "train_speed(iter/s)": 1.448769 }, { "epoch": 2.420204789854762, "grad_norm": 6.558166027069092, "learning_rate": 5.250579035117217e-05, "loss": 0.48492913246154784, "memory(GiB)": 70.5, "step": 56490, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.448766 }, { "epoch": 2.420419005184011, "grad_norm": 3.9468138217926025, "learning_rate": 5.249906901199091e-05, "loss": 0.35587201118469236, "memory(GiB)": 70.5, "step": 56495, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.448765 }, { "epoch": 2.4206332205132597, "grad_norm": 5.696479797363281, "learning_rate": 5.2492347627536686e-05, "loss": 0.9570916175842286, "memory(GiB)": 70.5, "step": 56500, "token_acc": 0.8149606299212598, "train_speed(iter/s)": 1.448769 }, { "epoch": 2.4206332205132597, "eval_loss": 2.4762532711029053, "eval_runtime": 13.4545, "eval_samples_per_second": 7.432, "eval_steps_per_second": 7.432, "eval_token_acc": 0.4435483870967742, "step": 56500 }, { "epoch": 2.420847435842509, "grad_norm": 7.150343418121338, "learning_rate": 5.248562619793124e-05, "loss": 0.595881175994873, "memory(GiB)": 70.5, "step": 56505, "token_acc": 0.5636540330417882, "train_speed(iter/s)": 1.448217 }, { "epoch": 2.4210616511717578, "grad_norm": 0.2951226830482483, "learning_rate": 5.247890472329634e-05, "loss": 0.3445908546447754, "memory(GiB)": 70.5, "step": 56510, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.448223 }, { "epoch": 2.4212758665010066, "grad_norm": 2.2398247718811035, "learning_rate": 5.247218320375376e-05, "loss": 0.3862424612045288, "memory(GiB)": 70.5, "step": 56515, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.448233 }, { "epoch": 2.421490081830256, "grad_norm": 3.302734136581421, "learning_rate": 5.246546163942526e-05, "loss": 0.28869736194610596, "memory(GiB)": 70.5, "step": 56520, "token_acc": 0.9525691699604744, "train_speed(iter/s)": 1.448227 }, { "epoch": 2.4217042971595046, "grad_norm": 1.9991284608840942, "learning_rate": 5.2458740030432595e-05, "loss": 0.23830883502960204, "memory(GiB)": 70.5, "step": 56525, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.448229 }, { "epoch": 2.4219185124887534, "grad_norm": 1.3386024236679077, "learning_rate": 5.245201837689757e-05, "loss": 0.4394369602203369, "memory(GiB)": 70.5, "step": 56530, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.448238 }, { "epoch": 2.4221327278180027, "grad_norm": 3.347008466720581, "learning_rate": 5.2445296678941935e-05, "loss": 0.17995870113372803, "memory(GiB)": 70.5, "step": 56535, "token_acc": 0.9651162790697675, "train_speed(iter/s)": 1.448244 }, { "epoch": 2.4223469431472515, "grad_norm": 2.0907716751098633, "learning_rate": 5.243857493668743e-05, "loss": 0.1879183053970337, "memory(GiB)": 70.5, "step": 56540, "token_acc": 0.9522388059701492, "train_speed(iter/s)": 1.448243 }, { "epoch": 2.4225611584765003, "grad_norm": 2.9270269870758057, "learning_rate": 5.2431853150255886e-05, "loss": 0.36714587211608884, "memory(GiB)": 70.5, "step": 56545, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.448253 }, { "epoch": 2.4227753738057496, "grad_norm": 5.220797538757324, "learning_rate": 5.2425131319769024e-05, "loss": 0.7798663139343261, "memory(GiB)": 70.5, "step": 56550, "token_acc": 0.8051470588235294, "train_speed(iter/s)": 1.448269 }, { "epoch": 2.4229895891349984, "grad_norm": 1.8095414638519287, "learning_rate": 5.241840944534862e-05, "loss": 0.4704314708709717, "memory(GiB)": 70.5, "step": 56555, "token_acc": 0.8868501529051988, "train_speed(iter/s)": 1.448283 }, { "epoch": 2.423203804464247, "grad_norm": 3.4328105449676514, "learning_rate": 5.241168752711648e-05, "loss": 0.4425013065338135, "memory(GiB)": 70.5, "step": 56560, "token_acc": 0.896, "train_speed(iter/s)": 1.448286 }, { "epoch": 2.4234180197934965, "grad_norm": 4.415077209472656, "learning_rate": 5.240496556519435e-05, "loss": 0.49169297218322755, "memory(GiB)": 70.5, "step": 56565, "token_acc": 0.8895522388059701, "train_speed(iter/s)": 1.448294 }, { "epoch": 2.4236322351227453, "grad_norm": 0.13139604032039642, "learning_rate": 5.239824355970401e-05, "loss": 0.10269489288330078, "memory(GiB)": 70.5, "step": 56570, "token_acc": 0.967391304347826, "train_speed(iter/s)": 1.448299 }, { "epoch": 2.4238464504519945, "grad_norm": 0.7733641862869263, "learning_rate": 5.239152151076725e-05, "loss": 0.13380199670791626, "memory(GiB)": 70.5, "step": 56575, "token_acc": 0.9712230215827338, "train_speed(iter/s)": 1.448295 }, { "epoch": 2.4240606657812434, "grad_norm": 3.031653881072998, "learning_rate": 5.2384799418505824e-05, "loss": 0.20992257595062255, "memory(GiB)": 70.5, "step": 56580, "token_acc": 0.9494584837545126, "train_speed(iter/s)": 1.448294 }, { "epoch": 2.424274881110492, "grad_norm": 3.7143805027008057, "learning_rate": 5.237807728304152e-05, "loss": 0.2954482793807983, "memory(GiB)": 70.5, "step": 56585, "token_acc": 0.92578125, "train_speed(iter/s)": 1.448297 }, { "epoch": 2.4244890964397414, "grad_norm": 2.8441526889801025, "learning_rate": 5.2371355104496125e-05, "loss": 0.3840787410736084, "memory(GiB)": 70.5, "step": 56590, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.448297 }, { "epoch": 2.4247033117689902, "grad_norm": 3.4041717052459717, "learning_rate": 5.236463288299139e-05, "loss": 0.2243553638458252, "memory(GiB)": 70.5, "step": 56595, "token_acc": 0.946969696969697, "train_speed(iter/s)": 1.44829 }, { "epoch": 2.424917527098239, "grad_norm": 1.8409174680709839, "learning_rate": 5.235791061864912e-05, "loss": 0.15853266716003417, "memory(GiB)": 70.5, "step": 56600, "token_acc": 0.9651567944250871, "train_speed(iter/s)": 1.448297 }, { "epoch": 2.4251317424274883, "grad_norm": 5.844822883605957, "learning_rate": 5.2351188311591093e-05, "loss": 0.44948477745056153, "memory(GiB)": 70.5, "step": 56605, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.448307 }, { "epoch": 2.425345957756737, "grad_norm": 3.8850724697113037, "learning_rate": 5.234446596193907e-05, "loss": 0.3057550430297852, "memory(GiB)": 70.5, "step": 56610, "token_acc": 0.9395973154362416, "train_speed(iter/s)": 1.448313 }, { "epoch": 2.425560173085986, "grad_norm": 3.722764015197754, "learning_rate": 5.233774356981487e-05, "loss": 0.589443302154541, "memory(GiB)": 70.5, "step": 56615, "token_acc": 0.8783382789317508, "train_speed(iter/s)": 1.448312 }, { "epoch": 2.425774388415235, "grad_norm": 2.6088855266571045, "learning_rate": 5.233102113534024e-05, "loss": 0.6726843833923339, "memory(GiB)": 70.5, "step": 56620, "token_acc": 0.8715953307392996, "train_speed(iter/s)": 1.448303 }, { "epoch": 2.425988603744484, "grad_norm": 5.297031879425049, "learning_rate": 5.2324298658636974e-05, "loss": 0.4924017906188965, "memory(GiB)": 70.5, "step": 56625, "token_acc": 0.897887323943662, "train_speed(iter/s)": 1.448311 }, { "epoch": 2.426202819073733, "grad_norm": 1.4685853719711304, "learning_rate": 5.231757613982686e-05, "loss": 0.2652392864227295, "memory(GiB)": 70.5, "step": 56630, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.44831 }, { "epoch": 2.426417034402982, "grad_norm": 4.596954345703125, "learning_rate": 5.2310853579031674e-05, "loss": 0.3882966279983521, "memory(GiB)": 70.5, "step": 56635, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.448315 }, { "epoch": 2.426631249732231, "grad_norm": 0.4417217969894409, "learning_rate": 5.23041309763732e-05, "loss": 0.3644073247909546, "memory(GiB)": 70.5, "step": 56640, "token_acc": 0.9322709163346613, "train_speed(iter/s)": 1.44831 }, { "epoch": 2.4268454650614797, "grad_norm": 4.258556365966797, "learning_rate": 5.229740833197325e-05, "loss": 0.5097434997558594, "memory(GiB)": 70.5, "step": 56645, "token_acc": 0.8868501529051988, "train_speed(iter/s)": 1.448314 }, { "epoch": 2.427059680390729, "grad_norm": 3.718078851699829, "learning_rate": 5.2290685645953597e-05, "loss": 0.4244943618774414, "memory(GiB)": 70.5, "step": 56650, "token_acc": 0.9029850746268657, "train_speed(iter/s)": 1.448312 }, { "epoch": 2.4272738957199778, "grad_norm": 3.9190125465393066, "learning_rate": 5.2283962918436014e-05, "loss": 0.26002719402313235, "memory(GiB)": 70.5, "step": 56655, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.448319 }, { "epoch": 2.4274881110492266, "grad_norm": 4.380746841430664, "learning_rate": 5.227724014954231e-05, "loss": 0.36347312927246095, "memory(GiB)": 70.5, "step": 56660, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.448326 }, { "epoch": 2.427702326378476, "grad_norm": 2.1761817932128906, "learning_rate": 5.227051733939425e-05, "loss": 0.46587376594543456, "memory(GiB)": 70.5, "step": 56665, "token_acc": 0.9144736842105263, "train_speed(iter/s)": 1.448322 }, { "epoch": 2.4279165417077246, "grad_norm": 2.2093684673309326, "learning_rate": 5.2263794488113635e-05, "loss": 0.27090744972229003, "memory(GiB)": 70.5, "step": 56670, "token_acc": 0.9542857142857143, "train_speed(iter/s)": 1.448327 }, { "epoch": 2.4281307570369735, "grad_norm": 3.7881946563720703, "learning_rate": 5.225707159582227e-05, "loss": 0.27616124153137206, "memory(GiB)": 70.5, "step": 56675, "token_acc": 0.9416666666666667, "train_speed(iter/s)": 1.448325 }, { "epoch": 2.4283449723662227, "grad_norm": 4.233097553253174, "learning_rate": 5.225034866264193e-05, "loss": 0.4729299545288086, "memory(GiB)": 70.5, "step": 56680, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.448322 }, { "epoch": 2.4285591876954715, "grad_norm": 7.156883716583252, "learning_rate": 5.224362568869442e-05, "loss": 0.35352792739868166, "memory(GiB)": 70.5, "step": 56685, "token_acc": 0.9267399267399268, "train_speed(iter/s)": 1.44832 }, { "epoch": 2.4287734030247203, "grad_norm": 1.7449836730957031, "learning_rate": 5.223690267410153e-05, "loss": 0.26649866104125974, "memory(GiB)": 70.5, "step": 56690, "token_acc": 0.9446640316205533, "train_speed(iter/s)": 1.44832 }, { "epoch": 2.4289876183539696, "grad_norm": 2.476670503616333, "learning_rate": 5.223017961898504e-05, "loss": 0.6501718997955322, "memory(GiB)": 70.5, "step": 56695, "token_acc": 0.8664259927797834, "train_speed(iter/s)": 1.448321 }, { "epoch": 2.4292018336832184, "grad_norm": 4.336485862731934, "learning_rate": 5.222345652346675e-05, "loss": 0.3353621244430542, "memory(GiB)": 70.5, "step": 56700, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.448343 }, { "epoch": 2.429416049012467, "grad_norm": 3.560464859008789, "learning_rate": 5.221673338766847e-05, "loss": 0.49829955101013185, "memory(GiB)": 70.5, "step": 56705, "token_acc": 0.9025157232704403, "train_speed(iter/s)": 1.448345 }, { "epoch": 2.4296302643417165, "grad_norm": 1.9422701597213745, "learning_rate": 5.221001021171198e-05, "loss": 0.45909790992736815, "memory(GiB)": 70.5, "step": 56710, "token_acc": 0.8990536277602523, "train_speed(iter/s)": 1.448357 }, { "epoch": 2.4298444796709653, "grad_norm": 5.138984680175781, "learning_rate": 5.220328699571908e-05, "loss": 0.4579333305358887, "memory(GiB)": 70.5, "step": 56715, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.448367 }, { "epoch": 2.430058695000214, "grad_norm": 2.926255702972412, "learning_rate": 5.219656373981158e-05, "loss": 0.40799660682678224, "memory(GiB)": 70.5, "step": 56720, "token_acc": 0.9093959731543624, "train_speed(iter/s)": 1.448365 }, { "epoch": 2.4302729103294634, "grad_norm": 6.59411096572876, "learning_rate": 5.218984044411126e-05, "loss": 0.4676806449890137, "memory(GiB)": 70.5, "step": 56725, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.448382 }, { "epoch": 2.430487125658712, "grad_norm": 1.5078972578048706, "learning_rate": 5.218311710873992e-05, "loss": 0.4594693183898926, "memory(GiB)": 70.5, "step": 56730, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.448385 }, { "epoch": 2.430701340987961, "grad_norm": 1.1715627908706665, "learning_rate": 5.2176393733819376e-05, "loss": 0.293573260307312, "memory(GiB)": 70.5, "step": 56735, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.448387 }, { "epoch": 2.4309155563172102, "grad_norm": 7.083529949188232, "learning_rate": 5.2169670319471406e-05, "loss": 0.40186057090759275, "memory(GiB)": 70.5, "step": 56740, "token_acc": 0.9063670411985019, "train_speed(iter/s)": 1.448397 }, { "epoch": 2.431129771646459, "grad_norm": 0.34995660185813904, "learning_rate": 5.216294686581783e-05, "loss": 0.3416635036468506, "memory(GiB)": 70.5, "step": 56745, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.448404 }, { "epoch": 2.431343986975708, "grad_norm": 3.7364425659179688, "learning_rate": 5.215622337298044e-05, "loss": 0.41409101486206057, "memory(GiB)": 70.5, "step": 56750, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.448425 }, { "epoch": 2.431558202304957, "grad_norm": 1.776629090309143, "learning_rate": 5.214949984108104e-05, "loss": 0.2303173065185547, "memory(GiB)": 70.5, "step": 56755, "token_acc": 0.9513888888888888, "train_speed(iter/s)": 1.448438 }, { "epoch": 2.431772417634206, "grad_norm": 4.4566168785095215, "learning_rate": 5.214277627024144e-05, "loss": 0.3365774154663086, "memory(GiB)": 70.5, "step": 56760, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.448442 }, { "epoch": 2.4319866329634547, "grad_norm": 3.106457233428955, "learning_rate": 5.2136052660583444e-05, "loss": 0.400194787979126, "memory(GiB)": 70.5, "step": 56765, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.44844 }, { "epoch": 2.432200848292704, "grad_norm": 0.6109474301338196, "learning_rate": 5.212932901222883e-05, "loss": 0.26446075439453126, "memory(GiB)": 70.5, "step": 56770, "token_acc": 0.9369085173501577, "train_speed(iter/s)": 1.448439 }, { "epoch": 2.432415063621953, "grad_norm": 2.7874093055725098, "learning_rate": 5.212260532529945e-05, "loss": 0.3496942758560181, "memory(GiB)": 70.5, "step": 56775, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.44844 }, { "epoch": 2.4326292789512016, "grad_norm": 3.9649486541748047, "learning_rate": 5.211588159991707e-05, "loss": 0.31619877815246583, "memory(GiB)": 70.5, "step": 56780, "token_acc": 0.9283387622149837, "train_speed(iter/s)": 1.448449 }, { "epoch": 2.432843494280451, "grad_norm": 3.536764144897461, "learning_rate": 5.210915783620349e-05, "loss": 0.49575209617614746, "memory(GiB)": 70.5, "step": 56785, "token_acc": 0.8867924528301887, "train_speed(iter/s)": 1.448453 }, { "epoch": 2.4330577096096997, "grad_norm": 3.9457309246063232, "learning_rate": 5.2102434034280566e-05, "loss": 0.3534814119338989, "memory(GiB)": 70.5, "step": 56790, "token_acc": 0.9251497005988024, "train_speed(iter/s)": 1.44845 }, { "epoch": 2.4332719249389485, "grad_norm": 2.296849012374878, "learning_rate": 5.2095710194270067e-05, "loss": 0.2899554491043091, "memory(GiB)": 70.5, "step": 56795, "token_acc": 0.9433198380566802, "train_speed(iter/s)": 1.448457 }, { "epoch": 2.4334861402681978, "grad_norm": 4.377267837524414, "learning_rate": 5.208898631629381e-05, "loss": 0.21058299541473388, "memory(GiB)": 70.5, "step": 56800, "token_acc": 0.9427480916030534, "train_speed(iter/s)": 1.448462 }, { "epoch": 2.4337003555974466, "grad_norm": 4.7649993896484375, "learning_rate": 5.208226240047362e-05, "loss": 0.38440356254577634, "memory(GiB)": 70.5, "step": 56805, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.448466 }, { "epoch": 2.4339145709266954, "grad_norm": 5.93859338760376, "learning_rate": 5.207553844693128e-05, "loss": 0.37232208251953125, "memory(GiB)": 70.5, "step": 56810, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.448472 }, { "epoch": 2.4341287862559446, "grad_norm": 1.6571828126907349, "learning_rate": 5.206881445578861e-05, "loss": 0.3343423128128052, "memory(GiB)": 70.5, "step": 56815, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.448469 }, { "epoch": 2.4343430015851935, "grad_norm": 4.519125938415527, "learning_rate": 5.206209042716742e-05, "loss": 0.4601716995239258, "memory(GiB)": 70.5, "step": 56820, "token_acc": 0.912, "train_speed(iter/s)": 1.448479 }, { "epoch": 2.4345572169144423, "grad_norm": 0.5034583210945129, "learning_rate": 5.205536636118955e-05, "loss": 0.38246452808380127, "memory(GiB)": 70.5, "step": 56825, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.448489 }, { "epoch": 2.4347714322436915, "grad_norm": 0.7149195671081543, "learning_rate": 5.204864225797676e-05, "loss": 0.1221287727355957, "memory(GiB)": 70.5, "step": 56830, "token_acc": 0.9699248120300752, "train_speed(iter/s)": 1.448485 }, { "epoch": 2.4349856475729403, "grad_norm": 3.1677072048187256, "learning_rate": 5.204191811765092e-05, "loss": 0.3453385829925537, "memory(GiB)": 70.5, "step": 56835, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.448485 }, { "epoch": 2.435199862902189, "grad_norm": 2.0008749961853027, "learning_rate": 5.203519394033382e-05, "loss": 0.24400105476379394, "memory(GiB)": 70.5, "step": 56840, "token_acc": 0.9323308270676691, "train_speed(iter/s)": 1.448486 }, { "epoch": 2.4354140782314384, "grad_norm": 2.495157241821289, "learning_rate": 5.202846972614726e-05, "loss": 0.26302547454833985, "memory(GiB)": 70.5, "step": 56845, "token_acc": 0.9431137724550899, "train_speed(iter/s)": 1.448489 }, { "epoch": 2.435628293560687, "grad_norm": 2.1651525497436523, "learning_rate": 5.2021745475213076e-05, "loss": 0.5120481491088867, "memory(GiB)": 70.5, "step": 56850, "token_acc": 0.906801007556675, "train_speed(iter/s)": 1.448488 }, { "epoch": 2.435842508889936, "grad_norm": 5.523556709289551, "learning_rate": 5.201502118765307e-05, "loss": 0.36449060440063474, "memory(GiB)": 70.5, "step": 56855, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.44849 }, { "epoch": 2.4360567242191853, "grad_norm": 0.14099812507629395, "learning_rate": 5.200829686358906e-05, "loss": 0.2680016756057739, "memory(GiB)": 70.5, "step": 56860, "token_acc": 0.9457142857142857, "train_speed(iter/s)": 1.448498 }, { "epoch": 2.436270939548434, "grad_norm": 1.6898950338363647, "learning_rate": 5.2001572503142873e-05, "loss": 0.48560256958007814, "memory(GiB)": 70.5, "step": 56865, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.448498 }, { "epoch": 2.436485154877683, "grad_norm": 3.3787007331848145, "learning_rate": 5.1994848106436334e-05, "loss": 0.2775826692581177, "memory(GiB)": 70.5, "step": 56870, "token_acc": 0.9391891891891891, "train_speed(iter/s)": 1.448499 }, { "epoch": 2.436699370206932, "grad_norm": 1.3520610332489014, "learning_rate": 5.198812367359123e-05, "loss": 0.5643519878387451, "memory(GiB)": 70.5, "step": 56875, "token_acc": 0.8885793871866295, "train_speed(iter/s)": 1.4485 }, { "epoch": 2.436913585536181, "grad_norm": 0.2658529281616211, "learning_rate": 5.198139920472942e-05, "loss": 0.14086084365844725, "memory(GiB)": 70.5, "step": 56880, "token_acc": 0.9560439560439561, "train_speed(iter/s)": 1.448513 }, { "epoch": 2.43712780086543, "grad_norm": 7.397921562194824, "learning_rate": 5.1974674699972684e-05, "loss": 0.7537872791290283, "memory(GiB)": 70.5, "step": 56885, "token_acc": 0.8487084870848709, "train_speed(iter/s)": 1.448521 }, { "epoch": 2.437342016194679, "grad_norm": 1.4973970651626587, "learning_rate": 5.196795015944288e-05, "loss": 0.7602391719818116, "memory(GiB)": 70.5, "step": 56890, "token_acc": 0.8187311178247734, "train_speed(iter/s)": 1.448538 }, { "epoch": 2.437556231523928, "grad_norm": 2.2041728496551514, "learning_rate": 5.19612255832618e-05, "loss": 0.3490167140960693, "memory(GiB)": 70.5, "step": 56895, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.448537 }, { "epoch": 2.4377704468531767, "grad_norm": 4.621360778808594, "learning_rate": 5.195450097155128e-05, "loss": 0.4110256195068359, "memory(GiB)": 70.5, "step": 56900, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.448548 }, { "epoch": 2.437984662182426, "grad_norm": 9.312942504882812, "learning_rate": 5.194777632443315e-05, "loss": 0.3526482582092285, "memory(GiB)": 70.5, "step": 56905, "token_acc": 0.9205776173285198, "train_speed(iter/s)": 1.448546 }, { "epoch": 2.4381988775116747, "grad_norm": 3.323967218399048, "learning_rate": 5.194105164202924e-05, "loss": 0.34964756965637206, "memory(GiB)": 70.5, "step": 56910, "token_acc": 0.93359375, "train_speed(iter/s)": 1.448545 }, { "epoch": 2.4384130928409236, "grad_norm": 1.2493937015533447, "learning_rate": 5.1934326924461326e-05, "loss": 0.1740780234336853, "memory(GiB)": 70.5, "step": 56915, "token_acc": 0.9637883008356546, "train_speed(iter/s)": 1.448547 }, { "epoch": 2.438627308170173, "grad_norm": 3.847851276397705, "learning_rate": 5.192760217185129e-05, "loss": 0.5197426795959472, "memory(GiB)": 70.5, "step": 56920, "token_acc": 0.8731884057971014, "train_speed(iter/s)": 1.448548 }, { "epoch": 2.4388415234994216, "grad_norm": 5.639786720275879, "learning_rate": 5.192087738432092e-05, "loss": 0.994933032989502, "memory(GiB)": 70.5, "step": 56925, "token_acc": 0.8038585209003215, "train_speed(iter/s)": 1.448557 }, { "epoch": 2.4390557388286704, "grad_norm": 2.8760182857513428, "learning_rate": 5.191415256199205e-05, "loss": 0.3375399589538574, "memory(GiB)": 70.5, "step": 56930, "token_acc": 0.9292307692307692, "train_speed(iter/s)": 1.448553 }, { "epoch": 2.4392699541579197, "grad_norm": 4.094425201416016, "learning_rate": 5.190742770498652e-05, "loss": 0.3077689647674561, "memory(GiB)": 70.5, "step": 56935, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.448557 }, { "epoch": 2.4394841694871685, "grad_norm": 6.18977689743042, "learning_rate": 5.190070281342615e-05, "loss": 0.3598968744277954, "memory(GiB)": 70.5, "step": 56940, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.448554 }, { "epoch": 2.4396983848164173, "grad_norm": 5.067746162414551, "learning_rate": 5.189397788743275e-05, "loss": 0.3462624788284302, "memory(GiB)": 70.5, "step": 56945, "token_acc": 0.9407407407407408, "train_speed(iter/s)": 1.448558 }, { "epoch": 2.4399126001456666, "grad_norm": 3.243006944656372, "learning_rate": 5.188725292712818e-05, "loss": 0.2595922231674194, "memory(GiB)": 70.5, "step": 56950, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.448557 }, { "epoch": 2.4401268154749154, "grad_norm": 3.467949390411377, "learning_rate": 5.188052793263426e-05, "loss": 0.5685307025909424, "memory(GiB)": 70.5, "step": 56955, "token_acc": 0.8905660377358491, "train_speed(iter/s)": 1.448567 }, { "epoch": 2.440341030804164, "grad_norm": 3.8163371086120605, "learning_rate": 5.1873802904072786e-05, "loss": 0.2986143589019775, "memory(GiB)": 70.5, "step": 56960, "token_acc": 0.933852140077821, "train_speed(iter/s)": 1.448566 }, { "epoch": 2.4405552461334135, "grad_norm": 2.076935291290283, "learning_rate": 5.1867077841565635e-05, "loss": 0.22722980976104737, "memory(GiB)": 70.5, "step": 56965, "token_acc": 0.961038961038961, "train_speed(iter/s)": 1.448586 }, { "epoch": 2.4407694614626623, "grad_norm": 2.3600635528564453, "learning_rate": 5.186035274523461e-05, "loss": 0.3513511657714844, "memory(GiB)": 70.5, "step": 56970, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.448588 }, { "epoch": 2.440983676791911, "grad_norm": 3.849141836166382, "learning_rate": 5.1853627615201536e-05, "loss": 0.15970182418823242, "memory(GiB)": 70.5, "step": 56975, "token_acc": 0.961038961038961, "train_speed(iter/s)": 1.448583 }, { "epoch": 2.4411978921211603, "grad_norm": 2.2628090381622314, "learning_rate": 5.184690245158829e-05, "loss": 0.32185912132263184, "memory(GiB)": 70.5, "step": 56980, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.448582 }, { "epoch": 2.441412107450409, "grad_norm": 2.1634321212768555, "learning_rate": 5.1840177254516666e-05, "loss": 0.37788803577423097, "memory(GiB)": 70.5, "step": 56985, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.448587 }, { "epoch": 2.441626322779658, "grad_norm": 2.570725679397583, "learning_rate": 5.183345202410849e-05, "loss": 0.3029686212539673, "memory(GiB)": 70.5, "step": 56990, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.448594 }, { "epoch": 2.441840538108907, "grad_norm": 6.6105780601501465, "learning_rate": 5.182672676048561e-05, "loss": 0.5242393493652344, "memory(GiB)": 70.5, "step": 56995, "token_acc": 0.8814814814814815, "train_speed(iter/s)": 1.448593 }, { "epoch": 2.442054753438156, "grad_norm": 1.8263847827911377, "learning_rate": 5.182000146376986e-05, "loss": 0.4684088706970215, "memory(GiB)": 70.5, "step": 57000, "token_acc": 0.9360902255639098, "train_speed(iter/s)": 1.448586 }, { "epoch": 2.442054753438156, "eval_loss": 2.337622880935669, "eval_runtime": 13.5194, "eval_samples_per_second": 7.397, "eval_steps_per_second": 7.397, "eval_token_acc": 0.46112600536193027, "step": 57000 }, { "epoch": 2.442268968767405, "grad_norm": 6.144504547119141, "learning_rate": 5.181327613408309e-05, "loss": 0.24183378219604493, "memory(GiB)": 70.5, "step": 57005, "token_acc": 0.5959692898272553, "train_speed(iter/s)": 1.448053 }, { "epoch": 2.442483184096654, "grad_norm": 1.335349678993225, "learning_rate": 5.1806550771547115e-05, "loss": 0.1728075385093689, "memory(GiB)": 70.5, "step": 57010, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.448056 }, { "epoch": 2.442697399425903, "grad_norm": 2.4555814266204834, "learning_rate": 5.179982537628378e-05, "loss": 0.5658479690551758, "memory(GiB)": 70.5, "step": 57015, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.448056 }, { "epoch": 2.4429116147551517, "grad_norm": 1.040018081665039, "learning_rate": 5.1793099948414925e-05, "loss": 0.4027434825897217, "memory(GiB)": 70.5, "step": 57020, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.448065 }, { "epoch": 2.443125830084401, "grad_norm": 4.92796516418457, "learning_rate": 5.1786374488062375e-05, "loss": 0.28519158363342284, "memory(GiB)": 70.5, "step": 57025, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.448069 }, { "epoch": 2.44334004541365, "grad_norm": 1.804032802581787, "learning_rate": 5.1779648995347975e-05, "loss": 0.142301607131958, "memory(GiB)": 70.5, "step": 57030, "token_acc": 0.9713114754098361, "train_speed(iter/s)": 1.448065 }, { "epoch": 2.4435542607428986, "grad_norm": 4.3377885818481445, "learning_rate": 5.177292347039358e-05, "loss": 0.4124687671661377, "memory(GiB)": 70.5, "step": 57035, "token_acc": 0.8960573476702509, "train_speed(iter/s)": 1.448063 }, { "epoch": 2.443768476072148, "grad_norm": 2.925485849380493, "learning_rate": 5.176619791332099e-05, "loss": 0.31764233112335205, "memory(GiB)": 70.5, "step": 57040, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.448064 }, { "epoch": 2.4439826914013967, "grad_norm": 4.367794036865234, "learning_rate": 5.175947232425207e-05, "loss": 0.4404900550842285, "memory(GiB)": 70.5, "step": 57045, "token_acc": 0.8940809968847352, "train_speed(iter/s)": 1.448077 }, { "epoch": 2.4441969067306455, "grad_norm": 5.157540321350098, "learning_rate": 5.1752746703308664e-05, "loss": 0.32676215171813966, "memory(GiB)": 70.5, "step": 57050, "token_acc": 0.9295392953929539, "train_speed(iter/s)": 1.448085 }, { "epoch": 2.4444111220598947, "grad_norm": 5.1797590255737305, "learning_rate": 5.174602105061262e-05, "loss": 0.5902118682861328, "memory(GiB)": 70.5, "step": 57055, "token_acc": 0.881578947368421, "train_speed(iter/s)": 1.448082 }, { "epoch": 2.4446253373891436, "grad_norm": 2.106867551803589, "learning_rate": 5.1739295366285745e-05, "loss": 0.3868129730224609, "memory(GiB)": 70.5, "step": 57060, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.448076 }, { "epoch": 2.4448395527183924, "grad_norm": 3.035968065261841, "learning_rate": 5.173256965044991e-05, "loss": 0.1077705979347229, "memory(GiB)": 70.5, "step": 57065, "token_acc": 0.9639344262295082, "train_speed(iter/s)": 1.448081 }, { "epoch": 2.4450537680476416, "grad_norm": 0.25868186354637146, "learning_rate": 5.1725843903226966e-05, "loss": 0.3204088926315308, "memory(GiB)": 70.5, "step": 57070, "token_acc": 0.9274447949526814, "train_speed(iter/s)": 1.448087 }, { "epoch": 2.4452679833768904, "grad_norm": 3.27046799659729, "learning_rate": 5.171911812473872e-05, "loss": 0.3323599576950073, "memory(GiB)": 70.5, "step": 57075, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.448123 }, { "epoch": 2.4454821987061393, "grad_norm": 2.8100695610046387, "learning_rate": 5.171239231510704e-05, "loss": 0.2608546257019043, "memory(GiB)": 70.5, "step": 57080, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.448123 }, { "epoch": 2.4456964140353885, "grad_norm": 5.271791934967041, "learning_rate": 5.1705666474453785e-05, "loss": 0.4280370235443115, "memory(GiB)": 70.5, "step": 57085, "token_acc": 0.9078498293515358, "train_speed(iter/s)": 1.448144 }, { "epoch": 2.4459106293646373, "grad_norm": 1.4096628427505493, "learning_rate": 5.1698940602900756e-05, "loss": 0.35230939388275145, "memory(GiB)": 70.5, "step": 57090, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.448151 }, { "epoch": 2.446124844693886, "grad_norm": 2.1621456146240234, "learning_rate": 5.169221470056984e-05, "loss": 0.3582402944564819, "memory(GiB)": 70.5, "step": 57095, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.448139 }, { "epoch": 2.4463390600231354, "grad_norm": 0.35775312781333923, "learning_rate": 5.168548876758288e-05, "loss": 0.2380190372467041, "memory(GiB)": 70.5, "step": 57100, "token_acc": 0.949685534591195, "train_speed(iter/s)": 1.448135 }, { "epoch": 2.446553275352384, "grad_norm": 2.0258853435516357, "learning_rate": 5.1678762804061685e-05, "loss": 0.2505171298980713, "memory(GiB)": 70.5, "step": 57105, "token_acc": 0.9493243243243243, "train_speed(iter/s)": 1.448142 }, { "epoch": 2.446767490681633, "grad_norm": 2.813734531402588, "learning_rate": 5.167203681012813e-05, "loss": 0.3606745958328247, "memory(GiB)": 70.5, "step": 57110, "token_acc": 0.9183006535947712, "train_speed(iter/s)": 1.448138 }, { "epoch": 2.4469817060108823, "grad_norm": 2.2597310543060303, "learning_rate": 5.1665310785904066e-05, "loss": 0.25474960803985597, "memory(GiB)": 70.5, "step": 57115, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.448139 }, { "epoch": 2.447195921340131, "grad_norm": 2.9739768505096436, "learning_rate": 5.165858473151133e-05, "loss": 0.3156658411026001, "memory(GiB)": 70.5, "step": 57120, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.448139 }, { "epoch": 2.44741013666938, "grad_norm": 4.254566669464111, "learning_rate": 5.165185864707178e-05, "loss": 0.25340938568115234, "memory(GiB)": 70.5, "step": 57125, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.448137 }, { "epoch": 2.447624351998629, "grad_norm": 7.658082962036133, "learning_rate": 5.164513253270727e-05, "loss": 0.7418982028961182, "memory(GiB)": 70.5, "step": 57130, "token_acc": 0.8621908127208481, "train_speed(iter/s)": 1.448135 }, { "epoch": 2.447838567327878, "grad_norm": 2.461824893951416, "learning_rate": 5.163840638853963e-05, "loss": 0.5576932907104493, "memory(GiB)": 70.5, "step": 57135, "token_acc": 0.8884892086330936, "train_speed(iter/s)": 1.448153 }, { "epoch": 2.448052782657127, "grad_norm": 4.729228496551514, "learning_rate": 5.163168021469073e-05, "loss": 0.3876772165298462, "memory(GiB)": 70.5, "step": 57140, "token_acc": 0.9489795918367347, "train_speed(iter/s)": 1.448152 }, { "epoch": 2.448266997986376, "grad_norm": 4.5476202964782715, "learning_rate": 5.16249540112824e-05, "loss": 0.2736964702606201, "memory(GiB)": 70.5, "step": 57145, "token_acc": 0.9430604982206405, "train_speed(iter/s)": 1.448154 }, { "epoch": 2.448481213315625, "grad_norm": 5.482770919799805, "learning_rate": 5.161822777843651e-05, "loss": 0.5452134132385253, "memory(GiB)": 70.5, "step": 57150, "token_acc": 0.8788732394366198, "train_speed(iter/s)": 1.448151 }, { "epoch": 2.4486954286448737, "grad_norm": 1.8832621574401855, "learning_rate": 5.1611501516274904e-05, "loss": 0.2564457893371582, "memory(GiB)": 70.5, "step": 57155, "token_acc": 0.9490909090909091, "train_speed(iter/s)": 1.448154 }, { "epoch": 2.448909643974123, "grad_norm": 3.9060311317443848, "learning_rate": 5.160477522491943e-05, "loss": 0.4757274627685547, "memory(GiB)": 70.5, "step": 57160, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.448153 }, { "epoch": 2.4491238593033717, "grad_norm": 3.2309885025024414, "learning_rate": 5.159804890449196e-05, "loss": 0.22482154369354249, "memory(GiB)": 70.5, "step": 57165, "token_acc": 0.9518900343642611, "train_speed(iter/s)": 1.448145 }, { "epoch": 2.4493380746326205, "grad_norm": 3.949028730392456, "learning_rate": 5.159132255511434e-05, "loss": 0.38207497596740725, "memory(GiB)": 70.5, "step": 57170, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.448144 }, { "epoch": 2.44955228996187, "grad_norm": 3.718838691711426, "learning_rate": 5.15845961769084e-05, "loss": 0.45941834449768065, "memory(GiB)": 70.5, "step": 57175, "token_acc": 0.9171974522292994, "train_speed(iter/s)": 1.448162 }, { "epoch": 2.4497665052911186, "grad_norm": 0.8816488981246948, "learning_rate": 5.157786976999602e-05, "loss": 0.2516125202178955, "memory(GiB)": 70.5, "step": 57180, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.448165 }, { "epoch": 2.4499807206203674, "grad_norm": 2.3906962871551514, "learning_rate": 5.157114333449906e-05, "loss": 0.4934391975402832, "memory(GiB)": 70.5, "step": 57185, "token_acc": 0.8977272727272727, "train_speed(iter/s)": 1.448169 }, { "epoch": 2.4501949359496167, "grad_norm": 3.526014804840088, "learning_rate": 5.1564416870539346e-05, "loss": 0.2576786279678345, "memory(GiB)": 70.5, "step": 57190, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.448176 }, { "epoch": 2.4504091512788655, "grad_norm": 5.763206958770752, "learning_rate": 5.155769037823876e-05, "loss": 0.21323716640472412, "memory(GiB)": 70.5, "step": 57195, "token_acc": 0.9426229508196722, "train_speed(iter/s)": 1.448177 }, { "epoch": 2.4506233666081143, "grad_norm": 3.9867773056030273, "learning_rate": 5.155096385771917e-05, "loss": 0.6590302944183349, "memory(GiB)": 70.5, "step": 57200, "token_acc": 0.8551236749116607, "train_speed(iter/s)": 1.44818 }, { "epoch": 2.4508375819373636, "grad_norm": 3.44386625289917, "learning_rate": 5.1544237309102395e-05, "loss": 0.4201654434204102, "memory(GiB)": 70.5, "step": 57205, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.448181 }, { "epoch": 2.4510517972666124, "grad_norm": 3.092787265777588, "learning_rate": 5.153751073251032e-05, "loss": 0.3629965305328369, "memory(GiB)": 70.5, "step": 57210, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.448206 }, { "epoch": 2.451266012595861, "grad_norm": 4.947149753570557, "learning_rate": 5.15307841280648e-05, "loss": 0.9124042510986328, "memory(GiB)": 70.5, "step": 57215, "token_acc": 0.8036175710594315, "train_speed(iter/s)": 1.448233 }, { "epoch": 2.4514802279251104, "grad_norm": 3.6263175010681152, "learning_rate": 5.152405749588768e-05, "loss": 0.2827276706695557, "memory(GiB)": 70.5, "step": 57220, "token_acc": 0.9308176100628931, "train_speed(iter/s)": 1.448232 }, { "epoch": 2.4516944432543593, "grad_norm": 2.3771677017211914, "learning_rate": 5.151733083610083e-05, "loss": 0.25628745555877686, "memory(GiB)": 70.5, "step": 57225, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.448239 }, { "epoch": 2.451908658583608, "grad_norm": 3.5029382705688477, "learning_rate": 5.15106041488261e-05, "loss": 0.3804809093475342, "memory(GiB)": 70.5, "step": 57230, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.448242 }, { "epoch": 2.4521228739128573, "grad_norm": 3.273998498916626, "learning_rate": 5.1503877434185366e-05, "loss": 0.6043426036834717, "memory(GiB)": 70.5, "step": 57235, "token_acc": 0.8929889298892989, "train_speed(iter/s)": 1.448261 }, { "epoch": 2.452337089242106, "grad_norm": 3.550443172454834, "learning_rate": 5.149715069230049e-05, "loss": 0.48622884750366213, "memory(GiB)": 70.5, "step": 57240, "token_acc": 0.8996960486322189, "train_speed(iter/s)": 1.448272 }, { "epoch": 2.452551304571355, "grad_norm": 4.146138668060303, "learning_rate": 5.149042392329333e-05, "loss": 0.33394978046417234, "memory(GiB)": 70.5, "step": 57245, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.448284 }, { "epoch": 2.452765519900604, "grad_norm": 5.634178161621094, "learning_rate": 5.148369712728572e-05, "loss": 0.45397200584411623, "memory(GiB)": 70.5, "step": 57250, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.448291 }, { "epoch": 2.452979735229853, "grad_norm": 4.171021938323975, "learning_rate": 5.1476970304399565e-05, "loss": 0.47551565170288085, "memory(GiB)": 70.5, "step": 57255, "token_acc": 0.8843537414965986, "train_speed(iter/s)": 1.44829 }, { "epoch": 2.453193950559102, "grad_norm": 5.587617874145508, "learning_rate": 5.1470243454756694e-05, "loss": 0.3858364105224609, "memory(GiB)": 70.5, "step": 57260, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.448294 }, { "epoch": 2.453408165888351, "grad_norm": 3.9496777057647705, "learning_rate": 5.146351657847898e-05, "loss": 0.3622389793395996, "memory(GiB)": 70.5, "step": 57265, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.448309 }, { "epoch": 2.4536223812176, "grad_norm": 3.154157876968384, "learning_rate": 5.14567896756883e-05, "loss": 0.29033479690551756, "memory(GiB)": 70.5, "step": 57270, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.448306 }, { "epoch": 2.4538365965468487, "grad_norm": 3.855154275894165, "learning_rate": 5.145006274650652e-05, "loss": 0.5654054641723633, "memory(GiB)": 70.5, "step": 57275, "token_acc": 0.8862876254180602, "train_speed(iter/s)": 1.448308 }, { "epoch": 2.454050811876098, "grad_norm": 4.873984336853027, "learning_rate": 5.144333579105547e-05, "loss": 0.2679815530776978, "memory(GiB)": 70.5, "step": 57280, "token_acc": 0.9507042253521126, "train_speed(iter/s)": 1.448297 }, { "epoch": 2.454265027205347, "grad_norm": 4.955414295196533, "learning_rate": 5.143660880945705e-05, "loss": 0.6150885581970215, "memory(GiB)": 70.5, "step": 57285, "token_acc": 0.8768115942028986, "train_speed(iter/s)": 1.448296 }, { "epoch": 2.4544792425345956, "grad_norm": 3.182302474975586, "learning_rate": 5.1429881801833116e-05, "loss": 0.29818115234375, "memory(GiB)": 70.5, "step": 57290, "token_acc": 0.9556451612903226, "train_speed(iter/s)": 1.448297 }, { "epoch": 2.454693457863845, "grad_norm": 3.328632116317749, "learning_rate": 5.1423154768305524e-05, "loss": 0.5018520832061768, "memory(GiB)": 70.5, "step": 57295, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.448304 }, { "epoch": 2.4549076731930937, "grad_norm": 1.3337379693984985, "learning_rate": 5.1416427708996154e-05, "loss": 0.4940495014190674, "memory(GiB)": 70.5, "step": 57300, "token_acc": 0.8855218855218855, "train_speed(iter/s)": 1.44832 }, { "epoch": 2.4551218885223425, "grad_norm": 4.709545135498047, "learning_rate": 5.1409700624026855e-05, "loss": 0.5250825405120849, "memory(GiB)": 70.5, "step": 57305, "token_acc": 0.9080882352941176, "train_speed(iter/s)": 1.448333 }, { "epoch": 2.4553361038515917, "grad_norm": 2.801558256149292, "learning_rate": 5.1402973513519527e-05, "loss": 0.6051984786987304, "memory(GiB)": 70.5, "step": 57310, "token_acc": 0.8932806324110671, "train_speed(iter/s)": 1.448345 }, { "epoch": 2.4555503191808405, "grad_norm": 3.4132633209228516, "learning_rate": 5.139624637759601e-05, "loss": 0.44390435218811036, "memory(GiB)": 70.5, "step": 57315, "token_acc": 0.8901098901098901, "train_speed(iter/s)": 1.44835 }, { "epoch": 2.4557645345100894, "grad_norm": 6.438333511352539, "learning_rate": 5.138951921637817e-05, "loss": 0.4446430206298828, "memory(GiB)": 70.5, "step": 57320, "token_acc": 0.8978328173374613, "train_speed(iter/s)": 1.448353 }, { "epoch": 2.4559787498393386, "grad_norm": 3.5518391132354736, "learning_rate": 5.1382792029987904e-05, "loss": 0.34377572536468504, "memory(GiB)": 70.5, "step": 57325, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.448355 }, { "epoch": 2.4561929651685874, "grad_norm": 1.894930362701416, "learning_rate": 5.137606481854705e-05, "loss": 0.4266801834106445, "memory(GiB)": 70.5, "step": 57330, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.448373 }, { "epoch": 2.4564071804978362, "grad_norm": 3.741417169570923, "learning_rate": 5.136933758217749e-05, "loss": 0.37605106830596924, "memory(GiB)": 70.5, "step": 57335, "token_acc": 0.92, "train_speed(iter/s)": 1.448373 }, { "epoch": 2.4566213958270855, "grad_norm": 2.9244461059570312, "learning_rate": 5.13626103210011e-05, "loss": 0.16047637462615966, "memory(GiB)": 70.5, "step": 57340, "token_acc": 0.9659090909090909, "train_speed(iter/s)": 1.448374 }, { "epoch": 2.4568356111563343, "grad_norm": 2.7852649688720703, "learning_rate": 5.135588303513975e-05, "loss": 0.23168425559997557, "memory(GiB)": 70.5, "step": 57345, "token_acc": 0.9543859649122807, "train_speed(iter/s)": 1.448376 }, { "epoch": 2.457049826485583, "grad_norm": 2.2606663703918457, "learning_rate": 5.1349155724715294e-05, "loss": 0.45886988639831544, "memory(GiB)": 70.5, "step": 57350, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.448382 }, { "epoch": 2.4572640418148324, "grad_norm": 2.3765628337860107, "learning_rate": 5.1342428389849626e-05, "loss": 0.25389719009399414, "memory(GiB)": 70.5, "step": 57355, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.448381 }, { "epoch": 2.457478257144081, "grad_norm": 2.1347835063934326, "learning_rate": 5.133570103066462e-05, "loss": 0.3848104476928711, "memory(GiB)": 70.5, "step": 57360, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.448408 }, { "epoch": 2.45769247247333, "grad_norm": 4.5239691734313965, "learning_rate": 5.1328973647282116e-05, "loss": 0.2631680011749268, "memory(GiB)": 70.5, "step": 57365, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.448402 }, { "epoch": 2.4579066878025793, "grad_norm": 3.230781316757202, "learning_rate": 5.1322246239824024e-05, "loss": 0.31971254348754885, "memory(GiB)": 70.5, "step": 57370, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.448398 }, { "epoch": 2.458120903131828, "grad_norm": 3.502159595489502, "learning_rate": 5.131551880841219e-05, "loss": 0.32837982177734376, "memory(GiB)": 70.5, "step": 57375, "token_acc": 0.9043824701195219, "train_speed(iter/s)": 1.448416 }, { "epoch": 2.458335118461077, "grad_norm": 2.6337080001831055, "learning_rate": 5.1308791353168484e-05, "loss": 0.5119675159454345, "memory(GiB)": 70.5, "step": 57380, "token_acc": 0.9034749034749034, "train_speed(iter/s)": 1.448429 }, { "epoch": 2.458549333790326, "grad_norm": 2.750481128692627, "learning_rate": 5.130206387421482e-05, "loss": 0.46085262298583984, "memory(GiB)": 70.5, "step": 57385, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.448446 }, { "epoch": 2.458763549119575, "grad_norm": 1.7483971118927002, "learning_rate": 5.1295336371673045e-05, "loss": 0.4210374355316162, "memory(GiB)": 70.5, "step": 57390, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.448456 }, { "epoch": 2.4589777644488238, "grad_norm": 3.512218713760376, "learning_rate": 5.1288608845665034e-05, "loss": 0.6180743217468262, "memory(GiB)": 70.5, "step": 57395, "token_acc": 0.8725868725868726, "train_speed(iter/s)": 1.448467 }, { "epoch": 2.459191979778073, "grad_norm": 4.392849922180176, "learning_rate": 5.128188129631266e-05, "loss": 0.4831393718719482, "memory(GiB)": 70.5, "step": 57400, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.448472 }, { "epoch": 2.459406195107322, "grad_norm": 0.13307121396064758, "learning_rate": 5.12751537237378e-05, "loss": 0.33095386028289797, "memory(GiB)": 70.5, "step": 57405, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.448486 }, { "epoch": 2.4596204104365706, "grad_norm": 4.319854736328125, "learning_rate": 5.126842612806234e-05, "loss": 0.35591554641723633, "memory(GiB)": 70.5, "step": 57410, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.448489 }, { "epoch": 2.45983462576582, "grad_norm": 0.690584659576416, "learning_rate": 5.1261698509408154e-05, "loss": 0.2808829307556152, "memory(GiB)": 70.5, "step": 57415, "token_acc": 0.9335443037974683, "train_speed(iter/s)": 1.44849 }, { "epoch": 2.4600488410950687, "grad_norm": 0.33131173253059387, "learning_rate": 5.125497086789711e-05, "loss": 0.4754952907562256, "memory(GiB)": 70.5, "step": 57420, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.448507 }, { "epoch": 2.460263056424318, "grad_norm": 3.8474037647247314, "learning_rate": 5.1248243203651094e-05, "loss": 0.36409337520599366, "memory(GiB)": 70.5, "step": 57425, "token_acc": 0.8916967509025271, "train_speed(iter/s)": 1.448503 }, { "epoch": 2.460477271753567, "grad_norm": 4.821377277374268, "learning_rate": 5.124151551679198e-05, "loss": 0.33535501956939695, "memory(GiB)": 70.5, "step": 57430, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.448509 }, { "epoch": 2.4606914870828156, "grad_norm": 3.6813952922821045, "learning_rate": 5.1234787807441655e-05, "loss": 0.20571620464324952, "memory(GiB)": 70.5, "step": 57435, "token_acc": 0.9405204460966543, "train_speed(iter/s)": 1.448525 }, { "epoch": 2.460905702412065, "grad_norm": 1.302983283996582, "learning_rate": 5.122806007572198e-05, "loss": 0.4556541442871094, "memory(GiB)": 70.5, "step": 57440, "token_acc": 0.9034267912772586, "train_speed(iter/s)": 1.448517 }, { "epoch": 2.4611199177413137, "grad_norm": 4.141223907470703, "learning_rate": 5.1221332321754855e-05, "loss": 0.3118530988693237, "memory(GiB)": 70.5, "step": 57445, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.448516 }, { "epoch": 2.4613341330705625, "grad_norm": 2.997743606567383, "learning_rate": 5.1214604545662135e-05, "loss": 0.3710240364074707, "memory(GiB)": 70.5, "step": 57450, "token_acc": 0.9153094462540716, "train_speed(iter/s)": 1.448517 }, { "epoch": 2.4615483483998117, "grad_norm": 5.682093143463135, "learning_rate": 5.120787674756573e-05, "loss": 0.20258989334106445, "memory(GiB)": 70.5, "step": 57455, "token_acc": 0.9691780821917808, "train_speed(iter/s)": 1.44852 }, { "epoch": 2.4617625637290605, "grad_norm": 1.3123506307601929, "learning_rate": 5.120114892758749e-05, "loss": 0.30606348514556886, "memory(GiB)": 70.5, "step": 57460, "token_acc": 0.9394812680115274, "train_speed(iter/s)": 1.448523 }, { "epoch": 2.4619767790583094, "grad_norm": 7.336658000946045, "learning_rate": 5.119442108584932e-05, "loss": 0.4541748046875, "memory(GiB)": 70.5, "step": 57465, "token_acc": 0.9094488188976378, "train_speed(iter/s)": 1.44853 }, { "epoch": 2.4621909943875586, "grad_norm": 2.516355514526367, "learning_rate": 5.11876932224731e-05, "loss": 0.3995200157165527, "memory(GiB)": 70.5, "step": 57470, "token_acc": 0.9120521172638436, "train_speed(iter/s)": 1.448535 }, { "epoch": 2.4624052097168074, "grad_norm": 5.464118957519531, "learning_rate": 5.11809653375807e-05, "loss": 0.3761723041534424, "memory(GiB)": 70.5, "step": 57475, "token_acc": 0.9296875, "train_speed(iter/s)": 1.448544 }, { "epoch": 2.4626194250460562, "grad_norm": 3.897918939590454, "learning_rate": 5.1174237431293994e-05, "loss": 0.2936774730682373, "memory(GiB)": 70.5, "step": 57480, "token_acc": 0.9328358208955224, "train_speed(iter/s)": 1.448549 }, { "epoch": 2.4628336403753055, "grad_norm": 0.7981911301612854, "learning_rate": 5.116750950373487e-05, "loss": 0.4814487934112549, "memory(GiB)": 70.5, "step": 57485, "token_acc": 0.8878504672897196, "train_speed(iter/s)": 1.448555 }, { "epoch": 2.4630478557045543, "grad_norm": 4.580133438110352, "learning_rate": 5.1160781555025225e-05, "loss": 0.3346451759338379, "memory(GiB)": 70.5, "step": 57490, "token_acc": 0.9461538461538461, "train_speed(iter/s)": 1.44856 }, { "epoch": 2.463262071033803, "grad_norm": 5.6508708000183105, "learning_rate": 5.115405358528693e-05, "loss": 0.5510771751403809, "memory(GiB)": 70.5, "step": 57495, "token_acc": 0.8915343915343915, "train_speed(iter/s)": 1.448574 }, { "epoch": 2.4634762863630524, "grad_norm": 0.6537868976593018, "learning_rate": 5.114732559464188e-05, "loss": 0.25106942653656006, "memory(GiB)": 70.5, "step": 57500, "token_acc": 0.940809968847352, "train_speed(iter/s)": 1.448572 }, { "epoch": 2.4634762863630524, "eval_loss": 2.226762294769287, "eval_runtime": 14.1431, "eval_samples_per_second": 7.071, "eval_steps_per_second": 7.071, "eval_token_acc": 0.4633821571238349, "step": 57500 }, { "epoch": 2.463690501692301, "grad_norm": 5.141593933105469, "learning_rate": 5.114059758321196e-05, "loss": 0.5656673431396484, "memory(GiB)": 70.5, "step": 57505, "token_acc": 0.5967153284671532, "train_speed(iter/s)": 1.448014 }, { "epoch": 2.46390471702155, "grad_norm": 3.721742630004883, "learning_rate": 5.1133869551119016e-05, "loss": 0.2736979961395264, "memory(GiB)": 70.5, "step": 57510, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.448015 }, { "epoch": 2.4641189323507993, "grad_norm": 2.8623461723327637, "learning_rate": 5.112714149848499e-05, "loss": 0.6348925590515136, "memory(GiB)": 70.5, "step": 57515, "token_acc": 0.904, "train_speed(iter/s)": 1.448017 }, { "epoch": 2.464333147680048, "grad_norm": 2.298549175262451, "learning_rate": 5.112041342543171e-05, "loss": 0.19450544118881224, "memory(GiB)": 70.5, "step": 57520, "token_acc": 0.9494584837545126, "train_speed(iter/s)": 1.448021 }, { "epoch": 2.464547363009297, "grad_norm": 2.084829092025757, "learning_rate": 5.1113685332081094e-05, "loss": 0.3198352098464966, "memory(GiB)": 70.5, "step": 57525, "token_acc": 0.9441176470588235, "train_speed(iter/s)": 1.448025 }, { "epoch": 2.464761578338546, "grad_norm": 2.7829604148864746, "learning_rate": 5.110695721855505e-05, "loss": 0.25968480110168457, "memory(GiB)": 70.5, "step": 57530, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.448028 }, { "epoch": 2.464975793667795, "grad_norm": 4.313577175140381, "learning_rate": 5.1100229084975424e-05, "loss": 0.42516179084777833, "memory(GiB)": 70.5, "step": 57535, "token_acc": 0.9036144578313253, "train_speed(iter/s)": 1.448034 }, { "epoch": 2.4651900089970438, "grad_norm": 6.713052749633789, "learning_rate": 5.109350093146411e-05, "loss": 0.4600831508636475, "memory(GiB)": 70.5, "step": 57540, "token_acc": 0.8984375, "train_speed(iter/s)": 1.448037 }, { "epoch": 2.465404224326293, "grad_norm": 4.533518314361572, "learning_rate": 5.108677275814301e-05, "loss": 0.32813096046447754, "memory(GiB)": 70.5, "step": 57545, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.448034 }, { "epoch": 2.465618439655542, "grad_norm": 2.270995855331421, "learning_rate": 5.108004456513399e-05, "loss": 0.18810672760009767, "memory(GiB)": 70.5, "step": 57550, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.448033 }, { "epoch": 2.4658326549847907, "grad_norm": 5.517618179321289, "learning_rate": 5.107331635255895e-05, "loss": 0.2597689628601074, "memory(GiB)": 70.5, "step": 57555, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.448049 }, { "epoch": 2.46604687031404, "grad_norm": 4.748538017272949, "learning_rate": 5.1066588120539785e-05, "loss": 0.2662620782852173, "memory(GiB)": 70.5, "step": 57560, "token_acc": 0.9486166007905138, "train_speed(iter/s)": 1.448069 }, { "epoch": 2.4662610856432887, "grad_norm": 0.1795201599597931, "learning_rate": 5.105985986919838e-05, "loss": 0.24135701656341552, "memory(GiB)": 70.5, "step": 57565, "token_acc": 0.9419354838709677, "train_speed(iter/s)": 1.448076 }, { "epoch": 2.4664753009725375, "grad_norm": 2.888547420501709, "learning_rate": 5.1053131598656614e-05, "loss": 0.48728437423706056, "memory(GiB)": 70.5, "step": 57570, "token_acc": 0.8882521489971347, "train_speed(iter/s)": 1.448071 }, { "epoch": 2.466689516301787, "grad_norm": 4.077663421630859, "learning_rate": 5.104640330903638e-05, "loss": 0.24443700313568115, "memory(GiB)": 70.5, "step": 57575, "token_acc": 0.9578651685393258, "train_speed(iter/s)": 1.448078 }, { "epoch": 2.4669037316310356, "grad_norm": 6.469759941101074, "learning_rate": 5.103967500045956e-05, "loss": 0.5081413269042969, "memory(GiB)": 70.5, "step": 57580, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.448079 }, { "epoch": 2.4671179469602844, "grad_norm": 4.494875907897949, "learning_rate": 5.1032946673048067e-05, "loss": 0.4226790428161621, "memory(GiB)": 70.5, "step": 57585, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.448081 }, { "epoch": 2.4673321622895337, "grad_norm": 3.0619664192199707, "learning_rate": 5.102621832692378e-05, "loss": 0.4612852096557617, "memory(GiB)": 70.5, "step": 57590, "token_acc": 0.8993506493506493, "train_speed(iter/s)": 1.448082 }, { "epoch": 2.4675463776187825, "grad_norm": 3.9224090576171875, "learning_rate": 5.1019489962208555e-05, "loss": 0.4343403816223145, "memory(GiB)": 70.5, "step": 57595, "token_acc": 0.9077809798270894, "train_speed(iter/s)": 1.448084 }, { "epoch": 2.4677605929480313, "grad_norm": 3.3029725551605225, "learning_rate": 5.101276157902434e-05, "loss": 0.1526494264602661, "memory(GiB)": 70.5, "step": 57600, "token_acc": 0.9609120521172638, "train_speed(iter/s)": 1.448077 }, { "epoch": 2.4679748082772806, "grad_norm": 16.864526748657227, "learning_rate": 5.100603317749299e-05, "loss": 0.4483452796936035, "memory(GiB)": 70.5, "step": 57605, "token_acc": 0.9017857142857143, "train_speed(iter/s)": 1.448093 }, { "epoch": 2.4681890236065294, "grad_norm": 0.016532177105545998, "learning_rate": 5.099930475773641e-05, "loss": 0.24435489177703856, "memory(GiB)": 70.5, "step": 57610, "token_acc": 0.9420849420849421, "train_speed(iter/s)": 1.448097 }, { "epoch": 2.468403238935778, "grad_norm": 4.017344951629639, "learning_rate": 5.099257631987648e-05, "loss": 0.22484233379364013, "memory(GiB)": 70.5, "step": 57615, "token_acc": 0.9547169811320755, "train_speed(iter/s)": 1.448094 }, { "epoch": 2.4686174542650274, "grad_norm": 2.2246880531311035, "learning_rate": 5.098584786403512e-05, "loss": 0.3118997573852539, "memory(GiB)": 70.5, "step": 57620, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.448088 }, { "epoch": 2.4688316695942762, "grad_norm": 0.4765454828739166, "learning_rate": 5.0979119390334175e-05, "loss": 0.3198455095291138, "memory(GiB)": 70.5, "step": 57625, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.448086 }, { "epoch": 2.469045884923525, "grad_norm": 4.370665550231934, "learning_rate": 5.097239089889558e-05, "loss": 0.45493717193603517, "memory(GiB)": 70.5, "step": 57630, "token_acc": 0.9065743944636678, "train_speed(iter/s)": 1.448086 }, { "epoch": 2.4692601002527743, "grad_norm": 1.7771892547607422, "learning_rate": 5.0965662389841196e-05, "loss": 0.21089978218078614, "memory(GiB)": 70.5, "step": 57635, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.448091 }, { "epoch": 2.469474315582023, "grad_norm": 1.1322224140167236, "learning_rate": 5.095893386329293e-05, "loss": 0.401251220703125, "memory(GiB)": 70.5, "step": 57640, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.448088 }, { "epoch": 2.469688530911272, "grad_norm": 2.2274117469787598, "learning_rate": 5.0952205319372706e-05, "loss": 0.35601975917816164, "memory(GiB)": 70.5, "step": 57645, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.448115 }, { "epoch": 2.469902746240521, "grad_norm": 2.631916046142578, "learning_rate": 5.094547675820237e-05, "loss": 0.20703794956207275, "memory(GiB)": 70.5, "step": 57650, "token_acc": 0.9628252788104089, "train_speed(iter/s)": 1.448116 }, { "epoch": 2.47011696156977, "grad_norm": 3.208961248397827, "learning_rate": 5.093874817990383e-05, "loss": 0.3639467477798462, "memory(GiB)": 70.5, "step": 57655, "token_acc": 0.9240924092409241, "train_speed(iter/s)": 1.448122 }, { "epoch": 2.470331176899019, "grad_norm": 2.897606134414673, "learning_rate": 5.0932019584599e-05, "loss": 0.19153220653533937, "memory(GiB)": 70.5, "step": 57660, "token_acc": 0.9578544061302682, "train_speed(iter/s)": 1.448115 }, { "epoch": 2.470545392228268, "grad_norm": 0.30485573410987854, "learning_rate": 5.092529097240976e-05, "loss": 0.15061994791030883, "memory(GiB)": 70.5, "step": 57665, "token_acc": 0.9759450171821306, "train_speed(iter/s)": 1.448115 }, { "epoch": 2.470759607557517, "grad_norm": 4.037619113922119, "learning_rate": 5.091856234345799e-05, "loss": 0.49530630111694335, "memory(GiB)": 70.5, "step": 57670, "token_acc": 0.8904109589041096, "train_speed(iter/s)": 1.448112 }, { "epoch": 2.4709738228867657, "grad_norm": 3.223482131958008, "learning_rate": 5.0911833697865607e-05, "loss": 0.1896933674812317, "memory(GiB)": 70.5, "step": 57675, "token_acc": 0.9528985507246377, "train_speed(iter/s)": 1.448111 }, { "epoch": 2.471188038216015, "grad_norm": 1.139317512512207, "learning_rate": 5.0905105035754516e-05, "loss": 0.32213854789733887, "memory(GiB)": 70.5, "step": 57680, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.448125 }, { "epoch": 2.4714022535452638, "grad_norm": 5.615582466125488, "learning_rate": 5.089837635724658e-05, "loss": 0.5486551284790039, "memory(GiB)": 70.5, "step": 57685, "token_acc": 0.8740458015267175, "train_speed(iter/s)": 1.448142 }, { "epoch": 2.4716164688745126, "grad_norm": 2.2609503269195557, "learning_rate": 5.0891647662463724e-05, "loss": 0.33198723793029783, "memory(GiB)": 70.5, "step": 57690, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.448142 }, { "epoch": 2.471830684203762, "grad_norm": 4.014735221862793, "learning_rate": 5.088491895152784e-05, "loss": 0.21634740829467775, "memory(GiB)": 70.5, "step": 57695, "token_acc": 0.9465020576131687, "train_speed(iter/s)": 1.448142 }, { "epoch": 2.4720448995330107, "grad_norm": 6.342479705810547, "learning_rate": 5.0878190224560804e-05, "loss": 0.19491324424743653, "memory(GiB)": 70.5, "step": 57700, "token_acc": 0.9466192170818505, "train_speed(iter/s)": 1.448142 }, { "epoch": 2.4722591148622595, "grad_norm": 3.2803385257720947, "learning_rate": 5.0871461481684546e-05, "loss": 0.16082704067230225, "memory(GiB)": 70.5, "step": 57705, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.448144 }, { "epoch": 2.4724733301915087, "grad_norm": 3.5789096355438232, "learning_rate": 5.0864732723020915e-05, "loss": 0.38210196495056153, "memory(GiB)": 70.5, "step": 57710, "token_acc": 0.9438596491228071, "train_speed(iter/s)": 1.448158 }, { "epoch": 2.4726875455207575, "grad_norm": 4.992557525634766, "learning_rate": 5.085800394869187e-05, "loss": 0.6534647941589355, "memory(GiB)": 70.5, "step": 57715, "token_acc": 0.8659420289855072, "train_speed(iter/s)": 1.448153 }, { "epoch": 2.4729017608500063, "grad_norm": 5.504053115844727, "learning_rate": 5.0851275158819264e-05, "loss": 0.5103494167327881, "memory(GiB)": 70.5, "step": 57720, "token_acc": 0.9153094462540716, "train_speed(iter/s)": 1.448147 }, { "epoch": 2.4731159761792556, "grad_norm": 0.2236536741256714, "learning_rate": 5.084454635352501e-05, "loss": 0.20838935375213624, "memory(GiB)": 70.5, "step": 57725, "token_acc": 0.9505703422053232, "train_speed(iter/s)": 1.448142 }, { "epoch": 2.4733301915085044, "grad_norm": 1.6235387325286865, "learning_rate": 5.083781753293102e-05, "loss": 0.12348778247833252, "memory(GiB)": 70.5, "step": 57730, "token_acc": 0.9701492537313433, "train_speed(iter/s)": 1.448142 }, { "epoch": 2.4735444068377532, "grad_norm": 2.560702085494995, "learning_rate": 5.083108869715918e-05, "loss": 0.34062347412109373, "memory(GiB)": 70.5, "step": 57735, "token_acc": 0.9233870967741935, "train_speed(iter/s)": 1.448157 }, { "epoch": 2.4737586221670025, "grad_norm": 8.288564682006836, "learning_rate": 5.082435984633137e-05, "loss": 0.6879231929779053, "memory(GiB)": 70.5, "step": 57740, "token_acc": 0.8644688644688645, "train_speed(iter/s)": 1.448155 }, { "epoch": 2.4739728374962513, "grad_norm": 1.5602251291275024, "learning_rate": 5.081763098056951e-05, "loss": 0.3509143114089966, "memory(GiB)": 70.5, "step": 57745, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.448148 }, { "epoch": 2.4741870528255, "grad_norm": 7.003243446350098, "learning_rate": 5.08109020999955e-05, "loss": 0.6695644378662109, "memory(GiB)": 70.5, "step": 57750, "token_acc": 0.8600682593856656, "train_speed(iter/s)": 1.448155 }, { "epoch": 2.4744012681547494, "grad_norm": 8.481746673583984, "learning_rate": 5.080417320473124e-05, "loss": 0.5123307228088378, "memory(GiB)": 70.5, "step": 57755, "token_acc": 0.9157894736842105, "train_speed(iter/s)": 1.448148 }, { "epoch": 2.474615483483998, "grad_norm": 3.320241928100586, "learning_rate": 5.0797444294898644e-05, "loss": 0.3713593721389771, "memory(GiB)": 70.5, "step": 57760, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.448148 }, { "epoch": 2.474829698813247, "grad_norm": 3.2255969047546387, "learning_rate": 5.079071537061959e-05, "loss": 0.4024803638458252, "memory(GiB)": 70.5, "step": 57765, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.448146 }, { "epoch": 2.4750439141424962, "grad_norm": 2.117729663848877, "learning_rate": 5.078398643201597e-05, "loss": 0.43294405937194824, "memory(GiB)": 70.5, "step": 57770, "token_acc": 0.9063444108761329, "train_speed(iter/s)": 1.44815 }, { "epoch": 2.475258129471745, "grad_norm": 2.1570982933044434, "learning_rate": 5.077725747920972e-05, "loss": 0.16863411664962769, "memory(GiB)": 70.5, "step": 57775, "token_acc": 0.9661016949152542, "train_speed(iter/s)": 1.448151 }, { "epoch": 2.475472344800994, "grad_norm": 1.0152091979980469, "learning_rate": 5.07705285123227e-05, "loss": 0.1779167413711548, "memory(GiB)": 70.5, "step": 57780, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.448153 }, { "epoch": 2.475686560130243, "grad_norm": 5.921262264251709, "learning_rate": 5.076379953147684e-05, "loss": 0.3118840456008911, "memory(GiB)": 70.5, "step": 57785, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.448155 }, { "epoch": 2.475900775459492, "grad_norm": 3.110541820526123, "learning_rate": 5.075707053679404e-05, "loss": 0.2043220043182373, "memory(GiB)": 70.5, "step": 57790, "token_acc": 0.9507042253521126, "train_speed(iter/s)": 1.448151 }, { "epoch": 2.4761149907887408, "grad_norm": 4.457925319671631, "learning_rate": 5.075034152839621e-05, "loss": 0.3023625612258911, "memory(GiB)": 70.5, "step": 57795, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.448147 }, { "epoch": 2.47632920611799, "grad_norm": 4.281535625457764, "learning_rate": 5.074361250640521e-05, "loss": 0.38251986503601076, "memory(GiB)": 70.5, "step": 57800, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.44815 }, { "epoch": 2.476543421447239, "grad_norm": 1.8090054988861084, "learning_rate": 5.0736883470942986e-05, "loss": 0.21448130607604982, "memory(GiB)": 70.5, "step": 57805, "token_acc": 0.9570957095709571, "train_speed(iter/s)": 1.448156 }, { "epoch": 2.4767576367764876, "grad_norm": 3.3664722442626953, "learning_rate": 5.0730154422131424e-05, "loss": 0.4634557723999023, "memory(GiB)": 70.5, "step": 57810, "token_acc": 0.9110320284697508, "train_speed(iter/s)": 1.448162 }, { "epoch": 2.476971852105737, "grad_norm": 1.7194995880126953, "learning_rate": 5.072342536009245e-05, "loss": 0.5181403636932373, "memory(GiB)": 70.5, "step": 57815, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.448158 }, { "epoch": 2.4771860674349857, "grad_norm": 5.165866374969482, "learning_rate": 5.071669628494792e-05, "loss": 0.5838961124420166, "memory(GiB)": 70.5, "step": 57820, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.44817 }, { "epoch": 2.4774002827642345, "grad_norm": 4.031564712524414, "learning_rate": 5.070996719681977e-05, "loss": 0.4997495174407959, "memory(GiB)": 70.5, "step": 57825, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.448183 }, { "epoch": 2.4776144980934838, "grad_norm": 3.0841140747070312, "learning_rate": 5.070323809582991e-05, "loss": 0.36577553749084474, "memory(GiB)": 70.5, "step": 57830, "token_acc": 0.9226006191950464, "train_speed(iter/s)": 1.448181 }, { "epoch": 2.4778287134227326, "grad_norm": 2.436889171600342, "learning_rate": 5.0696508982100225e-05, "loss": 0.5106980323791503, "memory(GiB)": 70.5, "step": 57835, "token_acc": 0.8781512605042017, "train_speed(iter/s)": 1.448183 }, { "epoch": 2.4780429287519814, "grad_norm": 4.704957485198975, "learning_rate": 5.0689779855752616e-05, "loss": 0.49423532485961913, "memory(GiB)": 70.5, "step": 57840, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.448195 }, { "epoch": 2.4782571440812307, "grad_norm": 4.427624702453613, "learning_rate": 5.068305071690902e-05, "loss": 0.4680153846740723, "memory(GiB)": 70.5, "step": 57845, "token_acc": 0.909433962264151, "train_speed(iter/s)": 1.44821 }, { "epoch": 2.4784713594104795, "grad_norm": 0.4787992537021637, "learning_rate": 5.067632156569131e-05, "loss": 0.30900299549102783, "memory(GiB)": 70.5, "step": 57850, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.448202 }, { "epoch": 2.4786855747397283, "grad_norm": 3.6225392818450928, "learning_rate": 5.066959240222138e-05, "loss": 0.32021512985229494, "memory(GiB)": 70.5, "step": 57855, "token_acc": 0.9495268138801262, "train_speed(iter/s)": 1.448204 }, { "epoch": 2.4788997900689775, "grad_norm": 2.7160630226135254, "learning_rate": 5.066286322662118e-05, "loss": 0.43352417945861815, "memory(GiB)": 70.5, "step": 57860, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.448208 }, { "epoch": 2.4791140053982264, "grad_norm": 3.8975117206573486, "learning_rate": 5.0656134039012593e-05, "loss": 0.5031645774841309, "memory(GiB)": 70.5, "step": 57865, "token_acc": 0.8854961832061069, "train_speed(iter/s)": 1.448205 }, { "epoch": 2.479328220727475, "grad_norm": 5.569057464599609, "learning_rate": 5.06494048395175e-05, "loss": 0.3610517501831055, "memory(GiB)": 70.5, "step": 57870, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.448208 }, { "epoch": 2.4795424360567244, "grad_norm": 1.5605252981185913, "learning_rate": 5.064267562825785e-05, "loss": 0.20415079593658447, "memory(GiB)": 70.5, "step": 57875, "token_acc": 0.95703125, "train_speed(iter/s)": 1.448209 }, { "epoch": 2.4797566513859732, "grad_norm": 4.2009992599487305, "learning_rate": 5.0635946405355525e-05, "loss": 0.27608556747436525, "memory(GiB)": 70.5, "step": 57880, "token_acc": 0.9368029739776952, "train_speed(iter/s)": 1.448226 }, { "epoch": 2.479970866715222, "grad_norm": 2.1549198627471924, "learning_rate": 5.062921717093243e-05, "loss": 0.3411139488220215, "memory(GiB)": 70.5, "step": 57885, "token_acc": 0.9283489096573209, "train_speed(iter/s)": 1.448234 }, { "epoch": 2.4801850820444713, "grad_norm": 3.6851730346679688, "learning_rate": 5.062248792511048e-05, "loss": 0.489317798614502, "memory(GiB)": 70.5, "step": 57890, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.448234 }, { "epoch": 2.48039929737372, "grad_norm": 3.2298247814178467, "learning_rate": 5.061575866801157e-05, "loss": 0.6305659770965576, "memory(GiB)": 70.5, "step": 57895, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.448254 }, { "epoch": 2.480613512702969, "grad_norm": 1.344207763671875, "learning_rate": 5.0609029399757615e-05, "loss": 0.14408282041549683, "memory(GiB)": 70.5, "step": 57900, "token_acc": 0.9659442724458205, "train_speed(iter/s)": 1.448262 }, { "epoch": 2.480827728032218, "grad_norm": 6.299310684204102, "learning_rate": 5.060230012047052e-05, "loss": 0.4464579582214355, "memory(GiB)": 70.5, "step": 57905, "token_acc": 0.8996138996138996, "train_speed(iter/s)": 1.448282 }, { "epoch": 2.481041943361467, "grad_norm": 4.062583923339844, "learning_rate": 5.059557083027221e-05, "loss": 0.38767390251159667, "memory(GiB)": 70.5, "step": 57910, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.448281 }, { "epoch": 2.481256158690716, "grad_norm": 2.4842631816864014, "learning_rate": 5.058884152928455e-05, "loss": 0.22927021980285645, "memory(GiB)": 70.5, "step": 57915, "token_acc": 0.9531772575250836, "train_speed(iter/s)": 1.448297 }, { "epoch": 2.481470374019965, "grad_norm": 3.3886702060699463, "learning_rate": 5.0582112217629494e-05, "loss": 0.39917612075805664, "memory(GiB)": 70.5, "step": 57920, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.448304 }, { "epoch": 2.481684589349214, "grad_norm": 3.3726518154144287, "learning_rate": 5.0575382895428914e-05, "loss": 0.3868204832077026, "memory(GiB)": 70.5, "step": 57925, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.448295 }, { "epoch": 2.4818988046784627, "grad_norm": 0.673920214176178, "learning_rate": 5.0568653562804734e-05, "loss": 0.19283438920974733, "memory(GiB)": 70.5, "step": 57930, "token_acc": 0.9563758389261745, "train_speed(iter/s)": 1.448304 }, { "epoch": 2.482113020007712, "grad_norm": 3.4440228939056396, "learning_rate": 5.056192421987888e-05, "loss": 0.3311715841293335, "memory(GiB)": 70.5, "step": 57935, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.448302 }, { "epoch": 2.4823272353369608, "grad_norm": 2.9458343982696533, "learning_rate": 5.055519486677322e-05, "loss": 0.3579057455062866, "memory(GiB)": 70.5, "step": 57940, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.448305 }, { "epoch": 2.4825414506662096, "grad_norm": 5.038288593292236, "learning_rate": 5.0548465503609697e-05, "loss": 0.3333864450454712, "memory(GiB)": 70.5, "step": 57945, "token_acc": 0.9191176470588235, "train_speed(iter/s)": 1.448296 }, { "epoch": 2.482755665995459, "grad_norm": 2.120593547821045, "learning_rate": 5.054173613051021e-05, "loss": 0.28450026512146, "memory(GiB)": 70.5, "step": 57950, "token_acc": 0.9382716049382716, "train_speed(iter/s)": 1.4483 }, { "epoch": 2.4829698813247076, "grad_norm": 3.570071220397949, "learning_rate": 5.0535006747596646e-05, "loss": 0.2659111022949219, "memory(GiB)": 70.5, "step": 57955, "token_acc": 0.9342105263157895, "train_speed(iter/s)": 1.4483 }, { "epoch": 2.4831840966539565, "grad_norm": 11.674662590026855, "learning_rate": 5.052827735499095e-05, "loss": 0.4510330677032471, "memory(GiB)": 70.5, "step": 57960, "token_acc": 0.8670886075949367, "train_speed(iter/s)": 1.448315 }, { "epoch": 2.4833983119832057, "grad_norm": 4.965940952301025, "learning_rate": 5.0521547952815005e-05, "loss": 0.36631026268005373, "memory(GiB)": 70.5, "step": 57965, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.448318 }, { "epoch": 2.4836125273124545, "grad_norm": 3.598508834838867, "learning_rate": 5.0514818541190745e-05, "loss": 0.2621077299118042, "memory(GiB)": 70.5, "step": 57970, "token_acc": 0.9413793103448276, "train_speed(iter/s)": 1.448319 }, { "epoch": 2.4838267426417033, "grad_norm": 2.4188621044158936, "learning_rate": 5.050808912024004e-05, "loss": 0.24543557167053223, "memory(GiB)": 70.5, "step": 57975, "token_acc": 0.9501779359430605, "train_speed(iter/s)": 1.44832 }, { "epoch": 2.4840409579709526, "grad_norm": 3.3372976779937744, "learning_rate": 5.050135969008485e-05, "loss": 0.33246328830718996, "memory(GiB)": 70.5, "step": 57980, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.448321 }, { "epoch": 2.4842551733002014, "grad_norm": 3.425675392150879, "learning_rate": 5.049463025084703e-05, "loss": 0.22016820907592774, "memory(GiB)": 70.5, "step": 57985, "token_acc": 0.949685534591195, "train_speed(iter/s)": 1.448333 }, { "epoch": 2.48446938862945, "grad_norm": 14.959691047668457, "learning_rate": 5.0487900802648544e-05, "loss": 0.43864688873291013, "memory(GiB)": 70.5, "step": 57990, "token_acc": 0.9033333333333333, "train_speed(iter/s)": 1.448346 }, { "epoch": 2.4846836039586995, "grad_norm": 3.4138286113739014, "learning_rate": 5.048117134561128e-05, "loss": 0.40889759063720704, "memory(GiB)": 70.5, "step": 57995, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.448339 }, { "epoch": 2.4848978192879483, "grad_norm": 2.4841935634613037, "learning_rate": 5.0474441879857125e-05, "loss": 0.2819089889526367, "memory(GiB)": 70.5, "step": 58000, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.448339 }, { "epoch": 2.4848978192879483, "eval_loss": 2.525204658508301, "eval_runtime": 13.2612, "eval_samples_per_second": 7.541, "eval_steps_per_second": 7.541, "eval_token_acc": 0.4610303830911493, "step": 58000 }, { "epoch": 2.485112034617197, "grad_norm": 3.5615322589874268, "learning_rate": 5.046771240550801e-05, "loss": 0.4144239902496338, "memory(GiB)": 70.5, "step": 58005, "token_acc": 0.5921787709497207, "train_speed(iter/s)": 1.447814 }, { "epoch": 2.4853262499464464, "grad_norm": 4.912576198577881, "learning_rate": 5.0460982922685865e-05, "loss": 0.4326457500457764, "memory(GiB)": 70.5, "step": 58010, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.447816 }, { "epoch": 2.485540465275695, "grad_norm": 4.670139789581299, "learning_rate": 5.045425343151255e-05, "loss": 0.2512061357498169, "memory(GiB)": 70.5, "step": 58015, "token_acc": 0.9477124183006536, "train_speed(iter/s)": 1.44783 }, { "epoch": 2.485754680604944, "grad_norm": 1.2623980045318604, "learning_rate": 5.044752393211004e-05, "loss": 0.15283788442611695, "memory(GiB)": 70.5, "step": 58020, "token_acc": 0.9659863945578231, "train_speed(iter/s)": 1.447831 }, { "epoch": 2.4859688959341932, "grad_norm": 4.146984100341797, "learning_rate": 5.044079442460021e-05, "loss": 0.2501493453979492, "memory(GiB)": 70.5, "step": 58025, "token_acc": 0.9413680781758957, "train_speed(iter/s)": 1.447836 }, { "epoch": 2.486183111263442, "grad_norm": 9.228221893310547, "learning_rate": 5.043406490910497e-05, "loss": 0.5813421249389649, "memory(GiB)": 70.5, "step": 58030, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.447843 }, { "epoch": 2.486397326592691, "grad_norm": 4.899510860443115, "learning_rate": 5.0427335385746244e-05, "loss": 0.6823376655578614, "memory(GiB)": 70.5, "step": 58035, "token_acc": 0.8818443804034583, "train_speed(iter/s)": 1.447846 }, { "epoch": 2.48661154192194, "grad_norm": 5.592764377593994, "learning_rate": 5.042060585464592e-05, "loss": 0.6560054302215577, "memory(GiB)": 70.5, "step": 58040, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.447868 }, { "epoch": 2.486825757251189, "grad_norm": 1.8296306133270264, "learning_rate": 5.0413876315925925e-05, "loss": 0.2164611339569092, "memory(GiB)": 70.5, "step": 58045, "token_acc": 0.9446153846153846, "train_speed(iter/s)": 1.447872 }, { "epoch": 2.4870399725804377, "grad_norm": 2.7461724281311035, "learning_rate": 5.040714676970819e-05, "loss": 0.3907480239868164, "memory(GiB)": 70.5, "step": 58050, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.447878 }, { "epoch": 2.487254187909687, "grad_norm": 2.466661214828491, "learning_rate": 5.04004172161146e-05, "loss": 0.44435667991638184, "memory(GiB)": 70.5, "step": 58055, "token_acc": 0.9083969465648855, "train_speed(iter/s)": 1.44789 }, { "epoch": 2.487468403238936, "grad_norm": 1.4530051946640015, "learning_rate": 5.039368765526708e-05, "loss": 0.4670563220977783, "memory(GiB)": 70.5, "step": 58060, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.447901 }, { "epoch": 2.4876826185681846, "grad_norm": 0.8356004953384399, "learning_rate": 5.038695808728754e-05, "loss": 0.15135031938552856, "memory(GiB)": 70.5, "step": 58065, "token_acc": 0.972318339100346, "train_speed(iter/s)": 1.447903 }, { "epoch": 2.487896833897434, "grad_norm": 3.0948617458343506, "learning_rate": 5.038022851229789e-05, "loss": 0.4650623321533203, "memory(GiB)": 70.5, "step": 58070, "token_acc": 0.8793103448275862, "train_speed(iter/s)": 1.447899 }, { "epoch": 2.4881110492266827, "grad_norm": 1.290993332862854, "learning_rate": 5.037349893042005e-05, "loss": 0.23324911594390868, "memory(GiB)": 70.5, "step": 58075, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.447907 }, { "epoch": 2.4883252645559315, "grad_norm": 2.0615110397338867, "learning_rate": 5.036676934177591e-05, "loss": 0.31691946983337405, "memory(GiB)": 70.5, "step": 58080, "token_acc": 0.9292307692307692, "train_speed(iter/s)": 1.447918 }, { "epoch": 2.4885394798851808, "grad_norm": 2.5442302227020264, "learning_rate": 5.036003974648741e-05, "loss": 0.27466464042663574, "memory(GiB)": 70.5, "step": 58085, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.447926 }, { "epoch": 2.4887536952144296, "grad_norm": 3.909088134765625, "learning_rate": 5.0353310144676445e-05, "loss": 0.6377777099609375, "memory(GiB)": 70.5, "step": 58090, "token_acc": 0.8910256410256411, "train_speed(iter/s)": 1.447924 }, { "epoch": 2.4889679105436784, "grad_norm": 2.121277093887329, "learning_rate": 5.034658053646495e-05, "loss": 0.29391729831695557, "memory(GiB)": 70.5, "step": 58095, "token_acc": 0.9263803680981595, "train_speed(iter/s)": 1.447927 }, { "epoch": 2.4891821258729276, "grad_norm": 4.233802318572998, "learning_rate": 5.033985092197481e-05, "loss": 0.49867701530456543, "memory(GiB)": 70.5, "step": 58100, "token_acc": 0.9, "train_speed(iter/s)": 1.447931 }, { "epoch": 2.4893963412021765, "grad_norm": 3.2582767009735107, "learning_rate": 5.033312130132796e-05, "loss": 0.39182727336883544, "memory(GiB)": 70.5, "step": 58105, "token_acc": 0.9144736842105263, "train_speed(iter/s)": 1.447933 }, { "epoch": 2.4896105565314253, "grad_norm": 0.8822880983352661, "learning_rate": 5.03263916746463e-05, "loss": 0.3140188455581665, "memory(GiB)": 70.5, "step": 58110, "token_acc": 0.9204892966360856, "train_speed(iter/s)": 1.447958 }, { "epoch": 2.4898247718606745, "grad_norm": 6.402811527252197, "learning_rate": 5.031966204205175e-05, "loss": 0.5472658157348633, "memory(GiB)": 70.5, "step": 58115, "token_acc": 0.8867924528301887, "train_speed(iter/s)": 1.447973 }, { "epoch": 2.4900389871899233, "grad_norm": 3.8917136192321777, "learning_rate": 5.0312932403666214e-05, "loss": 0.704986572265625, "memory(GiB)": 70.5, "step": 58120, "token_acc": 0.8275862068965517, "train_speed(iter/s)": 1.44798 }, { "epoch": 2.490253202519172, "grad_norm": 3.404987335205078, "learning_rate": 5.0306202759611634e-05, "loss": 0.3738913059234619, "memory(GiB)": 70.5, "step": 58125, "token_acc": 0.9017857142857143, "train_speed(iter/s)": 1.448003 }, { "epoch": 2.4904674178484214, "grad_norm": 5.4538702964782715, "learning_rate": 5.0299473110009876e-05, "loss": 0.6012606620788574, "memory(GiB)": 70.5, "step": 58130, "token_acc": 0.8788819875776398, "train_speed(iter/s)": 1.448015 }, { "epoch": 2.49068163317767, "grad_norm": 1.8098759651184082, "learning_rate": 5.0292743454982905e-05, "loss": 0.3407072782516479, "memory(GiB)": 70.5, "step": 58135, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.448032 }, { "epoch": 2.490895848506919, "grad_norm": 2.2439239025115967, "learning_rate": 5.028601379465261e-05, "loss": 0.3112943649291992, "memory(GiB)": 70.5, "step": 58140, "token_acc": 0.934375, "train_speed(iter/s)": 1.448031 }, { "epoch": 2.4911100638361683, "grad_norm": 4.009323596954346, "learning_rate": 5.0279284129140894e-05, "loss": 0.3397305965423584, "memory(GiB)": 70.5, "step": 58145, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.448028 }, { "epoch": 2.491324279165417, "grad_norm": 4.77995491027832, "learning_rate": 5.027255445856969e-05, "loss": 0.3978678464889526, "memory(GiB)": 70.5, "step": 58150, "token_acc": 0.9161849710982659, "train_speed(iter/s)": 1.448028 }, { "epoch": 2.491538494494666, "grad_norm": 1.452418565750122, "learning_rate": 5.0265824783060914e-05, "loss": 0.4963284969329834, "memory(GiB)": 70.5, "step": 58155, "token_acc": 0.8721311475409836, "train_speed(iter/s)": 1.448032 }, { "epoch": 2.491752709823915, "grad_norm": 2.7235355377197266, "learning_rate": 5.025909510273645e-05, "loss": 0.3268429279327393, "memory(GiB)": 70.5, "step": 58160, "token_acc": 0.9273504273504274, "train_speed(iter/s)": 1.44803 }, { "epoch": 2.491966925153164, "grad_norm": 2.0628857612609863, "learning_rate": 5.025236541771826e-05, "loss": 0.39425551891326904, "memory(GiB)": 70.5, "step": 58165, "token_acc": 0.9224489795918367, "train_speed(iter/s)": 1.448018 }, { "epoch": 2.492181140482413, "grad_norm": 2.8351871967315674, "learning_rate": 5.0245635728128225e-05, "loss": 0.5423977375030518, "memory(GiB)": 70.5, "step": 58170, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.448023 }, { "epoch": 2.492395355811662, "grad_norm": 1.747424602508545, "learning_rate": 5.023890603408827e-05, "loss": 0.31359641551971434, "memory(GiB)": 70.5, "step": 58175, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.448031 }, { "epoch": 2.492609571140911, "grad_norm": 2.159369707107544, "learning_rate": 5.0232176335720314e-05, "loss": 0.3227251052856445, "memory(GiB)": 70.5, "step": 58180, "token_acc": 0.9263565891472868, "train_speed(iter/s)": 1.448036 }, { "epoch": 2.4928237864701597, "grad_norm": 3.9651904106140137, "learning_rate": 5.0225446633146247e-05, "loss": 0.5251911640167236, "memory(GiB)": 70.5, "step": 58185, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.448043 }, { "epoch": 2.493038001799409, "grad_norm": 6.2718095779418945, "learning_rate": 5.021871692648801e-05, "loss": 0.26261334419250487, "memory(GiB)": 70.5, "step": 58190, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.448054 }, { "epoch": 2.4932522171286577, "grad_norm": 3.2178328037261963, "learning_rate": 5.0211987215867516e-05, "loss": 0.14038885831832887, "memory(GiB)": 70.5, "step": 58195, "token_acc": 0.9768339768339769, "train_speed(iter/s)": 1.448043 }, { "epoch": 2.4934664324579066, "grad_norm": 3.5306203365325928, "learning_rate": 5.020525750140668e-05, "loss": 0.34163966178894045, "memory(GiB)": 70.5, "step": 58200, "token_acc": 0.9096774193548387, "train_speed(iter/s)": 1.448044 }, { "epoch": 2.493680647787156, "grad_norm": 4.089174747467041, "learning_rate": 5.01985277832274e-05, "loss": 0.44449739456176757, "memory(GiB)": 70.5, "step": 58205, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.44805 }, { "epoch": 2.4938948631164046, "grad_norm": 5.325443267822266, "learning_rate": 5.0191798061451614e-05, "loss": 0.5127364158630371, "memory(GiB)": 70.5, "step": 58210, "token_acc": 0.8839590443686007, "train_speed(iter/s)": 1.448054 }, { "epoch": 2.4941090784456534, "grad_norm": 4.069386959075928, "learning_rate": 5.0185068336201215e-05, "loss": 0.27183146476745607, "memory(GiB)": 70.5, "step": 58215, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.448064 }, { "epoch": 2.4943232937749027, "grad_norm": 1.456366777420044, "learning_rate": 5.0178338607598144e-05, "loss": 0.3775474071502686, "memory(GiB)": 70.5, "step": 58220, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.448056 }, { "epoch": 2.4945375091041515, "grad_norm": 4.074248790740967, "learning_rate": 5.0171608875764286e-05, "loss": 0.20277223587036133, "memory(GiB)": 70.5, "step": 58225, "token_acc": 0.9588014981273408, "train_speed(iter/s)": 1.448047 }, { "epoch": 2.4947517244334003, "grad_norm": 4.0218610763549805, "learning_rate": 5.0164879140821586e-05, "loss": 0.48090085983276365, "memory(GiB)": 70.5, "step": 58230, "token_acc": 0.8996138996138996, "train_speed(iter/s)": 1.448043 }, { "epoch": 2.4949659397626496, "grad_norm": 2.432323455810547, "learning_rate": 5.015814940289193e-05, "loss": 0.3804868221282959, "memory(GiB)": 70.5, "step": 58235, "token_acc": 0.9233128834355828, "train_speed(iter/s)": 1.448048 }, { "epoch": 2.4951801550918984, "grad_norm": 3.281125783920288, "learning_rate": 5.015141966209726e-05, "loss": 0.5362614154815674, "memory(GiB)": 70.5, "step": 58240, "token_acc": 0.8990536277602523, "train_speed(iter/s)": 1.448059 }, { "epoch": 2.495394370421147, "grad_norm": 1.484585165977478, "learning_rate": 5.014468991855947e-05, "loss": 0.3260998010635376, "memory(GiB)": 70.5, "step": 58245, "token_acc": 0.9532374100719424, "train_speed(iter/s)": 1.448063 }, { "epoch": 2.4956085857503965, "grad_norm": 2.8264541625976562, "learning_rate": 5.01379601724005e-05, "loss": 0.5451457023620605, "memory(GiB)": 70.5, "step": 58250, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.448062 }, { "epoch": 2.4958228010796453, "grad_norm": 3.5189356803894043, "learning_rate": 5.013123042374226e-05, "loss": 0.3798977375030518, "memory(GiB)": 70.5, "step": 58255, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.448064 }, { "epoch": 2.496037016408894, "grad_norm": 1.1913162469863892, "learning_rate": 5.012450067270663e-05, "loss": 0.509389591217041, "memory(GiB)": 70.5, "step": 58260, "token_acc": 0.8821428571428571, "train_speed(iter/s)": 1.448075 }, { "epoch": 2.4962512317381433, "grad_norm": 3.0768730640411377, "learning_rate": 5.0117770919415566e-05, "loss": 0.2642954111099243, "memory(GiB)": 70.5, "step": 58265, "token_acc": 0.9390243902439024, "train_speed(iter/s)": 1.448074 }, { "epoch": 2.496465447067392, "grad_norm": 0.03239154815673828, "learning_rate": 5.011104116399098e-05, "loss": 0.4092276096343994, "memory(GiB)": 70.5, "step": 58270, "token_acc": 0.9164086687306502, "train_speed(iter/s)": 1.448075 }, { "epoch": 2.496679662396641, "grad_norm": 2.109123706817627, "learning_rate": 5.010431140655477e-05, "loss": 0.3383918523788452, "memory(GiB)": 70.5, "step": 58275, "token_acc": 0.9228486646884273, "train_speed(iter/s)": 1.44807 }, { "epoch": 2.49689387772589, "grad_norm": 0.7426810264587402, "learning_rate": 5.009758164722887e-05, "loss": 0.30419254302978516, "memory(GiB)": 70.5, "step": 58280, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.448075 }, { "epoch": 2.497108093055139, "grad_norm": 3.5431630611419678, "learning_rate": 5.0090851886135184e-05, "loss": 0.4205626010894775, "memory(GiB)": 70.5, "step": 58285, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.448071 }, { "epoch": 2.497322308384388, "grad_norm": 5.5162248611450195, "learning_rate": 5.0084122123395615e-05, "loss": 0.5568991661071777, "memory(GiB)": 70.5, "step": 58290, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.448081 }, { "epoch": 2.497536523713637, "grad_norm": 7.706772804260254, "learning_rate": 5.007739235913211e-05, "loss": 0.6286866188049316, "memory(GiB)": 70.5, "step": 58295, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.448093 }, { "epoch": 2.497750739042886, "grad_norm": 4.615623950958252, "learning_rate": 5.0070662593466564e-05, "loss": 0.278727388381958, "memory(GiB)": 70.5, "step": 58300, "token_acc": 0.9438596491228071, "train_speed(iter/s)": 1.448093 }, { "epoch": 2.4979649543721347, "grad_norm": 4.400970458984375, "learning_rate": 5.00639328265209e-05, "loss": 0.3347862482070923, "memory(GiB)": 70.5, "step": 58305, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.448106 }, { "epoch": 2.498179169701384, "grad_norm": 7.017396450042725, "learning_rate": 5.0057203058417044e-05, "loss": 0.38214712142944335, "memory(GiB)": 70.5, "step": 58310, "token_acc": 0.8955223880597015, "train_speed(iter/s)": 1.448125 }, { "epoch": 2.498393385030633, "grad_norm": 3.0798943042755127, "learning_rate": 5.00504732892769e-05, "loss": 0.35723731517791746, "memory(GiB)": 70.5, "step": 58315, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.448129 }, { "epoch": 2.4986076003598816, "grad_norm": 2.596384286880493, "learning_rate": 5.004374351922236e-05, "loss": 0.37736010551452637, "memory(GiB)": 70.5, "step": 58320, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.44814 }, { "epoch": 2.498821815689131, "grad_norm": 7.9264631271362305, "learning_rate": 5.003701374837539e-05, "loss": 0.4606457710266113, "memory(GiB)": 70.5, "step": 58325, "token_acc": 0.9127725856697819, "train_speed(iter/s)": 1.448137 }, { "epoch": 2.4990360310183797, "grad_norm": 1.3245102167129517, "learning_rate": 5.003028397685787e-05, "loss": 0.27433998584747316, "memory(GiB)": 70.5, "step": 58330, "token_acc": 0.9356223175965666, "train_speed(iter/s)": 1.448144 }, { "epoch": 2.4992502463476285, "grad_norm": 1.7104305028915405, "learning_rate": 5.0023554204791724e-05, "loss": 0.40608820915222166, "memory(GiB)": 70.5, "step": 58335, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.448157 }, { "epoch": 2.4994644616768777, "grad_norm": 1.575256109237671, "learning_rate": 5.001682443229888e-05, "loss": 0.5543204784393311, "memory(GiB)": 70.5, "step": 58340, "token_acc": 0.8895705521472392, "train_speed(iter/s)": 1.448165 }, { "epoch": 2.4996786770061266, "grad_norm": 5.293137073516846, "learning_rate": 5.001009465950125e-05, "loss": 0.31023263931274414, "memory(GiB)": 70.5, "step": 58345, "token_acc": 0.934375, "train_speed(iter/s)": 1.448162 }, { "epoch": 2.4998928923353754, "grad_norm": 3.4097847938537598, "learning_rate": 5.000336488652074e-05, "loss": 0.5272829055786132, "memory(GiB)": 70.5, "step": 58350, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.448164 }, { "epoch": 2.5001071076646246, "grad_norm": 3.1294682025909424, "learning_rate": 4.9996635113479265e-05, "loss": 0.405879545211792, "memory(GiB)": 70.5, "step": 58355, "token_acc": 0.896797153024911, "train_speed(iter/s)": 1.448168 }, { "epoch": 2.5003213229938734, "grad_norm": 3.038503646850586, "learning_rate": 4.9989905340498755e-05, "loss": 0.22734394073486328, "memory(GiB)": 70.5, "step": 58360, "token_acc": 0.9397590361445783, "train_speed(iter/s)": 1.448167 }, { "epoch": 2.5005355383231223, "grad_norm": 2.4757087230682373, "learning_rate": 4.998317556770111e-05, "loss": 0.5674033164978027, "memory(GiB)": 70.5, "step": 58365, "token_acc": 0.8807017543859649, "train_speed(iter/s)": 1.448177 }, { "epoch": 2.5007497536523715, "grad_norm": 2.4566822052001953, "learning_rate": 4.9976445795208274e-05, "loss": 0.19135011434555055, "memory(GiB)": 70.5, "step": 58370, "token_acc": 0.9588014981273408, "train_speed(iter/s)": 1.448171 }, { "epoch": 2.5009639689816203, "grad_norm": 1.818909764289856, "learning_rate": 4.996971602314214e-05, "loss": 0.17835628986358643, "memory(GiB)": 70.5, "step": 58375, "token_acc": 0.95625, "train_speed(iter/s)": 1.448179 }, { "epoch": 2.501178184310869, "grad_norm": 4.807403564453125, "learning_rate": 4.996298625162462e-05, "loss": 0.43394861221313474, "memory(GiB)": 70.5, "step": 58380, "token_acc": 0.8898071625344353, "train_speed(iter/s)": 1.44819 }, { "epoch": 2.5013923996401184, "grad_norm": 0.9165571928024292, "learning_rate": 4.9956256480777643e-05, "loss": 0.3144109010696411, "memory(GiB)": 70.5, "step": 58385, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.44819 }, { "epoch": 2.501606614969367, "grad_norm": 4.4611897468566895, "learning_rate": 4.994952671072312e-05, "loss": 0.2970869779586792, "memory(GiB)": 70.5, "step": 58390, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.448205 }, { "epoch": 2.501820830298616, "grad_norm": 2.220470666885376, "learning_rate": 4.994279694158297e-05, "loss": 0.504002046585083, "memory(GiB)": 70.5, "step": 58395, "token_acc": 0.8653846153846154, "train_speed(iter/s)": 1.448204 }, { "epoch": 2.5020350456278653, "grad_norm": 6.326952934265137, "learning_rate": 4.993606717347911e-05, "loss": 0.533213758468628, "memory(GiB)": 70.5, "step": 58400, "token_acc": 0.8877887788778878, "train_speed(iter/s)": 1.448217 }, { "epoch": 2.502249260957114, "grad_norm": 3.6657912731170654, "learning_rate": 4.992933740653345e-05, "loss": 0.2991269588470459, "memory(GiB)": 70.5, "step": 58405, "token_acc": 0.9380165289256198, "train_speed(iter/s)": 1.448222 }, { "epoch": 2.502463476286363, "grad_norm": 3.5209083557128906, "learning_rate": 4.99226076408679e-05, "loss": 0.39585447311401367, "memory(GiB)": 70.5, "step": 58410, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.448234 }, { "epoch": 2.502677691615612, "grad_norm": 2.3035995960235596, "learning_rate": 4.99158778766044e-05, "loss": 0.27551932334899903, "memory(GiB)": 70.5, "step": 58415, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.448228 }, { "epoch": 2.502891906944861, "grad_norm": 8.118408203125, "learning_rate": 4.990914811386484e-05, "loss": 0.4993506908416748, "memory(GiB)": 70.5, "step": 58420, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.448236 }, { "epoch": 2.50310612227411, "grad_norm": 0.4034695029258728, "learning_rate": 4.990241835277116e-05, "loss": 0.14012649059295654, "memory(GiB)": 70.5, "step": 58425, "token_acc": 0.9601449275362319, "train_speed(iter/s)": 1.448244 }, { "epoch": 2.503320337603359, "grad_norm": 2.0542659759521484, "learning_rate": 4.989568859344523e-05, "loss": 0.4128554821014404, "memory(GiB)": 70.5, "step": 58430, "token_acc": 0.91015625, "train_speed(iter/s)": 1.448251 }, { "epoch": 2.503534552932608, "grad_norm": 1.44866943359375, "learning_rate": 4.988895883600902e-05, "loss": 0.4091980457305908, "memory(GiB)": 70.5, "step": 58435, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.448256 }, { "epoch": 2.5037487682618567, "grad_norm": 2.3448643684387207, "learning_rate": 4.988222908058443e-05, "loss": 0.4954817295074463, "memory(GiB)": 70.5, "step": 58440, "token_acc": 0.8970099667774086, "train_speed(iter/s)": 1.448248 }, { "epoch": 2.503962983591106, "grad_norm": 3.786768913269043, "learning_rate": 4.987549932729337e-05, "loss": 0.3222055912017822, "memory(GiB)": 70.5, "step": 58445, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.448257 }, { "epoch": 2.5041771989203547, "grad_norm": 2.693272590637207, "learning_rate": 4.986876957625776e-05, "loss": 0.2051058292388916, "memory(GiB)": 70.5, "step": 58450, "token_acc": 0.9465020576131687, "train_speed(iter/s)": 1.448258 }, { "epoch": 2.5043914142496035, "grad_norm": 2.4201745986938477, "learning_rate": 4.98620398275995e-05, "loss": 0.3882559061050415, "memory(GiB)": 70.5, "step": 58455, "token_acc": 0.9075907590759076, "train_speed(iter/s)": 1.448263 }, { "epoch": 2.504605629578853, "grad_norm": 4.781452655792236, "learning_rate": 4.9855310081440535e-05, "loss": 0.3230464935302734, "memory(GiB)": 70.5, "step": 58460, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.448268 }, { "epoch": 2.5048198449081016, "grad_norm": 5.058071136474609, "learning_rate": 4.984858033790275e-05, "loss": 0.2954325437545776, "memory(GiB)": 70.5, "step": 58465, "token_acc": 0.9475409836065574, "train_speed(iter/s)": 1.448272 }, { "epoch": 2.5050340602373504, "grad_norm": 3.1975784301757812, "learning_rate": 4.984185059710808e-05, "loss": 0.32850806713104247, "memory(GiB)": 70.5, "step": 58470, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.448272 }, { "epoch": 2.5052482755665997, "grad_norm": 1.7963240146636963, "learning_rate": 4.983512085917843e-05, "loss": 0.29418106079101564, "memory(GiB)": 70.5, "step": 58475, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.448276 }, { "epoch": 2.5054624908958485, "grad_norm": 1.9475735425949097, "learning_rate": 4.9828391124235726e-05, "loss": 0.5832791328430176, "memory(GiB)": 70.5, "step": 58480, "token_acc": 0.8935483870967742, "train_speed(iter/s)": 1.448273 }, { "epoch": 2.5056767062250973, "grad_norm": 4.075699329376221, "learning_rate": 4.982166139240188e-05, "loss": 0.5196787357330322, "memory(GiB)": 70.5, "step": 58485, "token_acc": 0.898360655737705, "train_speed(iter/s)": 1.448282 }, { "epoch": 2.5058909215543466, "grad_norm": 2.09134840965271, "learning_rate": 4.98149316637988e-05, "loss": 0.18351829051971436, "memory(GiB)": 70.5, "step": 58490, "token_acc": 0.9695945945945946, "train_speed(iter/s)": 1.448282 }, { "epoch": 2.5061051368835954, "grad_norm": 3.6568830013275146, "learning_rate": 4.9808201938548404e-05, "loss": 0.2519824981689453, "memory(GiB)": 70.5, "step": 58495, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.448293 }, { "epoch": 2.506319352212844, "grad_norm": 2.3840742111206055, "learning_rate": 4.980147221677262e-05, "loss": 0.3262258768081665, "memory(GiB)": 70.5, "step": 58500, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.448295 }, { "epoch": 2.506319352212844, "eval_loss": 2.7483203411102295, "eval_runtime": 13.677, "eval_samples_per_second": 7.312, "eval_steps_per_second": 7.312, "eval_token_acc": 0.4309559939301973, "step": 58500 }, { "epoch": 2.5065335675420934, "grad_norm": 1.4123412370681763, "learning_rate": 4.979474249859333e-05, "loss": 0.2034893035888672, "memory(GiB)": 70.5, "step": 58505, "token_acc": 0.5931321540062435, "train_speed(iter/s)": 1.447772 }, { "epoch": 2.5067477828713423, "grad_norm": 2.4410712718963623, "learning_rate": 4.978801278413248e-05, "loss": 0.5382270336151123, "memory(GiB)": 70.5, "step": 58510, "token_acc": 0.8793103448275862, "train_speed(iter/s)": 1.447783 }, { "epoch": 2.506961998200591, "grad_norm": 6.281691074371338, "learning_rate": 4.9781283073511995e-05, "loss": 0.5185661315917969, "memory(GiB)": 70.5, "step": 58515, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.447787 }, { "epoch": 2.5071762135298403, "grad_norm": 2.3794264793395996, "learning_rate": 4.977455336685376e-05, "loss": 0.10395948886871338, "memory(GiB)": 70.5, "step": 58520, "token_acc": 0.976271186440678, "train_speed(iter/s)": 1.447787 }, { "epoch": 2.507390428859089, "grad_norm": 1.4468313455581665, "learning_rate": 4.97678236642797e-05, "loss": 0.06567342877388001, "memory(GiB)": 70.5, "step": 58525, "token_acc": 0.9879154078549849, "train_speed(iter/s)": 1.447796 }, { "epoch": 2.507604644188338, "grad_norm": 4.0089216232299805, "learning_rate": 4.9761093965911746e-05, "loss": 0.4924482345581055, "memory(GiB)": 70.5, "step": 58530, "token_acc": 0.9014084507042254, "train_speed(iter/s)": 1.447793 }, { "epoch": 2.507818859517587, "grad_norm": 1.6455928087234497, "learning_rate": 4.975436427187178e-05, "loss": 0.35918564796447755, "memory(GiB)": 70.5, "step": 58535, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.447794 }, { "epoch": 2.508033074846836, "grad_norm": 3.1892027854919434, "learning_rate": 4.974763458228175e-05, "loss": 0.664127779006958, "memory(GiB)": 70.5, "step": 58540, "token_acc": 0.8798701298701299, "train_speed(iter/s)": 1.447807 }, { "epoch": 2.508247290176085, "grad_norm": 4.029596328735352, "learning_rate": 4.9740904897263554e-05, "loss": 0.27950119972229004, "memory(GiB)": 70.5, "step": 58545, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.447806 }, { "epoch": 2.508461505505334, "grad_norm": 2.1736371517181396, "learning_rate": 4.97341752169391e-05, "loss": 0.42523980140686035, "memory(GiB)": 70.5, "step": 58550, "token_acc": 0.91015625, "train_speed(iter/s)": 1.447807 }, { "epoch": 2.508675720834583, "grad_norm": 1.1972453594207764, "learning_rate": 4.972744554143032e-05, "loss": 0.4808037757873535, "memory(GiB)": 70.5, "step": 58555, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.447819 }, { "epoch": 2.5088899361638317, "grad_norm": 2.3787925243377686, "learning_rate": 4.9720715870859124e-05, "loss": 0.5269412994384766, "memory(GiB)": 70.5, "step": 58560, "token_acc": 0.9098039215686274, "train_speed(iter/s)": 1.447827 }, { "epoch": 2.509104151493081, "grad_norm": 4.575430393218994, "learning_rate": 4.971398620534741e-05, "loss": 0.27805418968200685, "memory(GiB)": 70.5, "step": 58565, "token_acc": 0.9330543933054394, "train_speed(iter/s)": 1.447835 }, { "epoch": 2.50931836682233, "grad_norm": 2.0042223930358887, "learning_rate": 4.970725654501712e-05, "loss": 0.3376027822494507, "memory(GiB)": 70.5, "step": 58570, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.447832 }, { "epoch": 2.5095325821515786, "grad_norm": 4.169680595397949, "learning_rate": 4.970052688999012e-05, "loss": 0.22165563106536865, "memory(GiB)": 70.5, "step": 58575, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.447828 }, { "epoch": 2.509746797480828, "grad_norm": 4.801746368408203, "learning_rate": 4.969379724038837e-05, "loss": 0.3306748867034912, "memory(GiB)": 70.5, "step": 58580, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.447824 }, { "epoch": 2.5099610128100767, "grad_norm": 5.458007335662842, "learning_rate": 4.9687067596333784e-05, "loss": 0.2910116195678711, "memory(GiB)": 70.5, "step": 58585, "token_acc": 0.9348659003831418, "train_speed(iter/s)": 1.447827 }, { "epoch": 2.5101752281393255, "grad_norm": 3.47196102142334, "learning_rate": 4.968033795794827e-05, "loss": 0.6117453575134277, "memory(GiB)": 70.5, "step": 58590, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.447841 }, { "epoch": 2.5103894434685747, "grad_norm": 0.6650385856628418, "learning_rate": 4.967360832535371e-05, "loss": 0.21337389945983887, "memory(GiB)": 70.5, "step": 58595, "token_acc": 0.9407894736842105, "train_speed(iter/s)": 1.447832 }, { "epoch": 2.5106036587978235, "grad_norm": 3.8012688159942627, "learning_rate": 4.9666878698672054e-05, "loss": 0.2890504837036133, "memory(GiB)": 70.5, "step": 58600, "token_acc": 0.9339622641509434, "train_speed(iter/s)": 1.44784 }, { "epoch": 2.5108178741270724, "grad_norm": 4.200118541717529, "learning_rate": 4.96601490780252e-05, "loss": 0.372768759727478, "memory(GiB)": 70.5, "step": 58605, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.447839 }, { "epoch": 2.5110320894563216, "grad_norm": 4.129665374755859, "learning_rate": 4.965341946353506e-05, "loss": 0.4233363628387451, "memory(GiB)": 70.5, "step": 58610, "token_acc": 0.9033816425120773, "train_speed(iter/s)": 1.447857 }, { "epoch": 2.5112463047855704, "grad_norm": 2.7676234245300293, "learning_rate": 4.964668985532357e-05, "loss": 0.2722019672393799, "memory(GiB)": 70.5, "step": 58615, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.447859 }, { "epoch": 2.5114605201148192, "grad_norm": 0.7598388195037842, "learning_rate": 4.96399602535126e-05, "loss": 0.4779960155487061, "memory(GiB)": 70.5, "step": 58620, "token_acc": 0.9044943820224719, "train_speed(iter/s)": 1.447867 }, { "epoch": 2.5116747354440685, "grad_norm": 5.265523433685303, "learning_rate": 4.9633230658224106e-05, "loss": 0.4203805923461914, "memory(GiB)": 70.5, "step": 58625, "token_acc": 0.9068825910931174, "train_speed(iter/s)": 1.447871 }, { "epoch": 2.5118889507733173, "grad_norm": 2.3079326152801514, "learning_rate": 4.9626501069579975e-05, "loss": 0.3628975868225098, "memory(GiB)": 70.5, "step": 58630, "token_acc": 0.9288389513108615, "train_speed(iter/s)": 1.44787 }, { "epoch": 2.512103166102566, "grad_norm": 4.5716705322265625, "learning_rate": 4.961977148770212e-05, "loss": 0.27283854484558107, "memory(GiB)": 70.5, "step": 58635, "token_acc": 0.93359375, "train_speed(iter/s)": 1.447887 }, { "epoch": 2.5123173814318154, "grad_norm": 1.9725183248519897, "learning_rate": 4.961304191271248e-05, "loss": 0.408354663848877, "memory(GiB)": 70.5, "step": 58640, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.447887 }, { "epoch": 2.512531596761064, "grad_norm": 5.859982967376709, "learning_rate": 4.960631234473294e-05, "loss": 0.401885986328125, "memory(GiB)": 70.5, "step": 58645, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.447896 }, { "epoch": 2.512745812090313, "grad_norm": 2.2652432918548584, "learning_rate": 4.9599582783885404e-05, "loss": 0.4647829055786133, "memory(GiB)": 70.5, "step": 58650, "token_acc": 0.8960573476702509, "train_speed(iter/s)": 1.447902 }, { "epoch": 2.5129600274195623, "grad_norm": 3.4274375438690186, "learning_rate": 4.959285323029181e-05, "loss": 0.3295647144317627, "memory(GiB)": 70.5, "step": 58655, "token_acc": 0.9180887372013652, "train_speed(iter/s)": 1.447909 }, { "epoch": 2.513174242748811, "grad_norm": 0.6767504215240479, "learning_rate": 4.958612368407407e-05, "loss": 0.35027987957000734, "memory(GiB)": 70.5, "step": 58660, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.447916 }, { "epoch": 2.51338845807806, "grad_norm": 6.250726699829102, "learning_rate": 4.957939414535409e-05, "loss": 0.37266619205474855, "memory(GiB)": 70.5, "step": 58665, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.447913 }, { "epoch": 2.513602673407309, "grad_norm": 3.255220413208008, "learning_rate": 4.9572664614253774e-05, "loss": 0.368271541595459, "memory(GiB)": 70.5, "step": 58670, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.447908 }, { "epoch": 2.513816888736558, "grad_norm": 0.28916800022125244, "learning_rate": 4.9565935090895046e-05, "loss": 0.5092487335205078, "memory(GiB)": 70.5, "step": 58675, "token_acc": 0.9063829787234042, "train_speed(iter/s)": 1.44792 }, { "epoch": 2.5140311040658068, "grad_norm": 4.301886081695557, "learning_rate": 4.95592055753998e-05, "loss": 0.6435919284820557, "memory(GiB)": 70.5, "step": 58680, "token_acc": 0.8803680981595092, "train_speed(iter/s)": 1.447905 }, { "epoch": 2.514245319395056, "grad_norm": 1.3764773607254028, "learning_rate": 4.9552476067889967e-05, "loss": 0.49982562065124514, "memory(GiB)": 70.5, "step": 58685, "token_acc": 0.8964497041420119, "train_speed(iter/s)": 1.447904 }, { "epoch": 2.514459534724305, "grad_norm": 4.101471900939941, "learning_rate": 4.954574656848745e-05, "loss": 0.7152890205383301, "memory(GiB)": 70.5, "step": 58690, "token_acc": 0.8370607028753994, "train_speed(iter/s)": 1.447904 }, { "epoch": 2.5146737500535536, "grad_norm": 3.0584850311279297, "learning_rate": 4.953901707731415e-05, "loss": 0.2525299072265625, "memory(GiB)": 70.5, "step": 58695, "token_acc": 0.9568527918781726, "train_speed(iter/s)": 1.447907 }, { "epoch": 2.514887965382803, "grad_norm": 2.7116949558258057, "learning_rate": 4.9532287594491996e-05, "loss": 0.2901188850402832, "memory(GiB)": 70.5, "step": 58700, "token_acc": 0.9375, "train_speed(iter/s)": 1.447904 }, { "epoch": 2.5151021807120517, "grad_norm": 1.9368447065353394, "learning_rate": 4.95255581201429e-05, "loss": 0.14266457557678222, "memory(GiB)": 70.5, "step": 58705, "token_acc": 0.9701986754966887, "train_speed(iter/s)": 1.447908 }, { "epoch": 2.5153163960413005, "grad_norm": 4.395059585571289, "learning_rate": 4.951882865438875e-05, "loss": 0.42879419326782225, "memory(GiB)": 70.5, "step": 58710, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.447925 }, { "epoch": 2.51553061137055, "grad_norm": 2.9210100173950195, "learning_rate": 4.951209919735148e-05, "loss": 0.34883530139923097, "memory(GiB)": 70.5, "step": 58715, "token_acc": 0.9300291545189504, "train_speed(iter/s)": 1.447935 }, { "epoch": 2.5157448266997986, "grad_norm": 2.3298914432525635, "learning_rate": 4.9505369749152975e-05, "loss": 0.4781391143798828, "memory(GiB)": 70.5, "step": 58720, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.44795 }, { "epoch": 2.5159590420290474, "grad_norm": 3.896411418914795, "learning_rate": 4.9498640309915156e-05, "loss": 0.2648519515991211, "memory(GiB)": 70.5, "step": 58725, "token_acc": 0.9322709163346613, "train_speed(iter/s)": 1.447955 }, { "epoch": 2.5161732573582967, "grad_norm": 2.6048104763031006, "learning_rate": 4.9491910879759956e-05, "loss": 0.6436981201171875, "memory(GiB)": 70.5, "step": 58730, "token_acc": 0.8508474576271187, "train_speed(iter/s)": 1.447952 }, { "epoch": 2.5163874726875455, "grad_norm": 3.54612398147583, "learning_rate": 4.9485181458809273e-05, "loss": 0.5082064628601074, "memory(GiB)": 70.5, "step": 58735, "token_acc": 0.8909774436090225, "train_speed(iter/s)": 1.447962 }, { "epoch": 2.5166016880167943, "grad_norm": 0.2870037257671356, "learning_rate": 4.947845204718499e-05, "loss": 0.15071842670440674, "memory(GiB)": 70.5, "step": 58740, "token_acc": 0.959731543624161, "train_speed(iter/s)": 1.447956 }, { "epoch": 2.5168159033460435, "grad_norm": 2.7171897888183594, "learning_rate": 4.947172264500905e-05, "loss": 0.2913355350494385, "memory(GiB)": 70.5, "step": 58745, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.44796 }, { "epoch": 2.5170301186752924, "grad_norm": 3.3847708702087402, "learning_rate": 4.9464993252403366e-05, "loss": 0.5615349769592285, "memory(GiB)": 70.5, "step": 58750, "token_acc": 0.8892988929889298, "train_speed(iter/s)": 1.447967 }, { "epoch": 2.517244334004541, "grad_norm": 2.961794137954712, "learning_rate": 4.945826386948981e-05, "loss": 0.37916734218597414, "memory(GiB)": 70.5, "step": 58755, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.447977 }, { "epoch": 2.5174585493337904, "grad_norm": 2.222403049468994, "learning_rate": 4.9451534496390315e-05, "loss": 0.25006699562072754, "memory(GiB)": 70.5, "step": 58760, "token_acc": 0.9304347826086956, "train_speed(iter/s)": 1.447975 }, { "epoch": 2.5176727646630392, "grad_norm": 1.989259958267212, "learning_rate": 4.944480513322678e-05, "loss": 0.27332708835601804, "memory(GiB)": 70.5, "step": 58765, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.44799 }, { "epoch": 2.517886979992288, "grad_norm": 1.2248071432113647, "learning_rate": 4.943807578012114e-05, "loss": 0.34197134971618653, "memory(GiB)": 70.5, "step": 58770, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.447999 }, { "epoch": 2.5181011953215373, "grad_norm": 3.228273868560791, "learning_rate": 4.943134643719528e-05, "loss": 0.39010910987854003, "memory(GiB)": 70.5, "step": 58775, "token_acc": 0.9543859649122807, "train_speed(iter/s)": 1.447999 }, { "epoch": 2.518315410650786, "grad_norm": 2.630650758743286, "learning_rate": 4.94246171045711e-05, "loss": 0.25847930908203126, "memory(GiB)": 70.5, "step": 58780, "token_acc": 0.9510204081632653, "train_speed(iter/s)": 1.448 }, { "epoch": 2.518529625980035, "grad_norm": 2.851236581802368, "learning_rate": 4.941788778237053e-05, "loss": 0.3757209777832031, "memory(GiB)": 70.5, "step": 58785, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.448001 }, { "epoch": 2.518743841309284, "grad_norm": 1.6079667806625366, "learning_rate": 4.941115847071547e-05, "loss": 0.30015885829925537, "memory(GiB)": 70.5, "step": 58790, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.448017 }, { "epoch": 2.518958056638533, "grad_norm": 1.5499424934387207, "learning_rate": 4.9404429169727815e-05, "loss": 0.4352890968322754, "memory(GiB)": 70.5, "step": 58795, "token_acc": 0.9205776173285198, "train_speed(iter/s)": 1.448017 }, { "epoch": 2.519172271967782, "grad_norm": 4.70640754699707, "learning_rate": 4.9397699879529475e-05, "loss": 0.3444368362426758, "memory(GiB)": 70.5, "step": 58800, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.448029 }, { "epoch": 2.519386487297031, "grad_norm": 3.467071533203125, "learning_rate": 4.939097060024239e-05, "loss": 0.2430823564529419, "memory(GiB)": 70.5, "step": 58805, "token_acc": 0.9480122324159022, "train_speed(iter/s)": 1.448027 }, { "epoch": 2.51960070262628, "grad_norm": 0.7759162783622742, "learning_rate": 4.9384241331988444e-05, "loss": 0.39481167793273925, "memory(GiB)": 70.5, "step": 58810, "token_acc": 0.9022556390977443, "train_speed(iter/s)": 1.448029 }, { "epoch": 2.5198149179555287, "grad_norm": 3.7817556858062744, "learning_rate": 4.937751207488953e-05, "loss": 0.3152715444564819, "memory(GiB)": 70.5, "step": 58815, "token_acc": 0.93125, "train_speed(iter/s)": 1.448035 }, { "epoch": 2.520029133284778, "grad_norm": 2.790290355682373, "learning_rate": 4.937078282906758e-05, "loss": 0.22903094291687012, "memory(GiB)": 70.5, "step": 58820, "token_acc": 0.9476923076923077, "train_speed(iter/s)": 1.448047 }, { "epoch": 2.5202433486140268, "grad_norm": 3.4690370559692383, "learning_rate": 4.936405359464448e-05, "loss": 0.15675041675567628, "memory(GiB)": 70.5, "step": 58825, "token_acc": 0.9507575757575758, "train_speed(iter/s)": 1.448055 }, { "epoch": 2.5204575639432756, "grad_norm": 2.3843865394592285, "learning_rate": 4.9357324371742156e-05, "loss": 0.32299563884735105, "memory(GiB)": 70.5, "step": 58830, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.448059 }, { "epoch": 2.520671779272525, "grad_norm": 3.646042823791504, "learning_rate": 4.93505951604825e-05, "loss": 0.31419057846069337, "memory(GiB)": 70.5, "step": 58835, "token_acc": 0.9398496240601504, "train_speed(iter/s)": 1.448059 }, { "epoch": 2.5208859946017736, "grad_norm": 4.305171489715576, "learning_rate": 4.9343865960987425e-05, "loss": 0.7653546333312988, "memory(GiB)": 70.5, "step": 58840, "token_acc": 0.8609022556390977, "train_speed(iter/s)": 1.448082 }, { "epoch": 2.5211002099310225, "grad_norm": 3.3429465293884277, "learning_rate": 4.9337136773378836e-05, "loss": 0.5332307815551758, "memory(GiB)": 70.5, "step": 58845, "token_acc": 0.8862068965517241, "train_speed(iter/s)": 1.448078 }, { "epoch": 2.5213144252602717, "grad_norm": 4.075344085693359, "learning_rate": 4.933040759777863e-05, "loss": 0.251200795173645, "memory(GiB)": 70.5, "step": 58850, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.448076 }, { "epoch": 2.5215286405895205, "grad_norm": 3.369629144668579, "learning_rate": 4.9323678434308713e-05, "loss": 0.32784082889556887, "memory(GiB)": 70.5, "step": 58855, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.44808 }, { "epoch": 2.5217428559187693, "grad_norm": 1.806627869606018, "learning_rate": 4.9316949283091004e-05, "loss": 0.16780433654785157, "memory(GiB)": 70.5, "step": 58860, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.448088 }, { "epoch": 2.5219570712480186, "grad_norm": 2.4678027629852295, "learning_rate": 4.9310220144247396e-05, "loss": 0.2607236623764038, "memory(GiB)": 70.5, "step": 58865, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.448093 }, { "epoch": 2.5221712865772674, "grad_norm": 1.4347301721572876, "learning_rate": 4.930349101789978e-05, "loss": 0.45713090896606445, "memory(GiB)": 70.5, "step": 58870, "token_acc": 0.898876404494382, "train_speed(iter/s)": 1.448097 }, { "epoch": 2.5223855019065162, "grad_norm": 0.05418318510055542, "learning_rate": 4.9296761904170095e-05, "loss": 0.35408523082733157, "memory(GiB)": 70.5, "step": 58875, "token_acc": 0.9318885448916409, "train_speed(iter/s)": 1.448095 }, { "epoch": 2.5225997172357655, "grad_norm": 6.982428073883057, "learning_rate": 4.9290032803180234e-05, "loss": 0.49468564987182617, "memory(GiB)": 70.5, "step": 58880, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.448093 }, { "epoch": 2.5228139325650143, "grad_norm": 3.5002925395965576, "learning_rate": 4.928330371505208e-05, "loss": 0.3986830472946167, "memory(GiB)": 70.5, "step": 58885, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.448093 }, { "epoch": 2.523028147894263, "grad_norm": 4.23842716217041, "learning_rate": 4.927657463990757e-05, "loss": 0.4949055194854736, "memory(GiB)": 70.5, "step": 58890, "token_acc": 0.8933823529411765, "train_speed(iter/s)": 1.448099 }, { "epoch": 2.5232423632235124, "grad_norm": 5.50463342666626, "learning_rate": 4.926984557786859e-05, "loss": 0.634669303894043, "memory(GiB)": 70.5, "step": 58895, "token_acc": 0.8576271186440678, "train_speed(iter/s)": 1.448112 }, { "epoch": 2.523456578552761, "grad_norm": 3.6052258014678955, "learning_rate": 4.926311652905702e-05, "loss": 0.21661653518676757, "memory(GiB)": 70.5, "step": 58900, "token_acc": 0.96, "train_speed(iter/s)": 1.448114 }, { "epoch": 2.52367079388201, "grad_norm": 3.627699136734009, "learning_rate": 4.9256387493594805e-05, "loss": 0.3708128213882446, "memory(GiB)": 70.5, "step": 58905, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.448107 }, { "epoch": 2.5238850092112592, "grad_norm": 6.0818705558776855, "learning_rate": 4.924965847160381e-05, "loss": 0.6890861511230468, "memory(GiB)": 70.5, "step": 58910, "token_acc": 0.8540145985401459, "train_speed(iter/s)": 1.448123 }, { "epoch": 2.524099224540508, "grad_norm": 3.2669668197631836, "learning_rate": 4.924292946320597e-05, "loss": 0.41802067756652833, "memory(GiB)": 70.5, "step": 58915, "token_acc": 0.91, "train_speed(iter/s)": 1.448122 }, { "epoch": 2.524313439869757, "grad_norm": 3.162588596343994, "learning_rate": 4.923620046852318e-05, "loss": 0.3112445831298828, "memory(GiB)": 70.5, "step": 58920, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.44812 }, { "epoch": 2.524527655199006, "grad_norm": 2.4474568367004395, "learning_rate": 4.9229471487677316e-05, "loss": 0.37276554107666016, "memory(GiB)": 70.5, "step": 58925, "token_acc": 0.932, "train_speed(iter/s)": 1.448125 }, { "epoch": 2.524741870528255, "grad_norm": 4.540578365325928, "learning_rate": 4.9222742520790314e-05, "loss": 0.4140131950378418, "memory(GiB)": 70.5, "step": 58930, "token_acc": 0.9174041297935103, "train_speed(iter/s)": 1.448128 }, { "epoch": 2.5249560858575038, "grad_norm": 2.1798505783081055, "learning_rate": 4.9216013567984055e-05, "loss": 0.16476099491119384, "memory(GiB)": 70.5, "step": 58935, "token_acc": 0.9633333333333334, "train_speed(iter/s)": 1.44813 }, { "epoch": 2.525170301186753, "grad_norm": 3.962891101837158, "learning_rate": 4.920928462938044e-05, "loss": 0.4475104808807373, "memory(GiB)": 70.5, "step": 58940, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.44814 }, { "epoch": 2.525384516516002, "grad_norm": 5.507608890533447, "learning_rate": 4.9202555705101355e-05, "loss": 0.44591064453125, "memory(GiB)": 70.5, "step": 58945, "token_acc": 0.8896103896103896, "train_speed(iter/s)": 1.448141 }, { "epoch": 2.5255987318452506, "grad_norm": 9.782048225402832, "learning_rate": 4.919582679526875e-05, "loss": 0.5619991779327392, "memory(GiB)": 70.5, "step": 58950, "token_acc": 0.8844884488448845, "train_speed(iter/s)": 1.44814 }, { "epoch": 2.5258129471745, "grad_norm": 4.097883701324463, "learning_rate": 4.918909790000449e-05, "loss": 0.6445967197418213, "memory(GiB)": 70.5, "step": 58955, "token_acc": 0.8664596273291926, "train_speed(iter/s)": 1.44814 }, { "epoch": 2.5260271625037487, "grad_norm": 3.3517541885375977, "learning_rate": 4.918236901943049e-05, "loss": 0.38232927322387694, "memory(GiB)": 70.5, "step": 58960, "token_acc": 0.9118773946360154, "train_speed(iter/s)": 1.448134 }, { "epoch": 2.5262413778329975, "grad_norm": 3.197361707687378, "learning_rate": 4.917564015366864e-05, "loss": 0.3969565868377686, "memory(GiB)": 70.5, "step": 58965, "token_acc": 0.8908450704225352, "train_speed(iter/s)": 1.448133 }, { "epoch": 2.5264555931622468, "grad_norm": 0.5433465838432312, "learning_rate": 4.9168911302840834e-05, "loss": 0.22356815338134767, "memory(GiB)": 70.5, "step": 58970, "token_acc": 0.9439655172413793, "train_speed(iter/s)": 1.448142 }, { "epoch": 2.5266698084914956, "grad_norm": 2.85564923286438, "learning_rate": 4.916218246706899e-05, "loss": 0.3832036256790161, "memory(GiB)": 70.5, "step": 58975, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.448153 }, { "epoch": 2.5268840238207444, "grad_norm": 2.8711578845977783, "learning_rate": 4.9155453646475e-05, "loss": 0.4896231651306152, "memory(GiB)": 70.5, "step": 58980, "token_acc": 0.8952702702702703, "train_speed(iter/s)": 1.448151 }, { "epoch": 2.5270982391499937, "grad_norm": 1.5234606266021729, "learning_rate": 4.914872484118074e-05, "loss": 0.44480252265930176, "memory(GiB)": 70.5, "step": 58985, "token_acc": 0.9121338912133892, "train_speed(iter/s)": 1.448153 }, { "epoch": 2.5273124544792425, "grad_norm": 3.071272134780884, "learning_rate": 4.914199605130814e-05, "loss": 0.5242734432220459, "memory(GiB)": 70.5, "step": 58990, "token_acc": 0.9014925373134328, "train_speed(iter/s)": 1.448156 }, { "epoch": 2.5275266698084913, "grad_norm": 2.112407684326172, "learning_rate": 4.913526727697909e-05, "loss": 0.30191280841827395, "memory(GiB)": 70.5, "step": 58995, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.448161 }, { "epoch": 2.5277408851377405, "grad_norm": 4.005379676818848, "learning_rate": 4.912853851831547e-05, "loss": 0.4606758117675781, "memory(GiB)": 70.5, "step": 59000, "token_acc": 0.8941176470588236, "train_speed(iter/s)": 1.448172 }, { "epoch": 2.5277408851377405, "eval_loss": 2.4196512699127197, "eval_runtime": 13.3337, "eval_samples_per_second": 7.5, "eval_steps_per_second": 7.5, "eval_token_acc": 0.43239795918367346, "step": 59000 }, { "epoch": 2.5279551004669893, "grad_norm": 2.5390141010284424, "learning_rate": 4.9121809775439214e-05, "loss": 0.4370736122131348, "memory(GiB)": 70.5, "step": 59005, "token_acc": 0.545367717287488, "train_speed(iter/s)": 1.44765 }, { "epoch": 2.528169315796238, "grad_norm": 2.717944383621216, "learning_rate": 4.911508104847218e-05, "loss": 0.36449761390686036, "memory(GiB)": 70.5, "step": 59010, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.447651 }, { "epoch": 2.5283835311254874, "grad_norm": 2.189587354660034, "learning_rate": 4.910835233753629e-05, "loss": 0.4286686897277832, "memory(GiB)": 70.5, "step": 59015, "token_acc": 0.9031413612565445, "train_speed(iter/s)": 1.44765 }, { "epoch": 2.5285977464547362, "grad_norm": 2.4126620292663574, "learning_rate": 4.9101623642753416e-05, "loss": 0.64586501121521, "memory(GiB)": 70.5, "step": 59020, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.447662 }, { "epoch": 2.528811961783985, "grad_norm": 8.587518692016602, "learning_rate": 4.909489496424549e-05, "loss": 0.17564696073532104, "memory(GiB)": 70.5, "step": 59025, "token_acc": 0.9453781512605042, "train_speed(iter/s)": 1.447667 }, { "epoch": 2.5290261771132343, "grad_norm": 5.01985502243042, "learning_rate": 4.9088166302134385e-05, "loss": 0.6571852684020996, "memory(GiB)": 70.5, "step": 59030, "token_acc": 0.8547854785478548, "train_speed(iter/s)": 1.447667 }, { "epoch": 2.529240392442483, "grad_norm": 2.383903741836548, "learning_rate": 4.908143765654201e-05, "loss": 0.27752697467803955, "memory(GiB)": 70.5, "step": 59035, "token_acc": 0.945054945054945, "train_speed(iter/s)": 1.447668 }, { "epoch": 2.529454607771732, "grad_norm": 2.24817156791687, "learning_rate": 4.907470902759026e-05, "loss": 0.17322503328323363, "memory(GiB)": 70.5, "step": 59040, "token_acc": 0.9563492063492064, "train_speed(iter/s)": 1.447644 }, { "epoch": 2.529668823100981, "grad_norm": 4.541551113128662, "learning_rate": 4.906798041540101e-05, "loss": 0.6654854774475097, "memory(GiB)": 70.5, "step": 59045, "token_acc": 0.87890625, "train_speed(iter/s)": 1.447658 }, { "epoch": 2.52988303843023, "grad_norm": 2.1386265754699707, "learning_rate": 4.906125182009618e-05, "loss": 0.4121506690979004, "memory(GiB)": 70.5, "step": 59050, "token_acc": 0.8986928104575164, "train_speed(iter/s)": 1.447653 }, { "epoch": 2.530097253759479, "grad_norm": 2.4872655868530273, "learning_rate": 4.905452324179764e-05, "loss": 0.145536470413208, "memory(GiB)": 70.5, "step": 59055, "token_acc": 0.9765625, "train_speed(iter/s)": 1.447654 }, { "epoch": 2.530311469088728, "grad_norm": 3.2195944786071777, "learning_rate": 4.904779468062731e-05, "loss": 0.4111196517944336, "memory(GiB)": 70.5, "step": 59060, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.447669 }, { "epoch": 2.530525684417977, "grad_norm": 5.074983596801758, "learning_rate": 4.904106613670707e-05, "loss": 0.3333961725234985, "memory(GiB)": 70.5, "step": 59065, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.447666 }, { "epoch": 2.5307398997472257, "grad_norm": 1.6135588884353638, "learning_rate": 4.9034337610158815e-05, "loss": 0.19834024906158448, "memory(GiB)": 70.5, "step": 59070, "token_acc": 0.9629629629629629, "train_speed(iter/s)": 1.447679 }, { "epoch": 2.530954115076475, "grad_norm": 3.32763671875, "learning_rate": 4.9027609101104444e-05, "loss": 0.287885856628418, "memory(GiB)": 70.5, "step": 59075, "token_acc": 0.952, "train_speed(iter/s)": 1.447684 }, { "epoch": 2.5311683304057238, "grad_norm": 5.014883995056152, "learning_rate": 4.902088060966585e-05, "loss": 0.22803654670715331, "memory(GiB)": 70.5, "step": 59080, "token_acc": 0.9462915601023018, "train_speed(iter/s)": 1.447703 }, { "epoch": 2.5313825457349726, "grad_norm": 3.176724910736084, "learning_rate": 4.90141521359649e-05, "loss": 0.5737852573394775, "memory(GiB)": 70.5, "step": 59085, "token_acc": 0.8774928774928775, "train_speed(iter/s)": 1.447725 }, { "epoch": 2.531596761064222, "grad_norm": 2.0917797088623047, "learning_rate": 4.900742368012353e-05, "loss": 0.3293910980224609, "memory(GiB)": 70.5, "step": 59090, "token_acc": 0.9157088122605364, "train_speed(iter/s)": 1.44772 }, { "epoch": 2.5318109763934706, "grad_norm": 2.8381834030151367, "learning_rate": 4.9000695242263586e-05, "loss": 0.4488013744354248, "memory(GiB)": 70.5, "step": 59095, "token_acc": 0.8996960486322189, "train_speed(iter/s)": 1.447735 }, { "epoch": 2.5320251917227194, "grad_norm": 2.029451370239258, "learning_rate": 4.8993966822507006e-05, "loss": 0.23902637958526612, "memory(GiB)": 70.5, "step": 59100, "token_acc": 0.9598393574297188, "train_speed(iter/s)": 1.447733 }, { "epoch": 2.5322394070519687, "grad_norm": 6.024895668029785, "learning_rate": 4.898723842097566e-05, "loss": 0.4117762088775635, "memory(GiB)": 70.5, "step": 59105, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.447734 }, { "epoch": 2.5324536223812175, "grad_norm": 5.9803571701049805, "learning_rate": 4.898051003779144e-05, "loss": 0.47714786529541015, "memory(GiB)": 70.5, "step": 59110, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.447745 }, { "epoch": 2.5326678377104663, "grad_norm": 5.734964847564697, "learning_rate": 4.8973781673076236e-05, "loss": 0.3076761245727539, "memory(GiB)": 70.5, "step": 59115, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.447742 }, { "epoch": 2.5328820530397156, "grad_norm": 2.923720598220825, "learning_rate": 4.896705332695194e-05, "loss": 0.16880486011505128, "memory(GiB)": 70.5, "step": 59120, "token_acc": 0.9585987261146497, "train_speed(iter/s)": 1.447745 }, { "epoch": 2.5330962683689644, "grad_norm": 3.9532177448272705, "learning_rate": 4.896032499954045e-05, "loss": 0.2662095546722412, "memory(GiB)": 70.5, "step": 59125, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.447751 }, { "epoch": 2.533310483698213, "grad_norm": 3.533195734024048, "learning_rate": 4.895359669096363e-05, "loss": 0.6065425395965576, "memory(GiB)": 70.5, "step": 59130, "token_acc": 0.8700787401574803, "train_speed(iter/s)": 1.447769 }, { "epoch": 2.5335246990274625, "grad_norm": 3.8125109672546387, "learning_rate": 4.8946868401343404e-05, "loss": 0.3855398178100586, "memory(GiB)": 70.5, "step": 59135, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.447787 }, { "epoch": 2.5337389143567113, "grad_norm": 2.1476705074310303, "learning_rate": 4.894014013080163e-05, "loss": 0.24307565689086913, "memory(GiB)": 70.5, "step": 59140, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.447787 }, { "epoch": 2.53395312968596, "grad_norm": 3.232288122177124, "learning_rate": 4.893341187946022e-05, "loss": 0.3228936195373535, "memory(GiB)": 70.5, "step": 59145, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.447787 }, { "epoch": 2.5341673450152093, "grad_norm": 2.432175874710083, "learning_rate": 4.8926683647441065e-05, "loss": 0.1828052282333374, "memory(GiB)": 70.5, "step": 59150, "token_acc": 0.9675675675675676, "train_speed(iter/s)": 1.447797 }, { "epoch": 2.534381560344458, "grad_norm": 5.612984657287598, "learning_rate": 4.891995543486602e-05, "loss": 0.6008591651916504, "memory(GiB)": 70.5, "step": 59155, "token_acc": 0.8582089552238806, "train_speed(iter/s)": 1.447823 }, { "epoch": 2.534595775673707, "grad_norm": 5.478182792663574, "learning_rate": 4.8913227241857016e-05, "loss": 0.19764493703842162, "memory(GiB)": 70.5, "step": 59160, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.447826 }, { "epoch": 2.5348099910029562, "grad_norm": 1.0537967681884766, "learning_rate": 4.8906499068535917e-05, "loss": 0.5231286525726319, "memory(GiB)": 70.5, "step": 59165, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.447836 }, { "epoch": 2.535024206332205, "grad_norm": 3.067488670349121, "learning_rate": 4.889977091502459e-05, "loss": 0.357782769203186, "memory(GiB)": 70.5, "step": 59170, "token_acc": 0.900709219858156, "train_speed(iter/s)": 1.447846 }, { "epoch": 2.535238421661454, "grad_norm": 1.4149047136306763, "learning_rate": 4.889304278144495e-05, "loss": 0.10418986082077027, "memory(GiB)": 70.5, "step": 59175, "token_acc": 0.9735973597359736, "train_speed(iter/s)": 1.447843 }, { "epoch": 2.535452636990703, "grad_norm": 4.032512664794922, "learning_rate": 4.88863146679189e-05, "loss": 0.44106593132019045, "memory(GiB)": 70.5, "step": 59180, "token_acc": 0.9193083573487032, "train_speed(iter/s)": 1.447839 }, { "epoch": 2.535666852319952, "grad_norm": 3.6200549602508545, "learning_rate": 4.88795865745683e-05, "loss": 0.1590280771255493, "memory(GiB)": 70.5, "step": 59185, "token_acc": 0.9676258992805755, "train_speed(iter/s)": 1.447843 }, { "epoch": 2.5358810676492007, "grad_norm": 2.0708882808685303, "learning_rate": 4.887285850151503e-05, "loss": 0.5199456691741944, "memory(GiB)": 70.5, "step": 59190, "token_acc": 0.9027237354085603, "train_speed(iter/s)": 1.447857 }, { "epoch": 2.53609528297845, "grad_norm": 3.165858745574951, "learning_rate": 4.8866130448880995e-05, "loss": 0.3282433748245239, "memory(GiB)": 70.5, "step": 59195, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.447861 }, { "epoch": 2.536309498307699, "grad_norm": 3.77192759513855, "learning_rate": 4.885940241678806e-05, "loss": 0.3978853464126587, "memory(GiB)": 70.5, "step": 59200, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.447854 }, { "epoch": 2.5365237136369476, "grad_norm": 5.563136100769043, "learning_rate": 4.885267440535813e-05, "loss": 0.4899578094482422, "memory(GiB)": 70.5, "step": 59205, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.447869 }, { "epoch": 2.536737928966197, "grad_norm": 8.118234634399414, "learning_rate": 4.8845946414713076e-05, "loss": 0.5838555335998535, "memory(GiB)": 70.5, "step": 59210, "token_acc": 0.8690909090909091, "train_speed(iter/s)": 1.447865 }, { "epoch": 2.5369521442954457, "grad_norm": 0.3368057608604431, "learning_rate": 4.883921844497478e-05, "loss": 0.42038669586181643, "memory(GiB)": 70.5, "step": 59215, "token_acc": 0.921161825726141, "train_speed(iter/s)": 1.447866 }, { "epoch": 2.5371663596246945, "grad_norm": 0.454548716545105, "learning_rate": 4.883249049626514e-05, "loss": 0.17383211851119995, "memory(GiB)": 70.5, "step": 59220, "token_acc": 0.9527272727272728, "train_speed(iter/s)": 1.447873 }, { "epoch": 2.5373805749539438, "grad_norm": 3.5904698371887207, "learning_rate": 4.882576256870604e-05, "loss": 0.26928999423980715, "memory(GiB)": 70.5, "step": 59225, "token_acc": 0.9511400651465798, "train_speed(iter/s)": 1.447876 }, { "epoch": 2.5375947902831926, "grad_norm": 3.3946967124938965, "learning_rate": 4.8819034662419326e-05, "loss": 0.513791847229004, "memory(GiB)": 70.5, "step": 59230, "token_acc": 0.9003215434083601, "train_speed(iter/s)": 1.447894 }, { "epoch": 2.5378090056124414, "grad_norm": 3.8333775997161865, "learning_rate": 4.881230677752693e-05, "loss": 0.4037765026092529, "memory(GiB)": 70.5, "step": 59235, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.447891 }, { "epoch": 2.5380232209416906, "grad_norm": 2.9136526584625244, "learning_rate": 4.880557891415068e-05, "loss": 0.4130918025970459, "memory(GiB)": 70.5, "step": 59240, "token_acc": 0.896774193548387, "train_speed(iter/s)": 1.447913 }, { "epoch": 2.5382374362709395, "grad_norm": 1.0213927030563354, "learning_rate": 4.87988510724125e-05, "loss": 0.34643917083740233, "memory(GiB)": 70.5, "step": 59245, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.447927 }, { "epoch": 2.5384516516001883, "grad_norm": 7.9885053634643555, "learning_rate": 4.8792123252434276e-05, "loss": 0.26111769676208496, "memory(GiB)": 70.5, "step": 59250, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.447922 }, { "epoch": 2.5386658669294375, "grad_norm": 4.7541022300720215, "learning_rate": 4.878539545433787e-05, "loss": 0.3065299987792969, "memory(GiB)": 70.5, "step": 59255, "token_acc": 0.930835734870317, "train_speed(iter/s)": 1.447934 }, { "epoch": 2.5388800822586863, "grad_norm": 4.858361721038818, "learning_rate": 4.877866767824515e-05, "loss": 0.5336041450500488, "memory(GiB)": 70.5, "step": 59260, "token_acc": 0.8940809968847352, "train_speed(iter/s)": 1.447933 }, { "epoch": 2.539094297587935, "grad_norm": 7.071676731109619, "learning_rate": 4.877193992427803e-05, "loss": 0.5510545253753663, "memory(GiB)": 70.5, "step": 59265, "token_acc": 0.8606271777003485, "train_speed(iter/s)": 1.447931 }, { "epoch": 2.5393085129171844, "grad_norm": 3.0350918769836426, "learning_rate": 4.876521219255836e-05, "loss": 0.3652268171310425, "memory(GiB)": 70.5, "step": 59270, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.447937 }, { "epoch": 2.539522728246433, "grad_norm": 2.3156015872955322, "learning_rate": 4.8758484483208024e-05, "loss": 0.42890100479125975, "memory(GiB)": 70.5, "step": 59275, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.44794 }, { "epoch": 2.539736943575682, "grad_norm": 1.942608118057251, "learning_rate": 4.875175679634892e-05, "loss": 0.3514684200286865, "memory(GiB)": 70.5, "step": 59280, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.44795 }, { "epoch": 2.5399511589049313, "grad_norm": 6.506754398345947, "learning_rate": 4.87450291321029e-05, "loss": 0.5576425552368164, "memory(GiB)": 70.5, "step": 59285, "token_acc": 0.8804347826086957, "train_speed(iter/s)": 1.447987 }, { "epoch": 2.54016537423418, "grad_norm": 2.794224739074707, "learning_rate": 4.873830149059186e-05, "loss": 0.23311614990234375, "memory(GiB)": 70.5, "step": 59290, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.44799 }, { "epoch": 2.540379589563429, "grad_norm": 7.5165114402771, "learning_rate": 4.873157387193767e-05, "loss": 0.332576847076416, "memory(GiB)": 70.5, "step": 59295, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.447989 }, { "epoch": 2.540593804892678, "grad_norm": 2.47064208984375, "learning_rate": 4.872484627626221e-05, "loss": 0.216333270072937, "memory(GiB)": 70.5, "step": 59300, "token_acc": 0.9625, "train_speed(iter/s)": 1.447996 }, { "epoch": 2.540808020221927, "grad_norm": 0.882618248462677, "learning_rate": 4.871811870368736e-05, "loss": 0.28080775737762453, "memory(GiB)": 70.5, "step": 59305, "token_acc": 0.9429530201342282, "train_speed(iter/s)": 1.448001 }, { "epoch": 2.541022235551176, "grad_norm": 6.353883266448975, "learning_rate": 4.8711391154335e-05, "loss": 0.7432687759399415, "memory(GiB)": 70.5, "step": 59310, "token_acc": 0.8362573099415205, "train_speed(iter/s)": 1.448013 }, { "epoch": 2.541236450880425, "grad_norm": 3.1140832901000977, "learning_rate": 4.870466362832696e-05, "loss": 0.5825001716613769, "memory(GiB)": 70.5, "step": 59315, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.448025 }, { "epoch": 2.541450666209674, "grad_norm": 2.0459728240966797, "learning_rate": 4.869793612578518e-05, "loss": 0.6274656295776367, "memory(GiB)": 70.5, "step": 59320, "token_acc": 0.8636363636363636, "train_speed(iter/s)": 1.44804 }, { "epoch": 2.5416648815389227, "grad_norm": 2.4760591983795166, "learning_rate": 4.869120864683151e-05, "loss": 0.2313126802444458, "memory(GiB)": 70.5, "step": 59325, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.448037 }, { "epoch": 2.541879096868172, "grad_norm": 5.765625476837158, "learning_rate": 4.868448119158783e-05, "loss": 0.5057467460632324, "memory(GiB)": 70.5, "step": 59330, "token_acc": 0.8900709219858156, "train_speed(iter/s)": 1.448034 }, { "epoch": 2.5420933121974207, "grad_norm": 1.124779224395752, "learning_rate": 4.8677753760175995e-05, "loss": 0.2358386516571045, "memory(GiB)": 70.5, "step": 59335, "token_acc": 0.9596412556053812, "train_speed(iter/s)": 1.448052 }, { "epoch": 2.5423075275266696, "grad_norm": 1.5786714553833008, "learning_rate": 4.8671026352717895e-05, "loss": 0.2807002067565918, "memory(GiB)": 70.5, "step": 59340, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.448055 }, { "epoch": 2.542521742855919, "grad_norm": 5.257143497467041, "learning_rate": 4.866429896933539e-05, "loss": 0.43343257904052734, "memory(GiB)": 70.5, "step": 59345, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.448072 }, { "epoch": 2.5427359581851676, "grad_norm": 3.8641347885131836, "learning_rate": 4.865757161015038e-05, "loss": 0.5644326686859131, "memory(GiB)": 70.5, "step": 59350, "token_acc": 0.8668730650154799, "train_speed(iter/s)": 1.44808 }, { "epoch": 2.5429501735144164, "grad_norm": 2.4359049797058105, "learning_rate": 4.865084427528471e-05, "loss": 0.1649085760116577, "memory(GiB)": 70.5, "step": 59355, "token_acc": 0.958041958041958, "train_speed(iter/s)": 1.448085 }, { "epoch": 2.5431643888436657, "grad_norm": 4.000796318054199, "learning_rate": 4.864411696486026e-05, "loss": 0.5393502235412597, "memory(GiB)": 70.5, "step": 59360, "token_acc": 0.8825910931174089, "train_speed(iter/s)": 1.448086 }, { "epoch": 2.5433786041729145, "grad_norm": 7.503458499908447, "learning_rate": 4.863738967899891e-05, "loss": 0.39565396308898926, "memory(GiB)": 70.5, "step": 59365, "token_acc": 0.8843283582089553, "train_speed(iter/s)": 1.448102 }, { "epoch": 2.5435928195021633, "grad_norm": 3.3021175861358643, "learning_rate": 4.8630662417822526e-05, "loss": 0.42720942497253417, "memory(GiB)": 70.5, "step": 59370, "token_acc": 0.9158576051779935, "train_speed(iter/s)": 1.448107 }, { "epoch": 2.5438070348314126, "grad_norm": 5.499824523925781, "learning_rate": 4.8623935181452966e-05, "loss": 0.24418351650238038, "memory(GiB)": 70.5, "step": 59375, "token_acc": 0.9548872180451128, "train_speed(iter/s)": 1.448111 }, { "epoch": 2.5440212501606614, "grad_norm": 4.118307590484619, "learning_rate": 4.861720797001212e-05, "loss": 0.44422292709350586, "memory(GiB)": 70.5, "step": 59380, "token_acc": 0.9352941176470588, "train_speed(iter/s)": 1.44811 }, { "epoch": 2.54423546548991, "grad_norm": 3.3075711727142334, "learning_rate": 4.8610480783621835e-05, "loss": 0.6434619903564454, "memory(GiB)": 70.5, "step": 59385, "token_acc": 0.8434504792332268, "train_speed(iter/s)": 1.44811 }, { "epoch": 2.5444496808191595, "grad_norm": 5.194568634033203, "learning_rate": 4.860375362240399e-05, "loss": 0.4686881542205811, "memory(GiB)": 70.5, "step": 59390, "token_acc": 0.9127725856697819, "train_speed(iter/s)": 1.44812 }, { "epoch": 2.5446638961484083, "grad_norm": 4.58898401260376, "learning_rate": 4.859702648648047e-05, "loss": 0.38907995223999026, "memory(GiB)": 70.5, "step": 59395, "token_acc": 0.9014084507042254, "train_speed(iter/s)": 1.448132 }, { "epoch": 2.544878111477657, "grad_norm": 5.020251274108887, "learning_rate": 4.859029937597314e-05, "loss": 0.6481900691986084, "memory(GiB)": 70.5, "step": 59400, "token_acc": 0.857566765578635, "train_speed(iter/s)": 1.448129 }, { "epoch": 2.5450923268069063, "grad_norm": 2.5202810764312744, "learning_rate": 4.858357229100385e-05, "loss": 0.4843702793121338, "memory(GiB)": 70.5, "step": 59405, "token_acc": 0.8914285714285715, "train_speed(iter/s)": 1.448132 }, { "epoch": 2.545306542136155, "grad_norm": 1.633677363395691, "learning_rate": 4.857684523169449e-05, "loss": 0.5076782703399658, "memory(GiB)": 70.5, "step": 59410, "token_acc": 0.8777777777777778, "train_speed(iter/s)": 1.448134 }, { "epoch": 2.545520757465404, "grad_norm": 3.159020185470581, "learning_rate": 4.8570118198166896e-05, "loss": 0.6669922828674316, "memory(GiB)": 70.5, "step": 59415, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.448137 }, { "epoch": 2.545734972794653, "grad_norm": 2.585453987121582, "learning_rate": 4.8563391190542954e-05, "loss": 0.5105390548706055, "memory(GiB)": 70.5, "step": 59420, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.448152 }, { "epoch": 2.545949188123902, "grad_norm": 5.528973579406738, "learning_rate": 4.855666420894454e-05, "loss": 0.37211041450500487, "memory(GiB)": 70.5, "step": 59425, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.448163 }, { "epoch": 2.546163403453151, "grad_norm": 3.5127639770507812, "learning_rate": 4.85499372534935e-05, "loss": 0.2787866830825806, "memory(GiB)": 70.5, "step": 59430, "token_acc": 0.945054945054945, "train_speed(iter/s)": 1.448164 }, { "epoch": 2.5463776187824, "grad_norm": 2.4585540294647217, "learning_rate": 4.8543210324311704e-05, "loss": 0.7508837699890136, "memory(GiB)": 70.5, "step": 59435, "token_acc": 0.8296089385474861, "train_speed(iter/s)": 1.448178 }, { "epoch": 2.546591834111649, "grad_norm": 1.796675443649292, "learning_rate": 4.8536483421521025e-05, "loss": 0.4490503787994385, "memory(GiB)": 70.5, "step": 59440, "token_acc": 0.8820058997050148, "train_speed(iter/s)": 1.44819 }, { "epoch": 2.5468060494408977, "grad_norm": 0.5752619504928589, "learning_rate": 4.852975654524332e-05, "loss": 0.09454126954078675, "memory(GiB)": 70.5, "step": 59445, "token_acc": 0.9814126394052045, "train_speed(iter/s)": 1.448191 }, { "epoch": 2.547020264770147, "grad_norm": 3.753777265548706, "learning_rate": 4.852302969560046e-05, "loss": 0.34363656044006347, "memory(GiB)": 70.5, "step": 59450, "token_acc": 0.9163346613545816, "train_speed(iter/s)": 1.448197 }, { "epoch": 2.547234480099396, "grad_norm": 2.963494062423706, "learning_rate": 4.8516302872714295e-05, "loss": 0.32582085132598876, "memory(GiB)": 70.5, "step": 59455, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.448195 }, { "epoch": 2.5474486954286446, "grad_norm": 6.849099159240723, "learning_rate": 4.8509576076706695e-05, "loss": 0.411700439453125, "memory(GiB)": 70.5, "step": 59460, "token_acc": 0.905511811023622, "train_speed(iter/s)": 1.448198 }, { "epoch": 2.547662910757894, "grad_norm": 5.0922369956970215, "learning_rate": 4.8502849307699504e-05, "loss": 0.6163081169128418, "memory(GiB)": 70.5, "step": 59465, "token_acc": 0.8505747126436781, "train_speed(iter/s)": 1.448195 }, { "epoch": 2.5478771260871427, "grad_norm": 3.163170099258423, "learning_rate": 4.849612256581463e-05, "loss": 0.26701428890228274, "memory(GiB)": 70.5, "step": 59470, "token_acc": 0.9513888888888888, "train_speed(iter/s)": 1.448196 }, { "epoch": 2.5480913414163915, "grad_norm": 2.6607658863067627, "learning_rate": 4.8489395851173905e-05, "loss": 0.2719823598861694, "memory(GiB)": 70.5, "step": 59475, "token_acc": 0.9537815126050421, "train_speed(iter/s)": 1.4482 }, { "epoch": 2.5483055567456407, "grad_norm": 1.4077775478363037, "learning_rate": 4.848266916389918e-05, "loss": 0.5447182655334473, "memory(GiB)": 70.5, "step": 59480, "token_acc": 0.8754208754208754, "train_speed(iter/s)": 1.448201 }, { "epoch": 2.5485197720748896, "grad_norm": 5.19551420211792, "learning_rate": 4.847594250411234e-05, "loss": 0.7500299453735352, "memory(GiB)": 70.5, "step": 59485, "token_acc": 0.844311377245509, "train_speed(iter/s)": 1.448229 }, { "epoch": 2.5487339874041384, "grad_norm": 1.6267571449279785, "learning_rate": 4.8469215871935216e-05, "loss": 0.3494712352752686, "memory(GiB)": 70.5, "step": 59490, "token_acc": 0.9323843416370107, "train_speed(iter/s)": 1.448231 }, { "epoch": 2.5489482027333876, "grad_norm": 1.624430537223816, "learning_rate": 4.846248926748969e-05, "loss": 0.0843917727470398, "memory(GiB)": 70.5, "step": 59495, "token_acc": 0.9821428571428571, "train_speed(iter/s)": 1.448223 }, { "epoch": 2.5491624180626364, "grad_norm": 3.7649996280670166, "learning_rate": 4.845576269089762e-05, "loss": 0.2392256736755371, "memory(GiB)": 70.5, "step": 59500, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.448227 }, { "epoch": 2.5491624180626364, "eval_loss": 2.5696957111358643, "eval_runtime": 13.4677, "eval_samples_per_second": 7.425, "eval_steps_per_second": 7.425, "eval_token_acc": 0.4478442280945758, "step": 59500 }, { "epoch": 2.5493766333918852, "grad_norm": 3.3883144855499268, "learning_rate": 4.844903614228084e-05, "loss": 0.3842835664749146, "memory(GiB)": 70.5, "step": 59505, "token_acc": 0.5963912630579298, "train_speed(iter/s)": 1.447706 }, { "epoch": 2.5495908487211345, "grad_norm": 1.9546549320220947, "learning_rate": 4.844230962176124e-05, "loss": 0.35518512725830076, "memory(GiB)": 70.5, "step": 59510, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.447704 }, { "epoch": 2.5498050640503833, "grad_norm": 4.506442546844482, "learning_rate": 4.8435583129460666e-05, "loss": 0.5375897884368896, "memory(GiB)": 70.5, "step": 59515, "token_acc": 0.8842105263157894, "train_speed(iter/s)": 1.447703 }, { "epoch": 2.550019279379632, "grad_norm": 1.333950400352478, "learning_rate": 4.842885666550095e-05, "loss": 0.29455602169036865, "memory(GiB)": 70.5, "step": 59520, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.447722 }, { "epoch": 2.5502334947088814, "grad_norm": 4.731009006500244, "learning_rate": 4.842213023000399e-05, "loss": 0.29827070236206055, "memory(GiB)": 70.5, "step": 59525, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.447723 }, { "epoch": 2.55044771003813, "grad_norm": 1.379414439201355, "learning_rate": 4.841540382309161e-05, "loss": 0.2122262954711914, "memory(GiB)": 70.5, "step": 59530, "token_acc": 0.9559322033898305, "train_speed(iter/s)": 1.447727 }, { "epoch": 2.550661925367379, "grad_norm": 0.3963235318660736, "learning_rate": 4.8408677444885685e-05, "loss": 0.3103933811187744, "memory(GiB)": 70.5, "step": 59535, "token_acc": 0.9471544715447154, "train_speed(iter/s)": 1.447738 }, { "epoch": 2.5508761406966283, "grad_norm": 6.360567569732666, "learning_rate": 4.840195109550804e-05, "loss": 0.5649425029754639, "memory(GiB)": 70.5, "step": 59540, "token_acc": 0.8786885245901639, "train_speed(iter/s)": 1.447751 }, { "epoch": 2.551090356025877, "grad_norm": 2.0468454360961914, "learning_rate": 4.8395224775080574e-05, "loss": 0.6063085556030273, "memory(GiB)": 70.5, "step": 59545, "token_acc": 0.875, "train_speed(iter/s)": 1.447743 }, { "epoch": 2.5513045713551263, "grad_norm": 2.779343605041504, "learning_rate": 4.83884984837251e-05, "loss": 0.550475025177002, "memory(GiB)": 70.5, "step": 59550, "token_acc": 0.8904109589041096, "train_speed(iter/s)": 1.447749 }, { "epoch": 2.551518786684375, "grad_norm": 6.863926887512207, "learning_rate": 4.83817722215635e-05, "loss": 0.39963555335998535, "memory(GiB)": 70.5, "step": 59555, "token_acc": 0.9041916167664671, "train_speed(iter/s)": 1.44776 }, { "epoch": 2.551733002013624, "grad_norm": 4.308107852935791, "learning_rate": 4.837504598871762e-05, "loss": 0.2791240930557251, "memory(GiB)": 70.5, "step": 59560, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.447761 }, { "epoch": 2.551947217342873, "grad_norm": 2.2922146320343018, "learning_rate": 4.8368319785309285e-05, "loss": 0.36616692543029783, "memory(GiB)": 70.5, "step": 59565, "token_acc": 0.9256198347107438, "train_speed(iter/s)": 1.447768 }, { "epoch": 2.552161432672122, "grad_norm": 3.6156551837921143, "learning_rate": 4.836159361146038e-05, "loss": 0.561351728439331, "memory(GiB)": 70.5, "step": 59570, "token_acc": 0.8717105263157895, "train_speed(iter/s)": 1.447777 }, { "epoch": 2.552375648001371, "grad_norm": 1.4867196083068848, "learning_rate": 4.835486746729274e-05, "loss": 0.1662299394607544, "memory(GiB)": 70.5, "step": 59575, "token_acc": 0.9628099173553719, "train_speed(iter/s)": 1.447783 }, { "epoch": 2.55258986333062, "grad_norm": 5.098706245422363, "learning_rate": 4.834814135292822e-05, "loss": 0.4768869876861572, "memory(GiB)": 70.5, "step": 59580, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.447793 }, { "epoch": 2.552804078659869, "grad_norm": 2.662602186203003, "learning_rate": 4.834141526848868e-05, "loss": 0.5864229679107666, "memory(GiB)": 70.5, "step": 59585, "token_acc": 0.8794520547945206, "train_speed(iter/s)": 1.447801 }, { "epoch": 2.5530182939891177, "grad_norm": 1.0740177631378174, "learning_rate": 4.833468921409594e-05, "loss": 0.33481361865997317, "memory(GiB)": 70.5, "step": 59590, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.447802 }, { "epoch": 2.553232509318367, "grad_norm": 8.25197696685791, "learning_rate": 4.832796318987188e-05, "loss": 0.48386383056640625, "memory(GiB)": 70.5, "step": 59595, "token_acc": 0.8859649122807017, "train_speed(iter/s)": 1.447814 }, { "epoch": 2.553446724647616, "grad_norm": 5.341479778289795, "learning_rate": 4.832123719593834e-05, "loss": 0.6009721755981445, "memory(GiB)": 70.5, "step": 59600, "token_acc": 0.8668941979522184, "train_speed(iter/s)": 1.447842 }, { "epoch": 2.5536609399768646, "grad_norm": 2.3513998985290527, "learning_rate": 4.831451123241715e-05, "loss": 0.3301408767700195, "memory(GiB)": 70.5, "step": 59605, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.447851 }, { "epoch": 2.553875155306114, "grad_norm": 1.1586573123931885, "learning_rate": 4.8307785299430156e-05, "loss": 0.2040421962738037, "memory(GiB)": 70.5, "step": 59610, "token_acc": 0.9597855227882037, "train_speed(iter/s)": 1.447845 }, { "epoch": 2.5540893706353627, "grad_norm": 0.07761450856924057, "learning_rate": 4.830105939709924e-05, "loss": 0.25692222118377683, "memory(GiB)": 70.5, "step": 59615, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.44785 }, { "epoch": 2.5543035859646115, "grad_norm": 2.773683786392212, "learning_rate": 4.8294333525546234e-05, "loss": 0.35431063175201416, "memory(GiB)": 70.5, "step": 59620, "token_acc": 0.9256965944272446, "train_speed(iter/s)": 1.447846 }, { "epoch": 2.5545178012938607, "grad_norm": 1.7285358905792236, "learning_rate": 4.828760768489295e-05, "loss": 0.22576169967651366, "memory(GiB)": 70.5, "step": 59625, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.44785 }, { "epoch": 2.5547320166231096, "grad_norm": 7.6263747215271, "learning_rate": 4.8280881875261284e-05, "loss": 0.4621800422668457, "memory(GiB)": 70.5, "step": 59630, "token_acc": 0.8692307692307693, "train_speed(iter/s)": 1.44785 }, { "epoch": 2.5549462319523584, "grad_norm": 2.083186388015747, "learning_rate": 4.8274156096773046e-05, "loss": 0.48786358833312987, "memory(GiB)": 70.5, "step": 59635, "token_acc": 0.908745247148289, "train_speed(iter/s)": 1.44785 }, { "epoch": 2.5551604472816076, "grad_norm": 1.861074686050415, "learning_rate": 4.8267430349550094e-05, "loss": 0.33250038623809813, "memory(GiB)": 70.5, "step": 59640, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.447868 }, { "epoch": 2.5553746626108564, "grad_norm": 4.608387470245361, "learning_rate": 4.826070463371427e-05, "loss": 0.4491335391998291, "memory(GiB)": 70.5, "step": 59645, "token_acc": 0.905511811023622, "train_speed(iter/s)": 1.447863 }, { "epoch": 2.5555888779401053, "grad_norm": 9.06170654296875, "learning_rate": 4.8253978949387394e-05, "loss": 0.40519256591796876, "memory(GiB)": 70.5, "step": 59650, "token_acc": 0.9121338912133892, "train_speed(iter/s)": 1.447881 }, { "epoch": 2.5558030932693545, "grad_norm": 4.077483177185059, "learning_rate": 4.824725329669135e-05, "loss": 0.7050768852233886, "memory(GiB)": 70.5, "step": 59655, "token_acc": 0.8470588235294118, "train_speed(iter/s)": 1.447892 }, { "epoch": 2.5560173085986033, "grad_norm": 3.5801122188568115, "learning_rate": 4.824052767574795e-05, "loss": 0.34255430698394773, "memory(GiB)": 70.5, "step": 59660, "token_acc": 0.9150326797385621, "train_speed(iter/s)": 1.447892 }, { "epoch": 2.556231523927852, "grad_norm": 3.0205016136169434, "learning_rate": 4.823380208667903e-05, "loss": 0.35816664695739747, "memory(GiB)": 70.5, "step": 59665, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.447894 }, { "epoch": 2.5564457392571014, "grad_norm": 2.7499518394470215, "learning_rate": 4.8227076529606455e-05, "loss": 0.32712767124176023, "memory(GiB)": 70.5, "step": 59670, "token_acc": 0.9394812680115274, "train_speed(iter/s)": 1.447903 }, { "epoch": 2.55665995458635, "grad_norm": 0.40258464217185974, "learning_rate": 4.8220351004652036e-05, "loss": 0.32467973232269287, "memory(GiB)": 70.5, "step": 59675, "token_acc": 0.9298245614035088, "train_speed(iter/s)": 1.447908 }, { "epoch": 2.556874169915599, "grad_norm": 3.7844717502593994, "learning_rate": 4.8213625511937644e-05, "loss": 0.3515667200088501, "memory(GiB)": 70.5, "step": 59680, "token_acc": 0.9139344262295082, "train_speed(iter/s)": 1.447923 }, { "epoch": 2.5570883852448483, "grad_norm": 2.702881336212158, "learning_rate": 4.820690005158508e-05, "loss": 0.18180389404296876, "memory(GiB)": 70.5, "step": 59685, "token_acc": 0.9537037037037037, "train_speed(iter/s)": 1.447924 }, { "epoch": 2.557302600574097, "grad_norm": 0.5031715035438538, "learning_rate": 4.820017462371622e-05, "loss": 0.13924756050109863, "memory(GiB)": 70.5, "step": 59690, "token_acc": 0.9617834394904459, "train_speed(iter/s)": 1.447929 }, { "epoch": 2.557516815903346, "grad_norm": 2.2283997535705566, "learning_rate": 4.819344922845288e-05, "loss": 0.3457207202911377, "memory(GiB)": 70.5, "step": 59695, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.447935 }, { "epoch": 2.557731031232595, "grad_norm": 3.4762163162231445, "learning_rate": 4.818672386591691e-05, "loss": 0.4823345184326172, "memory(GiB)": 70.5, "step": 59700, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.447942 }, { "epoch": 2.557945246561844, "grad_norm": 3.996432065963745, "learning_rate": 4.817999853623014e-05, "loss": 0.2878229856491089, "memory(GiB)": 70.5, "step": 59705, "token_acc": 0.9568345323741008, "train_speed(iter/s)": 1.447942 }, { "epoch": 2.558159461891093, "grad_norm": 5.647372722625732, "learning_rate": 4.8173273239514396e-05, "loss": 0.5637889862060547, "memory(GiB)": 70.5, "step": 59710, "token_acc": 0.8717201166180758, "train_speed(iter/s)": 1.447938 }, { "epoch": 2.558373677220342, "grad_norm": 0.5163751244544983, "learning_rate": 4.816654797589153e-05, "loss": 0.3420542240142822, "memory(GiB)": 70.5, "step": 59715, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.447938 }, { "epoch": 2.558587892549591, "grad_norm": 5.486770153045654, "learning_rate": 4.815982274548335e-05, "loss": 0.47229843139648436, "memory(GiB)": 70.5, "step": 59720, "token_acc": 0.9308176100628931, "train_speed(iter/s)": 1.44794 }, { "epoch": 2.5588021078788397, "grad_norm": 4.09434700012207, "learning_rate": 4.815309754841172e-05, "loss": 0.35894510746002195, "memory(GiB)": 70.5, "step": 59725, "token_acc": 0.9413793103448276, "train_speed(iter/s)": 1.447956 }, { "epoch": 2.559016323208089, "grad_norm": 7.591298580169678, "learning_rate": 4.814637238479847e-05, "loss": 0.582444715499878, "memory(GiB)": 70.5, "step": 59730, "token_acc": 0.9069069069069069, "train_speed(iter/s)": 1.447956 }, { "epoch": 2.5592305385373377, "grad_norm": 2.6433913707733154, "learning_rate": 4.8139647254765404e-05, "loss": 0.5797430515289307, "memory(GiB)": 70.5, "step": 59735, "token_acc": 0.9030303030303031, "train_speed(iter/s)": 1.44796 }, { "epoch": 2.5594447538665865, "grad_norm": 4.770023822784424, "learning_rate": 4.8132922158434384e-05, "loss": 0.48235182762145995, "memory(GiB)": 70.5, "step": 59740, "token_acc": 0.9186991869918699, "train_speed(iter/s)": 1.44798 }, { "epoch": 2.559658969195836, "grad_norm": 0.2188655585050583, "learning_rate": 4.812619709592723e-05, "loss": 0.38612320423126223, "memory(GiB)": 70.5, "step": 59745, "token_acc": 0.9241379310344827, "train_speed(iter/s)": 1.447977 }, { "epoch": 2.5598731845250846, "grad_norm": 2.42170786857605, "learning_rate": 4.8119472067365766e-05, "loss": 0.24583377838134765, "memory(GiB)": 70.5, "step": 59750, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.447974 }, { "epoch": 2.5600873998543334, "grad_norm": 4.364752292633057, "learning_rate": 4.8112747072871836e-05, "loss": 0.49868106842041016, "memory(GiB)": 70.5, "step": 59755, "token_acc": 0.9044368600682594, "train_speed(iter/s)": 1.447972 }, { "epoch": 2.5603016151835827, "grad_norm": 3.600128412246704, "learning_rate": 4.8106022112567247e-05, "loss": 0.5274038314819336, "memory(GiB)": 70.5, "step": 59760, "token_acc": 0.8850574712643678, "train_speed(iter/s)": 1.447968 }, { "epoch": 2.5605158305128315, "grad_norm": 4.17810583114624, "learning_rate": 4.809929718657386e-05, "loss": 0.3343848705291748, "memory(GiB)": 70.5, "step": 59765, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.447971 }, { "epoch": 2.5607300458420803, "grad_norm": 4.271091461181641, "learning_rate": 4.809257229501348e-05, "loss": 0.4508793830871582, "memory(GiB)": 70.5, "step": 59770, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.447974 }, { "epoch": 2.5609442611713296, "grad_norm": 4.327223300933838, "learning_rate": 4.8085847438007955e-05, "loss": 0.3066711902618408, "memory(GiB)": 70.5, "step": 59775, "token_acc": 0.9274193548387096, "train_speed(iter/s)": 1.447982 }, { "epoch": 2.5611584765005784, "grad_norm": 5.182854175567627, "learning_rate": 4.807912261567908e-05, "loss": 0.5948902130126953, "memory(GiB)": 70.5, "step": 59780, "token_acc": 0.8960573476702509, "train_speed(iter/s)": 1.447978 }, { "epoch": 2.561372691829827, "grad_norm": 4.554727554321289, "learning_rate": 4.807239782814872e-05, "loss": 0.2834742546081543, "memory(GiB)": 70.5, "step": 59785, "token_acc": 0.9331103678929766, "train_speed(iter/s)": 1.447975 }, { "epoch": 2.5615869071590764, "grad_norm": 5.3155999183654785, "learning_rate": 4.8065673075538685e-05, "loss": 0.42073354721069334, "memory(GiB)": 70.5, "step": 59790, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.447966 }, { "epoch": 2.5618011224883253, "grad_norm": 2.112886905670166, "learning_rate": 4.805894835797078e-05, "loss": 0.35428922176361083, "memory(GiB)": 70.5, "step": 59795, "token_acc": 0.9147982062780269, "train_speed(iter/s)": 1.447965 }, { "epoch": 2.562015337817574, "grad_norm": 0.9821924567222595, "learning_rate": 4.805222367556685e-05, "loss": 0.2463534355163574, "memory(GiB)": 70.5, "step": 59800, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.447967 }, { "epoch": 2.5622295531468233, "grad_norm": 0.07259444147348404, "learning_rate": 4.804549902844873e-05, "loss": 0.1659997820854187, "memory(GiB)": 70.5, "step": 59805, "token_acc": 0.9672131147540983, "train_speed(iter/s)": 1.447975 }, { "epoch": 2.562443768476072, "grad_norm": 4.364706516265869, "learning_rate": 4.8038774416738205e-05, "loss": 0.4457086563110352, "memory(GiB)": 70.5, "step": 59810, "token_acc": 0.8694029850746269, "train_speed(iter/s)": 1.447981 }, { "epoch": 2.562657983805321, "grad_norm": 1.0139411687850952, "learning_rate": 4.803204984055714e-05, "loss": 0.22575559616088867, "memory(GiB)": 70.5, "step": 59815, "token_acc": 0.9614035087719298, "train_speed(iter/s)": 1.447995 }, { "epoch": 2.56287219913457, "grad_norm": 2.945647954940796, "learning_rate": 4.802532530002733e-05, "loss": 0.3220551013946533, "memory(GiB)": 70.5, "step": 59820, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.447997 }, { "epoch": 2.563086414463819, "grad_norm": 5.617775917053223, "learning_rate": 4.801860079527061e-05, "loss": 0.558200454711914, "memory(GiB)": 70.5, "step": 59825, "token_acc": 0.8679867986798679, "train_speed(iter/s)": 1.44801 }, { "epoch": 2.563300629793068, "grad_norm": 2.0984723567962646, "learning_rate": 4.8011876326408796e-05, "loss": 0.32879412174224854, "memory(GiB)": 70.5, "step": 59830, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.448013 }, { "epoch": 2.563514845122317, "grad_norm": 2.5113375186920166, "learning_rate": 4.8005151893563684e-05, "loss": 0.4408135890960693, "memory(GiB)": 70.5, "step": 59835, "token_acc": 0.9034749034749034, "train_speed(iter/s)": 1.448005 }, { "epoch": 2.563729060451566, "grad_norm": 3.6893577575683594, "learning_rate": 4.799842749685713e-05, "loss": 0.2574336290359497, "memory(GiB)": 70.5, "step": 59840, "token_acc": 0.9590443686006825, "train_speed(iter/s)": 1.44802 }, { "epoch": 2.5639432757808147, "grad_norm": 3.612748622894287, "learning_rate": 4.799170313641095e-05, "loss": 0.2859429121017456, "memory(GiB)": 70.5, "step": 59845, "token_acc": 0.9376947040498442, "train_speed(iter/s)": 1.448023 }, { "epoch": 2.564157491110064, "grad_norm": 2.6843903064727783, "learning_rate": 4.798497881234695e-05, "loss": 0.46192069053649903, "memory(GiB)": 70.5, "step": 59850, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.448031 }, { "epoch": 2.564371706439313, "grad_norm": 3.20641827583313, "learning_rate": 4.7978254524786935e-05, "loss": 0.3943788051605225, "memory(GiB)": 70.5, "step": 59855, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.448036 }, { "epoch": 2.5645859217685616, "grad_norm": 1.0374454259872437, "learning_rate": 4.7971530273852754e-05, "loss": 0.29730203151702883, "memory(GiB)": 70.5, "step": 59860, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.448033 }, { "epoch": 2.564800137097811, "grad_norm": 0.2911221385002136, "learning_rate": 4.796480605966619e-05, "loss": 0.17471543550491334, "memory(GiB)": 70.5, "step": 59865, "token_acc": 0.9624060150375939, "train_speed(iter/s)": 1.448036 }, { "epoch": 2.5650143524270597, "grad_norm": 4.456498146057129, "learning_rate": 4.795808188234909e-05, "loss": 0.350221061706543, "memory(GiB)": 70.5, "step": 59870, "token_acc": 0.9343065693430657, "train_speed(iter/s)": 1.448051 }, { "epoch": 2.5652285677563085, "grad_norm": 4.248950004577637, "learning_rate": 4.795135774202324e-05, "loss": 0.33110666275024414, "memory(GiB)": 70.5, "step": 59875, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.448052 }, { "epoch": 2.5654427830855577, "grad_norm": 4.937633037567139, "learning_rate": 4.794463363881047e-05, "loss": 0.45792832374572756, "memory(GiB)": 70.5, "step": 59880, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.44805 }, { "epoch": 2.5656569984148065, "grad_norm": 0.9009456038475037, "learning_rate": 4.793790957283259e-05, "loss": 0.1758955240249634, "memory(GiB)": 70.5, "step": 59885, "token_acc": 0.9584905660377359, "train_speed(iter/s)": 1.44805 }, { "epoch": 2.565871213744056, "grad_norm": 2.798630714416504, "learning_rate": 4.7931185544211416e-05, "loss": 0.28858470916748047, "memory(GiB)": 70.5, "step": 59890, "token_acc": 0.95, "train_speed(iter/s)": 1.448055 }, { "epoch": 2.5660854290733046, "grad_norm": 2.0210230350494385, "learning_rate": 4.7924461553068745e-05, "loss": 0.2711117506027222, "memory(GiB)": 70.5, "step": 59895, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.448049 }, { "epoch": 2.5662996444025534, "grad_norm": 1.1994293928146362, "learning_rate": 4.7917737599526415e-05, "loss": 0.4530925273895264, "memory(GiB)": 70.5, "step": 59900, "token_acc": 0.9021406727828746, "train_speed(iter/s)": 1.448054 }, { "epoch": 2.5665138597318027, "grad_norm": 0.9524989128112793, "learning_rate": 4.791101368370619e-05, "loss": 0.33599343299865725, "memory(GiB)": 70.5, "step": 59905, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.448072 }, { "epoch": 2.5667280750610515, "grad_norm": 2.2706387042999268, "learning_rate": 4.790428980572994e-05, "loss": 0.263342547416687, "memory(GiB)": 70.5, "step": 59910, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.448062 }, { "epoch": 2.5669422903903003, "grad_norm": 2.5851426124572754, "learning_rate": 4.789756596571944e-05, "loss": 0.25719990730285647, "memory(GiB)": 70.5, "step": 59915, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.448063 }, { "epoch": 2.5671565057195496, "grad_norm": 0.4263951778411865, "learning_rate": 4.789084216379651e-05, "loss": 0.406453275680542, "memory(GiB)": 70.5, "step": 59920, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.448085 }, { "epoch": 2.5673707210487984, "grad_norm": 2.311314344406128, "learning_rate": 4.788411840008294e-05, "loss": 0.26244940757751467, "memory(GiB)": 70.5, "step": 59925, "token_acc": 0.9503105590062112, "train_speed(iter/s)": 1.44809 }, { "epoch": 2.567584936378047, "grad_norm": 3.697009801864624, "learning_rate": 4.7877394674700564e-05, "loss": 0.6918257236480713, "memory(GiB)": 70.5, "step": 59930, "token_acc": 0.8368794326241135, "train_speed(iter/s)": 1.4481 }, { "epoch": 2.5677991517072964, "grad_norm": 2.991582155227661, "learning_rate": 4.787067098777117e-05, "loss": 0.38728172779083253, "memory(GiB)": 70.5, "step": 59935, "token_acc": 0.89419795221843, "train_speed(iter/s)": 1.448106 }, { "epoch": 2.5680133670365453, "grad_norm": 2.416715145111084, "learning_rate": 4.786394733941657e-05, "loss": 0.2688279390335083, "memory(GiB)": 70.5, "step": 59940, "token_acc": 0.9296636085626911, "train_speed(iter/s)": 1.448108 }, { "epoch": 2.568227582365794, "grad_norm": 2.2965245246887207, "learning_rate": 4.785722372975857e-05, "loss": 0.5917043209075927, "memory(GiB)": 70.5, "step": 59945, "token_acc": 0.8988095238095238, "train_speed(iter/s)": 1.448119 }, { "epoch": 2.5684417976950433, "grad_norm": 3.7120425701141357, "learning_rate": 4.785050015891897e-05, "loss": 0.37626471519470217, "memory(GiB)": 70.5, "step": 59950, "token_acc": 0.9323843416370107, "train_speed(iter/s)": 1.448133 }, { "epoch": 2.568656013024292, "grad_norm": 2.9719607830047607, "learning_rate": 4.7843776627019567e-05, "loss": 0.2030038833618164, "memory(GiB)": 70.5, "step": 59955, "token_acc": 0.9546925566343042, "train_speed(iter/s)": 1.448139 }, { "epoch": 2.568870228353541, "grad_norm": 3.667565107345581, "learning_rate": 4.7837053134182183e-05, "loss": 0.16624733209609985, "memory(GiB)": 70.5, "step": 59960, "token_acc": 0.9624060150375939, "train_speed(iter/s)": 1.448142 }, { "epoch": 2.56908444368279, "grad_norm": 2.4923243522644043, "learning_rate": 4.78303296805286e-05, "loss": 0.5576051712036133, "memory(GiB)": 70.5, "step": 59965, "token_acc": 0.8885714285714286, "train_speed(iter/s)": 1.448147 }, { "epoch": 2.569298659012039, "grad_norm": 7.0875139236450195, "learning_rate": 4.782360626618064e-05, "loss": 0.342251181602478, "memory(GiB)": 70.5, "step": 59970, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.448154 }, { "epoch": 2.569512874341288, "grad_norm": 3.3962533473968506, "learning_rate": 4.78168828912601e-05, "loss": 0.34715795516967773, "memory(GiB)": 70.5, "step": 59975, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.448173 }, { "epoch": 2.569727089670537, "grad_norm": 3.3974130153656006, "learning_rate": 4.781015955588875e-05, "loss": 0.5359176635742188, "memory(GiB)": 70.5, "step": 59980, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.448185 }, { "epoch": 2.569941304999786, "grad_norm": 4.8437910079956055, "learning_rate": 4.7803436260188425e-05, "loss": 0.3897327661514282, "memory(GiB)": 70.5, "step": 59985, "token_acc": 0.912, "train_speed(iter/s)": 1.448193 }, { "epoch": 2.5701555203290347, "grad_norm": 1.839425802230835, "learning_rate": 4.779671300428092e-05, "loss": 0.4272324562072754, "memory(GiB)": 70.5, "step": 59990, "token_acc": 0.93125, "train_speed(iter/s)": 1.448191 }, { "epoch": 2.570369735658284, "grad_norm": 6.209967613220215, "learning_rate": 4.778998978828803e-05, "loss": 0.5003615379333496, "memory(GiB)": 70.5, "step": 59995, "token_acc": 0.8859934853420195, "train_speed(iter/s)": 1.448191 }, { "epoch": 2.570583950987533, "grad_norm": 0.027519801631569862, "learning_rate": 4.7783266612331536e-05, "loss": 0.3855245590209961, "memory(GiB)": 70.5, "step": 60000, "token_acc": 0.9099378881987578, "train_speed(iter/s)": 1.448191 }, { "epoch": 2.570583950987533, "eval_loss": 2.485543966293335, "eval_runtime": 13.7382, "eval_samples_per_second": 7.279, "eval_steps_per_second": 7.279, "eval_token_acc": 0.4395750332005312, "step": 60000 }, { "epoch": 2.5707981663167816, "grad_norm": 1.1381382942199707, "learning_rate": 4.777654347653326e-05, "loss": 0.22443218231201173, "memory(GiB)": 70.5, "step": 60005, "token_acc": 0.5863723608445297, "train_speed(iter/s)": 1.447672 }, { "epoch": 2.571012381646031, "grad_norm": 1.8169747591018677, "learning_rate": 4.776982038101497e-05, "loss": 0.4408287525177002, "memory(GiB)": 70.5, "step": 60010, "token_acc": 0.8856209150326797, "train_speed(iter/s)": 1.447684 }, { "epoch": 2.5712265969752797, "grad_norm": 1.6088435649871826, "learning_rate": 4.776309732589849e-05, "loss": 0.3304621934890747, "memory(GiB)": 70.5, "step": 60015, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.447693 }, { "epoch": 2.5714408123045285, "grad_norm": 3.9175102710723877, "learning_rate": 4.775637431130559e-05, "loss": 0.48352870941162107, "memory(GiB)": 70.5, "step": 60020, "token_acc": 0.908256880733945, "train_speed(iter/s)": 1.447697 }, { "epoch": 2.5716550276337777, "grad_norm": 3.2655248641967773, "learning_rate": 4.774965133735808e-05, "loss": 0.3541638612747192, "memory(GiB)": 70.5, "step": 60025, "token_acc": 0.9320987654320988, "train_speed(iter/s)": 1.44769 }, { "epoch": 2.5718692429630265, "grad_norm": 2.8271894454956055, "learning_rate": 4.7742928404177746e-05, "loss": 0.5174476623535156, "memory(GiB)": 70.5, "step": 60030, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.447692 }, { "epoch": 2.5720834582922754, "grad_norm": 3.5721967220306396, "learning_rate": 4.773620551188638e-05, "loss": 0.25433690547943116, "memory(GiB)": 70.5, "step": 60035, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.447694 }, { "epoch": 2.5722976736215246, "grad_norm": 3.71600341796875, "learning_rate": 4.772948266060577e-05, "loss": 0.3170119524002075, "memory(GiB)": 70.5, "step": 60040, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.447698 }, { "epoch": 2.5725118889507734, "grad_norm": 3.3705546855926514, "learning_rate": 4.772275985045772e-05, "loss": 0.2885743618011475, "memory(GiB)": 70.5, "step": 60045, "token_acc": 0.933852140077821, "train_speed(iter/s)": 1.447712 }, { "epoch": 2.5727261042800222, "grad_norm": 3.6419620513916016, "learning_rate": 4.7716037081564004e-05, "loss": 0.27875490188598634, "memory(GiB)": 70.5, "step": 60050, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.447716 }, { "epoch": 2.5729403196092715, "grad_norm": 3.7369956970214844, "learning_rate": 4.7709314354046415e-05, "loss": 0.4204849720001221, "memory(GiB)": 70.5, "step": 60055, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.447721 }, { "epoch": 2.5731545349385203, "grad_norm": 3.706280469894409, "learning_rate": 4.7702591668026745e-05, "loss": 0.367569637298584, "memory(GiB)": 70.5, "step": 60060, "token_acc": 0.9147727272727273, "train_speed(iter/s)": 1.447717 }, { "epoch": 2.573368750267769, "grad_norm": 3.8305013179779053, "learning_rate": 4.769586902362679e-05, "loss": 0.44542760848999025, "memory(GiB)": 70.5, "step": 60065, "token_acc": 0.909967845659164, "train_speed(iter/s)": 1.447732 }, { "epoch": 2.5735829655970184, "grad_norm": 1.3875317573547363, "learning_rate": 4.768914642096833e-05, "loss": 0.2813662052154541, "memory(GiB)": 70.5, "step": 60070, "token_acc": 0.9306122448979591, "train_speed(iter/s)": 1.447748 }, { "epoch": 2.573797180926267, "grad_norm": 3.3030571937561035, "learning_rate": 4.768242386017315e-05, "loss": 0.3589199066162109, "memory(GiB)": 70.5, "step": 60075, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.447763 }, { "epoch": 2.574011396255516, "grad_norm": 4.693004608154297, "learning_rate": 4.767570134136304e-05, "loss": 0.21158759593963622, "memory(GiB)": 70.5, "step": 60080, "token_acc": 0.9633699633699634, "train_speed(iter/s)": 1.447769 }, { "epoch": 2.5742256115847653, "grad_norm": 1.227569580078125, "learning_rate": 4.766897886465977e-05, "loss": 0.33424789905548097, "memory(GiB)": 70.5, "step": 60085, "token_acc": 0.9197530864197531, "train_speed(iter/s)": 1.447772 }, { "epoch": 2.574439826914014, "grad_norm": 5.6646928787231445, "learning_rate": 4.766225643018514e-05, "loss": 0.35194690227508546, "memory(GiB)": 70.5, "step": 60090, "token_acc": 0.9348534201954397, "train_speed(iter/s)": 1.447779 }, { "epoch": 2.574654042243263, "grad_norm": 3.7881453037261963, "learning_rate": 4.765553403806094e-05, "loss": 0.20416855812072754, "memory(GiB)": 70.5, "step": 60095, "token_acc": 0.9467213114754098, "train_speed(iter/s)": 1.447775 }, { "epoch": 2.574868257572512, "grad_norm": 5.964277267456055, "learning_rate": 4.764881168840892e-05, "loss": 0.33675355911254884, "memory(GiB)": 70.5, "step": 60100, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.447782 }, { "epoch": 2.575082472901761, "grad_norm": 2.3206729888916016, "learning_rate": 4.7642089381350895e-05, "loss": 0.31956484317779543, "memory(GiB)": 70.5, "step": 60105, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.447796 }, { "epoch": 2.5752966882310098, "grad_norm": 2.7029953002929688, "learning_rate": 4.763536711700862e-05, "loss": 0.5436374187469483, "memory(GiB)": 70.5, "step": 60110, "token_acc": 0.892271662763466, "train_speed(iter/s)": 1.447794 }, { "epoch": 2.575510903560259, "grad_norm": 5.29086446762085, "learning_rate": 4.76286448955039e-05, "loss": 0.36979126930236816, "memory(GiB)": 70.5, "step": 60115, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.447799 }, { "epoch": 2.575725118889508, "grad_norm": 2.8965978622436523, "learning_rate": 4.76219227169585e-05, "loss": 0.3297891139984131, "memory(GiB)": 70.5, "step": 60120, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.447808 }, { "epoch": 2.5759393342187566, "grad_norm": 3.5996007919311523, "learning_rate": 4.7615200581494194e-05, "loss": 0.3904033899307251, "memory(GiB)": 70.5, "step": 60125, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.447817 }, { "epoch": 2.576153549548006, "grad_norm": 2.684722900390625, "learning_rate": 4.7608478489232756e-05, "loss": 0.6034920692443848, "memory(GiB)": 70.5, "step": 60130, "token_acc": 0.8909657320872274, "train_speed(iter/s)": 1.44782 }, { "epoch": 2.5763677648772547, "grad_norm": 5.065707683563232, "learning_rate": 4.760175644029599e-05, "loss": 0.5713113784790039, "memory(GiB)": 70.5, "step": 60135, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.447838 }, { "epoch": 2.5765819802065035, "grad_norm": 1.6584800481796265, "learning_rate": 4.759503443480566e-05, "loss": 0.39683923721313474, "memory(GiB)": 70.5, "step": 60140, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.447835 }, { "epoch": 2.576796195535753, "grad_norm": 3.07021427154541, "learning_rate": 4.758831247288353e-05, "loss": 0.3121178388595581, "memory(GiB)": 70.5, "step": 60145, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.447836 }, { "epoch": 2.5770104108650016, "grad_norm": 1.45337975025177, "learning_rate": 4.758159055465138e-05, "loss": 0.386388373374939, "memory(GiB)": 70.5, "step": 60150, "token_acc": 0.9065155807365439, "train_speed(iter/s)": 1.447846 }, { "epoch": 2.5772246261942504, "grad_norm": 3.80178165435791, "learning_rate": 4.757486868023099e-05, "loss": 0.18699262142181397, "memory(GiB)": 70.5, "step": 60155, "token_acc": 0.9606557377049181, "train_speed(iter/s)": 1.447851 }, { "epoch": 2.5774388415234997, "grad_norm": 3.4728643894195557, "learning_rate": 4.756814684974413e-05, "loss": 0.4727576732635498, "memory(GiB)": 70.5, "step": 60160, "token_acc": 0.8932926829268293, "train_speed(iter/s)": 1.447847 }, { "epoch": 2.5776530568527485, "grad_norm": 2.707319498062134, "learning_rate": 4.756142506331258e-05, "loss": 0.26353609561920166, "memory(GiB)": 70.5, "step": 60165, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.447846 }, { "epoch": 2.5778672721819973, "grad_norm": 4.827116966247559, "learning_rate": 4.755470332105808e-05, "loss": 0.6284004211425781, "memory(GiB)": 70.5, "step": 60170, "token_acc": 0.8671328671328671, "train_speed(iter/s)": 1.447856 }, { "epoch": 2.5780814875112466, "grad_norm": 1.8508152961730957, "learning_rate": 4.754798162310244e-05, "loss": 0.2751585006713867, "memory(GiB)": 70.5, "step": 60175, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.447851 }, { "epoch": 2.5782957028404954, "grad_norm": 3.247572183609009, "learning_rate": 4.7541259969567416e-05, "loss": 0.4391640186309814, "memory(GiB)": 70.5, "step": 60180, "token_acc": 0.8884892086330936, "train_speed(iter/s)": 1.447845 }, { "epoch": 2.578509918169744, "grad_norm": 6.647392749786377, "learning_rate": 4.753453836057476e-05, "loss": 0.4789111614227295, "memory(GiB)": 70.5, "step": 60185, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.447846 }, { "epoch": 2.5787241334989934, "grad_norm": 4.709986686706543, "learning_rate": 4.752781679624626e-05, "loss": 0.3820672512054443, "memory(GiB)": 70.5, "step": 60190, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.447859 }, { "epoch": 2.5789383488282422, "grad_norm": 7.168796062469482, "learning_rate": 4.7521095276703676e-05, "loss": 0.5937058925628662, "memory(GiB)": 70.5, "step": 60195, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.447869 }, { "epoch": 2.579152564157491, "grad_norm": 4.107027053833008, "learning_rate": 4.7514373802068786e-05, "loss": 0.208231782913208, "memory(GiB)": 70.5, "step": 60200, "token_acc": 0.9602888086642599, "train_speed(iter/s)": 1.44788 }, { "epoch": 2.5793667794867403, "grad_norm": 1.1706066131591797, "learning_rate": 4.750765237246332e-05, "loss": 0.32257418632507323, "memory(GiB)": 70.5, "step": 60205, "token_acc": 0.9362880886426593, "train_speed(iter/s)": 1.447882 }, { "epoch": 2.579580994815989, "grad_norm": 3.938340187072754, "learning_rate": 4.750093098800909e-05, "loss": 0.32186074256896974, "memory(GiB)": 70.5, "step": 60210, "token_acc": 0.927797833935018, "train_speed(iter/s)": 1.447888 }, { "epoch": 2.579795210145238, "grad_norm": 1.9333864450454712, "learning_rate": 4.749420964882783e-05, "loss": 0.4448897361755371, "memory(GiB)": 70.5, "step": 60215, "token_acc": 0.89937106918239, "train_speed(iter/s)": 1.447901 }, { "epoch": 2.580009425474487, "grad_norm": 4.461699485778809, "learning_rate": 4.748748835504133e-05, "loss": 0.43119988441467283, "memory(GiB)": 70.5, "step": 60220, "token_acc": 0.9163346613545816, "train_speed(iter/s)": 1.447911 }, { "epoch": 2.580223640803736, "grad_norm": 2.672562599182129, "learning_rate": 4.7480767106771326e-05, "loss": 0.25748820304870607, "memory(GiB)": 70.5, "step": 60225, "token_acc": 0.9438596491228071, "train_speed(iter/s)": 1.447906 }, { "epoch": 2.580437856132985, "grad_norm": 2.108863353729248, "learning_rate": 4.7474045904139586e-05, "loss": 0.16033059358596802, "memory(GiB)": 70.5, "step": 60230, "token_acc": 0.9598393574297188, "train_speed(iter/s)": 1.447904 }, { "epoch": 2.580652071462234, "grad_norm": 4.010164260864258, "learning_rate": 4.746732474726788e-05, "loss": 0.40812244415283205, "memory(GiB)": 70.5, "step": 60235, "token_acc": 0.899641577060932, "train_speed(iter/s)": 1.447901 }, { "epoch": 2.580866286791483, "grad_norm": 4.638872146606445, "learning_rate": 4.7460603636277956e-05, "loss": 0.34308204650878904, "memory(GiB)": 70.5, "step": 60240, "token_acc": 0.939209726443769, "train_speed(iter/s)": 1.447905 }, { "epoch": 2.5810805021207317, "grad_norm": 1.7302275896072388, "learning_rate": 4.7453882571291584e-05, "loss": 0.3614074468612671, "memory(GiB)": 70.5, "step": 60245, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.447899 }, { "epoch": 2.581294717449981, "grad_norm": 2.185835599899292, "learning_rate": 4.7447161552430526e-05, "loss": 0.3719921112060547, "memory(GiB)": 70.5, "step": 60250, "token_acc": 0.9498327759197325, "train_speed(iter/s)": 1.447901 }, { "epoch": 2.5815089327792298, "grad_norm": 1.9284125566482544, "learning_rate": 4.744044057981651e-05, "loss": 0.41980533599853515, "memory(GiB)": 70.5, "step": 60255, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.447901 }, { "epoch": 2.5817231481084786, "grad_norm": 3.5374372005462646, "learning_rate": 4.743371965357133e-05, "loss": 0.4420769691467285, "memory(GiB)": 70.5, "step": 60260, "token_acc": 0.9191616766467066, "train_speed(iter/s)": 1.44789 }, { "epoch": 2.581937363437728, "grad_norm": 4.022443771362305, "learning_rate": 4.742699877381673e-05, "loss": 0.522889232635498, "memory(GiB)": 70.5, "step": 60265, "token_acc": 0.8977635782747604, "train_speed(iter/s)": 1.447888 }, { "epoch": 2.5821515787669767, "grad_norm": 0.6864205598831177, "learning_rate": 4.7420277940674446e-05, "loss": 0.33846774101257326, "memory(GiB)": 70.5, "step": 60270, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.447886 }, { "epoch": 2.5823657940962255, "grad_norm": 3.242974281311035, "learning_rate": 4.741355715426623e-05, "loss": 0.4755284309387207, "memory(GiB)": 70.5, "step": 60275, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.447886 }, { "epoch": 2.5825800094254747, "grad_norm": 1.6700787544250488, "learning_rate": 4.7406836414713884e-05, "loss": 0.2544086456298828, "memory(GiB)": 70.5, "step": 60280, "token_acc": 0.9315589353612167, "train_speed(iter/s)": 1.447895 }, { "epoch": 2.5827942247547235, "grad_norm": 3.9087960720062256, "learning_rate": 4.7400115722139126e-05, "loss": 0.49435086250305177, "memory(GiB)": 70.5, "step": 60285, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.447896 }, { "epoch": 2.5830084400839723, "grad_norm": 2.6562886238098145, "learning_rate": 4.73933950766637e-05, "loss": 0.49962821006774905, "memory(GiB)": 70.5, "step": 60290, "token_acc": 0.8851351351351351, "train_speed(iter/s)": 1.447902 }, { "epoch": 2.5832226554132216, "grad_norm": 2.7710065841674805, "learning_rate": 4.738667447840938e-05, "loss": 0.4193872451782227, "memory(GiB)": 70.5, "step": 60295, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.447907 }, { "epoch": 2.5834368707424704, "grad_norm": 2.733945369720459, "learning_rate": 4.737995392749789e-05, "loss": 0.27397933006286623, "memory(GiB)": 70.5, "step": 60300, "token_acc": 0.9440559440559441, "train_speed(iter/s)": 1.447919 }, { "epoch": 2.5836510860717192, "grad_norm": 5.273605823516846, "learning_rate": 4.7373233424051e-05, "loss": 0.2678699493408203, "memory(GiB)": 70.5, "step": 60305, "token_acc": 0.9291666666666667, "train_speed(iter/s)": 1.447916 }, { "epoch": 2.5838653014009685, "grad_norm": 4.914461135864258, "learning_rate": 4.7366512968190454e-05, "loss": 0.2666964530944824, "memory(GiB)": 70.5, "step": 60310, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.447919 }, { "epoch": 2.5840795167302173, "grad_norm": 5.7336859703063965, "learning_rate": 4.735979256003798e-05, "loss": 0.6646286964416503, "memory(GiB)": 70.5, "step": 60315, "token_acc": 0.8510028653295129, "train_speed(iter/s)": 1.447922 }, { "epoch": 2.584293732059466, "grad_norm": 3.097414255142212, "learning_rate": 4.735307219971536e-05, "loss": 0.28466019630432127, "memory(GiB)": 70.5, "step": 60320, "token_acc": 0.9395770392749244, "train_speed(iter/s)": 1.447926 }, { "epoch": 2.5845079473887154, "grad_norm": 2.224999189376831, "learning_rate": 4.734635188734432e-05, "loss": 0.24528899192810058, "memory(GiB)": 70.5, "step": 60325, "token_acc": 0.9537366548042705, "train_speed(iter/s)": 1.447929 }, { "epoch": 2.584722162717964, "grad_norm": 4.774545192718506, "learning_rate": 4.7339631623046585e-05, "loss": 0.38866190910339354, "memory(GiB)": 70.5, "step": 60330, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.447932 }, { "epoch": 2.584936378047213, "grad_norm": 3.5021963119506836, "learning_rate": 4.7332911406943934e-05, "loss": 0.5829997062683105, "memory(GiB)": 70.5, "step": 60335, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.447942 }, { "epoch": 2.5851505933764622, "grad_norm": 3.051359176635742, "learning_rate": 4.732619123915809e-05, "loss": 0.7341436386108399, "memory(GiB)": 70.5, "step": 60340, "token_acc": 0.8477011494252874, "train_speed(iter/s)": 1.44794 }, { "epoch": 2.585364808705711, "grad_norm": 0.3335005044937134, "learning_rate": 4.7319471119810805e-05, "loss": 0.2635643482208252, "memory(GiB)": 70.5, "step": 60345, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.447946 }, { "epoch": 2.58557902403496, "grad_norm": 6.343498706817627, "learning_rate": 4.731275104902379e-05, "loss": 0.5091739177703858, "memory(GiB)": 70.5, "step": 60350, "token_acc": 0.8976608187134503, "train_speed(iter/s)": 1.447947 }, { "epoch": 2.585793239364209, "grad_norm": 8.01653003692627, "learning_rate": 4.730603102691884e-05, "loss": 0.5547235488891602, "memory(GiB)": 70.5, "step": 60355, "token_acc": 0.8738738738738738, "train_speed(iter/s)": 1.44795 }, { "epoch": 2.586007454693458, "grad_norm": 2.999973773956299, "learning_rate": 4.729931105361765e-05, "loss": 0.11593474149703979, "memory(GiB)": 70.5, "step": 60360, "token_acc": 0.9737704918032787, "train_speed(iter/s)": 1.44795 }, { "epoch": 2.5862216700227068, "grad_norm": 2.3081393241882324, "learning_rate": 4.7292591129241985e-05, "loss": 0.29404406547546386, "memory(GiB)": 70.5, "step": 60365, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.447961 }, { "epoch": 2.586435885351956, "grad_norm": 4.058884143829346, "learning_rate": 4.728587125391357e-05, "loss": 0.6605926036834717, "memory(GiB)": 70.5, "step": 60370, "token_acc": 0.8485915492957746, "train_speed(iter/s)": 1.447974 }, { "epoch": 2.586650100681205, "grad_norm": 3.030366897583008, "learning_rate": 4.727915142775414e-05, "loss": 0.5186580657958985, "memory(GiB)": 70.5, "step": 60375, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.447969 }, { "epoch": 2.5868643160104536, "grad_norm": 1.7996788024902344, "learning_rate": 4.7272431650885436e-05, "loss": 0.4500833034515381, "memory(GiB)": 70.5, "step": 60380, "token_acc": 0.889273356401384, "train_speed(iter/s)": 1.447976 }, { "epoch": 2.587078531339703, "grad_norm": 4.038987159729004, "learning_rate": 4.726571192342919e-05, "loss": 0.5077277183532715, "memory(GiB)": 70.5, "step": 60385, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.447988 }, { "epoch": 2.5872927466689517, "grad_norm": 1.9564566612243652, "learning_rate": 4.7258992245507134e-05, "loss": 0.3286271572113037, "memory(GiB)": 70.5, "step": 60390, "token_acc": 0.9169675090252708, "train_speed(iter/s)": 1.447988 }, { "epoch": 2.5875069619982005, "grad_norm": 4.410405158996582, "learning_rate": 4.725227261724101e-05, "loss": 0.3835892200469971, "memory(GiB)": 70.5, "step": 60395, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.447995 }, { "epoch": 2.5877211773274498, "grad_norm": 2.365983486175537, "learning_rate": 4.7245553038752535e-05, "loss": 0.524315071105957, "memory(GiB)": 70.5, "step": 60400, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.447998 }, { "epoch": 2.5879353926566986, "grad_norm": 4.99817419052124, "learning_rate": 4.7238833510163475e-05, "loss": 0.6506525993347168, "memory(GiB)": 70.5, "step": 60405, "token_acc": 0.8593272171253823, "train_speed(iter/s)": 1.448008 }, { "epoch": 2.5881496079859474, "grad_norm": 3.0099880695343018, "learning_rate": 4.723211403159552e-05, "loss": 0.3287004232406616, "memory(GiB)": 70.5, "step": 60410, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.448012 }, { "epoch": 2.5883638233151967, "grad_norm": 0.5817801356315613, "learning_rate": 4.722539460317041e-05, "loss": 0.2938045263290405, "memory(GiB)": 70.5, "step": 60415, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.448011 }, { "epoch": 2.5885780386444455, "grad_norm": 2.3794286251068115, "learning_rate": 4.721867522500989e-05, "loss": 0.4382997989654541, "memory(GiB)": 70.5, "step": 60420, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.448003 }, { "epoch": 2.5887922539736943, "grad_norm": 3.553734064102173, "learning_rate": 4.721195589723565e-05, "loss": 0.5794327735900879, "memory(GiB)": 70.5, "step": 60425, "token_acc": 0.8776978417266187, "train_speed(iter/s)": 1.448015 }, { "epoch": 2.5890064693029435, "grad_norm": 4.03762149810791, "learning_rate": 4.7205236619969474e-05, "loss": 0.4883126735687256, "memory(GiB)": 70.5, "step": 60430, "token_acc": 0.891156462585034, "train_speed(iter/s)": 1.448024 }, { "epoch": 2.5892206846321923, "grad_norm": 1.555511236190796, "learning_rate": 4.719851739333305e-05, "loss": 0.49085707664489747, "memory(GiB)": 70.5, "step": 60435, "token_acc": 0.8972602739726028, "train_speed(iter/s)": 1.448026 }, { "epoch": 2.589434899961441, "grad_norm": 1.5766478776931763, "learning_rate": 4.7191798217448115e-05, "loss": 0.1579961895942688, "memory(GiB)": 70.5, "step": 60440, "token_acc": 0.9562043795620438, "train_speed(iter/s)": 1.448022 }, { "epoch": 2.5896491152906904, "grad_norm": 3.8613152503967285, "learning_rate": 4.718507909243638e-05, "loss": 0.6547273635864258, "memory(GiB)": 70.5, "step": 60445, "token_acc": 0.8615384615384616, "train_speed(iter/s)": 1.448035 }, { "epoch": 2.5898633306199392, "grad_norm": 0.37722423672676086, "learning_rate": 4.7178360018419585e-05, "loss": 0.21776931285858153, "memory(GiB)": 70.5, "step": 60450, "token_acc": 0.9666666666666667, "train_speed(iter/s)": 1.448034 }, { "epoch": 2.590077545949188, "grad_norm": 2.233877658843994, "learning_rate": 4.717164099551945e-05, "loss": 0.22224733829498292, "memory(GiB)": 70.5, "step": 60455, "token_acc": 0.9578947368421052, "train_speed(iter/s)": 1.448046 }, { "epoch": 2.5902917612784373, "grad_norm": 5.721165657043457, "learning_rate": 4.7164922023857686e-05, "loss": 0.3255039691925049, "memory(GiB)": 70.5, "step": 60460, "token_acc": 0.9288389513108615, "train_speed(iter/s)": 1.448046 }, { "epoch": 2.590505976607686, "grad_norm": 1.6180912256240845, "learning_rate": 4.7158203103556026e-05, "loss": 0.2504727840423584, "memory(GiB)": 70.5, "step": 60465, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.448055 }, { "epoch": 2.590720191936935, "grad_norm": 0.3569997549057007, "learning_rate": 4.715148423473618e-05, "loss": 0.47603044509887693, "memory(GiB)": 70.5, "step": 60470, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.448065 }, { "epoch": 2.590934407266184, "grad_norm": 4.741114616394043, "learning_rate": 4.714476541751986e-05, "loss": 0.3079944610595703, "memory(GiB)": 70.5, "step": 60475, "token_acc": 0.9158249158249159, "train_speed(iter/s)": 1.448079 }, { "epoch": 2.591148622595433, "grad_norm": 3.7710187435150146, "learning_rate": 4.71380466520288e-05, "loss": 0.434136962890625, "memory(GiB)": 70.5, "step": 60480, "token_acc": 0.8723404255319149, "train_speed(iter/s)": 1.448103 }, { "epoch": 2.591362837924682, "grad_norm": 2.027892827987671, "learning_rate": 4.7131327938384706e-05, "loss": 0.46236701011657716, "memory(GiB)": 70.5, "step": 60485, "token_acc": 0.8909774436090225, "train_speed(iter/s)": 1.448117 }, { "epoch": 2.591577053253931, "grad_norm": 3.273721933364868, "learning_rate": 4.71246092767093e-05, "loss": 0.3173551082611084, "memory(GiB)": 70.5, "step": 60490, "token_acc": 0.9429657794676806, "train_speed(iter/s)": 1.448135 }, { "epoch": 2.59179126858318, "grad_norm": 2.8388357162475586, "learning_rate": 4.7117890667124306e-05, "loss": 0.46428093910217283, "memory(GiB)": 70.5, "step": 60495, "token_acc": 0.9115853658536586, "train_speed(iter/s)": 1.448134 }, { "epoch": 2.5920054839124287, "grad_norm": 1.9578522443771362, "learning_rate": 4.7111172109751394e-05, "loss": 0.2468388557434082, "memory(GiB)": 70.5, "step": 60500, "token_acc": 0.9467084639498433, "train_speed(iter/s)": 1.448141 }, { "epoch": 2.5920054839124287, "eval_loss": 2.6442084312438965, "eval_runtime": 13.4046, "eval_samples_per_second": 7.46, "eval_steps_per_second": 7.46, "eval_token_acc": 0.4451697127937337, "step": 60500 }, { "epoch": 2.592219699241678, "grad_norm": 5.455618858337402, "learning_rate": 4.7104453604712326e-05, "loss": 0.5210098266601563, "memory(GiB)": 70.5, "step": 60505, "token_acc": 0.560077519379845, "train_speed(iter/s)": 1.447647 }, { "epoch": 2.5924339145709268, "grad_norm": 1.9609270095825195, "learning_rate": 4.70977351521288e-05, "loss": 0.19270493984222412, "memory(GiB)": 70.5, "step": 60510, "token_acc": 0.9516129032258065, "train_speed(iter/s)": 1.447651 }, { "epoch": 2.5926481299001756, "grad_norm": 4.425224304199219, "learning_rate": 4.709101675212253e-05, "loss": 0.25781586170196535, "memory(GiB)": 70.5, "step": 60515, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.44765 }, { "epoch": 2.592862345229425, "grad_norm": 2.222116470336914, "learning_rate": 4.7084298404815206e-05, "loss": 0.3616279363632202, "memory(GiB)": 70.5, "step": 60520, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.447657 }, { "epoch": 2.5930765605586736, "grad_norm": 3.503779888153076, "learning_rate": 4.7077580110328566e-05, "loss": 0.3498535633087158, "memory(GiB)": 70.5, "step": 60525, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.447661 }, { "epoch": 2.5932907758879225, "grad_norm": 2.3220016956329346, "learning_rate": 4.7070861868784296e-05, "loss": 0.44051227569580076, "memory(GiB)": 70.5, "step": 60530, "token_acc": 0.8978328173374613, "train_speed(iter/s)": 1.447664 }, { "epoch": 2.5935049912171717, "grad_norm": 1.4216413497924805, "learning_rate": 4.706414368030412e-05, "loss": 0.3088659763336182, "memory(GiB)": 70.5, "step": 60535, "token_acc": 0.9486404833836858, "train_speed(iter/s)": 1.447666 }, { "epoch": 2.5937192065464205, "grad_norm": 4.3216938972473145, "learning_rate": 4.705742554500973e-05, "loss": 0.7728950023651123, "memory(GiB)": 70.5, "step": 60540, "token_acc": 0.8250950570342205, "train_speed(iter/s)": 1.447663 }, { "epoch": 2.5939334218756693, "grad_norm": 0.14809566736221313, "learning_rate": 4.705070746302283e-05, "loss": 0.29857447147369387, "memory(GiB)": 70.5, "step": 60545, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.44766 }, { "epoch": 2.5941476372049186, "grad_norm": 2.551111936569214, "learning_rate": 4.704398943446514e-05, "loss": 0.38049893379211425, "memory(GiB)": 70.5, "step": 60550, "token_acc": 0.90625, "train_speed(iter/s)": 1.447653 }, { "epoch": 2.5943618525341674, "grad_norm": 0.7078120708465576, "learning_rate": 4.703727145945836e-05, "loss": 0.4422726631164551, "memory(GiB)": 70.5, "step": 60555, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.447657 }, { "epoch": 2.594576067863416, "grad_norm": 1.32964289188385, "learning_rate": 4.7030553538124166e-05, "loss": 0.40230140686035154, "memory(GiB)": 70.5, "step": 60560, "token_acc": 0.9084507042253521, "train_speed(iter/s)": 1.447654 }, { "epoch": 2.5947902831926655, "grad_norm": 4.943514823913574, "learning_rate": 4.70238356705843e-05, "loss": 0.584286880493164, "memory(GiB)": 70.5, "step": 60565, "token_acc": 0.8848314606741573, "train_speed(iter/s)": 1.447666 }, { "epoch": 2.5950044985219143, "grad_norm": 3.8251378536224365, "learning_rate": 4.701711785696042e-05, "loss": 0.21801586151123048, "memory(GiB)": 70.5, "step": 60570, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.447669 }, { "epoch": 2.595218713851163, "grad_norm": 2.0930769443511963, "learning_rate": 4.7010400097374264e-05, "loss": 0.2556967258453369, "memory(GiB)": 70.5, "step": 60575, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.44767 }, { "epoch": 2.5954329291804124, "grad_norm": 4.63015604019165, "learning_rate": 4.7003682391947504e-05, "loss": 0.45989289283752444, "memory(GiB)": 70.5, "step": 60580, "token_acc": 0.9097744360902256, "train_speed(iter/s)": 1.44767 }, { "epoch": 2.595647144509661, "grad_norm": 4.47753381729126, "learning_rate": 4.699696474080186e-05, "loss": 0.48909816741943357, "memory(GiB)": 70.5, "step": 60585, "token_acc": 0.9039145907473309, "train_speed(iter/s)": 1.44766 }, { "epoch": 2.59586135983891, "grad_norm": 2.4120049476623535, "learning_rate": 4.699024714405901e-05, "loss": 0.6141268253326416, "memory(GiB)": 70.5, "step": 60590, "token_acc": 0.8673139158576052, "train_speed(iter/s)": 1.447663 }, { "epoch": 2.5960755751681592, "grad_norm": 5.437417030334473, "learning_rate": 4.698352960184067e-05, "loss": 0.4967046737670898, "memory(GiB)": 70.5, "step": 60595, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.447656 }, { "epoch": 2.596289790497408, "grad_norm": 1.0533580780029297, "learning_rate": 4.697681211426851e-05, "loss": 0.5386630535125733, "memory(GiB)": 70.5, "step": 60600, "token_acc": 0.8782051282051282, "train_speed(iter/s)": 1.447662 }, { "epoch": 2.596504005826657, "grad_norm": 1.635386347770691, "learning_rate": 4.697009468146423e-05, "loss": 0.1809403657913208, "memory(GiB)": 70.5, "step": 60605, "token_acc": 0.9633333333333334, "train_speed(iter/s)": 1.447671 }, { "epoch": 2.596718221155906, "grad_norm": 1.9664092063903809, "learning_rate": 4.6963377303549546e-05, "loss": 0.28527662754058836, "memory(GiB)": 70.5, "step": 60610, "token_acc": 0.9357429718875502, "train_speed(iter/s)": 1.447678 }, { "epoch": 2.596932436485155, "grad_norm": 1.3140555620193481, "learning_rate": 4.6956659980646125e-05, "loss": 0.3305500030517578, "memory(GiB)": 70.5, "step": 60615, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.447671 }, { "epoch": 2.5971466518144037, "grad_norm": 3.8656704425811768, "learning_rate": 4.6949942712875645e-05, "loss": 0.5263751983642578, "memory(GiB)": 70.5, "step": 60620, "token_acc": 0.875, "train_speed(iter/s)": 1.447682 }, { "epoch": 2.597360867143653, "grad_norm": 10.720698356628418, "learning_rate": 4.6943225500359834e-05, "loss": 0.289869499206543, "memory(GiB)": 70.5, "step": 60625, "token_acc": 0.9283387622149837, "train_speed(iter/s)": 1.447679 }, { "epoch": 2.597575082472902, "grad_norm": 2.902771234512329, "learning_rate": 4.693650834322034e-05, "loss": 0.2443085193634033, "memory(GiB)": 70.5, "step": 60630, "token_acc": 0.9507692307692308, "train_speed(iter/s)": 1.447681 }, { "epoch": 2.5977892978021506, "grad_norm": 4.407103538513184, "learning_rate": 4.6929791241578894e-05, "loss": 0.34438519477844237, "memory(GiB)": 70.5, "step": 60635, "token_acc": 0.9163346613545816, "train_speed(iter/s)": 1.447683 }, { "epoch": 2.5980035131314, "grad_norm": 3.4050419330596924, "learning_rate": 4.6923074195557146e-05, "loss": 0.4900501251220703, "memory(GiB)": 70.5, "step": 60640, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.447696 }, { "epoch": 2.5982177284606487, "grad_norm": 3.9710922241210938, "learning_rate": 4.691635720527679e-05, "loss": 0.3450563907623291, "memory(GiB)": 70.5, "step": 60645, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.447699 }, { "epoch": 2.5984319437898975, "grad_norm": 2.6441543102264404, "learning_rate": 4.690964027085951e-05, "loss": 0.3017172336578369, "memory(GiB)": 70.5, "step": 60650, "token_acc": 0.9470198675496688, "train_speed(iter/s)": 1.447699 }, { "epoch": 2.5986461591191468, "grad_norm": 5.007290840148926, "learning_rate": 4.6902923392427014e-05, "loss": 0.3038146734237671, "memory(GiB)": 70.5, "step": 60655, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.447714 }, { "epoch": 2.5988603744483956, "grad_norm": 6.709249973297119, "learning_rate": 4.689620657010097e-05, "loss": 0.6449096202850342, "memory(GiB)": 70.5, "step": 60660, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.447722 }, { "epoch": 2.5990745897776444, "grad_norm": 3.652158737182617, "learning_rate": 4.688948980400304e-05, "loss": 0.2705942153930664, "memory(GiB)": 70.5, "step": 60665, "token_acc": 0.9389067524115756, "train_speed(iter/s)": 1.447716 }, { "epoch": 2.5992888051068936, "grad_norm": 4.2619309425354, "learning_rate": 4.688277309425494e-05, "loss": 0.4591989517211914, "memory(GiB)": 70.5, "step": 60670, "token_acc": 0.9131944444444444, "train_speed(iter/s)": 1.447734 }, { "epoch": 2.5995030204361425, "grad_norm": 5.205097198486328, "learning_rate": 4.68760564409783e-05, "loss": 0.5499157905578613, "memory(GiB)": 70.5, "step": 60675, "token_acc": 0.8962536023054755, "train_speed(iter/s)": 1.447744 }, { "epoch": 2.5997172357653913, "grad_norm": 2.482200860977173, "learning_rate": 4.686933984429485e-05, "loss": 0.27013206481933594, "memory(GiB)": 70.5, "step": 60680, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.447745 }, { "epoch": 2.5999314510946405, "grad_norm": 3.5334665775299072, "learning_rate": 4.686262330432625e-05, "loss": 0.5267004013061524, "memory(GiB)": 70.5, "step": 60685, "token_acc": 0.8943396226415095, "train_speed(iter/s)": 1.447744 }, { "epoch": 2.6001456664238893, "grad_norm": 5.26168155670166, "learning_rate": 4.685590682119415e-05, "loss": 0.5887321949005127, "memory(GiB)": 70.5, "step": 60690, "token_acc": 0.872852233676976, "train_speed(iter/s)": 1.447743 }, { "epoch": 2.600359881753138, "grad_norm": 3.7986979484558105, "learning_rate": 4.684919039502027e-05, "loss": 0.44018120765686036, "memory(GiB)": 70.5, "step": 60695, "token_acc": 0.8984962406015038, "train_speed(iter/s)": 1.447755 }, { "epoch": 2.6005740970823874, "grad_norm": 4.495683670043945, "learning_rate": 4.6842474025926255e-05, "loss": 0.41028175354003904, "memory(GiB)": 70.5, "step": 60700, "token_acc": 0.924, "train_speed(iter/s)": 1.447757 }, { "epoch": 2.600788312411636, "grad_norm": 2.4619457721710205, "learning_rate": 4.683575771403377e-05, "loss": 0.2892031669616699, "memory(GiB)": 70.5, "step": 60705, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.447762 }, { "epoch": 2.601002527740885, "grad_norm": 0.7367138862609863, "learning_rate": 4.6829041459464525e-05, "loss": 0.16452488899230958, "memory(GiB)": 70.5, "step": 60710, "token_acc": 0.9495268138801262, "train_speed(iter/s)": 1.447759 }, { "epoch": 2.6012167430701343, "grad_norm": 1.2945938110351562, "learning_rate": 4.6822325262340147e-05, "loss": 0.6337702751159668, "memory(GiB)": 70.5, "step": 60715, "token_acc": 0.8495821727019499, "train_speed(iter/s)": 1.447747 }, { "epoch": 2.601430958399383, "grad_norm": 3.2140657901763916, "learning_rate": 4.6815609122782336e-05, "loss": 0.40223207473754885, "memory(GiB)": 70.5, "step": 60720, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.44776 }, { "epoch": 2.601645173728632, "grad_norm": 2.1076714992523193, "learning_rate": 4.680889304091275e-05, "loss": 0.18363965749740602, "memory(GiB)": 70.5, "step": 60725, "token_acc": 0.959731543624161, "train_speed(iter/s)": 1.44777 }, { "epoch": 2.601859389057881, "grad_norm": 5.327023029327393, "learning_rate": 4.680217701685306e-05, "loss": 0.5576025009155273, "memory(GiB)": 70.5, "step": 60730, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.447769 }, { "epoch": 2.60207360438713, "grad_norm": 3.61332631111145, "learning_rate": 4.679546105072494e-05, "loss": 0.5410922050476075, "memory(GiB)": 70.5, "step": 60735, "token_acc": 0.8817204301075269, "train_speed(iter/s)": 1.447768 }, { "epoch": 2.602287819716379, "grad_norm": 1.7338221073150635, "learning_rate": 4.678874514265005e-05, "loss": 0.22180099487304689, "memory(GiB)": 70.5, "step": 60740, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.447766 }, { "epoch": 2.602502035045628, "grad_norm": 3.031604766845703, "learning_rate": 4.6782029292750056e-05, "loss": 0.3083348751068115, "memory(GiB)": 70.5, "step": 60745, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.447761 }, { "epoch": 2.602716250374877, "grad_norm": 2.0370779037475586, "learning_rate": 4.677531350114661e-05, "loss": 0.34260537624359133, "memory(GiB)": 70.5, "step": 60750, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.447766 }, { "epoch": 2.6029304657041257, "grad_norm": 0.49238574504852295, "learning_rate": 4.6768597767961395e-05, "loss": 0.20678403377532958, "memory(GiB)": 70.5, "step": 60755, "token_acc": 0.954248366013072, "train_speed(iter/s)": 1.44777 }, { "epoch": 2.603144681033375, "grad_norm": 7.070004463195801, "learning_rate": 4.676188209331606e-05, "loss": 0.6888382911682129, "memory(GiB)": 70.5, "step": 60760, "token_acc": 0.8320610687022901, "train_speed(iter/s)": 1.447767 }, { "epoch": 2.6033588963626237, "grad_norm": 3.8945791721343994, "learning_rate": 4.675516647733226e-05, "loss": 0.36170191764831544, "memory(GiB)": 70.5, "step": 60765, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.447766 }, { "epoch": 2.6035731116918726, "grad_norm": 1.4458872079849243, "learning_rate": 4.674845092013168e-05, "loss": 0.3155383586883545, "memory(GiB)": 70.5, "step": 60770, "token_acc": 0.9191616766467066, "train_speed(iter/s)": 1.447765 }, { "epoch": 2.603787327021122, "grad_norm": 0.10186082124710083, "learning_rate": 4.674173542183594e-05, "loss": 0.3137260913848877, "memory(GiB)": 70.5, "step": 60775, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.447756 }, { "epoch": 2.6040015423503706, "grad_norm": 1.5183030366897583, "learning_rate": 4.673501998256674e-05, "loss": 0.3031401872634888, "memory(GiB)": 70.5, "step": 60780, "token_acc": 0.9378531073446328, "train_speed(iter/s)": 1.44776 }, { "epoch": 2.6042157576796194, "grad_norm": 2.4537510871887207, "learning_rate": 4.672830460244571e-05, "loss": 0.34311106204986574, "memory(GiB)": 70.5, "step": 60785, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.447756 }, { "epoch": 2.6044299730088687, "grad_norm": 4.552237033843994, "learning_rate": 4.6721589281594504e-05, "loss": 0.6056631088256836, "memory(GiB)": 70.5, "step": 60790, "token_acc": 0.8793774319066148, "train_speed(iter/s)": 1.447767 }, { "epoch": 2.6046441883381175, "grad_norm": 3.182973623275757, "learning_rate": 4.671487402013477e-05, "loss": 0.4031537532806396, "memory(GiB)": 70.5, "step": 60795, "token_acc": 0.9093851132686084, "train_speed(iter/s)": 1.447766 }, { "epoch": 2.6048584036673663, "grad_norm": 6.786487102508545, "learning_rate": 4.6708158818188195e-05, "loss": 0.25377497673034666, "memory(GiB)": 70.5, "step": 60800, "token_acc": 0.940677966101695, "train_speed(iter/s)": 1.447767 }, { "epoch": 2.6050726189966156, "grad_norm": 1.8015286922454834, "learning_rate": 4.670144367587642e-05, "loss": 0.4555095672607422, "memory(GiB)": 70.5, "step": 60805, "token_acc": 0.910958904109589, "train_speed(iter/s)": 1.447768 }, { "epoch": 2.6052868343258644, "grad_norm": 5.0275163650512695, "learning_rate": 4.669472859332106e-05, "loss": 0.6324678421020508, "memory(GiB)": 70.5, "step": 60810, "token_acc": 0.8291814946619217, "train_speed(iter/s)": 1.447768 }, { "epoch": 2.605501049655113, "grad_norm": 3.30828595161438, "learning_rate": 4.668801357064382e-05, "loss": 0.3066544055938721, "memory(GiB)": 70.5, "step": 60815, "token_acc": 0.9356913183279743, "train_speed(iter/s)": 1.447772 }, { "epoch": 2.6057152649843625, "grad_norm": 3.501913547515869, "learning_rate": 4.66812986079663e-05, "loss": 0.48357553482055665, "memory(GiB)": 70.5, "step": 60820, "token_acc": 0.9255014326647565, "train_speed(iter/s)": 1.447784 }, { "epoch": 2.6059294803136113, "grad_norm": 1.6611212491989136, "learning_rate": 4.667458370541018e-05, "loss": 0.3010150909423828, "memory(GiB)": 70.5, "step": 60825, "token_acc": 0.936026936026936, "train_speed(iter/s)": 1.447793 }, { "epoch": 2.60614369564286, "grad_norm": 3.157522439956665, "learning_rate": 4.6667868863097096e-05, "loss": 0.46028976440429686, "memory(GiB)": 70.5, "step": 60830, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.447785 }, { "epoch": 2.6063579109721093, "grad_norm": 2.0559728145599365, "learning_rate": 4.666115408114869e-05, "loss": 0.3169586658477783, "memory(GiB)": 70.5, "step": 60835, "token_acc": 0.9226006191950464, "train_speed(iter/s)": 1.447792 }, { "epoch": 2.606572126301358, "grad_norm": 3.592005491256714, "learning_rate": 4.665443935968662e-05, "loss": 0.5828954219818115, "memory(GiB)": 70.5, "step": 60840, "token_acc": 0.8810289389067524, "train_speed(iter/s)": 1.447808 }, { "epoch": 2.606786341630607, "grad_norm": 2.944977045059204, "learning_rate": 4.664772469883251e-05, "loss": 0.5570123672485352, "memory(GiB)": 70.5, "step": 60845, "token_acc": 0.8841059602649006, "train_speed(iter/s)": 1.447827 }, { "epoch": 2.607000556959856, "grad_norm": 4.093582630157471, "learning_rate": 4.6641010098708006e-05, "loss": 0.3378610610961914, "memory(GiB)": 70.5, "step": 60850, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.447828 }, { "epoch": 2.607214772289105, "grad_norm": 2.312037229537964, "learning_rate": 4.6634295559434756e-05, "loss": 0.32152981758117677, "memory(GiB)": 70.5, "step": 60855, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.447832 }, { "epoch": 2.607428987618354, "grad_norm": 5.2473859786987305, "learning_rate": 4.66275810811344e-05, "loss": 0.3803919792175293, "memory(GiB)": 70.5, "step": 60860, "token_acc": 0.9063545150501672, "train_speed(iter/s)": 1.447834 }, { "epoch": 2.607643202947603, "grad_norm": 2.344916820526123, "learning_rate": 4.662086666392859e-05, "loss": 0.2897505521774292, "memory(GiB)": 70.5, "step": 60865, "token_acc": 0.932806324110672, "train_speed(iter/s)": 1.447831 }, { "epoch": 2.607857418276852, "grad_norm": 1.085307240486145, "learning_rate": 4.661415230793892e-05, "loss": 0.4487272262573242, "memory(GiB)": 70.5, "step": 60870, "token_acc": 0.9057750759878419, "train_speed(iter/s)": 1.447846 }, { "epoch": 2.6080716336061007, "grad_norm": 1.8810433149337769, "learning_rate": 4.660743801328709e-05, "loss": 0.27440452575683594, "memory(GiB)": 70.5, "step": 60875, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.447844 }, { "epoch": 2.60828584893535, "grad_norm": 3.7924022674560547, "learning_rate": 4.6600723780094677e-05, "loss": 0.5172390937805176, "memory(GiB)": 70.5, "step": 60880, "token_acc": 0.9033232628398792, "train_speed(iter/s)": 1.447846 }, { "epoch": 2.608500064264599, "grad_norm": 2.473771095275879, "learning_rate": 4.659400960848336e-05, "loss": 0.32001686096191406, "memory(GiB)": 70.5, "step": 60885, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.44785 }, { "epoch": 2.6087142795938476, "grad_norm": 5.24526309967041, "learning_rate": 4.658729549857476e-05, "loss": 0.23298847675323486, "memory(GiB)": 70.5, "step": 60890, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.447848 }, { "epoch": 2.608928494923097, "grad_norm": 2.500899314880371, "learning_rate": 4.658058145049048e-05, "loss": 0.3936495065689087, "memory(GiB)": 70.5, "step": 60895, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.44785 }, { "epoch": 2.6091427102523457, "grad_norm": 5.952851295471191, "learning_rate": 4.65738674643522e-05, "loss": 0.33632917404174806, "memory(GiB)": 70.5, "step": 60900, "token_acc": 0.9280575539568345, "train_speed(iter/s)": 1.447849 }, { "epoch": 2.6093569255815945, "grad_norm": 3.7259738445281982, "learning_rate": 4.656715354028151e-05, "loss": 0.5246819496154785, "memory(GiB)": 70.5, "step": 60905, "token_acc": 0.9058441558441559, "train_speed(iter/s)": 1.447849 }, { "epoch": 2.6095711409108437, "grad_norm": 2.985167980194092, "learning_rate": 4.656043967840005e-05, "loss": 0.37523512840270995, "memory(GiB)": 70.5, "step": 60910, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.447852 }, { "epoch": 2.6097853562400926, "grad_norm": 5.155113697052002, "learning_rate": 4.655372587882946e-05, "loss": 0.504874849319458, "memory(GiB)": 70.5, "step": 60915, "token_acc": 0.8740740740740741, "train_speed(iter/s)": 1.44785 }, { "epoch": 2.6099995715693414, "grad_norm": 3.18463397026062, "learning_rate": 4.654701214169135e-05, "loss": 0.47247848510742185, "memory(GiB)": 70.5, "step": 60920, "token_acc": 0.9, "train_speed(iter/s)": 1.447857 }, { "epoch": 2.6102137868985906, "grad_norm": 5.03803014755249, "learning_rate": 4.654029846710737e-05, "loss": 0.45287518501281737, "memory(GiB)": 70.5, "step": 60925, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.44786 }, { "epoch": 2.6104280022278394, "grad_norm": 2.4391653537750244, "learning_rate": 4.653358485519912e-05, "loss": 0.467210578918457, "memory(GiB)": 70.5, "step": 60930, "token_acc": 0.8969072164948454, "train_speed(iter/s)": 1.447878 }, { "epoch": 2.6106422175570883, "grad_norm": 2.162118434906006, "learning_rate": 4.652687130608823e-05, "loss": 0.12419147491455078, "memory(GiB)": 70.5, "step": 60935, "token_acc": 0.972972972972973, "train_speed(iter/s)": 1.447877 }, { "epoch": 2.6108564328863375, "grad_norm": 4.262059688568115, "learning_rate": 4.652015781989631e-05, "loss": 0.1827659249305725, "memory(GiB)": 70.5, "step": 60940, "token_acc": 0.9633699633699634, "train_speed(iter/s)": 1.447885 }, { "epoch": 2.6110706482155863, "grad_norm": 2.880875825881958, "learning_rate": 4.651344439674501e-05, "loss": 0.39275124073028567, "memory(GiB)": 70.5, "step": 60945, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.447888 }, { "epoch": 2.611284863544835, "grad_norm": 0.837633490562439, "learning_rate": 4.650673103675594e-05, "loss": 0.2944060802459717, "memory(GiB)": 70.5, "step": 60950, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.44789 }, { "epoch": 2.6114990788740844, "grad_norm": 3.201894521713257, "learning_rate": 4.6500017740050706e-05, "loss": 0.21110835075378417, "memory(GiB)": 70.5, "step": 60955, "token_acc": 0.9568345323741008, "train_speed(iter/s)": 1.447888 }, { "epoch": 2.611713294203333, "grad_norm": 4.154411315917969, "learning_rate": 4.649330450675095e-05, "loss": 0.34996752738952636, "memory(GiB)": 70.5, "step": 60960, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.447899 }, { "epoch": 2.611927509532582, "grad_norm": 3.238879919052124, "learning_rate": 4.6486591336978246e-05, "loss": 0.5037455558776855, "memory(GiB)": 70.5, "step": 60965, "token_acc": 0.8664122137404581, "train_speed(iter/s)": 1.447911 }, { "epoch": 2.6121417248618313, "grad_norm": 0.6944129467010498, "learning_rate": 4.647987823085426e-05, "loss": 0.25643789768218994, "memory(GiB)": 70.5, "step": 60970, "token_acc": 0.946031746031746, "train_speed(iter/s)": 1.447913 }, { "epoch": 2.61235594019108, "grad_norm": 3.059786558151245, "learning_rate": 4.647316518850058e-05, "loss": 0.4083270072937012, "memory(GiB)": 70.5, "step": 60975, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.447927 }, { "epoch": 2.612570155520329, "grad_norm": 8.93424129486084, "learning_rate": 4.64664522100388e-05, "loss": 0.47240409851074217, "memory(GiB)": 70.5, "step": 60980, "token_acc": 0.9179331306990881, "train_speed(iter/s)": 1.447939 }, { "epoch": 2.612784370849578, "grad_norm": 4.825551509857178, "learning_rate": 4.645973929559058e-05, "loss": 0.46715087890625, "memory(GiB)": 70.5, "step": 60985, "token_acc": 0.8951048951048951, "train_speed(iter/s)": 1.44794 }, { "epoch": 2.612998586178827, "grad_norm": 3.7721621990203857, "learning_rate": 4.645302644527749e-05, "loss": 0.39283850193023684, "memory(GiB)": 70.5, "step": 60990, "token_acc": 0.934375, "train_speed(iter/s)": 1.447953 }, { "epoch": 2.6132128015080758, "grad_norm": 7.77725887298584, "learning_rate": 4.644631365922114e-05, "loss": 0.2526139736175537, "memory(GiB)": 70.5, "step": 60995, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.447952 }, { "epoch": 2.613427016837325, "grad_norm": 6.5873823165893555, "learning_rate": 4.643960093754318e-05, "loss": 0.60037202835083, "memory(GiB)": 70.5, "step": 61000, "token_acc": 0.8798449612403101, "train_speed(iter/s)": 1.447949 }, { "epoch": 2.613427016837325, "eval_loss": 2.461714267730713, "eval_runtime": 13.3333, "eval_samples_per_second": 7.5, "eval_steps_per_second": 7.5, "eval_token_acc": 0.4368131868131868, "step": 61000 }, { "epoch": 2.613641232166574, "grad_norm": 2.6111600399017334, "learning_rate": 4.643288828036517e-05, "loss": 0.5382713317871094, "memory(GiB)": 70.5, "step": 61005, "token_acc": 0.5736137667304015, "train_speed(iter/s)": 1.44745 }, { "epoch": 2.6138554474958227, "grad_norm": 3.1024057865142822, "learning_rate": 4.642617568780874e-05, "loss": 0.27578437328338623, "memory(GiB)": 70.5, "step": 61010, "token_acc": 0.9504504504504504, "train_speed(iter/s)": 1.44745 }, { "epoch": 2.614069662825072, "grad_norm": 6.908020973205566, "learning_rate": 4.6419463159995474e-05, "loss": 0.5480142593383789, "memory(GiB)": 70.5, "step": 61015, "token_acc": 0.890625, "train_speed(iter/s)": 1.447456 }, { "epoch": 2.6142838781543207, "grad_norm": 4.949659824371338, "learning_rate": 4.6412750697047014e-05, "loss": 0.7011131286621094, "memory(GiB)": 70.5, "step": 61020, "token_acc": 0.8743589743589744, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.6144980934835695, "grad_norm": 3.3422505855560303, "learning_rate": 4.640603829908493e-05, "loss": 0.36664843559265137, "memory(GiB)": 70.5, "step": 61025, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.447461 }, { "epoch": 2.614712308812819, "grad_norm": 0.6027071475982666, "learning_rate": 4.6399325966230835e-05, "loss": 0.25500025749206545, "memory(GiB)": 70.5, "step": 61030, "token_acc": 0.9392857142857143, "train_speed(iter/s)": 1.447466 }, { "epoch": 2.6149265241420676, "grad_norm": 3.85408878326416, "learning_rate": 4.6392613698606336e-05, "loss": 0.33144607543945315, "memory(GiB)": 70.5, "step": 61035, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.447467 }, { "epoch": 2.6151407394713164, "grad_norm": 5.336202144622803, "learning_rate": 4.638590149633301e-05, "loss": 0.4289818286895752, "memory(GiB)": 70.5, "step": 61040, "token_acc": 0.89937106918239, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.6153549548005657, "grad_norm": 0.08405853807926178, "learning_rate": 4.637918935953248e-05, "loss": 0.30386712551116946, "memory(GiB)": 70.5, "step": 61045, "token_acc": 0.93359375, "train_speed(iter/s)": 1.447472 }, { "epoch": 2.6155691701298145, "grad_norm": 3.3498878479003906, "learning_rate": 4.637247728832633e-05, "loss": 0.24630346298217773, "memory(GiB)": 70.5, "step": 61050, "token_acc": 0.9380804953560371, "train_speed(iter/s)": 1.44749 }, { "epoch": 2.6157833854590633, "grad_norm": 2.81543231010437, "learning_rate": 4.636576528283615e-05, "loss": 0.24284920692443848, "memory(GiB)": 70.5, "step": 61055, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.447497 }, { "epoch": 2.6159976007883126, "grad_norm": 3.7184712886810303, "learning_rate": 4.635905334318355e-05, "loss": 0.3060037851333618, "memory(GiB)": 70.5, "step": 61060, "token_acc": 0.9251968503937008, "train_speed(iter/s)": 1.447513 }, { "epoch": 2.6162118161175614, "grad_norm": 2.512726068496704, "learning_rate": 4.635234146949011e-05, "loss": 0.30994091033935545, "memory(GiB)": 70.5, "step": 61065, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.447509 }, { "epoch": 2.61642603144681, "grad_norm": 1.0270386934280396, "learning_rate": 4.6345629661877425e-05, "loss": 0.13326451778411866, "memory(GiB)": 70.5, "step": 61070, "token_acc": 0.9705882352941176, "train_speed(iter/s)": 1.447522 }, { "epoch": 2.6166402467760594, "grad_norm": 4.618277072906494, "learning_rate": 4.6338917920467096e-05, "loss": 0.27915329933166505, "memory(GiB)": 70.5, "step": 61075, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.447525 }, { "epoch": 2.6168544621053083, "grad_norm": 2.306737184524536, "learning_rate": 4.633220624538069e-05, "loss": 0.2820326805114746, "memory(GiB)": 70.5, "step": 61080, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.447526 }, { "epoch": 2.617068677434557, "grad_norm": 2.9499623775482178, "learning_rate": 4.632549463673981e-05, "loss": 0.3850246906280518, "memory(GiB)": 70.5, "step": 61085, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.44754 }, { "epoch": 2.6172828927638063, "grad_norm": 4.513622760772705, "learning_rate": 4.631878309466603e-05, "loss": 0.6449522018432617, "memory(GiB)": 70.5, "step": 61090, "token_acc": 0.8587896253602305, "train_speed(iter/s)": 1.447557 }, { "epoch": 2.617497108093055, "grad_norm": 2.218764066696167, "learning_rate": 4.6312071619280974e-05, "loss": 0.19780592918395995, "memory(GiB)": 70.5, "step": 61095, "token_acc": 0.955719557195572, "train_speed(iter/s)": 1.447553 }, { "epoch": 2.617711323422304, "grad_norm": 3.4889888763427734, "learning_rate": 4.630536021070618e-05, "loss": 0.46076068878173826, "memory(GiB)": 70.5, "step": 61100, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.447568 }, { "epoch": 2.617925538751553, "grad_norm": 3.520646810531616, "learning_rate": 4.629864886906326e-05, "loss": 0.3748875617980957, "memory(GiB)": 70.5, "step": 61105, "token_acc": 0.9188034188034188, "train_speed(iter/s)": 1.447567 }, { "epoch": 2.618139754080802, "grad_norm": 7.738424777984619, "learning_rate": 4.629193759447378e-05, "loss": 0.07544230222702027, "memory(GiB)": 70.5, "step": 61110, "token_acc": 0.9753521126760564, "train_speed(iter/s)": 1.447586 }, { "epoch": 2.618353969410051, "grad_norm": 7.525439739227295, "learning_rate": 4.628522638705934e-05, "loss": 0.25493407249450684, "memory(GiB)": 70.5, "step": 61115, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.447578 }, { "epoch": 2.6185681847393, "grad_norm": 4.958010196685791, "learning_rate": 4.6278515246941514e-05, "loss": 0.39889748096466066, "memory(GiB)": 70.5, "step": 61120, "token_acc": 0.9201388888888888, "train_speed(iter/s)": 1.447577 }, { "epoch": 2.618782400068549, "grad_norm": 2.4816713333129883, "learning_rate": 4.6271804174241864e-05, "loss": 0.14290163516998292, "memory(GiB)": 70.5, "step": 61125, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.447574 }, { "epoch": 2.6189966153977977, "grad_norm": 2.412806510925293, "learning_rate": 4.6265093169081986e-05, "loss": 0.23992767333984374, "memory(GiB)": 70.5, "step": 61130, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.447576 }, { "epoch": 2.619210830727047, "grad_norm": 4.461517333984375, "learning_rate": 4.625838223158345e-05, "loss": 0.3668162107467651, "memory(GiB)": 70.5, "step": 61135, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.447591 }, { "epoch": 2.619425046056296, "grad_norm": 2.872967481613159, "learning_rate": 4.6251671361867813e-05, "loss": 0.38693130016326904, "memory(GiB)": 70.5, "step": 61140, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.447587 }, { "epoch": 2.6196392613855446, "grad_norm": 3.1546249389648438, "learning_rate": 4.6244960560056686e-05, "loss": 0.4322534084320068, "memory(GiB)": 70.5, "step": 61145, "token_acc": 0.9274447949526814, "train_speed(iter/s)": 1.447593 }, { "epoch": 2.619853476714794, "grad_norm": 4.419399261474609, "learning_rate": 4.623824982627161e-05, "loss": 0.5643083095550537, "memory(GiB)": 70.5, "step": 61150, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.447599 }, { "epoch": 2.6200676920440427, "grad_norm": 2.526902437210083, "learning_rate": 4.623153916063417e-05, "loss": 0.32895348072052, "memory(GiB)": 70.5, "step": 61155, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.447589 }, { "epoch": 2.6202819073732915, "grad_norm": 4.858348369598389, "learning_rate": 4.6224828563265934e-05, "loss": 0.4167595863342285, "memory(GiB)": 70.5, "step": 61160, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.447602 }, { "epoch": 2.6204961227025407, "grad_norm": 1.1871503591537476, "learning_rate": 4.6218118034288456e-05, "loss": 0.19755810499191284, "memory(GiB)": 70.5, "step": 61165, "token_acc": 0.9529085872576177, "train_speed(iter/s)": 1.447603 }, { "epoch": 2.6207103380317895, "grad_norm": 3.7741103172302246, "learning_rate": 4.6211407573823326e-05, "loss": 0.393303656578064, "memory(GiB)": 70.5, "step": 61170, "token_acc": 0.9281437125748503, "train_speed(iter/s)": 1.447602 }, { "epoch": 2.6209245533610384, "grad_norm": 2.7356488704681396, "learning_rate": 4.620469718199211e-05, "loss": 0.2894193172454834, "memory(GiB)": 70.5, "step": 61175, "token_acc": 0.9188405797101449, "train_speed(iter/s)": 1.44761 }, { "epoch": 2.6211387686902876, "grad_norm": 2.7515976428985596, "learning_rate": 4.619798685891637e-05, "loss": 0.39373738765716554, "memory(GiB)": 70.5, "step": 61180, "token_acc": 0.9, "train_speed(iter/s)": 1.447614 }, { "epoch": 2.6213529840195364, "grad_norm": 2.151256561279297, "learning_rate": 4.6191276604717645e-05, "loss": 0.1543282151222229, "memory(GiB)": 70.5, "step": 61185, "token_acc": 0.9571984435797666, "train_speed(iter/s)": 1.447623 }, { "epoch": 2.6215671993487852, "grad_norm": 4.776877403259277, "learning_rate": 4.618456641951753e-05, "loss": 0.3121748447418213, "memory(GiB)": 70.5, "step": 61190, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.447635 }, { "epoch": 2.6217814146780345, "grad_norm": 4.848793029785156, "learning_rate": 4.617785630343757e-05, "loss": 0.6679186820983887, "memory(GiB)": 70.5, "step": 61195, "token_acc": 0.8533333333333334, "train_speed(iter/s)": 1.447647 }, { "epoch": 2.6219956300072833, "grad_norm": 8.620506286621094, "learning_rate": 4.617114625659932e-05, "loss": 0.3462846040725708, "memory(GiB)": 70.5, "step": 61200, "token_acc": 0.9197530864197531, "train_speed(iter/s)": 1.447641 }, { "epoch": 2.622209845336532, "grad_norm": 1.1203548908233643, "learning_rate": 4.6164436279124366e-05, "loss": 0.3764320135116577, "memory(GiB)": 70.5, "step": 61205, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.447651 }, { "epoch": 2.6224240606657814, "grad_norm": 3.275834083557129, "learning_rate": 4.615772637113423e-05, "loss": 0.29796297550201417, "memory(GiB)": 70.5, "step": 61210, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.447648 }, { "epoch": 2.62263827599503, "grad_norm": 2.415708541870117, "learning_rate": 4.61510165327505e-05, "loss": 0.34885492324829104, "memory(GiB)": 70.5, "step": 61215, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.447653 }, { "epoch": 2.622852491324279, "grad_norm": 3.5703694820404053, "learning_rate": 4.6144306764094714e-05, "loss": 0.1636880159378052, "memory(GiB)": 70.5, "step": 61220, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.447653 }, { "epoch": 2.6230667066535283, "grad_norm": 2.6125476360321045, "learning_rate": 4.613759706528842e-05, "loss": 0.3171872138977051, "memory(GiB)": 70.5, "step": 61225, "token_acc": 0.9314516129032258, "train_speed(iter/s)": 1.447651 }, { "epoch": 2.623280921982777, "grad_norm": 3.5805389881134033, "learning_rate": 4.613088743645318e-05, "loss": 0.4536094665527344, "memory(GiB)": 70.5, "step": 61230, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.44766 }, { "epoch": 2.623495137312026, "grad_norm": 4.039298057556152, "learning_rate": 4.612417787771055e-05, "loss": 0.19941155910491942, "memory(GiB)": 70.5, "step": 61235, "token_acc": 0.9400749063670412, "train_speed(iter/s)": 1.447664 }, { "epoch": 2.623709352641275, "grad_norm": 1.9940903186798096, "learning_rate": 4.6117468389182053e-05, "loss": 0.32521762847900393, "memory(GiB)": 70.5, "step": 61240, "token_acc": 0.9224489795918367, "train_speed(iter/s)": 1.447675 }, { "epoch": 2.623923567970524, "grad_norm": 3.3047726154327393, "learning_rate": 4.6110758970989264e-05, "loss": 0.27944657802581785, "memory(GiB)": 70.5, "step": 61245, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.447674 }, { "epoch": 2.6241377832997728, "grad_norm": 3.0350568294525146, "learning_rate": 4.610404962325374e-05, "loss": 0.3547337532043457, "memory(GiB)": 70.5, "step": 61250, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.447672 }, { "epoch": 2.624351998629022, "grad_norm": 5.566816329956055, "learning_rate": 4.6097340346097004e-05, "loss": 0.23143272399902343, "memory(GiB)": 70.5, "step": 61255, "token_acc": 0.9635258358662614, "train_speed(iter/s)": 1.447673 }, { "epoch": 2.624566213958271, "grad_norm": 2.43613600730896, "learning_rate": 4.609063113964061e-05, "loss": 0.3809990406036377, "memory(GiB)": 70.5, "step": 61260, "token_acc": 0.9186991869918699, "train_speed(iter/s)": 1.44768 }, { "epoch": 2.6247804292875196, "grad_norm": 3.6608526706695557, "learning_rate": 4.608392200400611e-05, "loss": 0.4996376037597656, "memory(GiB)": 70.5, "step": 61265, "token_acc": 0.8796296296296297, "train_speed(iter/s)": 1.447679 }, { "epoch": 2.624994644616769, "grad_norm": 8.440574645996094, "learning_rate": 4.607721293931502e-05, "loss": 0.30780644416809083, "memory(GiB)": 70.5, "step": 61270, "token_acc": 0.9288702928870293, "train_speed(iter/s)": 1.447688 }, { "epoch": 2.6252088599460177, "grad_norm": 5.852423667907715, "learning_rate": 4.607050394568891e-05, "loss": 0.23564231395721436, "memory(GiB)": 70.5, "step": 61275, "token_acc": 0.9502262443438914, "train_speed(iter/s)": 1.447696 }, { "epoch": 2.6254230752752665, "grad_norm": 7.746718406677246, "learning_rate": 4.6063795023249303e-05, "loss": 0.31722733974456785, "memory(GiB)": 70.5, "step": 61280, "token_acc": 0.9298245614035088, "train_speed(iter/s)": 1.447696 }, { "epoch": 2.625637290604516, "grad_norm": 4.058821201324463, "learning_rate": 4.605708617211774e-05, "loss": 0.19441094398498535, "memory(GiB)": 70.5, "step": 61285, "token_acc": 0.9517684887459807, "train_speed(iter/s)": 1.447698 }, { "epoch": 2.6258515059337646, "grad_norm": 4.198193550109863, "learning_rate": 4.605037739241576e-05, "loss": 0.3677348613739014, "memory(GiB)": 70.5, "step": 61290, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.447698 }, { "epoch": 2.6260657212630134, "grad_norm": 3.317779779434204, "learning_rate": 4.604366868426489e-05, "loss": 0.46806998252868653, "memory(GiB)": 70.5, "step": 61295, "token_acc": 0.8962962962962963, "train_speed(iter/s)": 1.44771 }, { "epoch": 2.6262799365922627, "grad_norm": 5.683505058288574, "learning_rate": 4.603696004778669e-05, "loss": 0.5702770233154297, "memory(GiB)": 70.5, "step": 61300, "token_acc": 0.8884892086330936, "train_speed(iter/s)": 1.447716 }, { "epoch": 2.6264941519215115, "grad_norm": 3.980203151702881, "learning_rate": 4.603025148310267e-05, "loss": 0.5325785636901855, "memory(GiB)": 70.5, "step": 61305, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.447738 }, { "epoch": 2.6267083672507603, "grad_norm": 1.6444588899612427, "learning_rate": 4.602354299033435e-05, "loss": 0.514310884475708, "memory(GiB)": 70.5, "step": 61310, "token_acc": 0.8958990536277602, "train_speed(iter/s)": 1.44775 }, { "epoch": 2.6269225825800095, "grad_norm": 3.5431947708129883, "learning_rate": 4.601683456960329e-05, "loss": 0.36957426071166993, "memory(GiB)": 70.5, "step": 61315, "token_acc": 0.9269662921348315, "train_speed(iter/s)": 1.447755 }, { "epoch": 2.6271367979092584, "grad_norm": 2.6261146068573, "learning_rate": 4.601012622103102e-05, "loss": 0.2423548936843872, "memory(GiB)": 70.5, "step": 61320, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.447761 }, { "epoch": 2.627351013238507, "grad_norm": 4.542873859405518, "learning_rate": 4.600341794473905e-05, "loss": 0.33140134811401367, "memory(GiB)": 70.5, "step": 61325, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.44777 }, { "epoch": 2.6275652285677564, "grad_norm": 2.5103797912597656, "learning_rate": 4.599670974084891e-05, "loss": 0.24002795219421386, "memory(GiB)": 70.5, "step": 61330, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.447772 }, { "epoch": 2.6277794438970052, "grad_norm": 1.6654372215270996, "learning_rate": 4.599000160948214e-05, "loss": 0.4216325283050537, "memory(GiB)": 70.5, "step": 61335, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.447777 }, { "epoch": 2.627993659226254, "grad_norm": 4.151538372039795, "learning_rate": 4.598329355076024e-05, "loss": 0.30303568840026857, "memory(GiB)": 70.5, "step": 61340, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.447783 }, { "epoch": 2.6282078745555033, "grad_norm": 3.6377053260803223, "learning_rate": 4.597658556480475e-05, "loss": 0.4145235061645508, "memory(GiB)": 70.5, "step": 61345, "token_acc": 0.916, "train_speed(iter/s)": 1.447784 }, { "epoch": 2.628422089884752, "grad_norm": 2.3340649604797363, "learning_rate": 4.5969877651737194e-05, "loss": 0.44047861099243163, "memory(GiB)": 70.5, "step": 61350, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.447789 }, { "epoch": 2.628636305214001, "grad_norm": 2.935323715209961, "learning_rate": 4.596316981167906e-05, "loss": 0.3603860855102539, "memory(GiB)": 70.5, "step": 61355, "token_acc": 0.913312693498452, "train_speed(iter/s)": 1.447792 }, { "epoch": 2.62885052054325, "grad_norm": 1.3486486673355103, "learning_rate": 4.595646204475192e-05, "loss": 0.2365210771560669, "memory(GiB)": 70.5, "step": 61360, "token_acc": 0.954954954954955, "train_speed(iter/s)": 1.447792 }, { "epoch": 2.629064735872499, "grad_norm": 2.7939846515655518, "learning_rate": 4.594975435107726e-05, "loss": 0.22652871608734132, "memory(GiB)": 70.5, "step": 61365, "token_acc": 0.9669117647058824, "train_speed(iter/s)": 1.447792 }, { "epoch": 2.629278951201748, "grad_norm": 3.061826467514038, "learning_rate": 4.594304673077658e-05, "loss": 0.267567777633667, "memory(GiB)": 70.5, "step": 61370, "token_acc": 0.9346590909090909, "train_speed(iter/s)": 1.447789 }, { "epoch": 2.629493166530997, "grad_norm": 5.2847065925598145, "learning_rate": 4.593633918397143e-05, "loss": 0.41947040557861326, "memory(GiB)": 70.5, "step": 61375, "token_acc": 0.9080118694362018, "train_speed(iter/s)": 1.447798 }, { "epoch": 2.629707381860246, "grad_norm": 2.2360448837280273, "learning_rate": 4.592963171078331e-05, "loss": 0.2866261720657349, "memory(GiB)": 70.5, "step": 61380, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.447797 }, { "epoch": 2.6299215971894947, "grad_norm": 0.3754637539386749, "learning_rate": 4.59229243113337e-05, "loss": 0.05782880187034607, "memory(GiB)": 70.5, "step": 61385, "token_acc": 0.9873949579831933, "train_speed(iter/s)": 1.44781 }, { "epoch": 2.630135812518744, "grad_norm": 1.939903974533081, "learning_rate": 4.5916216985744164e-05, "loss": 0.46329636573791505, "memory(GiB)": 70.5, "step": 61390, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.447809 }, { "epoch": 2.6303500278479928, "grad_norm": 0.9144127368927002, "learning_rate": 4.590950973413619e-05, "loss": 0.24540886878967286, "memory(GiB)": 70.5, "step": 61395, "token_acc": 0.9551569506726457, "train_speed(iter/s)": 1.447825 }, { "epoch": 2.6305642431772416, "grad_norm": 4.979918003082275, "learning_rate": 4.5902802556631275e-05, "loss": 0.437441349029541, "memory(GiB)": 70.5, "step": 61400, "token_acc": 0.9051987767584098, "train_speed(iter/s)": 1.447836 }, { "epoch": 2.630778458506491, "grad_norm": 6.291593551635742, "learning_rate": 4.589609545335095e-05, "loss": 0.6274298667907715, "memory(GiB)": 70.5, "step": 61405, "token_acc": 0.8755186721991701, "train_speed(iter/s)": 1.447837 }, { "epoch": 2.6309926738357396, "grad_norm": 5.523523807525635, "learning_rate": 4.58893884244167e-05, "loss": 0.8953953742980957, "memory(GiB)": 70.5, "step": 61410, "token_acc": 0.8064516129032258, "train_speed(iter/s)": 1.447846 }, { "epoch": 2.6312068891649885, "grad_norm": 0.5055042505264282, "learning_rate": 4.588268146995003e-05, "loss": 0.22902591228485109, "memory(GiB)": 70.5, "step": 61415, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.447853 }, { "epoch": 2.6314211044942377, "grad_norm": 6.477840900421143, "learning_rate": 4.587597459007246e-05, "loss": 0.41686525344848635, "memory(GiB)": 70.5, "step": 61420, "token_acc": 0.9028213166144201, "train_speed(iter/s)": 1.447858 }, { "epoch": 2.6316353198234865, "grad_norm": 1.3132541179656982, "learning_rate": 4.586926778490547e-05, "loss": 0.35986416339874266, "memory(GiB)": 70.5, "step": 61425, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.447862 }, { "epoch": 2.6318495351527353, "grad_norm": 6.469356060028076, "learning_rate": 4.586256105457056e-05, "loss": 0.32283830642700195, "memory(GiB)": 70.5, "step": 61430, "token_acc": 0.9457364341085271, "train_speed(iter/s)": 1.447861 }, { "epoch": 2.6320637504819846, "grad_norm": 2.3179619312286377, "learning_rate": 4.585585439918925e-05, "loss": 0.38277230262756345, "memory(GiB)": 70.5, "step": 61435, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.447868 }, { "epoch": 2.6322779658112334, "grad_norm": 4.5965776443481445, "learning_rate": 4.5849147818883017e-05, "loss": 0.4863728046417236, "memory(GiB)": 70.5, "step": 61440, "token_acc": 0.9076433121019108, "train_speed(iter/s)": 1.447873 }, { "epoch": 2.6324921811404822, "grad_norm": 4.076902866363525, "learning_rate": 4.584244131377338e-05, "loss": 0.5520884990692139, "memory(GiB)": 70.5, "step": 61445, "token_acc": 0.8786885245901639, "train_speed(iter/s)": 1.447882 }, { "epoch": 2.6327063964697315, "grad_norm": 2.34574556350708, "learning_rate": 4.5835734883981804e-05, "loss": 0.3781766891479492, "memory(GiB)": 70.5, "step": 61450, "token_acc": 0.9184397163120568, "train_speed(iter/s)": 1.447889 }, { "epoch": 2.6329206117989803, "grad_norm": 3.5890491008758545, "learning_rate": 4.5829028529629794e-05, "loss": 0.3574995994567871, "memory(GiB)": 70.5, "step": 61455, "token_acc": 0.916083916083916, "train_speed(iter/s)": 1.44789 }, { "epoch": 2.633134827128229, "grad_norm": 1.8861055374145508, "learning_rate": 4.5822322250838836e-05, "loss": 0.48732542991638184, "memory(GiB)": 70.5, "step": 61460, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.447883 }, { "epoch": 2.6333490424574784, "grad_norm": 2.201122522354126, "learning_rate": 4.5815616047730455e-05, "loss": 0.18884475231170655, "memory(GiB)": 70.5, "step": 61465, "token_acc": 0.9607843137254902, "train_speed(iter/s)": 1.447883 }, { "epoch": 2.633563257786727, "grad_norm": 7.763796806335449, "learning_rate": 4.580890992042611e-05, "loss": 0.5929625034332275, "memory(GiB)": 70.5, "step": 61470, "token_acc": 0.8874172185430463, "train_speed(iter/s)": 1.44788 }, { "epoch": 2.633777473115976, "grad_norm": 3.2763404846191406, "learning_rate": 4.580220386904728e-05, "loss": 0.18027902841567994, "memory(GiB)": 70.5, "step": 61475, "token_acc": 0.9598214285714286, "train_speed(iter/s)": 1.447882 }, { "epoch": 2.6339916884452252, "grad_norm": 3.47609806060791, "learning_rate": 4.579549789371548e-05, "loss": 0.5458856582641601, "memory(GiB)": 70.5, "step": 61480, "token_acc": 0.8744939271255061, "train_speed(iter/s)": 1.447899 }, { "epoch": 2.634205903774474, "grad_norm": 0.16181060671806335, "learning_rate": 4.578879199455216e-05, "loss": 0.25810463428497316, "memory(GiB)": 70.5, "step": 61485, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.447899 }, { "epoch": 2.634420119103723, "grad_norm": 6.021367073059082, "learning_rate": 4.5782086171678845e-05, "loss": 0.43481292724609377, "memory(GiB)": 70.5, "step": 61490, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.447902 }, { "epoch": 2.634634334432972, "grad_norm": 3.4224863052368164, "learning_rate": 4.577538042521699e-05, "loss": 0.3320859432220459, "memory(GiB)": 70.5, "step": 61495, "token_acc": 0.9085173501577287, "train_speed(iter/s)": 1.447898 }, { "epoch": 2.634848549762221, "grad_norm": 3.517878770828247, "learning_rate": 4.5768674755288074e-05, "loss": 0.522334098815918, "memory(GiB)": 70.5, "step": 61500, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.447914 }, { "epoch": 2.634848549762221, "eval_loss": 2.412055730819702, "eval_runtime": 12.8555, "eval_samples_per_second": 7.779, "eval_steps_per_second": 7.779, "eval_token_acc": 0.44429347826086957, "step": 61500 }, { "epoch": 2.6350627650914697, "grad_norm": 2.461435079574585, "learning_rate": 4.576196916201359e-05, "loss": 0.22640254497528076, "memory(GiB)": 70.5, "step": 61505, "token_acc": 0.5833333333333334, "train_speed(iter/s)": 1.447453 }, { "epoch": 2.635276980420719, "grad_norm": 0.3752152621746063, "learning_rate": 4.575526364551501e-05, "loss": 0.20357902050018312, "memory(GiB)": 70.5, "step": 61510, "token_acc": 0.9546742209631728, "train_speed(iter/s)": 1.447447 }, { "epoch": 2.635491195749968, "grad_norm": 3.8056607246398926, "learning_rate": 4.5748558205913814e-05, "loss": 0.3291881561279297, "memory(GiB)": 70.5, "step": 61515, "token_acc": 0.9294478527607362, "train_speed(iter/s)": 1.447452 }, { "epoch": 2.6357054110792166, "grad_norm": 3.2258739471435547, "learning_rate": 4.5741852843331466e-05, "loss": 0.34124050140380857, "memory(GiB)": 70.5, "step": 61520, "token_acc": 0.9098360655737705, "train_speed(iter/s)": 1.447461 }, { "epoch": 2.635919626408466, "grad_norm": 4.251286506652832, "learning_rate": 4.5735147557889445e-05, "loss": 0.289488697052002, "memory(GiB)": 70.5, "step": 61525, "token_acc": 0.9260450160771704, "train_speed(iter/s)": 1.447465 }, { "epoch": 2.6361338417377147, "grad_norm": 3.049596071243286, "learning_rate": 4.5728442349709244e-05, "loss": 0.3340094804763794, "memory(GiB)": 70.5, "step": 61530, "token_acc": 0.9388379204892966, "train_speed(iter/s)": 1.447469 }, { "epoch": 2.6363480570669635, "grad_norm": 1.6077907085418701, "learning_rate": 4.5721737218912285e-05, "loss": 0.49300594329833985, "memory(GiB)": 70.5, "step": 61535, "token_acc": 0.8993055555555556, "train_speed(iter/s)": 1.447468 }, { "epoch": 2.6365622723962128, "grad_norm": 2.965414524078369, "learning_rate": 4.57150321656201e-05, "loss": 0.6063614845275879, "memory(GiB)": 70.5, "step": 61540, "token_acc": 0.8653846153846154, "train_speed(iter/s)": 1.447465 }, { "epoch": 2.6367764877254616, "grad_norm": 2.651120185852051, "learning_rate": 4.5708327189954116e-05, "loss": 0.46217942237854004, "memory(GiB)": 70.5, "step": 61545, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.447467 }, { "epoch": 2.6369907030547104, "grad_norm": 2.8855948448181152, "learning_rate": 4.5701622292035815e-05, "loss": 0.34079155921936033, "memory(GiB)": 70.5, "step": 61550, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.447465 }, { "epoch": 2.6372049183839597, "grad_norm": 2.38625431060791, "learning_rate": 4.569491747198666e-05, "loss": 0.2869521141052246, "memory(GiB)": 70.5, "step": 61555, "token_acc": 0.9273504273504274, "train_speed(iter/s)": 1.447463 }, { "epoch": 2.6374191337132085, "grad_norm": 4.381589889526367, "learning_rate": 4.56882127299281e-05, "loss": 0.31996893882751465, "memory(GiB)": 70.5, "step": 61560, "token_acc": 0.9202453987730062, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.6376333490424573, "grad_norm": 4.983942031860352, "learning_rate": 4.568150806598163e-05, "loss": 0.5336143493652343, "memory(GiB)": 70.5, "step": 61565, "token_acc": 0.8808664259927798, "train_speed(iter/s)": 1.447468 }, { "epoch": 2.6378475643717065, "grad_norm": 1.943608045578003, "learning_rate": 4.567480348026869e-05, "loss": 0.30729892253875735, "memory(GiB)": 70.5, "step": 61570, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.447469 }, { "epoch": 2.6380617797009553, "grad_norm": 1.6030519008636475, "learning_rate": 4.566809897291073e-05, "loss": 0.3169023275375366, "memory(GiB)": 70.5, "step": 61575, "token_acc": 0.9356913183279743, "train_speed(iter/s)": 1.447468 }, { "epoch": 2.638275995030204, "grad_norm": 7.661046028137207, "learning_rate": 4.566139454402923e-05, "loss": 0.5195615768432618, "memory(GiB)": 70.5, "step": 61580, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.447466 }, { "epoch": 2.6384902103594534, "grad_norm": 4.996121406555176, "learning_rate": 4.565469019374563e-05, "loss": 0.5801838874816895, "memory(GiB)": 70.5, "step": 61585, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.447479 }, { "epoch": 2.6387044256887022, "grad_norm": 3.533048152923584, "learning_rate": 4.5647985922181405e-05, "loss": 0.5621556282043457, "memory(GiB)": 70.5, "step": 61590, "token_acc": 0.8932038834951457, "train_speed(iter/s)": 1.447479 }, { "epoch": 2.638918641017951, "grad_norm": 2.9299230575561523, "learning_rate": 4.564128172945799e-05, "loss": 0.37748355865478517, "memory(GiB)": 70.5, "step": 61595, "token_acc": 0.9357429718875502, "train_speed(iter/s)": 1.447474 }, { "epoch": 2.6391328563472003, "grad_norm": 4.254628658294678, "learning_rate": 4.563457761569685e-05, "loss": 0.5746063709259033, "memory(GiB)": 70.5, "step": 61600, "token_acc": 0.8842443729903537, "train_speed(iter/s)": 1.447477 }, { "epoch": 2.639347071676449, "grad_norm": 4.1840949058532715, "learning_rate": 4.56278735810194e-05, "loss": 0.3969834327697754, "memory(GiB)": 70.5, "step": 61605, "token_acc": 0.9098039215686274, "train_speed(iter/s)": 1.447477 }, { "epoch": 2.639561287005698, "grad_norm": 2.6185483932495117, "learning_rate": 4.562116962554716e-05, "loss": 0.3301081657409668, "memory(GiB)": 70.5, "step": 61610, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.447477 }, { "epoch": 2.639775502334947, "grad_norm": 3.593273162841797, "learning_rate": 4.561446574940153e-05, "loss": 0.26254639625549314, "memory(GiB)": 70.5, "step": 61615, "token_acc": 0.9498432601880877, "train_speed(iter/s)": 1.447491 }, { "epoch": 2.639989717664196, "grad_norm": 3.780306816101074, "learning_rate": 4.560776195270396e-05, "loss": 0.6180973052978516, "memory(GiB)": 70.5, "step": 61620, "token_acc": 0.8904593639575972, "train_speed(iter/s)": 1.447508 }, { "epoch": 2.640203932993445, "grad_norm": 2.176625967025757, "learning_rate": 4.5601058235575914e-05, "loss": 0.20591967105865477, "memory(GiB)": 70.5, "step": 61625, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.447509 }, { "epoch": 2.640418148322694, "grad_norm": 0.5674892067909241, "learning_rate": 4.559435459813881e-05, "loss": 0.2882565498352051, "memory(GiB)": 70.5, "step": 61630, "token_acc": 0.9547325102880658, "train_speed(iter/s)": 1.447513 }, { "epoch": 2.640632363651943, "grad_norm": 12.422568321228027, "learning_rate": 4.558765104051412e-05, "loss": 0.4681526184082031, "memory(GiB)": 70.5, "step": 61635, "token_acc": 0.9212328767123288, "train_speed(iter/s)": 1.447532 }, { "epoch": 2.6408465789811917, "grad_norm": 2.130526065826416, "learning_rate": 4.558094756282327e-05, "loss": 0.5061566829681396, "memory(GiB)": 70.5, "step": 61640, "token_acc": 0.909375, "train_speed(iter/s)": 1.44753 }, { "epoch": 2.641060794310441, "grad_norm": 1.3075767755508423, "learning_rate": 4.557424416518768e-05, "loss": 0.4304489612579346, "memory(GiB)": 70.5, "step": 61645, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.44753 }, { "epoch": 2.6412750096396898, "grad_norm": 0.9399771094322205, "learning_rate": 4.556754084772883e-05, "loss": 0.18796178102493286, "memory(GiB)": 70.5, "step": 61650, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.447521 }, { "epoch": 2.6414892249689386, "grad_norm": 3.8090062141418457, "learning_rate": 4.556083761056814e-05, "loss": 0.19445762634277344, "memory(GiB)": 70.5, "step": 61655, "token_acc": 0.9478827361563518, "train_speed(iter/s)": 1.447517 }, { "epoch": 2.641703440298188, "grad_norm": 2.6742045879364014, "learning_rate": 4.5554134453827024e-05, "loss": 0.37818431854248047, "memory(GiB)": 70.5, "step": 61660, "token_acc": 0.9139072847682119, "train_speed(iter/s)": 1.447525 }, { "epoch": 2.6419176556274366, "grad_norm": 2.1844875812530518, "learning_rate": 4.554743137762694e-05, "loss": 0.4525002479553223, "memory(GiB)": 70.5, "step": 61665, "token_acc": 0.9057971014492754, "train_speed(iter/s)": 1.447533 }, { "epoch": 2.6421318709566854, "grad_norm": 6.542811870574951, "learning_rate": 4.554072838208931e-05, "loss": 0.3769400119781494, "memory(GiB)": 70.5, "step": 61670, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.447535 }, { "epoch": 2.6423460862859347, "grad_norm": 5.116458415985107, "learning_rate": 4.553402546733557e-05, "loss": 0.5813519477844238, "memory(GiB)": 70.5, "step": 61675, "token_acc": 0.8703071672354948, "train_speed(iter/s)": 1.447534 }, { "epoch": 2.6425603016151835, "grad_norm": 2.21981143951416, "learning_rate": 4.552732263348713e-05, "loss": 0.43384761810302735, "memory(GiB)": 70.5, "step": 61680, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.44753 }, { "epoch": 2.6427745169444323, "grad_norm": 1.4639217853546143, "learning_rate": 4.5520619880665466e-05, "loss": 0.24800541400909423, "memory(GiB)": 70.5, "step": 61685, "token_acc": 0.9625850340136054, "train_speed(iter/s)": 1.447525 }, { "epoch": 2.6429887322736816, "grad_norm": 3.018535614013672, "learning_rate": 4.551391720899196e-05, "loss": 0.37110195159912107, "memory(GiB)": 70.5, "step": 61690, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.447533 }, { "epoch": 2.6432029476029304, "grad_norm": 2.7520089149475098, "learning_rate": 4.5507214618588055e-05, "loss": 0.3076845645904541, "memory(GiB)": 70.5, "step": 61695, "token_acc": 0.9431818181818182, "train_speed(iter/s)": 1.447547 }, { "epoch": 2.643417162932179, "grad_norm": 5.0172271728515625, "learning_rate": 4.5500512109575186e-05, "loss": 0.4185813903808594, "memory(GiB)": 70.5, "step": 61700, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.447555 }, { "epoch": 2.6436313782614285, "grad_norm": 4.614552021026611, "learning_rate": 4.549380968207474e-05, "loss": 0.4791110515594482, "memory(GiB)": 70.5, "step": 61705, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.447556 }, { "epoch": 2.6438455935906773, "grad_norm": 2.823154926300049, "learning_rate": 4.5487107336208166e-05, "loss": 0.200669264793396, "memory(GiB)": 70.5, "step": 61710, "token_acc": 0.951417004048583, "train_speed(iter/s)": 1.447567 }, { "epoch": 2.644059808919926, "grad_norm": 3.1636853218078613, "learning_rate": 4.548040507209689e-05, "loss": 0.36113929748535156, "memory(GiB)": 70.5, "step": 61715, "token_acc": 0.9292307692307692, "train_speed(iter/s)": 1.447573 }, { "epoch": 2.6442740242491753, "grad_norm": 3.198113203048706, "learning_rate": 4.547370288986229e-05, "loss": 0.2868433237075806, "memory(GiB)": 70.5, "step": 61720, "token_acc": 0.9343283582089552, "train_speed(iter/s)": 1.447572 }, { "epoch": 2.644488239578424, "grad_norm": 1.5984017848968506, "learning_rate": 4.546700078962582e-05, "loss": 0.23110063076019288, "memory(GiB)": 70.5, "step": 61725, "token_acc": 0.9483282674772037, "train_speed(iter/s)": 1.447576 }, { "epoch": 2.644702454907673, "grad_norm": 3.524688482284546, "learning_rate": 4.5460298771508885e-05, "loss": 0.3024256944656372, "memory(GiB)": 70.5, "step": 61730, "token_acc": 0.9486166007905138, "train_speed(iter/s)": 1.447582 }, { "epoch": 2.6449166702369222, "grad_norm": 2.2486536502838135, "learning_rate": 4.54535968356329e-05, "loss": 0.2127291202545166, "memory(GiB)": 70.5, "step": 61735, "token_acc": 0.9503311258278145, "train_speed(iter/s)": 1.447586 }, { "epoch": 2.645130885566171, "grad_norm": 0.7443834543228149, "learning_rate": 4.544689498211927e-05, "loss": 0.5230721473693848, "memory(GiB)": 70.5, "step": 61740, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.447601 }, { "epoch": 2.64534510089542, "grad_norm": 6.602192401885986, "learning_rate": 4.54401932110894e-05, "loss": 0.3551377773284912, "memory(GiB)": 70.5, "step": 61745, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.447601 }, { "epoch": 2.645559316224669, "grad_norm": 2.611539125442505, "learning_rate": 4.543349152266472e-05, "loss": 0.3158955335617065, "memory(GiB)": 70.5, "step": 61750, "token_acc": 0.9240506329113924, "train_speed(iter/s)": 1.4476 }, { "epoch": 2.645773531553918, "grad_norm": 4.033263683319092, "learning_rate": 4.54267899169666e-05, "loss": 0.36622159481048583, "memory(GiB)": 70.5, "step": 61755, "token_acc": 0.9123505976095617, "train_speed(iter/s)": 1.447598 }, { "epoch": 2.6459877468831667, "grad_norm": 2.268050193786621, "learning_rate": 4.54200883941165e-05, "loss": 0.5759293079376221, "memory(GiB)": 70.5, "step": 61760, "token_acc": 0.8551724137931035, "train_speed(iter/s)": 1.447599 }, { "epoch": 2.646201962212416, "grad_norm": 3.341790199279785, "learning_rate": 4.541338695423578e-05, "loss": 0.5484394073486328, "memory(GiB)": 70.5, "step": 61765, "token_acc": 0.8850174216027874, "train_speed(iter/s)": 1.447598 }, { "epoch": 2.646416177541665, "grad_norm": 6.018964767456055, "learning_rate": 4.5406685597445866e-05, "loss": 0.3176295042037964, "memory(GiB)": 70.5, "step": 61770, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.447595 }, { "epoch": 2.6466303928709136, "grad_norm": 6.782587051391602, "learning_rate": 4.539998432386814e-05, "loss": 0.5410772323608398, "memory(GiB)": 70.5, "step": 61775, "token_acc": 0.8711864406779661, "train_speed(iter/s)": 1.447591 }, { "epoch": 2.646844608200163, "grad_norm": 5.127666473388672, "learning_rate": 4.5393283133624026e-05, "loss": 0.19411921501159668, "memory(GiB)": 70.5, "step": 61780, "token_acc": 0.94921875, "train_speed(iter/s)": 1.447596 }, { "epoch": 2.6470588235294117, "grad_norm": 3.432521343231201, "learning_rate": 4.5386582026834906e-05, "loss": 0.44103713035583497, "memory(GiB)": 70.5, "step": 61785, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.447593 }, { "epoch": 2.6472730388586605, "grad_norm": 0.8241600394248962, "learning_rate": 4.5379881003622174e-05, "loss": 0.3440232753753662, "memory(GiB)": 70.5, "step": 61790, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.447596 }, { "epoch": 2.6474872541879098, "grad_norm": 4.59395170211792, "learning_rate": 4.537318006410724e-05, "loss": 0.23594017028808595, "memory(GiB)": 70.5, "step": 61795, "token_acc": 0.9507042253521126, "train_speed(iter/s)": 1.447588 }, { "epoch": 2.6477014695171586, "grad_norm": 5.28005838394165, "learning_rate": 4.53664792084115e-05, "loss": 0.6669252395629883, "memory(GiB)": 70.5, "step": 61800, "token_acc": 0.8548387096774194, "train_speed(iter/s)": 1.447584 }, { "epoch": 2.6479156848464074, "grad_norm": 5.439007759094238, "learning_rate": 4.535977843665631e-05, "loss": 0.5724901676177978, "memory(GiB)": 70.5, "step": 61805, "token_acc": 0.8789625360230547, "train_speed(iter/s)": 1.447581 }, { "epoch": 2.6481299001756566, "grad_norm": 5.23031759262085, "learning_rate": 4.53530777489631e-05, "loss": 0.5377740859985352, "memory(GiB)": 70.5, "step": 61810, "token_acc": 0.8883928571428571, "train_speed(iter/s)": 1.447585 }, { "epoch": 2.6483441155049054, "grad_norm": 5.360401153564453, "learning_rate": 4.534637714545324e-05, "loss": 0.3579395294189453, "memory(GiB)": 70.5, "step": 61815, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.447577 }, { "epoch": 2.6485583308341543, "grad_norm": 0.7323693037033081, "learning_rate": 4.533967662624813e-05, "loss": 0.31434359550476076, "memory(GiB)": 70.5, "step": 61820, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.447583 }, { "epoch": 2.6487725461634035, "grad_norm": 2.0621631145477295, "learning_rate": 4.533297619146915e-05, "loss": 0.4119595527648926, "memory(GiB)": 70.5, "step": 61825, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.447586 }, { "epoch": 2.6489867614926523, "grad_norm": 12.524123191833496, "learning_rate": 4.532627584123766e-05, "loss": 0.6590642929077148, "memory(GiB)": 70.5, "step": 61830, "token_acc": 0.8669064748201439, "train_speed(iter/s)": 1.447614 }, { "epoch": 2.649200976821901, "grad_norm": 3.0450568199157715, "learning_rate": 4.531957557567509e-05, "loss": 0.22728297710418702, "memory(GiB)": 70.5, "step": 61835, "token_acc": 0.9488817891373802, "train_speed(iter/s)": 1.447619 }, { "epoch": 2.6494151921511504, "grad_norm": 1.6804946660995483, "learning_rate": 4.5312875394902795e-05, "loss": 0.36830203533172606, "memory(GiB)": 70.5, "step": 61840, "token_acc": 0.9276729559748428, "train_speed(iter/s)": 1.447622 }, { "epoch": 2.649629407480399, "grad_norm": 4.561955451965332, "learning_rate": 4.530617529904218e-05, "loss": 0.4207762718200684, "memory(GiB)": 70.5, "step": 61845, "token_acc": 0.8958904109589041, "train_speed(iter/s)": 1.447622 }, { "epoch": 2.649843622809648, "grad_norm": 2.235570192337036, "learning_rate": 4.529947528821457e-05, "loss": 0.4551135540008545, "memory(GiB)": 70.5, "step": 61850, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.44763 }, { "epoch": 2.6500578381388973, "grad_norm": 2.9894652366638184, "learning_rate": 4.52927753625414e-05, "loss": 0.3946289777755737, "memory(GiB)": 70.5, "step": 61855, "token_acc": 0.9249329758713136, "train_speed(iter/s)": 1.447629 }, { "epoch": 2.650272053468146, "grad_norm": 2.86881160736084, "learning_rate": 4.528607552214401e-05, "loss": 0.498002290725708, "memory(GiB)": 70.5, "step": 61860, "token_acc": 0.8906752411575563, "train_speed(iter/s)": 1.447628 }, { "epoch": 2.650486268797395, "grad_norm": 3.005465269088745, "learning_rate": 4.527937576714378e-05, "loss": 0.5542258739471435, "memory(GiB)": 70.5, "step": 61865, "token_acc": 0.8921568627450981, "train_speed(iter/s)": 1.447618 }, { "epoch": 2.650700484126644, "grad_norm": 3.4489707946777344, "learning_rate": 4.527267609766209e-05, "loss": 0.5069169998168945, "memory(GiB)": 70.5, "step": 61870, "token_acc": 0.8804347826086957, "train_speed(iter/s)": 1.447614 }, { "epoch": 2.650914699455893, "grad_norm": 3.322847843170166, "learning_rate": 4.52659765138203e-05, "loss": 0.47964982986450194, "memory(GiB)": 70.5, "step": 61875, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.447614 }, { "epoch": 2.651128914785142, "grad_norm": 3.1377198696136475, "learning_rate": 4.5259277015739796e-05, "loss": 0.37734332084655764, "memory(GiB)": 70.5, "step": 61880, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.447608 }, { "epoch": 2.651343130114391, "grad_norm": 1.858526587486267, "learning_rate": 4.525257760354194e-05, "loss": 0.34901251792907717, "memory(GiB)": 70.5, "step": 61885, "token_acc": 0.9116465863453815, "train_speed(iter/s)": 1.447604 }, { "epoch": 2.65155734544364, "grad_norm": 2.7254104614257812, "learning_rate": 4.5245878277348084e-05, "loss": 0.24496827125549317, "memory(GiB)": 70.5, "step": 61890, "token_acc": 0.9380165289256198, "train_speed(iter/s)": 1.447603 }, { "epoch": 2.6517715607728887, "grad_norm": 2.1868069171905518, "learning_rate": 4.523917903727961e-05, "loss": 0.4473905563354492, "memory(GiB)": 70.5, "step": 61895, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.447611 }, { "epoch": 2.651985776102138, "grad_norm": 0.798876941204071, "learning_rate": 4.5232479883457865e-05, "loss": 0.45054874420166013, "memory(GiB)": 70.5, "step": 61900, "token_acc": 0.8996960486322189, "train_speed(iter/s)": 1.447612 }, { "epoch": 2.6521999914313867, "grad_norm": 2.8011605739593506, "learning_rate": 4.522578081600421e-05, "loss": 0.1952047109603882, "memory(GiB)": 70.5, "step": 61905, "token_acc": 0.9553903345724907, "train_speed(iter/s)": 1.447618 }, { "epoch": 2.6524142067606356, "grad_norm": 3.1155450344085693, "learning_rate": 4.521908183504002e-05, "loss": 0.6918132305145264, "memory(GiB)": 70.5, "step": 61910, "token_acc": 0.8283582089552238, "train_speed(iter/s)": 1.447622 }, { "epoch": 2.652628422089885, "grad_norm": 4.067039966583252, "learning_rate": 4.521238294068667e-05, "loss": 0.5355288982391357, "memory(GiB)": 70.5, "step": 61915, "token_acc": 0.8844884488448845, "train_speed(iter/s)": 1.447621 }, { "epoch": 2.6528426374191336, "grad_norm": 3.7300891876220703, "learning_rate": 4.520568413306547e-05, "loss": 0.4307830333709717, "memory(GiB)": 70.5, "step": 61920, "token_acc": 0.9140127388535032, "train_speed(iter/s)": 1.44762 }, { "epoch": 2.6530568527483824, "grad_norm": 4.879171848297119, "learning_rate": 4.519898541229781e-05, "loss": 0.45722060203552245, "memory(GiB)": 70.5, "step": 61925, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.447621 }, { "epoch": 2.6532710680776317, "grad_norm": 2.188586950302124, "learning_rate": 4.519228677850504e-05, "loss": 0.35382134914398194, "memory(GiB)": 70.5, "step": 61930, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.447623 }, { "epoch": 2.6534852834068805, "grad_norm": 4.501918792724609, "learning_rate": 4.5185588231808486e-05, "loss": 0.3405893564224243, "memory(GiB)": 70.5, "step": 61935, "token_acc": 0.9046153846153846, "train_speed(iter/s)": 1.447639 }, { "epoch": 2.6536994987361293, "grad_norm": 2.791771411895752, "learning_rate": 4.517888977232953e-05, "loss": 0.5824811935424805, "memory(GiB)": 70.5, "step": 61940, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.447637 }, { "epoch": 2.6539137140653786, "grad_norm": 1.8375006914138794, "learning_rate": 4.5172191400189515e-05, "loss": 0.2747244119644165, "memory(GiB)": 70.5, "step": 61945, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.447641 }, { "epoch": 2.6541279293946274, "grad_norm": 3.816554069519043, "learning_rate": 4.516549311550977e-05, "loss": 0.47806944847106936, "memory(GiB)": 70.5, "step": 61950, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.447653 }, { "epoch": 2.654342144723876, "grad_norm": 4.5717267990112305, "learning_rate": 4.515879491841166e-05, "loss": 0.3377795696258545, "memory(GiB)": 70.5, "step": 61955, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.447668 }, { "epoch": 2.6545563600531255, "grad_norm": 1.6245028972625732, "learning_rate": 4.515209680901651e-05, "loss": 0.47426528930664064, "memory(GiB)": 70.5, "step": 61960, "token_acc": 0.9, "train_speed(iter/s)": 1.447682 }, { "epoch": 2.6547705753823743, "grad_norm": 0.8550208210945129, "learning_rate": 4.514539878744568e-05, "loss": 0.19966109991073608, "memory(GiB)": 70.5, "step": 61965, "token_acc": 0.96415770609319, "train_speed(iter/s)": 1.447694 }, { "epoch": 2.654984790711623, "grad_norm": 4.214755058288574, "learning_rate": 4.5138700853820516e-05, "loss": 0.5378626823425293, "memory(GiB)": 70.5, "step": 61970, "token_acc": 0.9017341040462428, "train_speed(iter/s)": 1.447696 }, { "epoch": 2.6551990060408723, "grad_norm": 1.872337818145752, "learning_rate": 4.513200300826232e-05, "loss": 0.14181779623031615, "memory(GiB)": 70.5, "step": 61975, "token_acc": 0.9602446483180428, "train_speed(iter/s)": 1.447699 }, { "epoch": 2.655413221370121, "grad_norm": 7.91402530670166, "learning_rate": 4.512530525089246e-05, "loss": 0.22990975379943848, "memory(GiB)": 70.5, "step": 61980, "token_acc": 0.9375, "train_speed(iter/s)": 1.447712 }, { "epoch": 2.65562743669937, "grad_norm": 0.9903507232666016, "learning_rate": 4.511860758183229e-05, "loss": 0.24826087951660156, "memory(GiB)": 70.5, "step": 61985, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.44772 }, { "epoch": 2.655841652028619, "grad_norm": 4.321025848388672, "learning_rate": 4.511191000120312e-05, "loss": 0.3794630289077759, "memory(GiB)": 70.5, "step": 61990, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.447726 }, { "epoch": 2.656055867357868, "grad_norm": 1.890407919883728, "learning_rate": 4.510521250912627e-05, "loss": 0.181824791431427, "memory(GiB)": 70.5, "step": 61995, "token_acc": 0.9537953795379538, "train_speed(iter/s)": 1.447726 }, { "epoch": 2.656270082687117, "grad_norm": 2.8949079513549805, "learning_rate": 4.50985151057231e-05, "loss": 0.3622307777404785, "memory(GiB)": 70.5, "step": 62000, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.447721 }, { "epoch": 2.656270082687117, "eval_loss": 2.4479222297668457, "eval_runtime": 13.0015, "eval_samples_per_second": 7.691, "eval_steps_per_second": 7.691, "eval_token_acc": 0.45161290322580644, "step": 62000 }, { "epoch": 2.656484298016366, "grad_norm": 6.188794136047363, "learning_rate": 4.509181779111493e-05, "loss": 0.4362116813659668, "memory(GiB)": 70.5, "step": 62005, "token_acc": 0.5863770977295163, "train_speed(iter/s)": 1.447263 }, { "epoch": 2.656698513345615, "grad_norm": 2.930385112762451, "learning_rate": 4.508512056542307e-05, "loss": 0.33515353202819825, "memory(GiB)": 70.5, "step": 62010, "token_acc": 0.944, "train_speed(iter/s)": 1.447278 }, { "epoch": 2.6569127286748637, "grad_norm": 6.253642559051514, "learning_rate": 4.507842342876887e-05, "loss": 0.41450185775756837, "memory(GiB)": 70.5, "step": 62015, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.447285 }, { "epoch": 2.657126944004113, "grad_norm": 3.6871767044067383, "learning_rate": 4.507172638127364e-05, "loss": 0.37525343894958496, "memory(GiB)": 70.5, "step": 62020, "token_acc": 0.9063670411985019, "train_speed(iter/s)": 1.447288 }, { "epoch": 2.657341159333362, "grad_norm": 3.435983180999756, "learning_rate": 4.5065029423058726e-05, "loss": 0.22835855484008788, "memory(GiB)": 70.5, "step": 62025, "token_acc": 0.9480968858131488, "train_speed(iter/s)": 1.44729 }, { "epoch": 2.6575553746626106, "grad_norm": 7.841897010803223, "learning_rate": 4.505833255424543e-05, "loss": 0.34644601345062254, "memory(GiB)": 70.5, "step": 62030, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.447296 }, { "epoch": 2.65776958999186, "grad_norm": 3.4957785606384277, "learning_rate": 4.505163577495506e-05, "loss": 0.2834619998931885, "memory(GiB)": 70.5, "step": 62035, "token_acc": 0.9429429429429429, "train_speed(iter/s)": 1.447301 }, { "epoch": 2.6579838053211087, "grad_norm": 2.7655091285705566, "learning_rate": 4.504493908530896e-05, "loss": 0.2255397081375122, "memory(GiB)": 70.5, "step": 62040, "token_acc": 0.9457364341085271, "train_speed(iter/s)": 1.447312 }, { "epoch": 2.6581980206503575, "grad_norm": 2.2170956134796143, "learning_rate": 4.5038242485428436e-05, "loss": 0.12379798889160157, "memory(GiB)": 70.5, "step": 62045, "token_acc": 0.9862385321100917, "train_speed(iter/s)": 1.447315 }, { "epoch": 2.6584122359796067, "grad_norm": 3.158536672592163, "learning_rate": 4.503154597543479e-05, "loss": 0.4423708915710449, "memory(GiB)": 70.5, "step": 62050, "token_acc": 0.9212827988338192, "train_speed(iter/s)": 1.447319 }, { "epoch": 2.6586264513088556, "grad_norm": 3.0011682510375977, "learning_rate": 4.5024849555449353e-05, "loss": 0.45866665840148924, "memory(GiB)": 70.5, "step": 62055, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.447325 }, { "epoch": 2.6588406666381044, "grad_norm": 4.572297096252441, "learning_rate": 4.501815322559345e-05, "loss": 0.5289100646972656, "memory(GiB)": 70.5, "step": 62060, "token_acc": 0.875, "train_speed(iter/s)": 1.447338 }, { "epoch": 2.6590548819673536, "grad_norm": 7.501572608947754, "learning_rate": 4.501145698598836e-05, "loss": 0.458814001083374, "memory(GiB)": 70.5, "step": 62065, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.447368 }, { "epoch": 2.6592690972966024, "grad_norm": 4.163656711578369, "learning_rate": 4.500476083675542e-05, "loss": 0.6144587993621826, "memory(GiB)": 70.5, "step": 62070, "token_acc": 0.879245283018868, "train_speed(iter/s)": 1.447381 }, { "epoch": 2.6594833126258512, "grad_norm": 1.4386930465698242, "learning_rate": 4.499806477801592e-05, "loss": 0.3060781240463257, "memory(GiB)": 70.5, "step": 62075, "token_acc": 0.9409448818897638, "train_speed(iter/s)": 1.447384 }, { "epoch": 2.6596975279551005, "grad_norm": 2.068629264831543, "learning_rate": 4.499136880989116e-05, "loss": 0.20511746406555176, "memory(GiB)": 70.5, "step": 62080, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.447383 }, { "epoch": 2.6599117432843493, "grad_norm": 2.6252760887145996, "learning_rate": 4.498467293250246e-05, "loss": 0.23621025085449218, "memory(GiB)": 70.5, "step": 62085, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.44739 }, { "epoch": 2.660125958613598, "grad_norm": 4.683954238891602, "learning_rate": 4.497797714597112e-05, "loss": 0.67561354637146, "memory(GiB)": 70.5, "step": 62090, "token_acc": 0.8795180722891566, "train_speed(iter/s)": 1.447399 }, { "epoch": 2.6603401739428474, "grad_norm": 3.367122173309326, "learning_rate": 4.4971281450418425e-05, "loss": 0.22562475204467775, "memory(GiB)": 70.5, "step": 62095, "token_acc": 0.9393063583815029, "train_speed(iter/s)": 1.447402 }, { "epoch": 2.660554389272096, "grad_norm": 6.0975494384765625, "learning_rate": 4.496458584596569e-05, "loss": 0.37267913818359377, "memory(GiB)": 70.5, "step": 62100, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.447406 }, { "epoch": 2.660768604601345, "grad_norm": 1.219221591949463, "learning_rate": 4.495789033273419e-05, "loss": 0.46807408332824707, "memory(GiB)": 70.5, "step": 62105, "token_acc": 0.92, "train_speed(iter/s)": 1.44741 }, { "epoch": 2.6609828199305943, "grad_norm": 2.703113555908203, "learning_rate": 4.495119491084526e-05, "loss": 0.17241687774658204, "memory(GiB)": 70.5, "step": 62110, "token_acc": 0.9485530546623794, "train_speed(iter/s)": 1.447418 }, { "epoch": 2.661197035259843, "grad_norm": 2.2936196327209473, "learning_rate": 4.4944499580420166e-05, "loss": 0.271669864654541, "memory(GiB)": 70.5, "step": 62115, "token_acc": 0.9406528189910979, "train_speed(iter/s)": 1.447425 }, { "epoch": 2.661411250589092, "grad_norm": 4.713520526885986, "learning_rate": 4.4937804341580184e-05, "loss": 0.624672269821167, "memory(GiB)": 70.5, "step": 62120, "token_acc": 0.8866666666666667, "train_speed(iter/s)": 1.447435 }, { "epoch": 2.661625465918341, "grad_norm": 5.517458915710449, "learning_rate": 4.4931109194446624e-05, "loss": 0.2576157093048096, "memory(GiB)": 70.5, "step": 62125, "token_acc": 0.9490445859872612, "train_speed(iter/s)": 1.447442 }, { "epoch": 2.66183968124759, "grad_norm": 3.057835817337036, "learning_rate": 4.49244141391408e-05, "loss": 0.3207888126373291, "memory(GiB)": 70.5, "step": 62130, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.447441 }, { "epoch": 2.6620538965768388, "grad_norm": 8.213821411132812, "learning_rate": 4.4917719175783965e-05, "loss": 0.2501098871231079, "memory(GiB)": 70.5, "step": 62135, "token_acc": 0.956989247311828, "train_speed(iter/s)": 1.447444 }, { "epoch": 2.662268111906088, "grad_norm": 1.2636791467666626, "learning_rate": 4.491102430449741e-05, "loss": 0.21295928955078125, "memory(GiB)": 70.5, "step": 62140, "token_acc": 0.9581749049429658, "train_speed(iter/s)": 1.447439 }, { "epoch": 2.662482327235337, "grad_norm": 3.1308646202087402, "learning_rate": 4.490432952540243e-05, "loss": 0.268076753616333, "memory(GiB)": 70.5, "step": 62145, "token_acc": 0.9337748344370861, "train_speed(iter/s)": 1.447446 }, { "epoch": 2.6626965425645857, "grad_norm": 0.8392356634140015, "learning_rate": 4.489763483862031e-05, "loss": 0.22498054504394532, "memory(GiB)": 70.5, "step": 62150, "token_acc": 0.9556962025316456, "train_speed(iter/s)": 1.447461 }, { "epoch": 2.662910757893835, "grad_norm": 0.9999603033065796, "learning_rate": 4.4890940244272305e-05, "loss": 0.43745903968811034, "memory(GiB)": 70.5, "step": 62155, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.6631249732230837, "grad_norm": 3.2843410968780518, "learning_rate": 4.488424574247972e-05, "loss": 0.41991491317749025, "memory(GiB)": 70.5, "step": 62160, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.447456 }, { "epoch": 2.6633391885523325, "grad_norm": 0.5478345155715942, "learning_rate": 4.4877551333363814e-05, "loss": 0.4388895034790039, "memory(GiB)": 70.5, "step": 62165, "token_acc": 0.9051987767584098, "train_speed(iter/s)": 1.447472 }, { "epoch": 2.663553403881582, "grad_norm": 2.8743066787719727, "learning_rate": 4.487085701704588e-05, "loss": 0.3616382360458374, "memory(GiB)": 70.5, "step": 62170, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.447478 }, { "epoch": 2.6637676192108306, "grad_norm": 9.986801147460938, "learning_rate": 4.4864162793647184e-05, "loss": 0.30351901054382324, "memory(GiB)": 70.5, "step": 62175, "token_acc": 0.9433333333333334, "train_speed(iter/s)": 1.447475 }, { "epoch": 2.6639818345400794, "grad_norm": 4.64256477355957, "learning_rate": 4.4857468663288985e-05, "loss": 0.40125384330749514, "memory(GiB)": 70.5, "step": 62180, "token_acc": 0.9022222222222223, "train_speed(iter/s)": 1.447486 }, { "epoch": 2.6641960498693287, "grad_norm": 2.0142953395843506, "learning_rate": 4.485077462609258e-05, "loss": 0.2647902011871338, "memory(GiB)": 70.5, "step": 62185, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.447481 }, { "epoch": 2.6644102651985775, "grad_norm": 4.031135082244873, "learning_rate": 4.484408068217922e-05, "loss": 0.39009039402008056, "memory(GiB)": 70.5, "step": 62190, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.447482 }, { "epoch": 2.6646244805278263, "grad_norm": 2.006535768508911, "learning_rate": 4.4837386831670155e-05, "loss": 0.4520547866821289, "memory(GiB)": 70.5, "step": 62195, "token_acc": 0.9122137404580153, "train_speed(iter/s)": 1.447488 }, { "epoch": 2.6648386958570756, "grad_norm": 4.003526210784912, "learning_rate": 4.4830693074686675e-05, "loss": 0.4088658332824707, "memory(GiB)": 70.5, "step": 62200, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.447503 }, { "epoch": 2.6650529111863244, "grad_norm": 1.7813329696655273, "learning_rate": 4.482399941135005e-05, "loss": 0.3011425256729126, "memory(GiB)": 70.5, "step": 62205, "token_acc": 0.9475409836065574, "train_speed(iter/s)": 1.447503 }, { "epoch": 2.665267126515573, "grad_norm": 3.1795668601989746, "learning_rate": 4.481730584178153e-05, "loss": 0.565117597579956, "memory(GiB)": 70.5, "step": 62210, "token_acc": 0.8715277777777778, "train_speed(iter/s)": 1.447522 }, { "epoch": 2.6654813418448224, "grad_norm": 1.9905132055282593, "learning_rate": 4.481061236610238e-05, "loss": 0.15180439949035646, "memory(GiB)": 70.5, "step": 62215, "token_acc": 0.9715302491103203, "train_speed(iter/s)": 1.44753 }, { "epoch": 2.6656955571740713, "grad_norm": 1.9156246185302734, "learning_rate": 4.480391898443386e-05, "loss": 0.3886182069778442, "memory(GiB)": 70.5, "step": 62220, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.447539 }, { "epoch": 2.66590977250332, "grad_norm": 4.19224214553833, "learning_rate": 4.4797225696897205e-05, "loss": 0.38079047203063965, "memory(GiB)": 70.5, "step": 62225, "token_acc": 0.9338235294117647, "train_speed(iter/s)": 1.447541 }, { "epoch": 2.6661239878325693, "grad_norm": 5.043607711791992, "learning_rate": 4.4790532503613696e-05, "loss": 0.7618361473083496, "memory(GiB)": 70.5, "step": 62230, "token_acc": 0.8421052631578947, "train_speed(iter/s)": 1.447531 }, { "epoch": 2.666338203161818, "grad_norm": 4.006303310394287, "learning_rate": 4.4783839404704587e-05, "loss": 0.44088053703308105, "memory(GiB)": 70.5, "step": 62235, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.447523 }, { "epoch": 2.666552418491067, "grad_norm": 4.6570844650268555, "learning_rate": 4.47771464002911e-05, "loss": 0.2712312459945679, "memory(GiB)": 70.5, "step": 62240, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.447539 }, { "epoch": 2.666766633820316, "grad_norm": 4.354842185974121, "learning_rate": 4.4770453490494526e-05, "loss": 0.89688081741333, "memory(GiB)": 70.5, "step": 62245, "token_acc": 0.8308823529411765, "train_speed(iter/s)": 1.44756 }, { "epoch": 2.666980849149565, "grad_norm": 6.075008869171143, "learning_rate": 4.476376067543607e-05, "loss": 0.4759481430053711, "memory(GiB)": 70.5, "step": 62250, "token_acc": 0.893687707641196, "train_speed(iter/s)": 1.447559 }, { "epoch": 2.6671950644788143, "grad_norm": 3.2484171390533447, "learning_rate": 4.475706795523702e-05, "loss": 0.4639185905456543, "memory(GiB)": 70.5, "step": 62255, "token_acc": 0.9098039215686274, "train_speed(iter/s)": 1.447562 }, { "epoch": 2.667409279808063, "grad_norm": 1.8316489458084106, "learning_rate": 4.47503753300186e-05, "loss": 0.3672872304916382, "memory(GiB)": 70.5, "step": 62260, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.447559 }, { "epoch": 2.667623495137312, "grad_norm": 3.964089870452881, "learning_rate": 4.474368279990205e-05, "loss": 0.36299350261688235, "memory(GiB)": 70.5, "step": 62265, "token_acc": 0.9240924092409241, "train_speed(iter/s)": 1.447563 }, { "epoch": 2.667837710466561, "grad_norm": 0.6748683452606201, "learning_rate": 4.47369903650086e-05, "loss": 0.19935842752456664, "memory(GiB)": 70.5, "step": 62270, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.44757 }, { "epoch": 2.66805192579581, "grad_norm": 4.881246089935303, "learning_rate": 4.4730298025459536e-05, "loss": 0.5943005084991455, "memory(GiB)": 70.5, "step": 62275, "token_acc": 0.9012345679012346, "train_speed(iter/s)": 1.4476 }, { "epoch": 2.6682661411250588, "grad_norm": 5.013237476348877, "learning_rate": 4.4723605781376056e-05, "loss": 0.5519759654998779, "memory(GiB)": 70.5, "step": 62280, "token_acc": 0.8745644599303136, "train_speed(iter/s)": 1.447601 }, { "epoch": 2.668480356454308, "grad_norm": 5.1350531578063965, "learning_rate": 4.47169136328794e-05, "loss": 0.19264392852783202, "memory(GiB)": 70.5, "step": 62285, "token_acc": 0.951310861423221, "train_speed(iter/s)": 1.447614 }, { "epoch": 2.668694571783557, "grad_norm": 0.4157807230949402, "learning_rate": 4.471022158009082e-05, "loss": 0.17727173566818238, "memory(GiB)": 70.5, "step": 62290, "token_acc": 0.965034965034965, "train_speed(iter/s)": 1.447623 }, { "epoch": 2.6689087871128057, "grad_norm": 2.1172196865081787, "learning_rate": 4.470352962313154e-05, "loss": 0.33814690113067625, "memory(GiB)": 70.5, "step": 62295, "token_acc": 0.9385474860335196, "train_speed(iter/s)": 1.447635 }, { "epoch": 2.669123002442055, "grad_norm": 4.236176490783691, "learning_rate": 4.4696837762122777e-05, "loss": 0.4015036582946777, "memory(GiB)": 70.5, "step": 62300, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.447641 }, { "epoch": 2.6693372177713037, "grad_norm": 0.11673542857170105, "learning_rate": 4.469014599718579e-05, "loss": 0.3423347234725952, "memory(GiB)": 70.5, "step": 62305, "token_acc": 0.9172932330827067, "train_speed(iter/s)": 1.447638 }, { "epoch": 2.6695514331005525, "grad_norm": 0.4165034294128418, "learning_rate": 4.468345432844177e-05, "loss": 0.3770503759384155, "memory(GiB)": 70.5, "step": 62310, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.447639 }, { "epoch": 2.669765648429802, "grad_norm": 0.24340839684009552, "learning_rate": 4.4676762756011974e-05, "loss": 0.492935037612915, "memory(GiB)": 70.5, "step": 62315, "token_acc": 0.8696969696969697, "train_speed(iter/s)": 1.447638 }, { "epoch": 2.6699798637590506, "grad_norm": 5.654083251953125, "learning_rate": 4.467007128001762e-05, "loss": 0.3606488943099976, "memory(GiB)": 70.5, "step": 62320, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.447643 }, { "epoch": 2.6701940790882994, "grad_norm": 3.500302791595459, "learning_rate": 4.466337990057991e-05, "loss": 0.3748777866363525, "memory(GiB)": 70.5, "step": 62325, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.447645 }, { "epoch": 2.6704082944175487, "grad_norm": 2.648968458175659, "learning_rate": 4.4656688617820095e-05, "loss": 0.5630839347839356, "memory(GiB)": 70.5, "step": 62330, "token_acc": 0.856687898089172, "train_speed(iter/s)": 1.447647 }, { "epoch": 2.6706225097467975, "grad_norm": 4.108028888702393, "learning_rate": 4.464999743185937e-05, "loss": 0.3390624761581421, "memory(GiB)": 70.5, "step": 62335, "token_acc": 0.9336099585062241, "train_speed(iter/s)": 1.447642 }, { "epoch": 2.6708367250760463, "grad_norm": 0.3022342324256897, "learning_rate": 4.464330634281895e-05, "loss": 0.2710545778274536, "memory(GiB)": 70.5, "step": 62340, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.447645 }, { "epoch": 2.6710509404052956, "grad_norm": 2.3988196849823, "learning_rate": 4.4636615350820054e-05, "loss": 0.14255814552307128, "memory(GiB)": 70.5, "step": 62345, "token_acc": 0.9766666666666667, "train_speed(iter/s)": 1.447649 }, { "epoch": 2.6712651557345444, "grad_norm": 3.5349950790405273, "learning_rate": 4.462992445598392e-05, "loss": 0.4753734111785889, "memory(GiB)": 70.5, "step": 62350, "token_acc": 0.9129032258064517, "train_speed(iter/s)": 1.447655 }, { "epoch": 2.671479371063793, "grad_norm": 6.476291656494141, "learning_rate": 4.462323365843174e-05, "loss": 0.5088140964508057, "memory(GiB)": 70.5, "step": 62355, "token_acc": 0.9018181818181819, "train_speed(iter/s)": 1.447656 }, { "epoch": 2.6716935863930424, "grad_norm": 2.9321882724761963, "learning_rate": 4.4616542958284725e-05, "loss": 0.28748035430908203, "memory(GiB)": 70.5, "step": 62360, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.447662 }, { "epoch": 2.6719078017222913, "grad_norm": 7.791177272796631, "learning_rate": 4.460985235566409e-05, "loss": 0.27793643474578855, "memory(GiB)": 70.5, "step": 62365, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.44766 }, { "epoch": 2.67212201705154, "grad_norm": 0.062391966581344604, "learning_rate": 4.4603161850691025e-05, "loss": 0.3525341033935547, "memory(GiB)": 70.5, "step": 62370, "token_acc": 0.9258160237388724, "train_speed(iter/s)": 1.447655 }, { "epoch": 2.6723362323807893, "grad_norm": 3.9771525859832764, "learning_rate": 4.459647144348675e-05, "loss": 0.7660824298858643, "memory(GiB)": 70.5, "step": 62375, "token_acc": 0.8521739130434782, "train_speed(iter/s)": 1.447659 }, { "epoch": 2.672550447710038, "grad_norm": 3.972144842147827, "learning_rate": 4.458978113417248e-05, "loss": 0.20133619308471679, "memory(GiB)": 70.5, "step": 62380, "token_acc": 0.9586206896551724, "train_speed(iter/s)": 1.447663 }, { "epoch": 2.672764663039287, "grad_norm": 4.696717262268066, "learning_rate": 4.4583090922869375e-05, "loss": 0.45039615631103513, "memory(GiB)": 70.5, "step": 62385, "token_acc": 0.8848684210526315, "train_speed(iter/s)": 1.447663 }, { "epoch": 2.672978878368536, "grad_norm": 1.9007459878921509, "learning_rate": 4.457640080969868e-05, "loss": 0.24563612937927246, "memory(GiB)": 70.5, "step": 62390, "token_acc": 0.941358024691358, "train_speed(iter/s)": 1.447661 }, { "epoch": 2.673193093697785, "grad_norm": 3.1224498748779297, "learning_rate": 4.456971079478155e-05, "loss": 0.5925087928771973, "memory(GiB)": 70.5, "step": 62395, "token_acc": 0.8835616438356164, "train_speed(iter/s)": 1.447669 }, { "epoch": 2.673407309027034, "grad_norm": 5.533380508422852, "learning_rate": 4.456302087823922e-05, "loss": 0.34356322288513186, "memory(GiB)": 70.5, "step": 62400, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.447671 }, { "epoch": 2.673621524356283, "grad_norm": 2.393078565597534, "learning_rate": 4.455633106019287e-05, "loss": 0.4533069133758545, "memory(GiB)": 70.5, "step": 62405, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.447673 }, { "epoch": 2.673835739685532, "grad_norm": 0.07072113454341888, "learning_rate": 4.4549641340763676e-05, "loss": 0.4326894760131836, "memory(GiB)": 70.5, "step": 62410, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.447676 }, { "epoch": 2.6740499550147807, "grad_norm": 3.368162155151367, "learning_rate": 4.454295172007285e-05, "loss": 0.41844024658203127, "memory(GiB)": 70.5, "step": 62415, "token_acc": 0.9219653179190751, "train_speed(iter/s)": 1.447673 }, { "epoch": 2.67426417034403, "grad_norm": 6.23519229888916, "learning_rate": 4.4536262198241555e-05, "loss": 0.255311918258667, "memory(GiB)": 70.5, "step": 62420, "token_acc": 0.9606557377049181, "train_speed(iter/s)": 1.447674 }, { "epoch": 2.674478385673279, "grad_norm": 1.3809280395507812, "learning_rate": 4.4529572775391014e-05, "loss": 0.37593066692352295, "memory(GiB)": 70.5, "step": 62425, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.447676 }, { "epoch": 2.6746926010025276, "grad_norm": 2.3045454025268555, "learning_rate": 4.4522883451642386e-05, "loss": 0.24141299724578857, "memory(GiB)": 70.5, "step": 62430, "token_acc": 0.9493243243243243, "train_speed(iter/s)": 1.447676 }, { "epoch": 2.674906816331777, "grad_norm": 6.614663600921631, "learning_rate": 4.451619422711687e-05, "loss": 0.6233209609985352, "memory(GiB)": 70.5, "step": 62435, "token_acc": 0.8776758409785933, "train_speed(iter/s)": 1.447669 }, { "epoch": 2.6751210316610257, "grad_norm": 1.599133014678955, "learning_rate": 4.4509505101935636e-05, "loss": 0.4985771179199219, "memory(GiB)": 70.5, "step": 62440, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.447671 }, { "epoch": 2.6753352469902745, "grad_norm": 1.482329249382019, "learning_rate": 4.450281607621987e-05, "loss": 0.32178659439086915, "memory(GiB)": 70.5, "step": 62445, "token_acc": 0.9308176100628931, "train_speed(iter/s)": 1.44768 }, { "epoch": 2.6755494623195237, "grad_norm": 4.699796676635742, "learning_rate": 4.449612715009075e-05, "loss": 0.6197531700134278, "memory(GiB)": 70.5, "step": 62450, "token_acc": 0.8639240506329114, "train_speed(iter/s)": 1.447686 }, { "epoch": 2.6757636776487725, "grad_norm": 3.2139315605163574, "learning_rate": 4.4489438323669435e-05, "loss": 0.3811391592025757, "memory(GiB)": 70.5, "step": 62455, "token_acc": 0.9247648902821317, "train_speed(iter/s)": 1.44771 }, { "epoch": 2.6759778929780214, "grad_norm": 2.952954053878784, "learning_rate": 4.448274959707713e-05, "loss": 0.5657593250274658, "memory(GiB)": 70.5, "step": 62460, "token_acc": 0.8660130718954249, "train_speed(iter/s)": 1.447707 }, { "epoch": 2.6761921083072706, "grad_norm": 1.3432409763336182, "learning_rate": 4.447606097043499e-05, "loss": 0.216054105758667, "memory(GiB)": 70.5, "step": 62465, "token_acc": 0.9557522123893806, "train_speed(iter/s)": 1.44771 }, { "epoch": 2.6764063236365194, "grad_norm": 2.007343292236328, "learning_rate": 4.4469372443864185e-05, "loss": 0.3902971506118774, "memory(GiB)": 70.5, "step": 62470, "token_acc": 0.9112627986348123, "train_speed(iter/s)": 1.447715 }, { "epoch": 2.6766205389657682, "grad_norm": 2.974259853363037, "learning_rate": 4.4462684017485884e-05, "loss": 0.21773128509521483, "memory(GiB)": 70.5, "step": 62475, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.447727 }, { "epoch": 2.6768347542950175, "grad_norm": 1.4690396785736084, "learning_rate": 4.445599569142127e-05, "loss": 0.25847735404968264, "memory(GiB)": 70.5, "step": 62480, "token_acc": 0.9531772575250836, "train_speed(iter/s)": 1.447726 }, { "epoch": 2.6770489696242663, "grad_norm": 3.5662496089935303, "learning_rate": 4.444930746579147e-05, "loss": 0.4562369346618652, "memory(GiB)": 70.5, "step": 62485, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.447725 }, { "epoch": 2.677263184953515, "grad_norm": 1.7665257453918457, "learning_rate": 4.444261934071769e-05, "loss": 0.20178804397583008, "memory(GiB)": 70.5, "step": 62490, "token_acc": 0.9607843137254902, "train_speed(iter/s)": 1.447729 }, { "epoch": 2.6774774002827644, "grad_norm": 0.15136446058750153, "learning_rate": 4.443593131632105e-05, "loss": 0.38600869178771974, "memory(GiB)": 70.5, "step": 62495, "token_acc": 0.9119318181818182, "train_speed(iter/s)": 1.447741 }, { "epoch": 2.677691615612013, "grad_norm": 2.030184030532837, "learning_rate": 4.442924339272275e-05, "loss": 0.3512770652770996, "memory(GiB)": 70.5, "step": 62500, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.447763 }, { "epoch": 2.677691615612013, "eval_loss": 2.3742074966430664, "eval_runtime": 13.6968, "eval_samples_per_second": 7.301, "eval_steps_per_second": 7.301, "eval_token_acc": 0.4540540540540541, "step": 62500 }, { "epoch": 2.677905830941262, "grad_norm": 4.145299434661865, "learning_rate": 4.442255557004393e-05, "loss": 0.4555978775024414, "memory(GiB)": 70.5, "step": 62505, "token_acc": 0.5715695952615992, "train_speed(iter/s)": 1.447273 }, { "epoch": 2.6781200462705113, "grad_norm": 0.8118100762367249, "learning_rate": 4.441586784840576e-05, "loss": 0.07326777577400208, "memory(GiB)": 70.5, "step": 62510, "token_acc": 0.9834710743801653, "train_speed(iter/s)": 1.447274 }, { "epoch": 2.67833426159976, "grad_norm": 2.789337635040283, "learning_rate": 4.440918022792937e-05, "loss": 0.3287363052368164, "memory(GiB)": 70.5, "step": 62515, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.447289 }, { "epoch": 2.678548476929009, "grad_norm": 2.113752603530884, "learning_rate": 4.440249270873593e-05, "loss": 0.3011852025985718, "memory(GiB)": 70.5, "step": 62520, "token_acc": 0.9409448818897638, "train_speed(iter/s)": 1.447296 }, { "epoch": 2.678762692258258, "grad_norm": 3.925823211669922, "learning_rate": 4.439580529094659e-05, "loss": 0.3577587127685547, "memory(GiB)": 70.5, "step": 62525, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.447296 }, { "epoch": 2.678976907587507, "grad_norm": 3.3004872798919678, "learning_rate": 4.4389117974682484e-05, "loss": 0.5625218391418457, "memory(GiB)": 70.5, "step": 62530, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.447302 }, { "epoch": 2.6791911229167558, "grad_norm": 3.3272581100463867, "learning_rate": 4.4382430760064774e-05, "loss": 0.32215204238891604, "memory(GiB)": 70.5, "step": 62535, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.447306 }, { "epoch": 2.679405338246005, "grad_norm": 4.458067417144775, "learning_rate": 4.4375743647214596e-05, "loss": 0.3722968101501465, "memory(GiB)": 70.5, "step": 62540, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.447303 }, { "epoch": 2.679619553575254, "grad_norm": 5.582276344299316, "learning_rate": 4.436905663625311e-05, "loss": 0.5283791542053222, "memory(GiB)": 70.5, "step": 62545, "token_acc": 0.8785425101214575, "train_speed(iter/s)": 1.447309 }, { "epoch": 2.6798337689045026, "grad_norm": 3.107813596725464, "learning_rate": 4.436236972730144e-05, "loss": 0.5672366619110107, "memory(GiB)": 70.5, "step": 62550, "token_acc": 0.8560311284046692, "train_speed(iter/s)": 1.447306 }, { "epoch": 2.680047984233752, "grad_norm": 4.1710591316223145, "learning_rate": 4.435568292048072e-05, "loss": 0.4705760955810547, "memory(GiB)": 70.5, "step": 62555, "token_acc": 0.8943661971830986, "train_speed(iter/s)": 1.447318 }, { "epoch": 2.6802621995630007, "grad_norm": 2.6885411739349365, "learning_rate": 4.4348996215912114e-05, "loss": 0.3069709062576294, "memory(GiB)": 70.5, "step": 62560, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.447335 }, { "epoch": 2.6804764148922495, "grad_norm": 2.937551259994507, "learning_rate": 4.434230961371674e-05, "loss": 0.6702749252319335, "memory(GiB)": 70.5, "step": 62565, "token_acc": 0.8981723237597912, "train_speed(iter/s)": 1.447337 }, { "epoch": 2.680690630221499, "grad_norm": 4.211277484893799, "learning_rate": 4.433562311401571e-05, "loss": 0.4159238338470459, "memory(GiB)": 70.5, "step": 62570, "token_acc": 0.9079365079365079, "train_speed(iter/s)": 1.447335 }, { "epoch": 2.6809048455507476, "grad_norm": 2.659636974334717, "learning_rate": 4.43289367169302e-05, "loss": 0.3950314998626709, "memory(GiB)": 70.5, "step": 62575, "token_acc": 0.91875, "train_speed(iter/s)": 1.447337 }, { "epoch": 2.6811190608799964, "grad_norm": 3.4124321937561035, "learning_rate": 4.4322250422581326e-05, "loss": 0.24736375808715821, "memory(GiB)": 70.5, "step": 62580, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.447333 }, { "epoch": 2.6813332762092457, "grad_norm": 2.8699426651000977, "learning_rate": 4.431556423109021e-05, "loss": 0.4144914627075195, "memory(GiB)": 70.5, "step": 62585, "token_acc": 0.9069148936170213, "train_speed(iter/s)": 1.447344 }, { "epoch": 2.6815474915384945, "grad_norm": 0.2609194219112396, "learning_rate": 4.430887814257798e-05, "loss": 0.43549046516418455, "memory(GiB)": 70.5, "step": 62590, "token_acc": 0.9177631578947368, "train_speed(iter/s)": 1.447346 }, { "epoch": 2.6817617068677437, "grad_norm": 11.140837669372559, "learning_rate": 4.430219215716576e-05, "loss": 0.26322734355926514, "memory(GiB)": 70.5, "step": 62595, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.447345 }, { "epoch": 2.6819759221969925, "grad_norm": 3.288661241531372, "learning_rate": 4.429550627497467e-05, "loss": 0.2932554006576538, "memory(GiB)": 70.5, "step": 62600, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.44735 }, { "epoch": 2.6821901375262414, "grad_norm": 4.707742214202881, "learning_rate": 4.428882049612584e-05, "loss": 0.4940946102142334, "memory(GiB)": 70.5, "step": 62605, "token_acc": 0.8966565349544073, "train_speed(iter/s)": 1.447344 }, { "epoch": 2.6824043528554906, "grad_norm": 3.106504440307617, "learning_rate": 4.428213482074039e-05, "loss": 0.3149724960327148, "memory(GiB)": 70.5, "step": 62610, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.447339 }, { "epoch": 2.6826185681847394, "grad_norm": 3.8156683444976807, "learning_rate": 4.427544924893941e-05, "loss": 0.44593062400817873, "memory(GiB)": 70.5, "step": 62615, "token_acc": 0.937007874015748, "train_speed(iter/s)": 1.447344 }, { "epoch": 2.6828327835139882, "grad_norm": 4.160550117492676, "learning_rate": 4.426876378084406e-05, "loss": 0.4735736846923828, "memory(GiB)": 70.5, "step": 62620, "token_acc": 0.9283489096573209, "train_speed(iter/s)": 1.447349 }, { "epoch": 2.6830469988432375, "grad_norm": 2.1456682682037354, "learning_rate": 4.426207841657543e-05, "loss": 0.4026142120361328, "memory(GiB)": 70.5, "step": 62625, "token_acc": 0.8798586572438163, "train_speed(iter/s)": 1.447349 }, { "epoch": 2.6832612141724863, "grad_norm": 3.410428285598755, "learning_rate": 4.425539315625462e-05, "loss": 0.20670638084411622, "memory(GiB)": 70.5, "step": 62630, "token_acc": 0.947565543071161, "train_speed(iter/s)": 1.447367 }, { "epoch": 2.683475429501735, "grad_norm": 4.5037078857421875, "learning_rate": 4.424870800000276e-05, "loss": 0.35214216709136964, "memory(GiB)": 70.5, "step": 62635, "token_acc": 0.9218241042345277, "train_speed(iter/s)": 1.44738 }, { "epoch": 2.6836896448309844, "grad_norm": 6.571510314941406, "learning_rate": 4.424202294794093e-05, "loss": 0.3896992921829224, "memory(GiB)": 70.5, "step": 62640, "token_acc": 0.9174041297935103, "train_speed(iter/s)": 1.447381 }, { "epoch": 2.683903860160233, "grad_norm": 2.7848527431488037, "learning_rate": 4.423533800019026e-05, "loss": 0.377701997756958, "memory(GiB)": 70.5, "step": 62645, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.447384 }, { "epoch": 2.684118075489482, "grad_norm": 3.359452724456787, "learning_rate": 4.422865315687187e-05, "loss": 0.2985809803009033, "memory(GiB)": 70.5, "step": 62650, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.44739 }, { "epoch": 2.6843322908187313, "grad_norm": 2.0841283798217773, "learning_rate": 4.4221968418106844e-05, "loss": 0.29588093757629397, "memory(GiB)": 70.5, "step": 62655, "token_acc": 0.948220064724919, "train_speed(iter/s)": 1.44739 }, { "epoch": 2.68454650614798, "grad_norm": 0.2556278109550476, "learning_rate": 4.421528378401626e-05, "loss": 0.22843804359436035, "memory(GiB)": 70.5, "step": 62660, "token_acc": 0.9471698113207547, "train_speed(iter/s)": 1.447391 }, { "epoch": 2.684760721477229, "grad_norm": 4.95668888092041, "learning_rate": 4.420859925472125e-05, "loss": 0.5178972244262695, "memory(GiB)": 70.5, "step": 62665, "token_acc": 0.9099378881987578, "train_speed(iter/s)": 1.447404 }, { "epoch": 2.684974936806478, "grad_norm": 3.0805540084838867, "learning_rate": 4.42019148303429e-05, "loss": 0.753924036026001, "memory(GiB)": 70.5, "step": 62670, "token_acc": 0.8601823708206687, "train_speed(iter/s)": 1.447416 }, { "epoch": 2.685189152135727, "grad_norm": 2.0419833660125732, "learning_rate": 4.419523051100229e-05, "loss": 0.5576873779296875, "memory(GiB)": 70.5, "step": 62675, "token_acc": 0.8766233766233766, "train_speed(iter/s)": 1.44743 }, { "epoch": 2.6854033674649758, "grad_norm": 4.344888687133789, "learning_rate": 4.418854629682053e-05, "loss": 0.5703277111053466, "memory(GiB)": 70.5, "step": 62680, "token_acc": 0.8870431893687708, "train_speed(iter/s)": 1.447428 }, { "epoch": 2.685617582794225, "grad_norm": 0.5150325298309326, "learning_rate": 4.41818621879187e-05, "loss": 0.7700783252716065, "memory(GiB)": 70.5, "step": 62685, "token_acc": 0.8676923076923077, "train_speed(iter/s)": 1.447462 }, { "epoch": 2.685831798123474, "grad_norm": 2.055495262145996, "learning_rate": 4.41751781844179e-05, "loss": 0.07775970101356507, "memory(GiB)": 70.5, "step": 62690, "token_acc": 0.9818181818181818, "train_speed(iter/s)": 1.447463 }, { "epoch": 2.6860460134527226, "grad_norm": 1.3722928762435913, "learning_rate": 4.416849428643922e-05, "loss": 0.5592421054840088, "memory(GiB)": 70.5, "step": 62695, "token_acc": 0.8774373259052924, "train_speed(iter/s)": 1.447469 }, { "epoch": 2.686260228781972, "grad_norm": 1.6199452877044678, "learning_rate": 4.416181049410372e-05, "loss": 0.28538951873779295, "memory(GiB)": 70.5, "step": 62700, "token_acc": 0.9552238805970149, "train_speed(iter/s)": 1.447475 }, { "epoch": 2.6864744441112207, "grad_norm": 5.531240463256836, "learning_rate": 4.415512680753251e-05, "loss": 0.3770728588104248, "memory(GiB)": 70.5, "step": 62705, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.447477 }, { "epoch": 2.6866886594404695, "grad_norm": 2.096184492111206, "learning_rate": 4.414844322684667e-05, "loss": 0.3095993518829346, "memory(GiB)": 70.5, "step": 62710, "token_acc": 0.9321533923303835, "train_speed(iter/s)": 1.447473 }, { "epoch": 2.686902874769719, "grad_norm": 11.800418853759766, "learning_rate": 4.414175975216724e-05, "loss": 0.35005435943603513, "memory(GiB)": 70.5, "step": 62715, "token_acc": 0.92, "train_speed(iter/s)": 1.447486 }, { "epoch": 2.6871170900989676, "grad_norm": 2.22708797454834, "learning_rate": 4.413507638361534e-05, "loss": 0.41910924911499026, "memory(GiB)": 70.5, "step": 62720, "token_acc": 0.9170305676855895, "train_speed(iter/s)": 1.447493 }, { "epoch": 2.6873313054282164, "grad_norm": 3.320981740951538, "learning_rate": 4.412839312131204e-05, "loss": 0.5126067638397217, "memory(GiB)": 70.5, "step": 62725, "token_acc": 0.8901515151515151, "train_speed(iter/s)": 1.447506 }, { "epoch": 2.6875455207574657, "grad_norm": 1.8621397018432617, "learning_rate": 4.41217099653784e-05, "loss": 0.4814084529876709, "memory(GiB)": 70.5, "step": 62730, "token_acc": 0.8954248366013072, "train_speed(iter/s)": 1.447516 }, { "epoch": 2.6877597360867145, "grad_norm": 3.1632602214813232, "learning_rate": 4.411502691593551e-05, "loss": 0.4468897819519043, "memory(GiB)": 70.5, "step": 62735, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.447513 }, { "epoch": 2.6879739514159633, "grad_norm": 2.2920820713043213, "learning_rate": 4.410834397310443e-05, "loss": 0.25765442848205566, "memory(GiB)": 70.5, "step": 62740, "token_acc": 0.9579579579579579, "train_speed(iter/s)": 1.447521 }, { "epoch": 2.6881881667452125, "grad_norm": 2.7502663135528564, "learning_rate": 4.410166113700621e-05, "loss": 0.6375481605529785, "memory(GiB)": 70.5, "step": 62745, "token_acc": 0.8556701030927835, "train_speed(iter/s)": 1.447529 }, { "epoch": 2.6884023820744614, "grad_norm": 0.7795251607894897, "learning_rate": 4.4094978407761936e-05, "loss": 0.3453514099121094, "memory(GiB)": 70.5, "step": 62750, "token_acc": 0.916, "train_speed(iter/s)": 1.44753 }, { "epoch": 2.68861659740371, "grad_norm": 3.6598055362701416, "learning_rate": 4.408829578549268e-05, "loss": 0.2537302255630493, "memory(GiB)": 70.5, "step": 62755, "token_acc": 0.937007874015748, "train_speed(iter/s)": 1.447528 }, { "epoch": 2.6888308127329594, "grad_norm": 3.3972690105438232, "learning_rate": 4.4081613270319476e-05, "loss": 0.3418891429901123, "memory(GiB)": 70.5, "step": 62760, "token_acc": 0.910394265232975, "train_speed(iter/s)": 1.447522 }, { "epoch": 2.6890450280622082, "grad_norm": 2.679300546646118, "learning_rate": 4.407493086236341e-05, "loss": 0.35912585258483887, "memory(GiB)": 70.5, "step": 62765, "token_acc": 0.90234375, "train_speed(iter/s)": 1.447518 }, { "epoch": 2.689259243391457, "grad_norm": 3.471464157104492, "learning_rate": 4.406824856174552e-05, "loss": 0.2813483715057373, "memory(GiB)": 70.5, "step": 62770, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.447526 }, { "epoch": 2.6894734587207063, "grad_norm": 5.069715976715088, "learning_rate": 4.406156636858688e-05, "loss": 0.35245299339294434, "memory(GiB)": 70.5, "step": 62775, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.447525 }, { "epoch": 2.689687674049955, "grad_norm": 2.847865104675293, "learning_rate": 4.4054884283008534e-05, "loss": 0.3333984375, "memory(GiB)": 70.5, "step": 62780, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.447521 }, { "epoch": 2.689901889379204, "grad_norm": 2.9036197662353516, "learning_rate": 4.404820230513153e-05, "loss": 0.6692329406738281, "memory(GiB)": 70.5, "step": 62785, "token_acc": 0.8627450980392157, "train_speed(iter/s)": 1.447535 }, { "epoch": 2.690116104708453, "grad_norm": 3.0980613231658936, "learning_rate": 4.404152043507692e-05, "loss": 0.28593313694000244, "memory(GiB)": 70.5, "step": 62790, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.447543 }, { "epoch": 2.690330320037702, "grad_norm": 0.02017628401517868, "learning_rate": 4.4034838672965764e-05, "loss": 0.3849677801132202, "memory(GiB)": 70.5, "step": 62795, "token_acc": 0.9254658385093167, "train_speed(iter/s)": 1.447546 }, { "epoch": 2.690544535366951, "grad_norm": 4.203407287597656, "learning_rate": 4.4028157018919106e-05, "loss": 0.3528479814529419, "memory(GiB)": 70.5, "step": 62800, "token_acc": 0.909433962264151, "train_speed(iter/s)": 1.447539 }, { "epoch": 2.6907587506962, "grad_norm": 3.6012959480285645, "learning_rate": 4.4021475473057984e-05, "loss": 0.3061349391937256, "memory(GiB)": 70.5, "step": 62805, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.447539 }, { "epoch": 2.690972966025449, "grad_norm": 6.448614120483398, "learning_rate": 4.401479403550344e-05, "loss": 0.2852522850036621, "memory(GiB)": 70.5, "step": 62810, "token_acc": 0.9312977099236641, "train_speed(iter/s)": 1.447549 }, { "epoch": 2.6911871813546977, "grad_norm": 3.053342819213867, "learning_rate": 4.400811270637652e-05, "loss": 0.19390342235565186, "memory(GiB)": 70.5, "step": 62815, "token_acc": 0.9611940298507463, "train_speed(iter/s)": 1.447549 }, { "epoch": 2.691401396683947, "grad_norm": 3.524334669113159, "learning_rate": 4.400143148579826e-05, "loss": 0.665482473373413, "memory(GiB)": 70.5, "step": 62820, "token_acc": 0.875, "train_speed(iter/s)": 1.447546 }, { "epoch": 2.6916156120131958, "grad_norm": 0.8630567789077759, "learning_rate": 4.399475037388969e-05, "loss": 0.23399031162261963, "memory(GiB)": 70.5, "step": 62825, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.44755 }, { "epoch": 2.6918298273424446, "grad_norm": 1.3126429319381714, "learning_rate": 4.398806937077185e-05, "loss": 0.31683566570281985, "memory(GiB)": 70.5, "step": 62830, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.447549 }, { "epoch": 2.692044042671694, "grad_norm": 2.459117889404297, "learning_rate": 4.398138847656578e-05, "loss": 0.5016665458679199, "memory(GiB)": 70.5, "step": 62835, "token_acc": 0.8722741433021807, "train_speed(iter/s)": 1.447563 }, { "epoch": 2.6922582580009427, "grad_norm": 1.3627525568008423, "learning_rate": 4.3974707691392503e-05, "loss": 0.3127125263214111, "memory(GiB)": 70.5, "step": 62840, "token_acc": 0.9365558912386707, "train_speed(iter/s)": 1.447564 }, { "epoch": 2.6924724733301915, "grad_norm": 2.897458076477051, "learning_rate": 4.396802701537304e-05, "loss": 0.40622668266296386, "memory(GiB)": 70.5, "step": 62845, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.447565 }, { "epoch": 2.6926866886594407, "grad_norm": 2.681706190109253, "learning_rate": 4.396134644862844e-05, "loss": 0.33518352508544924, "memory(GiB)": 70.5, "step": 62850, "token_acc": 0.9488054607508533, "train_speed(iter/s)": 1.447574 }, { "epoch": 2.6929009039886895, "grad_norm": 6.306046485900879, "learning_rate": 4.395466599127971e-05, "loss": 0.4672982692718506, "memory(GiB)": 70.5, "step": 62855, "token_acc": 0.8905775075987842, "train_speed(iter/s)": 1.447576 }, { "epoch": 2.6931151193179383, "grad_norm": 4.770489692687988, "learning_rate": 4.3947985643447866e-05, "loss": 0.26877219676971437, "memory(GiB)": 70.5, "step": 62860, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.447594 }, { "epoch": 2.6933293346471876, "grad_norm": 3.2695963382720947, "learning_rate": 4.394130540525392e-05, "loss": 0.32145204544067385, "memory(GiB)": 70.5, "step": 62865, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.447594 }, { "epoch": 2.6935435499764364, "grad_norm": 4.191781520843506, "learning_rate": 4.393462527681894e-05, "loss": 0.4408303737640381, "memory(GiB)": 70.5, "step": 62870, "token_acc": 0.8831168831168831, "train_speed(iter/s)": 1.447608 }, { "epoch": 2.6937577653056852, "grad_norm": 7.554002285003662, "learning_rate": 4.39279452582639e-05, "loss": 0.43377046585083007, "memory(GiB)": 70.5, "step": 62875, "token_acc": 0.9228395061728395, "train_speed(iter/s)": 1.447624 }, { "epoch": 2.6939719806349345, "grad_norm": 3.4975521564483643, "learning_rate": 4.3921265349709844e-05, "loss": 0.39728431701660155, "memory(GiB)": 70.5, "step": 62880, "token_acc": 0.9105960264900662, "train_speed(iter/s)": 1.447621 }, { "epoch": 2.6941861959641833, "grad_norm": 6.700943946838379, "learning_rate": 4.391458555127777e-05, "loss": 0.3234791994094849, "memory(GiB)": 70.5, "step": 62885, "token_acc": 0.9283018867924528, "train_speed(iter/s)": 1.447624 }, { "epoch": 2.694400411293432, "grad_norm": 4.013463020324707, "learning_rate": 4.390790586308867e-05, "loss": 0.4613466262817383, "memory(GiB)": 70.5, "step": 62890, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.447648 }, { "epoch": 2.6946146266226814, "grad_norm": 4.478011131286621, "learning_rate": 4.390122628526358e-05, "loss": 0.8812132835388183, "memory(GiB)": 70.5, "step": 62895, "token_acc": 0.8361774744027304, "train_speed(iter/s)": 1.447665 }, { "epoch": 2.69482884195193, "grad_norm": 3.2755448818206787, "learning_rate": 4.38945468179235e-05, "loss": 0.3072239398956299, "memory(GiB)": 70.5, "step": 62900, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.447669 }, { "epoch": 2.695043057281179, "grad_norm": 4.341145992279053, "learning_rate": 4.3887867461189416e-05, "loss": 0.3014104127883911, "memory(GiB)": 70.5, "step": 62905, "token_acc": 0.9457364341085271, "train_speed(iter/s)": 1.447665 }, { "epoch": 2.6952572726104282, "grad_norm": 5.72363805770874, "learning_rate": 4.388118821518236e-05, "loss": 0.2478041648864746, "memory(GiB)": 70.5, "step": 62910, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.447675 }, { "epoch": 2.695471487939677, "grad_norm": 2.1828114986419678, "learning_rate": 4.3874509080023315e-05, "loss": 0.3195215702056885, "memory(GiB)": 70.5, "step": 62915, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.447674 }, { "epoch": 2.695685703268926, "grad_norm": 4.725935935974121, "learning_rate": 4.3867830055833284e-05, "loss": 0.49680542945861816, "memory(GiB)": 70.5, "step": 62920, "token_acc": 0.875, "train_speed(iter/s)": 1.447668 }, { "epoch": 2.695899918598175, "grad_norm": 4.044095039367676, "learning_rate": 4.386115114273328e-05, "loss": 0.41925849914550783, "memory(GiB)": 70.5, "step": 62925, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.447662 }, { "epoch": 2.696114133927424, "grad_norm": 6.429652690887451, "learning_rate": 4.385447234084426e-05, "loss": 0.40684189796447756, "memory(GiB)": 70.5, "step": 62930, "token_acc": 0.9288537549407114, "train_speed(iter/s)": 1.447666 }, { "epoch": 2.6963283492566728, "grad_norm": 0.6925517320632935, "learning_rate": 4.384779365028722e-05, "loss": 0.14581238031387328, "memory(GiB)": 70.5, "step": 62935, "token_acc": 0.9625468164794008, "train_speed(iter/s)": 1.447665 }, { "epoch": 2.696542564585922, "grad_norm": 2.2698352336883545, "learning_rate": 4.38411150711832e-05, "loss": 0.3435510635375977, "memory(GiB)": 70.5, "step": 62940, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.44767 }, { "epoch": 2.696756779915171, "grad_norm": 6.57139778137207, "learning_rate": 4.383443660365316e-05, "loss": 0.3891897201538086, "memory(GiB)": 70.5, "step": 62945, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.447666 }, { "epoch": 2.6969709952444196, "grad_norm": 2.77087664604187, "learning_rate": 4.3827758247818075e-05, "loss": 0.20450782775878906, "memory(GiB)": 70.5, "step": 62950, "token_acc": 0.954248366013072, "train_speed(iter/s)": 1.447661 }, { "epoch": 2.697185210573669, "grad_norm": 5.028747081756592, "learning_rate": 4.382108000379894e-05, "loss": 0.4343697547912598, "memory(GiB)": 70.5, "step": 62955, "token_acc": 0.8971631205673759, "train_speed(iter/s)": 1.447657 }, { "epoch": 2.6973994259029177, "grad_norm": 0.08896806091070175, "learning_rate": 4.381440187171675e-05, "loss": 0.3048293113708496, "memory(GiB)": 70.5, "step": 62960, "token_acc": 0.9337748344370861, "train_speed(iter/s)": 1.447657 }, { "epoch": 2.6976136412321665, "grad_norm": 5.543375492095947, "learning_rate": 4.380772385169245e-05, "loss": 0.5845470905303956, "memory(GiB)": 70.5, "step": 62965, "token_acc": 0.8576642335766423, "train_speed(iter/s)": 1.447668 }, { "epoch": 2.6978278565614158, "grad_norm": 1.4817026853561401, "learning_rate": 4.3801045943847064e-05, "loss": 0.22683207988739013, "memory(GiB)": 70.5, "step": 62970, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.447666 }, { "epoch": 2.6980420718906646, "grad_norm": 7.120710372924805, "learning_rate": 4.3794368148301525e-05, "loss": 0.49210367202758787, "memory(GiB)": 70.5, "step": 62975, "token_acc": 0.8804347826086957, "train_speed(iter/s)": 1.447668 }, { "epoch": 2.6982562872199134, "grad_norm": 2.2951016426086426, "learning_rate": 4.378769046517685e-05, "loss": 0.30957343578338625, "memory(GiB)": 70.5, "step": 62980, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.447666 }, { "epoch": 2.6984705025491627, "grad_norm": 3.5682291984558105, "learning_rate": 4.3781012894593975e-05, "loss": 0.30461099147796633, "memory(GiB)": 70.5, "step": 62985, "token_acc": 0.9363957597173145, "train_speed(iter/s)": 1.447668 }, { "epoch": 2.6986847178784115, "grad_norm": 3.3019087314605713, "learning_rate": 4.377433543667388e-05, "loss": 0.3384881496429443, "memory(GiB)": 70.5, "step": 62990, "token_acc": 0.9463722397476341, "train_speed(iter/s)": 1.447668 }, { "epoch": 2.6988989332076603, "grad_norm": 6.223130226135254, "learning_rate": 4.376765809153755e-05, "loss": 0.504764461517334, "memory(GiB)": 70.5, "step": 62995, "token_acc": 0.9030303030303031, "train_speed(iter/s)": 1.447663 }, { "epoch": 2.6991131485369095, "grad_norm": 4.839898586273193, "learning_rate": 4.376098085930594e-05, "loss": 0.4773582935333252, "memory(GiB)": 70.5, "step": 63000, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.447666 }, { "epoch": 2.6991131485369095, "eval_loss": 2.6757211685180664, "eval_runtime": 14.0026, "eval_samples_per_second": 7.142, "eval_steps_per_second": 7.142, "eval_token_acc": 0.4282178217821782, "step": 63000 }, { "epoch": 2.6993273638661583, "grad_norm": 4.911745071411133, "learning_rate": 4.375430374010001e-05, "loss": 0.40027856826782227, "memory(GiB)": 70.5, "step": 63005, "token_acc": 0.5459662288930581, "train_speed(iter/s)": 1.44718 }, { "epoch": 2.699541579195407, "grad_norm": 1.2048588991165161, "learning_rate": 4.3747626734040716e-05, "loss": 0.17251538038253783, "memory(GiB)": 70.5, "step": 63010, "token_acc": 0.9486404833836858, "train_speed(iter/s)": 1.447173 }, { "epoch": 2.6997557945246564, "grad_norm": 2.815121650695801, "learning_rate": 4.374094984124904e-05, "loss": 0.7350337505340576, "memory(GiB)": 70.5, "step": 63015, "token_acc": 0.8402555910543131, "train_speed(iter/s)": 1.447188 }, { "epoch": 2.6999700098539052, "grad_norm": 4.631516456604004, "learning_rate": 4.3734273061845926e-05, "loss": 0.42221593856811523, "memory(GiB)": 70.5, "step": 63020, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.447194 }, { "epoch": 2.700184225183154, "grad_norm": 2.4700801372528076, "learning_rate": 4.372759639595234e-05, "loss": 0.2731067419052124, "memory(GiB)": 70.5, "step": 63025, "token_acc": 0.9467680608365019, "train_speed(iter/s)": 1.447201 }, { "epoch": 2.7003984405124033, "grad_norm": 3.4681127071380615, "learning_rate": 4.3720919843689236e-05, "loss": 0.15780282020568848, "memory(GiB)": 70.5, "step": 63030, "token_acc": 0.9592592592592593, "train_speed(iter/s)": 1.447194 }, { "epoch": 2.700612655841652, "grad_norm": 0.08073275536298752, "learning_rate": 4.371424340517754e-05, "loss": 0.2566578149795532, "memory(GiB)": 70.5, "step": 63035, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.447173 }, { "epoch": 2.700826871170901, "grad_norm": 4.326213836669922, "learning_rate": 4.3707567080538235e-05, "loss": 0.35116479396820066, "memory(GiB)": 70.5, "step": 63040, "token_acc": 0.9338235294117647, "train_speed(iter/s)": 1.447183 }, { "epoch": 2.70104108650015, "grad_norm": 3.23868989944458, "learning_rate": 4.370089086989225e-05, "loss": 0.484804105758667, "memory(GiB)": 70.5, "step": 63045, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.447185 }, { "epoch": 2.701255301829399, "grad_norm": 4.342711448669434, "learning_rate": 4.369421477336054e-05, "loss": 0.3775313138961792, "memory(GiB)": 70.5, "step": 63050, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.447195 }, { "epoch": 2.701469517158648, "grad_norm": 0.6932234764099121, "learning_rate": 4.368753879106404e-05, "loss": 0.15355197191238404, "memory(GiB)": 70.5, "step": 63055, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.447201 }, { "epoch": 2.701683732487897, "grad_norm": 1.7476184368133545, "learning_rate": 4.368086292312369e-05, "loss": 0.35852599143981934, "memory(GiB)": 70.5, "step": 63060, "token_acc": 0.92, "train_speed(iter/s)": 1.447199 }, { "epoch": 2.701897947817146, "grad_norm": 2.707198143005371, "learning_rate": 4.367418716966045e-05, "loss": 0.33435373306274413, "memory(GiB)": 70.5, "step": 63065, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.447196 }, { "epoch": 2.7021121631463947, "grad_norm": 1.8437947034835815, "learning_rate": 4.366751153079525e-05, "loss": 0.20994036197662352, "memory(GiB)": 70.5, "step": 63070, "token_acc": 0.9628252788104089, "train_speed(iter/s)": 1.447204 }, { "epoch": 2.702326378475644, "grad_norm": 4.162490367889404, "learning_rate": 4.3660836006649e-05, "loss": 0.40768895149230955, "memory(GiB)": 70.5, "step": 63075, "token_acc": 0.9201277955271565, "train_speed(iter/s)": 1.447216 }, { "epoch": 2.7025405938048928, "grad_norm": 4.4268903732299805, "learning_rate": 4.365416059734266e-05, "loss": 0.34325194358825684, "memory(GiB)": 70.5, "step": 63080, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.447215 }, { "epoch": 2.7027548091341416, "grad_norm": 4.833093643188477, "learning_rate": 4.364748530299714e-05, "loss": 0.557254695892334, "memory(GiB)": 70.5, "step": 63085, "token_acc": 0.9068100358422939, "train_speed(iter/s)": 1.447219 }, { "epoch": 2.702969024463391, "grad_norm": 6.19686222076416, "learning_rate": 4.364081012373339e-05, "loss": 0.3484412431716919, "memory(GiB)": 70.5, "step": 63090, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.44722 }, { "epoch": 2.7031832397926396, "grad_norm": 4.593024253845215, "learning_rate": 4.363413505967233e-05, "loss": 0.3468609094619751, "memory(GiB)": 70.5, "step": 63095, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.447233 }, { "epoch": 2.7033974551218884, "grad_norm": 0.3892802298069, "learning_rate": 4.36274601109349e-05, "loss": 0.18218854665756226, "memory(GiB)": 70.5, "step": 63100, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.447229 }, { "epoch": 2.7036116704511377, "grad_norm": 1.1087315082550049, "learning_rate": 4.3620785277642004e-05, "loss": 0.21635661125183106, "memory(GiB)": 70.5, "step": 63105, "token_acc": 0.9540229885057471, "train_speed(iter/s)": 1.447229 }, { "epoch": 2.7038258857803865, "grad_norm": 3.458427906036377, "learning_rate": 4.3614110559914555e-05, "loss": 0.3509517669677734, "memory(GiB)": 70.5, "step": 63110, "token_acc": 0.933993399339934, "train_speed(iter/s)": 1.447232 }, { "epoch": 2.7040401011096353, "grad_norm": 3.0557100772857666, "learning_rate": 4.360743595787349e-05, "loss": 0.4397778034210205, "memory(GiB)": 70.5, "step": 63115, "token_acc": 0.8986928104575164, "train_speed(iter/s)": 1.44722 }, { "epoch": 2.7042543164388846, "grad_norm": 4.620411396026611, "learning_rate": 4.360076147163972e-05, "loss": 0.34035615921020507, "memory(GiB)": 70.5, "step": 63120, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.447213 }, { "epoch": 2.7044685317681334, "grad_norm": 5.604130268096924, "learning_rate": 4.3594087101334164e-05, "loss": 0.3975877046585083, "memory(GiB)": 70.5, "step": 63125, "token_acc": 0.9256965944272446, "train_speed(iter/s)": 1.447218 }, { "epoch": 2.704682747097382, "grad_norm": 3.9504542350769043, "learning_rate": 4.3587412847077726e-05, "loss": 0.5005442142486572, "memory(GiB)": 70.5, "step": 63130, "token_acc": 0.903125, "train_speed(iter/s)": 1.447226 }, { "epoch": 2.7048969624266315, "grad_norm": 2.3871257305145264, "learning_rate": 4.358073870899131e-05, "loss": 0.11631497144699096, "memory(GiB)": 70.5, "step": 63135, "token_acc": 0.9713375796178344, "train_speed(iter/s)": 1.447231 }, { "epoch": 2.7051111777558803, "grad_norm": 2.9666812419891357, "learning_rate": 4.3574064687195846e-05, "loss": 0.31323943138122556, "memory(GiB)": 70.5, "step": 63140, "token_acc": 0.9348659003831418, "train_speed(iter/s)": 1.447238 }, { "epoch": 2.705325393085129, "grad_norm": 3.550665855407715, "learning_rate": 4.356739078181223e-05, "loss": 0.30730609893798827, "memory(GiB)": 70.5, "step": 63145, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.447233 }, { "epoch": 2.7055396084143783, "grad_norm": 3.079688310623169, "learning_rate": 4.356071699296135e-05, "loss": 0.28371553421020507, "memory(GiB)": 70.5, "step": 63150, "token_acc": 0.95, "train_speed(iter/s)": 1.447247 }, { "epoch": 2.705753823743627, "grad_norm": 4.671093940734863, "learning_rate": 4.3554043320764134e-05, "loss": 0.5016465663909913, "memory(GiB)": 70.5, "step": 63155, "token_acc": 0.9, "train_speed(iter/s)": 1.447249 }, { "epoch": 2.705968039072876, "grad_norm": 2.204268455505371, "learning_rate": 4.354736976534145e-05, "loss": 0.42434234619140626, "memory(GiB)": 70.5, "step": 63160, "token_acc": 0.9108910891089109, "train_speed(iter/s)": 1.447251 }, { "epoch": 2.7061822544021252, "grad_norm": 1.8168326616287231, "learning_rate": 4.354069632681423e-05, "loss": 0.6371042251586914, "memory(GiB)": 70.5, "step": 63165, "token_acc": 0.8743016759776536, "train_speed(iter/s)": 1.447275 }, { "epoch": 2.706396469731374, "grad_norm": 4.034434795379639, "learning_rate": 4.353402300530336e-05, "loss": 0.4516075611114502, "memory(GiB)": 70.5, "step": 63170, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.447272 }, { "epoch": 2.706610685060623, "grad_norm": 3.790156602859497, "learning_rate": 4.352734980092973e-05, "loss": 0.5216668605804443, "memory(GiB)": 70.5, "step": 63175, "token_acc": 0.9011627906976745, "train_speed(iter/s)": 1.447285 }, { "epoch": 2.706824900389872, "grad_norm": 4.36681604385376, "learning_rate": 4.352067671381422e-05, "loss": 0.5129871845245362, "memory(GiB)": 70.5, "step": 63180, "token_acc": 0.9049295774647887, "train_speed(iter/s)": 1.4473 }, { "epoch": 2.707039115719121, "grad_norm": 3.076903820037842, "learning_rate": 4.3514003744077745e-05, "loss": 0.23592181205749513, "memory(GiB)": 70.5, "step": 63185, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.447303 }, { "epoch": 2.7072533310483697, "grad_norm": 6.116319179534912, "learning_rate": 4.350733089184117e-05, "loss": 0.25504231452941895, "memory(GiB)": 70.5, "step": 63190, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.447302 }, { "epoch": 2.707467546377619, "grad_norm": 8.163044929504395, "learning_rate": 4.3500658157225375e-05, "loss": 0.39110240936279295, "memory(GiB)": 70.5, "step": 63195, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.447301 }, { "epoch": 2.707681761706868, "grad_norm": 0.26342299580574036, "learning_rate": 4.3493985540351265e-05, "loss": 0.33121671676635744, "memory(GiB)": 70.5, "step": 63200, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.447302 }, { "epoch": 2.7078959770361166, "grad_norm": 1.835461974143982, "learning_rate": 4.34873130413397e-05, "loss": 0.33591859340667723, "memory(GiB)": 70.5, "step": 63205, "token_acc": 0.9448051948051948, "train_speed(iter/s)": 1.447306 }, { "epoch": 2.708110192365366, "grad_norm": 2.737363338470459, "learning_rate": 4.348064066031159e-05, "loss": 0.47972941398620605, "memory(GiB)": 70.5, "step": 63210, "token_acc": 0.8875, "train_speed(iter/s)": 1.447298 }, { "epoch": 2.7083244076946147, "grad_norm": 4.43945837020874, "learning_rate": 4.3473968397387774e-05, "loss": 0.3023594856262207, "memory(GiB)": 70.5, "step": 63215, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.447287 }, { "epoch": 2.7085386230238635, "grad_norm": 3.2999284267425537, "learning_rate": 4.3467296252689144e-05, "loss": 0.24434220790863037, "memory(GiB)": 70.5, "step": 63220, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.447284 }, { "epoch": 2.7087528383531128, "grad_norm": 1.298545002937317, "learning_rate": 4.3460624226336576e-05, "loss": 0.23886899948120116, "memory(GiB)": 70.5, "step": 63225, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.447288 }, { "epoch": 2.7089670536823616, "grad_norm": 5.507147789001465, "learning_rate": 4.345395231845094e-05, "loss": 0.5268346309661865, "memory(GiB)": 70.5, "step": 63230, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.447293 }, { "epoch": 2.7091812690116104, "grad_norm": 5.395366668701172, "learning_rate": 4.344728052915307e-05, "loss": 0.3226576089859009, "memory(GiB)": 70.5, "step": 63235, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.447285 }, { "epoch": 2.7093954843408596, "grad_norm": 2.7400753498077393, "learning_rate": 4.344060885856387e-05, "loss": 0.2315037727355957, "memory(GiB)": 70.5, "step": 63240, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.447275 }, { "epoch": 2.7096096996701085, "grad_norm": 2.813922643661499, "learning_rate": 4.343393730680421e-05, "loss": 0.2222830295562744, "memory(GiB)": 70.5, "step": 63245, "token_acc": 0.9511278195488722, "train_speed(iter/s)": 1.447282 }, { "epoch": 2.7098239149993573, "grad_norm": 5.465986251831055, "learning_rate": 4.3427265873994935e-05, "loss": 0.635386610031128, "memory(GiB)": 70.5, "step": 63250, "token_acc": 0.8623481781376519, "train_speed(iter/s)": 1.447279 }, { "epoch": 2.7100381303286065, "grad_norm": 0.40652868151664734, "learning_rate": 4.342059456025689e-05, "loss": 0.22955875396728515, "memory(GiB)": 70.5, "step": 63255, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.447279 }, { "epoch": 2.7102523456578553, "grad_norm": 3.083129644393921, "learning_rate": 4.341392336571096e-05, "loss": 0.4051100730895996, "memory(GiB)": 70.5, "step": 63260, "token_acc": 0.8875, "train_speed(iter/s)": 1.447294 }, { "epoch": 2.710466560987104, "grad_norm": 6.798651695251465, "learning_rate": 4.340725229047797e-05, "loss": 0.7617060661315918, "memory(GiB)": 70.5, "step": 63265, "token_acc": 0.8565891472868217, "train_speed(iter/s)": 1.447306 }, { "epoch": 2.7106807763163534, "grad_norm": 2.3345556259155273, "learning_rate": 4.3400581334678805e-05, "loss": 0.49373736381530764, "memory(GiB)": 70.5, "step": 63270, "token_acc": 0.8980392156862745, "train_speed(iter/s)": 1.44731 }, { "epoch": 2.710894991645602, "grad_norm": 4.784095287322998, "learning_rate": 4.33939104984343e-05, "loss": 0.5778711318969727, "memory(GiB)": 70.5, "step": 63275, "token_acc": 0.8870292887029289, "train_speed(iter/s)": 1.447319 }, { "epoch": 2.711109206974851, "grad_norm": 10.476367950439453, "learning_rate": 4.338723978186529e-05, "loss": 0.3944443464279175, "memory(GiB)": 70.5, "step": 63280, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.447319 }, { "epoch": 2.7113234223041003, "grad_norm": 3.3069252967834473, "learning_rate": 4.338056918509265e-05, "loss": 0.2422435998916626, "memory(GiB)": 70.5, "step": 63285, "token_acc": 0.9608540925266904, "train_speed(iter/s)": 1.447324 }, { "epoch": 2.711537637633349, "grad_norm": 4.561628818511963, "learning_rate": 4.33738987082372e-05, "loss": 0.36191110610961913, "memory(GiB)": 70.5, "step": 63290, "token_acc": 0.9025270758122743, "train_speed(iter/s)": 1.447322 }, { "epoch": 2.711751852962598, "grad_norm": 4.774181842803955, "learning_rate": 4.336722835141979e-05, "loss": 0.7002363204956055, "memory(GiB)": 70.5, "step": 63295, "token_acc": 0.871875, "train_speed(iter/s)": 1.447329 }, { "epoch": 2.711966068291847, "grad_norm": 2.9743661880493164, "learning_rate": 4.3360558114761266e-05, "loss": 0.37065820693969725, "memory(GiB)": 70.5, "step": 63300, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.44733 }, { "epoch": 2.712180283621096, "grad_norm": 4.3011064529418945, "learning_rate": 4.3353887998382444e-05, "loss": 0.39388892650604246, "memory(GiB)": 70.5, "step": 63305, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.447331 }, { "epoch": 2.712394498950345, "grad_norm": 2.430797815322876, "learning_rate": 4.334721800240418e-05, "loss": 0.31942176818847656, "memory(GiB)": 70.5, "step": 63310, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.44735 }, { "epoch": 2.712608714279594, "grad_norm": 4.454470634460449, "learning_rate": 4.3340548126947316e-05, "loss": 0.33848023414611816, "memory(GiB)": 70.5, "step": 63315, "token_acc": 0.9313304721030042, "train_speed(iter/s)": 1.447356 }, { "epoch": 2.712822929608843, "grad_norm": 3.037752866744995, "learning_rate": 4.333387837213267e-05, "loss": 0.43710908889770506, "memory(GiB)": 70.5, "step": 63320, "token_acc": 0.8967551622418879, "train_speed(iter/s)": 1.447362 }, { "epoch": 2.7130371449380917, "grad_norm": 5.213731288909912, "learning_rate": 4.332720873808106e-05, "loss": 0.47892584800720217, "memory(GiB)": 70.5, "step": 63325, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.447368 }, { "epoch": 2.713251360267341, "grad_norm": 3.9847629070281982, "learning_rate": 4.332053922491333e-05, "loss": 0.484360933303833, "memory(GiB)": 70.5, "step": 63330, "token_acc": 0.8921568627450981, "train_speed(iter/s)": 1.447374 }, { "epoch": 2.7134655755965897, "grad_norm": 0.15512599050998688, "learning_rate": 4.33138698327503e-05, "loss": 0.13870052099227906, "memory(GiB)": 70.5, "step": 63335, "token_acc": 0.9751552795031055, "train_speed(iter/s)": 1.447376 }, { "epoch": 2.7136797909258386, "grad_norm": 2.7393388748168945, "learning_rate": 4.3307200561712777e-05, "loss": 0.2813058614730835, "memory(GiB)": 70.5, "step": 63340, "token_acc": 0.91875, "train_speed(iter/s)": 1.447375 }, { "epoch": 2.713894006255088, "grad_norm": 4.162505149841309, "learning_rate": 4.330053141192161e-05, "loss": 0.3164257049560547, "memory(GiB)": 70.5, "step": 63345, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.447374 }, { "epoch": 2.7141082215843366, "grad_norm": 7.244020462036133, "learning_rate": 4.3293862383497594e-05, "loss": 0.5244711875915528, "memory(GiB)": 70.5, "step": 63350, "token_acc": 0.8837920489296636, "train_speed(iter/s)": 1.447377 }, { "epoch": 2.7143224369135854, "grad_norm": 1.4966429471969604, "learning_rate": 4.3287193476561555e-05, "loss": 0.3611196994781494, "memory(GiB)": 70.5, "step": 63355, "token_acc": 0.9190031152647975, "train_speed(iter/s)": 1.447374 }, { "epoch": 2.7145366522428347, "grad_norm": 0.21932768821716309, "learning_rate": 4.32805246912343e-05, "loss": 0.1652053713798523, "memory(GiB)": 70.5, "step": 63360, "token_acc": 0.9554317548746518, "train_speed(iter/s)": 1.447377 }, { "epoch": 2.7147508675720835, "grad_norm": 2.1013407707214355, "learning_rate": 4.327385602763664e-05, "loss": 0.39332821369171145, "memory(GiB)": 70.5, "step": 63365, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.447376 }, { "epoch": 2.7149650829013323, "grad_norm": 3.783369302749634, "learning_rate": 4.32671874858894e-05, "loss": 0.3342350244522095, "memory(GiB)": 70.5, "step": 63370, "token_acc": 0.9348534201954397, "train_speed(iter/s)": 1.447383 }, { "epoch": 2.7151792982305816, "grad_norm": 3.9622983932495117, "learning_rate": 4.326051906611337e-05, "loss": 0.4542851448059082, "memory(GiB)": 70.5, "step": 63375, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.447396 }, { "epoch": 2.7153935135598304, "grad_norm": 0.9275841116905212, "learning_rate": 4.325385076842934e-05, "loss": 0.15359079837799072, "memory(GiB)": 70.5, "step": 63380, "token_acc": 0.9689922480620154, "train_speed(iter/s)": 1.4474 }, { "epoch": 2.715607728889079, "grad_norm": 2.1068787574768066, "learning_rate": 4.3247182592958136e-05, "loss": 0.5807925224304199, "memory(GiB)": 70.5, "step": 63385, "token_acc": 0.9015873015873016, "train_speed(iter/s)": 1.447417 }, { "epoch": 2.7158219442183285, "grad_norm": 3.070470094680786, "learning_rate": 4.3240514539820574e-05, "loss": 0.33278689384460447, "memory(GiB)": 70.5, "step": 63390, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.447417 }, { "epoch": 2.7160361595475773, "grad_norm": 7.610358238220215, "learning_rate": 4.323384660913743e-05, "loss": 0.538487720489502, "memory(GiB)": 70.5, "step": 63395, "token_acc": 0.9149659863945578, "train_speed(iter/s)": 1.447423 }, { "epoch": 2.716250374876826, "grad_norm": 4.04231071472168, "learning_rate": 4.322717880102949e-05, "loss": 0.4324038505554199, "memory(GiB)": 70.5, "step": 63400, "token_acc": 0.9053497942386831, "train_speed(iter/s)": 1.447422 }, { "epoch": 2.7164645902060753, "grad_norm": 3.5368432998657227, "learning_rate": 4.3220511115617565e-05, "loss": 0.3954612731933594, "memory(GiB)": 70.5, "step": 63405, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.44742 }, { "epoch": 2.716678805535324, "grad_norm": 2.523486614227295, "learning_rate": 4.321384355302244e-05, "loss": 0.42231502532958987, "memory(GiB)": 70.5, "step": 63410, "token_acc": 0.889763779527559, "train_speed(iter/s)": 1.447413 }, { "epoch": 2.716893020864573, "grad_norm": 4.687292098999023, "learning_rate": 4.320717611336491e-05, "loss": 0.6610712051391602, "memory(GiB)": 70.5, "step": 63415, "token_acc": 0.8703071672354948, "train_speed(iter/s)": 1.447426 }, { "epoch": 2.717107236193822, "grad_norm": 2.514007806777954, "learning_rate": 4.320050879676575e-05, "loss": 0.09602398276329041, "memory(GiB)": 70.5, "step": 63420, "token_acc": 0.9754098360655737, "train_speed(iter/s)": 1.447433 }, { "epoch": 2.717321451523071, "grad_norm": 4.149065017700195, "learning_rate": 4.3193841603345755e-05, "loss": 0.3664699077606201, "memory(GiB)": 70.5, "step": 63425, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.447437 }, { "epoch": 2.71753566685232, "grad_norm": 2.073883295059204, "learning_rate": 4.318717453322571e-05, "loss": 0.2986651659011841, "memory(GiB)": 70.5, "step": 63430, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.447455 }, { "epoch": 2.717749882181569, "grad_norm": 3.9252090454101562, "learning_rate": 4.318050758652638e-05, "loss": 0.3984020233154297, "memory(GiB)": 70.5, "step": 63435, "token_acc": 0.926984126984127, "train_speed(iter/s)": 1.447451 }, { "epoch": 2.717964097510818, "grad_norm": 4.237318515777588, "learning_rate": 4.317384076336855e-05, "loss": 0.47126474380493166, "memory(GiB)": 70.5, "step": 63440, "token_acc": 0.9112627986348123, "train_speed(iter/s)": 1.447451 }, { "epoch": 2.7181783128400667, "grad_norm": 7.401182174682617, "learning_rate": 4.3167174063873004e-05, "loss": 0.40613422393798826, "memory(GiB)": 70.5, "step": 63445, "token_acc": 0.9186991869918699, "train_speed(iter/s)": 1.447445 }, { "epoch": 2.718392528169316, "grad_norm": 0.8761941194534302, "learning_rate": 4.3160507488160504e-05, "loss": 0.5329716205596924, "memory(GiB)": 70.5, "step": 63450, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.44745 }, { "epoch": 2.718606743498565, "grad_norm": 1.4616321325302124, "learning_rate": 4.3153841036351814e-05, "loss": 0.5584182739257812, "memory(GiB)": 70.5, "step": 63455, "token_acc": 0.8731343283582089, "train_speed(iter/s)": 1.447458 }, { "epoch": 2.7188209588278136, "grad_norm": 1.3787574768066406, "learning_rate": 4.314717470856774e-05, "loss": 0.19487109184265136, "memory(GiB)": 70.5, "step": 63460, "token_acc": 0.9608433734939759, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.719035174157063, "grad_norm": 2.733083963394165, "learning_rate": 4.314050850492902e-05, "loss": 0.6218723297119141, "memory(GiB)": 70.5, "step": 63465, "token_acc": 0.874251497005988, "train_speed(iter/s)": 1.447482 }, { "epoch": 2.7192493894863117, "grad_norm": 2.867466926574707, "learning_rate": 4.313384242555641e-05, "loss": 0.39052605628967285, "memory(GiB)": 70.5, "step": 63470, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.447479 }, { "epoch": 2.7194636048155605, "grad_norm": 6.095458507537842, "learning_rate": 4.31271764705707e-05, "loss": 0.5973820686340332, "memory(GiB)": 70.5, "step": 63475, "token_acc": 0.8451178451178452, "train_speed(iter/s)": 1.447487 }, { "epoch": 2.7196778201448097, "grad_norm": 3.061824083328247, "learning_rate": 4.3120510640092636e-05, "loss": 0.2771646976470947, "memory(GiB)": 70.5, "step": 63480, "token_acc": 0.9348659003831418, "train_speed(iter/s)": 1.44749 }, { "epoch": 2.7198920354740586, "grad_norm": 3.168989658355713, "learning_rate": 4.311384493424297e-05, "loss": 0.3605653285980225, "memory(GiB)": 70.5, "step": 63485, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.4475 }, { "epoch": 2.7201062508033074, "grad_norm": 2.567735433578491, "learning_rate": 4.310717935314247e-05, "loss": 0.27093617916107177, "memory(GiB)": 70.5, "step": 63490, "token_acc": 0.9307479224376731, "train_speed(iter/s)": 1.447497 }, { "epoch": 2.7203204661325566, "grad_norm": 3.240954875946045, "learning_rate": 4.3100513896911865e-05, "loss": 0.25567576885223386, "memory(GiB)": 70.5, "step": 63495, "token_acc": 0.9547038327526133, "train_speed(iter/s)": 1.447489 }, { "epoch": 2.7205346814618054, "grad_norm": 2.4474005699157715, "learning_rate": 4.309384856567194e-05, "loss": 0.5246871948242188, "memory(GiB)": 70.5, "step": 63500, "token_acc": 0.9194444444444444, "train_speed(iter/s)": 1.447491 }, { "epoch": 2.7205346814618054, "eval_loss": 2.436755418777466, "eval_runtime": 13.8731, "eval_samples_per_second": 7.208, "eval_steps_per_second": 7.208, "eval_token_acc": 0.43543956043956045, "step": 63500 }, { "epoch": 2.7207488967910542, "grad_norm": 2.254228115081787, "learning_rate": 4.3087183359543426e-05, "loss": 0.6216618537902832, "memory(GiB)": 70.5, "step": 63505, "token_acc": 0.557013118062563, "train_speed(iter/s)": 1.447013 }, { "epoch": 2.7209631121203035, "grad_norm": 6.9640302658081055, "learning_rate": 4.308051827864705e-05, "loss": 0.4501436710357666, "memory(GiB)": 70.5, "step": 63510, "token_acc": 0.9263565891472868, "train_speed(iter/s)": 1.447012 }, { "epoch": 2.7211773274495523, "grad_norm": 0.6108056306838989, "learning_rate": 4.3073853323103604e-05, "loss": 0.3076782703399658, "memory(GiB)": 70.5, "step": 63515, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.447028 }, { "epoch": 2.721391542778801, "grad_norm": 2.232466697692871, "learning_rate": 4.3067188493033796e-05, "loss": 0.5162205696105957, "memory(GiB)": 70.5, "step": 63520, "token_acc": 0.890282131661442, "train_speed(iter/s)": 1.447046 }, { "epoch": 2.7216057581080504, "grad_norm": 3.5204226970672607, "learning_rate": 4.3060523788558355e-05, "loss": 0.28899593353271485, "memory(GiB)": 70.5, "step": 63525, "token_acc": 0.9373134328358209, "train_speed(iter/s)": 1.44705 }, { "epoch": 2.721819973437299, "grad_norm": 3.48346209526062, "learning_rate": 4.3053859209798025e-05, "loss": 0.3979550838470459, "memory(GiB)": 70.5, "step": 63530, "token_acc": 0.9169675090252708, "train_speed(iter/s)": 1.447049 }, { "epoch": 2.722034188766548, "grad_norm": 5.722189426422119, "learning_rate": 4.3047194756873575e-05, "loss": 0.3066288471221924, "memory(GiB)": 70.5, "step": 63535, "token_acc": 0.9288389513108615, "train_speed(iter/s)": 1.447042 }, { "epoch": 2.7222484040957973, "grad_norm": 3.939880847930908, "learning_rate": 4.3040530429905715e-05, "loss": 0.5340934753417969, "memory(GiB)": 70.5, "step": 63540, "token_acc": 0.8745098039215686, "train_speed(iter/s)": 1.447045 }, { "epoch": 2.722462619425046, "grad_norm": 2.9286043643951416, "learning_rate": 4.3033866229015166e-05, "loss": 0.30812132358551025, "memory(GiB)": 70.5, "step": 63545, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.447038 }, { "epoch": 2.722676834754295, "grad_norm": 3.466496706008911, "learning_rate": 4.302720215432268e-05, "loss": 0.5473494529724121, "memory(GiB)": 70.5, "step": 63550, "token_acc": 0.903125, "train_speed(iter/s)": 1.447042 }, { "epoch": 2.722891050083544, "grad_norm": 4.540406227111816, "learning_rate": 4.302053820594895e-05, "loss": 0.34674811363220215, "memory(GiB)": 70.5, "step": 63555, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.447045 }, { "epoch": 2.723105265412793, "grad_norm": 3.987926959991455, "learning_rate": 4.301387438401473e-05, "loss": 0.5371238231658936, "memory(GiB)": 70.5, "step": 63560, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.447047 }, { "epoch": 2.7233194807420418, "grad_norm": 2.1184909343719482, "learning_rate": 4.300721068864073e-05, "loss": 0.3658773183822632, "memory(GiB)": 70.5, "step": 63565, "token_acc": 0.9123287671232877, "train_speed(iter/s)": 1.44706 }, { "epoch": 2.723533696071291, "grad_norm": 1.4582535028457642, "learning_rate": 4.3000547119947656e-05, "loss": 0.2094902753829956, "memory(GiB)": 70.5, "step": 63570, "token_acc": 0.9419354838709677, "train_speed(iter/s)": 1.447059 }, { "epoch": 2.72374791140054, "grad_norm": 1.813159465789795, "learning_rate": 4.2993883678056246e-05, "loss": 0.37538168430328367, "memory(GiB)": 70.5, "step": 63575, "token_acc": 0.9002770083102493, "train_speed(iter/s)": 1.44706 }, { "epoch": 2.7239621267297887, "grad_norm": 0.5982832312583923, "learning_rate": 4.298722036308721e-05, "loss": 0.19820359945297242, "memory(GiB)": 70.5, "step": 63580, "token_acc": 0.9618320610687023, "train_speed(iter/s)": 1.447068 }, { "epoch": 2.724176342059038, "grad_norm": 4.246555805206299, "learning_rate": 4.298055717516124e-05, "loss": 0.25366127490997314, "memory(GiB)": 70.5, "step": 63585, "token_acc": 0.960431654676259, "train_speed(iter/s)": 1.447068 }, { "epoch": 2.7243905573882867, "grad_norm": 1.6391431093215942, "learning_rate": 4.297389411439908e-05, "loss": 0.15423260927200316, "memory(GiB)": 70.5, "step": 63590, "token_acc": 0.9669669669669669, "train_speed(iter/s)": 1.447071 }, { "epoch": 2.7246047727175355, "grad_norm": 6.118751525878906, "learning_rate": 4.2967231180921395e-05, "loss": 0.32827277183532716, "memory(GiB)": 70.5, "step": 63595, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.447083 }, { "epoch": 2.724818988046785, "grad_norm": 4.028990268707275, "learning_rate": 4.296056837484894e-05, "loss": 0.26759800910949705, "memory(GiB)": 70.5, "step": 63600, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.447083 }, { "epoch": 2.7250332033760336, "grad_norm": 3.1350364685058594, "learning_rate": 4.295390569630236e-05, "loss": 0.3870536327362061, "memory(GiB)": 70.5, "step": 63605, "token_acc": 0.89375, "train_speed(iter/s)": 1.447082 }, { "epoch": 2.7252474187052824, "grad_norm": 3.611405372619629, "learning_rate": 4.294724314540241e-05, "loss": 0.18212423324584961, "memory(GiB)": 70.5, "step": 63610, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.447078 }, { "epoch": 2.7254616340345317, "grad_norm": 2.0281624794006348, "learning_rate": 4.294058072226976e-05, "loss": 0.139784574508667, "memory(GiB)": 70.5, "step": 63615, "token_acc": 0.9606741573033708, "train_speed(iter/s)": 1.447075 }, { "epoch": 2.7256758493637805, "grad_norm": 1.9775567054748535, "learning_rate": 4.293391842702513e-05, "loss": 0.35311226844787597, "memory(GiB)": 70.5, "step": 63620, "token_acc": 0.9239543726235742, "train_speed(iter/s)": 1.447083 }, { "epoch": 2.7258900646930293, "grad_norm": 0.4522967040538788, "learning_rate": 4.2927256259789184e-05, "loss": 0.5474803447723389, "memory(GiB)": 70.5, "step": 63625, "token_acc": 0.8938356164383562, "train_speed(iter/s)": 1.447108 }, { "epoch": 2.7261042800222786, "grad_norm": 0.4936099946498871, "learning_rate": 4.292059422068262e-05, "loss": 0.16295785903930665, "memory(GiB)": 70.5, "step": 63630, "token_acc": 0.967391304347826, "train_speed(iter/s)": 1.447112 }, { "epoch": 2.7263184953515274, "grad_norm": 3.1984105110168457, "learning_rate": 4.291393230982614e-05, "loss": 0.3867784023284912, "memory(GiB)": 70.5, "step": 63635, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.447113 }, { "epoch": 2.726532710680776, "grad_norm": 2.702345609664917, "learning_rate": 4.290727052734042e-05, "loss": 0.6766353130340577, "memory(GiB)": 70.5, "step": 63640, "token_acc": 0.861198738170347, "train_speed(iter/s)": 1.447114 }, { "epoch": 2.7267469260100254, "grad_norm": 2.078977584838867, "learning_rate": 4.290060887334616e-05, "loss": 0.2396895170211792, "memory(GiB)": 70.5, "step": 63645, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.447118 }, { "epoch": 2.7269611413392743, "grad_norm": 4.204092025756836, "learning_rate": 4.289394734796402e-05, "loss": 0.4931086540222168, "memory(GiB)": 70.5, "step": 63650, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.447127 }, { "epoch": 2.727175356668523, "grad_norm": 5.024911403656006, "learning_rate": 4.288728595131469e-05, "loss": 0.3871969938278198, "memory(GiB)": 70.5, "step": 63655, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.447142 }, { "epoch": 2.7273895719977723, "grad_norm": 3.2780964374542236, "learning_rate": 4.2880624683518844e-05, "loss": 0.3431878566741943, "memory(GiB)": 70.5, "step": 63660, "token_acc": 0.9138461538461539, "train_speed(iter/s)": 1.447147 }, { "epoch": 2.727603787327021, "grad_norm": 4.205585956573486, "learning_rate": 4.287396354469717e-05, "loss": 0.4683064460754395, "memory(GiB)": 70.5, "step": 63665, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.447159 }, { "epoch": 2.72781800265627, "grad_norm": 3.8435750007629395, "learning_rate": 4.286730253497032e-05, "loss": 0.3377640962600708, "memory(GiB)": 70.5, "step": 63670, "token_acc": 0.9307228915662651, "train_speed(iter/s)": 1.447163 }, { "epoch": 2.728032217985519, "grad_norm": 3.1966934204101562, "learning_rate": 4.286064165445896e-05, "loss": 0.25565409660339355, "memory(GiB)": 70.5, "step": 63675, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.447169 }, { "epoch": 2.728246433314768, "grad_norm": 4.5644917488098145, "learning_rate": 4.28539809032838e-05, "loss": 0.4781313896179199, "memory(GiB)": 70.5, "step": 63680, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.447186 }, { "epoch": 2.728460648644017, "grad_norm": 3.819481372833252, "learning_rate": 4.284732028156548e-05, "loss": 0.26508274078369143, "memory(GiB)": 70.5, "step": 63685, "token_acc": 0.9578544061302682, "train_speed(iter/s)": 1.447202 }, { "epoch": 2.728674863973266, "grad_norm": 4.339139461517334, "learning_rate": 4.284065978942465e-05, "loss": 0.42593965530395506, "memory(GiB)": 70.5, "step": 63690, "token_acc": 0.9202898550724637, "train_speed(iter/s)": 1.447201 }, { "epoch": 2.728889079302515, "grad_norm": 3.950843334197998, "learning_rate": 4.2833999426982e-05, "loss": 0.35227968692779543, "memory(GiB)": 70.5, "step": 63695, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.447201 }, { "epoch": 2.7291032946317637, "grad_norm": 3.7715885639190674, "learning_rate": 4.282733919435815e-05, "loss": 0.4615791320800781, "memory(GiB)": 70.5, "step": 63700, "token_acc": 0.8878205128205128, "train_speed(iter/s)": 1.447219 }, { "epoch": 2.729317509961013, "grad_norm": 3.6411731243133545, "learning_rate": 4.28206790916738e-05, "loss": 0.4126612186431885, "memory(GiB)": 70.5, "step": 63705, "token_acc": 0.8929765886287625, "train_speed(iter/s)": 1.447209 }, { "epoch": 2.729531725290262, "grad_norm": 3.9073657989501953, "learning_rate": 4.281401911904958e-05, "loss": 0.3841430187225342, "memory(GiB)": 70.5, "step": 63710, "token_acc": 0.924187725631769, "train_speed(iter/s)": 1.447211 }, { "epoch": 2.7297459406195106, "grad_norm": 2.7440602779388428, "learning_rate": 4.280735927660613e-05, "loss": 0.5026887416839599, "memory(GiB)": 70.5, "step": 63715, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.447238 }, { "epoch": 2.72996015594876, "grad_norm": 2.465942859649658, "learning_rate": 4.2800699564464134e-05, "loss": 0.15744903087615966, "memory(GiB)": 70.5, "step": 63720, "token_acc": 0.9630996309963099, "train_speed(iter/s)": 1.447248 }, { "epoch": 2.7301743712780087, "grad_norm": 3.316317319869995, "learning_rate": 4.279403998274421e-05, "loss": 0.4373174667358398, "memory(GiB)": 70.5, "step": 63725, "token_acc": 0.8872727272727273, "train_speed(iter/s)": 1.447256 }, { "epoch": 2.7303885866072575, "grad_norm": 2.9077565670013428, "learning_rate": 4.2787380531567e-05, "loss": 0.3251603603363037, "memory(GiB)": 70.5, "step": 63730, "token_acc": 0.9228395061728395, "train_speed(iter/s)": 1.447259 }, { "epoch": 2.7306028019365067, "grad_norm": 0.6026391386985779, "learning_rate": 4.278072121105318e-05, "loss": 0.40250697135925295, "memory(GiB)": 70.5, "step": 63735, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.447264 }, { "epoch": 2.7308170172657555, "grad_norm": 2.3489692211151123, "learning_rate": 4.277406202132335e-05, "loss": 0.6065227031707764, "memory(GiB)": 70.5, "step": 63740, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.447268 }, { "epoch": 2.7310312325950044, "grad_norm": 4.983358383178711, "learning_rate": 4.276740296249817e-05, "loss": 0.40277557373046874, "memory(GiB)": 70.5, "step": 63745, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.447291 }, { "epoch": 2.7312454479242536, "grad_norm": 3.7504494190216064, "learning_rate": 4.276074403469825e-05, "loss": 0.4164095401763916, "memory(GiB)": 70.5, "step": 63750, "token_acc": 0.91796875, "train_speed(iter/s)": 1.447295 }, { "epoch": 2.7314596632535024, "grad_norm": 0.3652341663837433, "learning_rate": 4.275408523804427e-05, "loss": 0.22440226078033448, "memory(GiB)": 70.5, "step": 63755, "token_acc": 0.9505703422053232, "train_speed(iter/s)": 1.447291 }, { "epoch": 2.7316738785827512, "grad_norm": 1.586742877960205, "learning_rate": 4.274742657265682e-05, "loss": 0.22504866123199463, "memory(GiB)": 70.5, "step": 63760, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.447291 }, { "epoch": 2.7318880939120005, "grad_norm": 1.4565062522888184, "learning_rate": 4.2740768038656546e-05, "loss": 0.3017827749252319, "memory(GiB)": 70.5, "step": 63765, "token_acc": 0.9301310043668122, "train_speed(iter/s)": 1.447288 }, { "epoch": 2.7321023092412493, "grad_norm": 2.194143056869507, "learning_rate": 4.2734109636164074e-05, "loss": 0.21719827651977539, "memory(GiB)": 70.5, "step": 63770, "token_acc": 0.9442724458204335, "train_speed(iter/s)": 1.447299 }, { "epoch": 2.732316524570498, "grad_norm": 3.7003014087677, "learning_rate": 4.2727451365300014e-05, "loss": 0.3753760099411011, "memory(GiB)": 70.5, "step": 63775, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.447303 }, { "epoch": 2.7325307398997474, "grad_norm": 0.8799652457237244, "learning_rate": 4.272079322618501e-05, "loss": 0.5823846340179444, "memory(GiB)": 70.5, "step": 63780, "token_acc": 0.8931034482758621, "train_speed(iter/s)": 1.447317 }, { "epoch": 2.732744955228996, "grad_norm": 3.988002300262451, "learning_rate": 4.271413521893965e-05, "loss": 0.335337495803833, "memory(GiB)": 70.5, "step": 63785, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.447324 }, { "epoch": 2.732959170558245, "grad_norm": 3.2943079471588135, "learning_rate": 4.270747734368457e-05, "loss": 0.32047698497772215, "memory(GiB)": 70.5, "step": 63790, "token_acc": 0.9233576642335767, "train_speed(iter/s)": 1.447328 }, { "epoch": 2.7331733858874943, "grad_norm": 4.8098602294921875, "learning_rate": 4.270081960054038e-05, "loss": 0.2359396457672119, "memory(GiB)": 70.5, "step": 63795, "token_acc": 0.9536231884057971, "train_speed(iter/s)": 1.44733 }, { "epoch": 2.733387601216743, "grad_norm": 1.5428937673568726, "learning_rate": 4.269416198962768e-05, "loss": 0.24677281379699706, "memory(GiB)": 70.5, "step": 63800, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.447319 }, { "epoch": 2.733601816545992, "grad_norm": 0.0616576224565506, "learning_rate": 4.26875045110671e-05, "loss": 0.25650749206542967, "memory(GiB)": 70.5, "step": 63805, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.44732 }, { "epoch": 2.733816031875241, "grad_norm": 6.584847927093506, "learning_rate": 4.268084716497924e-05, "loss": 0.3341339111328125, "memory(GiB)": 70.5, "step": 63810, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.44734 }, { "epoch": 2.73403024720449, "grad_norm": 2.52010440826416, "learning_rate": 4.267418995148468e-05, "loss": 0.4328632354736328, "memory(GiB)": 70.5, "step": 63815, "token_acc": 0.9141104294478528, "train_speed(iter/s)": 1.447355 }, { "epoch": 2.7342444625337388, "grad_norm": 1.9377225637435913, "learning_rate": 4.266753287070406e-05, "loss": 0.3990891695022583, "memory(GiB)": 70.5, "step": 63820, "token_acc": 0.897887323943662, "train_speed(iter/s)": 1.447361 }, { "epoch": 2.734458677862988, "grad_norm": 1.9031126499176025, "learning_rate": 4.266087592275794e-05, "loss": 0.48194184303283694, "memory(GiB)": 70.5, "step": 63825, "token_acc": 0.8978328173374613, "train_speed(iter/s)": 1.44736 }, { "epoch": 2.734672893192237, "grad_norm": 2.664780616760254, "learning_rate": 4.265421910776694e-05, "loss": 0.25226213932037356, "memory(GiB)": 70.5, "step": 63830, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.447361 }, { "epoch": 2.7348871085214856, "grad_norm": 0.3105071783065796, "learning_rate": 4.2647562425851666e-05, "loss": 0.2549535512924194, "memory(GiB)": 70.5, "step": 63835, "token_acc": 0.943089430894309, "train_speed(iter/s)": 1.447357 }, { "epoch": 2.735101323850735, "grad_norm": 1.2024283409118652, "learning_rate": 4.26409058771327e-05, "loss": 0.3383232831954956, "memory(GiB)": 70.5, "step": 63840, "token_acc": 0.9370860927152318, "train_speed(iter/s)": 1.447362 }, { "epoch": 2.7353155391799837, "grad_norm": 1.7536181211471558, "learning_rate": 4.2634249461730616e-05, "loss": 0.16095118522644042, "memory(GiB)": 70.5, "step": 63845, "token_acc": 0.9679715302491103, "train_speed(iter/s)": 1.447366 }, { "epoch": 2.7355297545092325, "grad_norm": 1.2627317905426025, "learning_rate": 4.262759317976602e-05, "loss": 0.21236779689788818, "memory(GiB)": 70.5, "step": 63850, "token_acc": 0.9533333333333334, "train_speed(iter/s)": 1.447366 }, { "epoch": 2.735743969838482, "grad_norm": 0.07331901043653488, "learning_rate": 4.262093703135949e-05, "loss": 0.5469540119171142, "memory(GiB)": 70.5, "step": 63855, "token_acc": 0.8761609907120743, "train_speed(iter/s)": 1.447369 }, { "epoch": 2.7359581851677306, "grad_norm": 4.700119972229004, "learning_rate": 4.26142810166316e-05, "loss": 0.24162690639495848, "memory(GiB)": 70.5, "step": 63860, "token_acc": 0.9486301369863014, "train_speed(iter/s)": 1.447378 }, { "epoch": 2.7361724004969794, "grad_norm": 5.054917335510254, "learning_rate": 4.260762513570294e-05, "loss": 0.3848599910736084, "memory(GiB)": 70.5, "step": 63865, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.447368 }, { "epoch": 2.7363866158262287, "grad_norm": 3.0197629928588867, "learning_rate": 4.260096938869409e-05, "loss": 0.3273277282714844, "memory(GiB)": 70.5, "step": 63870, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.447366 }, { "epoch": 2.7366008311554775, "grad_norm": 3.503068447113037, "learning_rate": 4.2594313775725616e-05, "loss": 0.3825935125350952, "memory(GiB)": 70.5, "step": 63875, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.447376 }, { "epoch": 2.7368150464847263, "grad_norm": 4.2593488693237305, "learning_rate": 4.25876582969181e-05, "loss": 0.45152816772460935, "memory(GiB)": 70.5, "step": 63880, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.447389 }, { "epoch": 2.7370292618139755, "grad_norm": 1.9009124040603638, "learning_rate": 4.258100295239209e-05, "loss": 0.35864834785461425, "memory(GiB)": 70.5, "step": 63885, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.44741 }, { "epoch": 2.7372434771432244, "grad_norm": 2.5433084964752197, "learning_rate": 4.25743477422682e-05, "loss": 0.2646726131439209, "memory(GiB)": 70.5, "step": 63890, "token_acc": 0.9452887537993921, "train_speed(iter/s)": 1.447421 }, { "epoch": 2.737457692472473, "grad_norm": 1.99435555934906, "learning_rate": 4.2567692666666945e-05, "loss": 0.2526238918304443, "memory(GiB)": 70.5, "step": 63895, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.447421 }, { "epoch": 2.7376719078017224, "grad_norm": 4.662778854370117, "learning_rate": 4.2561037725708904e-05, "loss": 0.13605034351348877, "memory(GiB)": 70.5, "step": 63900, "token_acc": 0.9741379310344828, "train_speed(iter/s)": 1.447426 }, { "epoch": 2.7378861231309712, "grad_norm": 5.414318084716797, "learning_rate": 4.2554382919514645e-05, "loss": 0.4961839199066162, "memory(GiB)": 70.5, "step": 63905, "token_acc": 0.9079365079365079, "train_speed(iter/s)": 1.44743 }, { "epoch": 2.73810033846022, "grad_norm": 0.2597482204437256, "learning_rate": 4.254772824820474e-05, "loss": 0.11105364561080933, "memory(GiB)": 70.5, "step": 63910, "token_acc": 0.975609756097561, "train_speed(iter/s)": 1.447446 }, { "epoch": 2.7383145537894693, "grad_norm": 4.398972988128662, "learning_rate": 4.254107371189973e-05, "loss": 0.23820953369140624, "memory(GiB)": 70.5, "step": 63915, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.447451 }, { "epoch": 2.738528769118718, "grad_norm": 6.118549346923828, "learning_rate": 4.253441931072015e-05, "loss": 0.401507568359375, "memory(GiB)": 70.5, "step": 63920, "token_acc": 0.9093959731543624, "train_speed(iter/s)": 1.44745 }, { "epoch": 2.738742984447967, "grad_norm": 0.10441772639751434, "learning_rate": 4.2527765044786576e-05, "loss": 0.5281307697296143, "memory(GiB)": 70.5, "step": 63925, "token_acc": 0.9164179104477612, "train_speed(iter/s)": 1.447454 }, { "epoch": 2.738957199777216, "grad_norm": 2.123169183731079, "learning_rate": 4.252111091421954e-05, "loss": 0.30581936836242674, "memory(GiB)": 70.5, "step": 63930, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.447455 }, { "epoch": 2.739171415106465, "grad_norm": 1.9047958850860596, "learning_rate": 4.251445691913961e-05, "loss": 0.29887259006500244, "memory(GiB)": 70.5, "step": 63935, "token_acc": 0.9396551724137931, "train_speed(iter/s)": 1.447455 }, { "epoch": 2.739385630435714, "grad_norm": 0.8516576886177063, "learning_rate": 4.250780305966731e-05, "loss": 0.1533438205718994, "memory(GiB)": 70.5, "step": 63940, "token_acc": 0.9688581314878892, "train_speed(iter/s)": 1.447457 }, { "epoch": 2.739599845764963, "grad_norm": 1.502975583076477, "learning_rate": 4.2501149335923176e-05, "loss": 0.16991833448410035, "memory(GiB)": 70.5, "step": 63945, "token_acc": 0.9625850340136054, "train_speed(iter/s)": 1.447458 }, { "epoch": 2.739814061094212, "grad_norm": 2.778970241546631, "learning_rate": 4.2494495748027776e-05, "loss": 0.2557036876678467, "memory(GiB)": 70.5, "step": 63950, "token_acc": 0.9432835820895522, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.7400282764234607, "grad_norm": 5.11224365234375, "learning_rate": 4.2487842296101615e-05, "loss": 0.30615081787109377, "memory(GiB)": 70.5, "step": 63955, "token_acc": 0.9196787148594378, "train_speed(iter/s)": 1.447467 }, { "epoch": 2.74024249175271, "grad_norm": 1.9087446928024292, "learning_rate": 4.248118898026523e-05, "loss": 0.2384951114654541, "memory(GiB)": 70.5, "step": 63960, "token_acc": 0.9375, "train_speed(iter/s)": 1.447467 }, { "epoch": 2.7404567070819588, "grad_norm": 4.852374076843262, "learning_rate": 4.247453580063917e-05, "loss": 0.5609960556030273, "memory(GiB)": 70.5, "step": 63965, "token_acc": 0.8684931506849315, "train_speed(iter/s)": 1.44746 }, { "epoch": 2.7406709224112076, "grad_norm": 3.1022226810455322, "learning_rate": 4.246788275734393e-05, "loss": 0.24230875968933105, "memory(GiB)": 70.5, "step": 63970, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.447476 }, { "epoch": 2.740885137740457, "grad_norm": 4.598024845123291, "learning_rate": 4.2461229850500075e-05, "loss": 0.6560426712036133, "memory(GiB)": 70.5, "step": 63975, "token_acc": 0.8636363636363636, "train_speed(iter/s)": 1.447474 }, { "epoch": 2.7410993530697056, "grad_norm": 2.8321053981781006, "learning_rate": 4.245457708022813e-05, "loss": 0.35940191745758054, "memory(GiB)": 70.5, "step": 63980, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.44747 }, { "epoch": 2.7413135683989545, "grad_norm": 6.34235954284668, "learning_rate": 4.244792444664859e-05, "loss": 0.3224613904953003, "memory(GiB)": 70.5, "step": 63985, "token_acc": 0.9319526627218935, "train_speed(iter/s)": 1.44747 }, { "epoch": 2.7415277837282037, "grad_norm": 3.5585460662841797, "learning_rate": 4.2441271949881975e-05, "loss": 0.3728576421737671, "memory(GiB)": 70.5, "step": 63990, "token_acc": 0.925531914893617, "train_speed(iter/s)": 1.447476 }, { "epoch": 2.7417419990574525, "grad_norm": 3.4094674587249756, "learning_rate": 4.2434619590048817e-05, "loss": 0.2550528049468994, "memory(GiB)": 70.5, "step": 63995, "token_acc": 0.9484126984126984, "train_speed(iter/s)": 1.447492 }, { "epoch": 2.7419562143867013, "grad_norm": 8.502140998840332, "learning_rate": 4.242796736726963e-05, "loss": 0.6555193424224853, "memory(GiB)": 70.5, "step": 64000, "token_acc": 0.8808664259927798, "train_speed(iter/s)": 1.447493 }, { "epoch": 2.7419562143867013, "eval_loss": 2.356330156326294, "eval_runtime": 13.1511, "eval_samples_per_second": 7.604, "eval_steps_per_second": 7.604, "eval_token_acc": 0.4607329842931937, "step": 64000 }, { "epoch": 2.7421704297159506, "grad_norm": 5.234201431274414, "learning_rate": 4.24213152816649e-05, "loss": 0.689316987991333, "memory(GiB)": 70.5, "step": 64005, "token_acc": 0.5663117134559535, "train_speed(iter/s)": 1.447037 }, { "epoch": 2.7423846450451994, "grad_norm": 4.1237969398498535, "learning_rate": 4.241466333335517e-05, "loss": 0.5017897605895996, "memory(GiB)": 70.5, "step": 64010, "token_acc": 0.8932806324110671, "train_speed(iter/s)": 1.447037 }, { "epoch": 2.742598860374448, "grad_norm": 4.889407634735107, "learning_rate": 4.240801152246091e-05, "loss": 0.32562599182128904, "memory(GiB)": 70.5, "step": 64015, "token_acc": 0.909433962264151, "train_speed(iter/s)": 1.447036 }, { "epoch": 2.7428130757036975, "grad_norm": 1.8664137125015259, "learning_rate": 4.2401359849102653e-05, "loss": 0.4960677146911621, "memory(GiB)": 70.5, "step": 64020, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.447037 }, { "epoch": 2.7430272910329463, "grad_norm": 3.156277894973755, "learning_rate": 4.2394708313400894e-05, "loss": 0.6060359954833985, "memory(GiB)": 70.5, "step": 64025, "token_acc": 0.8654353562005277, "train_speed(iter/s)": 1.447034 }, { "epoch": 2.743241506362195, "grad_norm": 5.700826168060303, "learning_rate": 4.238805691547612e-05, "loss": 0.45258069038391113, "memory(GiB)": 70.5, "step": 64030, "token_acc": 0.90625, "train_speed(iter/s)": 1.447041 }, { "epoch": 2.7434557216914444, "grad_norm": 2.147225856781006, "learning_rate": 4.238140565544885e-05, "loss": 0.1516585111618042, "memory(GiB)": 70.5, "step": 64035, "token_acc": 0.9626436781609196, "train_speed(iter/s)": 1.447036 }, { "epoch": 2.743669937020693, "grad_norm": 3.1151797771453857, "learning_rate": 4.237475453343955e-05, "loss": 0.8099445343017578, "memory(GiB)": 70.5, "step": 64040, "token_acc": 0.8454810495626822, "train_speed(iter/s)": 1.447047 }, { "epoch": 2.743884152349942, "grad_norm": 5.937415599822998, "learning_rate": 4.236810354956872e-05, "loss": 0.42780122756958006, "memory(GiB)": 70.5, "step": 64045, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.447047 }, { "epoch": 2.7440983676791912, "grad_norm": 1.790346384048462, "learning_rate": 4.236145270395685e-05, "loss": 0.3047647953033447, "memory(GiB)": 70.5, "step": 64050, "token_acc": 0.9287833827893175, "train_speed(iter/s)": 1.447049 }, { "epoch": 2.74431258300844, "grad_norm": 3.745429039001465, "learning_rate": 4.2354801996724444e-05, "loss": 0.2527316570281982, "memory(GiB)": 70.5, "step": 64055, "token_acc": 0.9498069498069498, "train_speed(iter/s)": 1.447041 }, { "epoch": 2.744526798337689, "grad_norm": 9.209460258483887, "learning_rate": 4.2348151427991974e-05, "loss": 0.29998552799224854, "memory(GiB)": 70.5, "step": 64060, "token_acc": 0.9425981873111783, "train_speed(iter/s)": 1.447055 }, { "epoch": 2.744741013666938, "grad_norm": 2.4422948360443115, "learning_rate": 4.234150099787991e-05, "loss": 0.41063637733459474, "memory(GiB)": 70.5, "step": 64065, "token_acc": 0.8923611111111112, "train_speed(iter/s)": 1.447067 }, { "epoch": 2.744955228996187, "grad_norm": 3.244544267654419, "learning_rate": 4.233485070650874e-05, "loss": 0.23713653087615966, "memory(GiB)": 70.5, "step": 64070, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.447061 }, { "epoch": 2.7451694443254357, "grad_norm": 2.2838523387908936, "learning_rate": 4.2328200553998944e-05, "loss": 0.1583117961883545, "memory(GiB)": 70.5, "step": 64075, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.447074 }, { "epoch": 2.745383659654685, "grad_norm": 1.8464423418045044, "learning_rate": 4.2321550540470996e-05, "loss": 0.28962907791137693, "memory(GiB)": 70.5, "step": 64080, "token_acc": 0.9451219512195121, "train_speed(iter/s)": 1.447074 }, { "epoch": 2.745597874983934, "grad_norm": 4.462709426879883, "learning_rate": 4.231490066604536e-05, "loss": 0.5141269207000733, "memory(GiB)": 70.5, "step": 64085, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.44709 }, { "epoch": 2.7458120903131826, "grad_norm": 3.7157177925109863, "learning_rate": 4.23082509308425e-05, "loss": 0.4260213375091553, "memory(GiB)": 70.5, "step": 64090, "token_acc": 0.8816793893129771, "train_speed(iter/s)": 1.447102 }, { "epoch": 2.746026305642432, "grad_norm": 4.9435882568359375, "learning_rate": 4.23016013349829e-05, "loss": 0.35941166877746583, "memory(GiB)": 70.5, "step": 64095, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.447116 }, { "epoch": 2.7462405209716807, "grad_norm": 4.961340427398682, "learning_rate": 4.229495187858701e-05, "loss": 0.4193395137786865, "memory(GiB)": 70.5, "step": 64100, "token_acc": 0.8923076923076924, "train_speed(iter/s)": 1.447118 }, { "epoch": 2.7464547363009295, "grad_norm": 1.027000904083252, "learning_rate": 4.2288302561775295e-05, "loss": 0.20475525856018068, "memory(GiB)": 70.5, "step": 64105, "token_acc": 0.9571428571428572, "train_speed(iter/s)": 1.447121 }, { "epoch": 2.7466689516301788, "grad_norm": 3.707071304321289, "learning_rate": 4.228165338466821e-05, "loss": 0.3381448745727539, "memory(GiB)": 70.5, "step": 64110, "token_acc": 0.9267605633802817, "train_speed(iter/s)": 1.447124 }, { "epoch": 2.7468831669594276, "grad_norm": 0.3154057562351227, "learning_rate": 4.227500434738622e-05, "loss": 0.43197131156921387, "memory(GiB)": 70.5, "step": 64115, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.447133 }, { "epoch": 2.7470973822886764, "grad_norm": 4.355676174163818, "learning_rate": 4.226835545004975e-05, "loss": 0.6952884197235107, "memory(GiB)": 70.5, "step": 64120, "token_acc": 0.8544061302681992, "train_speed(iter/s)": 1.447129 }, { "epoch": 2.7473115976179256, "grad_norm": 2.641995668411255, "learning_rate": 4.226170669277929e-05, "loss": 0.5974457740783692, "memory(GiB)": 70.5, "step": 64125, "token_acc": 0.8689138576779026, "train_speed(iter/s)": 1.447134 }, { "epoch": 2.7475258129471745, "grad_norm": 2.8492748737335205, "learning_rate": 4.225505807569529e-05, "loss": 0.484539270401001, "memory(GiB)": 70.5, "step": 64130, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.447137 }, { "epoch": 2.7477400282764233, "grad_norm": 2.9282448291778564, "learning_rate": 4.224840959891815e-05, "loss": 0.37242951393127444, "memory(GiB)": 70.5, "step": 64135, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.447134 }, { "epoch": 2.7479542436056725, "grad_norm": 3.0713088512420654, "learning_rate": 4.224176126256836e-05, "loss": 0.3685158729553223, "memory(GiB)": 70.5, "step": 64140, "token_acc": 0.9239766081871345, "train_speed(iter/s)": 1.447135 }, { "epoch": 2.7481684589349213, "grad_norm": 3.255164861679077, "learning_rate": 4.223511306676634e-05, "loss": 0.2650378942489624, "memory(GiB)": 70.5, "step": 64145, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.447142 }, { "epoch": 2.74838267426417, "grad_norm": 2.6378471851348877, "learning_rate": 4.222846501163253e-05, "loss": 0.30159883499145507, "memory(GiB)": 70.5, "step": 64150, "token_acc": 0.9537953795379538, "train_speed(iter/s)": 1.447149 }, { "epoch": 2.7485968895934194, "grad_norm": 2.477562189102173, "learning_rate": 4.222181709728736e-05, "loss": 0.4517836093902588, "memory(GiB)": 70.5, "step": 64155, "token_acc": 0.9007352941176471, "train_speed(iter/s)": 1.447165 }, { "epoch": 2.7488111049226682, "grad_norm": 3.0279853343963623, "learning_rate": 4.221516932385128e-05, "loss": 0.17286239862442015, "memory(GiB)": 70.5, "step": 64160, "token_acc": 0.964516129032258, "train_speed(iter/s)": 1.447163 }, { "epoch": 2.749025320251917, "grad_norm": 0.4168773889541626, "learning_rate": 4.220852169144471e-05, "loss": 0.4140986442565918, "memory(GiB)": 70.5, "step": 64165, "token_acc": 0.9043478260869565, "train_speed(iter/s)": 1.447161 }, { "epoch": 2.7492395355811663, "grad_norm": 3.7749009132385254, "learning_rate": 4.2201874200188074e-05, "loss": 0.25414111614227297, "memory(GiB)": 70.5, "step": 64170, "token_acc": 0.9470198675496688, "train_speed(iter/s)": 1.447158 }, { "epoch": 2.749453750910415, "grad_norm": 0.6360145807266235, "learning_rate": 4.2195226850201796e-05, "loss": 0.2575418949127197, "memory(GiB)": 70.5, "step": 64175, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.44717 }, { "epoch": 2.749667966239664, "grad_norm": 5.229663848876953, "learning_rate": 4.2188579641606316e-05, "loss": 0.5130187511444092, "memory(GiB)": 70.5, "step": 64180, "token_acc": 0.8892405063291139, "train_speed(iter/s)": 1.447169 }, { "epoch": 2.749882181568913, "grad_norm": 1.9810059070587158, "learning_rate": 4.218193257452204e-05, "loss": 0.4143965244293213, "memory(GiB)": 70.5, "step": 64185, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.447175 }, { "epoch": 2.750096396898162, "grad_norm": 1.5099080801010132, "learning_rate": 4.217528564906938e-05, "loss": 0.2960068941116333, "memory(GiB)": 70.5, "step": 64190, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.447179 }, { "epoch": 2.750310612227411, "grad_norm": 3.1403489112854004, "learning_rate": 4.2168638865368756e-05, "loss": 0.6273831844329834, "memory(GiB)": 70.5, "step": 64195, "token_acc": 0.8693181818181818, "train_speed(iter/s)": 1.447189 }, { "epoch": 2.75052482755666, "grad_norm": 3.271702527999878, "learning_rate": 4.2161992223540606e-05, "loss": 0.5167066097259522, "memory(GiB)": 70.5, "step": 64200, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.447204 }, { "epoch": 2.750739042885909, "grad_norm": 4.084289073944092, "learning_rate": 4.215534572370531e-05, "loss": 0.16831663846969605, "memory(GiB)": 70.5, "step": 64205, "token_acc": 0.9728997289972899, "train_speed(iter/s)": 1.447208 }, { "epoch": 2.7509532582151577, "grad_norm": 1.7818299531936646, "learning_rate": 4.214869936598329e-05, "loss": 0.3341160774230957, "memory(GiB)": 70.5, "step": 64210, "token_acc": 0.9505703422053232, "train_speed(iter/s)": 1.447206 }, { "epoch": 2.751167473544407, "grad_norm": 5.160538673400879, "learning_rate": 4.214205315049494e-05, "loss": 0.24617865085601806, "memory(GiB)": 70.5, "step": 64215, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.447206 }, { "epoch": 2.7513816888736558, "grad_norm": 3.3697612285614014, "learning_rate": 4.2135407077360667e-05, "loss": 0.6224346160888672, "memory(GiB)": 70.5, "step": 64220, "token_acc": 0.8554216867469879, "train_speed(iter/s)": 1.447208 }, { "epoch": 2.7515959042029046, "grad_norm": 3.025794744491577, "learning_rate": 4.2128761146700884e-05, "loss": 0.3127118110656738, "memory(GiB)": 70.5, "step": 64225, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.447213 }, { "epoch": 2.751810119532154, "grad_norm": 3.1108360290527344, "learning_rate": 4.2122115358635975e-05, "loss": 0.296408486366272, "memory(GiB)": 70.5, "step": 64230, "token_acc": 0.9407114624505929, "train_speed(iter/s)": 1.44722 }, { "epoch": 2.7520243348614026, "grad_norm": 0.1007022112607956, "learning_rate": 4.2115469713286325e-05, "loss": 0.28120102882385256, "memory(GiB)": 70.5, "step": 64235, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.447227 }, { "epoch": 2.7522385501906514, "grad_norm": 4.331879615783691, "learning_rate": 4.210882421077235e-05, "loss": 0.6418715476989746, "memory(GiB)": 70.5, "step": 64240, "token_acc": 0.8716417910447761, "train_speed(iter/s)": 1.447221 }, { "epoch": 2.7524527655199007, "grad_norm": 8.471258163452148, "learning_rate": 4.210217885121442e-05, "loss": 0.27043609619140624, "memory(GiB)": 70.5, "step": 64245, "token_acc": 0.9339622641509434, "train_speed(iter/s)": 1.447216 }, { "epoch": 2.7526669808491495, "grad_norm": 4.238602161407471, "learning_rate": 4.209553363473293e-05, "loss": 0.3701355218887329, "memory(GiB)": 70.5, "step": 64250, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.44722 }, { "epoch": 2.7528811961783983, "grad_norm": 4.8680291175842285, "learning_rate": 4.208888856144826e-05, "loss": 0.3940586566925049, "memory(GiB)": 70.5, "step": 64255, "token_acc": 0.9198813056379822, "train_speed(iter/s)": 1.447219 }, { "epoch": 2.7530954115076476, "grad_norm": 2.576105833053589, "learning_rate": 4.2082243631480795e-05, "loss": 0.2913111686706543, "memory(GiB)": 70.5, "step": 64260, "token_acc": 0.9438596491228071, "train_speed(iter/s)": 1.447222 }, { "epoch": 2.7533096268368964, "grad_norm": 5.844583988189697, "learning_rate": 4.207559884495092e-05, "loss": 0.4956247329711914, "memory(GiB)": 70.5, "step": 64265, "token_acc": 0.9045643153526971, "train_speed(iter/s)": 1.447226 }, { "epoch": 2.753523842166145, "grad_norm": 2.1192967891693115, "learning_rate": 4.2068954201978985e-05, "loss": 0.26144044399261473, "memory(GiB)": 70.5, "step": 64270, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.447233 }, { "epoch": 2.7537380574953945, "grad_norm": 3.906113624572754, "learning_rate": 4.2062309702685404e-05, "loss": 0.3883247375488281, "memory(GiB)": 70.5, "step": 64275, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.447249 }, { "epoch": 2.7539522728246433, "grad_norm": 2.5599873065948486, "learning_rate": 4.205566534719052e-05, "loss": 0.30041463375091554, "memory(GiB)": 70.5, "step": 64280, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.447247 }, { "epoch": 2.754166488153892, "grad_norm": 3.517768383026123, "learning_rate": 4.204902113561472e-05, "loss": 0.3461506128311157, "memory(GiB)": 70.5, "step": 64285, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.447259 }, { "epoch": 2.7543807034831413, "grad_norm": 1.597086787223816, "learning_rate": 4.204237706807836e-05, "loss": 0.6599215507507324, "memory(GiB)": 70.5, "step": 64290, "token_acc": 0.851963746223565, "train_speed(iter/s)": 1.447276 }, { "epoch": 2.75459491881239, "grad_norm": 2.812886953353882, "learning_rate": 4.20357331447018e-05, "loss": 0.5245871543884277, "memory(GiB)": 70.5, "step": 64295, "token_acc": 0.8994413407821229, "train_speed(iter/s)": 1.447277 }, { "epoch": 2.754809134141639, "grad_norm": 0.40455085039138794, "learning_rate": 4.202908936560541e-05, "loss": 0.42578749656677245, "memory(GiB)": 70.5, "step": 64300, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.447282 }, { "epoch": 2.7550233494708882, "grad_norm": 4.056870937347412, "learning_rate": 4.202244573090954e-05, "loss": 0.31157944202423093, "memory(GiB)": 70.5, "step": 64305, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.447282 }, { "epoch": 2.755237564800137, "grad_norm": 0.20379763841629028, "learning_rate": 4.2015802240734554e-05, "loss": 0.26846387386322024, "memory(GiB)": 70.5, "step": 64310, "token_acc": 0.9493243243243243, "train_speed(iter/s)": 1.447287 }, { "epoch": 2.755451780129386, "grad_norm": 2.1695361137390137, "learning_rate": 4.200915889520079e-05, "loss": 0.28403141498565676, "memory(GiB)": 70.5, "step": 64315, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.447296 }, { "epoch": 2.755665995458635, "grad_norm": 0.8568649291992188, "learning_rate": 4.2002515694428616e-05, "loss": 0.42678537368774416, "memory(GiB)": 70.5, "step": 64320, "token_acc": 0.921161825726141, "train_speed(iter/s)": 1.447301 }, { "epoch": 2.755880210787884, "grad_norm": 2.4218404293060303, "learning_rate": 4.1995872638538375e-05, "loss": 0.3117112636566162, "memory(GiB)": 70.5, "step": 64325, "token_acc": 0.9364161849710982, "train_speed(iter/s)": 1.447308 }, { "epoch": 2.7560944261171327, "grad_norm": 5.570744037628174, "learning_rate": 4.1989229727650416e-05, "loss": 0.49114260673522947, "memory(GiB)": 70.5, "step": 64330, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.447313 }, { "epoch": 2.756308641446382, "grad_norm": 4.874022960662842, "learning_rate": 4.1982586961885054e-05, "loss": 0.40097436904907224, "memory(GiB)": 70.5, "step": 64335, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.44731 }, { "epoch": 2.756522856775631, "grad_norm": 1.8916562795639038, "learning_rate": 4.1975944341362646e-05, "loss": 0.20904288291931153, "memory(GiB)": 70.5, "step": 64340, "token_acc": 0.95, "train_speed(iter/s)": 1.447313 }, { "epoch": 2.7567370721048796, "grad_norm": 3.9541051387786865, "learning_rate": 4.196930186620355e-05, "loss": 0.3687300682067871, "memory(GiB)": 70.5, "step": 64345, "token_acc": 0.9187192118226601, "train_speed(iter/s)": 1.447317 }, { "epoch": 2.756951287434129, "grad_norm": 0.6775537729263306, "learning_rate": 4.196265953652809e-05, "loss": 0.3274896860122681, "memory(GiB)": 70.5, "step": 64350, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.447326 }, { "epoch": 2.7571655027633777, "grad_norm": 8.297306060791016, "learning_rate": 4.195601735245658e-05, "loss": 0.7100034236907959, "memory(GiB)": 70.5, "step": 64355, "token_acc": 0.8550295857988166, "train_speed(iter/s)": 1.447336 }, { "epoch": 2.7573797180926265, "grad_norm": 5.865365982055664, "learning_rate": 4.194937531410937e-05, "loss": 0.37453982830047605, "memory(GiB)": 70.5, "step": 64360, "token_acc": 0.9441176470588235, "train_speed(iter/s)": 1.447363 }, { "epoch": 2.7575939334218758, "grad_norm": 2.12463116645813, "learning_rate": 4.1942733421606764e-05, "loss": 0.43750791549682616, "memory(GiB)": 70.5, "step": 64365, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.447363 }, { "epoch": 2.7578081487511246, "grad_norm": 1.0944751501083374, "learning_rate": 4.193609167506912e-05, "loss": 0.49049787521362304, "memory(GiB)": 70.5, "step": 64370, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.447362 }, { "epoch": 2.7580223640803734, "grad_norm": 2.8608176708221436, "learning_rate": 4.1929450074616734e-05, "loss": 0.471663236618042, "memory(GiB)": 70.5, "step": 64375, "token_acc": 0.8950437317784257, "train_speed(iter/s)": 1.447369 }, { "epoch": 2.7582365794096226, "grad_norm": 0.720837414264679, "learning_rate": 4.192280862036992e-05, "loss": 0.29541356563568116, "memory(GiB)": 70.5, "step": 64380, "token_acc": 0.9449838187702265, "train_speed(iter/s)": 1.447369 }, { "epoch": 2.7584507947388714, "grad_norm": 3.3965132236480713, "learning_rate": 4.1916167312449014e-05, "loss": 0.335845947265625, "memory(GiB)": 70.5, "step": 64385, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.447371 }, { "epoch": 2.7586650100681203, "grad_norm": 4.045267105102539, "learning_rate": 4.1909526150974326e-05, "loss": 0.16934466361999512, "memory(GiB)": 70.5, "step": 64390, "token_acc": 0.9512987012987013, "train_speed(iter/s)": 1.447373 }, { "epoch": 2.7588792253973695, "grad_norm": 2.699507713317871, "learning_rate": 4.190288513606615e-05, "loss": 0.13958394527435303, "memory(GiB)": 70.5, "step": 64395, "token_acc": 0.9698996655518395, "train_speed(iter/s)": 1.447376 }, { "epoch": 2.7590934407266183, "grad_norm": 0.5227271318435669, "learning_rate": 4.189624426784481e-05, "loss": 0.22514455318450927, "memory(GiB)": 70.5, "step": 64400, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.44737 }, { "epoch": 2.759307656055867, "grad_norm": 2.830543041229248, "learning_rate": 4.18896035464306e-05, "loss": 0.4722576141357422, "memory(GiB)": 70.5, "step": 64405, "token_acc": 0.8951310861423221, "train_speed(iter/s)": 1.447369 }, { "epoch": 2.7595218713851164, "grad_norm": 1.9185919761657715, "learning_rate": 4.188296297194384e-05, "loss": 0.07396316528320312, "memory(GiB)": 70.5, "step": 64410, "token_acc": 0.9771863117870723, "train_speed(iter/s)": 1.44738 }, { "epoch": 2.759736086714365, "grad_norm": 2.418405532836914, "learning_rate": 4.18763225445048e-05, "loss": 0.2585972785949707, "memory(GiB)": 70.5, "step": 64415, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.447383 }, { "epoch": 2.759950302043614, "grad_norm": 1.9479100704193115, "learning_rate": 4.1869682264233826e-05, "loss": 0.29885573387145997, "memory(GiB)": 70.5, "step": 64420, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.447397 }, { "epoch": 2.7601645173728633, "grad_norm": 2.0083351135253906, "learning_rate": 4.186304213125116e-05, "loss": 0.32686920166015626, "memory(GiB)": 70.5, "step": 64425, "token_acc": 0.91796875, "train_speed(iter/s)": 1.447402 }, { "epoch": 2.760378732702112, "grad_norm": 4.134090900421143, "learning_rate": 4.185640214567714e-05, "loss": 0.6171991348266601, "memory(GiB)": 70.5, "step": 64430, "token_acc": 0.8490566037735849, "train_speed(iter/s)": 1.447404 }, { "epoch": 2.760592948031361, "grad_norm": 2.885958671569824, "learning_rate": 4.184976230763203e-05, "loss": 0.2849236488342285, "memory(GiB)": 70.5, "step": 64435, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.447402 }, { "epoch": 2.76080716336061, "grad_norm": 2.3937363624572754, "learning_rate": 4.1843122617236106e-05, "loss": 0.4060819625854492, "memory(GiB)": 70.5, "step": 64440, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.447408 }, { "epoch": 2.761021378689859, "grad_norm": 3.775437116622925, "learning_rate": 4.183648307460969e-05, "loss": 0.3258893251419067, "memory(GiB)": 70.5, "step": 64445, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.447428 }, { "epoch": 2.761235594019108, "grad_norm": 5.188766956329346, "learning_rate": 4.182984367987302e-05, "loss": 0.45186190605163573, "memory(GiB)": 70.5, "step": 64450, "token_acc": 0.8951048951048951, "train_speed(iter/s)": 1.447444 }, { "epoch": 2.761449809348357, "grad_norm": 4.472078800201416, "learning_rate": 4.182320443314641e-05, "loss": 0.34918889999389646, "memory(GiB)": 70.5, "step": 64455, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.447444 }, { "epoch": 2.761664024677606, "grad_norm": 3.6090171337127686, "learning_rate": 4.181656533455013e-05, "loss": 0.5202494621276855, "memory(GiB)": 70.5, "step": 64460, "token_acc": 0.8961038961038961, "train_speed(iter/s)": 1.447443 }, { "epoch": 2.7618782400068547, "grad_norm": 4.033676624298096, "learning_rate": 4.1809926384204426e-05, "loss": 0.44240412712097166, "memory(GiB)": 70.5, "step": 64465, "token_acc": 0.8850574712643678, "train_speed(iter/s)": 1.447456 }, { "epoch": 2.762092455336104, "grad_norm": 2.78326153755188, "learning_rate": 4.18032875822296e-05, "loss": 0.30789010524749755, "memory(GiB)": 70.5, "step": 64470, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.447461 }, { "epoch": 2.7623066706653527, "grad_norm": 3.7348554134368896, "learning_rate": 4.179664892874591e-05, "loss": 0.4051199913024902, "memory(GiB)": 70.5, "step": 64475, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.447458 }, { "epoch": 2.7625208859946015, "grad_norm": 2.7712533473968506, "learning_rate": 4.17900104238736e-05, "loss": 0.41397881507873535, "memory(GiB)": 70.5, "step": 64480, "token_acc": 0.8842443729903537, "train_speed(iter/s)": 1.447461 }, { "epoch": 2.762735101323851, "grad_norm": 1.29079008102417, "learning_rate": 4.1783372067732977e-05, "loss": 0.48822813034057616, "memory(GiB)": 70.5, "step": 64485, "token_acc": 0.8859060402684564, "train_speed(iter/s)": 1.447479 }, { "epoch": 2.7629493166530996, "grad_norm": 0.25128626823425293, "learning_rate": 4.177673386044425e-05, "loss": 0.26605615615844724, "memory(GiB)": 70.5, "step": 64490, "token_acc": 0.9439655172413793, "train_speed(iter/s)": 1.447485 }, { "epoch": 2.7631635319823484, "grad_norm": 2.483635187149048, "learning_rate": 4.177009580212773e-05, "loss": 0.410248327255249, "memory(GiB)": 70.5, "step": 64495, "token_acc": 0.8969465648854962, "train_speed(iter/s)": 1.447499 }, { "epoch": 2.7633777473115977, "grad_norm": 3.9132144451141357, "learning_rate": 4.176345789290363e-05, "loss": 0.604255199432373, "memory(GiB)": 70.5, "step": 64500, "token_acc": 0.8783382789317508, "train_speed(iter/s)": 1.447497 }, { "epoch": 2.7633777473115977, "eval_loss": 2.5938799381256104, "eval_runtime": 12.752, "eval_samples_per_second": 7.842, "eval_steps_per_second": 7.842, "eval_token_acc": 0.4528061224489796, "step": 64500 }, { "epoch": 2.7635919626408465, "grad_norm": 3.3909194469451904, "learning_rate": 4.175682013289223e-05, "loss": 0.22704715728759767, "memory(GiB)": 70.5, "step": 64505, "token_acc": 0.5754082612872238, "train_speed(iter/s)": 1.44706 }, { "epoch": 2.7638061779700953, "grad_norm": 4.658672332763672, "learning_rate": 4.1750182522213745e-05, "loss": 0.4202091217041016, "memory(GiB)": 70.5, "step": 64510, "token_acc": 0.9036144578313253, "train_speed(iter/s)": 1.447077 }, { "epoch": 2.7640203932993446, "grad_norm": 4.664797782897949, "learning_rate": 4.174354506098847e-05, "loss": 0.5492270946502685, "memory(GiB)": 70.5, "step": 64515, "token_acc": 0.8896797153024911, "train_speed(iter/s)": 1.447078 }, { "epoch": 2.7642346086285934, "grad_norm": 3.524773120880127, "learning_rate": 4.1736907749336603e-05, "loss": 0.23233861923217775, "memory(GiB)": 70.5, "step": 64520, "token_acc": 0.954225352112676, "train_speed(iter/s)": 1.447086 }, { "epoch": 2.764448823957842, "grad_norm": 4.936490535736084, "learning_rate": 4.17302705873784e-05, "loss": 0.4504572868347168, "memory(GiB)": 70.5, "step": 64525, "token_acc": 0.9039145907473309, "train_speed(iter/s)": 1.447086 }, { "epoch": 2.7646630392870915, "grad_norm": 4.39907169342041, "learning_rate": 4.172363357523412e-05, "loss": 0.37729334831237793, "memory(GiB)": 70.5, "step": 64530, "token_acc": 0.8954248366013072, "train_speed(iter/s)": 1.447086 }, { "epoch": 2.7648772546163403, "grad_norm": 5.258445739746094, "learning_rate": 4.1716996713023976e-05, "loss": 0.3056314468383789, "memory(GiB)": 70.5, "step": 64535, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.447091 }, { "epoch": 2.765091469945589, "grad_norm": 2.3570480346679688, "learning_rate": 4.17103600008682e-05, "loss": 0.5240492820739746, "memory(GiB)": 70.5, "step": 64540, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.447117 }, { "epoch": 2.7653056852748383, "grad_norm": 3.910691022872925, "learning_rate": 4.170372343888703e-05, "loss": 0.40659055709838865, "memory(GiB)": 70.5, "step": 64545, "token_acc": 0.9116719242902208, "train_speed(iter/s)": 1.447122 }, { "epoch": 2.765519900604087, "grad_norm": 4.172811985015869, "learning_rate": 4.169708702720069e-05, "loss": 0.34165616035461427, "memory(GiB)": 70.5, "step": 64550, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.447138 }, { "epoch": 2.765734115933336, "grad_norm": 2.178283929824829, "learning_rate": 4.169045076592942e-05, "loss": 0.2027437686920166, "memory(GiB)": 70.5, "step": 64555, "token_acc": 0.9522184300341296, "train_speed(iter/s)": 1.447143 }, { "epoch": 2.765948331262585, "grad_norm": 1.4516924619674683, "learning_rate": 4.168381465519342e-05, "loss": 0.30718603134155276, "memory(GiB)": 70.5, "step": 64560, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.447151 }, { "epoch": 2.766162546591834, "grad_norm": 3.997983694076538, "learning_rate": 4.167717869511291e-05, "loss": 0.373547625541687, "memory(GiB)": 70.5, "step": 64565, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.44715 }, { "epoch": 2.766376761921083, "grad_norm": 3.704028367996216, "learning_rate": 4.167054288580812e-05, "loss": 0.31400165557861326, "memory(GiB)": 70.5, "step": 64570, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.447152 }, { "epoch": 2.766590977250332, "grad_norm": 0.57720947265625, "learning_rate": 4.166390722739926e-05, "loss": 0.4853196144104004, "memory(GiB)": 70.5, "step": 64575, "token_acc": 0.9083969465648855, "train_speed(iter/s)": 1.447152 }, { "epoch": 2.766805192579581, "grad_norm": 1.3825042247772217, "learning_rate": 4.165727172000655e-05, "loss": 0.5712833881378174, "memory(GiB)": 70.5, "step": 64580, "token_acc": 0.8986486486486487, "train_speed(iter/s)": 1.447169 }, { "epoch": 2.7670194079088297, "grad_norm": 2.686846971511841, "learning_rate": 4.165063636375018e-05, "loss": 0.38083834648132325, "memory(GiB)": 70.5, "step": 64585, "token_acc": 0.9336870026525199, "train_speed(iter/s)": 1.447166 }, { "epoch": 2.767233623238079, "grad_norm": 2.648146867752075, "learning_rate": 4.164400115875037e-05, "loss": 0.499160099029541, "memory(GiB)": 70.5, "step": 64590, "token_acc": 0.8706624605678234, "train_speed(iter/s)": 1.447162 }, { "epoch": 2.767447838567328, "grad_norm": 1.5507270097732544, "learning_rate": 4.1637366105127315e-05, "loss": 0.2310617446899414, "memory(GiB)": 70.5, "step": 64595, "token_acc": 0.9552715654952076, "train_speed(iter/s)": 1.447182 }, { "epoch": 2.7676620538965766, "grad_norm": 2.697460889816284, "learning_rate": 4.163073120300122e-05, "loss": 0.27885434627532957, "memory(GiB)": 70.5, "step": 64600, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.447181 }, { "epoch": 2.767876269225826, "grad_norm": 1.3462733030319214, "learning_rate": 4.162409645249228e-05, "loss": 0.22039551734924318, "memory(GiB)": 70.5, "step": 64605, "token_acc": 0.9528985507246377, "train_speed(iter/s)": 1.447188 }, { "epoch": 2.7680904845550747, "grad_norm": 0.4477658271789551, "learning_rate": 4.1617461853720685e-05, "loss": 0.4889101505279541, "memory(GiB)": 70.5, "step": 64610, "token_acc": 0.9267399267399268, "train_speed(iter/s)": 1.447197 }, { "epoch": 2.7683046998843235, "grad_norm": 3.747922897338867, "learning_rate": 4.161082740680664e-05, "loss": 0.5222739219665528, "memory(GiB)": 70.5, "step": 64615, "token_acc": 0.8797814207650273, "train_speed(iter/s)": 1.447202 }, { "epoch": 2.7685189152135727, "grad_norm": 3.24414324760437, "learning_rate": 4.160419311187033e-05, "loss": 0.5671875, "memory(GiB)": 70.5, "step": 64620, "token_acc": 0.8557046979865772, "train_speed(iter/s)": 1.447201 }, { "epoch": 2.7687331305428216, "grad_norm": 7.235586166381836, "learning_rate": 4.1597558969031924e-05, "loss": 0.20189988613128662, "memory(GiB)": 70.5, "step": 64625, "token_acc": 0.956989247311828, "train_speed(iter/s)": 1.447192 }, { "epoch": 2.7689473458720704, "grad_norm": 8.123555183410645, "learning_rate": 4.159092497841163e-05, "loss": 0.3075474739074707, "memory(GiB)": 70.5, "step": 64630, "token_acc": 0.9461538461538461, "train_speed(iter/s)": 1.447196 }, { "epoch": 2.7691615612013196, "grad_norm": 2.73297119140625, "learning_rate": 4.15842911401296e-05, "loss": 0.29380652904510496, "memory(GiB)": 70.5, "step": 64635, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.447215 }, { "epoch": 2.7693757765305684, "grad_norm": 2.1140921115875244, "learning_rate": 4.157765745430605e-05, "loss": 0.5138519287109375, "memory(GiB)": 70.5, "step": 64640, "token_acc": 0.8848920863309353, "train_speed(iter/s)": 1.447226 }, { "epoch": 2.7695899918598172, "grad_norm": 4.180202007293701, "learning_rate": 4.157102392106112e-05, "loss": 0.42391467094421387, "memory(GiB)": 70.5, "step": 64645, "token_acc": 0.8842105263157894, "train_speed(iter/s)": 1.447226 }, { "epoch": 2.7698042071890665, "grad_norm": 3.212494373321533, "learning_rate": 4.156439054051501e-05, "loss": 0.3758485555648804, "memory(GiB)": 70.5, "step": 64650, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.44722 }, { "epoch": 2.7700184225183153, "grad_norm": 2.735640048980713, "learning_rate": 4.1557757312787874e-05, "loss": 0.48067054748535154, "memory(GiB)": 70.5, "step": 64655, "token_acc": 0.9096209912536443, "train_speed(iter/s)": 1.447226 }, { "epoch": 2.770232637847564, "grad_norm": 2.398343563079834, "learning_rate": 4.1551124237999895e-05, "loss": 0.3100878238677979, "memory(GiB)": 70.5, "step": 64660, "token_acc": 0.9482071713147411, "train_speed(iter/s)": 1.44724 }, { "epoch": 2.7704468531768134, "grad_norm": 1.4659647941589355, "learning_rate": 4.1544491316271224e-05, "loss": 0.46073288917541505, "memory(GiB)": 70.5, "step": 64665, "token_acc": 0.9188311688311688, "train_speed(iter/s)": 1.447247 }, { "epoch": 2.770661068506062, "grad_norm": 7.9042863845825195, "learning_rate": 4.153785854772201e-05, "loss": 0.7685583114624024, "memory(GiB)": 70.5, "step": 64670, "token_acc": 0.8379204892966361, "train_speed(iter/s)": 1.447247 }, { "epoch": 2.770875283835311, "grad_norm": 0.23309653997421265, "learning_rate": 4.153122593247244e-05, "loss": 0.22178409099578858, "memory(GiB)": 70.5, "step": 64675, "token_acc": 0.9504373177842566, "train_speed(iter/s)": 1.447246 }, { "epoch": 2.7710894991645603, "grad_norm": 2.417905330657959, "learning_rate": 4.1524593470642656e-05, "loss": 0.3942133903503418, "memory(GiB)": 70.5, "step": 64680, "token_acc": 0.9081364829396326, "train_speed(iter/s)": 1.447243 }, { "epoch": 2.771303714493809, "grad_norm": 1.842147707939148, "learning_rate": 4.15179611623528e-05, "loss": 0.13138712644577027, "memory(GiB)": 70.5, "step": 64685, "token_acc": 0.9666666666666667, "train_speed(iter/s)": 1.447243 }, { "epoch": 2.771517929823058, "grad_norm": 3.2145869731903076, "learning_rate": 4.1511329007723046e-05, "loss": 0.34955592155456544, "memory(GiB)": 70.5, "step": 64690, "token_acc": 0.9191176470588235, "train_speed(iter/s)": 1.447237 }, { "epoch": 2.771732145152307, "grad_norm": 3.617306709289551, "learning_rate": 4.1504697006873524e-05, "loss": 0.2527862310409546, "memory(GiB)": 70.5, "step": 64695, "token_acc": 0.9405204460966543, "train_speed(iter/s)": 1.447241 }, { "epoch": 2.771946360481556, "grad_norm": 4.195311546325684, "learning_rate": 4.1498065159924394e-05, "loss": 0.21672587394714354, "memory(GiB)": 70.5, "step": 64700, "token_acc": 0.9522184300341296, "train_speed(iter/s)": 1.447242 }, { "epoch": 2.7721605758108048, "grad_norm": 3.529592514038086, "learning_rate": 4.149143346699579e-05, "loss": 0.32183313369750977, "memory(GiB)": 70.5, "step": 64705, "token_acc": 0.94140625, "train_speed(iter/s)": 1.447249 }, { "epoch": 2.772374791140054, "grad_norm": 4.510876655578613, "learning_rate": 4.1484801928207824e-05, "loss": 0.3843485593795776, "memory(GiB)": 70.5, "step": 64710, "token_acc": 0.9179331306990881, "train_speed(iter/s)": 1.447238 }, { "epoch": 2.772589006469303, "grad_norm": 5.611743927001953, "learning_rate": 4.1478170543680664e-05, "loss": 0.2822412967681885, "memory(GiB)": 70.5, "step": 64715, "token_acc": 0.953125, "train_speed(iter/s)": 1.447243 }, { "epoch": 2.7728032217985517, "grad_norm": 5.408237934112549, "learning_rate": 4.147153931353446e-05, "loss": 0.7725577354431152, "memory(GiB)": 70.5, "step": 64720, "token_acc": 0.8527397260273972, "train_speed(iter/s)": 1.447255 }, { "epoch": 2.773017437127801, "grad_norm": 3.0940282344818115, "learning_rate": 4.1464908237889324e-05, "loss": 0.2764521598815918, "memory(GiB)": 70.5, "step": 64725, "token_acc": 0.9482071713147411, "train_speed(iter/s)": 1.447259 }, { "epoch": 2.7732316524570497, "grad_norm": 4.313165664672852, "learning_rate": 4.145827731686536e-05, "loss": 0.2625635385513306, "memory(GiB)": 70.5, "step": 64730, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.447271 }, { "epoch": 2.7734458677862985, "grad_norm": 5.906126499176025, "learning_rate": 4.145164655058273e-05, "loss": 0.5787650108337402, "memory(GiB)": 70.5, "step": 64735, "token_acc": 0.8661971830985915, "train_speed(iter/s)": 1.447279 }, { "epoch": 2.773660083115548, "grad_norm": 5.032597541809082, "learning_rate": 4.144501593916154e-05, "loss": 0.18278589248657226, "memory(GiB)": 70.5, "step": 64740, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.447279 }, { "epoch": 2.7738742984447966, "grad_norm": 2.7044591903686523, "learning_rate": 4.1438385482721913e-05, "loss": 0.38160719871521, "memory(GiB)": 70.5, "step": 64745, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.447285 }, { "epoch": 2.7740885137740454, "grad_norm": 5.301630020141602, "learning_rate": 4.143175518138397e-05, "loss": 0.29948840141296384, "memory(GiB)": 70.5, "step": 64750, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.447278 }, { "epoch": 2.7743027291032947, "grad_norm": 5.285496711730957, "learning_rate": 4.14251250352678e-05, "loss": 0.5495663642883301, "memory(GiB)": 70.5, "step": 64755, "token_acc": 0.8943661971830986, "train_speed(iter/s)": 1.447276 }, { "epoch": 2.7745169444325435, "grad_norm": 2.504960298538208, "learning_rate": 4.141849504449355e-05, "loss": 0.5839006900787354, "memory(GiB)": 70.5, "step": 64760, "token_acc": 0.8621908127208481, "train_speed(iter/s)": 1.447291 }, { "epoch": 2.7747311597617923, "grad_norm": 4.319902420043945, "learning_rate": 4.141186520918132e-05, "loss": 0.4011786460876465, "memory(GiB)": 70.5, "step": 64765, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.447298 }, { "epoch": 2.7749453750910416, "grad_norm": 3.0023369789123535, "learning_rate": 4.140523552945118e-05, "loss": 0.43236145973205564, "memory(GiB)": 70.5, "step": 64770, "token_acc": 0.9, "train_speed(iter/s)": 1.447298 }, { "epoch": 2.7751595904202904, "grad_norm": 2.7904255390167236, "learning_rate": 4.1398606005423284e-05, "loss": 0.21042978763580322, "memory(GiB)": 70.5, "step": 64775, "token_acc": 0.9522292993630573, "train_speed(iter/s)": 1.447292 }, { "epoch": 2.775373805749539, "grad_norm": 4.137453556060791, "learning_rate": 4.13919766372177e-05, "loss": 0.5895178318023682, "memory(GiB)": 70.5, "step": 64780, "token_acc": 0.8791208791208791, "train_speed(iter/s)": 1.447294 }, { "epoch": 2.7755880210787884, "grad_norm": 3.896679162979126, "learning_rate": 4.1385347424954526e-05, "loss": 0.23463189601898193, "memory(GiB)": 70.5, "step": 64785, "token_acc": 0.9372937293729373, "train_speed(iter/s)": 1.447296 }, { "epoch": 2.7758022364080372, "grad_norm": 2.7868499755859375, "learning_rate": 4.137871836875387e-05, "loss": 0.4675302028656006, "memory(GiB)": 70.5, "step": 64790, "token_acc": 0.900990099009901, "train_speed(iter/s)": 1.44731 }, { "epoch": 2.776016451737286, "grad_norm": 4.884207248687744, "learning_rate": 4.137208946873582e-05, "loss": 0.39016094207763674, "memory(GiB)": 70.5, "step": 64795, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.447314 }, { "epoch": 2.7762306670665353, "grad_norm": 2.988022565841675, "learning_rate": 4.1365460725020466e-05, "loss": 0.21896796226501464, "memory(GiB)": 70.5, "step": 64800, "token_acc": 0.9624060150375939, "train_speed(iter/s)": 1.44734 }, { "epoch": 2.776444882395784, "grad_norm": 3.6900827884674072, "learning_rate": 4.135883213772789e-05, "loss": 0.5038420200347901, "memory(GiB)": 70.5, "step": 64805, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.44735 }, { "epoch": 2.776659097725033, "grad_norm": 4.502459526062012, "learning_rate": 4.1352203706978186e-05, "loss": 0.30875296592712403, "memory(GiB)": 70.5, "step": 64810, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.447349 }, { "epoch": 2.776873313054282, "grad_norm": 3.091966152191162, "learning_rate": 4.134557543289141e-05, "loss": 0.21999711990356446, "memory(GiB)": 70.5, "step": 64815, "token_acc": 0.9557739557739557, "train_speed(iter/s)": 1.44735 }, { "epoch": 2.777087528383531, "grad_norm": 1.9836987257003784, "learning_rate": 4.1338947315587664e-05, "loss": 0.16959307193756104, "memory(GiB)": 70.5, "step": 64820, "token_acc": 0.9635761589403974, "train_speed(iter/s)": 1.447355 }, { "epoch": 2.77730174371278, "grad_norm": 2.7940990924835205, "learning_rate": 4.133231935518701e-05, "loss": 0.349604058265686, "memory(GiB)": 70.5, "step": 64825, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.447353 }, { "epoch": 2.777515959042029, "grad_norm": 1.3738101720809937, "learning_rate": 4.132569155180951e-05, "loss": 0.44671082496643066, "memory(GiB)": 70.5, "step": 64830, "token_acc": 0.9106145251396648, "train_speed(iter/s)": 1.447358 }, { "epoch": 2.777730174371278, "grad_norm": 2.232062578201294, "learning_rate": 4.131906390557526e-05, "loss": 0.254518461227417, "memory(GiB)": 70.5, "step": 64835, "token_acc": 0.9429530201342282, "train_speed(iter/s)": 1.447355 }, { "epoch": 2.7779443897005267, "grad_norm": 3.6004254817962646, "learning_rate": 4.131243641660429e-05, "loss": 0.28875668048858644, "memory(GiB)": 70.5, "step": 64840, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.447357 }, { "epoch": 2.778158605029776, "grad_norm": 3.4352262020111084, "learning_rate": 4.130580908501671e-05, "loss": 0.4341428756713867, "memory(GiB)": 70.5, "step": 64845, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.447384 }, { "epoch": 2.7783728203590248, "grad_norm": 0.4371896982192993, "learning_rate": 4.129918191093254e-05, "loss": 0.37861027717590334, "memory(GiB)": 70.5, "step": 64850, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.447389 }, { "epoch": 2.7785870356882736, "grad_norm": 4.118015766143799, "learning_rate": 4.1292554894471847e-05, "loss": 0.38886189460754395, "memory(GiB)": 70.5, "step": 64855, "token_acc": 0.926984126984127, "train_speed(iter/s)": 1.44739 }, { "epoch": 2.778801251017523, "grad_norm": 1.0695202350616455, "learning_rate": 4.1285928035754684e-05, "loss": 0.19757663011550902, "memory(GiB)": 70.5, "step": 64860, "token_acc": 0.9518518518518518, "train_speed(iter/s)": 1.447394 }, { "epoch": 2.7790154663467717, "grad_norm": 1.8571826219558716, "learning_rate": 4.127930133490112e-05, "loss": 0.4682793617248535, "memory(GiB)": 70.5, "step": 64865, "token_acc": 0.8993288590604027, "train_speed(iter/s)": 1.447408 }, { "epoch": 2.7792296816760205, "grad_norm": 2.7603261470794678, "learning_rate": 4.12726747920312e-05, "loss": 0.3878632068634033, "memory(GiB)": 70.5, "step": 64870, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.447405 }, { "epoch": 2.7794438970052697, "grad_norm": 3.317103624343872, "learning_rate": 4.126604840726496e-05, "loss": 0.7620277881622315, "memory(GiB)": 70.5, "step": 64875, "token_acc": 0.8393574297188755, "train_speed(iter/s)": 1.447412 }, { "epoch": 2.7796581123345185, "grad_norm": 3.060573101043701, "learning_rate": 4.125942218072244e-05, "loss": 0.3025252103805542, "memory(GiB)": 70.5, "step": 64880, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.44741 }, { "epoch": 2.7798723276637674, "grad_norm": 3.1476831436157227, "learning_rate": 4.125279611252369e-05, "loss": 0.3223943948745728, "memory(GiB)": 70.5, "step": 64885, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.44741 }, { "epoch": 2.7800865429930166, "grad_norm": 4.828221321105957, "learning_rate": 4.124617020278875e-05, "loss": 0.28006443977355955, "memory(GiB)": 70.5, "step": 64890, "token_acc": 0.9449838187702265, "train_speed(iter/s)": 1.447411 }, { "epoch": 2.7803007583222654, "grad_norm": 1.4535952806472778, "learning_rate": 4.1239544451637646e-05, "loss": 0.3131978988647461, "memory(GiB)": 70.5, "step": 64895, "token_acc": 0.955719557195572, "train_speed(iter/s)": 1.447419 }, { "epoch": 2.7805149736515142, "grad_norm": 3.597987174987793, "learning_rate": 4.12329188591904e-05, "loss": 0.2687454462051392, "memory(GiB)": 70.5, "step": 64900, "token_acc": 0.9535603715170279, "train_speed(iter/s)": 1.447422 }, { "epoch": 2.7807291889807635, "grad_norm": 6.607239246368408, "learning_rate": 4.122629342556706e-05, "loss": 0.7134824752807617, "memory(GiB)": 70.5, "step": 64905, "token_acc": 0.8427835051546392, "train_speed(iter/s)": 1.447419 }, { "epoch": 2.7809434043100123, "grad_norm": 1.579677939414978, "learning_rate": 4.121966815088766e-05, "loss": 0.5193864345550537, "memory(GiB)": 70.5, "step": 64910, "token_acc": 0.915129151291513, "train_speed(iter/s)": 1.44742 }, { "epoch": 2.781157619639261, "grad_norm": 0.8404396772384644, "learning_rate": 4.1213043035272184e-05, "loss": 0.34959518909454346, "memory(GiB)": 70.5, "step": 64915, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.447427 }, { "epoch": 2.7813718349685104, "grad_norm": 2.582603931427002, "learning_rate": 4.12064180788407e-05, "loss": 0.15796160697937012, "memory(GiB)": 70.5, "step": 64920, "token_acc": 0.9688888888888889, "train_speed(iter/s)": 1.447428 }, { "epoch": 2.781586050297759, "grad_norm": 5.888064384460449, "learning_rate": 4.1199793281713176e-05, "loss": 0.48841042518615724, "memory(GiB)": 70.5, "step": 64925, "token_acc": 0.9075342465753424, "train_speed(iter/s)": 1.447441 }, { "epoch": 2.781800265627008, "grad_norm": 2.2963101863861084, "learning_rate": 4.119316864400967e-05, "loss": 0.21862392425537108, "memory(GiB)": 70.5, "step": 64930, "token_acc": 0.9367588932806324, "train_speed(iter/s)": 1.44744 }, { "epoch": 2.7820144809562573, "grad_norm": 3.4301040172576904, "learning_rate": 4.118654416585015e-05, "loss": 0.5008277893066406, "memory(GiB)": 70.5, "step": 64935, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.447444 }, { "epoch": 2.782228696285506, "grad_norm": 2.9857544898986816, "learning_rate": 4.117991984735468e-05, "loss": 0.4866320610046387, "memory(GiB)": 70.5, "step": 64940, "token_acc": 0.878698224852071, "train_speed(iter/s)": 1.447444 }, { "epoch": 2.782442911614755, "grad_norm": 2.7495787143707275, "learning_rate": 4.117329568864322e-05, "loss": 0.33805294036865235, "memory(GiB)": 70.5, "step": 64945, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.447447 }, { "epoch": 2.782657126944004, "grad_norm": 0.3812364935874939, "learning_rate": 4.11666716898358e-05, "loss": 0.2827253818511963, "memory(GiB)": 70.5, "step": 64950, "token_acc": 0.9246031746031746, "train_speed(iter/s)": 1.447444 }, { "epoch": 2.782871342273253, "grad_norm": 3.597952127456665, "learning_rate": 4.116004785105241e-05, "loss": 0.6536751747131347, "memory(GiB)": 70.5, "step": 64955, "token_acc": 0.8872727272727273, "train_speed(iter/s)": 1.447456 }, { "epoch": 2.7830855576025018, "grad_norm": 5.029714584350586, "learning_rate": 4.115342417241304e-05, "loss": 0.289227557182312, "memory(GiB)": 70.5, "step": 64960, "token_acc": 0.9349593495934959, "train_speed(iter/s)": 1.447461 }, { "epoch": 2.783299772931751, "grad_norm": 3.434293508529663, "learning_rate": 4.114680065403769e-05, "loss": 0.34526231288909914, "memory(GiB)": 70.5, "step": 64965, "token_acc": 0.925, "train_speed(iter/s)": 1.447465 }, { "epoch": 2.783513988261, "grad_norm": 1.8889940977096558, "learning_rate": 4.114017729604635e-05, "loss": 0.5252589225769043, "memory(GiB)": 70.5, "step": 64970, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.447477 }, { "epoch": 2.783728203590249, "grad_norm": 1.9018261432647705, "learning_rate": 4.113355409855901e-05, "loss": 0.15777170658111572, "memory(GiB)": 70.5, "step": 64975, "token_acc": 0.9636363636363636, "train_speed(iter/s)": 1.44748 }, { "epoch": 2.783942418919498, "grad_norm": 2.4065496921539307, "learning_rate": 4.1126931061695656e-05, "loss": 0.2429669141769409, "memory(GiB)": 70.5, "step": 64980, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.447479 }, { "epoch": 2.7841566342487467, "grad_norm": 2.436152935028076, "learning_rate": 4.112030818557626e-05, "loss": 0.5359068870544433, "memory(GiB)": 70.5, "step": 64985, "token_acc": 0.8416666666666667, "train_speed(iter/s)": 1.447481 }, { "epoch": 2.784370849577996, "grad_norm": 1.480725884437561, "learning_rate": 4.111368547032083e-05, "loss": 0.2701308012008667, "memory(GiB)": 70.5, "step": 64990, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.447488 }, { "epoch": 2.784585064907245, "grad_norm": 4.441800117492676, "learning_rate": 4.110706291604931e-05, "loss": 0.20483908653259278, "memory(GiB)": 70.5, "step": 64995, "token_acc": 0.9664179104477612, "train_speed(iter/s)": 1.447496 }, { "epoch": 2.7847992802364936, "grad_norm": 3.6579749584198, "learning_rate": 4.110044052288169e-05, "loss": 0.38043386936187745, "memory(GiB)": 70.5, "step": 65000, "token_acc": 0.9193083573487032, "train_speed(iter/s)": 1.4475 }, { "epoch": 2.7847992802364936, "eval_loss": 2.4592678546905518, "eval_runtime": 13.3801, "eval_samples_per_second": 7.474, "eval_steps_per_second": 7.474, "eval_token_acc": 0.4638472032742155, "step": 65000 }, { "epoch": 2.785013495565743, "grad_norm": 4.061110973358154, "learning_rate": 4.109381829093792e-05, "loss": 0.41044392585754397, "memory(GiB)": 70.5, "step": 65005, "token_acc": 0.5843253968253969, "train_speed(iter/s)": 1.447039 }, { "epoch": 2.7852277108949917, "grad_norm": 4.724144458770752, "learning_rate": 4.108719622033801e-05, "loss": 0.595648717880249, "memory(GiB)": 70.5, "step": 65010, "token_acc": 0.8662207357859532, "train_speed(iter/s)": 1.44705 }, { "epoch": 2.7854419262242405, "grad_norm": 4.578857898712158, "learning_rate": 4.10805743112019e-05, "loss": 0.36465792655944823, "memory(GiB)": 70.5, "step": 65015, "token_acc": 0.9224489795918367, "train_speed(iter/s)": 1.447059 }, { "epoch": 2.7856561415534897, "grad_norm": 4.199429035186768, "learning_rate": 4.1073952563649546e-05, "loss": 0.29490208625793457, "memory(GiB)": 70.5, "step": 65020, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.447073 }, { "epoch": 2.7858703568827385, "grad_norm": 0.2744632959365845, "learning_rate": 4.1067330977800924e-05, "loss": 0.36468636989593506, "memory(GiB)": 70.5, "step": 65025, "token_acc": 0.9291666666666667, "train_speed(iter/s)": 1.447071 }, { "epoch": 2.7860845722119874, "grad_norm": 3.3545267581939697, "learning_rate": 4.106070955377597e-05, "loss": 0.29489474296569823, "memory(GiB)": 70.5, "step": 65030, "token_acc": 0.9457364341085271, "train_speed(iter/s)": 1.447079 }, { "epoch": 2.7862987875412366, "grad_norm": 6.718740940093994, "learning_rate": 4.105408829169466e-05, "loss": 0.47815766334533694, "memory(GiB)": 70.5, "step": 65035, "token_acc": 0.9027237354085603, "train_speed(iter/s)": 1.447082 }, { "epoch": 2.7865130028704854, "grad_norm": 2.865877628326416, "learning_rate": 4.104746719167693e-05, "loss": 0.42421865463256836, "memory(GiB)": 70.5, "step": 65040, "token_acc": 0.9093851132686084, "train_speed(iter/s)": 1.447096 }, { "epoch": 2.7867272181997342, "grad_norm": 2.4726219177246094, "learning_rate": 4.104084625384272e-05, "loss": 0.16710200309753417, "memory(GiB)": 70.5, "step": 65045, "token_acc": 0.9647058823529412, "train_speed(iter/s)": 1.447098 }, { "epoch": 2.7869414335289835, "grad_norm": 1.2461167573928833, "learning_rate": 4.103422547831199e-05, "loss": 0.3984663009643555, "memory(GiB)": 70.5, "step": 65050, "token_acc": 0.9134615384615384, "train_speed(iter/s)": 1.447103 }, { "epoch": 2.7871556488582323, "grad_norm": 4.614664554595947, "learning_rate": 4.102760486520468e-05, "loss": 0.27661948204040526, "memory(GiB)": 70.5, "step": 65055, "token_acc": 0.925, "train_speed(iter/s)": 1.447109 }, { "epoch": 2.787369864187481, "grad_norm": 2.549321413040161, "learning_rate": 4.1020984414640716e-05, "loss": 0.25324244499206544, "memory(GiB)": 70.5, "step": 65060, "token_acc": 0.9656488549618321, "train_speed(iter/s)": 1.447123 }, { "epoch": 2.7875840795167304, "grad_norm": 3.4791970252990723, "learning_rate": 4.1014364126740056e-05, "loss": 0.38058953285217284, "memory(GiB)": 70.5, "step": 65065, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.447121 }, { "epoch": 2.787798294845979, "grad_norm": 3.3028554916381836, "learning_rate": 4.100774400162261e-05, "loss": 0.27556452751159666, "memory(GiB)": 70.5, "step": 65070, "token_acc": 0.948051948051948, "train_speed(iter/s)": 1.447123 }, { "epoch": 2.788012510175228, "grad_norm": 4.270534515380859, "learning_rate": 4.100112403940832e-05, "loss": 0.3254996299743652, "memory(GiB)": 70.5, "step": 65075, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.44715 }, { "epoch": 2.7882267255044773, "grad_norm": 2.6122608184814453, "learning_rate": 4.099450424021709e-05, "loss": 0.42832245826721194, "memory(GiB)": 70.5, "step": 65080, "token_acc": 0.9152542372881356, "train_speed(iter/s)": 1.447151 }, { "epoch": 2.788440940833726, "grad_norm": 1.8529683351516724, "learning_rate": 4.0987884604168886e-05, "loss": 0.7383793830871582, "memory(GiB)": 70.5, "step": 65085, "token_acc": 0.863013698630137, "train_speed(iter/s)": 1.447156 }, { "epoch": 2.788655156162975, "grad_norm": 3.683210849761963, "learning_rate": 4.09812651313836e-05, "loss": 0.2979586124420166, "memory(GiB)": 70.5, "step": 65090, "token_acc": 0.946843853820598, "train_speed(iter/s)": 1.447153 }, { "epoch": 2.788869371492224, "grad_norm": 3.5415878295898438, "learning_rate": 4.097464582198116e-05, "loss": 0.4436281681060791, "memory(GiB)": 70.5, "step": 65095, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.447163 }, { "epoch": 2.789083586821473, "grad_norm": 2.8948781490325928, "learning_rate": 4.0968026676081474e-05, "loss": 0.20174810886383057, "memory(GiB)": 70.5, "step": 65100, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.447164 }, { "epoch": 2.7892978021507218, "grad_norm": 2.343510627746582, "learning_rate": 4.096140769380445e-05, "loss": 0.2816267251968384, "memory(GiB)": 70.5, "step": 65105, "token_acc": 0.9303135888501742, "train_speed(iter/s)": 1.447164 }, { "epoch": 2.789512017479971, "grad_norm": 3.802795886993408, "learning_rate": 4.095478887527002e-05, "loss": 0.4312308311462402, "memory(GiB)": 70.5, "step": 65110, "token_acc": 0.915625, "train_speed(iter/s)": 1.447176 }, { "epoch": 2.78972623280922, "grad_norm": 3.681272268295288, "learning_rate": 4.094817022059806e-05, "loss": 0.4683330535888672, "memory(GiB)": 70.5, "step": 65115, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.447184 }, { "epoch": 2.7899404481384686, "grad_norm": 3.3453447818756104, "learning_rate": 4.09415517299085e-05, "loss": 0.41641788482666015, "memory(GiB)": 70.5, "step": 65120, "token_acc": 0.8983739837398373, "train_speed(iter/s)": 1.447193 }, { "epoch": 2.790154663467718, "grad_norm": 2.2669897079467773, "learning_rate": 4.0934933403321226e-05, "loss": 0.19046480655670167, "memory(GiB)": 70.5, "step": 65125, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.447206 }, { "epoch": 2.7903688787969667, "grad_norm": 1.7267236709594727, "learning_rate": 4.0928315240956134e-05, "loss": 0.26402623653411866, "memory(GiB)": 70.5, "step": 65130, "token_acc": 0.94140625, "train_speed(iter/s)": 1.447207 }, { "epoch": 2.7905830941262155, "grad_norm": 5.350974082946777, "learning_rate": 4.0921697242933125e-05, "loss": 0.40633444786071776, "memory(GiB)": 70.5, "step": 65135, "token_acc": 0.9276729559748428, "train_speed(iter/s)": 1.447219 }, { "epoch": 2.790797309455465, "grad_norm": 5.300814151763916, "learning_rate": 4.0915079409372094e-05, "loss": 0.3806781530380249, "memory(GiB)": 70.5, "step": 65140, "token_acc": 0.925, "train_speed(iter/s)": 1.447227 }, { "epoch": 2.7910115247847136, "grad_norm": 1.1849637031555176, "learning_rate": 4.09084617403929e-05, "loss": 0.27837743759155276, "memory(GiB)": 70.5, "step": 65145, "token_acc": 0.9315476190476191, "train_speed(iter/s)": 1.447226 }, { "epoch": 2.7912257401139624, "grad_norm": 1.9873937368392944, "learning_rate": 4.0901844236115464e-05, "loss": 0.19775410890579223, "memory(GiB)": 70.5, "step": 65150, "token_acc": 0.9566666666666667, "train_speed(iter/s)": 1.447228 }, { "epoch": 2.7914399554432117, "grad_norm": 1.9303464889526367, "learning_rate": 4.089522689665964e-05, "loss": 0.28887290954589845, "memory(GiB)": 70.5, "step": 65155, "token_acc": 0.9250720461095101, "train_speed(iter/s)": 1.447231 }, { "epoch": 2.7916541707724605, "grad_norm": 2.440824031829834, "learning_rate": 4.088860972214534e-05, "loss": 0.3914952754974365, "memory(GiB)": 70.5, "step": 65160, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.447235 }, { "epoch": 2.7918683861017093, "grad_norm": 4.724839687347412, "learning_rate": 4.088199271269241e-05, "loss": 0.5876166343688964, "memory(GiB)": 70.5, "step": 65165, "token_acc": 0.873015873015873, "train_speed(iter/s)": 1.447233 }, { "epoch": 2.7920826014309585, "grad_norm": 2.315429449081421, "learning_rate": 4.087537586842074e-05, "loss": 0.34972562789916994, "memory(GiB)": 70.5, "step": 65170, "token_acc": 0.906832298136646, "train_speed(iter/s)": 1.447236 }, { "epoch": 2.7922968167602074, "grad_norm": 1.3048852682113647, "learning_rate": 4.086875918945019e-05, "loss": 0.23476040363311768, "memory(GiB)": 70.5, "step": 65175, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.447239 }, { "epoch": 2.792511032089456, "grad_norm": 5.036691188812256, "learning_rate": 4.0862142675900645e-05, "loss": 0.500761604309082, "memory(GiB)": 70.5, "step": 65180, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.447253 }, { "epoch": 2.7927252474187054, "grad_norm": 0.4292367398738861, "learning_rate": 4.0855526327891956e-05, "loss": 0.0979434072971344, "memory(GiB)": 70.5, "step": 65185, "token_acc": 0.9802371541501976, "train_speed(iter/s)": 1.447269 }, { "epoch": 2.7929394627479542, "grad_norm": 1.7389988899230957, "learning_rate": 4.084891014554398e-05, "loss": 0.4357320308685303, "memory(GiB)": 70.5, "step": 65190, "token_acc": 0.9094650205761317, "train_speed(iter/s)": 1.447286 }, { "epoch": 2.793153678077203, "grad_norm": 2.9058542251586914, "learning_rate": 4.0842294128976586e-05, "loss": 0.4205146789550781, "memory(GiB)": 70.5, "step": 65195, "token_acc": 0.9254385964912281, "train_speed(iter/s)": 1.447284 }, { "epoch": 2.7933678934064523, "grad_norm": 2.5856099128723145, "learning_rate": 4.083567827830962e-05, "loss": 0.2692077398300171, "memory(GiB)": 70.5, "step": 65200, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.447286 }, { "epoch": 2.793582108735701, "grad_norm": 4.839263439178467, "learning_rate": 4.0829062593662944e-05, "loss": 0.7059619903564454, "memory(GiB)": 70.5, "step": 65205, "token_acc": 0.8438538205980066, "train_speed(iter/s)": 1.447293 }, { "epoch": 2.79379632406495, "grad_norm": 4.5880608558654785, "learning_rate": 4.08224470751564e-05, "loss": 0.48299360275268555, "memory(GiB)": 70.5, "step": 65210, "token_acc": 0.8990228013029316, "train_speed(iter/s)": 1.447307 }, { "epoch": 2.794010539394199, "grad_norm": 3.58577299118042, "learning_rate": 4.081583172290983e-05, "loss": 0.3507451057434082, "memory(GiB)": 70.5, "step": 65215, "token_acc": 0.9138461538461539, "train_speed(iter/s)": 1.447302 }, { "epoch": 2.794224754723448, "grad_norm": 3.971111297607422, "learning_rate": 4.080921653704309e-05, "loss": 0.4299809455871582, "memory(GiB)": 70.5, "step": 65220, "token_acc": 0.8777292576419214, "train_speed(iter/s)": 1.447299 }, { "epoch": 2.794438970052697, "grad_norm": 2.350179672241211, "learning_rate": 4.080260151767602e-05, "loss": 0.12587409019470214, "memory(GiB)": 70.5, "step": 65225, "token_acc": 0.9649805447470817, "train_speed(iter/s)": 1.447301 }, { "epoch": 2.794653185381946, "grad_norm": 2.078672170639038, "learning_rate": 4.079598666492843e-05, "loss": 0.47969489097595214, "memory(GiB)": 70.5, "step": 65230, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.447303 }, { "epoch": 2.794867400711195, "grad_norm": 6.421050548553467, "learning_rate": 4.0789371978920185e-05, "loss": 0.32059345245361326, "memory(GiB)": 70.5, "step": 65235, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.447293 }, { "epoch": 2.7950816160404437, "grad_norm": 0.8864030838012695, "learning_rate": 4.078275745977112e-05, "loss": 0.2824270486831665, "memory(GiB)": 70.5, "step": 65240, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.447291 }, { "epoch": 2.795295831369693, "grad_norm": 2.2251110076904297, "learning_rate": 4.0776143107601037e-05, "loss": 0.3151897668838501, "memory(GiB)": 70.5, "step": 65245, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.447308 }, { "epoch": 2.7955100466989418, "grad_norm": 4.933167934417725, "learning_rate": 4.076952892252977e-05, "loss": 0.5641922950744629, "memory(GiB)": 70.5, "step": 65250, "token_acc": 0.8676923076923077, "train_speed(iter/s)": 1.44731 }, { "epoch": 2.7957242620281906, "grad_norm": 2.552314519882202, "learning_rate": 4.0762914904677165e-05, "loss": 0.22024898529052733, "memory(GiB)": 70.5, "step": 65255, "token_acc": 0.9604863221884499, "train_speed(iter/s)": 1.447311 }, { "epoch": 2.79593847735744, "grad_norm": 3.391914129257202, "learning_rate": 4.0756301054163004e-05, "loss": 0.3218238830566406, "memory(GiB)": 70.5, "step": 65260, "token_acc": 0.943089430894309, "train_speed(iter/s)": 1.447311 }, { "epoch": 2.7961526926866886, "grad_norm": 1.7928359508514404, "learning_rate": 4.074968737110713e-05, "loss": 0.2470175266265869, "memory(GiB)": 70.5, "step": 65265, "token_acc": 0.9530685920577617, "train_speed(iter/s)": 1.447314 }, { "epoch": 2.7963669080159375, "grad_norm": 1.5529485940933228, "learning_rate": 4.0743073855629355e-05, "loss": 0.3972229719161987, "memory(GiB)": 70.5, "step": 65270, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.447307 }, { "epoch": 2.7965811233451867, "grad_norm": 5.2288923263549805, "learning_rate": 4.073646050784946e-05, "loss": 0.1909575343132019, "memory(GiB)": 70.5, "step": 65275, "token_acc": 0.952054794520548, "train_speed(iter/s)": 1.447313 }, { "epoch": 2.7967953386744355, "grad_norm": 5.515224933624268, "learning_rate": 4.072984732788729e-05, "loss": 0.6055114269256592, "memory(GiB)": 70.5, "step": 65280, "token_acc": 0.8566666666666667, "train_speed(iter/s)": 1.447339 }, { "epoch": 2.7970095540036843, "grad_norm": 2.583432197570801, "learning_rate": 4.072323431586263e-05, "loss": 0.494219446182251, "memory(GiB)": 70.5, "step": 65285, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.447336 }, { "epoch": 2.7972237693329336, "grad_norm": 3.9602243900299072, "learning_rate": 4.0716621471895275e-05, "loss": 0.3808135509490967, "memory(GiB)": 70.5, "step": 65290, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.447338 }, { "epoch": 2.7974379846621824, "grad_norm": 2.2115015983581543, "learning_rate": 4.0710008796105034e-05, "loss": 0.5067630767822265, "memory(GiB)": 70.5, "step": 65295, "token_acc": 0.9030303030303031, "train_speed(iter/s)": 1.447341 }, { "epoch": 2.797652199991431, "grad_norm": 2.4326882362365723, "learning_rate": 4.0703396288611694e-05, "loss": 0.1473292112350464, "memory(GiB)": 70.5, "step": 65300, "token_acc": 0.9560439560439561, "train_speed(iter/s)": 1.447345 }, { "epoch": 2.7978664153206805, "grad_norm": 2.0492899417877197, "learning_rate": 4.069678394953505e-05, "loss": 0.44561061859130857, "memory(GiB)": 70.5, "step": 65305, "token_acc": 0.9168900804289544, "train_speed(iter/s)": 1.447354 }, { "epoch": 2.7980806306499293, "grad_norm": 3.1452085971832275, "learning_rate": 4.069017177899489e-05, "loss": 0.4552005767822266, "memory(GiB)": 70.5, "step": 65310, "token_acc": 0.8798798798798799, "train_speed(iter/s)": 1.447355 }, { "epoch": 2.7982948459791785, "grad_norm": 2.184251070022583, "learning_rate": 4.0683559777111014e-05, "loss": 0.18176445960998536, "memory(GiB)": 70.5, "step": 65315, "token_acc": 0.9619377162629758, "train_speed(iter/s)": 1.447369 }, { "epoch": 2.7985090613084274, "grad_norm": 3.111306667327881, "learning_rate": 4.0676947944003175e-05, "loss": 0.3205218553543091, "memory(GiB)": 70.5, "step": 65320, "token_acc": 0.9379562043795621, "train_speed(iter/s)": 1.447372 }, { "epoch": 2.798723276637676, "grad_norm": 3.9730777740478516, "learning_rate": 4.0670336279791186e-05, "loss": 0.5713878154754639, "memory(GiB)": 70.5, "step": 65325, "token_acc": 0.9009287925696594, "train_speed(iter/s)": 1.447381 }, { "epoch": 2.7989374919669254, "grad_norm": 1.4721957445144653, "learning_rate": 4.066372478459481e-05, "loss": 0.37087211608886717, "memory(GiB)": 70.5, "step": 65330, "token_acc": 0.9395161290322581, "train_speed(iter/s)": 1.44738 }, { "epoch": 2.7991517072961742, "grad_norm": 3.065363645553589, "learning_rate": 4.06571134585338e-05, "loss": 0.35188295841217043, "memory(GiB)": 70.5, "step": 65335, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.447376 }, { "epoch": 2.799365922625423, "grad_norm": 0.04467729106545448, "learning_rate": 4.065050230172796e-05, "loss": 0.22762069702148438, "memory(GiB)": 70.5, "step": 65340, "token_acc": 0.9543726235741445, "train_speed(iter/s)": 1.447375 }, { "epoch": 2.7995801379546723, "grad_norm": 2.0564699172973633, "learning_rate": 4.064389131429704e-05, "loss": 0.4050747871398926, "memory(GiB)": 70.5, "step": 65345, "token_acc": 0.9150943396226415, "train_speed(iter/s)": 1.447378 }, { "epoch": 2.799794353283921, "grad_norm": 2.836165189743042, "learning_rate": 4.0637280496360795e-05, "loss": 0.31054461002349854, "memory(GiB)": 70.5, "step": 65350, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.447384 }, { "epoch": 2.80000856861317, "grad_norm": 4.740918159484863, "learning_rate": 4.0630669848039005e-05, "loss": 0.5263712406158447, "memory(GiB)": 70.5, "step": 65355, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.447388 }, { "epoch": 2.800222783942419, "grad_norm": 2.8751816749572754, "learning_rate": 4.0624059369451415e-05, "loss": 0.6307284355163574, "memory(GiB)": 70.5, "step": 65360, "token_acc": 0.8700361010830325, "train_speed(iter/s)": 1.447398 }, { "epoch": 2.800436999271668, "grad_norm": 0.8935974836349487, "learning_rate": 4.061744906071779e-05, "loss": 0.10794730186462402, "memory(GiB)": 70.5, "step": 65365, "token_acc": 0.9775280898876404, "train_speed(iter/s)": 1.447399 }, { "epoch": 2.800651214600917, "grad_norm": 1.3291949033737183, "learning_rate": 4.061083892195788e-05, "loss": 0.5285824298858642, "memory(GiB)": 70.5, "step": 65370, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.447387 }, { "epoch": 2.800865429930166, "grad_norm": 5.042686939239502, "learning_rate": 4.0604228953291404e-05, "loss": 0.39534845352172854, "memory(GiB)": 70.5, "step": 65375, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.447387 }, { "epoch": 2.801079645259415, "grad_norm": 2.6069557666778564, "learning_rate": 4.059761915483815e-05, "loss": 0.7030075550079345, "memory(GiB)": 70.5, "step": 65380, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.447402 }, { "epoch": 2.8012938605886637, "grad_norm": 0.2612636387348175, "learning_rate": 4.059100952671786e-05, "loss": 0.5105576992034913, "memory(GiB)": 70.5, "step": 65385, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.447408 }, { "epoch": 2.801508075917913, "grad_norm": 4.365941524505615, "learning_rate": 4.058440006905025e-05, "loss": 0.15441906452178955, "memory(GiB)": 70.5, "step": 65390, "token_acc": 0.952755905511811, "train_speed(iter/s)": 1.447413 }, { "epoch": 2.8017222912471618, "grad_norm": 1.234144687652588, "learning_rate": 4.057779078195506e-05, "loss": 0.24028642177581788, "memory(GiB)": 70.5, "step": 65395, "token_acc": 0.953125, "train_speed(iter/s)": 1.447416 }, { "epoch": 2.8019365065764106, "grad_norm": 2.396169900894165, "learning_rate": 4.0571181665552035e-05, "loss": 0.2668771743774414, "memory(GiB)": 70.5, "step": 65400, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.44742 }, { "epoch": 2.80215072190566, "grad_norm": 3.0904831886291504, "learning_rate": 4.05645727199609e-05, "loss": 0.4542675018310547, "memory(GiB)": 70.5, "step": 65405, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.447418 }, { "epoch": 2.8023649372349086, "grad_norm": 4.245720863342285, "learning_rate": 4.055796394530138e-05, "loss": 0.299834156036377, "memory(GiB)": 70.5, "step": 65410, "token_acc": 0.9315589353612167, "train_speed(iter/s)": 1.44742 }, { "epoch": 2.8025791525641575, "grad_norm": 2.8813765048980713, "learning_rate": 4.05513553416932e-05, "loss": 0.22031049728393554, "memory(GiB)": 70.5, "step": 65415, "token_acc": 0.949685534591195, "train_speed(iter/s)": 1.447435 }, { "epoch": 2.8027933678934067, "grad_norm": 1.8902325630187988, "learning_rate": 4.054474690925607e-05, "loss": 0.16527761220932008, "memory(GiB)": 70.5, "step": 65420, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.447439 }, { "epoch": 2.8030075832226555, "grad_norm": 6.8477373123168945, "learning_rate": 4.053813864810974e-05, "loss": 0.5375009059906006, "memory(GiB)": 70.5, "step": 65425, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.447462 }, { "epoch": 2.8032217985519043, "grad_norm": 4.156376361846924, "learning_rate": 4.05315305583739e-05, "loss": 0.5096409797668457, "memory(GiB)": 70.5, "step": 65430, "token_acc": 0.8904109589041096, "train_speed(iter/s)": 1.447475 }, { "epoch": 2.8034360138811536, "grad_norm": 2.7227964401245117, "learning_rate": 4.052492264016825e-05, "loss": 0.4367702484130859, "memory(GiB)": 70.5, "step": 65435, "token_acc": 0.9093959731543624, "train_speed(iter/s)": 1.447492 }, { "epoch": 2.8036502292104024, "grad_norm": 2.993776798248291, "learning_rate": 4.0518314893612535e-05, "loss": 0.5975071907043457, "memory(GiB)": 70.5, "step": 65440, "token_acc": 0.8687258687258688, "train_speed(iter/s)": 1.4475 }, { "epoch": 2.8038644445396512, "grad_norm": 1.3467299938201904, "learning_rate": 4.0511707318826426e-05, "loss": 0.23333203792572021, "memory(GiB)": 70.5, "step": 65445, "token_acc": 0.9197530864197531, "train_speed(iter/s)": 1.44751 }, { "epoch": 2.8040786598689005, "grad_norm": 2.0019307136535645, "learning_rate": 4.050509991592964e-05, "loss": 0.3400393009185791, "memory(GiB)": 70.5, "step": 65450, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.447512 }, { "epoch": 2.8042928751981493, "grad_norm": 2.286032199859619, "learning_rate": 4.049849268504187e-05, "loss": 0.30358152389526366, "memory(GiB)": 70.5, "step": 65455, "token_acc": 0.9293286219081273, "train_speed(iter/s)": 1.447529 }, { "epoch": 2.804507090527398, "grad_norm": 5.2249016761779785, "learning_rate": 4.0491885626282836e-05, "loss": 0.35160245895385744, "memory(GiB)": 70.5, "step": 65460, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.447534 }, { "epoch": 2.8047213058566474, "grad_norm": 3.2115511894226074, "learning_rate": 4.04852787397722e-05, "loss": 0.4058983325958252, "memory(GiB)": 70.5, "step": 65465, "token_acc": 0.896875, "train_speed(iter/s)": 1.447536 }, { "epoch": 2.804935521185896, "grad_norm": 2.6461009979248047, "learning_rate": 4.047867202562967e-05, "loss": 0.3824728727340698, "memory(GiB)": 70.5, "step": 65470, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.447555 }, { "epoch": 2.805149736515145, "grad_norm": 2.3565335273742676, "learning_rate": 4.0472065483974933e-05, "loss": 0.16277716159820557, "memory(GiB)": 70.5, "step": 65475, "token_acc": 0.968503937007874, "train_speed(iter/s)": 1.447557 }, { "epoch": 2.8053639518443942, "grad_norm": 0.04891785979270935, "learning_rate": 4.046545911492766e-05, "loss": 0.1885126233100891, "memory(GiB)": 70.5, "step": 65480, "token_acc": 0.940625, "train_speed(iter/s)": 1.447554 }, { "epoch": 2.805578167173643, "grad_norm": 4.535712718963623, "learning_rate": 4.0458852918607545e-05, "loss": 0.3993035316467285, "memory(GiB)": 70.5, "step": 65485, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.447567 }, { "epoch": 2.805792382502892, "grad_norm": 1.2075130939483643, "learning_rate": 4.0452246895134266e-05, "loss": 0.24976034164428712, "memory(GiB)": 70.5, "step": 65490, "token_acc": 0.9531772575250836, "train_speed(iter/s)": 1.447573 }, { "epoch": 2.806006597832141, "grad_norm": 2.4892630577087402, "learning_rate": 4.044564104462747e-05, "loss": 0.16194266080856323, "memory(GiB)": 70.5, "step": 65495, "token_acc": 0.9674267100977199, "train_speed(iter/s)": 1.447596 }, { "epoch": 2.80622081316139, "grad_norm": 4.754246711730957, "learning_rate": 4.0439035367206875e-05, "loss": 0.35838191509246825, "memory(GiB)": 70.5, "step": 65500, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.447602 }, { "epoch": 2.80622081316139, "eval_loss": 2.61106538772583, "eval_runtime": 12.7838, "eval_samples_per_second": 7.822, "eval_steps_per_second": 7.822, "eval_token_acc": 0.4450402144772118, "step": 65500 }, { "epoch": 2.8064350284906388, "grad_norm": 2.1217753887176514, "learning_rate": 4.04324298629921e-05, "loss": 0.41364588737487795, "memory(GiB)": 70.5, "step": 65505, "token_acc": 0.5674325674325674, "train_speed(iter/s)": 1.447165 }, { "epoch": 2.806649243819888, "grad_norm": 2.7716224193573, "learning_rate": 4.042582453210285e-05, "loss": 0.2500730514526367, "memory(GiB)": 70.5, "step": 65510, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.447169 }, { "epoch": 2.806863459149137, "grad_norm": 4.581140995025635, "learning_rate": 4.0419219374658766e-05, "loss": 0.3414626121520996, "memory(GiB)": 70.5, "step": 65515, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.447176 }, { "epoch": 2.8070776744783856, "grad_norm": 4.055304050445557, "learning_rate": 4.04126143907795e-05, "loss": 0.5748294830322266, "memory(GiB)": 70.5, "step": 65520, "token_acc": 0.8710801393728222, "train_speed(iter/s)": 1.447187 }, { "epoch": 2.807291889807635, "grad_norm": 4.594996452331543, "learning_rate": 4.040600958058471e-05, "loss": 0.3870840072631836, "memory(GiB)": 70.5, "step": 65525, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.447189 }, { "epoch": 2.8075061051368837, "grad_norm": 2.659824848175049, "learning_rate": 4.039940494419407e-05, "loss": 0.30927157402038574, "memory(GiB)": 70.5, "step": 65530, "token_acc": 0.9475806451612904, "train_speed(iter/s)": 1.447189 }, { "epoch": 2.8077203204661325, "grad_norm": 2.9789791107177734, "learning_rate": 4.0392800481727224e-05, "loss": 0.13736239671707154, "memory(GiB)": 70.5, "step": 65535, "token_acc": 0.9556451612903226, "train_speed(iter/s)": 1.447188 }, { "epoch": 2.8079345357953818, "grad_norm": 5.3989644050598145, "learning_rate": 4.03861961933038e-05, "loss": 0.3124204635620117, "memory(GiB)": 70.5, "step": 65540, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.447193 }, { "epoch": 2.8081487511246306, "grad_norm": 6.953273773193359, "learning_rate": 4.037959207904346e-05, "loss": 0.40552325248718263, "memory(GiB)": 70.5, "step": 65545, "token_acc": 0.9036144578313253, "train_speed(iter/s)": 1.447196 }, { "epoch": 2.8083629664538794, "grad_norm": 3.4184439182281494, "learning_rate": 4.0372988139065824e-05, "loss": 0.36038596630096437, "memory(GiB)": 70.5, "step": 65550, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.447206 }, { "epoch": 2.8085771817831287, "grad_norm": 1.5506142377853394, "learning_rate": 4.036638437349054e-05, "loss": 0.2236245632171631, "memory(GiB)": 70.5, "step": 65555, "token_acc": 0.9573643410852714, "train_speed(iter/s)": 1.447205 }, { "epoch": 2.8087913971123775, "grad_norm": 0.597811222076416, "learning_rate": 4.035978078243725e-05, "loss": 0.4485891819000244, "memory(GiB)": 70.5, "step": 65560, "token_acc": 0.9013157894736842, "train_speed(iter/s)": 1.447211 }, { "epoch": 2.8090056124416263, "grad_norm": 5.578658103942871, "learning_rate": 4.0353177366025565e-05, "loss": 0.6790993690490723, "memory(GiB)": 70.5, "step": 65565, "token_acc": 0.865625, "train_speed(iter/s)": 1.447226 }, { "epoch": 2.8092198277708755, "grad_norm": 2.9286370277404785, "learning_rate": 4.0346574124375126e-05, "loss": 0.5922324657440186, "memory(GiB)": 70.5, "step": 65570, "token_acc": 0.8683274021352313, "train_speed(iter/s)": 1.447238 }, { "epoch": 2.8094340431001243, "grad_norm": 3.873889684677124, "learning_rate": 4.033997105760555e-05, "loss": 0.4971031665802002, "memory(GiB)": 70.5, "step": 65575, "token_acc": 0.8909774436090225, "train_speed(iter/s)": 1.447239 }, { "epoch": 2.809648258429373, "grad_norm": 5.785956382751465, "learning_rate": 4.0333368165836456e-05, "loss": 0.6018098354339599, "memory(GiB)": 70.5, "step": 65580, "token_acc": 0.8661710037174721, "train_speed(iter/s)": 1.447244 }, { "epoch": 2.8098624737586224, "grad_norm": 3.2937934398651123, "learning_rate": 4.032676544918747e-05, "loss": 0.5646754264831543, "memory(GiB)": 70.5, "step": 65585, "token_acc": 0.8957528957528957, "train_speed(iter/s)": 1.447277 }, { "epoch": 2.8100766890878712, "grad_norm": 1.1718324422836304, "learning_rate": 4.0320162907778196e-05, "loss": 0.4056044101715088, "memory(GiB)": 70.5, "step": 65590, "token_acc": 0.9226190476190477, "train_speed(iter/s)": 1.447295 }, { "epoch": 2.81029090441712, "grad_norm": 3.739021062850952, "learning_rate": 4.031356054172826e-05, "loss": 0.450758171081543, "memory(GiB)": 70.5, "step": 65595, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.447311 }, { "epoch": 2.8105051197463693, "grad_norm": 3.4903602600097656, "learning_rate": 4.0306958351157245e-05, "loss": 0.38242230415344236, "memory(GiB)": 70.5, "step": 65600, "token_acc": 0.9320754716981132, "train_speed(iter/s)": 1.447317 }, { "epoch": 2.810719335075618, "grad_norm": 3.754523515701294, "learning_rate": 4.0300356336184786e-05, "loss": 0.36229040622711184, "memory(GiB)": 70.5, "step": 65605, "token_acc": 0.9196891191709845, "train_speed(iter/s)": 1.447317 }, { "epoch": 2.810933550404867, "grad_norm": 1.2235926389694214, "learning_rate": 4.029375449693047e-05, "loss": 0.28717315196990967, "memory(GiB)": 70.5, "step": 65610, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.447321 }, { "epoch": 2.811147765734116, "grad_norm": 2.807979106903076, "learning_rate": 4.02871528335139e-05, "loss": 0.47182493209838866, "memory(GiB)": 70.5, "step": 65615, "token_acc": 0.8893617021276595, "train_speed(iter/s)": 1.447329 }, { "epoch": 2.811361981063365, "grad_norm": 4.286454200744629, "learning_rate": 4.028055134605467e-05, "loss": 0.3874686241149902, "memory(GiB)": 70.5, "step": 65620, "token_acc": 0.9039735099337748, "train_speed(iter/s)": 1.447338 }, { "epoch": 2.811576196392614, "grad_norm": 4.29857063293457, "learning_rate": 4.0273950034672356e-05, "loss": 0.6466447830200195, "memory(GiB)": 70.5, "step": 65625, "token_acc": 0.8669354838709677, "train_speed(iter/s)": 1.447365 }, { "epoch": 2.811790411721863, "grad_norm": 4.272668361663818, "learning_rate": 4.026734889948657e-05, "loss": 0.4003013610839844, "memory(GiB)": 70.5, "step": 65630, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.447377 }, { "epoch": 2.812004627051112, "grad_norm": 2.714235544204712, "learning_rate": 4.02607479406169e-05, "loss": 0.5596010684967041, "memory(GiB)": 70.5, "step": 65635, "token_acc": 0.875, "train_speed(iter/s)": 1.447379 }, { "epoch": 2.8122188423803607, "grad_norm": 0.9737520813941956, "learning_rate": 4.0254147158182895e-05, "loss": 0.221573543548584, "memory(GiB)": 70.5, "step": 65640, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.447382 }, { "epoch": 2.81243305770961, "grad_norm": 1.6398098468780518, "learning_rate": 4.024754655230417e-05, "loss": 0.6445670604705811, "memory(GiB)": 70.5, "step": 65645, "token_acc": 0.8862068965517241, "train_speed(iter/s)": 1.447396 }, { "epoch": 2.8126472730388588, "grad_norm": 4.333428859710693, "learning_rate": 4.024094612310028e-05, "loss": 0.4044950008392334, "memory(GiB)": 70.5, "step": 65650, "token_acc": 0.9103139013452914, "train_speed(iter/s)": 1.447399 }, { "epoch": 2.8128614883681076, "grad_norm": 1.1214721202850342, "learning_rate": 4.023434587069081e-05, "loss": 0.47982068061828614, "memory(GiB)": 70.5, "step": 65655, "token_acc": 0.91005291005291, "train_speed(iter/s)": 1.447399 }, { "epoch": 2.813075703697357, "grad_norm": 4.4294538497924805, "learning_rate": 4.0227745795195335e-05, "loss": 0.2524839401245117, "memory(GiB)": 70.5, "step": 65660, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.447417 }, { "epoch": 2.8132899190266056, "grad_norm": 3.8918893337249756, "learning_rate": 4.02211458967334e-05, "loss": 0.30101020336151124, "memory(GiB)": 70.5, "step": 65665, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.447422 }, { "epoch": 2.8135041343558544, "grad_norm": 2.224165439605713, "learning_rate": 4.021454617542457e-05, "loss": 0.39116358757019043, "memory(GiB)": 70.5, "step": 65670, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.447437 }, { "epoch": 2.8137183496851037, "grad_norm": 3.922574043273926, "learning_rate": 4.0207946631388426e-05, "loss": 0.3454158306121826, "memory(GiB)": 70.5, "step": 65675, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.447436 }, { "epoch": 2.8139325650143525, "grad_norm": 2.100076198577881, "learning_rate": 4.0201347264744524e-05, "loss": 0.4519326210021973, "memory(GiB)": 70.5, "step": 65680, "token_acc": 0.9233870967741935, "train_speed(iter/s)": 1.447444 }, { "epoch": 2.8141467803436013, "grad_norm": 3.2297000885009766, "learning_rate": 4.0194748075612396e-05, "loss": 0.5913656711578369, "memory(GiB)": 70.5, "step": 65685, "token_acc": 0.867109634551495, "train_speed(iter/s)": 1.447449 }, { "epoch": 2.8143609956728506, "grad_norm": 2.584153175354004, "learning_rate": 4.0188149064111615e-05, "loss": 0.5880422592163086, "memory(GiB)": 70.5, "step": 65690, "token_acc": 0.8590604026845637, "train_speed(iter/s)": 1.447448 }, { "epoch": 2.8145752110020994, "grad_norm": 3.975919723510742, "learning_rate": 4.018155023036171e-05, "loss": 0.5264403343200683, "memory(GiB)": 70.5, "step": 65695, "token_acc": 0.8696969696969697, "train_speed(iter/s)": 1.447459 }, { "epoch": 2.814789426331348, "grad_norm": 0.7207290530204773, "learning_rate": 4.017495157448224e-05, "loss": 0.19659755229949952, "memory(GiB)": 70.5, "step": 65700, "token_acc": 0.9518072289156626, "train_speed(iter/s)": 1.447462 }, { "epoch": 2.8150036416605975, "grad_norm": 1.0837005376815796, "learning_rate": 4.0168353096592735e-05, "loss": 0.39782114028930665, "memory(GiB)": 70.5, "step": 65705, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.447463 }, { "epoch": 2.8152178569898463, "grad_norm": 3.238957166671753, "learning_rate": 4.0161754796812736e-05, "loss": 0.3363406419754028, "memory(GiB)": 70.5, "step": 65710, "token_acc": 0.9171597633136095, "train_speed(iter/s)": 1.447471 }, { "epoch": 2.815432072319095, "grad_norm": 4.058709144592285, "learning_rate": 4.0155156675261785e-05, "loss": 0.5748291015625, "memory(GiB)": 70.5, "step": 65715, "token_acc": 0.8945783132530121, "train_speed(iter/s)": 1.44747 }, { "epoch": 2.8156462876483443, "grad_norm": 2.0098016262054443, "learning_rate": 4.01485587320594e-05, "loss": 0.3010366916656494, "memory(GiB)": 70.5, "step": 65720, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.44747 }, { "epoch": 2.815860502977593, "grad_norm": 2.082310914993286, "learning_rate": 4.014196096732511e-05, "loss": 0.13481926918029785, "memory(GiB)": 70.5, "step": 65725, "token_acc": 0.9745762711864406, "train_speed(iter/s)": 1.447476 }, { "epoch": 2.816074718306842, "grad_norm": 0.2884879410266876, "learning_rate": 4.0135363381178454e-05, "loss": 0.34470601081848146, "memory(GiB)": 70.5, "step": 65730, "token_acc": 0.930327868852459, "train_speed(iter/s)": 1.44748 }, { "epoch": 2.8162889336360912, "grad_norm": 2.4506723880767822, "learning_rate": 4.012876597373893e-05, "loss": 0.28392724990844725, "memory(GiB)": 70.5, "step": 65735, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.447481 }, { "epoch": 2.81650314896534, "grad_norm": 3.062225103378296, "learning_rate": 4.012216874512609e-05, "loss": 0.34068903923034666, "memory(GiB)": 70.5, "step": 65740, "token_acc": 0.9283387622149837, "train_speed(iter/s)": 1.447498 }, { "epoch": 2.816717364294589, "grad_norm": 3.15340518951416, "learning_rate": 4.0115571695459396e-05, "loss": 0.7546008110046387, "memory(GiB)": 70.5, "step": 65745, "token_acc": 0.8436363636363636, "train_speed(iter/s)": 1.447504 }, { "epoch": 2.816931579623838, "grad_norm": 3.9223599433898926, "learning_rate": 4.0108974824858425e-05, "loss": 0.15568189620971679, "memory(GiB)": 70.5, "step": 65750, "token_acc": 0.9610894941634242, "train_speed(iter/s)": 1.44751 }, { "epoch": 2.817145794953087, "grad_norm": 5.46969747543335, "learning_rate": 4.010237813344264e-05, "loss": 0.21836724281311035, "memory(GiB)": 70.5, "step": 65755, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.447517 }, { "epoch": 2.8173600102823357, "grad_norm": 0.17792794108390808, "learning_rate": 4.0095781621331563e-05, "loss": 0.19510886669158936, "memory(GiB)": 70.5, "step": 65760, "token_acc": 0.96, "train_speed(iter/s)": 1.447522 }, { "epoch": 2.817574225611585, "grad_norm": 3.976799726486206, "learning_rate": 4.0089185288644706e-05, "loss": 0.27987892627716066, "memory(GiB)": 70.5, "step": 65765, "token_acc": 0.9324894514767933, "train_speed(iter/s)": 1.447523 }, { "epoch": 2.817788440940834, "grad_norm": 3.1028332710266113, "learning_rate": 4.008258913550153e-05, "loss": 0.37985868453979493, "memory(GiB)": 70.5, "step": 65770, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.447519 }, { "epoch": 2.8180026562700826, "grad_norm": 2.071793556213379, "learning_rate": 4.0075993162021575e-05, "loss": 0.1918225646018982, "memory(GiB)": 70.5, "step": 65775, "token_acc": 0.9621212121212122, "train_speed(iter/s)": 1.447525 }, { "epoch": 2.818216871599332, "grad_norm": 2.459130048751831, "learning_rate": 4.006939736832431e-05, "loss": 0.5563003540039062, "memory(GiB)": 70.5, "step": 65780, "token_acc": 0.8853503184713376, "train_speed(iter/s)": 1.447525 }, { "epoch": 2.8184310869285807, "grad_norm": 5.064753532409668, "learning_rate": 4.006280175452922e-05, "loss": 0.25032715797424315, "memory(GiB)": 70.5, "step": 65785, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.447523 }, { "epoch": 2.8186453022578295, "grad_norm": 5.609072208404541, "learning_rate": 4.0056206320755806e-05, "loss": 0.3041250705718994, "memory(GiB)": 70.5, "step": 65790, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.447517 }, { "epoch": 2.8188595175870788, "grad_norm": 0.9684300422668457, "learning_rate": 4.0049611067123526e-05, "loss": 0.19255608320236206, "memory(GiB)": 70.5, "step": 65795, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.447522 }, { "epoch": 2.8190737329163276, "grad_norm": 2.077866554260254, "learning_rate": 4.00430159937519e-05, "loss": 0.452288818359375, "memory(GiB)": 70.5, "step": 65800, "token_acc": 0.8996539792387543, "train_speed(iter/s)": 1.447525 }, { "epoch": 2.8192879482455764, "grad_norm": 3.8816754817962646, "learning_rate": 4.003642110076037e-05, "loss": 0.391249418258667, "memory(GiB)": 70.5, "step": 65805, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.447524 }, { "epoch": 2.8195021635748256, "grad_norm": 0.10942092537879944, "learning_rate": 4.002982638826841e-05, "loss": 0.204693603515625, "memory(GiB)": 70.5, "step": 65810, "token_acc": 0.9553903345724907, "train_speed(iter/s)": 1.447517 }, { "epoch": 2.8197163789040744, "grad_norm": 1.4451452493667603, "learning_rate": 4.0023231856395505e-05, "loss": 0.17195792198181153, "memory(GiB)": 70.5, "step": 65815, "token_acc": 0.9690402476780186, "train_speed(iter/s)": 1.447529 }, { "epoch": 2.8199305942333233, "grad_norm": 3.324211359024048, "learning_rate": 4.00166375052611e-05, "loss": 0.2927267074584961, "memory(GiB)": 70.5, "step": 65820, "token_acc": 0.941358024691358, "train_speed(iter/s)": 1.447531 }, { "epoch": 2.8201448095625725, "grad_norm": 0.3051203489303589, "learning_rate": 4.001004333498469e-05, "loss": 0.3352349281311035, "memory(GiB)": 70.5, "step": 65825, "token_acc": 0.9276595744680851, "train_speed(iter/s)": 1.447537 }, { "epoch": 2.8203590248918213, "grad_norm": 2.9657983779907227, "learning_rate": 4.0003449345685704e-05, "loss": 0.36175999641418455, "memory(GiB)": 70.5, "step": 65830, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.447547 }, { "epoch": 2.82057324022107, "grad_norm": 6.318264961242676, "learning_rate": 3.999685553748362e-05, "loss": 0.49136719703674314, "memory(GiB)": 70.5, "step": 65835, "token_acc": 0.8927536231884058, "train_speed(iter/s)": 1.447557 }, { "epoch": 2.8207874555503194, "grad_norm": 5.312129974365234, "learning_rate": 3.9990261910497876e-05, "loss": 0.4423810958862305, "memory(GiB)": 70.5, "step": 65840, "token_acc": 0.887459807073955, "train_speed(iter/s)": 1.447562 }, { "epoch": 2.821001670879568, "grad_norm": 3.6515443325042725, "learning_rate": 3.9983668464847935e-05, "loss": 0.5017635345458984, "memory(GiB)": 70.5, "step": 65845, "token_acc": 0.8763250883392226, "train_speed(iter/s)": 1.447568 }, { "epoch": 2.821215886208817, "grad_norm": 4.767198085784912, "learning_rate": 3.9977075200653234e-05, "loss": 0.6141701698303222, "memory(GiB)": 70.5, "step": 65850, "token_acc": 0.8867924528301887, "train_speed(iter/s)": 1.447579 }, { "epoch": 2.8214301015380663, "grad_norm": 2.3564260005950928, "learning_rate": 3.997048211803321e-05, "loss": 0.23367273807525635, "memory(GiB)": 70.5, "step": 65855, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.447577 }, { "epoch": 2.821644316867315, "grad_norm": 4.073162078857422, "learning_rate": 3.996388921710732e-05, "loss": 0.5777758598327637, "memory(GiB)": 70.5, "step": 65860, "token_acc": 0.894927536231884, "train_speed(iter/s)": 1.44758 }, { "epoch": 2.821858532196564, "grad_norm": 2.631847858428955, "learning_rate": 3.995729649799499e-05, "loss": 0.43934950828552244, "memory(GiB)": 70.5, "step": 65865, "token_acc": 0.9223880597014925, "train_speed(iter/s)": 1.44758 }, { "epoch": 2.822072747525813, "grad_norm": 3.551302433013916, "learning_rate": 3.995070396081565e-05, "loss": 0.3345721483230591, "memory(GiB)": 70.5, "step": 65870, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.44759 }, { "epoch": 2.822286962855062, "grad_norm": 4.0038042068481445, "learning_rate": 3.994411160568874e-05, "loss": 0.3705452442169189, "memory(GiB)": 70.5, "step": 65875, "token_acc": 0.9291338582677166, "train_speed(iter/s)": 1.447607 }, { "epoch": 2.822501178184311, "grad_norm": 2.4785895347595215, "learning_rate": 3.993751943273367e-05, "loss": 0.31236817836761477, "memory(GiB)": 70.5, "step": 65880, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.447604 }, { "epoch": 2.82271539351356, "grad_norm": 3.6710078716278076, "learning_rate": 3.9930927442069885e-05, "loss": 0.4261664390563965, "memory(GiB)": 70.5, "step": 65885, "token_acc": 0.9163346613545816, "train_speed(iter/s)": 1.447603 }, { "epoch": 2.822929608842809, "grad_norm": 2.3866469860076904, "learning_rate": 3.99243356338168e-05, "loss": 0.33325586318969724, "memory(GiB)": 70.5, "step": 65890, "token_acc": 0.9481481481481482, "train_speed(iter/s)": 1.447616 }, { "epoch": 2.8231438241720577, "grad_norm": 1.994363784790039, "learning_rate": 3.9917744008093806e-05, "loss": 0.2088099479675293, "memory(GiB)": 70.5, "step": 65895, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.447612 }, { "epoch": 2.823358039501307, "grad_norm": 3.0112292766571045, "learning_rate": 3.991115256502034e-05, "loss": 0.40438151359558105, "memory(GiB)": 70.5, "step": 65900, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.44761 }, { "epoch": 2.8235722548305557, "grad_norm": 3.104426860809326, "learning_rate": 3.9904561304715824e-05, "loss": 0.26966798305511475, "memory(GiB)": 70.5, "step": 65905, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.447612 }, { "epoch": 2.8237864701598046, "grad_norm": 4.517772197723389, "learning_rate": 3.989797022729966e-05, "loss": 0.375781774520874, "memory(GiB)": 70.5, "step": 65910, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.447616 }, { "epoch": 2.824000685489054, "grad_norm": 3.7424428462982178, "learning_rate": 3.9891379332891224e-05, "loss": 0.23707613945007325, "memory(GiB)": 70.5, "step": 65915, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.447628 }, { "epoch": 2.8242149008183026, "grad_norm": 2.136129856109619, "learning_rate": 3.9884788621609936e-05, "loss": 0.4658190250396729, "memory(GiB)": 70.5, "step": 65920, "token_acc": 0.9096774193548387, "train_speed(iter/s)": 1.447623 }, { "epoch": 2.8244291161475514, "grad_norm": 3.126932144165039, "learning_rate": 3.98781980935752e-05, "loss": 0.4726677417755127, "memory(GiB)": 70.5, "step": 65925, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.44763 }, { "epoch": 2.8246433314768007, "grad_norm": 2.0444881916046143, "learning_rate": 3.9871607748906395e-05, "loss": 0.4562983989715576, "memory(GiB)": 70.5, "step": 65930, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.447633 }, { "epoch": 2.8248575468060495, "grad_norm": 3.3959834575653076, "learning_rate": 3.9865017587722916e-05, "loss": 0.2507554292678833, "memory(GiB)": 70.5, "step": 65935, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.447639 }, { "epoch": 2.8250717621352983, "grad_norm": 4.223495006561279, "learning_rate": 3.985842761014414e-05, "loss": 0.44249815940856935, "memory(GiB)": 70.5, "step": 65940, "token_acc": 0.9050632911392406, "train_speed(iter/s)": 1.447645 }, { "epoch": 2.8252859774645476, "grad_norm": 2.978490114212036, "learning_rate": 3.9851837816289485e-05, "loss": 0.2162032127380371, "memory(GiB)": 70.5, "step": 65945, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.447652 }, { "epoch": 2.8255001927937964, "grad_norm": 0.5683831572532654, "learning_rate": 3.984524820627829e-05, "loss": 0.16647781133651735, "memory(GiB)": 70.5, "step": 65950, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.447657 }, { "epoch": 2.825714408123045, "grad_norm": 2.81089186668396, "learning_rate": 3.983865878022995e-05, "loss": 0.33897483348846436, "memory(GiB)": 70.5, "step": 65955, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.447678 }, { "epoch": 2.8259286234522945, "grad_norm": 3.276883602142334, "learning_rate": 3.983206953826385e-05, "loss": 0.3765129327774048, "memory(GiB)": 70.5, "step": 65960, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.447679 }, { "epoch": 2.8261428387815433, "grad_norm": 2.574774742126465, "learning_rate": 3.982548048049935e-05, "loss": 0.5041276931762695, "memory(GiB)": 70.5, "step": 65965, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.447699 }, { "epoch": 2.826357054110792, "grad_norm": 4.412144184112549, "learning_rate": 3.981889160705579e-05, "loss": 0.3333552122116089, "memory(GiB)": 70.5, "step": 65970, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.447705 }, { "epoch": 2.8265712694400413, "grad_norm": 4.568398952484131, "learning_rate": 3.981230291805257e-05, "loss": 0.5240084171295166, "memory(GiB)": 70.5, "step": 65975, "token_acc": 0.883495145631068, "train_speed(iter/s)": 1.447713 }, { "epoch": 2.82678548476929, "grad_norm": 0.4189091622829437, "learning_rate": 3.980571441360904e-05, "loss": 0.4849385738372803, "memory(GiB)": 70.5, "step": 65980, "token_acc": 0.8900343642611683, "train_speed(iter/s)": 1.447713 }, { "epoch": 2.826999700098539, "grad_norm": 4.8855881690979, "learning_rate": 3.979912609384456e-05, "loss": 0.6465899467468261, "memory(GiB)": 70.5, "step": 65985, "token_acc": 0.8641975308641975, "train_speed(iter/s)": 1.447711 }, { "epoch": 2.827213915427788, "grad_norm": 4.36391544342041, "learning_rate": 3.979253795887849e-05, "loss": 0.41993513107299807, "memory(GiB)": 70.5, "step": 65990, "token_acc": 0.8909090909090909, "train_speed(iter/s)": 1.447721 }, { "epoch": 2.827428130757037, "grad_norm": 1.5792783498764038, "learning_rate": 3.978595000883017e-05, "loss": 0.2016439914703369, "memory(GiB)": 70.5, "step": 65995, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.447719 }, { "epoch": 2.827642346086286, "grad_norm": 2.678069591522217, "learning_rate": 3.977936224381893e-05, "loss": 0.3228564500808716, "memory(GiB)": 70.5, "step": 66000, "token_acc": 0.9305019305019305, "train_speed(iter/s)": 1.447723 }, { "epoch": 2.827642346086286, "eval_loss": 2.5496714115142822, "eval_runtime": 13.3104, "eval_samples_per_second": 7.513, "eval_steps_per_second": 7.513, "eval_token_acc": 0.4623955431754875, "step": 66000 }, { "epoch": 2.827856561415535, "grad_norm": 2.6306920051574707, "learning_rate": 3.9772774663964145e-05, "loss": 0.36658999919891355, "memory(GiB)": 70.5, "step": 66005, "token_acc": 0.6115384615384616, "train_speed(iter/s)": 1.447268 }, { "epoch": 2.828070776744784, "grad_norm": 2.874450922012329, "learning_rate": 3.9766187269385144e-05, "loss": 0.2682220220565796, "memory(GiB)": 70.5, "step": 66010, "token_acc": 0.9470198675496688, "train_speed(iter/s)": 1.447276 }, { "epoch": 2.8282849920740327, "grad_norm": 3.26786732673645, "learning_rate": 3.9759600060201245e-05, "loss": 0.42403397560119627, "memory(GiB)": 70.5, "step": 66015, "token_acc": 0.9027355623100304, "train_speed(iter/s)": 1.447281 }, { "epoch": 2.828499207403282, "grad_norm": 1.8569695949554443, "learning_rate": 3.975301303653181e-05, "loss": 0.4194043636322021, "memory(GiB)": 70.5, "step": 66020, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.447283 }, { "epoch": 2.828713422732531, "grad_norm": 3.356198310852051, "learning_rate": 3.974642619849615e-05, "loss": 0.31343927383422854, "memory(GiB)": 70.5, "step": 66025, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.447286 }, { "epoch": 2.8289276380617796, "grad_norm": 1.2363712787628174, "learning_rate": 3.9739839546213596e-05, "loss": 0.42557196617126464, "memory(GiB)": 70.5, "step": 66030, "token_acc": 0.9115646258503401, "train_speed(iter/s)": 1.4473 }, { "epoch": 2.829141853391029, "grad_norm": 2.0232746601104736, "learning_rate": 3.9733253079803486e-05, "loss": 0.3307847023010254, "memory(GiB)": 70.5, "step": 66035, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.447308 }, { "epoch": 2.8293560687202777, "grad_norm": 5.024719715118408, "learning_rate": 3.9726666799385095e-05, "loss": 0.5698225975036622, "memory(GiB)": 70.5, "step": 66040, "token_acc": 0.8741935483870967, "train_speed(iter/s)": 1.447304 }, { "epoch": 2.8295702840495265, "grad_norm": 7.053027153015137, "learning_rate": 3.972008070507779e-05, "loss": 0.4823485851287842, "memory(GiB)": 70.5, "step": 66045, "token_acc": 0.907563025210084, "train_speed(iter/s)": 1.447305 }, { "epoch": 2.8297844993787757, "grad_norm": 0.7565944194793701, "learning_rate": 3.971349479700088e-05, "loss": 0.4803595066070557, "memory(GiB)": 70.5, "step": 66050, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.447307 }, { "epoch": 2.8299987147080246, "grad_norm": 4.279663562774658, "learning_rate": 3.970690907527366e-05, "loss": 0.5062711715698243, "memory(GiB)": 70.5, "step": 66055, "token_acc": 0.8953488372093024, "train_speed(iter/s)": 1.447318 }, { "epoch": 2.8302129300372734, "grad_norm": 2.779945135116577, "learning_rate": 3.970032354001542e-05, "loss": 0.2729729175567627, "memory(GiB)": 70.5, "step": 66060, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.447318 }, { "epoch": 2.8304271453665226, "grad_norm": 4.500383377075195, "learning_rate": 3.9693738191345495e-05, "loss": 0.386583137512207, "memory(GiB)": 70.5, "step": 66065, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.447313 }, { "epoch": 2.8306413606957714, "grad_norm": 2.1126224994659424, "learning_rate": 3.968715302938317e-05, "loss": 0.2627983808517456, "memory(GiB)": 70.5, "step": 66070, "token_acc": 0.9291553133514986, "train_speed(iter/s)": 1.447318 }, { "epoch": 2.8308555760250202, "grad_norm": 6.821784019470215, "learning_rate": 3.9680568054247744e-05, "loss": 0.48012189865112304, "memory(GiB)": 70.5, "step": 66075, "token_acc": 0.8992537313432836, "train_speed(iter/s)": 1.447316 }, { "epoch": 2.8310697913542695, "grad_norm": 4.382533550262451, "learning_rate": 3.9673983266058504e-05, "loss": 0.3308417797088623, "memory(GiB)": 70.5, "step": 66080, "token_acc": 0.9139344262295082, "train_speed(iter/s)": 1.447315 }, { "epoch": 2.8312840066835183, "grad_norm": 3.9530723094940186, "learning_rate": 3.9667398664934735e-05, "loss": 0.4421966552734375, "memory(GiB)": 70.5, "step": 66085, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.447325 }, { "epoch": 2.831498222012767, "grad_norm": 3.501617670059204, "learning_rate": 3.966081425099575e-05, "loss": 0.16654151678085327, "memory(GiB)": 70.5, "step": 66090, "token_acc": 0.952, "train_speed(iter/s)": 1.447325 }, { "epoch": 2.8317124373420164, "grad_norm": 2.8851287364959717, "learning_rate": 3.96542300243608e-05, "loss": 0.2382145881652832, "memory(GiB)": 70.5, "step": 66095, "token_acc": 0.9359756097560976, "train_speed(iter/s)": 1.447325 }, { "epoch": 2.831926652671265, "grad_norm": 2.281759023666382, "learning_rate": 3.9647645985149184e-05, "loss": 0.5194791793823242, "memory(GiB)": 70.5, "step": 66100, "token_acc": 0.8827361563517915, "train_speed(iter/s)": 1.447325 }, { "epoch": 2.832140868000514, "grad_norm": 3.594564914703369, "learning_rate": 3.964106213348017e-05, "loss": 0.213730525970459, "memory(GiB)": 70.5, "step": 66105, "token_acc": 0.9612903225806452, "train_speed(iter/s)": 1.447326 }, { "epoch": 2.8323550833297633, "grad_norm": 2.6096749305725098, "learning_rate": 3.963447846947304e-05, "loss": 0.5543881893157959, "memory(GiB)": 70.5, "step": 66110, "token_acc": 0.8745387453874539, "train_speed(iter/s)": 1.447345 }, { "epoch": 2.832569298659012, "grad_norm": 3.377002239227295, "learning_rate": 3.962789499324703e-05, "loss": 0.4318744659423828, "memory(GiB)": 70.5, "step": 66115, "token_acc": 0.9016949152542373, "train_speed(iter/s)": 1.447349 }, { "epoch": 2.832783513988261, "grad_norm": 4.131965637207031, "learning_rate": 3.962131170492145e-05, "loss": 0.6607012748718262, "memory(GiB)": 70.5, "step": 66120, "token_acc": 0.8828571428571429, "train_speed(iter/s)": 1.447372 }, { "epoch": 2.83299772931751, "grad_norm": 0.36935123801231384, "learning_rate": 3.961472860461555e-05, "loss": 0.4634706974029541, "memory(GiB)": 70.5, "step": 66125, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.44738 }, { "epoch": 2.833211944646759, "grad_norm": 4.314681529998779, "learning_rate": 3.9608145692448575e-05, "loss": 0.5434744834899903, "memory(GiB)": 70.5, "step": 66130, "token_acc": 0.870253164556962, "train_speed(iter/s)": 1.447391 }, { "epoch": 2.8334261599760078, "grad_norm": 2.2219982147216797, "learning_rate": 3.9601562968539796e-05, "loss": 0.5683823108673096, "memory(GiB)": 70.5, "step": 66135, "token_acc": 0.8605341246290801, "train_speed(iter/s)": 1.447399 }, { "epoch": 2.833640375305257, "grad_norm": 2.85251522064209, "learning_rate": 3.959498043300846e-05, "loss": 0.29082987308502195, "memory(GiB)": 70.5, "step": 66140, "token_acc": 0.9076433121019108, "train_speed(iter/s)": 1.447407 }, { "epoch": 2.833854590634506, "grad_norm": 3.84812331199646, "learning_rate": 3.958839808597381e-05, "loss": 0.6959355354309082, "memory(GiB)": 70.5, "step": 66145, "token_acc": 0.8737864077669902, "train_speed(iter/s)": 1.447407 }, { "epoch": 2.8340688059637547, "grad_norm": 5.292137145996094, "learning_rate": 3.95818159275551e-05, "loss": 0.321564769744873, "memory(GiB)": 70.5, "step": 66150, "token_acc": 0.9291666666666667, "train_speed(iter/s)": 1.447409 }, { "epoch": 2.834283021293004, "grad_norm": 5.454011917114258, "learning_rate": 3.957523395787156e-05, "loss": 0.5491637706756591, "memory(GiB)": 70.5, "step": 66155, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.447409 }, { "epoch": 2.8344972366222527, "grad_norm": 2.131324291229248, "learning_rate": 3.956865217704244e-05, "loss": 0.43274879455566406, "memory(GiB)": 70.5, "step": 66160, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.447419 }, { "epoch": 2.8347114519515015, "grad_norm": 3.559731960296631, "learning_rate": 3.956207058518697e-05, "loss": 0.34083631038665774, "memory(GiB)": 70.5, "step": 66165, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.447434 }, { "epoch": 2.834925667280751, "grad_norm": 4.544974327087402, "learning_rate": 3.955548918242438e-05, "loss": 0.44176297187805175, "memory(GiB)": 70.5, "step": 66170, "token_acc": 0.905829596412556, "train_speed(iter/s)": 1.447437 }, { "epoch": 2.8351398826099996, "grad_norm": 5.1930718421936035, "learning_rate": 3.954890796887391e-05, "loss": 0.4043581962585449, "memory(GiB)": 70.5, "step": 66175, "token_acc": 0.9010238907849829, "train_speed(iter/s)": 1.44745 }, { "epoch": 2.8353540979392484, "grad_norm": 4.500527858734131, "learning_rate": 3.9542326944654775e-05, "loss": 0.4125670909881592, "memory(GiB)": 70.5, "step": 66180, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.447448 }, { "epoch": 2.8355683132684977, "grad_norm": 2.3860344886779785, "learning_rate": 3.953574610988619e-05, "loss": 0.3922131538391113, "memory(GiB)": 70.5, "step": 66185, "token_acc": 0.9252873563218391, "train_speed(iter/s)": 1.44746 }, { "epoch": 2.8357825285977465, "grad_norm": 1.220852017402649, "learning_rate": 3.952916546468737e-05, "loss": 0.4194629669189453, "memory(GiB)": 70.5, "step": 66190, "token_acc": 0.9071207430340558, "train_speed(iter/s)": 1.447469 }, { "epoch": 2.8359967439269953, "grad_norm": 5.437346935272217, "learning_rate": 3.9522585009177554e-05, "loss": 0.41033115386962893, "memory(GiB)": 70.5, "step": 66195, "token_acc": 0.913312693498452, "train_speed(iter/s)": 1.447466 }, { "epoch": 2.8362109592562446, "grad_norm": 1.6807118654251099, "learning_rate": 3.951600474347594e-05, "loss": 0.19144123792648315, "memory(GiB)": 70.5, "step": 66200, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.8364251745854934, "grad_norm": 5.078380584716797, "learning_rate": 3.950942466770173e-05, "loss": 0.313324499130249, "memory(GiB)": 70.5, "step": 66205, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.836639389914742, "grad_norm": 1.9306877851486206, "learning_rate": 3.950284478197414e-05, "loss": 0.36316328048706054, "memory(GiB)": 70.5, "step": 66210, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.8368536052439914, "grad_norm": 3.781893014907837, "learning_rate": 3.9496265086412364e-05, "loss": 0.2557427644729614, "memory(GiB)": 70.5, "step": 66215, "token_acc": 0.9392857142857143, "train_speed(iter/s)": 1.44746 }, { "epoch": 2.8370678205732403, "grad_norm": 2.28578519821167, "learning_rate": 3.948968558113559e-05, "loss": 0.2455204963684082, "memory(GiB)": 70.5, "step": 66220, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.447463 }, { "epoch": 2.837282035902489, "grad_norm": 5.485607147216797, "learning_rate": 3.9483106266263036e-05, "loss": 0.3786125659942627, "memory(GiB)": 70.5, "step": 66225, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.447463 }, { "epoch": 2.8374962512317383, "grad_norm": 1.953791856765747, "learning_rate": 3.9476527141913866e-05, "loss": 0.32944183349609374, "memory(GiB)": 70.5, "step": 66230, "token_acc": 0.9347181008902077, "train_speed(iter/s)": 1.447461 }, { "epoch": 2.837710466560987, "grad_norm": 1.788444995880127, "learning_rate": 3.946994820820728e-05, "loss": 0.1590221643447876, "memory(GiB)": 70.5, "step": 66235, "token_acc": 0.9537815126050421, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.837924681890236, "grad_norm": 1.5950638055801392, "learning_rate": 3.9463369465262466e-05, "loss": 0.18363200426101683, "memory(GiB)": 70.5, "step": 66240, "token_acc": 0.9490196078431372, "train_speed(iter/s)": 1.447472 }, { "epoch": 2.838138897219485, "grad_norm": 2.186486005783081, "learning_rate": 3.945679091319859e-05, "loss": 0.293864107131958, "memory(GiB)": 70.5, "step": 66245, "token_acc": 0.9395973154362416, "train_speed(iter/s)": 1.447471 }, { "epoch": 2.838353112548734, "grad_norm": 2.1681466102600098, "learning_rate": 3.9450212552134845e-05, "loss": 0.092784583568573, "memory(GiB)": 70.5, "step": 66250, "token_acc": 0.9617021276595744, "train_speed(iter/s)": 1.447472 }, { "epoch": 2.838567327877983, "grad_norm": 7.381155490875244, "learning_rate": 3.9443634382190396e-05, "loss": 0.5260069847106934, "memory(GiB)": 70.5, "step": 66255, "token_acc": 0.8842443729903537, "train_speed(iter/s)": 1.447487 }, { "epoch": 2.838781543207232, "grad_norm": 3.742225170135498, "learning_rate": 3.9437056403484404e-05, "loss": 0.2531268119812012, "memory(GiB)": 70.5, "step": 66260, "token_acc": 0.9480122324159022, "train_speed(iter/s)": 1.447503 }, { "epoch": 2.838995758536481, "grad_norm": 3.5489211082458496, "learning_rate": 3.9430478616136036e-05, "loss": 0.42949490547180175, "memory(GiB)": 70.5, "step": 66265, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.447504 }, { "epoch": 2.8392099738657297, "grad_norm": 3.384443521499634, "learning_rate": 3.9423901020264474e-05, "loss": 0.4859020233154297, "memory(GiB)": 70.5, "step": 66270, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.447514 }, { "epoch": 2.839424189194979, "grad_norm": 2.9135968685150146, "learning_rate": 3.9417323615988864e-05, "loss": 0.37205982208251953, "memory(GiB)": 70.5, "step": 66275, "token_acc": 0.9052287581699346, "train_speed(iter/s)": 1.447522 }, { "epoch": 2.8396384045242278, "grad_norm": 3.752200126647949, "learning_rate": 3.941074640342838e-05, "loss": 0.3633903741836548, "memory(GiB)": 70.5, "step": 66280, "token_acc": 0.918429003021148, "train_speed(iter/s)": 1.44753 }, { "epoch": 2.8398526198534766, "grad_norm": 3.668041706085205, "learning_rate": 3.940416938270215e-05, "loss": 0.5304598331451416, "memory(GiB)": 70.5, "step": 66285, "token_acc": 0.8862745098039215, "train_speed(iter/s)": 1.44754 }, { "epoch": 2.840066835182726, "grad_norm": 2.9501171112060547, "learning_rate": 3.939759255392932e-05, "loss": 0.3508331298828125, "memory(GiB)": 70.5, "step": 66290, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.447541 }, { "epoch": 2.8402810505119747, "grad_norm": 3.3858091831207275, "learning_rate": 3.939101591722906e-05, "loss": 0.29598040580749513, "memory(GiB)": 70.5, "step": 66295, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.447549 }, { "epoch": 2.8404952658412235, "grad_norm": 2.5204687118530273, "learning_rate": 3.93844394727205e-05, "loss": 0.6426472663879395, "memory(GiB)": 70.5, "step": 66300, "token_acc": 0.8524590163934426, "train_speed(iter/s)": 1.447567 }, { "epoch": 2.8407094811704727, "grad_norm": 3.4478511810302734, "learning_rate": 3.937786322052276e-05, "loss": 0.4959681034088135, "memory(GiB)": 70.5, "step": 66305, "token_acc": 0.9026845637583892, "train_speed(iter/s)": 1.447572 }, { "epoch": 2.8409236964997215, "grad_norm": 1.9132519960403442, "learning_rate": 3.937128716075501e-05, "loss": 0.3019278526306152, "memory(GiB)": 70.5, "step": 66310, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.447583 }, { "epoch": 2.8411379118289704, "grad_norm": 9.9467134475708, "learning_rate": 3.936471129353635e-05, "loss": 0.35906038284301756, "memory(GiB)": 70.5, "step": 66315, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.447587 }, { "epoch": 2.8413521271582196, "grad_norm": 3.807467222213745, "learning_rate": 3.935813561898593e-05, "loss": 0.453034496307373, "memory(GiB)": 70.5, "step": 66320, "token_acc": 0.910941475826972, "train_speed(iter/s)": 1.447591 }, { "epoch": 2.8415663424874684, "grad_norm": 0.36075660586357117, "learning_rate": 3.935156013722287e-05, "loss": 0.5290376663208007, "memory(GiB)": 70.5, "step": 66325, "token_acc": 0.8996763754045307, "train_speed(iter/s)": 1.447608 }, { "epoch": 2.8417805578167172, "grad_norm": 1.925873041152954, "learning_rate": 3.934498484836627e-05, "loss": 0.6895061016082764, "memory(GiB)": 70.5, "step": 66330, "token_acc": 0.8586206896551725, "train_speed(iter/s)": 1.447612 }, { "epoch": 2.8419947731459665, "grad_norm": 2.4360415935516357, "learning_rate": 3.933840975253527e-05, "loss": 0.6327857971191406, "memory(GiB)": 70.5, "step": 66335, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.447629 }, { "epoch": 2.8422089884752153, "grad_norm": 4.2464470863342285, "learning_rate": 3.933183484984898e-05, "loss": 0.47487335205078124, "memory(GiB)": 70.5, "step": 66340, "token_acc": 0.9087719298245615, "train_speed(iter/s)": 1.447638 }, { "epoch": 2.842423203804464, "grad_norm": 1.8609896898269653, "learning_rate": 3.932526014042652e-05, "loss": 0.3883563756942749, "memory(GiB)": 70.5, "step": 66345, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.447638 }, { "epoch": 2.8426374191337134, "grad_norm": 4.444515705108643, "learning_rate": 3.9318685624386975e-05, "loss": 0.50501708984375, "memory(GiB)": 70.5, "step": 66350, "token_acc": 0.8847352024922118, "train_speed(iter/s)": 1.447638 }, { "epoch": 2.842851634462962, "grad_norm": 3.924525737762451, "learning_rate": 3.931211130184947e-05, "loss": 0.35718085765838625, "memory(GiB)": 70.5, "step": 66355, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.447638 }, { "epoch": 2.843065849792211, "grad_norm": 2.4954006671905518, "learning_rate": 3.9305537172933085e-05, "loss": 0.3885574579238892, "memory(GiB)": 70.5, "step": 66360, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.447638 }, { "epoch": 2.8432800651214603, "grad_norm": 3.330033779144287, "learning_rate": 3.9298963237756934e-05, "loss": 0.3229584455490112, "memory(GiB)": 70.5, "step": 66365, "token_acc": 0.9276018099547512, "train_speed(iter/s)": 1.447644 }, { "epoch": 2.843494280450709, "grad_norm": 2.5279762744903564, "learning_rate": 3.92923894964401e-05, "loss": 0.3371935606002808, "memory(GiB)": 70.5, "step": 66370, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.447653 }, { "epoch": 2.843708495779958, "grad_norm": 1.1592239141464233, "learning_rate": 3.9285815949101675e-05, "loss": 0.28736541271209715, "memory(GiB)": 70.5, "step": 66375, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.447656 }, { "epoch": 2.843922711109207, "grad_norm": 3.092885732650757, "learning_rate": 3.9279242595860746e-05, "loss": 0.3666357517242432, "memory(GiB)": 70.5, "step": 66380, "token_acc": 0.9179331306990881, "train_speed(iter/s)": 1.447665 }, { "epoch": 2.844136926438456, "grad_norm": 2.415493965148926, "learning_rate": 3.9272669436836395e-05, "loss": 0.4707963943481445, "memory(GiB)": 70.5, "step": 66385, "token_acc": 0.9049429657794676, "train_speed(iter/s)": 1.447667 }, { "epoch": 2.8443511417677048, "grad_norm": 1.9600204229354858, "learning_rate": 3.9266096472147694e-05, "loss": 0.26213314533233645, "memory(GiB)": 70.5, "step": 66390, "token_acc": 0.9343434343434344, "train_speed(iter/s)": 1.447672 }, { "epoch": 2.844565357096954, "grad_norm": 0.5670682787895203, "learning_rate": 3.925952370191373e-05, "loss": 0.43649864196777344, "memory(GiB)": 70.5, "step": 66395, "token_acc": 0.8970588235294118, "train_speed(iter/s)": 1.447675 }, { "epoch": 2.844779572426203, "grad_norm": 6.592013835906982, "learning_rate": 3.9252951126253565e-05, "loss": 0.4264083385467529, "memory(GiB)": 70.5, "step": 66400, "token_acc": 0.9063670411985019, "train_speed(iter/s)": 1.447677 }, { "epoch": 2.8449937877554516, "grad_norm": 2.0301742553710938, "learning_rate": 3.9246378745286276e-05, "loss": 0.5412710189819336, "memory(GiB)": 70.5, "step": 66405, "token_acc": 0.8813559322033898, "train_speed(iter/s)": 1.447711 }, { "epoch": 2.845208003084701, "grad_norm": 3.61047101020813, "learning_rate": 3.923980655913091e-05, "loss": 0.3454113960266113, "memory(GiB)": 70.5, "step": 66410, "token_acc": 0.9169675090252708, "train_speed(iter/s)": 1.447719 }, { "epoch": 2.8454222184139497, "grad_norm": 3.7598228454589844, "learning_rate": 3.923323456790656e-05, "loss": 0.5636613368988037, "memory(GiB)": 70.5, "step": 66415, "token_acc": 0.8636363636363636, "train_speed(iter/s)": 1.447721 }, { "epoch": 2.8456364337431985, "grad_norm": 4.080146789550781, "learning_rate": 3.9226662771732244e-05, "loss": 0.4147786140441895, "memory(GiB)": 70.5, "step": 66420, "token_acc": 0.9020979020979021, "train_speed(iter/s)": 1.447716 }, { "epoch": 2.845850649072448, "grad_norm": 2.1239514350891113, "learning_rate": 3.922009117072706e-05, "loss": 0.26835343837738035, "memory(GiB)": 70.5, "step": 66425, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.447714 }, { "epoch": 2.8460648644016966, "grad_norm": 2.2617592811584473, "learning_rate": 3.921351976501004e-05, "loss": 0.3193783760070801, "memory(GiB)": 70.5, "step": 66430, "token_acc": 0.9316546762589928, "train_speed(iter/s)": 1.447719 }, { "epoch": 2.8462790797309454, "grad_norm": 5.3666558265686035, "learning_rate": 3.920694855470021e-05, "loss": 0.6773103237152099, "memory(GiB)": 70.5, "step": 66435, "token_acc": 0.8655172413793103, "train_speed(iter/s)": 1.447718 }, { "epoch": 2.8464932950601947, "grad_norm": 3.184515953063965, "learning_rate": 3.920037753991665e-05, "loss": 0.5259048938751221, "memory(GiB)": 70.5, "step": 66440, "token_acc": 0.8861788617886179, "train_speed(iter/s)": 1.447735 }, { "epoch": 2.8467075103894435, "grad_norm": 5.991496562957764, "learning_rate": 3.919380672077838e-05, "loss": 0.7048182487487793, "memory(GiB)": 70.5, "step": 66445, "token_acc": 0.8316151202749141, "train_speed(iter/s)": 1.447744 }, { "epoch": 2.8469217257186923, "grad_norm": 3.0973703861236572, "learning_rate": 3.918723609740442e-05, "loss": 0.2840451240539551, "memory(GiB)": 70.5, "step": 66450, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.44775 }, { "epoch": 2.8471359410479415, "grad_norm": 3.8812811374664307, "learning_rate": 3.918066566991384e-05, "loss": 0.22237370014190674, "memory(GiB)": 70.5, "step": 66455, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.447756 }, { "epoch": 2.8473501563771904, "grad_norm": 1.4138044118881226, "learning_rate": 3.917409543842564e-05, "loss": 0.45844235420227053, "memory(GiB)": 70.5, "step": 66460, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.447783 }, { "epoch": 2.847564371706439, "grad_norm": 5.018619537353516, "learning_rate": 3.9167525403058855e-05, "loss": 0.6258168697357178, "memory(GiB)": 70.5, "step": 66465, "token_acc": 0.8718861209964412, "train_speed(iter/s)": 1.447791 }, { "epoch": 2.8477785870356884, "grad_norm": 4.1983723640441895, "learning_rate": 3.9160955563932524e-05, "loss": 0.2938662052154541, "memory(GiB)": 70.5, "step": 66470, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.447797 }, { "epoch": 2.8479928023649372, "grad_norm": 3.7373850345611572, "learning_rate": 3.9154385921165635e-05, "loss": 0.31088528633117674, "memory(GiB)": 70.5, "step": 66475, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.447803 }, { "epoch": 2.848207017694186, "grad_norm": 3.1768083572387695, "learning_rate": 3.9147816474877226e-05, "loss": 0.44684524536132814, "memory(GiB)": 70.5, "step": 66480, "token_acc": 0.9129032258064517, "train_speed(iter/s)": 1.447806 }, { "epoch": 2.8484212330234353, "grad_norm": 1.3141103982925415, "learning_rate": 3.914124722518628e-05, "loss": 0.34629054069519044, "memory(GiB)": 70.5, "step": 66485, "token_acc": 0.9287925696594427, "train_speed(iter/s)": 1.447823 }, { "epoch": 2.848635448352684, "grad_norm": 3.6249372959136963, "learning_rate": 3.913467817221185e-05, "loss": 0.4236114501953125, "memory(GiB)": 70.5, "step": 66490, "token_acc": 0.903114186851211, "train_speed(iter/s)": 1.447838 }, { "epoch": 2.848849663681933, "grad_norm": 4.899203300476074, "learning_rate": 3.912810931607291e-05, "loss": 0.4030588626861572, "memory(GiB)": 70.5, "step": 66495, "token_acc": 0.9087301587301587, "train_speed(iter/s)": 1.447858 }, { "epoch": 2.849063879011182, "grad_norm": 0.9348913431167603, "learning_rate": 3.9121540656888475e-05, "loss": 0.3698451519012451, "memory(GiB)": 70.5, "step": 66500, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.447858 }, { "epoch": 2.849063879011182, "eval_loss": 2.5411570072174072, "eval_runtime": 11.8384, "eval_samples_per_second": 8.447, "eval_steps_per_second": 8.447, "eval_token_acc": 0.4619289340101523, "step": 66500 }, { "epoch": 2.849278094340431, "grad_norm": 2.1292097568511963, "learning_rate": 3.9114972194777534e-05, "loss": 0.38214385509490967, "memory(GiB)": 70.5, "step": 66505, "token_acc": 0.5780510879848628, "train_speed(iter/s)": 1.447462 }, { "epoch": 2.84949230966968, "grad_norm": 1.8684775829315186, "learning_rate": 3.910840392985908e-05, "loss": 0.6555909633636474, "memory(GiB)": 70.5, "step": 66510, "token_acc": 0.8724137931034482, "train_speed(iter/s)": 1.447459 }, { "epoch": 2.849706524998929, "grad_norm": 3.3033814430236816, "learning_rate": 3.910183586225212e-05, "loss": 0.7999634742736816, "memory(GiB)": 70.5, "step": 66515, "token_acc": 0.8283828382838284, "train_speed(iter/s)": 1.44746 }, { "epoch": 2.849920740328178, "grad_norm": 2.3313708305358887, "learning_rate": 3.9095267992075604e-05, "loss": 0.4001220703125, "memory(GiB)": 70.5, "step": 66520, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.447469 }, { "epoch": 2.8501349556574267, "grad_norm": 4.026516914367676, "learning_rate": 3.908870031944856e-05, "loss": 0.40673151016235354, "memory(GiB)": 70.5, "step": 66525, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.447464 }, { "epoch": 2.850349170986676, "grad_norm": 2.3022546768188477, "learning_rate": 3.9082132844489935e-05, "loss": 0.34919157028198244, "memory(GiB)": 70.5, "step": 66530, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.447467 }, { "epoch": 2.8505633863159248, "grad_norm": 1.5866734981536865, "learning_rate": 3.907556556731871e-05, "loss": 0.4028487682342529, "memory(GiB)": 70.5, "step": 66535, "token_acc": 0.9015384615384615, "train_speed(iter/s)": 1.447471 }, { "epoch": 2.8507776016451736, "grad_norm": 7.752634525299072, "learning_rate": 3.906899848805387e-05, "loss": 0.6046825408935547, "memory(GiB)": 70.5, "step": 66540, "token_acc": 0.865625, "train_speed(iter/s)": 1.447472 }, { "epoch": 2.850991816974423, "grad_norm": 2.325608730316162, "learning_rate": 3.906243160681436e-05, "loss": 0.367592453956604, "memory(GiB)": 70.5, "step": 66545, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.447477 }, { "epoch": 2.8512060323036716, "grad_norm": 5.992132663726807, "learning_rate": 3.9055864923719185e-05, "loss": 0.4311245918273926, "memory(GiB)": 70.5, "step": 66550, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.447484 }, { "epoch": 2.8514202476329205, "grad_norm": 2.664475917816162, "learning_rate": 3.9049298438887276e-05, "loss": 0.5092780590057373, "memory(GiB)": 70.5, "step": 66555, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.447494 }, { "epoch": 2.8516344629621697, "grad_norm": 5.778769016265869, "learning_rate": 3.904273215243758e-05, "loss": 0.5615509510040283, "memory(GiB)": 70.5, "step": 66560, "token_acc": 0.8754208754208754, "train_speed(iter/s)": 1.447497 }, { "epoch": 2.8518486782914185, "grad_norm": 3.403287887573242, "learning_rate": 3.903616606448908e-05, "loss": 0.6279273509979248, "memory(GiB)": 70.5, "step": 66565, "token_acc": 0.8794326241134752, "train_speed(iter/s)": 1.447513 }, { "epoch": 2.8520628936206673, "grad_norm": 2.0931222438812256, "learning_rate": 3.902960017516073e-05, "loss": 0.31105608940124513, "memory(GiB)": 70.5, "step": 66570, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.447511 }, { "epoch": 2.8522771089499166, "grad_norm": 2.8533689975738525, "learning_rate": 3.902303448457146e-05, "loss": 0.26679632663726804, "memory(GiB)": 70.5, "step": 66575, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.447513 }, { "epoch": 2.8524913242791654, "grad_norm": 5.252408027648926, "learning_rate": 3.9016468992840203e-05, "loss": 0.5113158702850342, "memory(GiB)": 70.5, "step": 66580, "token_acc": 0.8895522388059701, "train_speed(iter/s)": 1.447515 }, { "epoch": 2.852705539608414, "grad_norm": 3.925121784210205, "learning_rate": 3.900990370008593e-05, "loss": 0.5048113346099854, "memory(GiB)": 70.5, "step": 66585, "token_acc": 0.9, "train_speed(iter/s)": 1.447529 }, { "epoch": 2.8529197549376635, "grad_norm": 1.1296405792236328, "learning_rate": 3.9003338606427564e-05, "loss": 0.4548217296600342, "memory(GiB)": 70.5, "step": 66590, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.447545 }, { "epoch": 2.8531339702669123, "grad_norm": 1.4458285570144653, "learning_rate": 3.899677371198402e-05, "loss": 0.12321004867553711, "memory(GiB)": 70.5, "step": 66595, "token_acc": 0.9701986754966887, "train_speed(iter/s)": 1.447549 }, { "epoch": 2.853348185596161, "grad_norm": 3.6095216274261475, "learning_rate": 3.8990209016874254e-05, "loss": 0.376297664642334, "memory(GiB)": 70.5, "step": 66600, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.447557 }, { "epoch": 2.8535624009254104, "grad_norm": 2.533310890197754, "learning_rate": 3.898364452121717e-05, "loss": 0.3041513204574585, "memory(GiB)": 70.5, "step": 66605, "token_acc": 0.9346938775510204, "train_speed(iter/s)": 1.447562 }, { "epoch": 2.853776616254659, "grad_norm": 0.8389014005661011, "learning_rate": 3.89770802251317e-05, "loss": 0.28148157596588136, "memory(GiB)": 70.5, "step": 66610, "token_acc": 0.946058091286307, "train_speed(iter/s)": 1.447564 }, { "epoch": 2.853990831583908, "grad_norm": 4.049251079559326, "learning_rate": 3.897051612873677e-05, "loss": 0.3119738817214966, "memory(GiB)": 70.5, "step": 66615, "token_acc": 0.9488188976377953, "train_speed(iter/s)": 1.447568 }, { "epoch": 2.8542050469131572, "grad_norm": 2.508737325668335, "learning_rate": 3.896395223215128e-05, "loss": 0.27198033332824706, "memory(GiB)": 70.5, "step": 66620, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.447559 }, { "epoch": 2.854419262242406, "grad_norm": 4.216947555541992, "learning_rate": 3.8957388535494156e-05, "loss": 0.6018841743469239, "memory(GiB)": 70.5, "step": 66625, "token_acc": 0.8758169934640523, "train_speed(iter/s)": 1.447575 }, { "epoch": 2.854633477571655, "grad_norm": 3.372205972671509, "learning_rate": 3.8950825038884296e-05, "loss": 0.45223193168640136, "memory(GiB)": 70.5, "step": 66630, "token_acc": 0.8938906752411575, "train_speed(iter/s)": 1.447578 }, { "epoch": 2.854847692900904, "grad_norm": 4.271628379821777, "learning_rate": 3.894426174244058e-05, "loss": 0.2374575614929199, "memory(GiB)": 70.5, "step": 66635, "token_acc": 0.951417004048583, "train_speed(iter/s)": 1.447584 }, { "epoch": 2.855061908230153, "grad_norm": 4.5956010818481445, "learning_rate": 3.893769864628195e-05, "loss": 0.3876180171966553, "memory(GiB)": 70.5, "step": 66640, "token_acc": 0.900990099009901, "train_speed(iter/s)": 1.447587 }, { "epoch": 2.8552761235594017, "grad_norm": 1.941763997077942, "learning_rate": 3.89311357505273e-05, "loss": 0.302054238319397, "memory(GiB)": 70.5, "step": 66645, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.447609 }, { "epoch": 2.855490338888651, "grad_norm": 9.175124168395996, "learning_rate": 3.892457305529549e-05, "loss": 0.4524231433868408, "memory(GiB)": 70.5, "step": 66650, "token_acc": 0.8940397350993378, "train_speed(iter/s)": 1.447607 }, { "epoch": 2.8557045542179, "grad_norm": 2.788764238357544, "learning_rate": 3.891801056070545e-05, "loss": 0.31786959171295165, "memory(GiB)": 70.5, "step": 66655, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.447609 }, { "epoch": 2.8559187695471486, "grad_norm": 3.465691566467285, "learning_rate": 3.891144826687603e-05, "loss": 0.35066344738006594, "memory(GiB)": 70.5, "step": 66660, "token_acc": 0.9389067524115756, "train_speed(iter/s)": 1.447613 }, { "epoch": 2.856132984876398, "grad_norm": 2.615539073944092, "learning_rate": 3.890488617392613e-05, "loss": 0.24680919647216798, "memory(GiB)": 70.5, "step": 66665, "token_acc": 0.9458689458689459, "train_speed(iter/s)": 1.447621 }, { "epoch": 2.8563472002056467, "grad_norm": 6.523156642913818, "learning_rate": 3.889832428197463e-05, "loss": 0.22087609767913818, "memory(GiB)": 70.5, "step": 66670, "token_acc": 0.946360153256705, "train_speed(iter/s)": 1.447624 }, { "epoch": 2.8565614155348955, "grad_norm": 2.1747448444366455, "learning_rate": 3.88917625911404e-05, "loss": 0.10959359407424926, "memory(GiB)": 70.5, "step": 66675, "token_acc": 0.963265306122449, "train_speed(iter/s)": 1.447633 }, { "epoch": 2.8567756308641448, "grad_norm": 0.7748593688011169, "learning_rate": 3.8885201101542304e-05, "loss": 0.2883943796157837, "memory(GiB)": 70.5, "step": 66680, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.44764 }, { "epoch": 2.8569898461933936, "grad_norm": 2.881211519241333, "learning_rate": 3.8878639813299214e-05, "loss": 0.2567914009094238, "memory(GiB)": 70.5, "step": 66685, "token_acc": 0.9399293286219081, "train_speed(iter/s)": 1.447646 }, { "epoch": 2.8572040615226424, "grad_norm": 4.218475818634033, "learning_rate": 3.887207872653e-05, "loss": 0.4636648178100586, "memory(GiB)": 70.5, "step": 66690, "token_acc": 0.879746835443038, "train_speed(iter/s)": 1.447644 }, { "epoch": 2.8574182768518916, "grad_norm": 0.5130957365036011, "learning_rate": 3.886551784135352e-05, "loss": 0.3582013607025146, "memory(GiB)": 70.5, "step": 66695, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.447659 }, { "epoch": 2.8576324921811405, "grad_norm": 3.0643951892852783, "learning_rate": 3.885895715788864e-05, "loss": 0.6383772373199463, "memory(GiB)": 70.5, "step": 66700, "token_acc": 0.8478260869565217, "train_speed(iter/s)": 1.44766 }, { "epoch": 2.8578467075103893, "grad_norm": 1.1279206275939941, "learning_rate": 3.8852396676254175e-05, "loss": 0.31574578285217286, "memory(GiB)": 70.5, "step": 66705, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.447665 }, { "epoch": 2.8580609228396385, "grad_norm": 2.856600761413574, "learning_rate": 3.8845836396569007e-05, "loss": 0.3068131923675537, "memory(GiB)": 70.5, "step": 66710, "token_acc": 0.9320754716981132, "train_speed(iter/s)": 1.447671 }, { "epoch": 2.8582751381688873, "grad_norm": 3.716294527053833, "learning_rate": 3.8839276318952e-05, "loss": 0.5498095512390136, "memory(GiB)": 70.5, "step": 66715, "token_acc": 0.8896797153024911, "train_speed(iter/s)": 1.44766 }, { "epoch": 2.858489353498136, "grad_norm": 1.8042184114456177, "learning_rate": 3.883271644352196e-05, "loss": 0.6305339336395264, "memory(GiB)": 70.5, "step": 66720, "token_acc": 0.852760736196319, "train_speed(iter/s)": 1.447666 }, { "epoch": 2.8587035688273854, "grad_norm": 2.6224794387817383, "learning_rate": 3.882615677039773e-05, "loss": 0.4635188102722168, "memory(GiB)": 70.5, "step": 66725, "token_acc": 0.8847457627118644, "train_speed(iter/s)": 1.44767 }, { "epoch": 2.8589177841566342, "grad_norm": 4.126102447509766, "learning_rate": 3.881959729969816e-05, "loss": 0.2620461225509644, "memory(GiB)": 70.5, "step": 66730, "token_acc": 0.946969696969697, "train_speed(iter/s)": 1.447689 }, { "epoch": 2.859131999485883, "grad_norm": 2.0425233840942383, "learning_rate": 3.881303803154208e-05, "loss": 0.3130959987640381, "memory(GiB)": 70.5, "step": 66735, "token_acc": 0.910394265232975, "train_speed(iter/s)": 1.4477 }, { "epoch": 2.8593462148151323, "grad_norm": 1.4322675466537476, "learning_rate": 3.8806478966048295e-05, "loss": 0.3488304138183594, "memory(GiB)": 70.5, "step": 66740, "token_acc": 0.9453125, "train_speed(iter/s)": 1.447702 }, { "epoch": 2.859560430144381, "grad_norm": 3.2545480728149414, "learning_rate": 3.879992010333565e-05, "loss": 0.3374790668487549, "memory(GiB)": 70.5, "step": 66745, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.447703 }, { "epoch": 2.85977464547363, "grad_norm": 4.238231658935547, "learning_rate": 3.879336144352296e-05, "loss": 0.5620427131652832, "memory(GiB)": 70.5, "step": 66750, "token_acc": 0.8932806324110671, "train_speed(iter/s)": 1.447704 }, { "epoch": 2.859988860802879, "grad_norm": 1.8402211666107178, "learning_rate": 3.878680298672904e-05, "loss": 0.37086942195892336, "memory(GiB)": 70.5, "step": 66755, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.447717 }, { "epoch": 2.860203076132128, "grad_norm": 2.256416082382202, "learning_rate": 3.8780244733072706e-05, "loss": 0.5108039855957032, "memory(GiB)": 70.5, "step": 66760, "token_acc": 0.8787878787878788, "train_speed(iter/s)": 1.447712 }, { "epoch": 2.860417291461377, "grad_norm": 4.195559501647949, "learning_rate": 3.8773686682672756e-05, "loss": 0.3560898780822754, "memory(GiB)": 70.5, "step": 66765, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.44771 }, { "epoch": 2.860631506790626, "grad_norm": 4.23447322845459, "learning_rate": 3.876712883564801e-05, "loss": 0.45772485733032225, "memory(GiB)": 70.5, "step": 66770, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.447721 }, { "epoch": 2.860845722119875, "grad_norm": 1.0056902170181274, "learning_rate": 3.876057119211727e-05, "loss": 0.25912506580352784, "memory(GiB)": 70.5, "step": 66775, "token_acc": 0.9461077844311377, "train_speed(iter/s)": 1.447722 }, { "epoch": 2.8610599374491237, "grad_norm": 0.20162400603294373, "learning_rate": 3.875401375219929e-05, "loss": 0.29789443016052247, "memory(GiB)": 70.5, "step": 66780, "token_acc": 0.9495798319327731, "train_speed(iter/s)": 1.447719 }, { "epoch": 2.861274152778373, "grad_norm": 4.743063926696777, "learning_rate": 3.8747456516012926e-05, "loss": 0.4463296413421631, "memory(GiB)": 70.5, "step": 66785, "token_acc": 0.903125, "train_speed(iter/s)": 1.447732 }, { "epoch": 2.8614883681076217, "grad_norm": 2.163632869720459, "learning_rate": 3.8740899483676946e-05, "loss": 0.3014801502227783, "memory(GiB)": 70.5, "step": 66790, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.447729 }, { "epoch": 2.8617025834368706, "grad_norm": 6.067420959472656, "learning_rate": 3.8734342655310124e-05, "loss": 0.35998175144195554, "memory(GiB)": 70.5, "step": 66795, "token_acc": 0.9174917491749175, "train_speed(iter/s)": 1.447733 }, { "epoch": 2.86191679876612, "grad_norm": 1.7543585300445557, "learning_rate": 3.872778603103126e-05, "loss": 0.3211528778076172, "memory(GiB)": 70.5, "step": 66800, "token_acc": 0.930327868852459, "train_speed(iter/s)": 1.447755 }, { "epoch": 2.8621310140953686, "grad_norm": 1.183144450187683, "learning_rate": 3.8721229610959136e-05, "loss": 0.4881251335144043, "memory(GiB)": 70.5, "step": 66805, "token_acc": 0.9064327485380117, "train_speed(iter/s)": 1.447755 }, { "epoch": 2.8623452294246174, "grad_norm": 1.6494024991989136, "learning_rate": 3.87146733952125e-05, "loss": 0.3457007646560669, "memory(GiB)": 70.5, "step": 66810, "token_acc": 0.919732441471572, "train_speed(iter/s)": 1.447761 }, { "epoch": 2.8625594447538667, "grad_norm": 0.617072582244873, "learning_rate": 3.870811738391015e-05, "loss": 0.36046152114868163, "memory(GiB)": 70.5, "step": 66815, "token_acc": 0.9197530864197531, "train_speed(iter/s)": 1.447775 }, { "epoch": 2.8627736600831155, "grad_norm": 0.198045015335083, "learning_rate": 3.8701561577170855e-05, "loss": 0.3860219717025757, "memory(GiB)": 70.5, "step": 66820, "token_acc": 0.9153225806451613, "train_speed(iter/s)": 1.447787 }, { "epoch": 2.8629878754123643, "grad_norm": 2.5291197299957275, "learning_rate": 3.8695005975113357e-05, "loss": 0.29847543239593505, "memory(GiB)": 70.5, "step": 66825, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.447801 }, { "epoch": 2.8632020907416136, "grad_norm": 3.0991077423095703, "learning_rate": 3.868845057785644e-05, "loss": 0.4011994361877441, "memory(GiB)": 70.5, "step": 66830, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.44781 }, { "epoch": 2.8634163060708624, "grad_norm": 2.243878126144409, "learning_rate": 3.868189538551884e-05, "loss": 0.5083224296569824, "memory(GiB)": 70.5, "step": 66835, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.447815 }, { "epoch": 2.863630521400111, "grad_norm": 2.5142037868499756, "learning_rate": 3.8675340398219335e-05, "loss": 0.2896485090255737, "memory(GiB)": 70.5, "step": 66840, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.447818 }, { "epoch": 2.8638447367293605, "grad_norm": 4.540031909942627, "learning_rate": 3.8668785616076665e-05, "loss": 0.38742947578430176, "memory(GiB)": 70.5, "step": 66845, "token_acc": 0.9057971014492754, "train_speed(iter/s)": 1.447824 }, { "epoch": 2.8640589520586093, "grad_norm": 2.7252843379974365, "learning_rate": 3.866223103920956e-05, "loss": 0.23499655723571777, "memory(GiB)": 70.5, "step": 66850, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.447828 }, { "epoch": 2.864273167387858, "grad_norm": 2.1932592391967773, "learning_rate": 3.865567666773676e-05, "loss": 0.3170801639556885, "memory(GiB)": 70.5, "step": 66855, "token_acc": 0.9407407407407408, "train_speed(iter/s)": 1.447845 }, { "epoch": 2.8644873827171073, "grad_norm": 1.0161571502685547, "learning_rate": 3.864912250177704e-05, "loss": 0.20864946842193605, "memory(GiB)": 70.5, "step": 66860, "token_acc": 0.9694915254237289, "train_speed(iter/s)": 1.447852 }, { "epoch": 2.864701598046356, "grad_norm": 0.30263635516166687, "learning_rate": 3.864256854144912e-05, "loss": 0.40090384483337405, "memory(GiB)": 70.5, "step": 66865, "token_acc": 0.9395973154362416, "train_speed(iter/s)": 1.44787 }, { "epoch": 2.864915813375605, "grad_norm": 6.209007263183594, "learning_rate": 3.863601478687171e-05, "loss": 0.3141413688659668, "memory(GiB)": 70.5, "step": 66870, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.447889 }, { "epoch": 2.8651300287048542, "grad_norm": 2.5749151706695557, "learning_rate": 3.862946123816357e-05, "loss": 0.30755550861358644, "memory(GiB)": 70.5, "step": 66875, "token_acc": 0.923943661971831, "train_speed(iter/s)": 1.447895 }, { "epoch": 2.865344244034103, "grad_norm": 1.5115841627120972, "learning_rate": 3.86229078954434e-05, "loss": 0.4303779125213623, "memory(GiB)": 70.5, "step": 66880, "token_acc": 0.9, "train_speed(iter/s)": 1.447898 }, { "epoch": 2.865558459363352, "grad_norm": 3.6656951904296875, "learning_rate": 3.8616354758829915e-05, "loss": 0.1730343818664551, "memory(GiB)": 70.5, "step": 66885, "token_acc": 0.9513888888888888, "train_speed(iter/s)": 1.4479 }, { "epoch": 2.865772674692601, "grad_norm": 5.388194561004639, "learning_rate": 3.8609801828441855e-05, "loss": 0.5864045143127441, "memory(GiB)": 70.5, "step": 66890, "token_acc": 0.8987730061349694, "train_speed(iter/s)": 1.447906 }, { "epoch": 2.86598689002185, "grad_norm": 4.282329082489014, "learning_rate": 3.860324910439791e-05, "loss": 0.42018518447875974, "memory(GiB)": 70.5, "step": 66895, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.447902 }, { "epoch": 2.8662011053510987, "grad_norm": 3.960721015930176, "learning_rate": 3.859669658681681e-05, "loss": 0.24214000701904298, "memory(GiB)": 70.5, "step": 66900, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.447911 }, { "epoch": 2.866415320680348, "grad_norm": 4.296505451202393, "learning_rate": 3.859014427581724e-05, "loss": 0.4856898307800293, "memory(GiB)": 70.5, "step": 66905, "token_acc": 0.9042145593869731, "train_speed(iter/s)": 1.447909 }, { "epoch": 2.866629536009597, "grad_norm": 3.5059428215026855, "learning_rate": 3.858359217151791e-05, "loss": 0.31721885204315187, "memory(GiB)": 70.5, "step": 66910, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.44792 }, { "epoch": 2.8668437513388456, "grad_norm": 3.3448944091796875, "learning_rate": 3.857704027403752e-05, "loss": 0.7840248107910156, "memory(GiB)": 70.5, "step": 66915, "token_acc": 0.8390092879256966, "train_speed(iter/s)": 1.447931 }, { "epoch": 2.867057966668095, "grad_norm": 4.003973484039307, "learning_rate": 3.8570488583494766e-05, "loss": 0.3361915349960327, "memory(GiB)": 70.5, "step": 66920, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.44794 }, { "epoch": 2.8672721819973437, "grad_norm": 2.013416290283203, "learning_rate": 3.856393710000832e-05, "loss": 0.3875916004180908, "memory(GiB)": 70.5, "step": 66925, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.447945 }, { "epoch": 2.8674863973265925, "grad_norm": 2.7103662490844727, "learning_rate": 3.8557385823696875e-05, "loss": 0.2849641561508179, "memory(GiB)": 70.5, "step": 66930, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.447948 }, { "epoch": 2.8677006126558418, "grad_norm": 2.238617181777954, "learning_rate": 3.855083475467913e-05, "loss": 0.4356387138366699, "memory(GiB)": 70.5, "step": 66935, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.447961 }, { "epoch": 2.8679148279850906, "grad_norm": 3.4166455268859863, "learning_rate": 3.854428389307376e-05, "loss": 0.26622886657714845, "memory(GiB)": 70.5, "step": 66940, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.447971 }, { "epoch": 2.8681290433143394, "grad_norm": 4.495809555053711, "learning_rate": 3.853773323899943e-05, "loss": 0.3638364315032959, "memory(GiB)": 70.5, "step": 66945, "token_acc": 0.9271137026239067, "train_speed(iter/s)": 1.44798 }, { "epoch": 2.8683432586435886, "grad_norm": 3.8037431240081787, "learning_rate": 3.853118279257482e-05, "loss": 0.36611170768737794, "memory(GiB)": 70.5, "step": 66950, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.447988 }, { "epoch": 2.8685574739728374, "grad_norm": 2.7051353454589844, "learning_rate": 3.852463255391858e-05, "loss": 0.29459729194641116, "memory(GiB)": 70.5, "step": 66955, "token_acc": 0.9463722397476341, "train_speed(iter/s)": 1.448 }, { "epoch": 2.8687716893020863, "grad_norm": 2.8191568851470947, "learning_rate": 3.85180825231494e-05, "loss": 0.5130410671234131, "memory(GiB)": 70.5, "step": 66960, "token_acc": 0.903448275862069, "train_speed(iter/s)": 1.448025 }, { "epoch": 2.8689859046313355, "grad_norm": 2.805218458175659, "learning_rate": 3.851153270038592e-05, "loss": 0.26826789379119875, "memory(GiB)": 70.5, "step": 66965, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.448025 }, { "epoch": 2.8692001199605843, "grad_norm": 1.8055000305175781, "learning_rate": 3.85049830857468e-05, "loss": 0.33157274723052976, "memory(GiB)": 70.5, "step": 66970, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.448052 }, { "epoch": 2.869414335289833, "grad_norm": 3.8577473163604736, "learning_rate": 3.84984336793507e-05, "loss": 0.6110603332519531, "memory(GiB)": 70.5, "step": 66975, "token_acc": 0.8839590443686007, "train_speed(iter/s)": 1.448054 }, { "epoch": 2.8696285506190824, "grad_norm": 2.261465072631836, "learning_rate": 3.849188448131625e-05, "loss": 0.29571781158447263, "memory(GiB)": 70.5, "step": 66980, "token_acc": 0.9297124600638977, "train_speed(iter/s)": 1.448051 }, { "epoch": 2.869842765948331, "grad_norm": 0.2861659824848175, "learning_rate": 3.8485335491762126e-05, "loss": 0.3798626661300659, "memory(GiB)": 70.5, "step": 66985, "token_acc": 0.925531914893617, "train_speed(iter/s)": 1.448059 }, { "epoch": 2.87005698127758, "grad_norm": 3.9111037254333496, "learning_rate": 3.8478786710806946e-05, "loss": 0.2657310009002686, "memory(GiB)": 70.5, "step": 66990, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.448058 }, { "epoch": 2.8702711966068293, "grad_norm": 6.0583343505859375, "learning_rate": 3.847223813856934e-05, "loss": 0.3883610486984253, "memory(GiB)": 70.5, "step": 66995, "token_acc": 0.9208860759493671, "train_speed(iter/s)": 1.448065 }, { "epoch": 2.870485411936078, "grad_norm": 3.785140037536621, "learning_rate": 3.846568977516797e-05, "loss": 0.43224287033081055, "memory(GiB)": 70.5, "step": 67000, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.448068 }, { "epoch": 2.870485411936078, "eval_loss": 2.4158496856689453, "eval_runtime": 12.0553, "eval_samples_per_second": 8.295, "eval_steps_per_second": 8.295, "eval_token_acc": 0.44814340588988477, "step": 67000 }, { "epoch": 2.870699627265327, "grad_norm": 3.4012222290039062, "learning_rate": 3.845914162072143e-05, "loss": 0.3884379148483276, "memory(GiB)": 70.5, "step": 67005, "token_acc": 0.5696682464454976, "train_speed(iter/s)": 1.447672 }, { "epoch": 2.870913842594576, "grad_norm": 2.875516176223755, "learning_rate": 3.845259367534838e-05, "loss": 0.37299747467041017, "memory(GiB)": 70.5, "step": 67010, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.447687 }, { "epoch": 2.871128057923825, "grad_norm": 1.3484715223312378, "learning_rate": 3.8446045939167425e-05, "loss": 0.4829817771911621, "memory(GiB)": 70.5, "step": 67015, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.447691 }, { "epoch": 2.871342273253074, "grad_norm": 4.326211929321289, "learning_rate": 3.8439498412297195e-05, "loss": 0.49010210037231444, "memory(GiB)": 70.5, "step": 67020, "token_acc": 0.8840579710144928, "train_speed(iter/s)": 1.44769 }, { "epoch": 2.871556488582323, "grad_norm": 8.232673645019531, "learning_rate": 3.843295109485629e-05, "loss": 0.716582202911377, "memory(GiB)": 70.5, "step": 67025, "token_acc": 0.8723404255319149, "train_speed(iter/s)": 1.447695 }, { "epoch": 2.871770703911572, "grad_norm": 0.5941632986068726, "learning_rate": 3.842640398696332e-05, "loss": 0.5215926170349121, "memory(GiB)": 70.5, "step": 67030, "token_acc": 0.8966666666666666, "train_speed(iter/s)": 1.447695 }, { "epoch": 2.8719849192408207, "grad_norm": 2.233085870742798, "learning_rate": 3.841985708873691e-05, "loss": 0.19619650840759278, "memory(GiB)": 70.5, "step": 67035, "token_acc": 0.959866220735786, "train_speed(iter/s)": 1.447704 }, { "epoch": 2.87219913457007, "grad_norm": 4.1796183586120605, "learning_rate": 3.8413310400295636e-05, "loss": 0.23776865005493164, "memory(GiB)": 70.5, "step": 67040, "token_acc": 0.9574468085106383, "train_speed(iter/s)": 1.447706 }, { "epoch": 2.8724133498993187, "grad_norm": 5.769886493682861, "learning_rate": 3.8406763921758124e-05, "loss": 0.35133683681488037, "memory(GiB)": 70.5, "step": 67045, "token_acc": 0.9256965944272446, "train_speed(iter/s)": 1.447716 }, { "epoch": 2.8726275652285675, "grad_norm": 1.1729305982589722, "learning_rate": 3.840021765324297e-05, "loss": 0.20674796104431153, "memory(GiB)": 70.5, "step": 67050, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.447732 }, { "epoch": 2.872841780557817, "grad_norm": 1.3422417640686035, "learning_rate": 3.839367159486874e-05, "loss": 0.2530949831008911, "memory(GiB)": 70.5, "step": 67055, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.44774 }, { "epoch": 2.8730559958870656, "grad_norm": 4.040317535400391, "learning_rate": 3.838712574675404e-05, "loss": 0.3842175006866455, "memory(GiB)": 70.5, "step": 67060, "token_acc": 0.9127906976744186, "train_speed(iter/s)": 1.447745 }, { "epoch": 2.8732702112163144, "grad_norm": 0.8231345415115356, "learning_rate": 3.838058010901746e-05, "loss": 0.2611249446868896, "memory(GiB)": 70.5, "step": 67065, "token_acc": 0.958041958041958, "train_speed(iter/s)": 1.447753 }, { "epoch": 2.8734844265455637, "grad_norm": 2.3504562377929688, "learning_rate": 3.8374034681777546e-05, "loss": 0.23003318309783935, "memory(GiB)": 70.5, "step": 67070, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.447756 }, { "epoch": 2.8736986418748125, "grad_norm": 3.95674991607666, "learning_rate": 3.8367489465152905e-05, "loss": 0.41397719383239745, "memory(GiB)": 70.5, "step": 67075, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.44776 }, { "epoch": 2.8739128572040613, "grad_norm": 2.0576961040496826, "learning_rate": 3.836094445926212e-05, "loss": 0.30470051765441897, "memory(GiB)": 70.5, "step": 67080, "token_acc": 0.9297297297297298, "train_speed(iter/s)": 1.447763 }, { "epoch": 2.8741270725333106, "grad_norm": 4.907217025756836, "learning_rate": 3.835439966422373e-05, "loss": 0.7268563270568847, "memory(GiB)": 70.5, "step": 67085, "token_acc": 0.8566308243727598, "train_speed(iter/s)": 1.447769 }, { "epoch": 2.8743412878625594, "grad_norm": 3.6707663536071777, "learning_rate": 3.834785508015633e-05, "loss": 0.24798853397369386, "memory(GiB)": 70.5, "step": 67090, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.447773 }, { "epoch": 2.874555503191808, "grad_norm": 3.9607675075531006, "learning_rate": 3.834131070717847e-05, "loss": 0.604449462890625, "memory(GiB)": 70.5, "step": 67095, "token_acc": 0.8628048780487805, "train_speed(iter/s)": 1.447793 }, { "epoch": 2.8747697185210574, "grad_norm": 1.7383619546890259, "learning_rate": 3.833476654540869e-05, "loss": 0.16426993608474733, "memory(GiB)": 70.5, "step": 67100, "token_acc": 0.9560439560439561, "train_speed(iter/s)": 1.447817 }, { "epoch": 2.8749839338503063, "grad_norm": 4.70010232925415, "learning_rate": 3.8328222594965555e-05, "loss": 0.4442389488220215, "memory(GiB)": 70.5, "step": 67105, "token_acc": 0.9114391143911439, "train_speed(iter/s)": 1.44783 }, { "epoch": 2.875198149179555, "grad_norm": 0.29538148641586304, "learning_rate": 3.832167885596763e-05, "loss": 0.311225152015686, "memory(GiB)": 70.5, "step": 67110, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.447825 }, { "epoch": 2.8754123645088043, "grad_norm": 1.3726603984832764, "learning_rate": 3.8315135328533435e-05, "loss": 0.15462535619735718, "memory(GiB)": 70.5, "step": 67115, "token_acc": 0.9713261648745519, "train_speed(iter/s)": 1.447832 }, { "epoch": 2.875626579838053, "grad_norm": 0.39649027585983276, "learning_rate": 3.830859201278153e-05, "loss": 0.27174863815307615, "memory(GiB)": 70.5, "step": 67120, "token_acc": 0.9413793103448276, "train_speed(iter/s)": 1.447827 }, { "epoch": 2.875840795167302, "grad_norm": 4.068038463592529, "learning_rate": 3.830204890883044e-05, "loss": 0.3677940845489502, "memory(GiB)": 70.5, "step": 67125, "token_acc": 0.9073359073359073, "train_speed(iter/s)": 1.447841 }, { "epoch": 2.876055010496551, "grad_norm": 2.377215623855591, "learning_rate": 3.829550601679872e-05, "loss": 0.6124482154846191, "memory(GiB)": 70.5, "step": 67130, "token_acc": 0.8765822784810127, "train_speed(iter/s)": 1.447855 }, { "epoch": 2.8762692258258, "grad_norm": 1.7841854095458984, "learning_rate": 3.828896333680489e-05, "loss": 0.1816679358482361, "memory(GiB)": 70.5, "step": 67135, "token_acc": 0.9585062240663901, "train_speed(iter/s)": 1.44786 }, { "epoch": 2.876483441155049, "grad_norm": 2.8784637451171875, "learning_rate": 3.828242086896746e-05, "loss": 0.33973121643066406, "memory(GiB)": 70.5, "step": 67140, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.447863 }, { "epoch": 2.876697656484298, "grad_norm": 1.117360234260559, "learning_rate": 3.827587861340498e-05, "loss": 0.29737081527709963, "memory(GiB)": 70.5, "step": 67145, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.447868 }, { "epoch": 2.876911871813547, "grad_norm": 3.6858911514282227, "learning_rate": 3.826933657023594e-05, "loss": 0.1542014718055725, "memory(GiB)": 70.5, "step": 67150, "token_acc": 0.9573170731707317, "train_speed(iter/s)": 1.44787 }, { "epoch": 2.8771260871427957, "grad_norm": 1.1585487127304077, "learning_rate": 3.826279473957889e-05, "loss": 0.7374300003051758, "memory(GiB)": 70.5, "step": 67155, "token_acc": 0.8074074074074075, "train_speed(iter/s)": 1.447876 }, { "epoch": 2.877340302472045, "grad_norm": 1.377804160118103, "learning_rate": 3.825625312155231e-05, "loss": 0.301504921913147, "memory(GiB)": 70.5, "step": 67160, "token_acc": 0.9434782608695652, "train_speed(iter/s)": 1.44788 }, { "epoch": 2.877554517801294, "grad_norm": 1.975433588027954, "learning_rate": 3.824971171627474e-05, "loss": 0.35050828456878663, "memory(GiB)": 70.5, "step": 67165, "token_acc": 0.9307228915662651, "train_speed(iter/s)": 1.447881 }, { "epoch": 2.8777687331305426, "grad_norm": 3.0812442302703857, "learning_rate": 3.8243170523864655e-05, "loss": 0.4391437530517578, "memory(GiB)": 70.5, "step": 67170, "token_acc": 0.916923076923077, "train_speed(iter/s)": 1.447889 }, { "epoch": 2.877982948459792, "grad_norm": 4.650461196899414, "learning_rate": 3.823662954444056e-05, "loss": 0.6598155498504639, "memory(GiB)": 70.5, "step": 67175, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.447889 }, { "epoch": 2.8781971637890407, "grad_norm": 9.48575496673584, "learning_rate": 3.823008877812097e-05, "loss": 0.5521164417266846, "memory(GiB)": 70.5, "step": 67180, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.447904 }, { "epoch": 2.8784113791182895, "grad_norm": 3.9245717525482178, "learning_rate": 3.822354822502434e-05, "loss": 0.26292884349823, "memory(GiB)": 70.5, "step": 67185, "token_acc": 0.9433333333333334, "train_speed(iter/s)": 1.447907 }, { "epoch": 2.8786255944475387, "grad_norm": 4.604609966278076, "learning_rate": 3.821700788526921e-05, "loss": 0.4015693187713623, "memory(GiB)": 70.5, "step": 67190, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.447908 }, { "epoch": 2.8788398097767876, "grad_norm": 3.9072048664093018, "learning_rate": 3.8210467758974026e-05, "loss": 0.29074156284332275, "memory(GiB)": 70.5, "step": 67195, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.447908 }, { "epoch": 2.8790540251060364, "grad_norm": 2.942148208618164, "learning_rate": 3.820392784625726e-05, "loss": 0.3475234270095825, "memory(GiB)": 70.5, "step": 67200, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.447909 }, { "epoch": 2.8792682404352856, "grad_norm": 2.5388355255126953, "learning_rate": 3.819738814723742e-05, "loss": 0.3204353332519531, "memory(GiB)": 70.5, "step": 67205, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.447911 }, { "epoch": 2.8794824557645344, "grad_norm": 1.736295461654663, "learning_rate": 3.8190848662032974e-05, "loss": 0.5259270191192627, "memory(GiB)": 70.5, "step": 67210, "token_acc": 0.9037037037037037, "train_speed(iter/s)": 1.44793 }, { "epoch": 2.8796966710937832, "grad_norm": 3.579225778579712, "learning_rate": 3.8184309390762364e-05, "loss": 0.4163031578063965, "memory(GiB)": 70.5, "step": 67215, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.447934 }, { "epoch": 2.8799108864230325, "grad_norm": 2.174175500869751, "learning_rate": 3.817777033354409e-05, "loss": 0.5065288066864013, "memory(GiB)": 70.5, "step": 67220, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.447939 }, { "epoch": 2.8801251017522813, "grad_norm": 3.6106975078582764, "learning_rate": 3.817123149049657e-05, "loss": 0.2556574821472168, "memory(GiB)": 70.5, "step": 67225, "token_acc": 0.9352941176470588, "train_speed(iter/s)": 1.447947 }, { "epoch": 2.88033931708153, "grad_norm": 4.137025356292725, "learning_rate": 3.81646928617383e-05, "loss": 0.3111577987670898, "memory(GiB)": 70.5, "step": 67230, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.447957 }, { "epoch": 2.8805535324107794, "grad_norm": 3.286949872970581, "learning_rate": 3.8158154447387737e-05, "loss": 0.32758393287658694, "memory(GiB)": 70.5, "step": 67235, "token_acc": 0.9161490683229814, "train_speed(iter/s)": 1.447958 }, { "epoch": 2.880767747740028, "grad_norm": 0.1576942801475525, "learning_rate": 3.81516162475633e-05, "loss": 0.1796610951423645, "memory(GiB)": 70.5, "step": 67240, "token_acc": 0.9610894941634242, "train_speed(iter/s)": 1.447961 }, { "epoch": 2.880981963069277, "grad_norm": 4.242079257965088, "learning_rate": 3.814507826238345e-05, "loss": 0.17104625701904297, "memory(GiB)": 70.5, "step": 67245, "token_acc": 0.9570957095709571, "train_speed(iter/s)": 1.447962 }, { "epoch": 2.8811961783985263, "grad_norm": 6.859060287475586, "learning_rate": 3.813854049196664e-05, "loss": 0.5584538459777832, "memory(GiB)": 70.5, "step": 67250, "token_acc": 0.8868613138686131, "train_speed(iter/s)": 1.447967 }, { "epoch": 2.881410393727775, "grad_norm": 4.6570024490356445, "learning_rate": 3.813200293643129e-05, "loss": 0.6491868972778321, "memory(GiB)": 70.5, "step": 67255, "token_acc": 0.8819444444444444, "train_speed(iter/s)": 1.447967 }, { "epoch": 2.881624609057024, "grad_norm": 1.9114004373550415, "learning_rate": 3.8125465595895834e-05, "loss": 0.3998169660568237, "memory(GiB)": 70.5, "step": 67260, "token_acc": 0.9068100358422939, "train_speed(iter/s)": 1.447973 }, { "epoch": 2.881838824386273, "grad_norm": 0.20275039970874786, "learning_rate": 3.811892847047872e-05, "loss": 0.23361868858337403, "memory(GiB)": 70.5, "step": 67265, "token_acc": 0.950381679389313, "train_speed(iter/s)": 1.447974 }, { "epoch": 2.882053039715522, "grad_norm": 0.8228528499603271, "learning_rate": 3.811239156029835e-05, "loss": 0.29470243453979494, "memory(GiB)": 70.5, "step": 67270, "token_acc": 0.9306122448979591, "train_speed(iter/s)": 1.447977 }, { "epoch": 2.8822672550447708, "grad_norm": 6.638169765472412, "learning_rate": 3.8105854865473164e-05, "loss": 0.5095582008361816, "memory(GiB)": 70.5, "step": 67275, "token_acc": 0.9150326797385621, "train_speed(iter/s)": 1.447985 }, { "epoch": 2.88248147037402, "grad_norm": 1.0998705625534058, "learning_rate": 3.809931838612159e-05, "loss": 0.18349567651748658, "memory(GiB)": 70.5, "step": 67280, "token_acc": 0.9601328903654485, "train_speed(iter/s)": 1.447996 }, { "epoch": 2.882695685703269, "grad_norm": 3.7823665142059326, "learning_rate": 3.8092782122362e-05, "loss": 0.35943899154663084, "memory(GiB)": 70.5, "step": 67285, "token_acc": 0.9225806451612903, "train_speed(iter/s)": 1.447997 }, { "epoch": 2.8829099010325177, "grad_norm": 4.3879523277282715, "learning_rate": 3.808624607431285e-05, "loss": 0.3897504091262817, "memory(GiB)": 70.5, "step": 67290, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.447999 }, { "epoch": 2.883124116361767, "grad_norm": 4.238631248474121, "learning_rate": 3.807971024209252e-05, "loss": 0.3057858467102051, "memory(GiB)": 70.5, "step": 67295, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.448002 }, { "epoch": 2.8833383316910157, "grad_norm": 3.4167137145996094, "learning_rate": 3.8073174625819406e-05, "loss": 0.296979022026062, "memory(GiB)": 70.5, "step": 67300, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.448008 }, { "epoch": 2.8835525470202645, "grad_norm": 1.1476459503173828, "learning_rate": 3.8066639225611934e-05, "loss": 0.23614723682403566, "memory(GiB)": 70.5, "step": 67305, "token_acc": 0.9431818181818182, "train_speed(iter/s)": 1.448011 }, { "epoch": 2.883766762349514, "grad_norm": 4.579334735870361, "learning_rate": 3.8060104041588494e-05, "loss": 0.28195600509643554, "memory(GiB)": 70.5, "step": 67310, "token_acc": 0.953307392996109, "train_speed(iter/s)": 1.448015 }, { "epoch": 2.8839809776787626, "grad_norm": 0.9790074825286865, "learning_rate": 3.805356907386747e-05, "loss": 0.2269373893737793, "memory(GiB)": 70.5, "step": 67315, "token_acc": 0.9541284403669725, "train_speed(iter/s)": 1.448024 }, { "epoch": 2.8841951930080114, "grad_norm": 1.692314863204956, "learning_rate": 3.8047034322567246e-05, "loss": 0.36444945335388185, "memory(GiB)": 70.5, "step": 67320, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.448028 }, { "epoch": 2.8844094083372607, "grad_norm": 2.304074287414551, "learning_rate": 3.804049978780621e-05, "loss": 0.49854159355163574, "memory(GiB)": 70.5, "step": 67325, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.448043 }, { "epoch": 2.8846236236665095, "grad_norm": 3.2984635829925537, "learning_rate": 3.803396546970273e-05, "loss": 0.34385526180267334, "memory(GiB)": 70.5, "step": 67330, "token_acc": 0.9283387622149837, "train_speed(iter/s)": 1.448051 }, { "epoch": 2.8848378389957583, "grad_norm": 2.0991668701171875, "learning_rate": 3.80274313683752e-05, "loss": 0.49622769355773927, "memory(GiB)": 70.5, "step": 67335, "token_acc": 0.8953068592057761, "train_speed(iter/s)": 1.44807 }, { "epoch": 2.8850520543250076, "grad_norm": 1.4454987049102783, "learning_rate": 3.802089748394199e-05, "loss": 0.6378312587738038, "memory(GiB)": 70.5, "step": 67340, "token_acc": 0.8983957219251337, "train_speed(iter/s)": 1.44808 }, { "epoch": 2.8852662696542564, "grad_norm": 2.9056098461151123, "learning_rate": 3.801436381652144e-05, "loss": 0.4835883617401123, "memory(GiB)": 70.5, "step": 67345, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.448084 }, { "epoch": 2.885480484983505, "grad_norm": 7.732537269592285, "learning_rate": 3.800783036623194e-05, "loss": 0.21866085529327392, "memory(GiB)": 70.5, "step": 67350, "token_acc": 0.9295154185022027, "train_speed(iter/s)": 1.448086 }, { "epoch": 2.8856947003127544, "grad_norm": 0.23018546402454376, "learning_rate": 3.8001297133191855e-05, "loss": 0.22921984195709227, "memory(GiB)": 70.5, "step": 67355, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.448091 }, { "epoch": 2.8859089156420032, "grad_norm": 2.183704137802124, "learning_rate": 3.799476411751951e-05, "loss": 0.47748093605041503, "memory(GiB)": 70.5, "step": 67360, "token_acc": 0.910958904109589, "train_speed(iter/s)": 1.448098 }, { "epoch": 2.886123130971252, "grad_norm": 3.8021841049194336, "learning_rate": 3.798823131933329e-05, "loss": 0.496826696395874, "memory(GiB)": 70.5, "step": 67365, "token_acc": 0.8913857677902621, "train_speed(iter/s)": 1.448099 }, { "epoch": 2.8863373463005013, "grad_norm": 1.2918108701705933, "learning_rate": 3.79816987387515e-05, "loss": 0.15397857427597045, "memory(GiB)": 70.5, "step": 67370, "token_acc": 0.9572368421052632, "train_speed(iter/s)": 1.448106 }, { "epoch": 2.88655156162975, "grad_norm": 2.600475549697876, "learning_rate": 3.797516637589252e-05, "loss": 0.15097616910934447, "memory(GiB)": 70.5, "step": 67375, "token_acc": 0.9780564263322884, "train_speed(iter/s)": 1.448118 }, { "epoch": 2.886765776958999, "grad_norm": 0.8217527270317078, "learning_rate": 3.79686342308747e-05, "loss": 0.24574759006500244, "memory(GiB)": 70.5, "step": 67380, "token_acc": 0.9567099567099567, "train_speed(iter/s)": 1.448132 }, { "epoch": 2.886979992288248, "grad_norm": 2.5676867961883545, "learning_rate": 3.796210230381635e-05, "loss": 0.33406116962432864, "memory(GiB)": 70.5, "step": 67385, "token_acc": 0.9252669039145908, "train_speed(iter/s)": 1.448136 }, { "epoch": 2.887194207617497, "grad_norm": 3.017960548400879, "learning_rate": 3.795557059483581e-05, "loss": 0.41210379600524905, "memory(GiB)": 70.5, "step": 67390, "token_acc": 0.9050632911392406, "train_speed(iter/s)": 1.448147 }, { "epoch": 2.887408422946746, "grad_norm": 1.2825945615768433, "learning_rate": 3.794903910405141e-05, "loss": 0.2531184196472168, "memory(GiB)": 70.5, "step": 67395, "token_acc": 0.9453125, "train_speed(iter/s)": 1.448144 }, { "epoch": 2.887622638275995, "grad_norm": 1.8744757175445557, "learning_rate": 3.794250783158148e-05, "loss": 0.14506765604019164, "memory(GiB)": 70.5, "step": 67400, "token_acc": 0.9580152671755725, "train_speed(iter/s)": 1.448163 }, { "epoch": 2.887836853605244, "grad_norm": 0.7455418109893799, "learning_rate": 3.793597677754431e-05, "loss": 0.37750051021575926, "memory(GiB)": 70.5, "step": 67405, "token_acc": 0.9186746987951807, "train_speed(iter/s)": 1.448185 }, { "epoch": 2.8880510689344927, "grad_norm": 0.5883642435073853, "learning_rate": 3.792944594205825e-05, "loss": 0.13135693073272706, "memory(GiB)": 70.5, "step": 67410, "token_acc": 0.9707792207792207, "train_speed(iter/s)": 1.448186 }, { "epoch": 2.888265284263742, "grad_norm": 3.3966898918151855, "learning_rate": 3.792291532524159e-05, "loss": 0.40906553268432616, "memory(GiB)": 70.5, "step": 67415, "token_acc": 0.9067796610169492, "train_speed(iter/s)": 1.44819 }, { "epoch": 2.8884794995929908, "grad_norm": 9.494683265686035, "learning_rate": 3.791638492721266e-05, "loss": 0.5484580039978028, "memory(GiB)": 70.5, "step": 67420, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.448194 }, { "epoch": 2.8886937149222396, "grad_norm": 0.5603793859481812, "learning_rate": 3.790985474808975e-05, "loss": 0.4130890369415283, "memory(GiB)": 70.5, "step": 67425, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.448204 }, { "epoch": 2.888907930251489, "grad_norm": 3.2717103958129883, "learning_rate": 3.790332478799115e-05, "loss": 0.22059993743896483, "memory(GiB)": 70.5, "step": 67430, "token_acc": 0.9609375, "train_speed(iter/s)": 1.44821 }, { "epoch": 2.8891221455807377, "grad_norm": 3.2947134971618652, "learning_rate": 3.789679504703519e-05, "loss": 0.24920058250427246, "memory(GiB)": 70.5, "step": 67435, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.448212 }, { "epoch": 2.8893363609099865, "grad_norm": 1.4928665161132812, "learning_rate": 3.789026552534013e-05, "loss": 0.4900031089782715, "memory(GiB)": 70.5, "step": 67440, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.448229 }, { "epoch": 2.8895505762392357, "grad_norm": 6.105503082275391, "learning_rate": 3.788373622302426e-05, "loss": 0.42433757781982423, "memory(GiB)": 70.5, "step": 67445, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.44824 }, { "epoch": 2.8897647915684845, "grad_norm": 1.814147710800171, "learning_rate": 3.787720714020587e-05, "loss": 0.3718827247619629, "memory(GiB)": 70.5, "step": 67450, "token_acc": 0.9382716049382716, "train_speed(iter/s)": 1.44824 }, { "epoch": 2.8899790068977333, "grad_norm": 4.731106758117676, "learning_rate": 3.787067827700327e-05, "loss": 0.9182277679443359, "memory(GiB)": 70.5, "step": 67455, "token_acc": 0.8136645962732919, "train_speed(iter/s)": 1.448261 }, { "epoch": 2.8901932222269826, "grad_norm": 3.6166253089904785, "learning_rate": 3.7864149633534704e-05, "loss": 0.33563079833984377, "memory(GiB)": 70.5, "step": 67460, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.448259 }, { "epoch": 2.8904074375562314, "grad_norm": 6.340978622436523, "learning_rate": 3.785762120991844e-05, "loss": 0.34903483390808104, "memory(GiB)": 70.5, "step": 67465, "token_acc": 0.9307692307692308, "train_speed(iter/s)": 1.448268 }, { "epoch": 2.8906216528854802, "grad_norm": 0.7728412747383118, "learning_rate": 3.7851093006272775e-05, "loss": 0.382841157913208, "memory(GiB)": 70.5, "step": 67470, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.448282 }, { "epoch": 2.8908358682147295, "grad_norm": 7.055837631225586, "learning_rate": 3.784456502271594e-05, "loss": 0.6180291652679444, "memory(GiB)": 70.5, "step": 67475, "token_acc": 0.8620689655172413, "train_speed(iter/s)": 1.448288 }, { "epoch": 2.8910500835439783, "grad_norm": 0.5650585293769836, "learning_rate": 3.783803725936622e-05, "loss": 0.33963422775268554, "memory(GiB)": 70.5, "step": 67480, "token_acc": 0.92, "train_speed(iter/s)": 1.448284 }, { "epoch": 2.891264298873227, "grad_norm": 5.578777313232422, "learning_rate": 3.783150971634187e-05, "loss": 0.6579584598541259, "memory(GiB)": 70.5, "step": 67485, "token_acc": 0.8699551569506726, "train_speed(iter/s)": 1.448295 }, { "epoch": 2.8914785142024764, "grad_norm": 3.5158348083496094, "learning_rate": 3.782498239376113e-05, "loss": 0.33508827686309817, "memory(GiB)": 70.5, "step": 67490, "token_acc": 0.92, "train_speed(iter/s)": 1.448307 }, { "epoch": 2.891692729531725, "grad_norm": 3.344911575317383, "learning_rate": 3.781845529174226e-05, "loss": 0.778463888168335, "memory(GiB)": 70.5, "step": 67495, "token_acc": 0.8294314381270903, "train_speed(iter/s)": 1.448313 }, { "epoch": 2.891906944860974, "grad_norm": 5.224198818206787, "learning_rate": 3.7811928410403495e-05, "loss": 0.4105800151824951, "memory(GiB)": 70.5, "step": 67500, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.448327 }, { "epoch": 2.891906944860974, "eval_loss": 2.438832998275757, "eval_runtime": 11.4423, "eval_samples_per_second": 8.74, "eval_steps_per_second": 8.74, "eval_token_acc": 0.43283582089552236, "step": 67500 }, { "epoch": 2.8921211601902233, "grad_norm": 6.300090312957764, "learning_rate": 3.780540174986308e-05, "loss": 0.4822094440460205, "memory(GiB)": 70.5, "step": 67505, "token_acc": 0.5541082164328658, "train_speed(iter/s)": 1.447942 }, { "epoch": 2.892335375519472, "grad_norm": 4.933126926422119, "learning_rate": 3.7798875310239254e-05, "loss": 0.42020673751831056, "memory(GiB)": 70.5, "step": 67510, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.447943 }, { "epoch": 2.892549590848721, "grad_norm": 3.099299192428589, "learning_rate": 3.779234909165024e-05, "loss": 0.3065709829330444, "memory(GiB)": 70.5, "step": 67515, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.447949 }, { "epoch": 2.89276380617797, "grad_norm": 0.09859393537044525, "learning_rate": 3.7785823094214265e-05, "loss": 0.30836670398712157, "memory(GiB)": 70.5, "step": 67520, "token_acc": 0.9440559440559441, "train_speed(iter/s)": 1.447952 }, { "epoch": 2.892978021507219, "grad_norm": 3.809713840484619, "learning_rate": 3.777929731804958e-05, "loss": 0.44652228355407714, "memory(GiB)": 70.5, "step": 67525, "token_acc": 0.915068493150685, "train_speed(iter/s)": 1.447955 }, { "epoch": 2.8931922368364678, "grad_norm": 0.4162311851978302, "learning_rate": 3.777277176327439e-05, "loss": 0.22716042995452881, "memory(GiB)": 70.5, "step": 67530, "token_acc": 0.9522292993630573, "train_speed(iter/s)": 1.447961 }, { "epoch": 2.893406452165717, "grad_norm": 4.103150367736816, "learning_rate": 3.7766246430006904e-05, "loss": 0.6378963947296142, "memory(GiB)": 70.5, "step": 67535, "token_acc": 0.8664259927797834, "train_speed(iter/s)": 1.447961 }, { "epoch": 2.893620667494966, "grad_norm": 3.5943117141723633, "learning_rate": 3.7759721318365344e-05, "loss": 0.5090227127075195, "memory(GiB)": 70.5, "step": 67540, "token_acc": 0.8745980707395499, "train_speed(iter/s)": 1.447978 }, { "epoch": 2.8938348828242146, "grad_norm": 4.804305076599121, "learning_rate": 3.7753196428467915e-05, "loss": 0.3050595998764038, "memory(GiB)": 70.5, "step": 67545, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.447989 }, { "epoch": 2.894049098153464, "grad_norm": 3.1854050159454346, "learning_rate": 3.774667176043281e-05, "loss": 0.2949427366256714, "memory(GiB)": 70.5, "step": 67550, "token_acc": 0.9343065693430657, "train_speed(iter/s)": 1.447992 }, { "epoch": 2.8942633134827127, "grad_norm": 0.5888026356697083, "learning_rate": 3.7740147314378254e-05, "loss": 0.20536413192749023, "memory(GiB)": 70.5, "step": 67555, "token_acc": 0.9543859649122807, "train_speed(iter/s)": 1.447997 }, { "epoch": 2.8944775288119615, "grad_norm": 0.42541825771331787, "learning_rate": 3.7733623090422424e-05, "loss": 0.3093162775039673, "memory(GiB)": 70.5, "step": 67560, "token_acc": 0.930327868852459, "train_speed(iter/s)": 1.447999 }, { "epoch": 2.8946917441412108, "grad_norm": 2.339534044265747, "learning_rate": 3.772709908868351e-05, "loss": 0.26126394271850584, "memory(GiB)": 70.5, "step": 67565, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.447997 }, { "epoch": 2.8949059594704596, "grad_norm": 3.1624224185943604, "learning_rate": 3.772057530927973e-05, "loss": 0.3927292346954346, "memory(GiB)": 70.5, "step": 67570, "token_acc": 0.915625, "train_speed(iter/s)": 1.448002 }, { "epoch": 2.8951201747997084, "grad_norm": 3.4579286575317383, "learning_rate": 3.771405175232923e-05, "loss": 0.3717378616333008, "memory(GiB)": 70.5, "step": 67575, "token_acc": 0.9339622641509434, "train_speed(iter/s)": 1.44801 }, { "epoch": 2.8953343901289577, "grad_norm": 4.5448431968688965, "learning_rate": 3.7707528417950215e-05, "loss": 0.37340381145477297, "memory(GiB)": 70.5, "step": 67580, "token_acc": 0.9273255813953488, "train_speed(iter/s)": 1.448022 }, { "epoch": 2.8955486054582065, "grad_norm": 3.4704830646514893, "learning_rate": 3.770100530626086e-05, "loss": 0.302825403213501, "memory(GiB)": 70.5, "step": 67585, "token_acc": 0.9315068493150684, "train_speed(iter/s)": 1.448019 }, { "epoch": 2.8957628207874553, "grad_norm": 1.9862253665924072, "learning_rate": 3.769448241737932e-05, "loss": 0.22992997169494628, "memory(GiB)": 70.5, "step": 67590, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.448025 }, { "epoch": 2.8959770361167045, "grad_norm": 2.6256041526794434, "learning_rate": 3.7687959751423755e-05, "loss": 0.14215142726898194, "memory(GiB)": 70.5, "step": 67595, "token_acc": 0.9614243323442137, "train_speed(iter/s)": 1.448024 }, { "epoch": 2.8961912514459534, "grad_norm": 2.628854274749756, "learning_rate": 3.768143730851238e-05, "loss": 0.427754545211792, "memory(GiB)": 70.5, "step": 67600, "token_acc": 0.9084507042253521, "train_speed(iter/s)": 1.448042 }, { "epoch": 2.896405466775202, "grad_norm": 1.5190863609313965, "learning_rate": 3.767491508876331e-05, "loss": 0.3458396911621094, "memory(GiB)": 70.5, "step": 67605, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.448056 }, { "epoch": 2.8966196821044514, "grad_norm": 2.717557907104492, "learning_rate": 3.7668393092294723e-05, "loss": 0.4567598342895508, "memory(GiB)": 70.5, "step": 67610, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.44806 }, { "epoch": 2.8968338974337002, "grad_norm": 3.265171527862549, "learning_rate": 3.766187131922476e-05, "loss": 0.7348355770111084, "memory(GiB)": 70.5, "step": 67615, "token_acc": 0.8461538461538461, "train_speed(iter/s)": 1.448072 }, { "epoch": 2.897048112762949, "grad_norm": 2.1602156162261963, "learning_rate": 3.765534976967157e-05, "loss": 0.1645612597465515, "memory(GiB)": 70.5, "step": 67620, "token_acc": 0.9554794520547946, "train_speed(iter/s)": 1.448074 }, { "epoch": 2.8972623280921983, "grad_norm": 3.484055757522583, "learning_rate": 3.7648828443753305e-05, "loss": 0.41821913719177245, "memory(GiB)": 70.5, "step": 67625, "token_acc": 0.9, "train_speed(iter/s)": 1.44809 }, { "epoch": 2.897476543421447, "grad_norm": 3.1719765663146973, "learning_rate": 3.76423073415881e-05, "loss": 0.2410193920135498, "memory(GiB)": 70.5, "step": 67630, "token_acc": 0.9507575757575758, "train_speed(iter/s)": 1.448097 }, { "epoch": 2.897690758750696, "grad_norm": 2.4501402378082275, "learning_rate": 3.763578646329408e-05, "loss": 0.2652501344680786, "memory(GiB)": 70.5, "step": 67635, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.448117 }, { "epoch": 2.897904974079945, "grad_norm": 1.0491771697998047, "learning_rate": 3.76292658089894e-05, "loss": 0.2891498804092407, "memory(GiB)": 70.5, "step": 67640, "token_acc": 0.9543726235741445, "train_speed(iter/s)": 1.44812 }, { "epoch": 2.898119189409194, "grad_norm": 2.5386545658111572, "learning_rate": 3.7622745378792156e-05, "loss": 0.5327938556671142, "memory(GiB)": 70.5, "step": 67645, "token_acc": 0.8851963746223565, "train_speed(iter/s)": 1.44814 }, { "epoch": 2.898333404738443, "grad_norm": 4.638432502746582, "learning_rate": 3.761622517282051e-05, "loss": 0.3847452878952026, "memory(GiB)": 70.5, "step": 67650, "token_acc": 0.9067164179104478, "train_speed(iter/s)": 1.448158 }, { "epoch": 2.898547620067692, "grad_norm": 0.4308196008205414, "learning_rate": 3.7609705191192555e-05, "loss": 0.26043877601623533, "memory(GiB)": 70.5, "step": 67655, "token_acc": 0.9363957597173145, "train_speed(iter/s)": 1.448166 }, { "epoch": 2.898761835396941, "grad_norm": 1.7840312719345093, "learning_rate": 3.760318543402641e-05, "loss": 0.13524845838546753, "memory(GiB)": 70.5, "step": 67660, "token_acc": 0.9754601226993865, "train_speed(iter/s)": 1.448177 }, { "epoch": 2.8989760507261897, "grad_norm": 2.096053123474121, "learning_rate": 3.75966659014402e-05, "loss": 0.2989544153213501, "memory(GiB)": 70.5, "step": 67665, "token_acc": 0.9173228346456693, "train_speed(iter/s)": 1.448183 }, { "epoch": 2.899190266055439, "grad_norm": 1.9285271167755127, "learning_rate": 3.759014659355199e-05, "loss": 0.34816708564758303, "memory(GiB)": 70.5, "step": 67670, "token_acc": 0.9169675090252708, "train_speed(iter/s)": 1.448176 }, { "epoch": 2.8994044813846878, "grad_norm": 3.786750555038452, "learning_rate": 3.758362751047995e-05, "loss": 0.4887354373931885, "memory(GiB)": 70.5, "step": 67675, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.448189 }, { "epoch": 2.8996186967139366, "grad_norm": 3.037551164627075, "learning_rate": 3.7577108652342124e-05, "loss": 0.32048988342285156, "memory(GiB)": 70.5, "step": 67680, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.448192 }, { "epoch": 2.899832912043186, "grad_norm": 1.765465497970581, "learning_rate": 3.7570590019256645e-05, "loss": 0.4005606651306152, "memory(GiB)": 70.5, "step": 67685, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.448199 }, { "epoch": 2.9000471273724346, "grad_norm": 2.3959343433380127, "learning_rate": 3.756407161134159e-05, "loss": 0.11496585607528687, "memory(GiB)": 70.5, "step": 67690, "token_acc": 0.9762845849802372, "train_speed(iter/s)": 1.448212 }, { "epoch": 2.900261342701684, "grad_norm": 2.588968515396118, "learning_rate": 3.7557553428715026e-05, "loss": 0.3740550518035889, "memory(GiB)": 70.5, "step": 67695, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.448214 }, { "epoch": 2.9004755580309327, "grad_norm": 5.264646530151367, "learning_rate": 3.7551035471495065e-05, "loss": 0.47301855087280276, "memory(GiB)": 70.5, "step": 67700, "token_acc": 0.8760330578512396, "train_speed(iter/s)": 1.448217 }, { "epoch": 2.9006897733601815, "grad_norm": 3.052757740020752, "learning_rate": 3.7544517739799775e-05, "loss": 0.5077502250671386, "memory(GiB)": 70.5, "step": 67705, "token_acc": 0.8901098901098901, "train_speed(iter/s)": 1.448217 }, { "epoch": 2.900903988689431, "grad_norm": 0.9594593048095703, "learning_rate": 3.753800023374723e-05, "loss": 0.40351691246032717, "memory(GiB)": 70.5, "step": 67710, "token_acc": 0.9303135888501742, "train_speed(iter/s)": 1.448212 }, { "epoch": 2.9011182040186796, "grad_norm": 1.9862213134765625, "learning_rate": 3.7531482953455504e-05, "loss": 0.35221812725067136, "memory(GiB)": 70.5, "step": 67715, "token_acc": 0.9218106995884774, "train_speed(iter/s)": 1.448216 }, { "epoch": 2.9013324193479284, "grad_norm": 1.3302202224731445, "learning_rate": 3.752496589904265e-05, "loss": 0.19836781024932862, "memory(GiB)": 70.5, "step": 67720, "token_acc": 0.965034965034965, "train_speed(iter/s)": 1.448225 }, { "epoch": 2.9015466346771777, "grad_norm": 4.386117458343506, "learning_rate": 3.751844907062675e-05, "loss": 0.26109867095947265, "memory(GiB)": 70.5, "step": 67725, "token_acc": 0.958041958041958, "train_speed(iter/s)": 1.448243 }, { "epoch": 2.9017608500064265, "grad_norm": 4.508217811584473, "learning_rate": 3.7511932468325855e-05, "loss": 0.19665526151657103, "memory(GiB)": 70.5, "step": 67730, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.44826 }, { "epoch": 2.9019750653356753, "grad_norm": 2.7999212741851807, "learning_rate": 3.750541609225801e-05, "loss": 0.34843978881835935, "memory(GiB)": 70.5, "step": 67735, "token_acc": 0.9176029962546817, "train_speed(iter/s)": 1.448265 }, { "epoch": 2.9021892806649245, "grad_norm": 1.6105663776397705, "learning_rate": 3.749889994254126e-05, "loss": 0.4226216793060303, "memory(GiB)": 70.5, "step": 67740, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.448274 }, { "epoch": 2.9024034959941734, "grad_norm": 3.7846009731292725, "learning_rate": 3.7492384019293686e-05, "loss": 0.4584004878997803, "memory(GiB)": 70.5, "step": 67745, "token_acc": 0.9013605442176871, "train_speed(iter/s)": 1.448272 }, { "epoch": 2.902617711323422, "grad_norm": 1.8991121053695679, "learning_rate": 3.74858683226333e-05, "loss": 0.26105608940124514, "memory(GiB)": 70.5, "step": 67750, "token_acc": 0.9496644295302014, "train_speed(iter/s)": 1.448278 }, { "epoch": 2.9028319266526714, "grad_norm": 3.4212028980255127, "learning_rate": 3.747935285267816e-05, "loss": 0.3279649496078491, "memory(GiB)": 70.5, "step": 67755, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.448289 }, { "epoch": 2.9030461419819202, "grad_norm": 1.5401973724365234, "learning_rate": 3.747283760954629e-05, "loss": 0.3219198226928711, "memory(GiB)": 70.5, "step": 67760, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.448293 }, { "epoch": 2.903260357311169, "grad_norm": 0.42089512944221497, "learning_rate": 3.74663225933557e-05, "loss": 0.38065285682678224, "memory(GiB)": 70.5, "step": 67765, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.4483 }, { "epoch": 2.9034745726404183, "grad_norm": 1.5702241659164429, "learning_rate": 3.745980780422445e-05, "loss": 0.4592452049255371, "memory(GiB)": 70.5, "step": 67770, "token_acc": 0.89375, "train_speed(iter/s)": 1.448306 }, { "epoch": 2.903688787969667, "grad_norm": 4.578491687774658, "learning_rate": 3.7453293242270536e-05, "loss": 0.4362764835357666, "memory(GiB)": 70.5, "step": 67775, "token_acc": 0.8925619834710744, "train_speed(iter/s)": 1.44831 }, { "epoch": 2.903903003298916, "grad_norm": 2.710162878036499, "learning_rate": 3.744677890761198e-05, "loss": 0.7530575752258301, "memory(GiB)": 70.5, "step": 67780, "token_acc": 0.8546712802768166, "train_speed(iter/s)": 1.44832 }, { "epoch": 2.904117218628165, "grad_norm": 2.8845374584198, "learning_rate": 3.7440264800366815e-05, "loss": 0.5456203937530517, "memory(GiB)": 70.5, "step": 67785, "token_acc": 0.8780487804878049, "train_speed(iter/s)": 1.448325 }, { "epoch": 2.904331433957414, "grad_norm": 4.168403625488281, "learning_rate": 3.743375092065302e-05, "loss": 0.35518133640289307, "memory(GiB)": 70.5, "step": 67790, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.448336 }, { "epoch": 2.904545649286663, "grad_norm": 1.1309080123901367, "learning_rate": 3.7427237268588626e-05, "loss": 0.5198388576507569, "memory(GiB)": 70.5, "step": 67795, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.448357 }, { "epoch": 2.904759864615912, "grad_norm": 1.1953383684158325, "learning_rate": 3.7420723844291625e-05, "loss": 0.21716248989105225, "memory(GiB)": 70.5, "step": 67800, "token_acc": 0.9421221864951769, "train_speed(iter/s)": 1.448361 }, { "epoch": 2.904974079945161, "grad_norm": 0.37838563323020935, "learning_rate": 3.7414210647879996e-05, "loss": 0.33099842071533203, "memory(GiB)": 70.5, "step": 67805, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.448362 }, { "epoch": 2.9051882952744097, "grad_norm": 3.776196002960205, "learning_rate": 3.740769767947177e-05, "loss": 0.6958366394042969, "memory(GiB)": 70.5, "step": 67810, "token_acc": 0.8611111111111112, "train_speed(iter/s)": 1.44836 }, { "epoch": 2.905402510603659, "grad_norm": 1.841601014137268, "learning_rate": 3.740118493918489e-05, "loss": 0.23645904064178466, "memory(GiB)": 70.5, "step": 67815, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.448355 }, { "epoch": 2.9056167259329078, "grad_norm": 1.642853021621704, "learning_rate": 3.739467242713737e-05, "loss": 0.37792344093322755, "memory(GiB)": 70.5, "step": 67820, "token_acc": 0.9025270758122743, "train_speed(iter/s)": 1.448375 }, { "epoch": 2.9058309412621566, "grad_norm": 3.7296228408813477, "learning_rate": 3.7388160143447184e-05, "loss": 0.4739550590515137, "memory(GiB)": 70.5, "step": 67825, "token_acc": 0.8996282527881041, "train_speed(iter/s)": 1.448375 }, { "epoch": 2.906045156591406, "grad_norm": 2.293100357055664, "learning_rate": 3.738164808823231e-05, "loss": 0.21986565589904786, "memory(GiB)": 70.5, "step": 67830, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.448379 }, { "epoch": 2.9062593719206546, "grad_norm": 5.4700093269348145, "learning_rate": 3.737513626161073e-05, "loss": 0.7116509437561035, "memory(GiB)": 70.5, "step": 67835, "token_acc": 0.8520710059171598, "train_speed(iter/s)": 1.448388 }, { "epoch": 2.9064735872499035, "grad_norm": 1.853797435760498, "learning_rate": 3.736862466370038e-05, "loss": 0.3748700857162476, "memory(GiB)": 70.5, "step": 67840, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.448392 }, { "epoch": 2.9066878025791527, "grad_norm": 0.838111400604248, "learning_rate": 3.7362113294619264e-05, "loss": 0.1859172463417053, "memory(GiB)": 70.5, "step": 67845, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.448392 }, { "epoch": 2.9069020179084015, "grad_norm": 3.719611406326294, "learning_rate": 3.73556021544853e-05, "loss": 0.4582816123962402, "memory(GiB)": 70.5, "step": 67850, "token_acc": 0.8779661016949153, "train_speed(iter/s)": 1.448397 }, { "epoch": 2.9071162332376503, "grad_norm": 2.637657642364502, "learning_rate": 3.734909124341648e-05, "loss": 0.25450663566589354, "memory(GiB)": 70.5, "step": 67855, "token_acc": 0.9529411764705882, "train_speed(iter/s)": 1.448399 }, { "epoch": 2.9073304485668996, "grad_norm": 4.926001071929932, "learning_rate": 3.734258056153074e-05, "loss": 0.6502505779266358, "memory(GiB)": 70.5, "step": 67860, "token_acc": 0.8592375366568915, "train_speed(iter/s)": 1.448415 }, { "epoch": 2.9075446638961484, "grad_norm": 2.654332160949707, "learning_rate": 3.7336070108946016e-05, "loss": 0.39586827754974363, "memory(GiB)": 70.5, "step": 67865, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.448418 }, { "epoch": 2.907758879225397, "grad_norm": 4.348986625671387, "learning_rate": 3.732955988578027e-05, "loss": 0.37709980010986327, "memory(GiB)": 70.5, "step": 67870, "token_acc": 0.908745247148289, "train_speed(iter/s)": 1.448436 }, { "epoch": 2.9079730945546465, "grad_norm": 3.001940965652466, "learning_rate": 3.732304989215143e-05, "loss": 0.4826672554016113, "memory(GiB)": 70.5, "step": 67875, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.44845 }, { "epoch": 2.9081873098838953, "grad_norm": 7.137911319732666, "learning_rate": 3.731654012817743e-05, "loss": 0.6612221240997315, "memory(GiB)": 70.5, "step": 67880, "token_acc": 0.8591065292096219, "train_speed(iter/s)": 1.448461 }, { "epoch": 2.908401525213144, "grad_norm": 1.7334749698638916, "learning_rate": 3.7310030593976205e-05, "loss": 0.45337648391723634, "memory(GiB)": 70.5, "step": 67885, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.44847 }, { "epoch": 2.9086157405423934, "grad_norm": 1.9932074546813965, "learning_rate": 3.730352128966567e-05, "loss": 0.4890146255493164, "memory(GiB)": 70.5, "step": 67890, "token_acc": 0.8933333333333333, "train_speed(iter/s)": 1.448466 }, { "epoch": 2.908829955871642, "grad_norm": 7.451509475708008, "learning_rate": 3.7297012215363756e-05, "loss": 0.3733770132064819, "memory(GiB)": 70.5, "step": 67895, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.448471 }, { "epoch": 2.909044171200891, "grad_norm": 6.76476526260376, "learning_rate": 3.7290503371188404e-05, "loss": 0.7234157085418701, "memory(GiB)": 70.5, "step": 67900, "token_acc": 0.8599348534201955, "train_speed(iter/s)": 1.448482 }, { "epoch": 2.9092583865301402, "grad_norm": 3.769362688064575, "learning_rate": 3.72839947572575e-05, "loss": 0.3374791145324707, "memory(GiB)": 70.5, "step": 67905, "token_acc": 0.94, "train_speed(iter/s)": 1.448483 }, { "epoch": 2.909472601859389, "grad_norm": 1.3254145383834839, "learning_rate": 3.727748637368895e-05, "loss": 0.11239835023880004, "memory(GiB)": 70.5, "step": 67910, "token_acc": 0.9763513513513513, "train_speed(iter/s)": 1.448487 }, { "epoch": 2.909686817188638, "grad_norm": 2.117814302444458, "learning_rate": 3.727097822060068e-05, "loss": 0.4842940330505371, "memory(GiB)": 70.5, "step": 67915, "token_acc": 0.9037037037037037, "train_speed(iter/s)": 1.448484 }, { "epoch": 2.909901032517887, "grad_norm": 2.4196674823760986, "learning_rate": 3.7264470298110576e-05, "loss": 0.1897873878479004, "memory(GiB)": 70.5, "step": 67920, "token_acc": 0.9554794520547946, "train_speed(iter/s)": 1.448482 }, { "epoch": 2.910115247847136, "grad_norm": 3.2198212146759033, "learning_rate": 3.725796260633653e-05, "loss": 0.3758732318878174, "memory(GiB)": 70.5, "step": 67925, "token_acc": 0.9326599326599326, "train_speed(iter/s)": 1.448484 }, { "epoch": 2.9103294631763847, "grad_norm": 4.233128070831299, "learning_rate": 3.725145514539646e-05, "loss": 0.32414045333862307, "memory(GiB)": 70.5, "step": 67930, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.44849 }, { "epoch": 2.910543678505634, "grad_norm": 3.342904806137085, "learning_rate": 3.724494791540823e-05, "loss": 0.45610871315002444, "memory(GiB)": 70.5, "step": 67935, "token_acc": 0.90234375, "train_speed(iter/s)": 1.448487 }, { "epoch": 2.910757893834883, "grad_norm": 8.65498161315918, "learning_rate": 3.723844091648974e-05, "loss": 0.5506688594818115, "memory(GiB)": 70.5, "step": 67940, "token_acc": 0.8673835125448028, "train_speed(iter/s)": 1.448486 }, { "epoch": 2.9109721091641316, "grad_norm": 5.689695358276367, "learning_rate": 3.7231934148758865e-05, "loss": 0.2860711097717285, "memory(GiB)": 70.5, "step": 67945, "token_acc": 0.9423868312757202, "train_speed(iter/s)": 1.448493 }, { "epoch": 2.911186324493381, "grad_norm": 2.7540626525878906, "learning_rate": 3.7225427612333465e-05, "loss": 0.4624819278717041, "memory(GiB)": 70.5, "step": 67950, "token_acc": 0.9173553719008265, "train_speed(iter/s)": 1.448507 }, { "epoch": 2.9114005398226297, "grad_norm": 0.1807931810617447, "learning_rate": 3.7218921307331446e-05, "loss": 0.2738448143005371, "memory(GiB)": 70.5, "step": 67955, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.44851 }, { "epoch": 2.9116147551518785, "grad_norm": 1.9033808708190918, "learning_rate": 3.7212415233870646e-05, "loss": 0.2882621049880981, "memory(GiB)": 70.5, "step": 67960, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.448513 }, { "epoch": 2.9118289704811278, "grad_norm": 0.2934836745262146, "learning_rate": 3.7205909392068926e-05, "loss": 0.23726367950439453, "memory(GiB)": 70.5, "step": 67965, "token_acc": 0.9450867052023122, "train_speed(iter/s)": 1.448516 }, { "epoch": 2.9120431858103766, "grad_norm": 5.200509071350098, "learning_rate": 3.719940378204417e-05, "loss": 0.6000066280364991, "memory(GiB)": 70.5, "step": 67970, "token_acc": 0.8956521739130435, "train_speed(iter/s)": 1.448533 }, { "epoch": 2.9122574011396254, "grad_norm": 2.7145328521728516, "learning_rate": 3.7192898403914236e-05, "loss": 0.43511233329772947, "memory(GiB)": 70.5, "step": 67975, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.44855 }, { "epoch": 2.9124716164688746, "grad_norm": 0.42174267768859863, "learning_rate": 3.718639325779696e-05, "loss": 0.3524658679962158, "memory(GiB)": 70.5, "step": 67980, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.448547 }, { "epoch": 2.9126858317981235, "grad_norm": 5.464214324951172, "learning_rate": 3.717988834381017e-05, "loss": 0.44236130714416505, "memory(GiB)": 70.5, "step": 67985, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.448548 }, { "epoch": 2.9129000471273723, "grad_norm": 3.097761869430542, "learning_rate": 3.717338366207175e-05, "loss": 0.549461841583252, "memory(GiB)": 70.5, "step": 67990, "token_acc": 0.8925925925925926, "train_speed(iter/s)": 1.448565 }, { "epoch": 2.9131142624566215, "grad_norm": 3.605491876602173, "learning_rate": 3.71668792126995e-05, "loss": 0.6798386096954345, "memory(GiB)": 70.5, "step": 67995, "token_acc": 0.8355704697986577, "train_speed(iter/s)": 1.448561 }, { "epoch": 2.9133284777858703, "grad_norm": 2.5156283378601074, "learning_rate": 3.716037499581129e-05, "loss": 0.6109892368316651, "memory(GiB)": 70.5, "step": 68000, "token_acc": 0.8719723183391004, "train_speed(iter/s)": 1.448565 }, { "epoch": 2.9133284777858703, "eval_loss": 2.5067713260650635, "eval_runtime": 11.9646, "eval_samples_per_second": 8.358, "eval_steps_per_second": 8.358, "eval_token_acc": 0.4492131616595136, "step": 68000 }, { "epoch": 2.913542693115119, "grad_norm": 3.0858805179595947, "learning_rate": 3.7153871011524924e-05, "loss": 0.35081164836883544, "memory(GiB)": 70.5, "step": 68005, "token_acc": 0.5896414342629482, "train_speed(iter/s)": 1.448166 }, { "epoch": 2.9137569084443684, "grad_norm": 3.322232723236084, "learning_rate": 3.714736725995823e-05, "loss": 0.31016275882720945, "memory(GiB)": 70.5, "step": 68010, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.448189 }, { "epoch": 2.9139711237736172, "grad_norm": 2.6821789741516113, "learning_rate": 3.714086374122904e-05, "loss": 0.4240316867828369, "memory(GiB)": 70.5, "step": 68015, "token_acc": 0.9169960474308301, "train_speed(iter/s)": 1.44819 }, { "epoch": 2.914185339102866, "grad_norm": 2.791144609451294, "learning_rate": 3.713436045545518e-05, "loss": 0.5160602569580078, "memory(GiB)": 70.5, "step": 68020, "token_acc": 0.8896551724137931, "train_speed(iter/s)": 1.448208 }, { "epoch": 2.9143995544321153, "grad_norm": 2.0545568466186523, "learning_rate": 3.712785740275442e-05, "loss": 0.44081816673278806, "memory(GiB)": 70.5, "step": 68025, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.448213 }, { "epoch": 2.914613769761364, "grad_norm": 1.9982486963272095, "learning_rate": 3.7121354583244615e-05, "loss": 0.30972259044647216, "memory(GiB)": 70.5, "step": 68030, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.448216 }, { "epoch": 2.9148279850906134, "grad_norm": 4.46485710144043, "learning_rate": 3.711485199704353e-05, "loss": 0.5211190223693848, "memory(GiB)": 70.5, "step": 68035, "token_acc": 0.8876404494382022, "train_speed(iter/s)": 1.448234 }, { "epoch": 2.915042200419862, "grad_norm": 1.8860366344451904, "learning_rate": 3.710834964426901e-05, "loss": 0.24091558456420897, "memory(GiB)": 70.5, "step": 68040, "token_acc": 0.9455128205128205, "train_speed(iter/s)": 1.44825 }, { "epoch": 2.915256415749111, "grad_norm": 2.915290117263794, "learning_rate": 3.710184752503883e-05, "loss": 0.25520992279052734, "memory(GiB)": 70.5, "step": 68045, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.448257 }, { "epoch": 2.9154706310783602, "grad_norm": 4.466165065765381, "learning_rate": 3.709534563947078e-05, "loss": 0.20879440307617186, "memory(GiB)": 70.5, "step": 68050, "token_acc": 0.9466192170818505, "train_speed(iter/s)": 1.448258 }, { "epoch": 2.915684846407609, "grad_norm": 3.282010793685913, "learning_rate": 3.708884398768265e-05, "loss": 0.34175050258636475, "memory(GiB)": 70.5, "step": 68055, "token_acc": 0.9014598540145985, "train_speed(iter/s)": 1.44826 }, { "epoch": 2.915899061736858, "grad_norm": 2.947662353515625, "learning_rate": 3.708234256979223e-05, "loss": 0.40434832572937013, "memory(GiB)": 70.5, "step": 68060, "token_acc": 0.9145569620253164, "train_speed(iter/s)": 1.448275 }, { "epoch": 2.916113277066107, "grad_norm": 0.4763175845146179, "learning_rate": 3.7075841385917285e-05, "loss": 0.3846880435943604, "memory(GiB)": 70.5, "step": 68065, "token_acc": 0.9182389937106918, "train_speed(iter/s)": 1.448289 }, { "epoch": 2.916327492395356, "grad_norm": 1.2338873147964478, "learning_rate": 3.706934043617559e-05, "loss": 0.3090908765792847, "memory(GiB)": 70.5, "step": 68070, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.448289 }, { "epoch": 2.9165417077246047, "grad_norm": 2.9462060928344727, "learning_rate": 3.7062839720684926e-05, "loss": 0.37055678367614747, "memory(GiB)": 70.5, "step": 68075, "token_acc": 0.9143835616438356, "train_speed(iter/s)": 1.448291 }, { "epoch": 2.916755923053854, "grad_norm": 2.0588319301605225, "learning_rate": 3.705633923956306e-05, "loss": 0.5929404258728027, "memory(GiB)": 70.5, "step": 68080, "token_acc": 0.8724137931034482, "train_speed(iter/s)": 1.448293 }, { "epoch": 2.916970138383103, "grad_norm": 0.6783577799797058, "learning_rate": 3.7049838992927755e-05, "loss": 0.35113158226013186, "memory(GiB)": 70.5, "step": 68085, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.4483 }, { "epoch": 2.9171843537123516, "grad_norm": 2.121365547180176, "learning_rate": 3.704333898089676e-05, "loss": 0.15848034620285034, "memory(GiB)": 70.5, "step": 68090, "token_acc": 0.9638009049773756, "train_speed(iter/s)": 1.448327 }, { "epoch": 2.917398569041601, "grad_norm": 3.0275721549987793, "learning_rate": 3.703683920358783e-05, "loss": 0.4769552707672119, "memory(GiB)": 70.5, "step": 68095, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.448331 }, { "epoch": 2.9176127843708497, "grad_norm": 6.088558673858643, "learning_rate": 3.7030339661118726e-05, "loss": 0.7109987735748291, "memory(GiB)": 70.5, "step": 68100, "token_acc": 0.8543689320388349, "train_speed(iter/s)": 1.448327 }, { "epoch": 2.9178269997000985, "grad_norm": 2.933516263961792, "learning_rate": 3.7023840353607184e-05, "loss": 0.3469375133514404, "memory(GiB)": 70.5, "step": 68105, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.448326 }, { "epoch": 2.9180412150293478, "grad_norm": 2.6551690101623535, "learning_rate": 3.701734128117093e-05, "loss": 0.400360107421875, "memory(GiB)": 70.5, "step": 68110, "token_acc": 0.9161490683229814, "train_speed(iter/s)": 1.448333 }, { "epoch": 2.9182554303585966, "grad_norm": 0.9366979598999023, "learning_rate": 3.701084244392773e-05, "loss": 0.33584506511688234, "memory(GiB)": 70.5, "step": 68115, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.448346 }, { "epoch": 2.9184696456878454, "grad_norm": 6.327762603759766, "learning_rate": 3.70043438419953e-05, "loss": 0.5151030063629151, "memory(GiB)": 70.5, "step": 68120, "token_acc": 0.8919753086419753, "train_speed(iter/s)": 1.448358 }, { "epoch": 2.9186838610170946, "grad_norm": 3.785200357437134, "learning_rate": 3.699784547549139e-05, "loss": 0.4449461460113525, "memory(GiB)": 70.5, "step": 68125, "token_acc": 0.9022801302931596, "train_speed(iter/s)": 1.448365 }, { "epoch": 2.9188980763463435, "grad_norm": 3.1271004676818848, "learning_rate": 3.699134734453369e-05, "loss": 0.1811869740486145, "memory(GiB)": 70.5, "step": 68130, "token_acc": 0.9589905362776026, "train_speed(iter/s)": 1.448358 }, { "epoch": 2.9191122916755923, "grad_norm": 3.26046085357666, "learning_rate": 3.698484944923994e-05, "loss": 0.27257437705993653, "memory(GiB)": 70.5, "step": 68135, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.448362 }, { "epoch": 2.9193265070048415, "grad_norm": 2.4315643310546875, "learning_rate": 3.697835178972785e-05, "loss": 0.5807651042938232, "memory(GiB)": 70.5, "step": 68140, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.448361 }, { "epoch": 2.9195407223340903, "grad_norm": 8.239988327026367, "learning_rate": 3.6971854366115136e-05, "loss": 0.17075841426849364, "memory(GiB)": 70.5, "step": 68145, "token_acc": 0.9666666666666667, "train_speed(iter/s)": 1.448364 }, { "epoch": 2.919754937663339, "grad_norm": 4.324324607849121, "learning_rate": 3.69653571785195e-05, "loss": 0.4387547492980957, "memory(GiB)": 70.5, "step": 68150, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.448367 }, { "epoch": 2.9199691529925884, "grad_norm": 4.253852367401123, "learning_rate": 3.695886022705865e-05, "loss": 0.6175371170043945, "memory(GiB)": 70.5, "step": 68155, "token_acc": 0.867741935483871, "train_speed(iter/s)": 1.448368 }, { "epoch": 2.9201833683218372, "grad_norm": 1.2673723697662354, "learning_rate": 3.6952363511850274e-05, "loss": 0.3210650205612183, "memory(GiB)": 70.5, "step": 68160, "token_acc": 0.932, "train_speed(iter/s)": 1.448381 }, { "epoch": 2.920397583651086, "grad_norm": 3.2401962280273438, "learning_rate": 3.6945867033012084e-05, "loss": 0.4155735969543457, "memory(GiB)": 70.5, "step": 68165, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.448382 }, { "epoch": 2.9206117989803353, "grad_norm": 3.02972149848938, "learning_rate": 3.6939370790661745e-05, "loss": 0.48031005859375, "memory(GiB)": 70.5, "step": 68170, "token_acc": 0.8956228956228957, "train_speed(iter/s)": 1.448384 }, { "epoch": 2.920826014309584, "grad_norm": 1.1650118827819824, "learning_rate": 3.693287478491696e-05, "loss": 0.44468121528625487, "memory(GiB)": 70.5, "step": 68175, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.448399 }, { "epoch": 2.921040229638833, "grad_norm": 2.3010647296905518, "learning_rate": 3.6926379015895395e-05, "loss": 0.19987728595733642, "memory(GiB)": 70.5, "step": 68180, "token_acc": 0.9456869009584664, "train_speed(iter/s)": 1.448401 }, { "epoch": 2.921254444968082, "grad_norm": 3.488132953643799, "learning_rate": 3.691988348371473e-05, "loss": 0.2535308122634888, "memory(GiB)": 70.5, "step": 68185, "token_acc": 0.9375, "train_speed(iter/s)": 1.448402 }, { "epoch": 2.921468660297331, "grad_norm": 3.627194404602051, "learning_rate": 3.691338818849268e-05, "loss": 0.31006181240081787, "memory(GiB)": 70.5, "step": 68190, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.448407 }, { "epoch": 2.92168287562658, "grad_norm": 2.417560577392578, "learning_rate": 3.690689313034686e-05, "loss": 0.3683509826660156, "memory(GiB)": 70.5, "step": 68195, "token_acc": 0.9202279202279202, "train_speed(iter/s)": 1.448412 }, { "epoch": 2.921897090955829, "grad_norm": 2.7517032623291016, "learning_rate": 3.690039830939496e-05, "loss": 0.2483459234237671, "memory(GiB)": 70.5, "step": 68200, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.448425 }, { "epoch": 2.922111306285078, "grad_norm": 1.6790443658828735, "learning_rate": 3.689390372575463e-05, "loss": 0.20403244495391845, "memory(GiB)": 70.5, "step": 68205, "token_acc": 0.9453125, "train_speed(iter/s)": 1.448427 }, { "epoch": 2.9223255216143267, "grad_norm": 3.3660871982574463, "learning_rate": 3.688740937954354e-05, "loss": 0.22735223770141602, "memory(GiB)": 70.5, "step": 68210, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.448433 }, { "epoch": 2.922539736943576, "grad_norm": 4.074370861053467, "learning_rate": 3.6880915270879316e-05, "loss": 0.455535888671875, "memory(GiB)": 70.5, "step": 68215, "token_acc": 0.8917910447761194, "train_speed(iter/s)": 1.448434 }, { "epoch": 2.9227539522728248, "grad_norm": 1.162590742111206, "learning_rate": 3.6874421399879624e-05, "loss": 0.39028220176696776, "memory(GiB)": 70.5, "step": 68220, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.44844 }, { "epoch": 2.9229681676020736, "grad_norm": 2.2566428184509277, "learning_rate": 3.686792776666209e-05, "loss": 0.4559317111968994, "memory(GiB)": 70.5, "step": 68225, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.44845 }, { "epoch": 2.923182382931323, "grad_norm": 4.168013095855713, "learning_rate": 3.686143437134438e-05, "loss": 0.45693159103393555, "memory(GiB)": 70.5, "step": 68230, "token_acc": 0.8941605839416058, "train_speed(iter/s)": 1.448455 }, { "epoch": 2.9233965982605716, "grad_norm": 6.864307880401611, "learning_rate": 3.6854941214044105e-05, "loss": 0.5409673690795899, "memory(GiB)": 70.5, "step": 68235, "token_acc": 0.8490566037735849, "train_speed(iter/s)": 1.448459 }, { "epoch": 2.9236108135898204, "grad_norm": 3.3926374912261963, "learning_rate": 3.68484482948789e-05, "loss": 0.30749287605285647, "memory(GiB)": 70.5, "step": 68240, "token_acc": 0.9217391304347826, "train_speed(iter/s)": 1.448464 }, { "epoch": 2.9238250289190697, "grad_norm": 2.5674307346343994, "learning_rate": 3.684195561396639e-05, "loss": 0.37383835315704345, "memory(GiB)": 70.5, "step": 68245, "token_acc": 0.900398406374502, "train_speed(iter/s)": 1.448465 }, { "epoch": 2.9240392442483185, "grad_norm": 2.6016342639923096, "learning_rate": 3.6835463171424204e-05, "loss": 0.36697371006011964, "memory(GiB)": 70.5, "step": 68250, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.448471 }, { "epoch": 2.9242534595775673, "grad_norm": 3.0437285900115967, "learning_rate": 3.682897096736994e-05, "loss": 0.42805495262146, "memory(GiB)": 70.5, "step": 68255, "token_acc": 0.9273255813953488, "train_speed(iter/s)": 1.44847 }, { "epoch": 2.9244676749068166, "grad_norm": 4.621859550476074, "learning_rate": 3.682247900192121e-05, "loss": 0.5005968093872071, "memory(GiB)": 70.5, "step": 68260, "token_acc": 0.8931623931623932, "train_speed(iter/s)": 1.44847 }, { "epoch": 2.9246818902360654, "grad_norm": 1.861860990524292, "learning_rate": 3.6815987275195664e-05, "loss": 0.2754696846008301, "memory(GiB)": 70.5, "step": 68265, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.448475 }, { "epoch": 2.924896105565314, "grad_norm": 6.835525035858154, "learning_rate": 3.680949578731086e-05, "loss": 0.39803328514099123, "memory(GiB)": 70.5, "step": 68270, "token_acc": 0.9064516129032258, "train_speed(iter/s)": 1.448477 }, { "epoch": 2.9251103208945635, "grad_norm": 4.619340419769287, "learning_rate": 3.6803004538384425e-05, "loss": 0.40099177360534666, "memory(GiB)": 70.5, "step": 68275, "token_acc": 0.9154518950437318, "train_speed(iter/s)": 1.448482 }, { "epoch": 2.9253245362238123, "grad_norm": 2.71587872505188, "learning_rate": 3.679651352853394e-05, "loss": 0.29263863563537595, "memory(GiB)": 70.5, "step": 68280, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.448485 }, { "epoch": 2.925538751553061, "grad_norm": 2.82327938079834, "learning_rate": 3.679002275787698e-05, "loss": 0.4698473453521729, "memory(GiB)": 70.5, "step": 68285, "token_acc": 0.9015873015873016, "train_speed(iter/s)": 1.448483 }, { "epoch": 2.9257529668823103, "grad_norm": 5.4098005294799805, "learning_rate": 3.678353222653117e-05, "loss": 0.5389093399047852, "memory(GiB)": 70.5, "step": 68290, "token_acc": 0.8881987577639752, "train_speed(iter/s)": 1.448495 }, { "epoch": 2.925967182211559, "grad_norm": 2.578627109527588, "learning_rate": 3.6777041934614076e-05, "loss": 0.2834716796875, "memory(GiB)": 70.5, "step": 68295, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.448512 }, { "epoch": 2.926181397540808, "grad_norm": 4.402219772338867, "learning_rate": 3.6770551882243256e-05, "loss": 0.4915529727935791, "memory(GiB)": 70.5, "step": 68300, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.448515 }, { "epoch": 2.9263956128700572, "grad_norm": 2.6236956119537354, "learning_rate": 3.676406206953631e-05, "loss": 0.5410245895385742, "memory(GiB)": 70.5, "step": 68305, "token_acc": 0.8814102564102564, "train_speed(iter/s)": 1.448533 }, { "epoch": 2.926609828199306, "grad_norm": 0.33251503109931946, "learning_rate": 3.6757572496610794e-05, "loss": 0.15952955484390258, "memory(GiB)": 70.5, "step": 68310, "token_acc": 0.9601328903654485, "train_speed(iter/s)": 1.448536 }, { "epoch": 2.926824043528555, "grad_norm": 0.9630196690559387, "learning_rate": 3.675108316358426e-05, "loss": 0.20551471710205077, "memory(GiB)": 70.5, "step": 68315, "token_acc": 0.9566787003610109, "train_speed(iter/s)": 1.448538 }, { "epoch": 2.927038258857804, "grad_norm": 0.7640734314918518, "learning_rate": 3.67445940705743e-05, "loss": 0.14705803394317626, "memory(GiB)": 70.5, "step": 68320, "token_acc": 0.9669117647058824, "train_speed(iter/s)": 1.448539 }, { "epoch": 2.927252474187053, "grad_norm": 4.500670433044434, "learning_rate": 3.673810521769844e-05, "loss": 0.7081148147583007, "memory(GiB)": 70.5, "step": 68325, "token_acc": 0.8384146341463414, "train_speed(iter/s)": 1.448532 }, { "epoch": 2.9274666895163017, "grad_norm": 4.986722946166992, "learning_rate": 3.673161660507426e-05, "loss": 0.46341962814331056, "memory(GiB)": 70.5, "step": 68330, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.448545 }, { "epoch": 2.927680904845551, "grad_norm": 2.5816519260406494, "learning_rate": 3.672512823281926e-05, "loss": 0.07708157896995545, "memory(GiB)": 70.5, "step": 68335, "token_acc": 0.979757085020243, "train_speed(iter/s)": 1.448544 }, { "epoch": 2.9278951201748, "grad_norm": 3.7711267471313477, "learning_rate": 3.671864010105105e-05, "loss": 0.26838486194610595, "memory(GiB)": 70.5, "step": 68340, "token_acc": 0.9415204678362573, "train_speed(iter/s)": 1.448545 }, { "epoch": 2.9281093355040486, "grad_norm": 4.59187650680542, "learning_rate": 3.671215220988711e-05, "loss": 0.48224630355834963, "memory(GiB)": 70.5, "step": 68345, "token_acc": 0.9013605442176871, "train_speed(iter/s)": 1.448563 }, { "epoch": 2.928323550833298, "grad_norm": 1.7187634706497192, "learning_rate": 3.670566455944502e-05, "loss": 0.4637118339538574, "memory(GiB)": 70.5, "step": 68350, "token_acc": 0.9068100358422939, "train_speed(iter/s)": 1.448559 }, { "epoch": 2.9285377661625467, "grad_norm": 3.4699394702911377, "learning_rate": 3.6699177149842276e-05, "loss": 0.14228726625442506, "memory(GiB)": 70.5, "step": 68355, "token_acc": 0.9703703703703703, "train_speed(iter/s)": 1.44856 }, { "epoch": 2.9287519814917955, "grad_norm": 2.999743700027466, "learning_rate": 3.669268998119642e-05, "loss": 0.5138145446777344, "memory(GiB)": 70.5, "step": 68360, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.448566 }, { "epoch": 2.9289661968210448, "grad_norm": 2.0485873222351074, "learning_rate": 3.668620305362496e-05, "loss": 0.36344194412231445, "memory(GiB)": 70.5, "step": 68365, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.448565 }, { "epoch": 2.9291804121502936, "grad_norm": 2.8810372352600098, "learning_rate": 3.667971636724542e-05, "loss": 0.4613185405731201, "memory(GiB)": 70.5, "step": 68370, "token_acc": 0.9118541033434651, "train_speed(iter/s)": 1.448574 }, { "epoch": 2.9293946274795424, "grad_norm": 3.434542179107666, "learning_rate": 3.667322992217532e-05, "loss": 0.43770761489868165, "memory(GiB)": 70.5, "step": 68375, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.448579 }, { "epoch": 2.9296088428087916, "grad_norm": 2.588374614715576, "learning_rate": 3.666674371853217e-05, "loss": 0.13481065034866332, "memory(GiB)": 70.5, "step": 68380, "token_acc": 0.9639344262295082, "train_speed(iter/s)": 1.448592 }, { "epoch": 2.9298230581380404, "grad_norm": 5.413761615753174, "learning_rate": 3.6660257756433444e-05, "loss": 0.30007328987121584, "memory(GiB)": 70.5, "step": 68385, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.448594 }, { "epoch": 2.9300372734672893, "grad_norm": 1.9305607080459595, "learning_rate": 3.665377203599668e-05, "loss": 0.3986092805862427, "memory(GiB)": 70.5, "step": 68390, "token_acc": 0.913946587537092, "train_speed(iter/s)": 1.448593 }, { "epoch": 2.9302514887965385, "grad_norm": 2.479886531829834, "learning_rate": 3.664728655733936e-05, "loss": 0.32939133644104, "memory(GiB)": 70.5, "step": 68395, "token_acc": 0.9375, "train_speed(iter/s)": 1.448608 }, { "epoch": 2.9304657041257873, "grad_norm": 4.745147228240967, "learning_rate": 3.664080132057896e-05, "loss": 0.3736583709716797, "memory(GiB)": 70.5, "step": 68400, "token_acc": 0.910958904109589, "train_speed(iter/s)": 1.448611 }, { "epoch": 2.930679919455036, "grad_norm": 3.278843641281128, "learning_rate": 3.6634316325832954e-05, "loss": 0.4173441886901855, "memory(GiB)": 70.5, "step": 68405, "token_acc": 0.89568345323741, "train_speed(iter/s)": 1.448624 }, { "epoch": 2.9308941347842854, "grad_norm": 1.3406174182891846, "learning_rate": 3.662783157321887e-05, "loss": 0.11759241819381713, "memory(GiB)": 70.5, "step": 68410, "token_acc": 0.9770491803278688, "train_speed(iter/s)": 1.448628 }, { "epoch": 2.931108350113534, "grad_norm": 4.814792633056641, "learning_rate": 3.6621347062854164e-05, "loss": 0.7160830974578858, "memory(GiB)": 70.5, "step": 68415, "token_acc": 0.8541033434650456, "train_speed(iter/s)": 1.448639 }, { "epoch": 2.931322565442783, "grad_norm": 2.636530637741089, "learning_rate": 3.661486279485629e-05, "loss": 0.36199629306793213, "memory(GiB)": 70.5, "step": 68420, "token_acc": 0.9241379310344827, "train_speed(iter/s)": 1.448642 }, { "epoch": 2.9315367807720323, "grad_norm": 3.268195152282715, "learning_rate": 3.6608378769342746e-05, "loss": 0.4267704486846924, "memory(GiB)": 70.5, "step": 68425, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.448641 }, { "epoch": 2.931750996101281, "grad_norm": 0.8660794496536255, "learning_rate": 3.660189498643097e-05, "loss": 0.32707130908966064, "memory(GiB)": 70.5, "step": 68430, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.448647 }, { "epoch": 2.93196521143053, "grad_norm": 2.926863670349121, "learning_rate": 3.659541144623846e-05, "loss": 0.42340664863586425, "memory(GiB)": 70.5, "step": 68435, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.44865 }, { "epoch": 2.932179426759779, "grad_norm": 1.125244140625, "learning_rate": 3.658892814888264e-05, "loss": 0.2808080196380615, "memory(GiB)": 70.5, "step": 68440, "token_acc": 0.9342105263157895, "train_speed(iter/s)": 1.448656 }, { "epoch": 2.932393642089028, "grad_norm": 6.422170162200928, "learning_rate": 3.658244509448094e-05, "loss": 0.32743282318115235, "memory(GiB)": 70.5, "step": 68445, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.448668 }, { "epoch": 2.932607857418277, "grad_norm": 1.803343653678894, "learning_rate": 3.657596228315086e-05, "loss": 0.2865495204925537, "memory(GiB)": 70.5, "step": 68450, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.448659 }, { "epoch": 2.932822072747526, "grad_norm": 4.360405445098877, "learning_rate": 3.656947971500982e-05, "loss": 0.38343000411987305, "memory(GiB)": 70.5, "step": 68455, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.448663 }, { "epoch": 2.933036288076775, "grad_norm": 2.997400999069214, "learning_rate": 3.656299739017523e-05, "loss": 0.33374016284942626, "memory(GiB)": 70.5, "step": 68460, "token_acc": 0.9341692789968652, "train_speed(iter/s)": 1.448664 }, { "epoch": 2.9332505034060237, "grad_norm": 2.0809991359710693, "learning_rate": 3.655651530876456e-05, "loss": 0.28005595207214357, "memory(GiB)": 70.5, "step": 68465, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.448659 }, { "epoch": 2.933464718735273, "grad_norm": 3.854039192199707, "learning_rate": 3.6550033470895225e-05, "loss": 0.23237941265106202, "memory(GiB)": 70.5, "step": 68470, "token_acc": 0.9575163398692811, "train_speed(iter/s)": 1.448667 }, { "epoch": 2.9336789340645217, "grad_norm": 4.483768939971924, "learning_rate": 3.6543551876684656e-05, "loss": 0.41251797676086427, "memory(GiB)": 70.5, "step": 68475, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.448669 }, { "epoch": 2.9338931493937705, "grad_norm": 4.371690273284912, "learning_rate": 3.6537070526250244e-05, "loss": 0.49405660629272463, "memory(GiB)": 70.5, "step": 68480, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.448678 }, { "epoch": 2.93410736472302, "grad_norm": 3.97340989112854, "learning_rate": 3.653058941970945e-05, "loss": 0.3517828702926636, "memory(GiB)": 70.5, "step": 68485, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.448678 }, { "epoch": 2.9343215800522686, "grad_norm": 3.8515548706054688, "learning_rate": 3.652410855717966e-05, "loss": 0.2993659734725952, "memory(GiB)": 70.5, "step": 68490, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.448685 }, { "epoch": 2.9345357953815174, "grad_norm": 1.6249785423278809, "learning_rate": 3.651762793877829e-05, "loss": 0.31638078689575194, "memory(GiB)": 70.5, "step": 68495, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.448694 }, { "epoch": 2.9347500107107667, "grad_norm": 1.5625033378601074, "learning_rate": 3.6511147564622736e-05, "loss": 0.40027151107788084, "memory(GiB)": 70.5, "step": 68500, "token_acc": 0.9190031152647975, "train_speed(iter/s)": 1.448701 }, { "epoch": 2.9347500107107667, "eval_loss": 2.4306399822235107, "eval_runtime": 11.586, "eval_samples_per_second": 8.631, "eval_steps_per_second": 8.631, "eval_token_acc": 0.46371463714637146, "step": 68500 }, { "epoch": 2.9349642260400155, "grad_norm": 5.041499137878418, "learning_rate": 3.6504667434830394e-05, "loss": 0.30156726837158204, "memory(GiB)": 70.5, "step": 68505, "token_acc": 0.5803489439853077, "train_speed(iter/s)": 1.448313 }, { "epoch": 2.9351784413692643, "grad_norm": 0.07346577197313309, "learning_rate": 3.649818754951866e-05, "loss": 0.16443066596984862, "memory(GiB)": 70.5, "step": 68510, "token_acc": 0.9666666666666667, "train_speed(iter/s)": 1.448311 }, { "epoch": 2.9353926566985136, "grad_norm": 2.801046371459961, "learning_rate": 3.6491707908804926e-05, "loss": 0.2875931978225708, "memory(GiB)": 70.5, "step": 68515, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.448328 }, { "epoch": 2.9356068720277624, "grad_norm": 2.6963918209075928, "learning_rate": 3.648522851280658e-05, "loss": 0.28492145538330077, "memory(GiB)": 70.5, "step": 68520, "token_acc": 0.9181286549707602, "train_speed(iter/s)": 1.448336 }, { "epoch": 2.935821087357011, "grad_norm": 2.537327766418457, "learning_rate": 3.6478749361640996e-05, "loss": 0.46505441665649416, "memory(GiB)": 70.5, "step": 68525, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.448342 }, { "epoch": 2.9360353026862605, "grad_norm": 3.044809341430664, "learning_rate": 3.647227045542554e-05, "loss": 0.317980694770813, "memory(GiB)": 70.5, "step": 68530, "token_acc": 0.9098039215686274, "train_speed(iter/s)": 1.44834 }, { "epoch": 2.9362495180155093, "grad_norm": 4.473881721496582, "learning_rate": 3.646579179427761e-05, "loss": 0.5672770500183105, "memory(GiB)": 70.5, "step": 68535, "token_acc": 0.8778625954198473, "train_speed(iter/s)": 1.448359 }, { "epoch": 2.936463733344758, "grad_norm": 1.6647549867630005, "learning_rate": 3.6459313378314554e-05, "loss": 0.24354891777038573, "memory(GiB)": 70.5, "step": 68540, "token_acc": 0.9397590361445783, "train_speed(iter/s)": 1.448359 }, { "epoch": 2.9366779486740073, "grad_norm": 3.8480608463287354, "learning_rate": 3.645283520765373e-05, "loss": 0.3205977439880371, "memory(GiB)": 70.5, "step": 68545, "token_acc": 0.9338235294117647, "train_speed(iter/s)": 1.448362 }, { "epoch": 2.936892164003256, "grad_norm": 4.358080863952637, "learning_rate": 3.6446357282412514e-05, "loss": 0.40442347526550293, "memory(GiB)": 70.5, "step": 68550, "token_acc": 0.9044368600682594, "train_speed(iter/s)": 1.44837 }, { "epoch": 2.937106379332505, "grad_norm": 6.000283718109131, "learning_rate": 3.6439879602708224e-05, "loss": 0.5772449493408203, "memory(GiB)": 70.5, "step": 68555, "token_acc": 0.8906752411575563, "train_speed(iter/s)": 1.448386 }, { "epoch": 2.937320594661754, "grad_norm": 3.178584575653076, "learning_rate": 3.643340216865827e-05, "loss": 0.3838934421539307, "memory(GiB)": 70.5, "step": 68560, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.448388 }, { "epoch": 2.937534809991003, "grad_norm": 1.0844451189041138, "learning_rate": 3.642692498037994e-05, "loss": 0.2605139255523682, "memory(GiB)": 70.5, "step": 68565, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.44839 }, { "epoch": 2.937749025320252, "grad_norm": 0.28696706891059875, "learning_rate": 3.642044803799061e-05, "loss": 0.5277731418609619, "memory(GiB)": 70.5, "step": 68570, "token_acc": 0.8905775075987842, "train_speed(iter/s)": 1.448391 }, { "epoch": 2.937963240649501, "grad_norm": 2.25272536277771, "learning_rate": 3.6413971341607596e-05, "loss": 0.21955199241638185, "memory(GiB)": 70.5, "step": 68575, "token_acc": 0.9556962025316456, "train_speed(iter/s)": 1.448404 }, { "epoch": 2.93817745597875, "grad_norm": 1.1183253526687622, "learning_rate": 3.6407494891348244e-05, "loss": 0.3089275598526001, "memory(GiB)": 70.5, "step": 68580, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.448412 }, { "epoch": 2.9383916713079987, "grad_norm": 4.174056529998779, "learning_rate": 3.640101868732987e-05, "loss": 0.38869051933288573, "memory(GiB)": 70.5, "step": 68585, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.448427 }, { "epoch": 2.938605886637248, "grad_norm": 3.468385696411133, "learning_rate": 3.639454272966979e-05, "loss": 0.5762107849121094, "memory(GiB)": 70.5, "step": 68590, "token_acc": 0.8866666666666667, "train_speed(iter/s)": 1.44843 }, { "epoch": 2.938820101966497, "grad_norm": 5.445931434631348, "learning_rate": 3.6388067018485337e-05, "loss": 0.3603563070297241, "memory(GiB)": 70.5, "step": 68595, "token_acc": 0.912621359223301, "train_speed(iter/s)": 1.448435 }, { "epoch": 2.9390343172957456, "grad_norm": 3.557446002960205, "learning_rate": 3.638159155389383e-05, "loss": 0.3657860517501831, "memory(GiB)": 70.5, "step": 68600, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.448435 }, { "epoch": 2.939248532624995, "grad_norm": 3.7689695358276367, "learning_rate": 3.637511633601255e-05, "loss": 0.25400547981262206, "memory(GiB)": 70.5, "step": 68605, "token_acc": 0.9201520912547528, "train_speed(iter/s)": 1.448445 }, { "epoch": 2.9394627479542437, "grad_norm": 1.9595894813537598, "learning_rate": 3.636864136495883e-05, "loss": 0.24862475395202638, "memory(GiB)": 70.5, "step": 68610, "token_acc": 0.9504643962848297, "train_speed(iter/s)": 1.448453 }, { "epoch": 2.9396769632834925, "grad_norm": 7.789882183074951, "learning_rate": 3.636216664084995e-05, "loss": 0.34262571334838865, "memory(GiB)": 70.5, "step": 68615, "token_acc": 0.9148148148148149, "train_speed(iter/s)": 1.44846 }, { "epoch": 2.9398911786127417, "grad_norm": 4.106037616729736, "learning_rate": 3.635569216380322e-05, "loss": 0.4171159744262695, "memory(GiB)": 70.5, "step": 68620, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.448477 }, { "epoch": 2.9401053939419906, "grad_norm": 2.541008949279785, "learning_rate": 3.634921793393593e-05, "loss": 0.3011286735534668, "memory(GiB)": 70.5, "step": 68625, "token_acc": 0.9238095238095239, "train_speed(iter/s)": 1.448493 }, { "epoch": 2.9403196092712394, "grad_norm": 4.17448091506958, "learning_rate": 3.634274395136534e-05, "loss": 0.29750375747680663, "memory(GiB)": 70.5, "step": 68630, "token_acc": 0.92, "train_speed(iter/s)": 1.448495 }, { "epoch": 2.9405338246004886, "grad_norm": 4.048852920532227, "learning_rate": 3.633627021620877e-05, "loss": 0.3990520238876343, "memory(GiB)": 70.5, "step": 68635, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.448497 }, { "epoch": 2.9407480399297374, "grad_norm": 3.251354455947876, "learning_rate": 3.632979672858349e-05, "loss": 0.17404183149337768, "memory(GiB)": 70.5, "step": 68640, "token_acc": 0.9501661129568106, "train_speed(iter/s)": 1.448503 }, { "epoch": 2.9409622552589862, "grad_norm": 3.973600387573242, "learning_rate": 3.632332348860676e-05, "loss": 0.3049734354019165, "memory(GiB)": 70.5, "step": 68645, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.448506 }, { "epoch": 2.9411764705882355, "grad_norm": 0.38006147742271423, "learning_rate": 3.631685049639586e-05, "loss": 0.24090261459350587, "memory(GiB)": 70.5, "step": 68650, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.448511 }, { "epoch": 2.9413906859174843, "grad_norm": 2.792397975921631, "learning_rate": 3.6310377752068046e-05, "loss": 0.5245539665222168, "memory(GiB)": 70.5, "step": 68655, "token_acc": 0.9009287925696594, "train_speed(iter/s)": 1.448517 }, { "epoch": 2.941604901246733, "grad_norm": 2.461583137512207, "learning_rate": 3.6303905255740575e-05, "loss": 0.22740557193756103, "memory(GiB)": 70.5, "step": 68660, "token_acc": 0.9451476793248945, "train_speed(iter/s)": 1.448521 }, { "epoch": 2.9418191165759824, "grad_norm": 4.929706573486328, "learning_rate": 3.629743300753072e-05, "loss": 0.2850672721862793, "memory(GiB)": 70.5, "step": 68665, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.448528 }, { "epoch": 2.942033331905231, "grad_norm": 4.031632423400879, "learning_rate": 3.629096100755571e-05, "loss": 0.2881304740905762, "memory(GiB)": 70.5, "step": 68670, "token_acc": 0.9413489736070382, "train_speed(iter/s)": 1.448526 }, { "epoch": 2.94224754723448, "grad_norm": 2.71815824508667, "learning_rate": 3.6284489255932805e-05, "loss": 0.45978007316589353, "memory(GiB)": 70.5, "step": 68675, "token_acc": 0.9128787878787878, "train_speed(iter/s)": 1.448523 }, { "epoch": 2.9424617625637293, "grad_norm": 2.323263168334961, "learning_rate": 3.627801775277925e-05, "loss": 0.6784826755523682, "memory(GiB)": 70.5, "step": 68680, "token_acc": 0.8781163434903048, "train_speed(iter/s)": 1.448522 }, { "epoch": 2.942675977892978, "grad_norm": 3.0083765983581543, "learning_rate": 3.627154649821227e-05, "loss": 0.4679677486419678, "memory(GiB)": 70.5, "step": 68685, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.448531 }, { "epoch": 2.942890193222227, "grad_norm": 12.525483131408691, "learning_rate": 3.62650754923491e-05, "loss": 0.27561798095703127, "memory(GiB)": 70.5, "step": 68690, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.448535 }, { "epoch": 2.943104408551476, "grad_norm": 5.914331436157227, "learning_rate": 3.625860473530698e-05, "loss": 0.3809031963348389, "memory(GiB)": 70.5, "step": 68695, "token_acc": 0.9122257053291536, "train_speed(iter/s)": 1.448551 }, { "epoch": 2.943318623880725, "grad_norm": 3.049741744995117, "learning_rate": 3.62521342272031e-05, "loss": 0.39025282859802246, "memory(GiB)": 70.5, "step": 68700, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.448548 }, { "epoch": 2.9435328392099738, "grad_norm": 2.4266397953033447, "learning_rate": 3.624566396815473e-05, "loss": 0.26652159690856936, "memory(GiB)": 70.5, "step": 68705, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.448563 }, { "epoch": 2.943747054539223, "grad_norm": 1.4939494132995605, "learning_rate": 3.623919395827905e-05, "loss": 0.43891253471374514, "memory(GiB)": 70.5, "step": 68710, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.448565 }, { "epoch": 2.943961269868472, "grad_norm": 2.8893871307373047, "learning_rate": 3.623272419769329e-05, "loss": 0.37819409370422363, "memory(GiB)": 70.5, "step": 68715, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.448567 }, { "epoch": 2.9441754851977207, "grad_norm": 1.089648723602295, "learning_rate": 3.622625468651463e-05, "loss": 0.38320324420928953, "memory(GiB)": 70.5, "step": 68720, "token_acc": 0.8973384030418251, "train_speed(iter/s)": 1.448574 }, { "epoch": 2.94438970052697, "grad_norm": 2.7595102787017822, "learning_rate": 3.62197854248603e-05, "loss": 0.31622364521026614, "memory(GiB)": 70.5, "step": 68725, "token_acc": 0.9433333333333334, "train_speed(iter/s)": 1.448578 }, { "epoch": 2.9446039158562187, "grad_norm": 3.326239585876465, "learning_rate": 3.621331641284749e-05, "loss": 0.23570401668548585, "memory(GiB)": 70.5, "step": 68730, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.448576 }, { "epoch": 2.9448181311854675, "grad_norm": 3.127066135406494, "learning_rate": 3.620684765059337e-05, "loss": 0.4141895294189453, "memory(GiB)": 70.5, "step": 68735, "token_acc": 0.9136690647482014, "train_speed(iter/s)": 1.448581 }, { "epoch": 2.945032346514717, "grad_norm": 3.7951128482818604, "learning_rate": 3.620037913821516e-05, "loss": 0.3217073202133179, "memory(GiB)": 70.5, "step": 68740, "token_acc": 0.9389312977099237, "train_speed(iter/s)": 1.448588 }, { "epoch": 2.9452465618439656, "grad_norm": 3.649179458618164, "learning_rate": 3.619391087583002e-05, "loss": 0.3877542972564697, "memory(GiB)": 70.5, "step": 68745, "token_acc": 0.9029850746268657, "train_speed(iter/s)": 1.448586 }, { "epoch": 2.9454607771732144, "grad_norm": 1.7651010751724243, "learning_rate": 3.618744286355513e-05, "loss": 0.29950652122497556, "memory(GiB)": 70.5, "step": 68750, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.448589 }, { "epoch": 2.9456749925024637, "grad_norm": 2.6232006549835205, "learning_rate": 3.618097510150768e-05, "loss": 0.35355348587036134, "memory(GiB)": 70.5, "step": 68755, "token_acc": 0.9190283400809717, "train_speed(iter/s)": 1.448595 }, { "epoch": 2.9458892078317125, "grad_norm": 1.84345281124115, "learning_rate": 3.617450758980482e-05, "loss": 0.31662750244140625, "memory(GiB)": 70.5, "step": 68760, "token_acc": 0.9440559440559441, "train_speed(iter/s)": 1.448596 }, { "epoch": 2.9461034231609613, "grad_norm": 5.299279689788818, "learning_rate": 3.616804032856373e-05, "loss": 0.28591227531433105, "memory(GiB)": 70.5, "step": 68765, "token_acc": 0.9416666666666667, "train_speed(iter/s)": 1.448598 }, { "epoch": 2.9463176384902106, "grad_norm": 3.476170539855957, "learning_rate": 3.6161573317901564e-05, "loss": 0.6127137184143067, "memory(GiB)": 70.5, "step": 68770, "token_acc": 0.8733031674208145, "train_speed(iter/s)": 1.448605 }, { "epoch": 2.9465318538194594, "grad_norm": 7.341230392456055, "learning_rate": 3.6155106557935464e-05, "loss": 0.31979777812957766, "memory(GiB)": 70.5, "step": 68775, "token_acc": 0.9437751004016064, "train_speed(iter/s)": 1.448611 }, { "epoch": 2.946746069148708, "grad_norm": 1.2902593612670898, "learning_rate": 3.6148640048782604e-05, "loss": 0.22043471336364745, "memory(GiB)": 70.5, "step": 68780, "token_acc": 0.9588607594936709, "train_speed(iter/s)": 1.448619 }, { "epoch": 2.9469602844779574, "grad_norm": 2.1941921710968018, "learning_rate": 3.614217379056013e-05, "loss": 0.2518929958343506, "memory(GiB)": 70.5, "step": 68785, "token_acc": 0.9442724458204335, "train_speed(iter/s)": 1.448624 }, { "epoch": 2.9471744998072062, "grad_norm": 5.964407444000244, "learning_rate": 3.6135707783385183e-05, "loss": 0.6615216732025146, "memory(GiB)": 70.5, "step": 68790, "token_acc": 0.8819444444444444, "train_speed(iter/s)": 1.448632 }, { "epoch": 2.947388715136455, "grad_norm": 2.4202771186828613, "learning_rate": 3.6129242027374886e-05, "loss": 0.4708219051361084, "memory(GiB)": 70.5, "step": 68795, "token_acc": 0.9122257053291536, "train_speed(iter/s)": 1.448629 }, { "epoch": 2.9476029304657043, "grad_norm": 4.040325164794922, "learning_rate": 3.6122776522646386e-05, "loss": 0.5990802764892578, "memory(GiB)": 70.5, "step": 68800, "token_acc": 0.8662790697674418, "train_speed(iter/s)": 1.448628 }, { "epoch": 2.947817145794953, "grad_norm": 3.804215908050537, "learning_rate": 3.6116311269316804e-05, "loss": 0.49060893058776855, "memory(GiB)": 70.5, "step": 68805, "token_acc": 0.9192708333333334, "train_speed(iter/s)": 1.448629 }, { "epoch": 2.948031361124202, "grad_norm": 3.464890480041504, "learning_rate": 3.610984626750327e-05, "loss": 0.223305606842041, "memory(GiB)": 70.5, "step": 68810, "token_acc": 0.9518518518518518, "train_speed(iter/s)": 1.448637 }, { "epoch": 2.948245576453451, "grad_norm": 2.353804349899292, "learning_rate": 3.6103381517322905e-05, "loss": 0.5646705627441406, "memory(GiB)": 70.5, "step": 68815, "token_acc": 0.8821138211382114, "train_speed(iter/s)": 1.448645 }, { "epoch": 2.9484597917827, "grad_norm": 0.1495540291070938, "learning_rate": 3.609691701889281e-05, "loss": 0.35000569820404054, "memory(GiB)": 70.5, "step": 68820, "token_acc": 0.9429657794676806, "train_speed(iter/s)": 1.448649 }, { "epoch": 2.948674007111949, "grad_norm": 0.4471798837184906, "learning_rate": 3.6090452772330115e-05, "loss": 0.20046374797821045, "memory(GiB)": 70.5, "step": 68825, "token_acc": 0.9606557377049181, "train_speed(iter/s)": 1.448668 }, { "epoch": 2.948888222441198, "grad_norm": 2.8004159927368164, "learning_rate": 3.6083988777751916e-05, "loss": 0.34934267997741697, "memory(GiB)": 70.5, "step": 68830, "token_acc": 0.9287749287749287, "train_speed(iter/s)": 1.448672 }, { "epoch": 2.949102437770447, "grad_norm": 6.511341571807861, "learning_rate": 3.60775250352753e-05, "loss": 0.6652964591979981, "memory(GiB)": 70.5, "step": 68835, "token_acc": 0.8463855421686747, "train_speed(iter/s)": 1.448675 }, { "epoch": 2.9493166530996957, "grad_norm": 2.342341423034668, "learning_rate": 3.6071061545017395e-05, "loss": 0.5122045993804931, "memory(GiB)": 70.5, "step": 68840, "token_acc": 0.8737201365187713, "train_speed(iter/s)": 1.448675 }, { "epoch": 2.949530868428945, "grad_norm": 0.843584418296814, "learning_rate": 3.6064598307095266e-05, "loss": 0.2448648691177368, "memory(GiB)": 70.5, "step": 68845, "token_acc": 0.948905109489051, "train_speed(iter/s)": 1.448675 }, { "epoch": 2.9497450837581938, "grad_norm": 4.340083122253418, "learning_rate": 3.6058135321625995e-05, "loss": 0.3528769493103027, "memory(GiB)": 70.5, "step": 68850, "token_acc": 0.908745247148289, "train_speed(iter/s)": 1.448681 }, { "epoch": 2.9499592990874426, "grad_norm": 3.168048143386841, "learning_rate": 3.60516725887267e-05, "loss": 0.42436957359313965, "memory(GiB)": 70.5, "step": 68855, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.448697 }, { "epoch": 2.950173514416692, "grad_norm": 1.3541160821914673, "learning_rate": 3.604521010851445e-05, "loss": 0.4254271984100342, "memory(GiB)": 70.5, "step": 68860, "token_acc": 0.8957528957528957, "train_speed(iter/s)": 1.4487 }, { "epoch": 2.9503877297459407, "grad_norm": 3.6646337509155273, "learning_rate": 3.60387478811063e-05, "loss": 0.3790867805480957, "memory(GiB)": 70.5, "step": 68865, "token_acc": 0.9102564102564102, "train_speed(iter/s)": 1.448702 }, { "epoch": 2.9506019450751895, "grad_norm": 4.948216915130615, "learning_rate": 3.603228590661933e-05, "loss": 0.5552419185638428, "memory(GiB)": 70.5, "step": 68870, "token_acc": 0.85546875, "train_speed(iter/s)": 1.448718 }, { "epoch": 2.9508161604044387, "grad_norm": 1.8584794998168945, "learning_rate": 3.602582418517061e-05, "loss": 0.46191978454589844, "memory(GiB)": 70.5, "step": 68875, "token_acc": 0.8909090909090909, "train_speed(iter/s)": 1.448724 }, { "epoch": 2.9510303757336875, "grad_norm": 1.3187328577041626, "learning_rate": 3.601936271687718e-05, "loss": 0.29245948791503906, "memory(GiB)": 70.5, "step": 68880, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.448729 }, { "epoch": 2.9512445910629364, "grad_norm": 3.1321187019348145, "learning_rate": 3.601290150185612e-05, "loss": 0.31244540214538574, "memory(GiB)": 70.5, "step": 68885, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.448728 }, { "epoch": 2.9514588063921856, "grad_norm": 2.070000648498535, "learning_rate": 3.6006440540224465e-05, "loss": 0.21208245754241944, "memory(GiB)": 70.5, "step": 68890, "token_acc": 0.9478260869565217, "train_speed(iter/s)": 1.44873 }, { "epoch": 2.9516730217214344, "grad_norm": 5.03693962097168, "learning_rate": 3.599997983209927e-05, "loss": 0.38823232650756834, "memory(GiB)": 70.5, "step": 68895, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.448735 }, { "epoch": 2.9518872370506832, "grad_norm": 4.602385520935059, "learning_rate": 3.5993519377597576e-05, "loss": 0.6718157768249512, "memory(GiB)": 70.5, "step": 68900, "token_acc": 0.8809523809523809, "train_speed(iter/s)": 1.448753 }, { "epoch": 2.9521014523799325, "grad_norm": 4.1587114334106445, "learning_rate": 3.5987059176836404e-05, "loss": 0.2848172664642334, "memory(GiB)": 70.5, "step": 68905, "token_acc": 0.9486301369863014, "train_speed(iter/s)": 1.448765 }, { "epoch": 2.9523156677091813, "grad_norm": 2.133251667022705, "learning_rate": 3.598059922993282e-05, "loss": 0.3321115732192993, "memory(GiB)": 70.5, "step": 68910, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.448761 }, { "epoch": 2.95252988303843, "grad_norm": 2.8190481662750244, "learning_rate": 3.597413953700382e-05, "loss": 0.24955480098724364, "memory(GiB)": 70.5, "step": 68915, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.448776 }, { "epoch": 2.9527440983676794, "grad_norm": 1.7465879917144775, "learning_rate": 3.596768009816644e-05, "loss": 0.23136765956878663, "memory(GiB)": 70.5, "step": 68920, "token_acc": 0.9467680608365019, "train_speed(iter/s)": 1.448788 }, { "epoch": 2.952958313696928, "grad_norm": 3.1468305587768555, "learning_rate": 3.5961220913537683e-05, "loss": 0.3847119092941284, "memory(GiB)": 70.5, "step": 68925, "token_acc": 0.916083916083916, "train_speed(iter/s)": 1.448795 }, { "epoch": 2.953172529026177, "grad_norm": 4.3745436668396, "learning_rate": 3.5954761983234595e-05, "loss": 0.46385650634765624, "memory(GiB)": 70.5, "step": 68930, "token_acc": 0.9054441260744985, "train_speed(iter/s)": 1.448794 }, { "epoch": 2.9533867443554263, "grad_norm": 2.7824900150299072, "learning_rate": 3.594830330737417e-05, "loss": 0.6395976543426514, "memory(GiB)": 70.5, "step": 68935, "token_acc": 0.8611111111111112, "train_speed(iter/s)": 1.448812 }, { "epoch": 2.953600959684675, "grad_norm": 2.4373674392700195, "learning_rate": 3.5941844886073416e-05, "loss": 0.35182340145111085, "memory(GiB)": 70.5, "step": 68940, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.448818 }, { "epoch": 2.953815175013924, "grad_norm": 1.664318561553955, "learning_rate": 3.5935386719449324e-05, "loss": 0.6227609634399414, "memory(GiB)": 70.5, "step": 68945, "token_acc": 0.8754863813229572, "train_speed(iter/s)": 1.448834 }, { "epoch": 2.954029390343173, "grad_norm": 1.8290131092071533, "learning_rate": 3.5928928807618896e-05, "loss": 0.3528812646865845, "memory(GiB)": 70.5, "step": 68950, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.44884 }, { "epoch": 2.954243605672422, "grad_norm": 1.6828086376190186, "learning_rate": 3.592247115069913e-05, "loss": 0.47335214614868165, "memory(GiB)": 70.5, "step": 68955, "token_acc": 0.8957528957528957, "train_speed(iter/s)": 1.44886 }, { "epoch": 2.9544578210016708, "grad_norm": 4.037561416625977, "learning_rate": 3.5916013748807e-05, "loss": 0.3100311756134033, "memory(GiB)": 70.5, "step": 68960, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.448871 }, { "epoch": 2.95467203633092, "grad_norm": 2.2130110263824463, "learning_rate": 3.590955660205948e-05, "loss": 0.3941420793533325, "memory(GiB)": 70.5, "step": 68965, "token_acc": 0.912621359223301, "train_speed(iter/s)": 1.448878 }, { "epoch": 2.954886251660169, "grad_norm": 5.702152252197266, "learning_rate": 3.590309971057358e-05, "loss": 0.5547599792480469, "memory(GiB)": 70.5, "step": 68970, "token_acc": 0.8829787234042553, "train_speed(iter/s)": 1.44888 }, { "epoch": 2.9551004669894176, "grad_norm": 2.7223446369171143, "learning_rate": 3.5896643074466246e-05, "loss": 0.5236962318420411, "memory(GiB)": 70.5, "step": 68975, "token_acc": 0.8664259927797834, "train_speed(iter/s)": 1.448894 }, { "epoch": 2.955314682318667, "grad_norm": 2.3847103118896484, "learning_rate": 3.5890186693854444e-05, "loss": 0.36154866218566895, "memory(GiB)": 70.5, "step": 68980, "token_acc": 0.919732441471572, "train_speed(iter/s)": 1.448894 }, { "epoch": 2.9555288976479157, "grad_norm": 0.33095651865005493, "learning_rate": 3.5883730568855156e-05, "loss": 0.32874877452850343, "memory(GiB)": 70.5, "step": 68985, "token_acc": 0.9465648854961832, "train_speed(iter/s)": 1.448917 }, { "epoch": 2.9557431129771645, "grad_norm": 1.6629760265350342, "learning_rate": 3.587727469958532e-05, "loss": 0.49701681137084963, "memory(GiB)": 70.5, "step": 68990, "token_acc": 0.9211267605633803, "train_speed(iter/s)": 1.44893 }, { "epoch": 2.955957328306414, "grad_norm": 3.6660234928131104, "learning_rate": 3.5870819086161913e-05, "loss": 0.6729121685028077, "memory(GiB)": 70.5, "step": 68995, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.448945 }, { "epoch": 2.9561715436356626, "grad_norm": 4.887330532073975, "learning_rate": 3.5864363728701855e-05, "loss": 0.3406259536743164, "memory(GiB)": 70.5, "step": 69000, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.448948 }, { "epoch": 2.9561715436356626, "eval_loss": 2.5882532596588135, "eval_runtime": 11.0179, "eval_samples_per_second": 9.076, "eval_steps_per_second": 9.076, "eval_token_acc": 0.47955974842767296, "step": 69000 }, { "epoch": 2.9563857589649114, "grad_norm": 2.773432970046997, "learning_rate": 3.5857908627322124e-05, "loss": 0.22473454475402832, "memory(GiB)": 70.5, "step": 69005, "token_acc": 0.646998982706002, "train_speed(iter/s)": 1.448591 }, { "epoch": 2.9565999742941607, "grad_norm": 1.8912965059280396, "learning_rate": 3.585145378213963e-05, "loss": 0.29583988189697263, "memory(GiB)": 70.5, "step": 69010, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.448593 }, { "epoch": 2.9568141896234095, "grad_norm": 2.3957431316375732, "learning_rate": 3.584499919327135e-05, "loss": 0.43692784309387206, "memory(GiB)": 70.5, "step": 69015, "token_acc": 0.912621359223301, "train_speed(iter/s)": 1.448595 }, { "epoch": 2.9570284049526583, "grad_norm": 1.679940104484558, "learning_rate": 3.583854486083417e-05, "loss": 0.4783369541168213, "memory(GiB)": 70.5, "step": 69020, "token_acc": 0.8770764119601329, "train_speed(iter/s)": 1.448618 }, { "epoch": 2.9572426202819075, "grad_norm": 7.914979457855225, "learning_rate": 3.583209078494503e-05, "loss": 0.37102856636047366, "memory(GiB)": 70.5, "step": 69025, "token_acc": 0.92, "train_speed(iter/s)": 1.448634 }, { "epoch": 2.9574568356111564, "grad_norm": 3.655667543411255, "learning_rate": 3.582563696572087e-05, "loss": 0.3606879472732544, "memory(GiB)": 70.5, "step": 69030, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.448631 }, { "epoch": 2.957671050940405, "grad_norm": 5.671210289001465, "learning_rate": 3.581918340327858e-05, "loss": 0.3778620719909668, "memory(GiB)": 70.5, "step": 69035, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.448635 }, { "epoch": 2.9578852662696544, "grad_norm": 4.2183098793029785, "learning_rate": 3.581273009773509e-05, "loss": 0.48813672065734864, "memory(GiB)": 70.5, "step": 69040, "token_acc": 0.8843537414965986, "train_speed(iter/s)": 1.448651 }, { "epoch": 2.9580994815989032, "grad_norm": 2.683342218399048, "learning_rate": 3.5806277049207315e-05, "loss": 0.4466232776641846, "memory(GiB)": 70.5, "step": 69045, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.448657 }, { "epoch": 2.958313696928152, "grad_norm": 3.09743332862854, "learning_rate": 3.579982425781213e-05, "loss": 0.2374246597290039, "memory(GiB)": 70.5, "step": 69050, "token_acc": 0.9601593625498008, "train_speed(iter/s)": 1.448663 }, { "epoch": 2.9585279122574013, "grad_norm": 5.110622406005859, "learning_rate": 3.579337172366646e-05, "loss": 0.4190669536590576, "memory(GiB)": 70.5, "step": 69055, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.448666 }, { "epoch": 2.95874212758665, "grad_norm": 3.3250021934509277, "learning_rate": 3.578691944688719e-05, "loss": 0.4164477825164795, "memory(GiB)": 70.5, "step": 69060, "token_acc": 0.9083969465648855, "train_speed(iter/s)": 1.448685 }, { "epoch": 2.958956342915899, "grad_norm": 2.5656909942626953, "learning_rate": 3.5780467427591194e-05, "loss": 0.4286682605743408, "memory(GiB)": 70.5, "step": 69065, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.448692 }, { "epoch": 2.959170558245148, "grad_norm": 2.684544086456299, "learning_rate": 3.577401566589535e-05, "loss": 0.3242535352706909, "memory(GiB)": 70.5, "step": 69070, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.448695 }, { "epoch": 2.959384773574397, "grad_norm": 1.9919320344924927, "learning_rate": 3.576756416191659e-05, "loss": 0.17568353414535523, "memory(GiB)": 70.5, "step": 69075, "token_acc": 0.9560117302052786, "train_speed(iter/s)": 1.448703 }, { "epoch": 2.959598988903646, "grad_norm": 2.9152801036834717, "learning_rate": 3.5761112915771756e-05, "loss": 0.3437760353088379, "memory(GiB)": 70.5, "step": 69080, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.448707 }, { "epoch": 2.959813204232895, "grad_norm": 3.5551114082336426, "learning_rate": 3.57546619275777e-05, "loss": 0.2953130483627319, "memory(GiB)": 70.5, "step": 69085, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.448714 }, { "epoch": 2.960027419562144, "grad_norm": 1.775273084640503, "learning_rate": 3.574821119745133e-05, "loss": 0.12572847604751586, "memory(GiB)": 70.5, "step": 69090, "token_acc": 0.966804979253112, "train_speed(iter/s)": 1.448718 }, { "epoch": 2.9602416348913927, "grad_norm": 2.5432817935943604, "learning_rate": 3.5741760725509464e-05, "loss": 0.35720961093902587, "memory(GiB)": 70.5, "step": 69095, "token_acc": 0.9085545722713865, "train_speed(iter/s)": 1.448727 }, { "epoch": 2.960455850220642, "grad_norm": 3.6390891075134277, "learning_rate": 3.5735310511868994e-05, "loss": 0.30710687637329104, "memory(GiB)": 70.5, "step": 69100, "token_acc": 0.9212121212121213, "train_speed(iter/s)": 1.448732 }, { "epoch": 2.9606700655498908, "grad_norm": 3.997138261795044, "learning_rate": 3.572886055664675e-05, "loss": 0.6149721622467041, "memory(GiB)": 70.5, "step": 69105, "token_acc": 0.8797250859106529, "train_speed(iter/s)": 1.44875 }, { "epoch": 2.9608842808791396, "grad_norm": 4.747529983520508, "learning_rate": 3.5722410859959574e-05, "loss": 0.4261322498321533, "memory(GiB)": 70.5, "step": 69110, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.448771 }, { "epoch": 2.961098496208389, "grad_norm": 3.9227187633514404, "learning_rate": 3.571596142192433e-05, "loss": 0.4577311992645264, "memory(GiB)": 70.5, "step": 69115, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.448771 }, { "epoch": 2.9613127115376376, "grad_norm": 0.9685602784156799, "learning_rate": 3.570951224265785e-05, "loss": 0.6424359798431396, "memory(GiB)": 70.5, "step": 69120, "token_acc": 0.8315789473684211, "train_speed(iter/s)": 1.448779 }, { "epoch": 2.9615269268668865, "grad_norm": 2.235456705093384, "learning_rate": 3.570306332227694e-05, "loss": 0.2714141607284546, "memory(GiB)": 70.5, "step": 69125, "token_acc": 0.9381107491856677, "train_speed(iter/s)": 1.448786 }, { "epoch": 2.9617411421961357, "grad_norm": 3.645343542098999, "learning_rate": 3.5696614660898465e-05, "loss": 0.35677447319030764, "memory(GiB)": 70.5, "step": 69130, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.448791 }, { "epoch": 2.9619553575253845, "grad_norm": 7.363175868988037, "learning_rate": 3.5690166258639226e-05, "loss": 0.5860123634338379, "memory(GiB)": 70.5, "step": 69135, "token_acc": 0.8790322580645161, "train_speed(iter/s)": 1.448796 }, { "epoch": 2.9621695728546333, "grad_norm": 4.107384204864502, "learning_rate": 3.568371811561606e-05, "loss": 0.2222529172897339, "memory(GiB)": 70.5, "step": 69140, "token_acc": 0.9316546762589928, "train_speed(iter/s)": 1.448808 }, { "epoch": 2.9623837881838826, "grad_norm": 4.890659809112549, "learning_rate": 3.5677270231945745e-05, "loss": 0.28700096607208253, "memory(GiB)": 70.5, "step": 69145, "token_acc": 0.9201680672268907, "train_speed(iter/s)": 1.448812 }, { "epoch": 2.9625980035131314, "grad_norm": 2.430370330810547, "learning_rate": 3.5670822607745134e-05, "loss": 0.3745074510574341, "memory(GiB)": 70.5, "step": 69150, "token_acc": 0.9272030651340997, "train_speed(iter/s)": 1.448814 }, { "epoch": 2.96281221884238, "grad_norm": 2.376940965652466, "learning_rate": 3.566437524313101e-05, "loss": 0.28591241836547854, "memory(GiB)": 70.5, "step": 69155, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.448815 }, { "epoch": 2.9630264341716295, "grad_norm": 2.4438562393188477, "learning_rate": 3.5657928138220184e-05, "loss": 0.359742259979248, "memory(GiB)": 70.5, "step": 69160, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.448823 }, { "epoch": 2.9632406495008783, "grad_norm": 4.772762298583984, "learning_rate": 3.565148129312944e-05, "loss": 0.3528266429901123, "memory(GiB)": 70.5, "step": 69165, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.44882 }, { "epoch": 2.963454864830127, "grad_norm": 0.7515957951545715, "learning_rate": 3.564503470797556e-05, "loss": 0.32032132148742676, "memory(GiB)": 70.5, "step": 69170, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.448825 }, { "epoch": 2.9636690801593764, "grad_norm": 6.611395359039307, "learning_rate": 3.563858838287536e-05, "loss": 0.4356846809387207, "memory(GiB)": 70.5, "step": 69175, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.448832 }, { "epoch": 2.963883295488625, "grad_norm": 3.093428134918213, "learning_rate": 3.5632142317945596e-05, "loss": 0.2671316146850586, "memory(GiB)": 70.5, "step": 69180, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.448835 }, { "epoch": 2.964097510817874, "grad_norm": 2.5967071056365967, "learning_rate": 3.562569651330305e-05, "loss": 0.3461653470993042, "memory(GiB)": 70.5, "step": 69185, "token_acc": 0.9375, "train_speed(iter/s)": 1.448838 }, { "epoch": 2.9643117261471232, "grad_norm": 1.7339856624603271, "learning_rate": 3.561925096906451e-05, "loss": 0.19429349899291992, "memory(GiB)": 70.5, "step": 69190, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.44884 }, { "epoch": 2.964525941476372, "grad_norm": 7.620383262634277, "learning_rate": 3.56128056853467e-05, "loss": 0.511896800994873, "memory(GiB)": 70.5, "step": 69195, "token_acc": 0.902834008097166, "train_speed(iter/s)": 1.448847 }, { "epoch": 2.964740156805621, "grad_norm": 4.217465877532959, "learning_rate": 3.560636066226644e-05, "loss": 0.554145336151123, "memory(GiB)": 70.5, "step": 69200, "token_acc": 0.8764044943820225, "train_speed(iter/s)": 1.44886 }, { "epoch": 2.96495437213487, "grad_norm": 1.6904761791229248, "learning_rate": 3.5599915899940454e-05, "loss": 0.1572213053703308, "memory(GiB)": 70.5, "step": 69205, "token_acc": 0.9527272727272728, "train_speed(iter/s)": 1.448864 }, { "epoch": 2.965168587464119, "grad_norm": 0.7102317810058594, "learning_rate": 3.559347139848549e-05, "loss": 0.14545931816101074, "memory(GiB)": 70.5, "step": 69210, "token_acc": 0.9690140845070423, "train_speed(iter/s)": 1.448869 }, { "epoch": 2.9653828027933677, "grad_norm": 1.9150196313858032, "learning_rate": 3.558702715801832e-05, "loss": 0.3081878662109375, "memory(GiB)": 70.5, "step": 69215, "token_acc": 0.939209726443769, "train_speed(iter/s)": 1.448878 }, { "epoch": 2.965597018122617, "grad_norm": 2.852882146835327, "learning_rate": 3.5580583178655637e-05, "loss": 0.3368422746658325, "memory(GiB)": 70.5, "step": 69220, "token_acc": 0.9299610894941635, "train_speed(iter/s)": 1.448879 }, { "epoch": 2.965811233451866, "grad_norm": 3.144104480743408, "learning_rate": 3.557413946051425e-05, "loss": 0.3043376922607422, "memory(GiB)": 70.5, "step": 69225, "token_acc": 0.936, "train_speed(iter/s)": 1.448879 }, { "epoch": 2.9660254487811146, "grad_norm": 2.475142002105713, "learning_rate": 3.556769600371084e-05, "loss": 0.40980377197265627, "memory(GiB)": 70.5, "step": 69230, "token_acc": 0.9161290322580645, "train_speed(iter/s)": 1.448891 }, { "epoch": 2.966239664110364, "grad_norm": 3.78151273727417, "learning_rate": 3.5561252808362176e-05, "loss": 0.4552724838256836, "memory(GiB)": 70.5, "step": 69235, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.448894 }, { "epoch": 2.9664538794396127, "grad_norm": 2.7319858074188232, "learning_rate": 3.555480987458495e-05, "loss": 0.3216728210449219, "memory(GiB)": 70.5, "step": 69240, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.448912 }, { "epoch": 2.9666680947688615, "grad_norm": 3.0859498977661133, "learning_rate": 3.5548367202495894e-05, "loss": 0.44575977325439453, "memory(GiB)": 70.5, "step": 69245, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.448926 }, { "epoch": 2.9668823100981108, "grad_norm": 2.5963737964630127, "learning_rate": 3.554192479221173e-05, "loss": 0.4792964458465576, "memory(GiB)": 70.5, "step": 69250, "token_acc": 0.91015625, "train_speed(iter/s)": 1.44893 }, { "epoch": 2.9670965254273596, "grad_norm": 3.453392267227173, "learning_rate": 3.5535482643849153e-05, "loss": 0.16939162015914916, "memory(GiB)": 70.5, "step": 69255, "token_acc": 0.953125, "train_speed(iter/s)": 1.448935 }, { "epoch": 2.9673107407566084, "grad_norm": 7.687924861907959, "learning_rate": 3.552904075752489e-05, "loss": 0.32506532669067384, "memory(GiB)": 70.5, "step": 69260, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.448942 }, { "epoch": 2.9675249560858576, "grad_norm": 2.213505744934082, "learning_rate": 3.552259913335562e-05, "loss": 0.27340521812438967, "memory(GiB)": 70.5, "step": 69265, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.44895 }, { "epoch": 2.9677391714151065, "grad_norm": 1.727288842201233, "learning_rate": 3.5516157771458045e-05, "loss": 0.27587969303131105, "memory(GiB)": 70.5, "step": 69270, "token_acc": 0.9468599033816425, "train_speed(iter/s)": 1.448977 }, { "epoch": 2.9679533867443553, "grad_norm": 2.5520429611206055, "learning_rate": 3.550971667194886e-05, "loss": 0.6269798278808594, "memory(GiB)": 70.5, "step": 69275, "token_acc": 0.88, "train_speed(iter/s)": 1.448988 }, { "epoch": 2.9681676020736045, "grad_norm": 1.7455812692642212, "learning_rate": 3.550327583494475e-05, "loss": 0.5484203338623047, "memory(GiB)": 70.5, "step": 69280, "token_acc": 0.8628158844765343, "train_speed(iter/s)": 1.448989 }, { "epoch": 2.9683818174028533, "grad_norm": 0.519455075263977, "learning_rate": 3.54968352605624e-05, "loss": 0.16391600370407106, "memory(GiB)": 70.5, "step": 69285, "token_acc": 0.9605734767025089, "train_speed(iter/s)": 1.448992 }, { "epoch": 2.968596032732102, "grad_norm": 4.339543342590332, "learning_rate": 3.549039494891849e-05, "loss": 0.30791587829589845, "memory(GiB)": 70.5, "step": 69290, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.448997 }, { "epoch": 2.9688102480613514, "grad_norm": 1.686437964439392, "learning_rate": 3.548395490012966e-05, "loss": 0.2775006055831909, "memory(GiB)": 70.5, "step": 69295, "token_acc": 0.9425287356321839, "train_speed(iter/s)": 1.449002 }, { "epoch": 2.9690244633906, "grad_norm": 1.3230793476104736, "learning_rate": 3.547751511431262e-05, "loss": 0.2659328460693359, "memory(GiB)": 70.5, "step": 69300, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.449004 }, { "epoch": 2.969238678719849, "grad_norm": 3.3595118522644043, "learning_rate": 3.5471075591584024e-05, "loss": 0.27877302169799806, "memory(GiB)": 70.5, "step": 69305, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.449004 }, { "epoch": 2.9694528940490983, "grad_norm": 2.138981342315674, "learning_rate": 3.546463633206052e-05, "loss": 0.2200462579727173, "memory(GiB)": 70.5, "step": 69310, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.449006 }, { "epoch": 2.969667109378347, "grad_norm": 1.096839427947998, "learning_rate": 3.545819733585876e-05, "loss": 0.34502995014190674, "memory(GiB)": 70.5, "step": 69315, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.449007 }, { "epoch": 2.969881324707596, "grad_norm": 3.0834741592407227, "learning_rate": 3.5451758603095404e-05, "loss": 0.44173388481140136, "memory(GiB)": 70.5, "step": 69320, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.449011 }, { "epoch": 2.970095540036845, "grad_norm": 3.177708864212036, "learning_rate": 3.5445320133887075e-05, "loss": 0.4772975444793701, "memory(GiB)": 70.5, "step": 69325, "token_acc": 0.9023569023569024, "train_speed(iter/s)": 1.449015 }, { "epoch": 2.970309755366094, "grad_norm": 6.229069709777832, "learning_rate": 3.543888192835044e-05, "loss": 0.6099040985107422, "memory(GiB)": 70.5, "step": 69330, "token_acc": 0.8992248062015504, "train_speed(iter/s)": 1.449016 }, { "epoch": 2.970523970695343, "grad_norm": 4.408082485198975, "learning_rate": 3.543244398660212e-05, "loss": 0.8511798858642579, "memory(GiB)": 70.5, "step": 69335, "token_acc": 0.8176470588235294, "train_speed(iter/s)": 1.449019 }, { "epoch": 2.970738186024592, "grad_norm": 2.879591226577759, "learning_rate": 3.542600630875873e-05, "loss": 0.2786805868148804, "memory(GiB)": 70.5, "step": 69340, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.449022 }, { "epoch": 2.970952401353841, "grad_norm": 6.699375629425049, "learning_rate": 3.541956889493692e-05, "loss": 0.22342753410339355, "memory(GiB)": 70.5, "step": 69345, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.449026 }, { "epoch": 2.9711666166830897, "grad_norm": 0.2536490261554718, "learning_rate": 3.541313174525329e-05, "loss": 0.31709110736846924, "memory(GiB)": 70.5, "step": 69350, "token_acc": 0.9258064516129032, "train_speed(iter/s)": 1.449031 }, { "epoch": 2.971380832012339, "grad_norm": 2.5238068103790283, "learning_rate": 3.540669485982445e-05, "loss": 0.3431115627288818, "memory(GiB)": 70.5, "step": 69355, "token_acc": 0.9233870967741935, "train_speed(iter/s)": 1.449035 }, { "epoch": 2.9715950473415877, "grad_norm": 2.2261126041412354, "learning_rate": 3.540025823876704e-05, "loss": 0.3799097537994385, "memory(GiB)": 70.5, "step": 69360, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.449034 }, { "epoch": 2.9718092626708366, "grad_norm": 3.7917776107788086, "learning_rate": 3.539382188219764e-05, "loss": 0.3738841533660889, "memory(GiB)": 70.5, "step": 69365, "token_acc": 0.9221183800623053, "train_speed(iter/s)": 1.449042 }, { "epoch": 2.972023478000086, "grad_norm": 3.430891275405884, "learning_rate": 3.5387385790232854e-05, "loss": 0.25093653202056887, "memory(GiB)": 70.5, "step": 69370, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.449052 }, { "epoch": 2.9722376933293346, "grad_norm": 5.10992956161499, "learning_rate": 3.538094996298928e-05, "loss": 0.38561043739318845, "memory(GiB)": 70.5, "step": 69375, "token_acc": 0.94140625, "train_speed(iter/s)": 1.449056 }, { "epoch": 2.9724519086585834, "grad_norm": 2.3038012981414795, "learning_rate": 3.537451440058353e-05, "loss": 0.7128846168518066, "memory(GiB)": 70.5, "step": 69380, "token_acc": 0.8615916955017301, "train_speed(iter/s)": 1.449056 }, { "epoch": 2.9726661239878327, "grad_norm": 0.3915634751319885, "learning_rate": 3.5368079103132143e-05, "loss": 0.8063953399658204, "memory(GiB)": 70.5, "step": 69385, "token_acc": 0.8324022346368715, "train_speed(iter/s)": 1.449059 }, { "epoch": 2.9728803393170815, "grad_norm": 2.0973963737487793, "learning_rate": 3.536164407075175e-05, "loss": 0.34764947891235354, "memory(GiB)": 70.5, "step": 69390, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.449067 }, { "epoch": 2.9730945546463303, "grad_norm": 2.89494252204895, "learning_rate": 3.535520930355891e-05, "loss": 0.5745623588562012, "memory(GiB)": 70.5, "step": 69395, "token_acc": 0.8881789137380192, "train_speed(iter/s)": 1.449089 }, { "epoch": 2.9733087699755796, "grad_norm": 2.484997510910034, "learning_rate": 3.5348774801670184e-05, "loss": 0.5181981086730957, "memory(GiB)": 70.5, "step": 69400, "token_acc": 0.896, "train_speed(iter/s)": 1.449101 }, { "epoch": 2.9735229853048284, "grad_norm": 0.37214696407318115, "learning_rate": 3.5342340565202146e-05, "loss": 0.2663832187652588, "memory(GiB)": 70.5, "step": 69405, "token_acc": 0.9560439560439561, "train_speed(iter/s)": 1.449112 }, { "epoch": 2.973737200634077, "grad_norm": 2.7978992462158203, "learning_rate": 3.533590659427137e-05, "loss": 0.27620601654052734, "memory(GiB)": 70.5, "step": 69410, "token_acc": 0.9307958477508651, "train_speed(iter/s)": 1.449125 }, { "epoch": 2.9739514159633265, "grad_norm": 6.462371826171875, "learning_rate": 3.532947288899439e-05, "loss": 0.8762405395507813, "memory(GiB)": 70.5, "step": 69415, "token_acc": 0.8366013071895425, "train_speed(iter/s)": 1.449138 }, { "epoch": 2.9741656312925753, "grad_norm": 1.4558749198913574, "learning_rate": 3.532303944948777e-05, "loss": 0.371069598197937, "memory(GiB)": 70.5, "step": 69420, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.449144 }, { "epoch": 2.974379846621824, "grad_norm": 2.0732173919677734, "learning_rate": 3.5316606275868056e-05, "loss": 0.3816851615905762, "memory(GiB)": 70.5, "step": 69425, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.449152 }, { "epoch": 2.9745940619510733, "grad_norm": 4.282186985015869, "learning_rate": 3.5310173368251794e-05, "loss": 0.5010400772094726, "memory(GiB)": 70.5, "step": 69430, "token_acc": 0.8858267716535433, "train_speed(iter/s)": 1.449154 }, { "epoch": 2.974808277280322, "grad_norm": 7.790096759796143, "learning_rate": 3.5303740726755544e-05, "loss": 0.40584626197814944, "memory(GiB)": 70.5, "step": 69435, "token_acc": 0.9039145907473309, "train_speed(iter/s)": 1.449158 }, { "epoch": 2.975022492609571, "grad_norm": 3.495918035507202, "learning_rate": 3.529730835149577e-05, "loss": 0.4448524475097656, "memory(GiB)": 70.5, "step": 69440, "token_acc": 0.9147540983606557, "train_speed(iter/s)": 1.449154 }, { "epoch": 2.9752367079388202, "grad_norm": 5.463122844696045, "learning_rate": 3.5290876242589076e-05, "loss": 0.3770363092422485, "memory(GiB)": 70.5, "step": 69445, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.449152 }, { "epoch": 2.975450923268069, "grad_norm": 0.2518969178199768, "learning_rate": 3.528444440015196e-05, "loss": 0.2061060905456543, "memory(GiB)": 70.5, "step": 69450, "token_acc": 0.948905109489051, "train_speed(iter/s)": 1.449157 }, { "epoch": 2.975665138597318, "grad_norm": 2.9754390716552734, "learning_rate": 3.5278012824300944e-05, "loss": 0.24500761032104493, "memory(GiB)": 70.5, "step": 69455, "token_acc": 0.948905109489051, "train_speed(iter/s)": 1.449153 }, { "epoch": 2.975879353926567, "grad_norm": 3.4499759674072266, "learning_rate": 3.527158151515252e-05, "loss": 0.17260103225708007, "memory(GiB)": 70.5, "step": 69460, "token_acc": 0.9556451612903226, "train_speed(iter/s)": 1.449164 }, { "epoch": 2.976093569255816, "grad_norm": 1.4123040437698364, "learning_rate": 3.526515047282323e-05, "loss": 0.5018493175506592, "memory(GiB)": 70.5, "step": 69465, "token_acc": 0.8914473684210527, "train_speed(iter/s)": 1.449177 }, { "epoch": 2.9763077845850647, "grad_norm": 0.8135066628456116, "learning_rate": 3.525871969742954e-05, "loss": 0.22358713150024415, "memory(GiB)": 70.5, "step": 69470, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.449181 }, { "epoch": 2.976521999914314, "grad_norm": 6.295461177825928, "learning_rate": 3.525228918908799e-05, "loss": 0.5893518447875976, "memory(GiB)": 70.5, "step": 69475, "token_acc": 0.8825503355704698, "train_speed(iter/s)": 1.44918 }, { "epoch": 2.976736215243563, "grad_norm": 1.5937045812606812, "learning_rate": 3.524585894791506e-05, "loss": 0.41501760482788086, "memory(GiB)": 70.5, "step": 69480, "token_acc": 0.9009287925696594, "train_speed(iter/s)": 1.449186 }, { "epoch": 2.9769504305728116, "grad_norm": 7.20565938949585, "learning_rate": 3.523942897402721e-05, "loss": 0.6066811561584473, "memory(GiB)": 70.5, "step": 69485, "token_acc": 0.8627450980392157, "train_speed(iter/s)": 1.449202 }, { "epoch": 2.977164645902061, "grad_norm": 3.802213430404663, "learning_rate": 3.5232999267540964e-05, "loss": 0.5054030895233155, "memory(GiB)": 70.5, "step": 69490, "token_acc": 0.8789808917197452, "train_speed(iter/s)": 1.449205 }, { "epoch": 2.9773788612313097, "grad_norm": 1.1789519786834717, "learning_rate": 3.52265698285728e-05, "loss": 0.22252705097198486, "memory(GiB)": 70.5, "step": 69495, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.449211 }, { "epoch": 2.9775930765605585, "grad_norm": 2.9062345027923584, "learning_rate": 3.5220140657239164e-05, "loss": 0.260744571685791, "memory(GiB)": 70.5, "step": 69500, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.449209 }, { "epoch": 2.9775930765605585, "eval_loss": 2.5833165645599365, "eval_runtime": 11.7907, "eval_samples_per_second": 8.481, "eval_steps_per_second": 8.481, "eval_token_acc": 0.4104979811574697, "step": 69500 }, { "epoch": 2.9778072918898078, "grad_norm": 0.8578397631645203, "learning_rate": 3.5213711753656565e-05, "loss": 0.25577945709228517, "memory(GiB)": 70.5, "step": 69505, "token_acc": 0.5472081218274112, "train_speed(iter/s)": 1.448823 }, { "epoch": 2.9780215072190566, "grad_norm": 3.3331801891326904, "learning_rate": 3.520728311794143e-05, "loss": 0.4235753059387207, "memory(GiB)": 70.5, "step": 69510, "token_acc": 0.9053497942386831, "train_speed(iter/s)": 1.44884 }, { "epoch": 2.9782357225483054, "grad_norm": 5.17130184173584, "learning_rate": 3.5200854750210235e-05, "loss": 0.35670881271362304, "memory(GiB)": 70.5, "step": 69515, "token_acc": 0.9407407407407408, "train_speed(iter/s)": 1.448845 }, { "epoch": 2.9784499378775546, "grad_norm": 3.077836513519287, "learning_rate": 3.5194426650579445e-05, "loss": 0.23967399597167968, "memory(GiB)": 70.5, "step": 69520, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.448853 }, { "epoch": 2.9786641532068034, "grad_norm": 2.1718459129333496, "learning_rate": 3.518799881916551e-05, "loss": 0.2843461275100708, "memory(GiB)": 70.5, "step": 69525, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.448859 }, { "epoch": 2.9788783685360523, "grad_norm": 3.0487945079803467, "learning_rate": 3.518157125608487e-05, "loss": 0.4415945053100586, "memory(GiB)": 70.5, "step": 69530, "token_acc": 0.8951841359773371, "train_speed(iter/s)": 1.448862 }, { "epoch": 2.9790925838653015, "grad_norm": 2.1709866523742676, "learning_rate": 3.5175143961453965e-05, "loss": 0.13343753814697265, "memory(GiB)": 70.5, "step": 69535, "token_acc": 0.975177304964539, "train_speed(iter/s)": 1.448865 }, { "epoch": 2.9793067991945503, "grad_norm": 2.7673885822296143, "learning_rate": 3.516871693538924e-05, "loss": 0.2176825523376465, "memory(GiB)": 70.5, "step": 69540, "token_acc": 0.9447004608294931, "train_speed(iter/s)": 1.448863 }, { "epoch": 2.979521014523799, "grad_norm": 2.921459197998047, "learning_rate": 3.516229017800711e-05, "loss": 0.2697928428649902, "memory(GiB)": 70.5, "step": 69545, "token_acc": 0.9480968858131488, "train_speed(iter/s)": 1.448869 }, { "epoch": 2.9797352298530484, "grad_norm": 2.0099167823791504, "learning_rate": 3.515586368942402e-05, "loss": 0.8681355476379394, "memory(GiB)": 70.5, "step": 69550, "token_acc": 0.7977099236641222, "train_speed(iter/s)": 1.44889 }, { "epoch": 2.979949445182297, "grad_norm": 3.1403706073760986, "learning_rate": 3.514943746975639e-05, "loss": 0.14247757196426392, "memory(GiB)": 70.5, "step": 69555, "token_acc": 0.9645669291338582, "train_speed(iter/s)": 1.448895 }, { "epoch": 2.980163660511546, "grad_norm": 3.366570472717285, "learning_rate": 3.514301151912062e-05, "loss": 0.3946083545684814, "memory(GiB)": 70.5, "step": 69560, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.4489 }, { "epoch": 2.9803778758407953, "grad_norm": 2.793968439102173, "learning_rate": 3.513658583763314e-05, "loss": 0.6287849426269532, "memory(GiB)": 70.5, "step": 69565, "token_acc": 0.9013605442176871, "train_speed(iter/s)": 1.448918 }, { "epoch": 2.980592091170044, "grad_norm": 2.023543357849121, "learning_rate": 3.513016042541034e-05, "loss": 0.5445969581604004, "memory(GiB)": 70.5, "step": 69570, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.44892 }, { "epoch": 2.980806306499293, "grad_norm": 3.8269965648651123, "learning_rate": 3.5123735282568646e-05, "loss": 0.3445282936096191, "memory(GiB)": 70.5, "step": 69575, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.448927 }, { "epoch": 2.981020521828542, "grad_norm": 6.447869300842285, "learning_rate": 3.5117310409224443e-05, "loss": 0.3723740100860596, "memory(GiB)": 70.5, "step": 69580, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.448934 }, { "epoch": 2.981234737157791, "grad_norm": 3.8837883472442627, "learning_rate": 3.5110885805494115e-05, "loss": 0.3043516635894775, "memory(GiB)": 70.5, "step": 69585, "token_acc": 0.9298245614035088, "train_speed(iter/s)": 1.448942 }, { "epoch": 2.98144895248704, "grad_norm": 3.153458595275879, "learning_rate": 3.510446147149404e-05, "loss": 0.4448305606842041, "memory(GiB)": 70.5, "step": 69590, "token_acc": 0.8925925925925926, "train_speed(iter/s)": 1.448955 }, { "epoch": 2.981663167816289, "grad_norm": 2.9391865730285645, "learning_rate": 3.509803740734065e-05, "loss": 0.4088016986846924, "memory(GiB)": 70.5, "step": 69595, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.448965 }, { "epoch": 2.981877383145538, "grad_norm": 2.7705869674682617, "learning_rate": 3.509161361315028e-05, "loss": 0.49507646560668944, "memory(GiB)": 70.5, "step": 69600, "token_acc": 0.8768656716417911, "train_speed(iter/s)": 1.448965 }, { "epoch": 2.9820915984747867, "grad_norm": 5.155707359313965, "learning_rate": 3.508519008903931e-05, "loss": 0.5112536430358887, "memory(GiB)": 70.5, "step": 69605, "token_acc": 0.8996865203761756, "train_speed(iter/s)": 1.448977 }, { "epoch": 2.982305813804036, "grad_norm": 3.9702305793762207, "learning_rate": 3.507876683512412e-05, "loss": 0.6387495994567871, "memory(GiB)": 70.5, "step": 69610, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.448979 }, { "epoch": 2.9825200291332847, "grad_norm": 0.6655214428901672, "learning_rate": 3.507234385152106e-05, "loss": 0.43091607093811035, "memory(GiB)": 70.5, "step": 69615, "token_acc": 0.9147727272727273, "train_speed(iter/s)": 1.448975 }, { "epoch": 2.9827342444625335, "grad_norm": 2.3625435829162598, "learning_rate": 3.5065921138346504e-05, "loss": 0.3859349250793457, "memory(GiB)": 70.5, "step": 69620, "token_acc": 0.9314285714285714, "train_speed(iter/s)": 1.448977 }, { "epoch": 2.982948459791783, "grad_norm": 5.13886022567749, "learning_rate": 3.505949869571679e-05, "loss": 0.4402755260467529, "memory(GiB)": 70.5, "step": 69625, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.448978 }, { "epoch": 2.9831626751210316, "grad_norm": 2.7584869861602783, "learning_rate": 3.505307652374827e-05, "loss": 0.2643409013748169, "memory(GiB)": 70.5, "step": 69630, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.448983 }, { "epoch": 2.9833768904502804, "grad_norm": 3.1571614742279053, "learning_rate": 3.5046654622557295e-05, "loss": 0.3250680208206177, "memory(GiB)": 70.5, "step": 69635, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.448984 }, { "epoch": 2.9835911057795297, "grad_norm": 1.6068021059036255, "learning_rate": 3.50402329922602e-05, "loss": 0.26287641525268557, "memory(GiB)": 70.5, "step": 69640, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.448982 }, { "epoch": 2.9838053211087785, "grad_norm": 7.634116172790527, "learning_rate": 3.5033811632973315e-05, "loss": 0.37288947105407716, "memory(GiB)": 70.5, "step": 69645, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.448992 }, { "epoch": 2.9840195364380273, "grad_norm": 2.9285356998443604, "learning_rate": 3.502739054481297e-05, "loss": 0.20943353176116944, "memory(GiB)": 70.5, "step": 69650, "token_acc": 0.9543859649122807, "train_speed(iter/s)": 1.448999 }, { "epoch": 2.9842337517672766, "grad_norm": 1.982748031616211, "learning_rate": 3.5020969727895484e-05, "loss": 0.26486060619354246, "memory(GiB)": 70.5, "step": 69655, "token_acc": 0.9539007092198581, "train_speed(iter/s)": 1.449006 }, { "epoch": 2.9844479670965254, "grad_norm": 1.70413076877594, "learning_rate": 3.50145491823372e-05, "loss": 0.48985934257507324, "memory(GiB)": 70.5, "step": 69660, "token_acc": 0.9274447949526814, "train_speed(iter/s)": 1.449018 }, { "epoch": 2.984662182425774, "grad_norm": 2.155055284500122, "learning_rate": 3.500812890825439e-05, "loss": 0.36579735279083253, "memory(GiB)": 70.5, "step": 69665, "token_acc": 0.9142011834319527, "train_speed(iter/s)": 1.44902 }, { "epoch": 2.9848763977550234, "grad_norm": 2.3943777084350586, "learning_rate": 3.50017089057634e-05, "loss": 0.47864809036254885, "memory(GiB)": 70.5, "step": 69670, "token_acc": 0.8921933085501859, "train_speed(iter/s)": 1.449033 }, { "epoch": 2.9850906130842723, "grad_norm": 2.81363582611084, "learning_rate": 3.499528917498053e-05, "loss": 0.19301798343658447, "memory(GiB)": 70.5, "step": 69675, "token_acc": 0.944, "train_speed(iter/s)": 1.449039 }, { "epoch": 2.985304828413521, "grad_norm": 3.7639448642730713, "learning_rate": 3.4988869716022065e-05, "loss": 0.6310288906097412, "memory(GiB)": 70.5, "step": 69680, "token_acc": 0.8913738019169329, "train_speed(iter/s)": 1.449049 }, { "epoch": 2.9855190437427703, "grad_norm": 4.03608512878418, "learning_rate": 3.498245052900432e-05, "loss": 0.40221943855285647, "memory(GiB)": 70.5, "step": 69685, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.449051 }, { "epoch": 2.985733259072019, "grad_norm": 2.561411142349243, "learning_rate": 3.4976031614043555e-05, "loss": 0.23416435718536377, "memory(GiB)": 70.5, "step": 69690, "token_acc": 0.9367088607594937, "train_speed(iter/s)": 1.449055 }, { "epoch": 2.985947474401268, "grad_norm": 2.0033681392669678, "learning_rate": 3.496961297125608e-05, "loss": 0.6240200996398926, "memory(GiB)": 70.5, "step": 69695, "token_acc": 0.8629283489096573, "train_speed(iter/s)": 1.449057 }, { "epoch": 2.986161689730517, "grad_norm": 0.9432464838027954, "learning_rate": 3.4963194600758166e-05, "loss": 0.331189227104187, "memory(GiB)": 70.5, "step": 69700, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.449058 }, { "epoch": 2.986375905059766, "grad_norm": 4.271366119384766, "learning_rate": 3.4956776502666076e-05, "loss": 0.2146904945373535, "memory(GiB)": 70.5, "step": 69705, "token_acc": 0.9493243243243243, "train_speed(iter/s)": 1.449072 }, { "epoch": 2.986590120389015, "grad_norm": 0.058593787252902985, "learning_rate": 3.4950358677096103e-05, "loss": 0.21750741004943847, "memory(GiB)": 70.5, "step": 69710, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.449079 }, { "epoch": 2.986804335718264, "grad_norm": 4.222776889801025, "learning_rate": 3.4943941124164494e-05, "loss": 0.20769309997558594, "memory(GiB)": 70.5, "step": 69715, "token_acc": 0.950530035335689, "train_speed(iter/s)": 1.449087 }, { "epoch": 2.987018551047513, "grad_norm": 2.6656596660614014, "learning_rate": 3.493752384398753e-05, "loss": 0.443255090713501, "memory(GiB)": 70.5, "step": 69720, "token_acc": 0.9227642276422764, "train_speed(iter/s)": 1.449088 }, { "epoch": 2.9872327663767617, "grad_norm": 0.7597995400428772, "learning_rate": 3.493110683668144e-05, "loss": 0.2398970603942871, "memory(GiB)": 70.5, "step": 69725, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.449093 }, { "epoch": 2.987446981706011, "grad_norm": 3.9070982933044434, "learning_rate": 3.4924690102362475e-05, "loss": 0.4986997127532959, "memory(GiB)": 70.5, "step": 69730, "token_acc": 0.8919753086419753, "train_speed(iter/s)": 1.449102 }, { "epoch": 2.98766119703526, "grad_norm": 2.513915777206421, "learning_rate": 3.491827364114689e-05, "loss": 0.4823263645172119, "memory(GiB)": 70.5, "step": 69735, "token_acc": 0.8932384341637011, "train_speed(iter/s)": 1.449109 }, { "epoch": 2.9878754123645086, "grad_norm": 2.4375343322753906, "learning_rate": 3.491185745315094e-05, "loss": 0.4627972602844238, "memory(GiB)": 70.5, "step": 69740, "token_acc": 0.8907849829351536, "train_speed(iter/s)": 1.449108 }, { "epoch": 2.988089627693758, "grad_norm": 1.235072135925293, "learning_rate": 3.490544153849085e-05, "loss": 0.17318222522735596, "memory(GiB)": 70.5, "step": 69745, "token_acc": 0.9609375, "train_speed(iter/s)": 1.449108 }, { "epoch": 2.9883038430230067, "grad_norm": 1.037312388420105, "learning_rate": 3.489902589728283e-05, "loss": 0.23384346961975097, "memory(GiB)": 70.5, "step": 69750, "token_acc": 0.9581749049429658, "train_speed(iter/s)": 1.449119 }, { "epoch": 2.9885180583522555, "grad_norm": 5.970566749572754, "learning_rate": 3.4892610529643135e-05, "loss": 0.5797299385070801, "memory(GiB)": 70.5, "step": 69755, "token_acc": 0.891566265060241, "train_speed(iter/s)": 1.449113 }, { "epoch": 2.9887322736815047, "grad_norm": 3.3552427291870117, "learning_rate": 3.488619543568796e-05, "loss": 0.37422704696655273, "memory(GiB)": 70.5, "step": 69760, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.449118 }, { "epoch": 2.9889464890107535, "grad_norm": 4.310719966888428, "learning_rate": 3.487978061553355e-05, "loss": 0.4979377746582031, "memory(GiB)": 70.5, "step": 69765, "token_acc": 0.8813559322033898, "train_speed(iter/s)": 1.449124 }, { "epoch": 2.9891607043400024, "grad_norm": 2.6424295902252197, "learning_rate": 3.4873366069296095e-05, "loss": 0.23877418041229248, "memory(GiB)": 70.5, "step": 69770, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.449127 }, { "epoch": 2.9893749196692516, "grad_norm": 2.342857837677002, "learning_rate": 3.4866951797091786e-05, "loss": 0.1957413673400879, "memory(GiB)": 70.5, "step": 69775, "token_acc": 0.9516728624535316, "train_speed(iter/s)": 1.449123 }, { "epoch": 2.9895891349985004, "grad_norm": 2.024437189102173, "learning_rate": 3.486053779903686e-05, "loss": 0.21991729736328125, "memory(GiB)": 70.5, "step": 69780, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.449121 }, { "epoch": 2.9898033503277492, "grad_norm": 5.127934455871582, "learning_rate": 3.485412407524749e-05, "loss": 0.7116131782531738, "memory(GiB)": 70.5, "step": 69785, "token_acc": 0.8381877022653722, "train_speed(iter/s)": 1.449131 }, { "epoch": 2.9900175656569985, "grad_norm": 5.100478649139404, "learning_rate": 3.484771062583986e-05, "loss": 0.3869349479675293, "memory(GiB)": 70.5, "step": 69790, "token_acc": 0.9098039215686274, "train_speed(iter/s)": 1.449133 }, { "epoch": 2.9902317809862473, "grad_norm": 1.98316490650177, "learning_rate": 3.484129745093018e-05, "loss": 0.11865609884262085, "memory(GiB)": 70.5, "step": 69795, "token_acc": 0.9743589743589743, "train_speed(iter/s)": 1.449141 }, { "epoch": 2.990445996315496, "grad_norm": 8.152923583984375, "learning_rate": 3.48348845506346e-05, "loss": 0.6255522727966308, "memory(GiB)": 70.5, "step": 69800, "token_acc": 0.8863636363636364, "train_speed(iter/s)": 1.449149 }, { "epoch": 2.9906602116447454, "grad_norm": 2.226710796356201, "learning_rate": 3.482847192506933e-05, "loss": 0.43470087051391604, "memory(GiB)": 70.5, "step": 69805, "token_acc": 0.9186746987951807, "train_speed(iter/s)": 1.449153 }, { "epoch": 2.990874426973994, "grad_norm": 3.96224308013916, "learning_rate": 3.48220595743505e-05, "loss": 0.2514703035354614, "memory(GiB)": 70.5, "step": 69810, "token_acc": 0.956, "train_speed(iter/s)": 1.44915 }, { "epoch": 2.991088642303243, "grad_norm": 4.121099948883057, "learning_rate": 3.481564749859431e-05, "loss": 0.6290458679199219, "memory(GiB)": 70.5, "step": 69815, "token_acc": 0.8526645768025078, "train_speed(iter/s)": 1.449149 }, { "epoch": 2.9913028576324923, "grad_norm": 2.7721898555755615, "learning_rate": 3.480923569791691e-05, "loss": 0.25195975303649903, "memory(GiB)": 70.5, "step": 69820, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.449153 }, { "epoch": 2.991517072961741, "grad_norm": 2.8836565017700195, "learning_rate": 3.480282417243446e-05, "loss": 0.392657995223999, "memory(GiB)": 70.5, "step": 69825, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.449161 }, { "epoch": 2.99173128829099, "grad_norm": 2.4508237838745117, "learning_rate": 3.479641292226311e-05, "loss": 0.28517780303955076, "memory(GiB)": 70.5, "step": 69830, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.449156 }, { "epoch": 2.991945503620239, "grad_norm": 4.084285736083984, "learning_rate": 3.479000194751899e-05, "loss": 0.514182710647583, "memory(GiB)": 70.5, "step": 69835, "token_acc": 0.8906752411575563, "train_speed(iter/s)": 1.449165 }, { "epoch": 2.992159718949488, "grad_norm": 3.567345142364502, "learning_rate": 3.478359124831827e-05, "loss": 0.28413140773773193, "memory(GiB)": 70.5, "step": 69840, "token_acc": 0.9442622950819672, "train_speed(iter/s)": 1.449168 }, { "epoch": 2.9923739342787368, "grad_norm": 0.9773494601249695, "learning_rate": 3.4777180824777057e-05, "loss": 0.21460611820220948, "memory(GiB)": 70.5, "step": 69845, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.449172 }, { "epoch": 2.992588149607986, "grad_norm": 3.326352596282959, "learning_rate": 3.477077067701149e-05, "loss": 0.2818249940872192, "memory(GiB)": 70.5, "step": 69850, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.449179 }, { "epoch": 2.992802364937235, "grad_norm": 2.2533130645751953, "learning_rate": 3.476436080513771e-05, "loss": 0.42839322090148924, "memory(GiB)": 70.5, "step": 69855, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.449205 }, { "epoch": 2.9930165802664837, "grad_norm": 1.177463412284851, "learning_rate": 3.475795120927181e-05, "loss": 0.23373031616210938, "memory(GiB)": 70.5, "step": 69860, "token_acc": 0.9519774011299436, "train_speed(iter/s)": 1.449212 }, { "epoch": 2.993230795595733, "grad_norm": 0.4534831643104553, "learning_rate": 3.475154188952994e-05, "loss": 0.25145840644836426, "memory(GiB)": 70.5, "step": 69865, "token_acc": 0.9452887537993921, "train_speed(iter/s)": 1.449218 }, { "epoch": 2.9934450109249817, "grad_norm": 5.514671325683594, "learning_rate": 3.4745132846028185e-05, "loss": 0.30152335166931155, "memory(GiB)": 70.5, "step": 69870, "token_acc": 0.933933933933934, "train_speed(iter/s)": 1.449211 }, { "epoch": 2.9936592262542305, "grad_norm": 3.1967885494232178, "learning_rate": 3.473872407888266e-05, "loss": 0.43024625778198244, "memory(GiB)": 70.5, "step": 69875, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.44921 }, { "epoch": 2.99387344158348, "grad_norm": 2.319355010986328, "learning_rate": 3.473231558820946e-05, "loss": 0.3094393968582153, "memory(GiB)": 70.5, "step": 69880, "token_acc": 0.9510204081632653, "train_speed(iter/s)": 1.449225 }, { "epoch": 2.9940876569127286, "grad_norm": 1.3872102499008179, "learning_rate": 3.472590737412467e-05, "loss": 0.28546187877655027, "memory(GiB)": 70.5, "step": 69885, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.449231 }, { "epoch": 2.9943018722419774, "grad_norm": 1.9232628345489502, "learning_rate": 3.471949943674442e-05, "loss": 0.3390068054199219, "memory(GiB)": 70.5, "step": 69890, "token_acc": 0.9251497005988024, "train_speed(iter/s)": 1.449234 }, { "epoch": 2.9945160875712267, "grad_norm": 3.6586549282073975, "learning_rate": 3.471309177618476e-05, "loss": 0.32270872592926025, "memory(GiB)": 70.5, "step": 69895, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.449243 }, { "epoch": 2.9947303029004755, "grad_norm": 2.853086471557617, "learning_rate": 3.47066843925618e-05, "loss": 0.3115828275680542, "memory(GiB)": 70.5, "step": 69900, "token_acc": 0.9267399267399268, "train_speed(iter/s)": 1.449243 }, { "epoch": 2.9949445182297243, "grad_norm": 1.7497376203536987, "learning_rate": 3.4700277285991575e-05, "loss": 0.446243143081665, "memory(GiB)": 70.5, "step": 69905, "token_acc": 0.8980891719745223, "train_speed(iter/s)": 1.449241 }, { "epoch": 2.9951587335589736, "grad_norm": 3.473118543624878, "learning_rate": 3.469387045659019e-05, "loss": 0.4280991077423096, "memory(GiB)": 70.5, "step": 69910, "token_acc": 0.9152542372881356, "train_speed(iter/s)": 1.449242 }, { "epoch": 2.9953729488882224, "grad_norm": 4.569370269775391, "learning_rate": 3.4687463904473716e-05, "loss": 0.34952311515808104, "memory(GiB)": 70.5, "step": 69915, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.449259 }, { "epoch": 2.995587164217471, "grad_norm": 0.8012757301330566, "learning_rate": 3.468105762975817e-05, "loss": 0.26681220531463623, "memory(GiB)": 70.5, "step": 69920, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.449259 }, { "epoch": 2.9958013795467204, "grad_norm": 3.9596364498138428, "learning_rate": 3.467465163255966e-05, "loss": 0.3553952217102051, "memory(GiB)": 70.5, "step": 69925, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.449257 }, { "epoch": 2.9960155948759692, "grad_norm": 0.1888272911310196, "learning_rate": 3.46682459129942e-05, "loss": 0.3860813856124878, "memory(GiB)": 70.5, "step": 69930, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.449268 }, { "epoch": 2.996229810205218, "grad_norm": 1.517104983329773, "learning_rate": 3.466184047117784e-05, "loss": 0.578999137878418, "memory(GiB)": 70.5, "step": 69935, "token_acc": 0.8740458015267175, "train_speed(iter/s)": 1.449283 }, { "epoch": 2.9964440255344673, "grad_norm": 2.0513668060302734, "learning_rate": 3.4655435307226645e-05, "loss": 0.17950005531311036, "memory(GiB)": 70.5, "step": 69940, "token_acc": 0.9580838323353293, "train_speed(iter/s)": 1.449279 }, { "epoch": 2.996658240863716, "grad_norm": 4.597748279571533, "learning_rate": 3.4649030421256625e-05, "loss": 0.3515637874603271, "memory(GiB)": 70.5, "step": 69945, "token_acc": 0.9369369369369369, "train_speed(iter/s)": 1.449288 }, { "epoch": 2.996872456192965, "grad_norm": 1.9216856956481934, "learning_rate": 3.464262581338382e-05, "loss": 0.4755709648132324, "memory(GiB)": 70.5, "step": 69950, "token_acc": 0.8966789667896679, "train_speed(iter/s)": 1.449296 }, { "epoch": 2.997086671522214, "grad_norm": 7.402816295623779, "learning_rate": 3.463622148372426e-05, "loss": 0.44165792465209963, "memory(GiB)": 70.5, "step": 69955, "token_acc": 0.8955823293172691, "train_speed(iter/s)": 1.449296 }, { "epoch": 2.997300886851463, "grad_norm": 3.323026418685913, "learning_rate": 3.462981743239394e-05, "loss": 0.38926947116851807, "memory(GiB)": 70.5, "step": 69960, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.449302 }, { "epoch": 2.997515102180712, "grad_norm": 1.5634368658065796, "learning_rate": 3.46234136595089e-05, "loss": 0.37364811897277833, "memory(GiB)": 70.5, "step": 69965, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.449302 }, { "epoch": 2.997729317509961, "grad_norm": 0.7970244884490967, "learning_rate": 3.461701016518516e-05, "loss": 0.3683542490005493, "memory(GiB)": 70.5, "step": 69970, "token_acc": 0.9031007751937985, "train_speed(iter/s)": 1.449297 }, { "epoch": 2.99794353283921, "grad_norm": 1.312032699584961, "learning_rate": 3.461060694953871e-05, "loss": 0.37275333404541017, "memory(GiB)": 70.5, "step": 69975, "token_acc": 0.9295039164490861, "train_speed(iter/s)": 1.449305 }, { "epoch": 2.9981577481684587, "grad_norm": 0.46260589361190796, "learning_rate": 3.4604204012685546e-05, "loss": 0.4927990436553955, "memory(GiB)": 70.5, "step": 69980, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.449323 }, { "epoch": 2.998371963497708, "grad_norm": 3.329115152359009, "learning_rate": 3.459780135474168e-05, "loss": 0.5110931396484375, "memory(GiB)": 70.5, "step": 69985, "token_acc": 0.9017857142857143, "train_speed(iter/s)": 1.449327 }, { "epoch": 2.9985861788269568, "grad_norm": 3.7869341373443604, "learning_rate": 3.4591398975823084e-05, "loss": 0.3825204372406006, "memory(GiB)": 70.5, "step": 69990, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.44933 }, { "epoch": 2.9988003941562056, "grad_norm": 3.8151602745056152, "learning_rate": 3.458499687604575e-05, "loss": 0.33081612586975095, "memory(GiB)": 70.5, "step": 69995, "token_acc": 0.9360902255639098, "train_speed(iter/s)": 1.44934 }, { "epoch": 2.999014609485455, "grad_norm": 4.166277885437012, "learning_rate": 3.457859505552565e-05, "loss": 0.26952056884765624, "memory(GiB)": 70.5, "step": 70000, "token_acc": 0.9616613418530351, "train_speed(iter/s)": 1.449347 }, { "epoch": 2.999014609485455, "eval_loss": 2.4707815647125244, "eval_runtime": 12.4314, "eval_samples_per_second": 8.044, "eval_steps_per_second": 8.044, "eval_token_acc": 0.45007235890014474, "step": 70000 }, { "epoch": 2.9992288248147037, "grad_norm": 0.4950673282146454, "learning_rate": 3.457219351437877e-05, "loss": 0.2612674951553345, "memory(GiB)": 70.5, "step": 70005, "token_acc": 0.6208842897460018, "train_speed(iter/s)": 1.44896 }, { "epoch": 2.9994430401439525, "grad_norm": 9.471419334411621, "learning_rate": 3.456579225272109e-05, "loss": 0.3526965618133545, "memory(GiB)": 70.5, "step": 70010, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.448974 }, { "epoch": 2.9996572554732017, "grad_norm": 1.6679025888442993, "learning_rate": 3.4559391270668554e-05, "loss": 0.20439956188201905, "memory(GiB)": 70.5, "step": 70015, "token_acc": 0.9644012944983819, "train_speed(iter/s)": 1.448981 }, { "epoch": 2.9998714708024505, "grad_norm": 2.476112127304077, "learning_rate": 3.4552990568337124e-05, "loss": 0.2944769382476807, "memory(GiB)": 70.5, "step": 70020, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.448982 }, { "epoch": 3.0000856861317, "grad_norm": 3.8696885108947754, "learning_rate": 3.454659014584277e-05, "loss": 0.3392754077911377, "memory(GiB)": 70.5, "step": 70025, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.448995 }, { "epoch": 3.0002999014609486, "grad_norm": 4.508812427520752, "learning_rate": 3.4540190003301434e-05, "loss": 0.2966590642929077, "memory(GiB)": 70.5, "step": 70030, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.449 }, { "epoch": 3.0005141167901974, "grad_norm": 2.7108707427978516, "learning_rate": 3.453379014082902e-05, "loss": 0.22948045730590821, "memory(GiB)": 70.5, "step": 70035, "token_acc": 0.9430604982206405, "train_speed(iter/s)": 1.449006 }, { "epoch": 3.0007283321194467, "grad_norm": 7.326935768127441, "learning_rate": 3.452739055854154e-05, "loss": 0.4958390235900879, "memory(GiB)": 70.5, "step": 70040, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.449011 }, { "epoch": 3.0009425474486955, "grad_norm": 2.657395124435425, "learning_rate": 3.4520991256554895e-05, "loss": 0.2622621297836304, "memory(GiB)": 70.5, "step": 70045, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.449022 }, { "epoch": 3.0011567627779443, "grad_norm": 1.821048617362976, "learning_rate": 3.4514592234985006e-05, "loss": 0.5631406307220459, "memory(GiB)": 70.5, "step": 70050, "token_acc": 0.8908450704225352, "train_speed(iter/s)": 1.449044 }, { "epoch": 3.0013709781071936, "grad_norm": 3.219217538833618, "learning_rate": 3.4508193493947816e-05, "loss": 0.29300708770751954, "memory(GiB)": 70.5, "step": 70055, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.449053 }, { "epoch": 3.0015851934364424, "grad_norm": 2.4516115188598633, "learning_rate": 3.4501795033559224e-05, "loss": 0.4560965061187744, "memory(GiB)": 70.5, "step": 70060, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.449062 }, { "epoch": 3.001799408765691, "grad_norm": 1.7122242450714111, "learning_rate": 3.449539685393516e-05, "loss": 0.328051233291626, "memory(GiB)": 70.5, "step": 70065, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.449065 }, { "epoch": 3.0020136240949404, "grad_norm": 2.706934690475464, "learning_rate": 3.448899895519152e-05, "loss": 0.2566171884536743, "memory(GiB)": 70.5, "step": 70070, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.449064 }, { "epoch": 3.0022278394241892, "grad_norm": 1.6026099920272827, "learning_rate": 3.448260133744422e-05, "loss": 0.2414872646331787, "memory(GiB)": 70.5, "step": 70075, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.449062 }, { "epoch": 3.002442054753438, "grad_norm": 4.562417507171631, "learning_rate": 3.4476204000809145e-05, "loss": 0.3697316646575928, "memory(GiB)": 70.5, "step": 70080, "token_acc": 0.9262295081967213, "train_speed(iter/s)": 1.44906 }, { "epoch": 3.0026562700826873, "grad_norm": 9.416361808776855, "learning_rate": 3.446980694540221e-05, "loss": 0.7113440990447998, "memory(GiB)": 70.5, "step": 70085, "token_acc": 0.8561151079136691, "train_speed(iter/s)": 1.449075 }, { "epoch": 3.002870485411936, "grad_norm": 1.482348084449768, "learning_rate": 3.4463410171339275e-05, "loss": 0.4616390705108643, "memory(GiB)": 70.5, "step": 70090, "token_acc": 0.8916666666666667, "train_speed(iter/s)": 1.449072 }, { "epoch": 3.003084700741185, "grad_norm": 5.6302103996276855, "learning_rate": 3.445701367873625e-05, "loss": 0.28217430114746095, "memory(GiB)": 70.5, "step": 70095, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.449071 }, { "epoch": 3.003298916070434, "grad_norm": 6.643440246582031, "learning_rate": 3.445061746770901e-05, "loss": 0.3551944255828857, "memory(GiB)": 70.5, "step": 70100, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.449079 }, { "epoch": 3.003513131399683, "grad_norm": 4.714687347412109, "learning_rate": 3.444422153837339e-05, "loss": 0.4552613735198975, "memory(GiB)": 70.5, "step": 70105, "token_acc": 0.9137380191693291, "train_speed(iter/s)": 1.449098 }, { "epoch": 3.003727346728932, "grad_norm": 5.62851095199585, "learning_rate": 3.443782589084531e-05, "loss": 0.3890983581542969, "memory(GiB)": 70.5, "step": 70110, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.449101 }, { "epoch": 3.003941562058181, "grad_norm": 4.421396732330322, "learning_rate": 3.443143052524062e-05, "loss": 0.7382236003875733, "memory(GiB)": 70.5, "step": 70115, "token_acc": 0.8714285714285714, "train_speed(iter/s)": 1.449116 }, { "epoch": 3.00415577738743, "grad_norm": 2.9117743968963623, "learning_rate": 3.4425035441675165e-05, "loss": 0.4397408485412598, "memory(GiB)": 70.5, "step": 70120, "token_acc": 0.9039735099337748, "train_speed(iter/s)": 1.449112 }, { "epoch": 3.0043699927166787, "grad_norm": 1.0290496349334717, "learning_rate": 3.441864064026479e-05, "loss": 0.3223510026931763, "memory(GiB)": 70.5, "step": 70125, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.449127 }, { "epoch": 3.004584208045928, "grad_norm": 3.6548774242401123, "learning_rate": 3.441224612112538e-05, "loss": 0.35512428283691405, "memory(GiB)": 70.5, "step": 70130, "token_acc": 0.9426751592356688, "train_speed(iter/s)": 1.449129 }, { "epoch": 3.0047984233751768, "grad_norm": 3.4254934787750244, "learning_rate": 3.440585188437273e-05, "loss": 0.44026947021484375, "memory(GiB)": 70.5, "step": 70135, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.449157 }, { "epoch": 3.0050126387044256, "grad_norm": 2.5497684478759766, "learning_rate": 3.439945793012272e-05, "loss": 0.4997386455535889, "memory(GiB)": 70.5, "step": 70140, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.449158 }, { "epoch": 3.005226854033675, "grad_norm": 2.5575125217437744, "learning_rate": 3.439306425849116e-05, "loss": 0.427642822265625, "memory(GiB)": 70.5, "step": 70145, "token_acc": 0.8840125391849529, "train_speed(iter/s)": 1.44916 }, { "epoch": 3.0054410693629237, "grad_norm": 1.6778268814086914, "learning_rate": 3.438667086959388e-05, "loss": 0.22439355850219728, "memory(GiB)": 70.5, "step": 70150, "token_acc": 0.9504373177842566, "train_speed(iter/s)": 1.449156 }, { "epoch": 3.0056552846921725, "grad_norm": 2.689460039138794, "learning_rate": 3.438027776354671e-05, "loss": 0.41068143844604493, "memory(GiB)": 70.5, "step": 70155, "token_acc": 0.9054054054054054, "train_speed(iter/s)": 1.449156 }, { "epoch": 3.0058695000214217, "grad_norm": 4.275911808013916, "learning_rate": 3.437388494046545e-05, "loss": 0.23390092849731445, "memory(GiB)": 70.5, "step": 70160, "token_acc": 0.9390681003584229, "train_speed(iter/s)": 1.449163 }, { "epoch": 3.0060837153506705, "grad_norm": 0.910779595375061, "learning_rate": 3.4367492400465926e-05, "loss": 0.3882563829421997, "memory(GiB)": 70.5, "step": 70165, "token_acc": 0.9149659863945578, "train_speed(iter/s)": 1.449177 }, { "epoch": 3.0062979306799194, "grad_norm": 4.198979377746582, "learning_rate": 3.436110014366395e-05, "loss": 0.3504700899124146, "memory(GiB)": 70.5, "step": 70170, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.449188 }, { "epoch": 3.0065121460091686, "grad_norm": 0.6156359910964966, "learning_rate": 3.4354708170175296e-05, "loss": 0.3847563982009888, "memory(GiB)": 70.5, "step": 70175, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.449208 }, { "epoch": 3.0067263613384174, "grad_norm": 2.081590175628662, "learning_rate": 3.4348316480115786e-05, "loss": 0.4013237476348877, "memory(GiB)": 70.5, "step": 70180, "token_acc": 0.9131944444444444, "train_speed(iter/s)": 1.449212 }, { "epoch": 3.0069405766676662, "grad_norm": 2.749084234237671, "learning_rate": 3.4341925073601206e-05, "loss": 0.43490986824035643, "memory(GiB)": 70.5, "step": 70185, "token_acc": 0.9029850746268657, "train_speed(iter/s)": 1.44922 }, { "epoch": 3.0071547919969155, "grad_norm": 4.816529750823975, "learning_rate": 3.433553395074735e-05, "loss": 0.4981724739074707, "memory(GiB)": 70.5, "step": 70190, "token_acc": 0.8929889298892989, "train_speed(iter/s)": 1.449224 }, { "epoch": 3.0073690073261643, "grad_norm": 3.7480580806732178, "learning_rate": 3.432914311166998e-05, "loss": 0.23621795177459717, "memory(GiB)": 70.5, "step": 70195, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.449221 }, { "epoch": 3.007583222655413, "grad_norm": 2.765289545059204, "learning_rate": 3.4322752556484896e-05, "loss": 0.30035374164581297, "memory(GiB)": 70.5, "step": 70200, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.449224 }, { "epoch": 3.0077974379846624, "grad_norm": 2.708510160446167, "learning_rate": 3.4316362285307854e-05, "loss": 0.32959280014038084, "memory(GiB)": 70.5, "step": 70205, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.449227 }, { "epoch": 3.008011653313911, "grad_norm": 2.292823076248169, "learning_rate": 3.430997229825462e-05, "loss": 0.3684540748596191, "memory(GiB)": 70.5, "step": 70210, "token_acc": 0.9131652661064426, "train_speed(iter/s)": 1.449226 }, { "epoch": 3.00822586864316, "grad_norm": 3.683666467666626, "learning_rate": 3.430358259544095e-05, "loss": 0.2548251152038574, "memory(GiB)": 70.5, "step": 70215, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.44923 }, { "epoch": 3.0084400839724093, "grad_norm": 2.6410250663757324, "learning_rate": 3.429719317698262e-05, "loss": 0.5160991668701171, "memory(GiB)": 70.5, "step": 70220, "token_acc": 0.8845070422535212, "train_speed(iter/s)": 1.449238 }, { "epoch": 3.008654299301658, "grad_norm": 4.475489616394043, "learning_rate": 3.429080404299535e-05, "loss": 0.7672794818878174, "memory(GiB)": 70.5, "step": 70225, "token_acc": 0.8327402135231317, "train_speed(iter/s)": 1.449241 }, { "epoch": 3.008868514630907, "grad_norm": 2.9463694095611572, "learning_rate": 3.428441519359491e-05, "loss": 0.2415137767791748, "memory(GiB)": 70.5, "step": 70230, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.449248 }, { "epoch": 3.009082729960156, "grad_norm": 11.273773193359375, "learning_rate": 3.4278026628897024e-05, "loss": 0.2512695074081421, "memory(GiB)": 70.5, "step": 70235, "token_acc": 0.9337539432176656, "train_speed(iter/s)": 1.449248 }, { "epoch": 3.009296945289405, "grad_norm": 5.337444305419922, "learning_rate": 3.427163834901744e-05, "loss": 0.2638846874237061, "memory(GiB)": 70.5, "step": 70240, "token_acc": 0.941747572815534, "train_speed(iter/s)": 1.449256 }, { "epoch": 3.0095111606186538, "grad_norm": 3.557936429977417, "learning_rate": 3.426525035407189e-05, "loss": 0.316574239730835, "memory(GiB)": 70.5, "step": 70245, "token_acc": 0.9347181008902077, "train_speed(iter/s)": 1.449276 }, { "epoch": 3.009725375947903, "grad_norm": 2.702331781387329, "learning_rate": 3.425886264417607e-05, "loss": 0.4667813301086426, "memory(GiB)": 70.5, "step": 70250, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.449298 }, { "epoch": 3.009939591277152, "grad_norm": 1.012547492980957, "learning_rate": 3.425247521944572e-05, "loss": 0.3875896692276001, "memory(GiB)": 70.5, "step": 70255, "token_acc": 0.9438596491228071, "train_speed(iter/s)": 1.449322 }, { "epoch": 3.0101538066064006, "grad_norm": 2.608840227127075, "learning_rate": 3.424608807999656e-05, "loss": 0.4999388694763184, "memory(GiB)": 70.5, "step": 70260, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.449327 }, { "epoch": 3.01036802193565, "grad_norm": 0.0792476162314415, "learning_rate": 3.423970122594431e-05, "loss": 0.21458892822265624, "memory(GiB)": 70.5, "step": 70265, "token_acc": 0.9431818181818182, "train_speed(iter/s)": 1.44933 }, { "epoch": 3.0105822372648987, "grad_norm": 1.6878767013549805, "learning_rate": 3.423331465740463e-05, "loss": 0.3215686559677124, "memory(GiB)": 70.5, "step": 70270, "token_acc": 0.927797833935018, "train_speed(iter/s)": 1.449334 }, { "epoch": 3.0107964525941475, "grad_norm": 0.8255846500396729, "learning_rate": 3.4226928374493275e-05, "loss": 0.19720306396484374, "memory(GiB)": 70.5, "step": 70275, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.449342 }, { "epoch": 3.0110106679233968, "grad_norm": 3.593550682067871, "learning_rate": 3.422054237732588e-05, "loss": 0.471142053604126, "memory(GiB)": 70.5, "step": 70280, "token_acc": 0.907185628742515, "train_speed(iter/s)": 1.449345 }, { "epoch": 3.0112248832526456, "grad_norm": 3.3334853649139404, "learning_rate": 3.421415666601818e-05, "loss": 0.6976078510284424, "memory(GiB)": 70.5, "step": 70285, "token_acc": 0.8293515358361775, "train_speed(iter/s)": 1.449369 }, { "epoch": 3.0114390985818944, "grad_norm": 6.0684614181518555, "learning_rate": 3.420777124068584e-05, "loss": 0.5113982677459716, "memory(GiB)": 70.5, "step": 70290, "token_acc": 0.8949044585987261, "train_speed(iter/s)": 1.449379 }, { "epoch": 3.0116533139111437, "grad_norm": 2.3700718879699707, "learning_rate": 3.4201386101444524e-05, "loss": 0.2922682285308838, "memory(GiB)": 70.5, "step": 70295, "token_acc": 0.9309309309309309, "train_speed(iter/s)": 1.449379 }, { "epoch": 3.0118675292403925, "grad_norm": 0.6475743055343628, "learning_rate": 3.419500124840994e-05, "loss": 0.430401611328125, "memory(GiB)": 70.5, "step": 70300, "token_acc": 0.8850574712643678, "train_speed(iter/s)": 1.449378 }, { "epoch": 3.0120817445696413, "grad_norm": 5.104653358459473, "learning_rate": 3.418861668169773e-05, "loss": 0.21427621841430664, "memory(GiB)": 70.5, "step": 70305, "token_acc": 0.9204892966360856, "train_speed(iter/s)": 1.449377 }, { "epoch": 3.0122959598988905, "grad_norm": 3.318659782409668, "learning_rate": 3.4182232401423544e-05, "loss": 0.3986325740814209, "memory(GiB)": 70.5, "step": 70310, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.449375 }, { "epoch": 3.0125101752281394, "grad_norm": 2.617499351501465, "learning_rate": 3.417584840770307e-05, "loss": 0.32375140190124513, "memory(GiB)": 70.5, "step": 70315, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.449414 }, { "epoch": 3.012724390557388, "grad_norm": 1.4143368005752563, "learning_rate": 3.416946470065193e-05, "loss": 0.13687047958374024, "memory(GiB)": 70.5, "step": 70320, "token_acc": 0.96415770609319, "train_speed(iter/s)": 1.449416 }, { "epoch": 3.0129386058866374, "grad_norm": 4.591506004333496, "learning_rate": 3.416308128038581e-05, "loss": 0.4509040355682373, "memory(GiB)": 70.5, "step": 70325, "token_acc": 0.9169139465875371, "train_speed(iter/s)": 1.449437 }, { "epoch": 3.0131528212158862, "grad_norm": 0.6790614724159241, "learning_rate": 3.415669814702031e-05, "loss": 0.2253645896911621, "memory(GiB)": 70.5, "step": 70330, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.449442 }, { "epoch": 3.013367036545135, "grad_norm": 6.076555252075195, "learning_rate": 3.415031530067109e-05, "loss": 0.2893320322036743, "memory(GiB)": 70.5, "step": 70335, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.449442 }, { "epoch": 3.0135812518743843, "grad_norm": 2.122633457183838, "learning_rate": 3.414393274145378e-05, "loss": 0.2631540060043335, "memory(GiB)": 70.5, "step": 70340, "token_acc": 0.9435215946843853, "train_speed(iter/s)": 1.449442 }, { "epoch": 3.013795467203633, "grad_norm": 4.690642833709717, "learning_rate": 3.4137550469484e-05, "loss": 0.3094993352890015, "memory(GiB)": 70.5, "step": 70345, "token_acc": 0.9330543933054394, "train_speed(iter/s)": 1.449445 }, { "epoch": 3.014009682532882, "grad_norm": 0.37947383522987366, "learning_rate": 3.4131168484877385e-05, "loss": 0.3379755735397339, "memory(GiB)": 70.5, "step": 70350, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.449461 }, { "epoch": 3.014223897862131, "grad_norm": 0.8652358651161194, "learning_rate": 3.412478678774952e-05, "loss": 0.3286309719085693, "memory(GiB)": 70.5, "step": 70355, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.449497 }, { "epoch": 3.01443811319138, "grad_norm": 7.012746810913086, "learning_rate": 3.4118405378216056e-05, "loss": 0.5886387825012207, "memory(GiB)": 70.5, "step": 70360, "token_acc": 0.881578947368421, "train_speed(iter/s)": 1.449495 }, { "epoch": 3.014652328520629, "grad_norm": 3.0213301181793213, "learning_rate": 3.411202425639257e-05, "loss": 0.32210052013397217, "memory(GiB)": 70.5, "step": 70365, "token_acc": 0.9153846153846154, "train_speed(iter/s)": 1.449499 }, { "epoch": 3.014866543849878, "grad_norm": 2.072995662689209, "learning_rate": 3.410564342239466e-05, "loss": 0.4133634090423584, "memory(GiB)": 70.5, "step": 70370, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.449505 }, { "epoch": 3.015080759179127, "grad_norm": 1.2702866792678833, "learning_rate": 3.409926287633793e-05, "loss": 0.313975191116333, "memory(GiB)": 70.5, "step": 70375, "token_acc": 0.9260450160771704, "train_speed(iter/s)": 1.449509 }, { "epoch": 3.0152949745083757, "grad_norm": 2.973428249359131, "learning_rate": 3.409288261833797e-05, "loss": 0.3067629814147949, "memory(GiB)": 70.5, "step": 70380, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.449518 }, { "epoch": 3.015509189837625, "grad_norm": 3.0751497745513916, "learning_rate": 3.408650264851038e-05, "loss": 0.3618943214416504, "memory(GiB)": 70.5, "step": 70385, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.449516 }, { "epoch": 3.0157234051668738, "grad_norm": 2.5677008628845215, "learning_rate": 3.4080122966970705e-05, "loss": 0.2884681701660156, "memory(GiB)": 70.5, "step": 70390, "token_acc": 0.938953488372093, "train_speed(iter/s)": 1.449513 }, { "epoch": 3.0159376204961226, "grad_norm": 4.124676704406738, "learning_rate": 3.4073743573834535e-05, "loss": 0.468072509765625, "memory(GiB)": 70.5, "step": 70395, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.449521 }, { "epoch": 3.016151835825372, "grad_norm": 10.353225708007812, "learning_rate": 3.406736446921743e-05, "loss": 0.46955041885375975, "memory(GiB)": 70.5, "step": 70400, "token_acc": 0.9083333333333333, "train_speed(iter/s)": 1.449535 }, { "epoch": 3.0163660511546206, "grad_norm": 5.0717291831970215, "learning_rate": 3.406098565323499e-05, "loss": 0.4505149841308594, "memory(GiB)": 70.5, "step": 70405, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.449551 }, { "epoch": 3.0165802664838695, "grad_norm": 5.809447288513184, "learning_rate": 3.4054607126002734e-05, "loss": 0.2838095426559448, "memory(GiB)": 70.5, "step": 70410, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.449556 }, { "epoch": 3.0167944818131187, "grad_norm": 4.036857604980469, "learning_rate": 3.404822888763623e-05, "loss": 0.24555978775024415, "memory(GiB)": 70.5, "step": 70415, "token_acc": 0.9360902255639098, "train_speed(iter/s)": 1.449573 }, { "epoch": 3.0170086971423675, "grad_norm": 2.86064076423645, "learning_rate": 3.404185093825103e-05, "loss": 0.607453727722168, "memory(GiB)": 70.5, "step": 70420, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.449573 }, { "epoch": 3.0172229124716163, "grad_norm": 4.690732479095459, "learning_rate": 3.403547327796266e-05, "loss": 0.30834188461303713, "memory(GiB)": 70.5, "step": 70425, "token_acc": 0.9467084639498433, "train_speed(iter/s)": 1.449578 }, { "epoch": 3.0174371278008656, "grad_norm": 1.773501992225647, "learning_rate": 3.4029095906886676e-05, "loss": 0.26253390312194824, "memory(GiB)": 70.5, "step": 70430, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.44958 }, { "epoch": 3.0176513431301144, "grad_norm": 2.037996768951416, "learning_rate": 3.40227188251386e-05, "loss": 0.18847419023513795, "memory(GiB)": 70.5, "step": 70435, "token_acc": 0.9566666666666667, "train_speed(iter/s)": 1.449595 }, { "epoch": 3.017865558459363, "grad_norm": 2.835047960281372, "learning_rate": 3.401634203283395e-05, "loss": 0.6443010330200195, "memory(GiB)": 70.5, "step": 70440, "token_acc": 0.8565891472868217, "train_speed(iter/s)": 1.449605 }, { "epoch": 3.0180797737886125, "grad_norm": 2.616880416870117, "learning_rate": 3.4009965530088274e-05, "loss": 0.28308155536651614, "memory(GiB)": 70.5, "step": 70445, "token_acc": 0.9243697478991597, "train_speed(iter/s)": 1.449621 }, { "epoch": 3.0182939891178613, "grad_norm": 3.2157466411590576, "learning_rate": 3.4003589317017074e-05, "loss": 0.3926701545715332, "memory(GiB)": 70.5, "step": 70450, "token_acc": 0.9235474006116208, "train_speed(iter/s)": 1.449614 }, { "epoch": 3.01850820444711, "grad_norm": 3.7816200256347656, "learning_rate": 3.399721339373584e-05, "loss": 0.34666850566864016, "memory(GiB)": 70.5, "step": 70455, "token_acc": 0.9145299145299145, "train_speed(iter/s)": 1.449629 }, { "epoch": 3.0187224197763594, "grad_norm": 2.4959566593170166, "learning_rate": 3.3990837760360106e-05, "loss": 0.4360086917877197, "memory(GiB)": 70.5, "step": 70460, "token_acc": 0.9039039039039038, "train_speed(iter/s)": 1.449632 }, { "epoch": 3.018936635105608, "grad_norm": 1.7150770425796509, "learning_rate": 3.398446241700536e-05, "loss": 0.36735217571258544, "memory(GiB)": 70.5, "step": 70465, "token_acc": 0.9287925696594427, "train_speed(iter/s)": 1.449644 }, { "epoch": 3.019150850434857, "grad_norm": 1.9935675859451294, "learning_rate": 3.397808736378711e-05, "loss": 0.43543457984924316, "memory(GiB)": 70.5, "step": 70470, "token_acc": 0.8876811594202898, "train_speed(iter/s)": 1.44965 }, { "epoch": 3.0193650657641062, "grad_norm": 1.5519392490386963, "learning_rate": 3.397171260082082e-05, "loss": 0.28419055938720705, "memory(GiB)": 70.5, "step": 70475, "token_acc": 0.9277566539923955, "train_speed(iter/s)": 1.449648 }, { "epoch": 3.019579281093355, "grad_norm": 8.211771011352539, "learning_rate": 3.3965338128222e-05, "loss": 0.2952070713043213, "memory(GiB)": 70.5, "step": 70480, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.449659 }, { "epoch": 3.019793496422604, "grad_norm": 4.846519470214844, "learning_rate": 3.3958963946106126e-05, "loss": 0.4218573570251465, "memory(GiB)": 70.5, "step": 70485, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.449662 }, { "epoch": 3.020007711751853, "grad_norm": 2.1833629608154297, "learning_rate": 3.3952590054588675e-05, "loss": 0.4641568183898926, "memory(GiB)": 70.5, "step": 70490, "token_acc": 0.8935361216730038, "train_speed(iter/s)": 1.449669 }, { "epoch": 3.020221927081102, "grad_norm": 4.975493431091309, "learning_rate": 3.394621645378511e-05, "loss": 0.5910791397094727, "memory(GiB)": 70.5, "step": 70495, "token_acc": 0.8708609271523179, "train_speed(iter/s)": 1.449679 }, { "epoch": 3.0204361424103507, "grad_norm": 0.8360894918441772, "learning_rate": 3.3939843143810874e-05, "loss": 0.18357405662536622, "memory(GiB)": 70.5, "step": 70500, "token_acc": 0.9429657794676806, "train_speed(iter/s)": 1.449689 }, { "epoch": 3.0204361424103507, "eval_loss": 2.4050495624542236, "eval_runtime": 11.3095, "eval_samples_per_second": 8.842, "eval_steps_per_second": 8.842, "eval_token_acc": 0.45860709592641263, "step": 70500 }, { "epoch": 3.0206503577396, "grad_norm": 3.805227518081665, "learning_rate": 3.3933470124781466e-05, "loss": 0.2439131736755371, "memory(GiB)": 70.5, "step": 70505, "token_acc": 0.5852713178294574, "train_speed(iter/s)": 1.449321 }, { "epoch": 3.020864573068849, "grad_norm": 3.314934253692627, "learning_rate": 3.3927097396812304e-05, "loss": 0.4360383987426758, "memory(GiB)": 70.5, "step": 70510, "token_acc": 0.9061371841155235, "train_speed(iter/s)": 1.449324 }, { "epoch": 3.0210787883980976, "grad_norm": 1.1006669998168945, "learning_rate": 3.3920724960018856e-05, "loss": 0.49227347373962405, "memory(GiB)": 70.5, "step": 70515, "token_acc": 0.8954703832752613, "train_speed(iter/s)": 1.44932 }, { "epoch": 3.021293003727347, "grad_norm": 3.564971685409546, "learning_rate": 3.3914352814516556e-05, "loss": 0.45485563278198243, "memory(GiB)": 70.5, "step": 70520, "token_acc": 0.8932926829268293, "train_speed(iter/s)": 1.449331 }, { "epoch": 3.0215072190565957, "grad_norm": 4.114902019500732, "learning_rate": 3.390798096042085e-05, "loss": 0.35423669815063474, "memory(GiB)": 70.5, "step": 70525, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.449338 }, { "epoch": 3.0217214343858445, "grad_norm": 3.4866387844085693, "learning_rate": 3.390160939784716e-05, "loss": 0.6280174732208252, "memory(GiB)": 70.5, "step": 70530, "token_acc": 0.8754325259515571, "train_speed(iter/s)": 1.449353 }, { "epoch": 3.0219356497150938, "grad_norm": 4.422408103942871, "learning_rate": 3.3895238126910924e-05, "loss": 0.6231905460357666, "memory(GiB)": 70.5, "step": 70535, "token_acc": 0.8662790697674418, "train_speed(iter/s)": 1.449356 }, { "epoch": 3.0221498650443426, "grad_norm": 5.551548004150391, "learning_rate": 3.388886714772754e-05, "loss": 0.3972719669342041, "memory(GiB)": 70.5, "step": 70540, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.449359 }, { "epoch": 3.0223640803735914, "grad_norm": 4.326183795928955, "learning_rate": 3.388249646041246e-05, "loss": 0.7130504608154297, "memory(GiB)": 70.5, "step": 70545, "token_acc": 0.8666666666666667, "train_speed(iter/s)": 1.449382 }, { "epoch": 3.0225782957028406, "grad_norm": 0.7297135591506958, "learning_rate": 3.387612606508105e-05, "loss": 0.21407535076141357, "memory(GiB)": 70.5, "step": 70550, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.449384 }, { "epoch": 3.0227925110320895, "grad_norm": 0.28670522570610046, "learning_rate": 3.386975596184877e-05, "loss": 0.6168899536132812, "memory(GiB)": 70.5, "step": 70555, "token_acc": 0.8881987577639752, "train_speed(iter/s)": 1.449395 }, { "epoch": 3.0230067263613383, "grad_norm": 4.212818145751953, "learning_rate": 3.3863386150830974e-05, "loss": 0.31611220836639403, "memory(GiB)": 70.5, "step": 70560, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.449398 }, { "epoch": 3.0232209416905875, "grad_norm": 2.302762508392334, "learning_rate": 3.385701663214309e-05, "loss": 0.294032883644104, "memory(GiB)": 70.5, "step": 70565, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.449405 }, { "epoch": 3.0234351570198363, "grad_norm": 0.20201076567173004, "learning_rate": 3.385064740590048e-05, "loss": 0.285203742980957, "memory(GiB)": 70.5, "step": 70570, "token_acc": 0.9422222222222222, "train_speed(iter/s)": 1.449417 }, { "epoch": 3.023649372349085, "grad_norm": 2.725642442703247, "learning_rate": 3.384427847221855e-05, "loss": 0.49435739517211913, "memory(GiB)": 70.5, "step": 70575, "token_acc": 0.8879781420765027, "train_speed(iter/s)": 1.449424 }, { "epoch": 3.0238635876783344, "grad_norm": 1.3901491165161133, "learning_rate": 3.383790983121267e-05, "loss": 0.4196625709533691, "memory(GiB)": 70.5, "step": 70580, "token_acc": 0.9226666666666666, "train_speed(iter/s)": 1.449424 }, { "epoch": 3.024077803007583, "grad_norm": 2.9642977714538574, "learning_rate": 3.3831541482998205e-05, "loss": 0.3774879455566406, "memory(GiB)": 70.5, "step": 70585, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.449433 }, { "epoch": 3.024292018336832, "grad_norm": 5.43295955657959, "learning_rate": 3.3825173427690546e-05, "loss": 0.5706794738769532, "memory(GiB)": 70.5, "step": 70590, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.449447 }, { "epoch": 3.0245062336660813, "grad_norm": 2.661449432373047, "learning_rate": 3.381880566540505e-05, "loss": 0.46262273788452146, "memory(GiB)": 70.5, "step": 70595, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.44945 }, { "epoch": 3.02472044899533, "grad_norm": 4.311666488647461, "learning_rate": 3.381243819625705e-05, "loss": 0.22862367630004882, "memory(GiB)": 70.5, "step": 70600, "token_acc": 0.9419354838709677, "train_speed(iter/s)": 1.449453 }, { "epoch": 3.024934664324579, "grad_norm": 0.10888306051492691, "learning_rate": 3.380607102036193e-05, "loss": 0.18146179914474486, "memory(GiB)": 70.5, "step": 70605, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.449449 }, { "epoch": 3.025148879653828, "grad_norm": 3.2226502895355225, "learning_rate": 3.3799704137835014e-05, "loss": 0.3290407180786133, "memory(GiB)": 70.5, "step": 70610, "token_acc": 0.9119718309859155, "train_speed(iter/s)": 1.449454 }, { "epoch": 3.025363094983077, "grad_norm": 4.1622796058654785, "learning_rate": 3.3793337548791666e-05, "loss": 0.45887956619262693, "memory(GiB)": 70.5, "step": 70615, "token_acc": 0.8954703832752613, "train_speed(iter/s)": 1.449457 }, { "epoch": 3.025577310312326, "grad_norm": 4.268404960632324, "learning_rate": 3.378697125334721e-05, "loss": 0.8454802513122559, "memory(GiB)": 70.5, "step": 70620, "token_acc": 0.8265682656826568, "train_speed(iter/s)": 1.449462 }, { "epoch": 3.025791525641575, "grad_norm": 4.393484115600586, "learning_rate": 3.378060525161696e-05, "loss": 0.3013388395309448, "memory(GiB)": 70.5, "step": 70625, "token_acc": 0.930327868852459, "train_speed(iter/s)": 1.44947 }, { "epoch": 3.026005740970824, "grad_norm": 4.77984619140625, "learning_rate": 3.377423954371627e-05, "loss": 0.46789937019348143, "memory(GiB)": 70.5, "step": 70630, "token_acc": 0.9060402684563759, "train_speed(iter/s)": 1.449473 }, { "epoch": 3.0262199563000727, "grad_norm": 4.265554428100586, "learning_rate": 3.3767874129760455e-05, "loss": 0.5651033401489258, "memory(GiB)": 70.5, "step": 70635, "token_acc": 0.8556701030927835, "train_speed(iter/s)": 1.449487 }, { "epoch": 3.026434171629322, "grad_norm": 6.989786148071289, "learning_rate": 3.3761509009864834e-05, "loss": 0.4789254665374756, "memory(GiB)": 70.5, "step": 70640, "token_acc": 0.899390243902439, "train_speed(iter/s)": 1.4495 }, { "epoch": 3.0266483869585707, "grad_norm": 3.413773775100708, "learning_rate": 3.3755144184144704e-05, "loss": 0.1996924877166748, "memory(GiB)": 70.5, "step": 70645, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.449503 }, { "epoch": 3.0268626022878196, "grad_norm": 3.0238990783691406, "learning_rate": 3.374877965271538e-05, "loss": 0.4546391487121582, "memory(GiB)": 70.5, "step": 70650, "token_acc": 0.9026845637583892, "train_speed(iter/s)": 1.449501 }, { "epoch": 3.027076817617069, "grad_norm": 5.039183616638184, "learning_rate": 3.374241541569216e-05, "loss": 0.2206845760345459, "memory(GiB)": 70.5, "step": 70655, "token_acc": 0.9447852760736196, "train_speed(iter/s)": 1.449508 }, { "epoch": 3.0272910329463176, "grad_norm": 2.1847729682922363, "learning_rate": 3.3736051473190324e-05, "loss": 0.7996976375579834, "memory(GiB)": 70.5, "step": 70660, "token_acc": 0.858433734939759, "train_speed(iter/s)": 1.449517 }, { "epoch": 3.0275052482755664, "grad_norm": 4.079316139221191, "learning_rate": 3.372968782532519e-05, "loss": 0.4120847225189209, "memory(GiB)": 70.5, "step": 70665, "token_acc": 0.9235474006116208, "train_speed(iter/s)": 1.449524 }, { "epoch": 3.0277194636048157, "grad_norm": 0.054799437522888184, "learning_rate": 3.3723324472212e-05, "loss": 0.43191752433776853, "memory(GiB)": 70.5, "step": 70670, "token_acc": 0.9097744360902256, "train_speed(iter/s)": 1.449533 }, { "epoch": 3.0279336789340645, "grad_norm": 4.795840263366699, "learning_rate": 3.371696141396608e-05, "loss": 0.2978111505508423, "memory(GiB)": 70.5, "step": 70675, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.449538 }, { "epoch": 3.0281478942633133, "grad_norm": 0.9034021496772766, "learning_rate": 3.371059865070266e-05, "loss": 0.16992136240005493, "memory(GiB)": 70.5, "step": 70680, "token_acc": 0.9427480916030534, "train_speed(iter/s)": 1.449541 }, { "epoch": 3.0283621095925626, "grad_norm": 2.851041316986084, "learning_rate": 3.370423618253703e-05, "loss": 0.35426530838012693, "memory(GiB)": 70.5, "step": 70685, "token_acc": 0.916030534351145, "train_speed(iter/s)": 1.449547 }, { "epoch": 3.0285763249218114, "grad_norm": 4.586489677429199, "learning_rate": 3.369787400958446e-05, "loss": 0.32856295108795164, "memory(GiB)": 70.5, "step": 70690, "token_acc": 0.9326599326599326, "train_speed(iter/s)": 1.449548 }, { "epoch": 3.02879054025106, "grad_norm": 4.4781365394592285, "learning_rate": 3.369151213196019e-05, "loss": 0.501804256439209, "memory(GiB)": 70.5, "step": 70695, "token_acc": 0.8840579710144928, "train_speed(iter/s)": 1.449562 }, { "epoch": 3.0290047555803095, "grad_norm": 1.8471401929855347, "learning_rate": 3.368515054977945e-05, "loss": 0.30791027545928956, "memory(GiB)": 70.5, "step": 70700, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.449559 }, { "epoch": 3.0292189709095583, "grad_norm": 3.569230318069458, "learning_rate": 3.3678789263157534e-05, "loss": 0.24691619873046874, "memory(GiB)": 70.5, "step": 70705, "token_acc": 0.9537037037037037, "train_speed(iter/s)": 1.449563 }, { "epoch": 3.029433186238807, "grad_norm": 3.8857626914978027, "learning_rate": 3.367242827220967e-05, "loss": 0.3464330196380615, "memory(GiB)": 70.5, "step": 70710, "token_acc": 0.9093406593406593, "train_speed(iter/s)": 1.449565 }, { "epoch": 3.0296474015680563, "grad_norm": 3.2587907314300537, "learning_rate": 3.3666067577051064e-05, "loss": 0.2355710744857788, "memory(GiB)": 70.5, "step": 70715, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.449569 }, { "epoch": 3.029861616897305, "grad_norm": 4.2076416015625, "learning_rate": 3.3659707177796984e-05, "loss": 0.3136449337005615, "memory(GiB)": 70.5, "step": 70720, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.449576 }, { "epoch": 3.030075832226554, "grad_norm": 0.4581911265850067, "learning_rate": 3.365334707456264e-05, "loss": 0.09119341969490051, "memory(GiB)": 70.5, "step": 70725, "token_acc": 0.9810606060606061, "train_speed(iter/s)": 1.449583 }, { "epoch": 3.0302900475558032, "grad_norm": 0.018073663115501404, "learning_rate": 3.364698726746323e-05, "loss": 0.24485113620758056, "memory(GiB)": 70.5, "step": 70730, "token_acc": 0.9558823529411765, "train_speed(iter/s)": 1.449591 }, { "epoch": 3.030504262885052, "grad_norm": 4.116087436676025, "learning_rate": 3.364062775661401e-05, "loss": 0.3893077850341797, "memory(GiB)": 70.5, "step": 70735, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.449592 }, { "epoch": 3.030718478214301, "grad_norm": 2.8227832317352295, "learning_rate": 3.3634268542130146e-05, "loss": 0.19192962646484374, "memory(GiB)": 70.5, "step": 70740, "token_acc": 0.9548872180451128, "train_speed(iter/s)": 1.449594 }, { "epoch": 3.03093269354355, "grad_norm": 1.156556248664856, "learning_rate": 3.3627909624126864e-05, "loss": 0.14465032815933226, "memory(GiB)": 70.5, "step": 70745, "token_acc": 0.9666666666666667, "train_speed(iter/s)": 1.449598 }, { "epoch": 3.031146908872799, "grad_norm": 3.136187791824341, "learning_rate": 3.362155100271936e-05, "loss": 0.24695172309875488, "memory(GiB)": 70.5, "step": 70750, "token_acc": 0.9396551724137931, "train_speed(iter/s)": 1.449602 }, { "epoch": 3.0313611242020477, "grad_norm": 5.780299663543701, "learning_rate": 3.361519267802281e-05, "loss": 0.30298514366149903, "memory(GiB)": 70.5, "step": 70755, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.449596 }, { "epoch": 3.031575339531297, "grad_norm": 6.735627174377441, "learning_rate": 3.360883465015243e-05, "loss": 0.41071228981018065, "memory(GiB)": 70.5, "step": 70760, "token_acc": 0.9083665338645418, "train_speed(iter/s)": 1.449594 }, { "epoch": 3.031789554860546, "grad_norm": 5.242948055267334, "learning_rate": 3.3602476919223393e-05, "loss": 0.5768772602081299, "memory(GiB)": 70.5, "step": 70765, "token_acc": 0.8766666666666667, "train_speed(iter/s)": 1.449597 }, { "epoch": 3.0320037701897946, "grad_norm": 7.721283912658691, "learning_rate": 3.359611948535084e-05, "loss": 0.41194725036621094, "memory(GiB)": 70.5, "step": 70770, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.449609 }, { "epoch": 3.032217985519044, "grad_norm": 4.86944580078125, "learning_rate": 3.358976234864999e-05, "loss": 0.5081912040710449, "memory(GiB)": 70.5, "step": 70775, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.449606 }, { "epoch": 3.0324322008482927, "grad_norm": 5.755221366882324, "learning_rate": 3.3583405509235986e-05, "loss": 0.3254260063171387, "memory(GiB)": 70.5, "step": 70780, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.449606 }, { "epoch": 3.0326464161775415, "grad_norm": 2.838853359222412, "learning_rate": 3.3577048967224e-05, "loss": 0.3084479570388794, "memory(GiB)": 70.5, "step": 70785, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.449622 }, { "epoch": 3.0328606315067907, "grad_norm": 5.075512409210205, "learning_rate": 3.3570692722729167e-05, "loss": 0.38902835845947265, "memory(GiB)": 70.5, "step": 70790, "token_acc": 0.8929577464788733, "train_speed(iter/s)": 1.449625 }, { "epoch": 3.0330748468360396, "grad_norm": 6.308437347412109, "learning_rate": 3.3564336775866654e-05, "loss": 0.512471342086792, "memory(GiB)": 70.5, "step": 70795, "token_acc": 0.8996282527881041, "train_speed(iter/s)": 1.449634 }, { "epoch": 3.0332890621652884, "grad_norm": 4.143803596496582, "learning_rate": 3.35579811267516e-05, "loss": 0.2730919361114502, "memory(GiB)": 70.5, "step": 70800, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.449629 }, { "epoch": 3.0335032774945376, "grad_norm": 3.7269084453582764, "learning_rate": 3.355162577549913e-05, "loss": 0.35512707233428953, "memory(GiB)": 70.5, "step": 70805, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.44963 }, { "epoch": 3.0337174928237864, "grad_norm": 4.433131694793701, "learning_rate": 3.35452707222244e-05, "loss": 0.39910092353820803, "memory(GiB)": 70.5, "step": 70810, "token_acc": 0.9232954545454546, "train_speed(iter/s)": 1.44963 }, { "epoch": 3.0339317081530353, "grad_norm": 0.1524468958377838, "learning_rate": 3.353891596704252e-05, "loss": 0.22273879051208495, "memory(GiB)": 70.5, "step": 70815, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.449637 }, { "epoch": 3.0341459234822845, "grad_norm": 5.452152252197266, "learning_rate": 3.353256151006863e-05, "loss": 0.5160528182983398, "memory(GiB)": 70.5, "step": 70820, "token_acc": 0.8986486486486487, "train_speed(iter/s)": 1.44964 }, { "epoch": 3.0343601388115333, "grad_norm": 3.471271514892578, "learning_rate": 3.3526207351417825e-05, "loss": 0.4475379467010498, "memory(GiB)": 70.5, "step": 70825, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.449642 }, { "epoch": 3.034574354140782, "grad_norm": 2.598069429397583, "learning_rate": 3.351985349120523e-05, "loss": 0.5515995979309082, "memory(GiB)": 70.5, "step": 70830, "token_acc": 0.8872727272727273, "train_speed(iter/s)": 1.449652 }, { "epoch": 3.0347885694700314, "grad_norm": 3.6459364891052246, "learning_rate": 3.351349992954595e-05, "loss": 0.3481677293777466, "memory(GiB)": 70.5, "step": 70835, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.449655 }, { "epoch": 3.03500278479928, "grad_norm": 1.4024618864059448, "learning_rate": 3.350714666655509e-05, "loss": 0.5022400856018067, "memory(GiB)": 70.5, "step": 70840, "token_acc": 0.9152046783625731, "train_speed(iter/s)": 1.449663 }, { "epoch": 3.035217000128529, "grad_norm": 3.2859771251678467, "learning_rate": 3.3500793702347716e-05, "loss": 0.24667253494262695, "memory(GiB)": 70.5, "step": 70845, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.449674 }, { "epoch": 3.0354312154577783, "grad_norm": 5.3726115226745605, "learning_rate": 3.349444103703895e-05, "loss": 0.4253093719482422, "memory(GiB)": 70.5, "step": 70850, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.449689 }, { "epoch": 3.035645430787027, "grad_norm": 5.233246803283691, "learning_rate": 3.3488088670743886e-05, "loss": 0.21613457202911376, "memory(GiB)": 70.5, "step": 70855, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.449697 }, { "epoch": 3.035859646116276, "grad_norm": 4.995153903961182, "learning_rate": 3.3481736603577565e-05, "loss": 0.5667738437652587, "memory(GiB)": 70.5, "step": 70860, "token_acc": 0.8637992831541219, "train_speed(iter/s)": 1.44971 }, { "epoch": 3.036073861445525, "grad_norm": 3.609783887863159, "learning_rate": 3.34753848356551e-05, "loss": 0.24305391311645508, "memory(GiB)": 70.5, "step": 70865, "token_acc": 0.9293286219081273, "train_speed(iter/s)": 1.449708 }, { "epoch": 3.036288076774774, "grad_norm": 5.565057754516602, "learning_rate": 3.3469033367091534e-05, "loss": 0.4419837474822998, "memory(GiB)": 70.5, "step": 70870, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.449715 }, { "epoch": 3.036502292104023, "grad_norm": 2.1255009174346924, "learning_rate": 3.3462682198001925e-05, "loss": 0.2746814966201782, "memory(GiB)": 70.5, "step": 70875, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.449711 }, { "epoch": 3.036716507433272, "grad_norm": 2.7788195610046387, "learning_rate": 3.345633132850136e-05, "loss": 0.22620165348052979, "memory(GiB)": 70.5, "step": 70880, "token_acc": 0.9525547445255474, "train_speed(iter/s)": 1.44972 }, { "epoch": 3.036930722762521, "grad_norm": 4.124180316925049, "learning_rate": 3.344998075870487e-05, "loss": 0.4122814655303955, "memory(GiB)": 70.5, "step": 70885, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.449732 }, { "epoch": 3.0371449380917697, "grad_norm": 4.672472953796387, "learning_rate": 3.344363048872748e-05, "loss": 0.5049108982086181, "memory(GiB)": 70.5, "step": 70890, "token_acc": 0.9063545150501672, "train_speed(iter/s)": 1.449748 }, { "epoch": 3.037359153421019, "grad_norm": 4.431844711303711, "learning_rate": 3.3437280518684275e-05, "loss": 0.3907366514205933, "memory(GiB)": 70.5, "step": 70895, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.449749 }, { "epoch": 3.0375733687502677, "grad_norm": 3.343125820159912, "learning_rate": 3.3430930848690266e-05, "loss": 0.4290459632873535, "memory(GiB)": 70.5, "step": 70900, "token_acc": 0.9173228346456693, "train_speed(iter/s)": 1.449754 }, { "epoch": 3.0377875840795165, "grad_norm": 4.531734466552734, "learning_rate": 3.3424581478860495e-05, "loss": 0.48937430381774905, "memory(GiB)": 70.5, "step": 70905, "token_acc": 0.8756756756756757, "train_speed(iter/s)": 1.449759 }, { "epoch": 3.038001799408766, "grad_norm": 5.619024276733398, "learning_rate": 3.3418232409309975e-05, "loss": 0.20166680812835694, "memory(GiB)": 70.5, "step": 70910, "token_acc": 0.954983922829582, "train_speed(iter/s)": 1.449766 }, { "epoch": 3.0382160147380146, "grad_norm": 3.2186288833618164, "learning_rate": 3.341188364015372e-05, "loss": 0.3157422304153442, "memory(GiB)": 70.5, "step": 70915, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.449776 }, { "epoch": 3.0384302300672634, "grad_norm": 3.5172078609466553, "learning_rate": 3.340553517150674e-05, "loss": 0.14630062580108644, "memory(GiB)": 70.5, "step": 70920, "token_acc": 0.9704641350210971, "train_speed(iter/s)": 1.449781 }, { "epoch": 3.0386444453965127, "grad_norm": 3.385392189025879, "learning_rate": 3.339918700348409e-05, "loss": 0.1917886734008789, "memory(GiB)": 70.5, "step": 70925, "token_acc": 0.9480968858131488, "train_speed(iter/s)": 1.449786 }, { "epoch": 3.0388586607257615, "grad_norm": 4.127102851867676, "learning_rate": 3.339283913620073e-05, "loss": 0.33962535858154297, "memory(GiB)": 70.5, "step": 70930, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.4498 }, { "epoch": 3.0390728760550103, "grad_norm": 2.8964126110076904, "learning_rate": 3.338649156977166e-05, "loss": 0.44672622680664065, "memory(GiB)": 70.5, "step": 70935, "token_acc": 0.8902077151335311, "train_speed(iter/s)": 1.449816 }, { "epoch": 3.0392870913842596, "grad_norm": 0.34681349992752075, "learning_rate": 3.338014430431189e-05, "loss": 0.2153319835662842, "memory(GiB)": 70.5, "step": 70940, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.449818 }, { "epoch": 3.0395013067135084, "grad_norm": 3.6531357765197754, "learning_rate": 3.33737973399364e-05, "loss": 0.4646427154541016, "memory(GiB)": 70.5, "step": 70945, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.449821 }, { "epoch": 3.039715522042757, "grad_norm": 1.429765224456787, "learning_rate": 3.336745067676015e-05, "loss": 0.2680067539215088, "memory(GiB)": 70.5, "step": 70950, "token_acc": 0.9379562043795621, "train_speed(iter/s)": 1.449822 }, { "epoch": 3.0399297373720064, "grad_norm": 2.940812826156616, "learning_rate": 3.336110431489815e-05, "loss": 0.19051146507263184, "memory(GiB)": 70.5, "step": 70955, "token_acc": 0.9572953736654805, "train_speed(iter/s)": 1.449828 }, { "epoch": 3.0401439527012553, "grad_norm": 4.481710433959961, "learning_rate": 3.3354758254465346e-05, "loss": 0.21495041847229004, "memory(GiB)": 70.5, "step": 70960, "token_acc": 0.9691780821917808, "train_speed(iter/s)": 1.449832 }, { "epoch": 3.040358168030504, "grad_norm": 2.6841890811920166, "learning_rate": 3.334841249557672e-05, "loss": 0.41379218101501464, "memory(GiB)": 70.5, "step": 70965, "token_acc": 0.9077380952380952, "train_speed(iter/s)": 1.449836 }, { "epoch": 3.0405723833597533, "grad_norm": 1.5429673194885254, "learning_rate": 3.334206703834721e-05, "loss": 0.2292252779006958, "memory(GiB)": 70.5, "step": 70970, "token_acc": 0.9517045454545454, "train_speed(iter/s)": 1.449849 }, { "epoch": 3.040786598689002, "grad_norm": 2.723637104034424, "learning_rate": 3.333572188289179e-05, "loss": 0.33142948150634766, "memory(GiB)": 70.5, "step": 70975, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.449851 }, { "epoch": 3.041000814018251, "grad_norm": 2.9077281951904297, "learning_rate": 3.3329377029325396e-05, "loss": 0.5083657741546631, "memory(GiB)": 70.5, "step": 70980, "token_acc": 0.8856088560885609, "train_speed(iter/s)": 1.44986 }, { "epoch": 3.0412150293475, "grad_norm": 0.6437382102012634, "learning_rate": 3.3323032477762984e-05, "loss": 0.2423677921295166, "memory(GiB)": 70.5, "step": 70985, "token_acc": 0.9516129032258065, "train_speed(iter/s)": 1.449875 }, { "epoch": 3.041429244676749, "grad_norm": 3.3376166820526123, "learning_rate": 3.331668822831947e-05, "loss": 0.44129328727722167, "memory(GiB)": 70.5, "step": 70990, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.44988 }, { "epoch": 3.041643460005998, "grad_norm": 3.451761245727539, "learning_rate": 3.331034428110979e-05, "loss": 0.5048965454101563, "memory(GiB)": 70.5, "step": 70995, "token_acc": 0.8833333333333333, "train_speed(iter/s)": 1.449893 }, { "epoch": 3.041857675335247, "grad_norm": 2.063727378845215, "learning_rate": 3.33040006362489e-05, "loss": 0.3366199493408203, "memory(GiB)": 70.5, "step": 71000, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.449899 }, { "epoch": 3.041857675335247, "eval_loss": 2.4458425045013428, "eval_runtime": 11.9615, "eval_samples_per_second": 8.36, "eval_steps_per_second": 8.36, "eval_token_acc": 0.4567901234567901, "step": 71000 }, { "epoch": 3.042071890664496, "grad_norm": 3.54724383354187, "learning_rate": 3.329765729385169e-05, "loss": 0.3204677104949951, "memory(GiB)": 70.5, "step": 71005, "token_acc": 0.5868263473053892, "train_speed(iter/s)": 1.449515 }, { "epoch": 3.0422861059937447, "grad_norm": 2.2553482055664062, "learning_rate": 3.329131425403309e-05, "loss": 0.185563063621521, "memory(GiB)": 70.5, "step": 71010, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.449524 }, { "epoch": 3.042500321322994, "grad_norm": 3.4661831855773926, "learning_rate": 3.3284971516908014e-05, "loss": 0.3684095859527588, "memory(GiB)": 70.5, "step": 71015, "token_acc": 0.9235474006116208, "train_speed(iter/s)": 1.449523 }, { "epoch": 3.042714536652243, "grad_norm": 4.770281791687012, "learning_rate": 3.327862908259135e-05, "loss": 0.6370392322540284, "memory(GiB)": 70.5, "step": 71020, "token_acc": 0.8851851851851852, "train_speed(iter/s)": 1.44952 }, { "epoch": 3.0429287519814916, "grad_norm": 2.5096395015716553, "learning_rate": 3.327228695119801e-05, "loss": 0.23633501529693604, "memory(GiB)": 70.5, "step": 71025, "token_acc": 0.9572649572649573, "train_speed(iter/s)": 1.449523 }, { "epoch": 3.043142967310741, "grad_norm": 0.2219717800617218, "learning_rate": 3.326594512284289e-05, "loss": 0.21105589866638183, "memory(GiB)": 70.5, "step": 71030, "token_acc": 0.9559748427672956, "train_speed(iter/s)": 1.449527 }, { "epoch": 3.0433571826399897, "grad_norm": 2.947420835494995, "learning_rate": 3.325960359764085e-05, "loss": 0.2838409900665283, "memory(GiB)": 70.5, "step": 71035, "token_acc": 0.936026936026936, "train_speed(iter/s)": 1.449528 }, { "epoch": 3.0435713979692385, "grad_norm": 2.537553548812866, "learning_rate": 3.3253262375706815e-05, "loss": 0.36295907497406005, "memory(GiB)": 70.5, "step": 71040, "token_acc": 0.933993399339934, "train_speed(iter/s)": 1.449534 }, { "epoch": 3.0437856132984877, "grad_norm": 4.1117072105407715, "learning_rate": 3.3246921457155634e-05, "loss": 0.34757318496704104, "memory(GiB)": 70.5, "step": 71045, "token_acc": 0.9451612903225807, "train_speed(iter/s)": 1.449548 }, { "epoch": 3.0439998286277365, "grad_norm": 4.6851983070373535, "learning_rate": 3.3240580842102186e-05, "loss": 0.17811832427978516, "memory(GiB)": 70.5, "step": 71050, "token_acc": 0.9671532846715328, "train_speed(iter/s)": 1.449549 }, { "epoch": 3.0442140439569854, "grad_norm": 4.809726238250732, "learning_rate": 3.323424053066135e-05, "loss": 0.620262336730957, "memory(GiB)": 70.5, "step": 71055, "token_acc": 0.8993288590604027, "train_speed(iter/s)": 1.449548 }, { "epoch": 3.0444282592862346, "grad_norm": 5.374495029449463, "learning_rate": 3.322790052294796e-05, "loss": 0.5254961013793945, "memory(GiB)": 70.5, "step": 71060, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.449545 }, { "epoch": 3.0446424746154834, "grad_norm": 3.790814161300659, "learning_rate": 3.3221560819076894e-05, "loss": 0.3597207546234131, "memory(GiB)": 70.5, "step": 71065, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.449549 }, { "epoch": 3.0448566899447322, "grad_norm": 0.6888819932937622, "learning_rate": 3.3215221419162966e-05, "loss": 0.43225250244140623, "memory(GiB)": 70.5, "step": 71070, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.449553 }, { "epoch": 3.0450709052739815, "grad_norm": 3.1746346950531006, "learning_rate": 3.320888232332108e-05, "loss": 0.3226178169250488, "memory(GiB)": 70.5, "step": 71075, "token_acc": 0.9386503067484663, "train_speed(iter/s)": 1.44955 }, { "epoch": 3.0452851206032303, "grad_norm": 2.7813291549682617, "learning_rate": 3.320254353166602e-05, "loss": 0.33702027797698975, "memory(GiB)": 70.5, "step": 71080, "token_acc": 0.9197707736389685, "train_speed(iter/s)": 1.449558 }, { "epoch": 3.045499335932479, "grad_norm": 2.8282783031463623, "learning_rate": 3.3196205044312664e-05, "loss": 0.45302481651306153, "memory(GiB)": 70.5, "step": 71085, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.44956 }, { "epoch": 3.0457135512617284, "grad_norm": 1.9256523847579956, "learning_rate": 3.318986686137581e-05, "loss": 0.27975120544433596, "memory(GiB)": 70.5, "step": 71090, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.449566 }, { "epoch": 3.045927766590977, "grad_norm": 1.914352536201477, "learning_rate": 3.318352898297027e-05, "loss": 0.5086607456207275, "memory(GiB)": 70.5, "step": 71095, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.449572 }, { "epoch": 3.046141981920226, "grad_norm": 4.782687664031982, "learning_rate": 3.31771914092109e-05, "loss": 0.4219198703765869, "memory(GiB)": 70.5, "step": 71100, "token_acc": 0.8901098901098901, "train_speed(iter/s)": 1.449576 }, { "epoch": 3.0463561972494753, "grad_norm": 4.1030192375183105, "learning_rate": 3.317085414021247e-05, "loss": 0.3704395771026611, "memory(GiB)": 70.5, "step": 71105, "token_acc": 0.9262820512820513, "train_speed(iter/s)": 1.44958 }, { "epoch": 3.046570412578724, "grad_norm": 1.1666128635406494, "learning_rate": 3.316451717608982e-05, "loss": 0.2829681634902954, "memory(GiB)": 70.5, "step": 71110, "token_acc": 0.9349593495934959, "train_speed(iter/s)": 1.44958 }, { "epoch": 3.046784627907973, "grad_norm": 5.25005578994751, "learning_rate": 3.3158180516957736e-05, "loss": 0.20232744216918946, "memory(GiB)": 70.5, "step": 71115, "token_acc": 0.9465648854961832, "train_speed(iter/s)": 1.4496 }, { "epoch": 3.046998843237222, "grad_norm": 3.730238914489746, "learning_rate": 3.315184416293099e-05, "loss": 0.24775917530059816, "memory(GiB)": 70.5, "step": 71120, "token_acc": 0.9485530546623794, "train_speed(iter/s)": 1.449601 }, { "epoch": 3.047213058566471, "grad_norm": 1.071425437927246, "learning_rate": 3.314550811412441e-05, "loss": 0.32556777000427245, "memory(GiB)": 70.5, "step": 71125, "token_acc": 0.9151515151515152, "train_speed(iter/s)": 1.449609 }, { "epoch": 3.0474272738957198, "grad_norm": 1.201185703277588, "learning_rate": 3.313917237065277e-05, "loss": 0.2115487813949585, "memory(GiB)": 70.5, "step": 71130, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.44961 }, { "epoch": 3.047641489224969, "grad_norm": 2.702467679977417, "learning_rate": 3.313283693263082e-05, "loss": 0.16440892219543457, "memory(GiB)": 70.5, "step": 71135, "token_acc": 0.9652777777777778, "train_speed(iter/s)": 1.449614 }, { "epoch": 3.047855704554218, "grad_norm": 4.960114479064941, "learning_rate": 3.3126501800173334e-05, "loss": 0.3541442394256592, "memory(GiB)": 70.5, "step": 71140, "token_acc": 0.9256965944272446, "train_speed(iter/s)": 1.44962 }, { "epoch": 3.0480699198834666, "grad_norm": 2.122457981109619, "learning_rate": 3.312016697339513e-05, "loss": 0.2218721866607666, "memory(GiB)": 70.5, "step": 71145, "token_acc": 0.9453376205787781, "train_speed(iter/s)": 1.449646 }, { "epoch": 3.048284135212716, "grad_norm": 4.374231338500977, "learning_rate": 3.311383245241092e-05, "loss": 0.39212450981140134, "memory(GiB)": 70.5, "step": 71150, "token_acc": 0.9221789883268483, "train_speed(iter/s)": 1.44965 }, { "epoch": 3.0484983505419647, "grad_norm": 5.196154594421387, "learning_rate": 3.3107498237335476e-05, "loss": 0.543922233581543, "memory(GiB)": 70.5, "step": 71155, "token_acc": 0.8903654485049833, "train_speed(iter/s)": 1.44966 }, { "epoch": 3.0487125658712135, "grad_norm": 1.5131430625915527, "learning_rate": 3.3101164328283566e-05, "loss": 0.5778104782104492, "memory(GiB)": 70.5, "step": 71160, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.449662 }, { "epoch": 3.048926781200463, "grad_norm": 7.821019649505615, "learning_rate": 3.3094830725369894e-05, "loss": 0.37799415588378904, "memory(GiB)": 70.5, "step": 71165, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.449665 }, { "epoch": 3.0491409965297116, "grad_norm": 3.057478427886963, "learning_rate": 3.308849742870923e-05, "loss": 0.29924860000610354, "memory(GiB)": 70.5, "step": 71170, "token_acc": 0.9287833827893175, "train_speed(iter/s)": 1.449667 }, { "epoch": 3.0493552118589604, "grad_norm": 2.311943769454956, "learning_rate": 3.3082164438416306e-05, "loss": 0.38976640701293946, "memory(GiB)": 70.5, "step": 71175, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.449671 }, { "epoch": 3.0495694271882097, "grad_norm": 3.9380180835723877, "learning_rate": 3.3075831754605834e-05, "loss": 0.3187044620513916, "memory(GiB)": 70.5, "step": 71180, "token_acc": 0.9353846153846154, "train_speed(iter/s)": 1.449682 }, { "epoch": 3.0497836425174585, "grad_norm": 6.845571041107178, "learning_rate": 3.306949937739255e-05, "loss": 0.34322304725646974, "memory(GiB)": 70.5, "step": 71185, "token_acc": 0.9418604651162791, "train_speed(iter/s)": 1.449689 }, { "epoch": 3.0499978578467073, "grad_norm": 3.880753993988037, "learning_rate": 3.3063167306891165e-05, "loss": 0.6742908477783203, "memory(GiB)": 70.5, "step": 71190, "token_acc": 0.875, "train_speed(iter/s)": 1.449689 }, { "epoch": 3.0502120731759566, "grad_norm": 3.2876594066619873, "learning_rate": 3.30568355432164e-05, "loss": 0.19115068912506103, "memory(GiB)": 70.5, "step": 71195, "token_acc": 0.9461538461538461, "train_speed(iter/s)": 1.449692 }, { "epoch": 3.0504262885052054, "grad_norm": 3.2319564819335938, "learning_rate": 3.305050408648295e-05, "loss": 0.32229089736938477, "memory(GiB)": 70.5, "step": 71200, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.449694 }, { "epoch": 3.050640503834454, "grad_norm": 4.296942710876465, "learning_rate": 3.3044172936805505e-05, "loss": 0.27929935455322263, "memory(GiB)": 70.5, "step": 71205, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.449702 }, { "epoch": 3.0508547191637034, "grad_norm": 4.146852493286133, "learning_rate": 3.3037842094298795e-05, "loss": 0.45490121841430664, "memory(GiB)": 70.5, "step": 71210, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.449717 }, { "epoch": 3.0510689344929522, "grad_norm": 2.2843220233917236, "learning_rate": 3.303151155907746e-05, "loss": 0.1326514959335327, "memory(GiB)": 70.5, "step": 71215, "token_acc": 0.9721254355400697, "train_speed(iter/s)": 1.449712 }, { "epoch": 3.0512831498222015, "grad_norm": 1.3923814296722412, "learning_rate": 3.302518133125623e-05, "loss": 0.4515354156494141, "memory(GiB)": 70.5, "step": 71220, "token_acc": 0.8906752411575563, "train_speed(iter/s)": 1.449713 }, { "epoch": 3.0514973651514503, "grad_norm": 7.270052909851074, "learning_rate": 3.301885141094975e-05, "loss": 0.3631463527679443, "memory(GiB)": 70.5, "step": 71225, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.449714 }, { "epoch": 3.051711580480699, "grad_norm": 2.0523383617401123, "learning_rate": 3.3012521798272723e-05, "loss": 0.330565881729126, "memory(GiB)": 70.5, "step": 71230, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.449719 }, { "epoch": 3.0519257958099484, "grad_norm": 3.0091116428375244, "learning_rate": 3.3006192493339785e-05, "loss": 0.47198824882507323, "memory(GiB)": 70.5, "step": 71235, "token_acc": 0.9087136929460581, "train_speed(iter/s)": 1.449727 }, { "epoch": 3.052140011139197, "grad_norm": 0.03631063923239708, "learning_rate": 3.299986349626563e-05, "loss": 0.5203381061553956, "memory(GiB)": 70.5, "step": 71240, "token_acc": 0.9085173501577287, "train_speed(iter/s)": 1.449748 }, { "epoch": 3.052354226468446, "grad_norm": 6.7350873947143555, "learning_rate": 3.299353480716488e-05, "loss": 0.39736261367797854, "memory(GiB)": 70.5, "step": 71245, "token_acc": 0.910394265232975, "train_speed(iter/s)": 1.449751 }, { "epoch": 3.0525684417976953, "grad_norm": 3.5137014389038086, "learning_rate": 3.298720642615221e-05, "loss": 0.46876039505004885, "memory(GiB)": 70.5, "step": 71250, "token_acc": 0.875, "train_speed(iter/s)": 1.449756 }, { "epoch": 3.052782657126944, "grad_norm": 3.8344597816467285, "learning_rate": 3.298087835334225e-05, "loss": 0.34276509284973145, "memory(GiB)": 70.5, "step": 71255, "token_acc": 0.9212328767123288, "train_speed(iter/s)": 1.449758 }, { "epoch": 3.052996872456193, "grad_norm": 5.781467914581299, "learning_rate": 3.2974550588849645e-05, "loss": 0.39135658740997314, "memory(GiB)": 70.5, "step": 71260, "token_acc": 0.894927536231884, "train_speed(iter/s)": 1.449772 }, { "epoch": 3.053211087785442, "grad_norm": 3.283494234085083, "learning_rate": 3.296822313278902e-05, "loss": 0.41733899116516116, "memory(GiB)": 70.5, "step": 71265, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.449785 }, { "epoch": 3.053425303114691, "grad_norm": 0.9128319025039673, "learning_rate": 3.296189598527502e-05, "loss": 0.23237137794494628, "memory(GiB)": 70.5, "step": 71270, "token_acc": 0.9621848739495799, "train_speed(iter/s)": 1.449785 }, { "epoch": 3.0536395184439398, "grad_norm": 3.100032329559326, "learning_rate": 3.295556914642226e-05, "loss": 0.27533907890319825, "memory(GiB)": 70.5, "step": 71275, "token_acc": 0.9408284023668639, "train_speed(iter/s)": 1.449798 }, { "epoch": 3.053853733773189, "grad_norm": 7.214890956878662, "learning_rate": 3.294924261634534e-05, "loss": 0.5095433235168457, "memory(GiB)": 70.5, "step": 71280, "token_acc": 0.9111747851002865, "train_speed(iter/s)": 1.449821 }, { "epoch": 3.054067949102438, "grad_norm": 3.138288974761963, "learning_rate": 3.29429163951589e-05, "loss": 0.36711506843566893, "memory(GiB)": 70.5, "step": 71285, "token_acc": 0.9172932330827067, "train_speed(iter/s)": 1.449821 }, { "epoch": 3.0542821644316867, "grad_norm": 2.2557055950164795, "learning_rate": 3.29365904829775e-05, "loss": 0.34484434127807617, "memory(GiB)": 70.5, "step": 71290, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.449829 }, { "epoch": 3.054496379760936, "grad_norm": 2.384674549102783, "learning_rate": 3.293026487991579e-05, "loss": 0.44994163513183594, "memory(GiB)": 70.5, "step": 71295, "token_acc": 0.9159159159159159, "train_speed(iter/s)": 1.449841 }, { "epoch": 3.0547105950901847, "grad_norm": 7.937963485717773, "learning_rate": 3.2923939586088346e-05, "loss": 0.4790966033935547, "memory(GiB)": 70.5, "step": 71300, "token_acc": 0.9104938271604939, "train_speed(iter/s)": 1.449845 }, { "epoch": 3.0549248104194335, "grad_norm": 4.306896209716797, "learning_rate": 3.291761460160976e-05, "loss": 0.3672076463699341, "memory(GiB)": 70.5, "step": 71305, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.449847 }, { "epoch": 3.055139025748683, "grad_norm": 3.0139119625091553, "learning_rate": 3.2911289926594604e-05, "loss": 0.3967352151870728, "memory(GiB)": 70.5, "step": 71310, "token_acc": 0.9082969432314411, "train_speed(iter/s)": 1.449845 }, { "epoch": 3.0553532410779316, "grad_norm": 3.2674484252929688, "learning_rate": 3.2904965561157466e-05, "loss": 0.46221446990966797, "memory(GiB)": 70.5, "step": 71315, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.449849 }, { "epoch": 3.0555674564071804, "grad_norm": 3.675490379333496, "learning_rate": 3.289864150541292e-05, "loss": 0.5294981956481933, "memory(GiB)": 70.5, "step": 71320, "token_acc": 0.8858858858858859, "train_speed(iter/s)": 1.44987 }, { "epoch": 3.0557816717364297, "grad_norm": 5.12636661529541, "learning_rate": 3.289231775947551e-05, "loss": 0.746318244934082, "memory(GiB)": 70.5, "step": 71325, "token_acc": 0.8543956043956044, "train_speed(iter/s)": 1.449879 }, { "epoch": 3.0559958870656785, "grad_norm": 0.5827620625495911, "learning_rate": 3.2885994323459826e-05, "loss": 0.12481495141983032, "memory(GiB)": 70.5, "step": 71330, "token_acc": 0.9681528662420382, "train_speed(iter/s)": 1.449882 }, { "epoch": 3.0562101023949273, "grad_norm": 3.755095958709717, "learning_rate": 3.287967119748039e-05, "loss": 0.27704801559448244, "memory(GiB)": 70.5, "step": 71335, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.449886 }, { "epoch": 3.0564243177241766, "grad_norm": 2.0654103755950928, "learning_rate": 3.287334838165179e-05, "loss": 0.20120830535888673, "memory(GiB)": 70.5, "step": 71340, "token_acc": 0.9671641791044776, "train_speed(iter/s)": 1.449897 }, { "epoch": 3.0566385330534254, "grad_norm": 0.770018458366394, "learning_rate": 3.286702587608854e-05, "loss": 0.2631214141845703, "memory(GiB)": 70.5, "step": 71345, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.449901 }, { "epoch": 3.056852748382674, "grad_norm": 3.9539051055908203, "learning_rate": 3.286070368090519e-05, "loss": 0.29162211418151857, "memory(GiB)": 70.5, "step": 71350, "token_acc": 0.9405204460966543, "train_speed(iter/s)": 1.449898 }, { "epoch": 3.0570669637119234, "grad_norm": 5.211418628692627, "learning_rate": 3.285438179621627e-05, "loss": 0.6379844665527343, "memory(GiB)": 70.5, "step": 71355, "token_acc": 0.8619402985074627, "train_speed(iter/s)": 1.4499 }, { "epoch": 3.0572811790411722, "grad_norm": 12.723352432250977, "learning_rate": 3.284806022213632e-05, "loss": 0.390471887588501, "memory(GiB)": 70.5, "step": 71360, "token_acc": 0.9235294117647059, "train_speed(iter/s)": 1.449904 }, { "epoch": 3.057495394370421, "grad_norm": 1.516343593597412, "learning_rate": 3.284173895877982e-05, "loss": 0.3071139335632324, "memory(GiB)": 70.5, "step": 71365, "token_acc": 0.9386973180076629, "train_speed(iter/s)": 1.449904 }, { "epoch": 3.0577096096996703, "grad_norm": 1.0808955430984497, "learning_rate": 3.283541800626132e-05, "loss": 0.20356481075286864, "memory(GiB)": 70.5, "step": 71370, "token_acc": 0.9603960396039604, "train_speed(iter/s)": 1.449903 }, { "epoch": 3.057923825028919, "grad_norm": 2.8421528339385986, "learning_rate": 3.282909736469535e-05, "loss": 0.2806876182556152, "memory(GiB)": 70.5, "step": 71375, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.449906 }, { "epoch": 3.058138040358168, "grad_norm": 1.8477500677108765, "learning_rate": 3.2822777034196364e-05, "loss": 0.1973728895187378, "memory(GiB)": 70.5, "step": 71380, "token_acc": 0.9457627118644067, "train_speed(iter/s)": 1.449916 }, { "epoch": 3.058352255687417, "grad_norm": 1.8536553382873535, "learning_rate": 3.28164570148789e-05, "loss": 0.287847638130188, "memory(GiB)": 70.5, "step": 71385, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.449919 }, { "epoch": 3.058566471016666, "grad_norm": 7.3657941818237305, "learning_rate": 3.281013730685744e-05, "loss": 0.5342071056365967, "memory(GiB)": 70.5, "step": 71390, "token_acc": 0.8980891719745223, "train_speed(iter/s)": 1.449911 }, { "epoch": 3.058780686345915, "grad_norm": 6.727114200592041, "learning_rate": 3.280381791024646e-05, "loss": 0.23459417819976808, "memory(GiB)": 70.5, "step": 71395, "token_acc": 0.9609120521172638, "train_speed(iter/s)": 1.449915 }, { "epoch": 3.058994901675164, "grad_norm": 4.156139850616455, "learning_rate": 3.2797498825160456e-05, "loss": 0.6153350830078125, "memory(GiB)": 70.5, "step": 71400, "token_acc": 0.8565022421524664, "train_speed(iter/s)": 1.449917 }, { "epoch": 3.059209117004413, "grad_norm": 2.9127514362335205, "learning_rate": 3.279118005171389e-05, "loss": 0.4059562683105469, "memory(GiB)": 70.5, "step": 71405, "token_acc": 0.9110320284697508, "train_speed(iter/s)": 1.44992 }, { "epoch": 3.0594233323336617, "grad_norm": 2.2592506408691406, "learning_rate": 3.278486159002124e-05, "loss": 0.27956862449645997, "memory(GiB)": 70.5, "step": 71410, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.449924 }, { "epoch": 3.059637547662911, "grad_norm": 4.448282718658447, "learning_rate": 3.277854344019699e-05, "loss": 0.3196086883544922, "memory(GiB)": 70.5, "step": 71415, "token_acc": 0.9263565891472868, "train_speed(iter/s)": 1.449925 }, { "epoch": 3.0598517629921598, "grad_norm": 2.6345696449279785, "learning_rate": 3.2772225602355554e-05, "loss": 0.30381264686584475, "memory(GiB)": 70.5, "step": 71420, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.449949 }, { "epoch": 3.0600659783214086, "grad_norm": 1.3202028274536133, "learning_rate": 3.2765908076611426e-05, "loss": 0.2282790422439575, "memory(GiB)": 70.5, "step": 71425, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.449941 }, { "epoch": 3.060280193650658, "grad_norm": 5.649299144744873, "learning_rate": 3.275959086307905e-05, "loss": 0.42544031143188477, "memory(GiB)": 70.5, "step": 71430, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.44996 }, { "epoch": 3.0604944089799067, "grad_norm": 2.017620325088501, "learning_rate": 3.2753273961872824e-05, "loss": 0.5067818641662598, "memory(GiB)": 70.5, "step": 71435, "token_acc": 0.8969072164948454, "train_speed(iter/s)": 1.449966 }, { "epoch": 3.0607086243091555, "grad_norm": 2.6842808723449707, "learning_rate": 3.274695737310723e-05, "loss": 0.5569127559661865, "memory(GiB)": 70.5, "step": 71440, "token_acc": 0.8877887788778878, "train_speed(iter/s)": 1.449964 }, { "epoch": 3.0609228396384047, "grad_norm": 4.735574722290039, "learning_rate": 3.2740641096896715e-05, "loss": 0.23121745586395265, "memory(GiB)": 70.5, "step": 71445, "token_acc": 0.9619377162629758, "train_speed(iter/s)": 1.449965 }, { "epoch": 3.0611370549676535, "grad_norm": 1.0332005023956299, "learning_rate": 3.273432513335566e-05, "loss": 0.22225663661956788, "memory(GiB)": 70.5, "step": 71450, "token_acc": 0.9386973180076629, "train_speed(iter/s)": 1.449969 }, { "epoch": 3.0613512702969023, "grad_norm": 2.2108871936798096, "learning_rate": 3.27280094825985e-05, "loss": 0.3072463274002075, "memory(GiB)": 70.5, "step": 71455, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.449969 }, { "epoch": 3.0615654856261516, "grad_norm": 2.7512576580047607, "learning_rate": 3.2721694144739654e-05, "loss": 0.12235260009765625, "memory(GiB)": 70.5, "step": 71460, "token_acc": 0.9777158774373259, "train_speed(iter/s)": 1.449981 }, { "epoch": 3.0617797009554004, "grad_norm": 4.835775852203369, "learning_rate": 3.2715379119893534e-05, "loss": 0.7025155067443848, "memory(GiB)": 70.5, "step": 71465, "token_acc": 0.8554216867469879, "train_speed(iter/s)": 1.449991 }, { "epoch": 3.0619939162846492, "grad_norm": 1.9899848699569702, "learning_rate": 3.270906440817453e-05, "loss": 0.31758224964141846, "memory(GiB)": 70.5, "step": 71470, "token_acc": 0.94140625, "train_speed(iter/s)": 1.449995 }, { "epoch": 3.0622081316138985, "grad_norm": 0.17913982272148132, "learning_rate": 3.270275000969704e-05, "loss": 0.2982417821884155, "memory(GiB)": 70.5, "step": 71475, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.450002 }, { "epoch": 3.0624223469431473, "grad_norm": 1.3535594940185547, "learning_rate": 3.2696435924575456e-05, "loss": 0.516306734085083, "memory(GiB)": 70.5, "step": 71480, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.450016 }, { "epoch": 3.062636562272396, "grad_norm": 3.614039421081543, "learning_rate": 3.269012215292418e-05, "loss": 0.4416768550872803, "memory(GiB)": 70.5, "step": 71485, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.450028 }, { "epoch": 3.0628507776016454, "grad_norm": 4.621480464935303, "learning_rate": 3.268380869485758e-05, "loss": 0.47977194786071775, "memory(GiB)": 70.5, "step": 71490, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.45003 }, { "epoch": 3.063064992930894, "grad_norm": 2.3557050228118896, "learning_rate": 3.267749555049002e-05, "loss": 0.31049749851226804, "memory(GiB)": 70.5, "step": 71495, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.450028 }, { "epoch": 3.063279208260143, "grad_norm": 4.571683883666992, "learning_rate": 3.2671182719935874e-05, "loss": 0.18052160739898682, "memory(GiB)": 70.5, "step": 71500, "token_acc": 0.9622641509433962, "train_speed(iter/s)": 1.450027 }, { "epoch": 3.063279208260143, "eval_loss": 2.595233678817749, "eval_runtime": 12.0939, "eval_samples_per_second": 8.269, "eval_steps_per_second": 8.269, "eval_token_acc": 0.43529411764705883, "step": 71500 }, { "epoch": 3.0634934235893923, "grad_norm": 2.2178304195404053, "learning_rate": 3.266487020330953e-05, "loss": 0.16081924438476564, "memory(GiB)": 70.5, "step": 71505, "token_acc": 0.5857544517338332, "train_speed(iter/s)": 1.449651 }, { "epoch": 3.063707638918641, "grad_norm": 3.672724485397339, "learning_rate": 3.2658558000725295e-05, "loss": 0.22813220024108888, "memory(GiB)": 70.5, "step": 71510, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.449665 }, { "epoch": 3.06392185424789, "grad_norm": 3.358431577682495, "learning_rate": 3.265224611229755e-05, "loss": 0.25980849266052247, "memory(GiB)": 70.5, "step": 71515, "token_acc": 0.9537953795379538, "train_speed(iter/s)": 1.449672 }, { "epoch": 3.064136069577139, "grad_norm": 2.0913050174713135, "learning_rate": 3.264593453814066e-05, "loss": 0.30590894222259524, "memory(GiB)": 70.5, "step": 71520, "token_acc": 0.9372937293729373, "train_speed(iter/s)": 1.449677 }, { "epoch": 3.064350284906388, "grad_norm": 3.054814338684082, "learning_rate": 3.2639623278368936e-05, "loss": 0.3315309524536133, "memory(GiB)": 70.5, "step": 71525, "token_acc": 0.9128787878787878, "train_speed(iter/s)": 1.449679 }, { "epoch": 3.0645645002356368, "grad_norm": 2.862893581390381, "learning_rate": 3.263331233309673e-05, "loss": 0.2611210584640503, "memory(GiB)": 70.5, "step": 71530, "token_acc": 0.9455128205128205, "train_speed(iter/s)": 1.449673 }, { "epoch": 3.064778715564886, "grad_norm": 3.3026280403137207, "learning_rate": 3.262700170243836e-05, "loss": 0.1805781126022339, "memory(GiB)": 70.5, "step": 71535, "token_acc": 0.9504643962848297, "train_speed(iter/s)": 1.449681 }, { "epoch": 3.064992930894135, "grad_norm": 4.471858501434326, "learning_rate": 3.262069138650814e-05, "loss": 0.4550000190734863, "memory(GiB)": 70.5, "step": 71540, "token_acc": 0.8935361216730038, "train_speed(iter/s)": 1.449694 }, { "epoch": 3.0652071462233836, "grad_norm": 2.9769673347473145, "learning_rate": 3.261438138542041e-05, "loss": 0.20692553520202636, "memory(GiB)": 70.5, "step": 71545, "token_acc": 0.9413680781758957, "train_speed(iter/s)": 1.449696 }, { "epoch": 3.065421361552633, "grad_norm": 3.5394227504730225, "learning_rate": 3.260807169928948e-05, "loss": 0.42508735656738283, "memory(GiB)": 70.5, "step": 71550, "token_acc": 0.9161676646706587, "train_speed(iter/s)": 1.4497 }, { "epoch": 3.0656355768818817, "grad_norm": 2.3034543991088867, "learning_rate": 3.2601762328229624e-05, "loss": 0.756819486618042, "memory(GiB)": 70.5, "step": 71555, "token_acc": 0.8549382716049383, "train_speed(iter/s)": 1.449706 }, { "epoch": 3.0658497922111305, "grad_norm": 0.5475966334342957, "learning_rate": 3.259545327235518e-05, "loss": 0.4572866916656494, "memory(GiB)": 70.5, "step": 71560, "token_acc": 0.900990099009901, "train_speed(iter/s)": 1.449709 }, { "epoch": 3.0660640075403798, "grad_norm": 4.880006790161133, "learning_rate": 3.258914453178041e-05, "loss": 0.21363575458526612, "memory(GiB)": 70.5, "step": 71565, "token_acc": 0.9546827794561934, "train_speed(iter/s)": 1.449714 }, { "epoch": 3.0662782228696286, "grad_norm": 2.6907732486724854, "learning_rate": 3.2582836106619636e-05, "loss": 0.25587828159332277, "memory(GiB)": 70.5, "step": 71570, "token_acc": 0.9313725490196079, "train_speed(iter/s)": 1.449715 }, { "epoch": 3.0664924381988774, "grad_norm": 4.814002513885498, "learning_rate": 3.257652799698712e-05, "loss": 0.249724817276001, "memory(GiB)": 70.5, "step": 71575, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.44972 }, { "epoch": 3.0667066535281267, "grad_norm": 3.583427667617798, "learning_rate": 3.2570220202997134e-05, "loss": 0.40204548835754395, "memory(GiB)": 70.5, "step": 71580, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.44972 }, { "epoch": 3.0669208688573755, "grad_norm": 4.783653736114502, "learning_rate": 3.256391272476396e-05, "loss": 0.301775860786438, "memory(GiB)": 70.5, "step": 71585, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.449724 }, { "epoch": 3.0671350841866243, "grad_norm": 1.3294304609298706, "learning_rate": 3.255760556240187e-05, "loss": 0.3000039100646973, "memory(GiB)": 70.5, "step": 71590, "token_acc": 0.9252873563218391, "train_speed(iter/s)": 1.449739 }, { "epoch": 3.0673492995158735, "grad_norm": 2.3158745765686035, "learning_rate": 3.255129871602512e-05, "loss": 0.19140721559524537, "memory(GiB)": 70.5, "step": 71595, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.44975 }, { "epoch": 3.0675635148451224, "grad_norm": 1.980510950088501, "learning_rate": 3.2544992185747955e-05, "loss": 0.1562623381614685, "memory(GiB)": 70.5, "step": 71600, "token_acc": 0.9607843137254902, "train_speed(iter/s)": 1.449756 }, { "epoch": 3.067777730174371, "grad_norm": 3.4732329845428467, "learning_rate": 3.2538685971684646e-05, "loss": 0.45214138031005857, "memory(GiB)": 70.5, "step": 71605, "token_acc": 0.9034749034749034, "train_speed(iter/s)": 1.449757 }, { "epoch": 3.0679919455036204, "grad_norm": 2.997429609298706, "learning_rate": 3.253238007394942e-05, "loss": 0.4747790813446045, "memory(GiB)": 70.5, "step": 71610, "token_acc": 0.8962264150943396, "train_speed(iter/s)": 1.449753 }, { "epoch": 3.0682061608328692, "grad_norm": 4.493565559387207, "learning_rate": 3.2526074492656496e-05, "loss": 0.3142727851867676, "memory(GiB)": 70.5, "step": 71615, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.449752 }, { "epoch": 3.068420376162118, "grad_norm": 4.044743537902832, "learning_rate": 3.2519769227920135e-05, "loss": 0.30366108417510984, "memory(GiB)": 70.5, "step": 71620, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.44975 }, { "epoch": 3.0686345914913673, "grad_norm": 3.360025644302368, "learning_rate": 3.251346427985455e-05, "loss": 0.4183413028717041, "memory(GiB)": 70.5, "step": 71625, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.449751 }, { "epoch": 3.068848806820616, "grad_norm": 1.1940865516662598, "learning_rate": 3.250715964857397e-05, "loss": 0.2805870532989502, "memory(GiB)": 70.5, "step": 71630, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.449753 }, { "epoch": 3.069063022149865, "grad_norm": 3.1786396503448486, "learning_rate": 3.25008553341926e-05, "loss": 0.26912527084350585, "memory(GiB)": 70.5, "step": 71635, "token_acc": 0.941358024691358, "train_speed(iter/s)": 1.449749 }, { "epoch": 3.069277237479114, "grad_norm": 4.8220672607421875, "learning_rate": 3.249455133682464e-05, "loss": 0.41452903747558595, "memory(GiB)": 70.5, "step": 71640, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.449749 }, { "epoch": 3.069491452808363, "grad_norm": 2.7125394344329834, "learning_rate": 3.248824765658432e-05, "loss": 0.6431235313415528, "memory(GiB)": 70.5, "step": 71645, "token_acc": 0.8538205980066446, "train_speed(iter/s)": 1.449747 }, { "epoch": 3.069705668137612, "grad_norm": 3.89691162109375, "learning_rate": 3.248194429358582e-05, "loss": 0.44181203842163086, "memory(GiB)": 70.5, "step": 71650, "token_acc": 0.9014084507042254, "train_speed(iter/s)": 1.449762 }, { "epoch": 3.069919883466861, "grad_norm": 0.27784672379493713, "learning_rate": 3.247564124794332e-05, "loss": 0.5228743553161621, "memory(GiB)": 70.5, "step": 71655, "token_acc": 0.8921161825726142, "train_speed(iter/s)": 1.449765 }, { "epoch": 3.07013409879611, "grad_norm": 0.5197327136993408, "learning_rate": 3.246933851977101e-05, "loss": 0.19956870079040528, "memory(GiB)": 70.5, "step": 71660, "token_acc": 0.952247191011236, "train_speed(iter/s)": 1.449764 }, { "epoch": 3.0703483141253587, "grad_norm": 2.5122411251068115, "learning_rate": 3.246303610918309e-05, "loss": 0.5789984226226806, "memory(GiB)": 70.5, "step": 71665, "token_acc": 0.8838951310861424, "train_speed(iter/s)": 1.449775 }, { "epoch": 3.070562529454608, "grad_norm": 4.300606727600098, "learning_rate": 3.245673401629372e-05, "loss": 0.342710018157959, "memory(GiB)": 70.5, "step": 71670, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.449778 }, { "epoch": 3.0707767447838568, "grad_norm": 1.8902124166488647, "learning_rate": 3.245043224121708e-05, "loss": 0.3556164979934692, "memory(GiB)": 70.5, "step": 71675, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.449782 }, { "epoch": 3.0709909601131056, "grad_norm": 1.846895456314087, "learning_rate": 3.2444130784067315e-05, "loss": 0.46110901832580564, "memory(GiB)": 70.5, "step": 71680, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.449786 }, { "epoch": 3.071205175442355, "grad_norm": 1.4253637790679932, "learning_rate": 3.243782964495859e-05, "loss": 0.3890009164810181, "memory(GiB)": 70.5, "step": 71685, "token_acc": 0.9282868525896414, "train_speed(iter/s)": 1.449789 }, { "epoch": 3.0714193907716036, "grad_norm": 2.266624689102173, "learning_rate": 3.243152882400506e-05, "loss": 0.3536843776702881, "memory(GiB)": 70.5, "step": 71690, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.449788 }, { "epoch": 3.0716336061008525, "grad_norm": 2.273050546646118, "learning_rate": 3.2425228321320863e-05, "loss": 0.2550879240036011, "memory(GiB)": 70.5, "step": 71695, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.449794 }, { "epoch": 3.0718478214301017, "grad_norm": 3.8078176975250244, "learning_rate": 3.241892813702014e-05, "loss": 0.2130880117416382, "memory(GiB)": 70.5, "step": 71700, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.449805 }, { "epoch": 3.0720620367593505, "grad_norm": 5.280353546142578, "learning_rate": 3.2412628271217034e-05, "loss": 0.2653702735900879, "memory(GiB)": 70.5, "step": 71705, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.449809 }, { "epoch": 3.0722762520885993, "grad_norm": 3.409792900085449, "learning_rate": 3.240632872402565e-05, "loss": 0.306410551071167, "memory(GiB)": 70.5, "step": 71710, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.44981 }, { "epoch": 3.0724904674178486, "grad_norm": 0.8094643950462341, "learning_rate": 3.240002949556014e-05, "loss": 0.27801406383514404, "memory(GiB)": 70.5, "step": 71715, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.44981 }, { "epoch": 3.0727046827470974, "grad_norm": 4.051098346710205, "learning_rate": 3.23937305859346e-05, "loss": 0.3798035144805908, "memory(GiB)": 70.5, "step": 71720, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.449806 }, { "epoch": 3.072918898076346, "grad_norm": 1.8720428943634033, "learning_rate": 3.238743199526314e-05, "loss": 0.22402617931365967, "memory(GiB)": 70.5, "step": 71725, "token_acc": 0.9633333333333334, "train_speed(iter/s)": 1.44981 }, { "epoch": 3.0731331134055955, "grad_norm": 0.11345012485980988, "learning_rate": 3.2381133723659886e-05, "loss": 0.24989118576049804, "memory(GiB)": 70.5, "step": 71730, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.449815 }, { "epoch": 3.0733473287348443, "grad_norm": 4.191550254821777, "learning_rate": 3.23748357712389e-05, "loss": 0.6379828929901123, "memory(GiB)": 70.5, "step": 71735, "token_acc": 0.8778877887788779, "train_speed(iter/s)": 1.449813 }, { "epoch": 3.073561544064093, "grad_norm": 1.7271130084991455, "learning_rate": 3.236853813811431e-05, "loss": 0.4711876392364502, "memory(GiB)": 70.5, "step": 71740, "token_acc": 0.912621359223301, "train_speed(iter/s)": 1.449829 }, { "epoch": 3.0737757593933424, "grad_norm": 4.280358791351318, "learning_rate": 3.236224082440019e-05, "loss": 0.3762217044830322, "memory(GiB)": 70.5, "step": 71745, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.449833 }, { "epoch": 3.073989974722591, "grad_norm": 0.21755968034267426, "learning_rate": 3.235594383021063e-05, "loss": 0.2731450319290161, "memory(GiB)": 70.5, "step": 71750, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.449837 }, { "epoch": 3.07420419005184, "grad_norm": 2.5490200519561768, "learning_rate": 3.23496471556597e-05, "loss": 0.32067251205444336, "memory(GiB)": 70.5, "step": 71755, "token_acc": 0.9251968503937008, "train_speed(iter/s)": 1.44984 }, { "epoch": 3.0744184053810892, "grad_norm": 7.104940414428711, "learning_rate": 3.234335080086146e-05, "loss": 0.8888607025146484, "memory(GiB)": 70.5, "step": 71760, "token_acc": 0.82421875, "train_speed(iter/s)": 1.449844 }, { "epoch": 3.074632620710338, "grad_norm": 1.3933073282241821, "learning_rate": 3.233705476592999e-05, "loss": 0.4646648406982422, "memory(GiB)": 70.5, "step": 71765, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.449845 }, { "epoch": 3.074846836039587, "grad_norm": 1.1844826936721802, "learning_rate": 3.233075905097933e-05, "loss": 0.32055437564849854, "memory(GiB)": 70.5, "step": 71770, "token_acc": 0.9025270758122743, "train_speed(iter/s)": 1.449865 }, { "epoch": 3.075061051368836, "grad_norm": 3.167907238006592, "learning_rate": 3.232446365612356e-05, "loss": 0.3470921516418457, "memory(GiB)": 70.5, "step": 71775, "token_acc": 0.9237804878048781, "train_speed(iter/s)": 1.449877 }, { "epoch": 3.075275266698085, "grad_norm": 3.0191309452056885, "learning_rate": 3.231816858147672e-05, "loss": 0.4179695606231689, "memory(GiB)": 70.5, "step": 71780, "token_acc": 0.9148264984227129, "train_speed(iter/s)": 1.449875 }, { "epoch": 3.0754894820273337, "grad_norm": 1.2066431045532227, "learning_rate": 3.231187382715282e-05, "loss": 0.24595909118652343, "memory(GiB)": 70.5, "step": 71785, "token_acc": 0.946031746031746, "train_speed(iter/s)": 1.449882 }, { "epoch": 3.075703697356583, "grad_norm": 3.079944610595703, "learning_rate": 3.2305579393265926e-05, "loss": 0.32366418838500977, "memory(GiB)": 70.5, "step": 71790, "token_acc": 0.95, "train_speed(iter/s)": 1.449883 }, { "epoch": 3.075917912685832, "grad_norm": 1.2541950941085815, "learning_rate": 3.229928527993007e-05, "loss": 0.2605990171432495, "memory(GiB)": 70.5, "step": 71795, "token_acc": 0.9383116883116883, "train_speed(iter/s)": 1.449889 }, { "epoch": 3.0761321280150806, "grad_norm": 6.33047342300415, "learning_rate": 3.229299148725925e-05, "loss": 0.5041448593139648, "memory(GiB)": 70.5, "step": 71800, "token_acc": 0.8832684824902723, "train_speed(iter/s)": 1.449892 }, { "epoch": 3.07634634334433, "grad_norm": 2.3625428676605225, "learning_rate": 3.228669801536749e-05, "loss": 0.21516938209533693, "memory(GiB)": 70.5, "step": 71805, "token_acc": 0.9475806451612904, "train_speed(iter/s)": 1.449895 }, { "epoch": 3.0765605586735787, "grad_norm": 0.4461083710193634, "learning_rate": 3.228040486436884e-05, "loss": 0.31092162132263185, "memory(GiB)": 70.5, "step": 71810, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.449892 }, { "epoch": 3.0767747740028275, "grad_norm": 6.849164009094238, "learning_rate": 3.2274112034377255e-05, "loss": 0.43694357872009276, "memory(GiB)": 70.5, "step": 71815, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.449893 }, { "epoch": 3.0769889893320768, "grad_norm": 2.109119176864624, "learning_rate": 3.2267819525506784e-05, "loss": 0.2841574907302856, "memory(GiB)": 70.5, "step": 71820, "token_acc": 0.9432624113475178, "train_speed(iter/s)": 1.449893 }, { "epoch": 3.0772032046613256, "grad_norm": 3.094090461730957, "learning_rate": 3.226152733787138e-05, "loss": 0.3554091215133667, "memory(GiB)": 70.5, "step": 71825, "token_acc": 0.9134615384615384, "train_speed(iter/s)": 1.449899 }, { "epoch": 3.0774174199905744, "grad_norm": 0.517930269241333, "learning_rate": 3.225523547158505e-05, "loss": 0.16520652770996094, "memory(GiB)": 70.5, "step": 71830, "token_acc": 0.9627118644067797, "train_speed(iter/s)": 1.449915 }, { "epoch": 3.0776316353198236, "grad_norm": 3.6166117191314697, "learning_rate": 3.224894392676178e-05, "loss": 0.3408321142196655, "memory(GiB)": 70.5, "step": 71835, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.449918 }, { "epoch": 3.0778458506490725, "grad_norm": 0.035168617963790894, "learning_rate": 3.2242652703515545e-05, "loss": 0.36118245124816895, "memory(GiB)": 70.5, "step": 71840, "token_acc": 0.897887323943662, "train_speed(iter/s)": 1.449924 }, { "epoch": 3.0780600659783213, "grad_norm": 1.8731689453125, "learning_rate": 3.22363618019603e-05, "loss": 0.2976234436035156, "memory(GiB)": 70.5, "step": 71845, "token_acc": 0.9395973154362416, "train_speed(iter/s)": 1.449938 }, { "epoch": 3.0782742813075705, "grad_norm": 4.988794326782227, "learning_rate": 3.223007122221004e-05, "loss": 0.32525999546051027, "memory(GiB)": 70.5, "step": 71850, "token_acc": 0.948339483394834, "train_speed(iter/s)": 1.449946 }, { "epoch": 3.0784884966368193, "grad_norm": 4.739455223083496, "learning_rate": 3.22237809643787e-05, "loss": 0.3195915937423706, "memory(GiB)": 70.5, "step": 71855, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.449944 }, { "epoch": 3.078702711966068, "grad_norm": 4.020862102508545, "learning_rate": 3.221749102858025e-05, "loss": 0.440263557434082, "memory(GiB)": 70.5, "step": 71860, "token_acc": 0.8990825688073395, "train_speed(iter/s)": 1.449943 }, { "epoch": 3.0789169272953174, "grad_norm": 5.20256233215332, "learning_rate": 3.2211201414928634e-05, "loss": 0.5055536270141602, "memory(GiB)": 70.5, "step": 71865, "token_acc": 0.8973607038123167, "train_speed(iter/s)": 1.449949 }, { "epoch": 3.079131142624566, "grad_norm": 2.012552499771118, "learning_rate": 3.220491212353778e-05, "loss": 0.14717509746551513, "memory(GiB)": 70.5, "step": 71870, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.449958 }, { "epoch": 3.079345357953815, "grad_norm": 2.7107834815979004, "learning_rate": 3.219862315452164e-05, "loss": 0.30121028423309326, "memory(GiB)": 70.5, "step": 71875, "token_acc": 0.9442815249266863, "train_speed(iter/s)": 1.44996 }, { "epoch": 3.0795595732830643, "grad_norm": 3.572327136993408, "learning_rate": 3.2192334507994125e-05, "loss": 0.20666427612304689, "memory(GiB)": 70.5, "step": 71880, "token_acc": 0.946360153256705, "train_speed(iter/s)": 1.449975 }, { "epoch": 3.079773788612313, "grad_norm": 2.0885844230651855, "learning_rate": 3.2186046184069196e-05, "loss": 0.32459614276885984, "memory(GiB)": 70.5, "step": 71885, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.449987 }, { "epoch": 3.079988003941562, "grad_norm": 1.2684646844863892, "learning_rate": 3.2179758182860744e-05, "loss": 0.4099857807159424, "memory(GiB)": 70.5, "step": 71890, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.44999 }, { "epoch": 3.080202219270811, "grad_norm": 3.5641376972198486, "learning_rate": 3.21734705044827e-05, "loss": 0.43624420166015626, "memory(GiB)": 70.5, "step": 71895, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.450004 }, { "epoch": 3.08041643460006, "grad_norm": 6.388017177581787, "learning_rate": 3.216718314904895e-05, "loss": 0.6749660968780518, "memory(GiB)": 70.5, "step": 71900, "token_acc": 0.8774834437086093, "train_speed(iter/s)": 1.450017 }, { "epoch": 3.080630649929309, "grad_norm": 1.8492563962936401, "learning_rate": 3.216089611667341e-05, "loss": 0.47862658500671384, "memory(GiB)": 70.5, "step": 71905, "token_acc": 0.8825622775800712, "train_speed(iter/s)": 1.450033 }, { "epoch": 3.080844865258558, "grad_norm": 0.9743871688842773, "learning_rate": 3.215460940746997e-05, "loss": 0.36658611297607424, "memory(GiB)": 70.5, "step": 71910, "token_acc": 0.909967845659164, "train_speed(iter/s)": 1.450044 }, { "epoch": 3.081059080587807, "grad_norm": 0.3357701003551483, "learning_rate": 3.214832302155252e-05, "loss": 0.09709672331809997, "memory(GiB)": 70.5, "step": 71915, "token_acc": 0.9734848484848485, "train_speed(iter/s)": 1.450043 }, { "epoch": 3.0812732959170557, "grad_norm": 0.6599858403205872, "learning_rate": 3.214203695903495e-05, "loss": 0.24129810333251953, "memory(GiB)": 70.5, "step": 71920, "token_acc": 0.9420849420849421, "train_speed(iter/s)": 1.450046 }, { "epoch": 3.081487511246305, "grad_norm": 4.8673553466796875, "learning_rate": 3.213575122003113e-05, "loss": 0.4482258796691895, "memory(GiB)": 70.5, "step": 71925, "token_acc": 0.9020172910662824, "train_speed(iter/s)": 1.45005 }, { "epoch": 3.0817017265755537, "grad_norm": 3.4829177856445312, "learning_rate": 3.212946580465493e-05, "loss": 0.26018452644348145, "memory(GiB)": 70.5, "step": 71930, "token_acc": 0.9580645161290322, "train_speed(iter/s)": 1.45005 }, { "epoch": 3.0819159419048026, "grad_norm": 2.3189682960510254, "learning_rate": 3.212318071302023e-05, "loss": 0.4218435764312744, "memory(GiB)": 70.5, "step": 71935, "token_acc": 0.9171974522292994, "train_speed(iter/s)": 1.450064 }, { "epoch": 3.082130157234052, "grad_norm": 4.381964683532715, "learning_rate": 3.211689594524088e-05, "loss": 0.2778221845626831, "memory(GiB)": 70.5, "step": 71940, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.450075 }, { "epoch": 3.0823443725633006, "grad_norm": 1.5046316385269165, "learning_rate": 3.211061150143072e-05, "loss": 0.5088245391845703, "memory(GiB)": 70.5, "step": 71945, "token_acc": 0.9025974025974026, "train_speed(iter/s)": 1.450084 }, { "epoch": 3.0825585878925494, "grad_norm": 3.8818976879119873, "learning_rate": 3.210432738170363e-05, "loss": 0.24576034545898437, "memory(GiB)": 70.5, "step": 71950, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.450081 }, { "epoch": 3.0827728032217987, "grad_norm": 3.1130049228668213, "learning_rate": 3.209804358617342e-05, "loss": 0.2445775270462036, "memory(GiB)": 70.5, "step": 71955, "token_acc": 0.9351851851851852, "train_speed(iter/s)": 1.450086 }, { "epoch": 3.0829870185510475, "grad_norm": 3.6635468006134033, "learning_rate": 3.2091760114953945e-05, "loss": 0.41423864364624025, "memory(GiB)": 70.5, "step": 71960, "token_acc": 0.9111747851002865, "train_speed(iter/s)": 1.45009 }, { "epoch": 3.0832012338802963, "grad_norm": 2.0715672969818115, "learning_rate": 3.208547696815905e-05, "loss": 0.106837797164917, "memory(GiB)": 70.5, "step": 71965, "token_acc": 0.9770992366412213, "train_speed(iter/s)": 1.450097 }, { "epoch": 3.0834154492095456, "grad_norm": 4.877560138702393, "learning_rate": 3.207919414590255e-05, "loss": 0.5776120662689209, "memory(GiB)": 70.5, "step": 71970, "token_acc": 0.884180790960452, "train_speed(iter/s)": 1.450107 }, { "epoch": 3.0836296645387944, "grad_norm": 2.5956950187683105, "learning_rate": 3.207291164829824e-05, "loss": 0.2909021615982056, "memory(GiB)": 70.5, "step": 71975, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.450108 }, { "epoch": 3.083843879868043, "grad_norm": 1.570043921470642, "learning_rate": 3.2066629475459964e-05, "loss": 0.10869038105010986, "memory(GiB)": 70.5, "step": 71980, "token_acc": 0.9748427672955975, "train_speed(iter/s)": 1.450113 }, { "epoch": 3.0840580951972925, "grad_norm": 2.723996877670288, "learning_rate": 3.206034762750152e-05, "loss": 0.4121073246002197, "memory(GiB)": 70.5, "step": 71985, "token_acc": 0.9006849315068494, "train_speed(iter/s)": 1.450122 }, { "epoch": 3.0842723105265413, "grad_norm": 1.8899182081222534, "learning_rate": 3.20540661045367e-05, "loss": 0.30558195114135744, "memory(GiB)": 70.5, "step": 71990, "token_acc": 0.948, "train_speed(iter/s)": 1.450127 }, { "epoch": 3.08448652585579, "grad_norm": 4.220983505249023, "learning_rate": 3.204778490667931e-05, "loss": 0.3572458505630493, "memory(GiB)": 70.5, "step": 71995, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.450132 }, { "epoch": 3.0847007411850393, "grad_norm": 2.748910427093506, "learning_rate": 3.2041504034043134e-05, "loss": 0.3138303756713867, "memory(GiB)": 70.5, "step": 72000, "token_acc": 0.9419354838709677, "train_speed(iter/s)": 1.450136 }, { "epoch": 3.0847007411850393, "eval_loss": 2.4731180667877197, "eval_runtime": 11.7253, "eval_samples_per_second": 8.529, "eval_steps_per_second": 8.529, "eval_token_acc": 0.4562146892655367, "step": 72000 }, { "epoch": 3.084914956514288, "grad_norm": 3.921515941619873, "learning_rate": 3.203522348674197e-05, "loss": 0.3115487575531006, "memory(GiB)": 70.5, "step": 72005, "token_acc": 0.5897177419354839, "train_speed(iter/s)": 1.449768 }, { "epoch": 3.085129171843537, "grad_norm": 2.914583444595337, "learning_rate": 3.202894326488958e-05, "loss": 0.3670539379119873, "memory(GiB)": 70.5, "step": 72010, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.44977 }, { "epoch": 3.0853433871727862, "grad_norm": 3.9664411544799805, "learning_rate": 3.2022663368599736e-05, "loss": 0.3514024019241333, "memory(GiB)": 70.5, "step": 72015, "token_acc": 0.9137254901960784, "train_speed(iter/s)": 1.44977 }, { "epoch": 3.085557602502035, "grad_norm": 3.5649290084838867, "learning_rate": 3.201638379798622e-05, "loss": 0.19372320175170898, "memory(GiB)": 70.5, "step": 72020, "token_acc": 0.9595588235294118, "train_speed(iter/s)": 1.449772 }, { "epoch": 3.085771817831284, "grad_norm": 3.2322380542755127, "learning_rate": 3.2010104553162765e-05, "loss": 0.32059626579284667, "memory(GiB)": 70.5, "step": 72025, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.449777 }, { "epoch": 3.085986033160533, "grad_norm": 2.784275770187378, "learning_rate": 3.200382563424314e-05, "loss": 0.2988420486450195, "memory(GiB)": 70.5, "step": 72030, "token_acc": 0.9506578947368421, "train_speed(iter/s)": 1.44979 }, { "epoch": 3.086200248489782, "grad_norm": 2.9945321083068848, "learning_rate": 3.1997547041341105e-05, "loss": 0.3317952871322632, "memory(GiB)": 70.5, "step": 72035, "token_acc": 0.9049295774647887, "train_speed(iter/s)": 1.449793 }, { "epoch": 3.0864144638190307, "grad_norm": 2.6717183589935303, "learning_rate": 3.1991268774570396e-05, "loss": 0.24925656318664552, "memory(GiB)": 70.5, "step": 72040, "token_acc": 0.960431654676259, "train_speed(iter/s)": 1.449802 }, { "epoch": 3.08662867914828, "grad_norm": 4.2367424964904785, "learning_rate": 3.1984990834044746e-05, "loss": 0.262900710105896, "memory(GiB)": 70.5, "step": 72045, "token_acc": 0.946875, "train_speed(iter/s)": 1.449807 }, { "epoch": 3.086842894477529, "grad_norm": 1.072620153427124, "learning_rate": 3.1978713219877876e-05, "loss": 0.35761520862579343, "memory(GiB)": 70.5, "step": 72050, "token_acc": 0.911504424778761, "train_speed(iter/s)": 1.44981 }, { "epoch": 3.0870571098067776, "grad_norm": 11.878381729125977, "learning_rate": 3.197243593218353e-05, "loss": 0.6787012100219727, "memory(GiB)": 70.5, "step": 72055, "token_acc": 0.8299120234604106, "train_speed(iter/s)": 1.449805 }, { "epoch": 3.087271325136027, "grad_norm": 2.9534974098205566, "learning_rate": 3.196615897107542e-05, "loss": 0.2505557775497437, "memory(GiB)": 70.5, "step": 72060, "token_acc": 0.9423631123919308, "train_speed(iter/s)": 1.449811 }, { "epoch": 3.0874855404652757, "grad_norm": 3.5761735439300537, "learning_rate": 3.1959882336667254e-05, "loss": 0.36669230461120605, "memory(GiB)": 70.5, "step": 72065, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.44981 }, { "epoch": 3.0876997557945245, "grad_norm": 1.9142132997512817, "learning_rate": 3.1953606029072754e-05, "loss": 0.12959352731704712, "memory(GiB)": 70.5, "step": 72070, "token_acc": 0.9725085910652921, "train_speed(iter/s)": 1.449814 }, { "epoch": 3.0879139711237737, "grad_norm": 0.0631110668182373, "learning_rate": 3.194733004840559e-05, "loss": 0.3647179126739502, "memory(GiB)": 70.5, "step": 72075, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.449824 }, { "epoch": 3.0881281864530226, "grad_norm": 4.409379959106445, "learning_rate": 3.194105439477949e-05, "loss": 0.3199739694595337, "memory(GiB)": 70.5, "step": 72080, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.449839 }, { "epoch": 3.0883424017822714, "grad_norm": 5.569331169128418, "learning_rate": 3.193477906830813e-05, "loss": 0.5056793212890625, "memory(GiB)": 70.5, "step": 72085, "token_acc": 0.8905109489051095, "train_speed(iter/s)": 1.449849 }, { "epoch": 3.0885566171115206, "grad_norm": 3.2271904945373535, "learning_rate": 3.192850406910519e-05, "loss": 0.3904532194137573, "memory(GiB)": 70.5, "step": 72090, "token_acc": 0.9021739130434783, "train_speed(iter/s)": 1.449847 }, { "epoch": 3.0887708324407694, "grad_norm": 2.832735061645508, "learning_rate": 3.1922229397284355e-05, "loss": 0.379850959777832, "memory(GiB)": 70.5, "step": 72095, "token_acc": 0.9409722222222222, "train_speed(iter/s)": 1.449843 }, { "epoch": 3.0889850477700183, "grad_norm": 4.341202735900879, "learning_rate": 3.191595505295927e-05, "loss": 0.28164381980895997, "memory(GiB)": 70.5, "step": 72100, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.449849 }, { "epoch": 3.0891992630992675, "grad_norm": 2.83276629447937, "learning_rate": 3.1909681036243644e-05, "loss": 0.2094271183013916, "memory(GiB)": 70.5, "step": 72105, "token_acc": 0.9371069182389937, "train_speed(iter/s)": 1.449853 }, { "epoch": 3.0894134784285163, "grad_norm": 2.9010066986083984, "learning_rate": 3.1903407347251114e-05, "loss": 0.3240758180618286, "memory(GiB)": 70.5, "step": 72110, "token_acc": 0.9358490566037736, "train_speed(iter/s)": 1.449856 }, { "epoch": 3.089627693757765, "grad_norm": 3.542545795440674, "learning_rate": 3.189713398609534e-05, "loss": 0.5502250671386719, "memory(GiB)": 70.5, "step": 72115, "token_acc": 0.8753799392097265, "train_speed(iter/s)": 1.449873 }, { "epoch": 3.0898419090870144, "grad_norm": 3.344891309738159, "learning_rate": 3.1890860952889955e-05, "loss": 0.4282047271728516, "memory(GiB)": 70.5, "step": 72120, "token_acc": 0.8910505836575876, "train_speed(iter/s)": 1.449878 }, { "epoch": 3.090056124416263, "grad_norm": 3.585886001586914, "learning_rate": 3.188458824774862e-05, "loss": 0.46486434936523435, "memory(GiB)": 70.5, "step": 72125, "token_acc": 0.913946587537092, "train_speed(iter/s)": 1.449889 }, { "epoch": 3.090270339745512, "grad_norm": 1.1825350522994995, "learning_rate": 3.1878315870784975e-05, "loss": 0.23446979522705078, "memory(GiB)": 70.5, "step": 72130, "token_acc": 0.9371069182389937, "train_speed(iter/s)": 1.449893 }, { "epoch": 3.0904845550747613, "grad_norm": 1.9192047119140625, "learning_rate": 3.187204382211262e-05, "loss": 0.3400691270828247, "memory(GiB)": 70.5, "step": 72135, "token_acc": 0.9226006191950464, "train_speed(iter/s)": 1.449905 }, { "epoch": 3.09069877040401, "grad_norm": 2.7975101470947266, "learning_rate": 3.18657721018452e-05, "loss": 0.27067506313323975, "memory(GiB)": 70.5, "step": 72140, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.449904 }, { "epoch": 3.090912985733259, "grad_norm": 2.1594815254211426, "learning_rate": 3.185950071009633e-05, "loss": 0.3359628200531006, "memory(GiB)": 70.5, "step": 72145, "token_acc": 0.9212827988338192, "train_speed(iter/s)": 1.449907 }, { "epoch": 3.091127201062508, "grad_norm": 3.6089682579040527, "learning_rate": 3.185322964697963e-05, "loss": 0.5103726863861084, "memory(GiB)": 70.5, "step": 72150, "token_acc": 0.8923076923076924, "train_speed(iter/s)": 1.449901 }, { "epoch": 3.091341416391757, "grad_norm": 0.9019078016281128, "learning_rate": 3.184695891260869e-05, "loss": 0.32133915424346926, "memory(GiB)": 70.5, "step": 72155, "token_acc": 0.91796875, "train_speed(iter/s)": 1.449914 }, { "epoch": 3.091555631721006, "grad_norm": 3.298520803451538, "learning_rate": 3.184068850709711e-05, "loss": 0.33662288188934325, "memory(GiB)": 70.5, "step": 72160, "token_acc": 0.9256198347107438, "train_speed(iter/s)": 1.449922 }, { "epoch": 3.091769847050255, "grad_norm": 3.5231752395629883, "learning_rate": 3.183441843055851e-05, "loss": 0.1660441279411316, "memory(GiB)": 70.5, "step": 72165, "token_acc": 0.9570552147239264, "train_speed(iter/s)": 1.449932 }, { "epoch": 3.091984062379504, "grad_norm": 3.15330171585083, "learning_rate": 3.182814868310645e-05, "loss": 0.39726393222808837, "memory(GiB)": 70.5, "step": 72170, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.44994 }, { "epoch": 3.0921982777087527, "grad_norm": 3.7912721633911133, "learning_rate": 3.1821879264854514e-05, "loss": 0.5410212516784668, "memory(GiB)": 70.5, "step": 72175, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.44994 }, { "epoch": 3.092412493038002, "grad_norm": 2.881013870239258, "learning_rate": 3.181561017591629e-05, "loss": 0.5505449771881104, "memory(GiB)": 70.5, "step": 72180, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.449954 }, { "epoch": 3.0926267083672507, "grad_norm": 2.4946842193603516, "learning_rate": 3.180934141640536e-05, "loss": 0.5301424503326416, "memory(GiB)": 70.5, "step": 72185, "token_acc": 0.8803418803418803, "train_speed(iter/s)": 1.449968 }, { "epoch": 3.0928409236964995, "grad_norm": 1.6814677715301514, "learning_rate": 3.1803072986435276e-05, "loss": 0.20614368915557862, "memory(GiB)": 70.5, "step": 72190, "token_acc": 0.9501557632398754, "train_speed(iter/s)": 1.449965 }, { "epoch": 3.093055139025749, "grad_norm": 7.002566814422607, "learning_rate": 3.179680488611958e-05, "loss": 0.2626950263977051, "memory(GiB)": 70.5, "step": 72195, "token_acc": 0.9570957095709571, "train_speed(iter/s)": 1.44997 }, { "epoch": 3.0932693543549976, "grad_norm": 5.450211524963379, "learning_rate": 3.179053711557185e-05, "loss": 0.6183314323425293, "memory(GiB)": 70.5, "step": 72200, "token_acc": 0.8757961783439491, "train_speed(iter/s)": 1.449978 }, { "epoch": 3.0934835696842464, "grad_norm": 1.1715205907821655, "learning_rate": 3.178426967490562e-05, "loss": 0.22670562267303468, "memory(GiB)": 70.5, "step": 72205, "token_acc": 0.9426229508196722, "train_speed(iter/s)": 1.449987 }, { "epoch": 3.0936977850134957, "grad_norm": 4.251866340637207, "learning_rate": 3.1778002564234435e-05, "loss": 0.23005094528198242, "memory(GiB)": 70.5, "step": 72210, "token_acc": 0.952054794520548, "train_speed(iter/s)": 1.449989 }, { "epoch": 3.0939120003427445, "grad_norm": 6.39274263381958, "learning_rate": 3.177173578367183e-05, "loss": 0.5850098133087158, "memory(GiB)": 70.5, "step": 72215, "token_acc": 0.8808510638297873, "train_speed(iter/s)": 1.449998 }, { "epoch": 3.0941262156719933, "grad_norm": 3.1683244705200195, "learning_rate": 3.176546933333132e-05, "loss": 0.28982815742492674, "memory(GiB)": 70.5, "step": 72220, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.449996 }, { "epoch": 3.0943404310012426, "grad_norm": 3.5372869968414307, "learning_rate": 3.175920321332644e-05, "loss": 0.3099156618118286, "memory(GiB)": 70.5, "step": 72225, "token_acc": 0.9291784702549575, "train_speed(iter/s)": 1.449992 }, { "epoch": 3.0945546463304914, "grad_norm": 3.675954818725586, "learning_rate": 3.175293742377072e-05, "loss": 0.36700010299682617, "memory(GiB)": 70.5, "step": 72230, "token_acc": 0.9225806451612903, "train_speed(iter/s)": 1.449993 }, { "epoch": 3.09476886165974, "grad_norm": 1.5851383209228516, "learning_rate": 3.1746671964777635e-05, "loss": 0.17654627561569214, "memory(GiB)": 70.5, "step": 72235, "token_acc": 0.9540636042402827, "train_speed(iter/s)": 1.449996 }, { "epoch": 3.0949830769889894, "grad_norm": 3.1643013954162598, "learning_rate": 3.1740406836460724e-05, "loss": 0.36389172077178955, "memory(GiB)": 70.5, "step": 72240, "token_acc": 0.9375, "train_speed(iter/s)": 1.450001 }, { "epoch": 3.0951972923182383, "grad_norm": 3.1174161434173584, "learning_rate": 3.173414203893346e-05, "loss": 0.34289135932922366, "memory(GiB)": 70.5, "step": 72245, "token_acc": 0.944954128440367, "train_speed(iter/s)": 1.450008 }, { "epoch": 3.095411507647487, "grad_norm": 0.7762568593025208, "learning_rate": 3.172787757230934e-05, "loss": 0.19553376436233522, "memory(GiB)": 70.5, "step": 72250, "token_acc": 0.9589905362776026, "train_speed(iter/s)": 1.450021 }, { "epoch": 3.0956257229767363, "grad_norm": 1.355108618736267, "learning_rate": 3.172161343670188e-05, "loss": 0.2927064895629883, "memory(GiB)": 70.5, "step": 72255, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.450031 }, { "epoch": 3.095839938305985, "grad_norm": 2.276758909225464, "learning_rate": 3.1715349632224536e-05, "loss": 0.35223267078399656, "memory(GiB)": 70.5, "step": 72260, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.450034 }, { "epoch": 3.096054153635234, "grad_norm": 5.723140239715576, "learning_rate": 3.1709086158990774e-05, "loss": 0.33581743240356443, "memory(GiB)": 70.5, "step": 72265, "token_acc": 0.8925619834710744, "train_speed(iter/s)": 1.450039 }, { "epoch": 3.096268368964483, "grad_norm": 2.120798110961914, "learning_rate": 3.170282301711409e-05, "loss": 0.32944426536560056, "memory(GiB)": 70.5, "step": 72270, "token_acc": 0.9144736842105263, "train_speed(iter/s)": 1.450052 }, { "epoch": 3.096482584293732, "grad_norm": 3.476271867752075, "learning_rate": 3.169656020670793e-05, "loss": 0.1811241865158081, "memory(GiB)": 70.5, "step": 72275, "token_acc": 0.9621848739495799, "train_speed(iter/s)": 1.450056 }, { "epoch": 3.096696799622981, "grad_norm": 0.6553789377212524, "learning_rate": 3.1690297727885744e-05, "loss": 0.19377351999282838, "memory(GiB)": 70.5, "step": 72280, "token_acc": 0.9568627450980393, "train_speed(iter/s)": 1.450071 }, { "epoch": 3.09691101495223, "grad_norm": 5.710888862609863, "learning_rate": 3.1684035580761e-05, "loss": 0.3448661804199219, "memory(GiB)": 70.5, "step": 72285, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.450074 }, { "epoch": 3.097125230281479, "grad_norm": 4.354156017303467, "learning_rate": 3.1677773765447116e-05, "loss": 0.4079193115234375, "memory(GiB)": 70.5, "step": 72290, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.450072 }, { "epoch": 3.0973394456107277, "grad_norm": 4.1461358070373535, "learning_rate": 3.1671512282057556e-05, "loss": 0.47207136154174806, "memory(GiB)": 70.5, "step": 72295, "token_acc": 0.8893617021276595, "train_speed(iter/s)": 1.450072 }, { "epoch": 3.097553660939977, "grad_norm": 0.9477460980415344, "learning_rate": 3.166525113070575e-05, "loss": 0.27716965675354005, "memory(GiB)": 70.5, "step": 72300, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.450073 }, { "epoch": 3.097767876269226, "grad_norm": 0.2908230721950531, "learning_rate": 3.165899031150512e-05, "loss": 0.1237373948097229, "memory(GiB)": 70.5, "step": 72305, "token_acc": 0.9742647058823529, "train_speed(iter/s)": 1.450087 }, { "epoch": 3.0979820915984746, "grad_norm": 3.709801197052002, "learning_rate": 3.165272982456908e-05, "loss": 0.39800252914428713, "memory(GiB)": 70.5, "step": 72310, "token_acc": 0.9003436426116839, "train_speed(iter/s)": 1.450093 }, { "epoch": 3.098196306927724, "grad_norm": 0.15580220520496368, "learning_rate": 3.164646967001106e-05, "loss": 0.40797224044799807, "memory(GiB)": 70.5, "step": 72315, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.450107 }, { "epoch": 3.0984105222569727, "grad_norm": 3.3888063430786133, "learning_rate": 3.164020984794444e-05, "loss": 0.647221851348877, "memory(GiB)": 70.5, "step": 72320, "token_acc": 0.8601190476190477, "train_speed(iter/s)": 1.450112 }, { "epoch": 3.0986247375862215, "grad_norm": 1.0367556810379028, "learning_rate": 3.163395035848263e-05, "loss": 0.35288712978363035, "memory(GiB)": 70.5, "step": 72325, "token_acc": 0.9198813056379822, "train_speed(iter/s)": 1.450123 }, { "epoch": 3.0988389529154707, "grad_norm": 3.254272222518921, "learning_rate": 3.1627691201739075e-05, "loss": 0.37104620933532717, "memory(GiB)": 70.5, "step": 72330, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.450137 }, { "epoch": 3.0990531682447195, "grad_norm": 4.872398853302002, "learning_rate": 3.162143237782711e-05, "loss": 0.5394598484039307, "memory(GiB)": 70.5, "step": 72335, "token_acc": 0.875, "train_speed(iter/s)": 1.450146 }, { "epoch": 3.0992673835739684, "grad_norm": 4.410943984985352, "learning_rate": 3.161517388686014e-05, "loss": 0.43532605171203614, "memory(GiB)": 70.5, "step": 72340, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.450156 }, { "epoch": 3.0994815989032176, "grad_norm": 2.3876001834869385, "learning_rate": 3.160891572895155e-05, "loss": 0.17123310565948485, "memory(GiB)": 70.5, "step": 72345, "token_acc": 0.9607142857142857, "train_speed(iter/s)": 1.450157 }, { "epoch": 3.0996958142324664, "grad_norm": 3.58986234664917, "learning_rate": 3.160265790421469e-05, "loss": 0.2779965877532959, "memory(GiB)": 70.5, "step": 72350, "token_acc": 0.9462809917355371, "train_speed(iter/s)": 1.450162 }, { "epoch": 3.0999100295617152, "grad_norm": 3.803253412246704, "learning_rate": 3.159640041276295e-05, "loss": 0.3191555976867676, "memory(GiB)": 70.5, "step": 72355, "token_acc": 0.9296636085626911, "train_speed(iter/s)": 1.450165 }, { "epoch": 3.1001242448909645, "grad_norm": 0.9214876294136047, "learning_rate": 3.1590143254709684e-05, "loss": 0.3811555624008179, "memory(GiB)": 70.5, "step": 72360, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.450183 }, { "epoch": 3.1003384602202133, "grad_norm": 4.278130531311035, "learning_rate": 3.158388643016823e-05, "loss": 0.5344477653503418, "memory(GiB)": 70.5, "step": 72365, "token_acc": 0.9180887372013652, "train_speed(iter/s)": 1.450193 }, { "epoch": 3.100552675549462, "grad_norm": 3.1651813983917236, "learning_rate": 3.157762993925196e-05, "loss": 0.14538075923919677, "memory(GiB)": 70.5, "step": 72370, "token_acc": 0.9661654135338346, "train_speed(iter/s)": 1.450204 }, { "epoch": 3.1007668908787114, "grad_norm": 1.953881025314331, "learning_rate": 3.15713737820742e-05, "loss": 0.510716438293457, "memory(GiB)": 70.5, "step": 72375, "token_acc": 0.896797153024911, "train_speed(iter/s)": 1.450208 }, { "epoch": 3.10098110620796, "grad_norm": 0.925270140171051, "learning_rate": 3.156511795874829e-05, "loss": 0.3128093481063843, "memory(GiB)": 70.5, "step": 72380, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.450208 }, { "epoch": 3.101195321537209, "grad_norm": 3.6015195846557617, "learning_rate": 3.155886246938756e-05, "loss": 0.5107683658599853, "memory(GiB)": 70.5, "step": 72385, "token_acc": 0.8850174216027874, "train_speed(iter/s)": 1.450209 }, { "epoch": 3.1014095368664583, "grad_norm": 1.882399320602417, "learning_rate": 3.155260731410534e-05, "loss": 0.18931350708007813, "memory(GiB)": 70.5, "step": 72390, "token_acc": 0.9540229885057471, "train_speed(iter/s)": 1.450205 }, { "epoch": 3.101623752195707, "grad_norm": 2.520320177078247, "learning_rate": 3.1546352493014946e-05, "loss": 0.6724167346954346, "memory(GiB)": 70.5, "step": 72395, "token_acc": 0.8754208754208754, "train_speed(iter/s)": 1.450217 }, { "epoch": 3.101837967524956, "grad_norm": 5.6448655128479, "learning_rate": 3.154009800622967e-05, "loss": 0.5368038177490234, "memory(GiB)": 70.5, "step": 72400, "token_acc": 0.8817567567567568, "train_speed(iter/s)": 1.450231 }, { "epoch": 3.102052182854205, "grad_norm": 14.400019645690918, "learning_rate": 3.1533843853862856e-05, "loss": 0.19797039031982422, "memory(GiB)": 70.5, "step": 72405, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.450246 }, { "epoch": 3.102266398183454, "grad_norm": 2.807063579559326, "learning_rate": 3.1527590036027766e-05, "loss": 0.2670663595199585, "memory(GiB)": 70.5, "step": 72410, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.450252 }, { "epoch": 3.1024806135127028, "grad_norm": 2.312368631362915, "learning_rate": 3.152133655283773e-05, "loss": 0.30935609340667725, "memory(GiB)": 70.5, "step": 72415, "token_acc": 0.9382716049382716, "train_speed(iter/s)": 1.450252 }, { "epoch": 3.102694828841952, "grad_norm": 0.4019133746623993, "learning_rate": 3.151508340440601e-05, "loss": 0.46421451568603517, "memory(GiB)": 70.5, "step": 72420, "token_acc": 0.9031007751937985, "train_speed(iter/s)": 1.450249 }, { "epoch": 3.102909044171201, "grad_norm": 1.4174926280975342, "learning_rate": 3.150883059084588e-05, "loss": 0.19279974699020386, "memory(GiB)": 70.5, "step": 72425, "token_acc": 0.9584775086505191, "train_speed(iter/s)": 1.450264 }, { "epoch": 3.1031232595004496, "grad_norm": 2.884856700897217, "learning_rate": 3.150257811227065e-05, "loss": 0.2271991491317749, "memory(GiB)": 70.5, "step": 72430, "token_acc": 0.9397905759162304, "train_speed(iter/s)": 1.450268 }, { "epoch": 3.103337474829699, "grad_norm": 1.4148550033569336, "learning_rate": 3.149632596879356e-05, "loss": 0.279423189163208, "memory(GiB)": 70.5, "step": 72435, "token_acc": 0.9337539432176656, "train_speed(iter/s)": 1.450267 }, { "epoch": 3.1035516901589477, "grad_norm": 0.9961145520210266, "learning_rate": 3.149007416052789e-05, "loss": 0.12357563972473144, "memory(GiB)": 70.5, "step": 72440, "token_acc": 0.97, "train_speed(iter/s)": 1.450265 }, { "epoch": 3.1037659054881965, "grad_norm": 0.5872426629066467, "learning_rate": 3.148382268758689e-05, "loss": 0.41448063850402833, "memory(GiB)": 70.5, "step": 72445, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.450263 }, { "epoch": 3.103980120817446, "grad_norm": 1.537461757659912, "learning_rate": 3.14775715500838e-05, "loss": 0.3282250165939331, "memory(GiB)": 70.5, "step": 72450, "token_acc": 0.9358974358974359, "train_speed(iter/s)": 1.450277 }, { "epoch": 3.1041943361466946, "grad_norm": 3.569100856781006, "learning_rate": 3.147132074813189e-05, "loss": 0.41808743476867677, "memory(GiB)": 70.5, "step": 72455, "token_acc": 0.9107142857142857, "train_speed(iter/s)": 1.450293 }, { "epoch": 3.1044085514759434, "grad_norm": 2.213252067565918, "learning_rate": 3.146507028184438e-05, "loss": 0.15486435890197753, "memory(GiB)": 70.5, "step": 72460, "token_acc": 0.9713261648745519, "train_speed(iter/s)": 1.450293 }, { "epoch": 3.1046227668051927, "grad_norm": 2.0985302925109863, "learning_rate": 3.1458820151334504e-05, "loss": 0.3396519660949707, "memory(GiB)": 70.5, "step": 72465, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.450288 }, { "epoch": 3.1048369821344415, "grad_norm": 2.9352684020996094, "learning_rate": 3.145257035671548e-05, "loss": 0.33759136199951173, "memory(GiB)": 70.5, "step": 72470, "token_acc": 0.9300911854103343, "train_speed(iter/s)": 1.450287 }, { "epoch": 3.1050511974636903, "grad_norm": 3.4966232776641846, "learning_rate": 3.144632089810057e-05, "loss": 0.2417300224304199, "memory(GiB)": 70.5, "step": 72475, "token_acc": 0.9463722397476341, "train_speed(iter/s)": 1.450295 }, { "epoch": 3.1052654127929396, "grad_norm": 3.713627576828003, "learning_rate": 3.144007177560295e-05, "loss": 0.32644755840301515, "memory(GiB)": 70.5, "step": 72480, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.450304 }, { "epoch": 3.1054796281221884, "grad_norm": 1.1061186790466309, "learning_rate": 3.143382298933585e-05, "loss": 0.26905815601348876, "memory(GiB)": 70.5, "step": 72485, "token_acc": 0.9154078549848943, "train_speed(iter/s)": 1.450308 }, { "epoch": 3.105693843451437, "grad_norm": 2.6376612186431885, "learning_rate": 3.142757453941246e-05, "loss": 0.16301276683807372, "memory(GiB)": 70.5, "step": 72490, "token_acc": 0.968503937007874, "train_speed(iter/s)": 1.450312 }, { "epoch": 3.1059080587806864, "grad_norm": 2.9311716556549072, "learning_rate": 3.142132642594597e-05, "loss": 0.433198356628418, "memory(GiB)": 70.5, "step": 72495, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.450322 }, { "epoch": 3.1061222741099352, "grad_norm": 4.513182640075684, "learning_rate": 3.141507864904959e-05, "loss": 0.22942323684692384, "memory(GiB)": 70.5, "step": 72500, "token_acc": 0.9465020576131687, "train_speed(iter/s)": 1.450328 }, { "epoch": 3.1061222741099352, "eval_loss": 2.5206661224365234, "eval_runtime": 11.347, "eval_samples_per_second": 8.813, "eval_steps_per_second": 8.813, "eval_token_acc": 0.4444444444444444, "step": 72500 }, { "epoch": 3.106336489439184, "grad_norm": 1.0908310413360596, "learning_rate": 3.1408831208836496e-05, "loss": 0.4050902843475342, "memory(GiB)": 70.5, "step": 72505, "token_acc": 0.5690168818272096, "train_speed(iter/s)": 1.449977 }, { "epoch": 3.1065507047684333, "grad_norm": 6.617101669311523, "learning_rate": 3.140258410541985e-05, "loss": 0.4939249038696289, "memory(GiB)": 70.5, "step": 72510, "token_acc": 0.8859649122807017, "train_speed(iter/s)": 1.449984 }, { "epoch": 3.106764920097682, "grad_norm": 4.133399486541748, "learning_rate": 3.139633733891285e-05, "loss": 0.169075608253479, "memory(GiB)": 70.5, "step": 72515, "token_acc": 0.9678571428571429, "train_speed(iter/s)": 1.449983 }, { "epoch": 3.106979135426931, "grad_norm": 3.0969440937042236, "learning_rate": 3.139009090942865e-05, "loss": 0.2322669506072998, "memory(GiB)": 70.5, "step": 72520, "token_acc": 0.9548872180451128, "train_speed(iter/s)": 1.449989 }, { "epoch": 3.10719335075618, "grad_norm": 2.005704879760742, "learning_rate": 3.138384481708041e-05, "loss": 0.1584375262260437, "memory(GiB)": 70.5, "step": 72525, "token_acc": 0.967391304347826, "train_speed(iter/s)": 1.449991 }, { "epoch": 3.107407566085429, "grad_norm": 3.381958246231079, "learning_rate": 3.137759906198129e-05, "loss": 0.43033900260925295, "memory(GiB)": 70.5, "step": 72530, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.449989 }, { "epoch": 3.107621781414678, "grad_norm": 3.456819772720337, "learning_rate": 3.137135364424442e-05, "loss": 0.5314296245574951, "memory(GiB)": 70.5, "step": 72535, "token_acc": 0.8821752265861027, "train_speed(iter/s)": 1.449994 }, { "epoch": 3.107835996743927, "grad_norm": 8.818217277526855, "learning_rate": 3.136510856398297e-05, "loss": 0.5999717712402344, "memory(GiB)": 70.5, "step": 72540, "token_acc": 0.8804347826086957, "train_speed(iter/s)": 1.450023 }, { "epoch": 3.108050212073176, "grad_norm": 0.03712321072816849, "learning_rate": 3.135886382131003e-05, "loss": 0.21321263313293456, "memory(GiB)": 70.5, "step": 72545, "token_acc": 0.9548872180451128, "train_speed(iter/s)": 1.450031 }, { "epoch": 3.1082644274024247, "grad_norm": 1.8311684131622314, "learning_rate": 3.135261941633878e-05, "loss": 0.23181686401367188, "memory(GiB)": 70.5, "step": 72550, "token_acc": 0.9573863636363636, "train_speed(iter/s)": 1.450034 }, { "epoch": 3.108478642731674, "grad_norm": 3.9737746715545654, "learning_rate": 3.134637534918231e-05, "loss": 0.24115538597106934, "memory(GiB)": 70.5, "step": 72555, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.450042 }, { "epoch": 3.1086928580609228, "grad_norm": 2.0471367835998535, "learning_rate": 3.134013161995377e-05, "loss": 0.622397518157959, "memory(GiB)": 70.5, "step": 72560, "token_acc": 0.8875739644970414, "train_speed(iter/s)": 1.450067 }, { "epoch": 3.1089070733901716, "grad_norm": 5.108049392700195, "learning_rate": 3.133388822876624e-05, "loss": 0.49580111503601076, "memory(GiB)": 70.5, "step": 72565, "token_acc": 0.8944281524926686, "train_speed(iter/s)": 1.450081 }, { "epoch": 3.109121288719421, "grad_norm": 0.9765637516975403, "learning_rate": 3.1327645175732826e-05, "loss": 0.5344418048858642, "memory(GiB)": 70.5, "step": 72570, "token_acc": 0.8984375, "train_speed(iter/s)": 1.450082 }, { "epoch": 3.1093355040486697, "grad_norm": 0.7726131677627563, "learning_rate": 3.132140246096665e-05, "loss": 0.31422982215881345, "memory(GiB)": 70.5, "step": 72575, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.450088 }, { "epoch": 3.1095497193779185, "grad_norm": 0.14385506510734558, "learning_rate": 3.131516008458077e-05, "loss": 0.14127297401428224, "memory(GiB)": 70.5, "step": 72580, "token_acc": 0.9652996845425867, "train_speed(iter/s)": 1.450091 }, { "epoch": 3.1097639347071677, "grad_norm": 1.0451626777648926, "learning_rate": 3.130891804668832e-05, "loss": 0.4275080680847168, "memory(GiB)": 70.5, "step": 72585, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.450093 }, { "epoch": 3.1099781500364165, "grad_norm": 0.32351529598236084, "learning_rate": 3.1302676347402346e-05, "loss": 0.054647743701934814, "memory(GiB)": 70.5, "step": 72590, "token_acc": 0.9958333333333333, "train_speed(iter/s)": 1.450096 }, { "epoch": 3.1101923653656653, "grad_norm": 5.806795120239258, "learning_rate": 3.129643498683592e-05, "loss": 0.5817899703979492, "memory(GiB)": 70.5, "step": 72595, "token_acc": 0.8819672131147541, "train_speed(iter/s)": 1.450097 }, { "epoch": 3.1104065806949146, "grad_norm": 0.17504674196243286, "learning_rate": 3.129019396510212e-05, "loss": 0.19550241231918336, "memory(GiB)": 70.5, "step": 72600, "token_acc": 0.9601328903654485, "train_speed(iter/s)": 1.450103 }, { "epoch": 3.1106207960241634, "grad_norm": 4.341559410095215, "learning_rate": 3.128395328231401e-05, "loss": 0.18760616779327394, "memory(GiB)": 70.5, "step": 72605, "token_acc": 0.9700374531835206, "train_speed(iter/s)": 1.450109 }, { "epoch": 3.1108350113534122, "grad_norm": 3.714444875717163, "learning_rate": 3.1277712938584646e-05, "loss": 0.15841047763824462, "memory(GiB)": 70.5, "step": 72610, "token_acc": 0.9533333333333334, "train_speed(iter/s)": 1.450113 }, { "epoch": 3.1110492266826615, "grad_norm": 2.111168622970581, "learning_rate": 3.127147293402708e-05, "loss": 0.27705495357513427, "memory(GiB)": 70.5, "step": 72615, "token_acc": 0.9249146757679181, "train_speed(iter/s)": 1.450122 }, { "epoch": 3.1112634420119103, "grad_norm": 0.38275057077407837, "learning_rate": 3.126523326875432e-05, "loss": 0.31577348709106445, "memory(GiB)": 70.5, "step": 72620, "token_acc": 0.9233038348082596, "train_speed(iter/s)": 1.450124 }, { "epoch": 3.111477657341159, "grad_norm": 2.8276760578155518, "learning_rate": 3.1258993942879456e-05, "loss": 0.299195384979248, "memory(GiB)": 70.5, "step": 72625, "token_acc": 0.9307692307692308, "train_speed(iter/s)": 1.450123 }, { "epoch": 3.1116918726704084, "grad_norm": 4.036513328552246, "learning_rate": 3.125275495651551e-05, "loss": 0.29549827575683596, "memory(GiB)": 70.5, "step": 72630, "token_acc": 0.9449838187702265, "train_speed(iter/s)": 1.45014 }, { "epoch": 3.111906087999657, "grad_norm": 5.382920742034912, "learning_rate": 3.1246516309775484e-05, "loss": 0.4772514820098877, "memory(GiB)": 70.5, "step": 72635, "token_acc": 0.914179104477612, "train_speed(iter/s)": 1.450139 }, { "epoch": 3.112120303328906, "grad_norm": 2.990969181060791, "learning_rate": 3.12402780027724e-05, "loss": 0.5058940410614013, "memory(GiB)": 70.5, "step": 72640, "token_acc": 0.8934169278996865, "train_speed(iter/s)": 1.450139 }, { "epoch": 3.1123345186581552, "grad_norm": 3.5034496784210205, "learning_rate": 3.123404003561929e-05, "loss": 0.1864193320274353, "memory(GiB)": 70.5, "step": 72645, "token_acc": 0.9591836734693877, "train_speed(iter/s)": 1.450152 }, { "epoch": 3.112548733987404, "grad_norm": 4.287528991699219, "learning_rate": 3.122780240842915e-05, "loss": 0.3437450408935547, "memory(GiB)": 70.5, "step": 72650, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.450156 }, { "epoch": 3.112762949316653, "grad_norm": 1.6854544878005981, "learning_rate": 3.122156512131497e-05, "loss": 0.15466644763946533, "memory(GiB)": 70.5, "step": 72655, "token_acc": 0.9624060150375939, "train_speed(iter/s)": 1.450161 }, { "epoch": 3.112977164645902, "grad_norm": 4.366186141967773, "learning_rate": 3.1215328174389754e-05, "loss": 0.3588210344314575, "memory(GiB)": 70.5, "step": 72660, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.450167 }, { "epoch": 3.113191379975151, "grad_norm": 3.993802785873413, "learning_rate": 3.1209091567766484e-05, "loss": 0.4692076206207275, "memory(GiB)": 70.5, "step": 72665, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.450187 }, { "epoch": 3.1134055953044, "grad_norm": 2.957137107849121, "learning_rate": 3.120285530155816e-05, "loss": 0.3129724025726318, "memory(GiB)": 70.5, "step": 72670, "token_acc": 0.9329268292682927, "train_speed(iter/s)": 1.450192 }, { "epoch": 3.113619810633649, "grad_norm": 4.819007873535156, "learning_rate": 3.1196619375877746e-05, "loss": 0.4841763019561768, "memory(GiB)": 70.5, "step": 72675, "token_acc": 0.9017857142857143, "train_speed(iter/s)": 1.450201 }, { "epoch": 3.113834025962898, "grad_norm": 2.5300872325897217, "learning_rate": 3.11903837908382e-05, "loss": 0.33637175559997556, "memory(GiB)": 70.5, "step": 72680, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.450205 }, { "epoch": 3.114048241292147, "grad_norm": 3.0469753742218018, "learning_rate": 3.1184148546552505e-05, "loss": 0.37166786193847656, "memory(GiB)": 70.5, "step": 72685, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.450219 }, { "epoch": 3.114262456621396, "grad_norm": 2.379962205886841, "learning_rate": 3.117791364313361e-05, "loss": 0.18144602775573732, "memory(GiB)": 70.5, "step": 72690, "token_acc": 0.9548192771084337, "train_speed(iter/s)": 1.450217 }, { "epoch": 3.1144766719506447, "grad_norm": 2.386495590209961, "learning_rate": 3.117167908069445e-05, "loss": 0.13638323545455933, "memory(GiB)": 70.5, "step": 72695, "token_acc": 0.9635036496350365, "train_speed(iter/s)": 1.450217 }, { "epoch": 3.114690887279894, "grad_norm": 0.051362134516239166, "learning_rate": 3.116544485934799e-05, "loss": 0.34643054008483887, "memory(GiB)": 70.5, "step": 72700, "token_acc": 0.9297520661157025, "train_speed(iter/s)": 1.450221 }, { "epoch": 3.1149051026091428, "grad_norm": 3.02669095993042, "learning_rate": 3.115921097920718e-05, "loss": 0.12955853939056397, "memory(GiB)": 70.5, "step": 72705, "token_acc": 0.9695817490494296, "train_speed(iter/s)": 1.450227 }, { "epoch": 3.1151193179383916, "grad_norm": 2.3156704902648926, "learning_rate": 3.1152977440384927e-05, "loss": 0.14932955503463746, "memory(GiB)": 70.5, "step": 72710, "token_acc": 0.9637681159420289, "train_speed(iter/s)": 1.450231 }, { "epoch": 3.115333533267641, "grad_norm": 4.069611549377441, "learning_rate": 3.114674424299416e-05, "loss": 0.5413649082183838, "memory(GiB)": 70.5, "step": 72715, "token_acc": 0.8736462093862816, "train_speed(iter/s)": 1.45024 }, { "epoch": 3.1155477485968897, "grad_norm": 4.943265914916992, "learning_rate": 3.114051138714783e-05, "loss": 0.14831509590148925, "memory(GiB)": 70.5, "step": 72720, "token_acc": 0.9671532846715328, "train_speed(iter/s)": 1.450245 }, { "epoch": 3.1157619639261385, "grad_norm": 7.391343116760254, "learning_rate": 3.1134278872958814e-05, "loss": 0.3400416374206543, "memory(GiB)": 70.5, "step": 72725, "token_acc": 0.9352226720647774, "train_speed(iter/s)": 1.450241 }, { "epoch": 3.1159761792553877, "grad_norm": 3.107956647872925, "learning_rate": 3.112804670054004e-05, "loss": 0.47616009712219237, "memory(GiB)": 70.5, "step": 72730, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.450244 }, { "epoch": 3.1161903945846365, "grad_norm": 3.212742567062378, "learning_rate": 3.1121814870004395e-05, "loss": 0.2833714485168457, "memory(GiB)": 70.5, "step": 72735, "token_acc": 0.9396551724137931, "train_speed(iter/s)": 1.450253 }, { "epoch": 3.1164046099138853, "grad_norm": 3.3170716762542725, "learning_rate": 3.111558338146479e-05, "loss": 0.4174058437347412, "memory(GiB)": 70.5, "step": 72740, "token_acc": 0.9018867924528302, "train_speed(iter/s)": 1.450267 }, { "epoch": 3.1166188252431346, "grad_norm": 0.9381963610649109, "learning_rate": 3.11093522350341e-05, "loss": 0.40324935913085935, "memory(GiB)": 70.5, "step": 72745, "token_acc": 0.9164086687306502, "train_speed(iter/s)": 1.450277 }, { "epoch": 3.1168330405723834, "grad_norm": 2.415817975997925, "learning_rate": 3.1103121430825224e-05, "loss": 0.30118253231048586, "memory(GiB)": 70.5, "step": 72750, "token_acc": 0.9453924914675768, "train_speed(iter/s)": 1.450295 }, { "epoch": 3.1170472559016322, "grad_norm": 0.7563248872756958, "learning_rate": 3.109689096895102e-05, "loss": 0.17914390563964844, "memory(GiB)": 70.5, "step": 72755, "token_acc": 0.9522292993630573, "train_speed(iter/s)": 1.450295 }, { "epoch": 3.1172614712308815, "grad_norm": 4.408317565917969, "learning_rate": 3.109066084952438e-05, "loss": 0.4418308258056641, "memory(GiB)": 70.5, "step": 72760, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.450304 }, { "epoch": 3.1174756865601303, "grad_norm": 4.579488754272461, "learning_rate": 3.1084431072658135e-05, "loss": 0.3818065643310547, "memory(GiB)": 70.5, "step": 72765, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.450308 }, { "epoch": 3.117689901889379, "grad_norm": 4.5747480392456055, "learning_rate": 3.1078201638465164e-05, "loss": 0.519390344619751, "memory(GiB)": 70.5, "step": 72770, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.450306 }, { "epoch": 3.1179041172186284, "grad_norm": 4.979640007019043, "learning_rate": 3.107197254705835e-05, "loss": 0.18536741733551027, "memory(GiB)": 70.5, "step": 72775, "token_acc": 0.9598540145985401, "train_speed(iter/s)": 1.450307 }, { "epoch": 3.118118332547877, "grad_norm": 5.696773529052734, "learning_rate": 3.1065743798550496e-05, "loss": 0.39096047878265383, "memory(GiB)": 70.5, "step": 72780, "token_acc": 0.9054054054054054, "train_speed(iter/s)": 1.450306 }, { "epoch": 3.118332547877126, "grad_norm": 1.9743432998657227, "learning_rate": 3.105951539305445e-05, "loss": 0.3411376476287842, "memory(GiB)": 70.5, "step": 72785, "token_acc": 0.9080882352941176, "train_speed(iter/s)": 1.450307 }, { "epoch": 3.1185467632063752, "grad_norm": 0.8963715434074402, "learning_rate": 3.105328733068306e-05, "loss": 0.3366502523422241, "memory(GiB)": 70.5, "step": 72790, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.45031 }, { "epoch": 3.118760978535624, "grad_norm": 2.5082082748413086, "learning_rate": 3.104705961154914e-05, "loss": 0.46527976989746095, "memory(GiB)": 70.5, "step": 72795, "token_acc": 0.8931297709923665, "train_speed(iter/s)": 1.450312 }, { "epoch": 3.118975193864873, "grad_norm": 4.8278679847717285, "learning_rate": 3.1040832235765506e-05, "loss": 0.1733018636703491, "memory(GiB)": 70.5, "step": 72800, "token_acc": 0.9416909620991254, "train_speed(iter/s)": 1.450318 }, { "epoch": 3.119189409194122, "grad_norm": 3.76668643951416, "learning_rate": 3.103460520344499e-05, "loss": 0.08184295892715454, "memory(GiB)": 70.5, "step": 72805, "token_acc": 0.9776951672862454, "train_speed(iter/s)": 1.450324 }, { "epoch": 3.119403624523371, "grad_norm": 5.733508586883545, "learning_rate": 3.1028378514700386e-05, "loss": 0.321073579788208, "memory(GiB)": 70.5, "step": 72810, "token_acc": 0.9141630901287554, "train_speed(iter/s)": 1.450348 }, { "epoch": 3.1196178398526198, "grad_norm": 3.85040545463562, "learning_rate": 3.1022152169644515e-05, "loss": 0.3996045827865601, "memory(GiB)": 70.5, "step": 72815, "token_acc": 0.8992537313432836, "train_speed(iter/s)": 1.450353 }, { "epoch": 3.119832055181869, "grad_norm": 3.523994207382202, "learning_rate": 3.101592616839015e-05, "loss": 0.3673156976699829, "memory(GiB)": 70.5, "step": 72820, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.450356 }, { "epoch": 3.120046270511118, "grad_norm": 6.474807262420654, "learning_rate": 3.100970051105009e-05, "loss": 0.3396455764770508, "memory(GiB)": 70.5, "step": 72825, "token_acc": 0.8867924528301887, "train_speed(iter/s)": 1.450357 }, { "epoch": 3.1202604858403666, "grad_norm": 4.513150691986084, "learning_rate": 3.1003475197737125e-05, "loss": 0.44489450454711915, "memory(GiB)": 70.5, "step": 72830, "token_acc": 0.8963963963963963, "train_speed(iter/s)": 1.450358 }, { "epoch": 3.120474701169616, "grad_norm": 2.974208354949951, "learning_rate": 3.0997250228564026e-05, "loss": 0.41469011306762693, "memory(GiB)": 70.5, "step": 72835, "token_acc": 0.903010033444816, "train_speed(iter/s)": 1.450364 }, { "epoch": 3.1206889164988647, "grad_norm": 1.383023977279663, "learning_rate": 3.0991025603643556e-05, "loss": 0.2999792814254761, "memory(GiB)": 70.5, "step": 72840, "token_acc": 0.9156118143459916, "train_speed(iter/s)": 1.450359 }, { "epoch": 3.1209031318281135, "grad_norm": 6.577667236328125, "learning_rate": 3.0984801323088484e-05, "loss": 0.5586103916168212, "memory(GiB)": 70.5, "step": 72845, "token_acc": 0.8876404494382022, "train_speed(iter/s)": 1.450363 }, { "epoch": 3.1211173471573628, "grad_norm": 1.7594525814056396, "learning_rate": 3.097857738701159e-05, "loss": 0.29082062244415285, "memory(GiB)": 70.5, "step": 72850, "token_acc": 0.9410029498525073, "train_speed(iter/s)": 1.450369 }, { "epoch": 3.1213315624866116, "grad_norm": 3.543339967727661, "learning_rate": 3.097235379552561e-05, "loss": 0.4700503349304199, "memory(GiB)": 70.5, "step": 72855, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.450385 }, { "epoch": 3.1215457778158604, "grad_norm": 3.732567071914673, "learning_rate": 3.096613054874328e-05, "loss": 0.2522943735122681, "memory(GiB)": 70.5, "step": 72860, "token_acc": 0.9513888888888888, "train_speed(iter/s)": 1.450389 }, { "epoch": 3.1217599931451097, "grad_norm": 1.0825339555740356, "learning_rate": 3.0959907646777364e-05, "loss": 0.3750370740890503, "memory(GiB)": 70.5, "step": 72865, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.450405 }, { "epoch": 3.1219742084743585, "grad_norm": 2.2253336906433105, "learning_rate": 3.0953685089740566e-05, "loss": 0.24552996158599855, "memory(GiB)": 70.5, "step": 72870, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.450411 }, { "epoch": 3.1221884238036073, "grad_norm": 5.245238780975342, "learning_rate": 3.094746287774564e-05, "loss": 0.6150742530822754, "memory(GiB)": 70.5, "step": 72875, "token_acc": 0.8742138364779874, "train_speed(iter/s)": 1.450411 }, { "epoch": 3.1224026391328565, "grad_norm": 3.9274203777313232, "learning_rate": 3.09412410109053e-05, "loss": 0.28340442180633546, "memory(GiB)": 70.5, "step": 72880, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.450411 }, { "epoch": 3.1226168544621054, "grad_norm": 1.9998849630355835, "learning_rate": 3.093501948933225e-05, "loss": 0.3093850612640381, "memory(GiB)": 70.5, "step": 72885, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.450414 }, { "epoch": 3.122831069791354, "grad_norm": 5.870053291320801, "learning_rate": 3.0928798313139206e-05, "loss": 0.49124393463134763, "memory(GiB)": 70.5, "step": 72890, "token_acc": 0.889763779527559, "train_speed(iter/s)": 1.450432 }, { "epoch": 3.1230452851206034, "grad_norm": 3.3960864543914795, "learning_rate": 3.092257748243888e-05, "loss": 0.41794767379760744, "memory(GiB)": 70.5, "step": 72895, "token_acc": 0.9119718309859155, "train_speed(iter/s)": 1.450447 }, { "epoch": 3.1232595004498522, "grad_norm": 3.1232383251190186, "learning_rate": 3.0916356997343945e-05, "loss": 0.5239591121673584, "memory(GiB)": 70.5, "step": 72900, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.450462 }, { "epoch": 3.123473715779101, "grad_norm": 0.6879779696464539, "learning_rate": 3.091013685796712e-05, "loss": 0.07999642491340637, "memory(GiB)": 70.5, "step": 72905, "token_acc": 0.9628252788104089, "train_speed(iter/s)": 1.450463 }, { "epoch": 3.1236879311083503, "grad_norm": 7.630321979522705, "learning_rate": 3.0903917064421064e-05, "loss": 0.3314600944519043, "memory(GiB)": 70.5, "step": 72910, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.450464 }, { "epoch": 3.123902146437599, "grad_norm": 2.8000011444091797, "learning_rate": 3.0897697616818444e-05, "loss": 0.34306514263153076, "memory(GiB)": 70.5, "step": 72915, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.450467 }, { "epoch": 3.124116361766848, "grad_norm": 3.99639630317688, "learning_rate": 3.0891478515271975e-05, "loss": 0.48184618949890134, "memory(GiB)": 70.5, "step": 72920, "token_acc": 0.8970588235294118, "train_speed(iter/s)": 1.450479 }, { "epoch": 3.124330577096097, "grad_norm": 6.115474224090576, "learning_rate": 3.0885259759894306e-05, "loss": 0.28213503360748293, "memory(GiB)": 70.5, "step": 72925, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.450491 }, { "epoch": 3.124544792425346, "grad_norm": 4.501350402832031, "learning_rate": 3.0879041350798064e-05, "loss": 0.6085989952087403, "memory(GiB)": 70.5, "step": 72930, "token_acc": 0.8861788617886179, "train_speed(iter/s)": 1.450513 }, { "epoch": 3.124759007754595, "grad_norm": 5.626298427581787, "learning_rate": 3.0872823288095946e-05, "loss": 0.4808017730712891, "memory(GiB)": 70.5, "step": 72935, "token_acc": 0.8655172413793103, "train_speed(iter/s)": 1.450517 }, { "epoch": 3.124973223083844, "grad_norm": 4.270545482635498, "learning_rate": 3.086660557190057e-05, "loss": 0.32144238948822024, "memory(GiB)": 70.5, "step": 72940, "token_acc": 0.9134078212290503, "train_speed(iter/s)": 1.450518 }, { "epoch": 3.125187438413093, "grad_norm": 3.0900332927703857, "learning_rate": 3.086038820232458e-05, "loss": 0.4965977191925049, "memory(GiB)": 70.5, "step": 72945, "token_acc": 0.9076305220883534, "train_speed(iter/s)": 1.450519 }, { "epoch": 3.1254016537423417, "grad_norm": 2.272613286972046, "learning_rate": 3.085417117948062e-05, "loss": 0.2498617172241211, "memory(GiB)": 70.5, "step": 72950, "token_acc": 0.9588014981273408, "train_speed(iter/s)": 1.450521 }, { "epoch": 3.125615869071591, "grad_norm": 1.907772421836853, "learning_rate": 3.08479545034813e-05, "loss": 0.40628862380981445, "memory(GiB)": 70.5, "step": 72955, "token_acc": 0.90625, "train_speed(iter/s)": 1.450523 }, { "epoch": 3.1258300844008398, "grad_norm": 2.321406841278076, "learning_rate": 3.084173817443925e-05, "loss": 0.2153618812561035, "memory(GiB)": 70.5, "step": 72960, "token_acc": 0.9552238805970149, "train_speed(iter/s)": 1.450524 }, { "epoch": 3.1260442997300886, "grad_norm": 4.044538974761963, "learning_rate": 3.0835522192467095e-05, "loss": 0.32226722240447997, "memory(GiB)": 70.5, "step": 72965, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.450543 }, { "epoch": 3.126258515059338, "grad_norm": 3.2536160945892334, "learning_rate": 3.082930655767742e-05, "loss": 0.39254074096679686, "memory(GiB)": 70.5, "step": 72970, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.450553 }, { "epoch": 3.1264727303885866, "grad_norm": 1.6288875341415405, "learning_rate": 3.082309127018285e-05, "loss": 0.17521071434020996, "memory(GiB)": 70.5, "step": 72975, "token_acc": 0.952076677316294, "train_speed(iter/s)": 1.450557 }, { "epoch": 3.1266869457178355, "grad_norm": 2.8429439067840576, "learning_rate": 3.081687633009598e-05, "loss": 0.5452378749847412, "memory(GiB)": 70.5, "step": 72980, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.450572 }, { "epoch": 3.1269011610470847, "grad_norm": 6.554471492767334, "learning_rate": 3.081066173752937e-05, "loss": 0.6130585670471191, "memory(GiB)": 70.5, "step": 72985, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.450593 }, { "epoch": 3.1271153763763335, "grad_norm": 0.8267459273338318, "learning_rate": 3.080444749259561e-05, "loss": 0.21957619190216066, "memory(GiB)": 70.5, "step": 72990, "token_acc": 0.9546742209631728, "train_speed(iter/s)": 1.4506 }, { "epoch": 3.1273295917055823, "grad_norm": 4.8592095375061035, "learning_rate": 3.079823359540732e-05, "loss": 0.332131552696228, "memory(GiB)": 70.5, "step": 72995, "token_acc": 0.9320754716981132, "train_speed(iter/s)": 1.450607 }, { "epoch": 3.1275438070348316, "grad_norm": 6.665001392364502, "learning_rate": 3.079202004607704e-05, "loss": 0.5168676853179932, "memory(GiB)": 70.5, "step": 73000, "token_acc": 0.8628048780487805, "train_speed(iter/s)": 1.450605 }, { "epoch": 3.1275438070348316, "eval_loss": 2.5810251235961914, "eval_runtime": 11.9218, "eval_samples_per_second": 8.388, "eval_steps_per_second": 8.388, "eval_token_acc": 0.4550989345509893, "step": 73000 }, { "epoch": 3.1277580223640804, "grad_norm": 6.783167362213135, "learning_rate": 3.0785806844717324e-05, "loss": 0.7738393306732178, "memory(GiB)": 70.5, "step": 73005, "token_acc": 0.5628476084538376, "train_speed(iter/s)": 1.450194 }, { "epoch": 3.127972237693329, "grad_norm": 4.068376064300537, "learning_rate": 3.077959399144075e-05, "loss": 0.4600196838378906, "memory(GiB)": 70.5, "step": 73010, "token_acc": 0.8875, "train_speed(iter/s)": 1.450204 }, { "epoch": 3.1281864530225785, "grad_norm": 0.2856307029724121, "learning_rate": 3.0773381486359854e-05, "loss": 0.4653192043304443, "memory(GiB)": 70.5, "step": 73015, "token_acc": 0.9058441558441559, "train_speed(iter/s)": 1.450205 }, { "epoch": 3.1284006683518273, "grad_norm": 3.166783571243286, "learning_rate": 3.0767169329587195e-05, "loss": 0.2546437978744507, "memory(GiB)": 70.5, "step": 73020, "token_acc": 0.9480968858131488, "train_speed(iter/s)": 1.450209 }, { "epoch": 3.128614883681076, "grad_norm": 3.788447141647339, "learning_rate": 3.07609575212353e-05, "loss": 0.2858676195144653, "memory(GiB)": 70.5, "step": 73025, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.450225 }, { "epoch": 3.1288290990103254, "grad_norm": 4.101943492889404, "learning_rate": 3.0754746061416704e-05, "loss": 0.3190129041671753, "memory(GiB)": 70.5, "step": 73030, "token_acc": 0.9271137026239067, "train_speed(iter/s)": 1.450224 }, { "epoch": 3.129043314339574, "grad_norm": 2.403258800506592, "learning_rate": 3.074853495024395e-05, "loss": 0.35803685188293455, "memory(GiB)": 70.5, "step": 73035, "token_acc": 0.9330708661417323, "train_speed(iter/s)": 1.450226 }, { "epoch": 3.129257529668823, "grad_norm": 2.837510824203491, "learning_rate": 3.074232418782954e-05, "loss": 0.5214105129241944, "memory(GiB)": 70.5, "step": 73040, "token_acc": 0.8757763975155279, "train_speed(iter/s)": 1.450245 }, { "epoch": 3.1294717449980722, "grad_norm": 1.8691452741622925, "learning_rate": 3.0736113774285984e-05, "loss": 0.26155211925506594, "memory(GiB)": 70.5, "step": 73045, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.450256 }, { "epoch": 3.129685960327321, "grad_norm": 4.001856327056885, "learning_rate": 3.07299037097258e-05, "loss": 0.4160614013671875, "memory(GiB)": 70.5, "step": 73050, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.450264 }, { "epoch": 3.12990017565657, "grad_norm": 5.274770259857178, "learning_rate": 3.0723693994261496e-05, "loss": 0.33020195960998533, "memory(GiB)": 70.5, "step": 73055, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.450275 }, { "epoch": 3.130114390985819, "grad_norm": 4.006399154663086, "learning_rate": 3.0717484628005556e-05, "loss": 0.5194272994995117, "memory(GiB)": 70.5, "step": 73060, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.450295 }, { "epoch": 3.130328606315068, "grad_norm": 3.2292065620422363, "learning_rate": 3.071127561107044e-05, "loss": 0.5140131950378418, "memory(GiB)": 70.5, "step": 73065, "token_acc": 0.8849557522123894, "train_speed(iter/s)": 1.450303 }, { "epoch": 3.1305428216443167, "grad_norm": 6.062214374542236, "learning_rate": 3.07050669435687e-05, "loss": 0.49599852561950686, "memory(GiB)": 70.5, "step": 73070, "token_acc": 0.8878504672897196, "train_speed(iter/s)": 1.450303 }, { "epoch": 3.130757036973566, "grad_norm": 4.843158721923828, "learning_rate": 3.069885862561275e-05, "loss": 0.24959535598754884, "memory(GiB)": 70.5, "step": 73075, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.450308 }, { "epoch": 3.130971252302815, "grad_norm": 4.233370304107666, "learning_rate": 3.0692650657315106e-05, "loss": 0.367972207069397, "memory(GiB)": 70.5, "step": 73080, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.450306 }, { "epoch": 3.1311854676320636, "grad_norm": 2.239231824874878, "learning_rate": 3.0686443038788196e-05, "loss": 0.21105008125305175, "memory(GiB)": 70.5, "step": 73085, "token_acc": 0.9562043795620438, "train_speed(iter/s)": 1.450311 }, { "epoch": 3.131399682961313, "grad_norm": 4.2564616203308105, "learning_rate": 3.0680235770144474e-05, "loss": 0.2840015172958374, "memory(GiB)": 70.5, "step": 73090, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.450315 }, { "epoch": 3.1316138982905617, "grad_norm": 4.725430965423584, "learning_rate": 3.067402885149642e-05, "loss": 0.44568214416503904, "memory(GiB)": 70.5, "step": 73095, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.450327 }, { "epoch": 3.1318281136198105, "grad_norm": 2.237497091293335, "learning_rate": 3.066782228295645e-05, "loss": 0.34843735694885253, "memory(GiB)": 70.5, "step": 73100, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.450338 }, { "epoch": 3.1320423289490598, "grad_norm": 1.1672230958938599, "learning_rate": 3.0661616064637025e-05, "loss": 0.2607722282409668, "memory(GiB)": 70.5, "step": 73105, "token_acc": 0.9526813880126183, "train_speed(iter/s)": 1.45035 }, { "epoch": 3.1322565442783086, "grad_norm": 2.652182102203369, "learning_rate": 3.065541019665057e-05, "loss": 0.5748930454254151, "memory(GiB)": 70.5, "step": 73110, "token_acc": 0.872852233676976, "train_speed(iter/s)": 1.450355 }, { "epoch": 3.1324707596075574, "grad_norm": 4.933310508728027, "learning_rate": 3.0649204679109494e-05, "loss": 0.2662128686904907, "memory(GiB)": 70.5, "step": 73115, "token_acc": 0.932, "train_speed(iter/s)": 1.450369 }, { "epoch": 3.1326849749368066, "grad_norm": 0.16147381067276, "learning_rate": 3.064299951212624e-05, "loss": 0.3835946798324585, "memory(GiB)": 70.5, "step": 73120, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.45038 }, { "epoch": 3.1328991902660555, "grad_norm": 0.9958739280700684, "learning_rate": 3.06367946958132e-05, "loss": 0.2900132894515991, "memory(GiB)": 70.5, "step": 73125, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.450382 }, { "epoch": 3.1331134055953043, "grad_norm": 1.5248829126358032, "learning_rate": 3.0630590230282783e-05, "loss": 0.3754369020462036, "memory(GiB)": 70.5, "step": 73130, "token_acc": 0.9212328767123288, "train_speed(iter/s)": 1.450396 }, { "epoch": 3.1333276209245535, "grad_norm": 1.147213101387024, "learning_rate": 3.0624386115647386e-05, "loss": 0.2668358087539673, "memory(GiB)": 70.5, "step": 73135, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.450404 }, { "epoch": 3.1335418362538023, "grad_norm": 2.587137460708618, "learning_rate": 3.0618182352019434e-05, "loss": 0.4467640399932861, "memory(GiB)": 70.5, "step": 73140, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.450413 }, { "epoch": 3.133756051583051, "grad_norm": 3.2139840126037598, "learning_rate": 3.061197893951128e-05, "loss": 0.4436334133148193, "memory(GiB)": 70.5, "step": 73145, "token_acc": 0.900990099009901, "train_speed(iter/s)": 1.450413 }, { "epoch": 3.1339702669123004, "grad_norm": 1.9677220582962036, "learning_rate": 3.060577587823531e-05, "loss": 0.1749498128890991, "memory(GiB)": 70.5, "step": 73150, "token_acc": 0.967032967032967, "train_speed(iter/s)": 1.450417 }, { "epoch": 3.134184482241549, "grad_norm": 1.1829649209976196, "learning_rate": 3.059957316830391e-05, "loss": 0.4338232040405273, "memory(GiB)": 70.5, "step": 73155, "token_acc": 0.9148264984227129, "train_speed(iter/s)": 1.45042 }, { "epoch": 3.134398697570798, "grad_norm": 2.258877754211426, "learning_rate": 3.059337080982943e-05, "loss": 0.3069628715515137, "memory(GiB)": 70.5, "step": 73160, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.450426 }, { "epoch": 3.1346129129000473, "grad_norm": 1.2364379167556763, "learning_rate": 3.0587168802924266e-05, "loss": 0.1916717529296875, "memory(GiB)": 70.5, "step": 73165, "token_acc": 0.9613899613899614, "train_speed(iter/s)": 1.450437 }, { "epoch": 3.134827128229296, "grad_norm": 4.709557056427002, "learning_rate": 3.058096714770074e-05, "loss": 0.6890845775604248, "memory(GiB)": 70.5, "step": 73170, "token_acc": 0.8717948717948718, "train_speed(iter/s)": 1.450444 }, { "epoch": 3.135041343558545, "grad_norm": 4.269022464752197, "learning_rate": 3.05747658442712e-05, "loss": 0.3099186658859253, "memory(GiB)": 70.5, "step": 73175, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.450447 }, { "epoch": 3.135255558887794, "grad_norm": 6.7904953956604, "learning_rate": 3.0568564892748005e-05, "loss": 0.3615701913833618, "memory(GiB)": 70.5, "step": 73180, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.450464 }, { "epoch": 3.135469774217043, "grad_norm": 0.31687837839126587, "learning_rate": 3.0562364293243496e-05, "loss": 0.2260668992996216, "memory(GiB)": 70.5, "step": 73185, "token_acc": 0.9485530546623794, "train_speed(iter/s)": 1.450468 }, { "epoch": 3.135683989546292, "grad_norm": 1.5993090867996216, "learning_rate": 3.055616404586998e-05, "loss": 0.3232187032699585, "memory(GiB)": 70.5, "step": 73190, "token_acc": 0.9433198380566802, "train_speed(iter/s)": 1.450475 }, { "epoch": 3.135898204875541, "grad_norm": 4.604470729827881, "learning_rate": 3.05499641507398e-05, "loss": 0.34834249019622804, "memory(GiB)": 70.5, "step": 73195, "token_acc": 0.9152046783625731, "train_speed(iter/s)": 1.450472 }, { "epoch": 3.13611242020479, "grad_norm": 5.736226558685303, "learning_rate": 3.0543764607965256e-05, "loss": 0.3907604694366455, "memory(GiB)": 70.5, "step": 73200, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.450477 }, { "epoch": 3.1363266355340387, "grad_norm": 2.345639705657959, "learning_rate": 3.053756541765867e-05, "loss": 0.39138197898864746, "memory(GiB)": 70.5, "step": 73205, "token_acc": 0.9176136363636364, "train_speed(iter/s)": 1.450481 }, { "epoch": 3.136540850863288, "grad_norm": 2.236595392227173, "learning_rate": 3.053136657993233e-05, "loss": 0.2813680410385132, "memory(GiB)": 70.5, "step": 73210, "token_acc": 0.9496124031007752, "train_speed(iter/s)": 1.450488 }, { "epoch": 3.1367550661925367, "grad_norm": 1.4574528932571411, "learning_rate": 3.0525168094898566e-05, "loss": 0.2037731170654297, "memory(GiB)": 70.5, "step": 73215, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.450494 }, { "epoch": 3.1369692815217856, "grad_norm": 2.1806142330169678, "learning_rate": 3.051896996266964e-05, "loss": 0.23568975925445557, "memory(GiB)": 70.5, "step": 73220, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.450503 }, { "epoch": 3.137183496851035, "grad_norm": 3.4154744148254395, "learning_rate": 3.0512772183357856e-05, "loss": 0.3896113634109497, "memory(GiB)": 70.5, "step": 73225, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.450511 }, { "epoch": 3.1373977121802836, "grad_norm": 3.0360543727874756, "learning_rate": 3.050657475707549e-05, "loss": 0.24932920932769775, "memory(GiB)": 70.5, "step": 73230, "token_acc": 0.9343629343629344, "train_speed(iter/s)": 1.450518 }, { "epoch": 3.1376119275095324, "grad_norm": 1.3091641664505005, "learning_rate": 3.0500377683934796e-05, "loss": 0.18758909702301024, "memory(GiB)": 70.5, "step": 73235, "token_acc": 0.9517684887459807, "train_speed(iter/s)": 1.450526 }, { "epoch": 3.1378261428387817, "grad_norm": 7.315410614013672, "learning_rate": 3.0494180964048057e-05, "loss": 0.2637534856796265, "memory(GiB)": 70.5, "step": 73240, "token_acc": 0.9515151515151515, "train_speed(iter/s)": 1.450528 }, { "epoch": 3.1380403581680305, "grad_norm": 2.780618190765381, "learning_rate": 3.048798459752753e-05, "loss": 0.351548171043396, "memory(GiB)": 70.5, "step": 73245, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.450532 }, { "epoch": 3.1382545734972793, "grad_norm": 1.0494658946990967, "learning_rate": 3.0481788584485467e-05, "loss": 0.12470699548721313, "memory(GiB)": 70.5, "step": 73250, "token_acc": 0.9648562300319489, "train_speed(iter/s)": 1.450535 }, { "epoch": 3.1384687888265286, "grad_norm": 2.847287654876709, "learning_rate": 3.0475592925034112e-05, "loss": 0.5340160369873047, "memory(GiB)": 70.5, "step": 73255, "token_acc": 0.8927444794952681, "train_speed(iter/s)": 1.450551 }, { "epoch": 3.1386830041557774, "grad_norm": 0.7313413619995117, "learning_rate": 3.04693976192857e-05, "loss": 0.6416889190673828, "memory(GiB)": 70.5, "step": 73260, "token_acc": 0.889589905362776, "train_speed(iter/s)": 1.450559 }, { "epoch": 3.138897219485026, "grad_norm": 2.1597070693969727, "learning_rate": 3.0463202667352487e-05, "loss": 0.24860281944274903, "memory(GiB)": 70.5, "step": 73265, "token_acc": 0.9338235294117647, "train_speed(iter/s)": 1.450576 }, { "epoch": 3.1391114348142755, "grad_norm": 4.64398717880249, "learning_rate": 3.0457008069346677e-05, "loss": 0.3473460912704468, "memory(GiB)": 70.5, "step": 73270, "token_acc": 0.94140625, "train_speed(iter/s)": 1.450582 }, { "epoch": 3.1393256501435243, "grad_norm": 3.058621883392334, "learning_rate": 3.0450813825380487e-05, "loss": 0.26201653480529785, "memory(GiB)": 70.5, "step": 73275, "token_acc": 0.94, "train_speed(iter/s)": 1.450583 }, { "epoch": 3.139539865472773, "grad_norm": 6.2876458168029785, "learning_rate": 3.044461993556616e-05, "loss": 0.36849424839019773, "memory(GiB)": 70.5, "step": 73280, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.450588 }, { "epoch": 3.1397540808020223, "grad_norm": 0.17465870082378387, "learning_rate": 3.0438426400015864e-05, "loss": 0.0995989203453064, "memory(GiB)": 70.5, "step": 73285, "token_acc": 0.9754385964912281, "train_speed(iter/s)": 1.450593 }, { "epoch": 3.139968296131271, "grad_norm": 2.7094318866729736, "learning_rate": 3.0432233218841845e-05, "loss": 0.4995311737060547, "memory(GiB)": 70.5, "step": 73290, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.450611 }, { "epoch": 3.14018251146052, "grad_norm": 3.9998385906219482, "learning_rate": 3.0426040392156264e-05, "loss": 0.35295977592468264, "memory(GiB)": 70.5, "step": 73295, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.450624 }, { "epoch": 3.140396726789769, "grad_norm": 2.2830772399902344, "learning_rate": 3.0419847920071333e-05, "loss": 0.13942337036132812, "memory(GiB)": 70.5, "step": 73300, "token_acc": 0.9710610932475884, "train_speed(iter/s)": 1.45063 }, { "epoch": 3.140610942119018, "grad_norm": 3.15389347076416, "learning_rate": 3.0413655802699216e-05, "loss": 0.3991215705871582, "memory(GiB)": 70.5, "step": 73305, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.45063 }, { "epoch": 3.140825157448267, "grad_norm": 1.05165696144104, "learning_rate": 3.040746404015211e-05, "loss": 0.28418607711791993, "memory(GiB)": 70.5, "step": 73310, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.450629 }, { "epoch": 3.141039372777516, "grad_norm": 1.9944770336151123, "learning_rate": 3.0401272632542172e-05, "loss": 0.30993826389312745, "memory(GiB)": 70.5, "step": 73315, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.450632 }, { "epoch": 3.141253588106765, "grad_norm": 1.8868454694747925, "learning_rate": 3.0395081579981556e-05, "loss": 0.645773458480835, "memory(GiB)": 70.5, "step": 73320, "token_acc": 0.8745247148288974, "train_speed(iter/s)": 1.450644 }, { "epoch": 3.1414678034360137, "grad_norm": 1.9824943542480469, "learning_rate": 3.038889088258243e-05, "loss": 0.5418727397918701, "memory(GiB)": 70.5, "step": 73325, "token_acc": 0.8850267379679144, "train_speed(iter/s)": 1.450648 }, { "epoch": 3.141682018765263, "grad_norm": 0.08671123534440994, "learning_rate": 3.0382700540456954e-05, "loss": 0.30212674140930174, "memory(GiB)": 70.5, "step": 73330, "token_acc": 0.9351032448377581, "train_speed(iter/s)": 1.450649 }, { "epoch": 3.141896234094512, "grad_norm": 3.9832262992858887, "learning_rate": 3.0376510553717242e-05, "loss": 0.6857322216033935, "memory(GiB)": 70.5, "step": 73335, "token_acc": 0.861878453038674, "train_speed(iter/s)": 1.450649 }, { "epoch": 3.1421104494237606, "grad_norm": 0.041135769337415695, "learning_rate": 3.0370320922475466e-05, "loss": 0.1752207636833191, "memory(GiB)": 70.5, "step": 73340, "token_acc": 0.9614035087719298, "train_speed(iter/s)": 1.450654 }, { "epoch": 3.14232466475301, "grad_norm": 4.746490478515625, "learning_rate": 3.0364131646843725e-05, "loss": 0.5932612895965577, "memory(GiB)": 70.5, "step": 73345, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.450669 }, { "epoch": 3.1425388800822587, "grad_norm": 3.511265993118286, "learning_rate": 3.0357942726934173e-05, "loss": 0.3503153324127197, "memory(GiB)": 70.5, "step": 73350, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.450668 }, { "epoch": 3.1427530954115075, "grad_norm": 3.113703966140747, "learning_rate": 3.0351754162858913e-05, "loss": 0.3415482521057129, "memory(GiB)": 70.5, "step": 73355, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.450666 }, { "epoch": 3.1429673107407567, "grad_norm": 3.748004674911499, "learning_rate": 3.0345565954730036e-05, "loss": 0.30535140037536623, "memory(GiB)": 70.5, "step": 73360, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.450672 }, { "epoch": 3.1431815260700056, "grad_norm": 6.317716121673584, "learning_rate": 3.0339378102659678e-05, "loss": 0.39778842926025393, "memory(GiB)": 70.5, "step": 73365, "token_acc": 0.906832298136646, "train_speed(iter/s)": 1.450669 }, { "epoch": 3.1433957413992544, "grad_norm": 4.259237289428711, "learning_rate": 3.033319060675994e-05, "loss": 0.3080374479293823, "memory(GiB)": 70.5, "step": 73370, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.450682 }, { "epoch": 3.1436099567285036, "grad_norm": 8.837984085083008, "learning_rate": 3.0327003467142907e-05, "loss": 0.2820885181427002, "memory(GiB)": 70.5, "step": 73375, "token_acc": 0.9363957597173145, "train_speed(iter/s)": 1.450683 }, { "epoch": 3.1438241720577524, "grad_norm": 2.6730823516845703, "learning_rate": 3.032081668392065e-05, "loss": 0.23891620635986327, "memory(GiB)": 70.5, "step": 73380, "token_acc": 0.9522292993630573, "train_speed(iter/s)": 1.450679 }, { "epoch": 3.1440383873870013, "grad_norm": 2.6357474327087402, "learning_rate": 3.0314630257205266e-05, "loss": 0.2547128200531006, "memory(GiB)": 70.5, "step": 73385, "token_acc": 0.9432624113475178, "train_speed(iter/s)": 1.450684 }, { "epoch": 3.1442526027162505, "grad_norm": 4.042063236236572, "learning_rate": 3.030844418710881e-05, "loss": 0.18470441102981566, "memory(GiB)": 70.5, "step": 73390, "token_acc": 0.9598214285714286, "train_speed(iter/s)": 1.450692 }, { "epoch": 3.1444668180454993, "grad_norm": 1.6373586654663086, "learning_rate": 3.0302258473743374e-05, "loss": 0.4262528896331787, "memory(GiB)": 70.5, "step": 73395, "token_acc": 0.9110169491525424, "train_speed(iter/s)": 1.450707 }, { "epoch": 3.144681033374748, "grad_norm": 1.706063151359558, "learning_rate": 3.0296073117221004e-05, "loss": 0.3580707311630249, "memory(GiB)": 70.5, "step": 73400, "token_acc": 0.9264069264069265, "train_speed(iter/s)": 1.450706 }, { "epoch": 3.1448952487039974, "grad_norm": 3.957083225250244, "learning_rate": 3.028988811765374e-05, "loss": 0.22337806224822998, "memory(GiB)": 70.5, "step": 73405, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.45072 }, { "epoch": 3.145109464033246, "grad_norm": 5.195244789123535, "learning_rate": 3.0283703475153656e-05, "loss": 0.3568262577056885, "memory(GiB)": 70.5, "step": 73410, "token_acc": 0.9372822299651568, "train_speed(iter/s)": 1.450724 }, { "epoch": 3.145323679362495, "grad_norm": 2.08925199508667, "learning_rate": 3.0277519189832777e-05, "loss": 0.35995566844940186, "memory(GiB)": 70.5, "step": 73415, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.450722 }, { "epoch": 3.1455378946917443, "grad_norm": 0.9142557978630066, "learning_rate": 3.0271335261803124e-05, "loss": 0.4572728157043457, "memory(GiB)": 70.5, "step": 73420, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.450741 }, { "epoch": 3.145752110020993, "grad_norm": 2.9426679611206055, "learning_rate": 3.0265151691176756e-05, "loss": 0.5012942790985108, "memory(GiB)": 70.5, "step": 73425, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.45074 }, { "epoch": 3.145966325350242, "grad_norm": 3.5887629985809326, "learning_rate": 3.0258968478065665e-05, "loss": 0.25911214351654055, "memory(GiB)": 70.5, "step": 73430, "token_acc": 0.963963963963964, "train_speed(iter/s)": 1.450747 }, { "epoch": 3.146180540679491, "grad_norm": 4.508548259735107, "learning_rate": 3.025278562258188e-05, "loss": 0.4044854164123535, "memory(GiB)": 70.5, "step": 73435, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.450745 }, { "epoch": 3.14639475600874, "grad_norm": 2.3363609313964844, "learning_rate": 3.0246603124837414e-05, "loss": 0.38802714347839357, "memory(GiB)": 70.5, "step": 73440, "token_acc": 0.9105431309904153, "train_speed(iter/s)": 1.450746 }, { "epoch": 3.146608971337989, "grad_norm": 1.9065219163894653, "learning_rate": 3.024042098494426e-05, "loss": 0.24788985252380372, "memory(GiB)": 70.5, "step": 73445, "token_acc": 0.953307392996109, "train_speed(iter/s)": 1.450754 }, { "epoch": 3.146823186667238, "grad_norm": 2.2733170986175537, "learning_rate": 3.0234239203014413e-05, "loss": 0.34004998207092285, "memory(GiB)": 70.5, "step": 73450, "token_acc": 0.9296636085626911, "train_speed(iter/s)": 1.450768 }, { "epoch": 3.147037401996487, "grad_norm": 5.217135906219482, "learning_rate": 3.022805777915988e-05, "loss": 0.3126883029937744, "memory(GiB)": 70.5, "step": 73455, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.450772 }, { "epoch": 3.1472516173257357, "grad_norm": 0.6823260188102722, "learning_rate": 3.022187671349262e-05, "loss": 0.34251060485839846, "memory(GiB)": 70.5, "step": 73460, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.450779 }, { "epoch": 3.147465832654985, "grad_norm": 2.320727586746216, "learning_rate": 3.0215696006124606e-05, "loss": 0.2734535694122314, "memory(GiB)": 70.5, "step": 73465, "token_acc": 0.942652329749104, "train_speed(iter/s)": 1.450786 }, { "epoch": 3.1476800479842337, "grad_norm": 2.200155258178711, "learning_rate": 3.020951565716783e-05, "loss": 0.22131986618041993, "memory(GiB)": 70.5, "step": 73470, "token_acc": 0.9580645161290322, "train_speed(iter/s)": 1.450792 }, { "epoch": 3.1478942633134825, "grad_norm": 5.986841201782227, "learning_rate": 3.0203335666734244e-05, "loss": 0.7493361473083496, "memory(GiB)": 70.5, "step": 73475, "token_acc": 0.8836363636363637, "train_speed(iter/s)": 1.450805 }, { "epoch": 3.148108478642732, "grad_norm": 3.67688250541687, "learning_rate": 3.0197156034935792e-05, "loss": 0.3006605625152588, "memory(GiB)": 70.5, "step": 73480, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.450816 }, { "epoch": 3.1483226939719806, "grad_norm": 6.450192928314209, "learning_rate": 3.019097676188445e-05, "loss": 0.5056018352508544, "memory(GiB)": 70.5, "step": 73485, "token_acc": 0.9046153846153846, "train_speed(iter/s)": 1.45083 }, { "epoch": 3.1485369093012294, "grad_norm": 1.5776317119598389, "learning_rate": 3.0184797847692126e-05, "loss": 0.3478051424026489, "memory(GiB)": 70.5, "step": 73490, "token_acc": 0.9298701298701298, "train_speed(iter/s)": 1.450842 }, { "epoch": 3.1487511246304787, "grad_norm": 3.2061493396759033, "learning_rate": 3.01786192924708e-05, "loss": 0.27843284606933594, "memory(GiB)": 70.5, "step": 73495, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.450841 }, { "epoch": 3.1489653399597275, "grad_norm": 3.5802841186523438, "learning_rate": 3.017244109633237e-05, "loss": 0.39336309432983396, "memory(GiB)": 70.5, "step": 73500, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.450856 }, { "epoch": 3.1489653399597275, "eval_loss": 2.4917232990264893, "eval_runtime": 11.2011, "eval_samples_per_second": 8.928, "eval_steps_per_second": 8.928, "eval_token_acc": 0.44933920704845814, "step": 73500 }, { "epoch": 3.1491795552889763, "grad_norm": 4.157712936401367, "learning_rate": 3.016626325938875e-05, "loss": 0.4980586051940918, "memory(GiB)": 70.5, "step": 73505, "token_acc": 0.5856410256410256, "train_speed(iter/s)": 1.450503 }, { "epoch": 3.1493937706182256, "grad_norm": 2.818147659301758, "learning_rate": 3.016008578175189e-05, "loss": 0.3917638063430786, "memory(GiB)": 70.5, "step": 73510, "token_acc": 0.9043478260869565, "train_speed(iter/s)": 1.450497 }, { "epoch": 3.1496079859474744, "grad_norm": 6.821324825286865, "learning_rate": 3.0153908663533693e-05, "loss": 0.624757957458496, "memory(GiB)": 70.5, "step": 73515, "token_acc": 0.8803680981595092, "train_speed(iter/s)": 1.450501 }, { "epoch": 3.149822201276723, "grad_norm": 3.412435293197632, "learning_rate": 3.0147731904846067e-05, "loss": 0.41910762786865235, "memory(GiB)": 70.5, "step": 73520, "token_acc": 0.9153605015673981, "train_speed(iter/s)": 1.450504 }, { "epoch": 3.1500364166059724, "grad_norm": 4.228204727172852, "learning_rate": 3.014155550580088e-05, "loss": 0.38220288753509524, "memory(GiB)": 70.5, "step": 73525, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.450506 }, { "epoch": 3.1502506319352213, "grad_norm": 2.0454745292663574, "learning_rate": 3.0135379466510062e-05, "loss": 0.2625147581100464, "memory(GiB)": 70.5, "step": 73530, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.450514 }, { "epoch": 3.15046484726447, "grad_norm": 2.432882308959961, "learning_rate": 3.012920378708546e-05, "loss": 0.47804789543151854, "memory(GiB)": 70.5, "step": 73535, "token_acc": 0.8875, "train_speed(iter/s)": 1.450538 }, { "epoch": 3.1506790625937193, "grad_norm": 3.553889751434326, "learning_rate": 3.0123028467638992e-05, "loss": 0.19413448572158815, "memory(GiB)": 70.5, "step": 73540, "token_acc": 0.9552238805970149, "train_speed(iter/s)": 1.450545 }, { "epoch": 3.150893277922968, "grad_norm": 6.603944301605225, "learning_rate": 3.01168535082825e-05, "loss": 0.38190770149230957, "memory(GiB)": 70.5, "step": 73545, "token_acc": 0.9177215189873418, "train_speed(iter/s)": 1.450548 }, { "epoch": 3.151107493252217, "grad_norm": 3.3246052265167236, "learning_rate": 3.0110678909127855e-05, "loss": 0.4444310665130615, "memory(GiB)": 70.5, "step": 73550, "token_acc": 0.9055793991416309, "train_speed(iter/s)": 1.450551 }, { "epoch": 3.151321708581466, "grad_norm": 2.0558338165283203, "learning_rate": 3.0104504670286926e-05, "loss": 0.4156855583190918, "memory(GiB)": 70.5, "step": 73555, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.450566 }, { "epoch": 3.151535923910715, "grad_norm": 3.8383142948150635, "learning_rate": 3.009833079187156e-05, "loss": 0.2668982982635498, "memory(GiB)": 70.5, "step": 73560, "token_acc": 0.9323076923076923, "train_speed(iter/s)": 1.450582 }, { "epoch": 3.151750139239964, "grad_norm": 3.245245933532715, "learning_rate": 3.009215727399359e-05, "loss": 0.48227458000183104, "memory(GiB)": 70.5, "step": 73565, "token_acc": 0.8906752411575563, "train_speed(iter/s)": 1.450584 }, { "epoch": 3.151964354569213, "grad_norm": 2.4862964153289795, "learning_rate": 3.0085984116764877e-05, "loss": 0.44351348876953123, "memory(GiB)": 70.5, "step": 73570, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.450588 }, { "epoch": 3.152178569898462, "grad_norm": 4.466121673583984, "learning_rate": 3.0079811320297235e-05, "loss": 0.5630940914154052, "memory(GiB)": 70.5, "step": 73575, "token_acc": 0.8784313725490196, "train_speed(iter/s)": 1.450603 }, { "epoch": 3.1523927852277107, "grad_norm": 1.610608458518982, "learning_rate": 3.007363888470249e-05, "loss": 0.39224722385406496, "memory(GiB)": 70.5, "step": 73580, "token_acc": 0.9119496855345912, "train_speed(iter/s)": 1.450605 }, { "epoch": 3.15260700055696, "grad_norm": 3.328469753265381, "learning_rate": 3.0067466810092475e-05, "loss": 0.3280014514923096, "memory(GiB)": 70.5, "step": 73585, "token_acc": 0.913946587537092, "train_speed(iter/s)": 1.450608 }, { "epoch": 3.152821215886209, "grad_norm": 1.92730712890625, "learning_rate": 3.0061295096579008e-05, "loss": 0.13420588970184327, "memory(GiB)": 70.5, "step": 73590, "token_acc": 0.963855421686747, "train_speed(iter/s)": 1.450622 }, { "epoch": 3.1530354312154576, "grad_norm": 5.753725051879883, "learning_rate": 3.005512374427387e-05, "loss": 0.4218998908996582, "memory(GiB)": 70.5, "step": 73595, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.450624 }, { "epoch": 3.153249646544707, "grad_norm": 1.7879010438919067, "learning_rate": 3.0048952753288882e-05, "loss": 0.5176285743713379, "memory(GiB)": 70.5, "step": 73600, "token_acc": 0.898876404494382, "train_speed(iter/s)": 1.45062 }, { "epoch": 3.1534638618739557, "grad_norm": 0.7984970808029175, "learning_rate": 3.004278212373584e-05, "loss": 0.39300785064697263, "memory(GiB)": 70.5, "step": 73605, "token_acc": 0.9003690036900369, "train_speed(iter/s)": 1.450623 }, { "epoch": 3.1536780772032045, "grad_norm": 3.8859994411468506, "learning_rate": 3.003661185572651e-05, "loss": 0.6208965301513671, "memory(GiB)": 70.5, "step": 73610, "token_acc": 0.8765060240963856, "train_speed(iter/s)": 1.45063 }, { "epoch": 3.1538922925324537, "grad_norm": 7.594021320343018, "learning_rate": 3.003044194937269e-05, "loss": 0.5453869342803955, "memory(GiB)": 70.5, "step": 73615, "token_acc": 0.8916083916083916, "train_speed(iter/s)": 1.450629 }, { "epoch": 3.1541065078617025, "grad_norm": 1.8504774570465088, "learning_rate": 3.0024272404786146e-05, "loss": 0.37041985988616943, "memory(GiB)": 70.5, "step": 73620, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.450635 }, { "epoch": 3.1543207231909514, "grad_norm": 3.896853446960449, "learning_rate": 3.0018103222078643e-05, "loss": 0.27727532386779785, "memory(GiB)": 70.5, "step": 73625, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.450632 }, { "epoch": 3.1545349385202006, "grad_norm": 1.4698309898376465, "learning_rate": 3.001193440136194e-05, "loss": 0.2592029571533203, "memory(GiB)": 70.5, "step": 73630, "token_acc": 0.9375, "train_speed(iter/s)": 1.450628 }, { "epoch": 3.1547491538494494, "grad_norm": 1.2863587141036987, "learning_rate": 3.00057659427478e-05, "loss": 0.4293821811676025, "memory(GiB)": 70.5, "step": 73635, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.450633 }, { "epoch": 3.1549633691786982, "grad_norm": 4.299642086029053, "learning_rate": 2.999959784634797e-05, "loss": 0.298313570022583, "memory(GiB)": 70.5, "step": 73640, "token_acc": 0.9182879377431906, "train_speed(iter/s)": 1.450647 }, { "epoch": 3.1551775845079475, "grad_norm": 2.2782464027404785, "learning_rate": 2.999343011227419e-05, "loss": 0.2942802906036377, "memory(GiB)": 70.5, "step": 73645, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.450652 }, { "epoch": 3.1553917998371963, "grad_norm": 4.16910457611084, "learning_rate": 2.9987262740638178e-05, "loss": 0.4765586853027344, "memory(GiB)": 70.5, "step": 73650, "token_acc": 0.8996763754045307, "train_speed(iter/s)": 1.450654 }, { "epoch": 3.155606015166445, "grad_norm": 5.790886402130127, "learning_rate": 2.9981095731551666e-05, "loss": 0.38767714500427247, "memory(GiB)": 70.5, "step": 73655, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.450662 }, { "epoch": 3.1558202304956944, "grad_norm": 2.3426201343536377, "learning_rate": 2.99749290851264e-05, "loss": 0.13232579231262206, "memory(GiB)": 70.5, "step": 73660, "token_acc": 0.9714285714285714, "train_speed(iter/s)": 1.450667 }, { "epoch": 3.156034445824943, "grad_norm": 0.3715740442276001, "learning_rate": 2.9968762801474087e-05, "loss": 0.5381460189819336, "memory(GiB)": 70.5, "step": 73665, "token_acc": 0.9007936507936508, "train_speed(iter/s)": 1.450672 }, { "epoch": 3.156248661154192, "grad_norm": 3.868090867996216, "learning_rate": 2.9962596880706413e-05, "loss": 0.3056839942932129, "memory(GiB)": 70.5, "step": 73670, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.450673 }, { "epoch": 3.1564628764834413, "grad_norm": 3.8264167308807373, "learning_rate": 2.9956431322935104e-05, "loss": 0.22808594703674318, "memory(GiB)": 70.5, "step": 73675, "token_acc": 0.9305019305019305, "train_speed(iter/s)": 1.450674 }, { "epoch": 3.15667709181269, "grad_norm": 2.234189033508301, "learning_rate": 2.995026612827183e-05, "loss": 0.1961223602294922, "memory(GiB)": 70.5, "step": 73680, "token_acc": 0.9526813880126183, "train_speed(iter/s)": 1.450675 }, { "epoch": 3.156891307141939, "grad_norm": 1.657736897468567, "learning_rate": 2.994410129682831e-05, "loss": 0.5483423709869385, "memory(GiB)": 70.5, "step": 73685, "token_acc": 0.8801261829652997, "train_speed(iter/s)": 1.450672 }, { "epoch": 3.157105522471188, "grad_norm": 2.3023436069488525, "learning_rate": 2.99379368287162e-05, "loss": 0.3473632335662842, "memory(GiB)": 70.5, "step": 73690, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.450673 }, { "epoch": 3.157319737800437, "grad_norm": 3.177722930908203, "learning_rate": 2.993177272404718e-05, "loss": 0.43468713760375977, "memory(GiB)": 70.5, "step": 73695, "token_acc": 0.8942307692307693, "train_speed(iter/s)": 1.450671 }, { "epoch": 3.1575339531296858, "grad_norm": 1.720363974571228, "learning_rate": 2.9925608982932928e-05, "loss": 0.23571810722351075, "memory(GiB)": 70.5, "step": 73700, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.450675 }, { "epoch": 3.157748168458935, "grad_norm": 0.7730898857116699, "learning_rate": 2.9919445605485107e-05, "loss": 0.38688862323760986, "memory(GiB)": 70.5, "step": 73705, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.45068 }, { "epoch": 3.157962383788184, "grad_norm": 3.714487075805664, "learning_rate": 2.9913282591815352e-05, "loss": 0.4934700012207031, "memory(GiB)": 70.5, "step": 73710, "token_acc": 0.8908450704225352, "train_speed(iter/s)": 1.450683 }, { "epoch": 3.1581765991174326, "grad_norm": 2.575284719467163, "learning_rate": 2.9907119942035338e-05, "loss": 0.39403705596923827, "memory(GiB)": 70.5, "step": 73715, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.450686 }, { "epoch": 3.158390814446682, "grad_norm": 4.576959609985352, "learning_rate": 2.9900957656256685e-05, "loss": 0.37491779327392577, "memory(GiB)": 70.5, "step": 73720, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.450688 }, { "epoch": 3.1586050297759307, "grad_norm": 3.091804027557373, "learning_rate": 2.989479573459104e-05, "loss": 0.345388126373291, "memory(GiB)": 70.5, "step": 73725, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.450694 }, { "epoch": 3.1588192451051795, "grad_norm": 3.810310125350952, "learning_rate": 2.9888634177150022e-05, "loss": 0.22818927764892577, "memory(GiB)": 70.5, "step": 73730, "token_acc": 0.9406528189910979, "train_speed(iter/s)": 1.450702 }, { "epoch": 3.159033460434429, "grad_norm": 1.4884207248687744, "learning_rate": 2.9882472984045277e-05, "loss": 0.5151957511901856, "memory(GiB)": 70.5, "step": 73735, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.45071 }, { "epoch": 3.1592476757636776, "grad_norm": 3.1570260524749756, "learning_rate": 2.9876312155388398e-05, "loss": 0.4389228343963623, "memory(GiB)": 70.5, "step": 73740, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.450715 }, { "epoch": 3.1594618910929264, "grad_norm": 0.21663159132003784, "learning_rate": 2.987015169129101e-05, "loss": 0.3273011207580566, "memory(GiB)": 70.5, "step": 73745, "token_acc": 0.9244186046511628, "train_speed(iter/s)": 1.450728 }, { "epoch": 3.1596761064221757, "grad_norm": 3.24324369430542, "learning_rate": 2.9863991591864705e-05, "loss": 0.6081544399261475, "memory(GiB)": 70.5, "step": 73750, "token_acc": 0.8798701298701299, "train_speed(iter/s)": 1.450732 }, { "epoch": 3.1598903217514245, "grad_norm": 5.1927409172058105, "learning_rate": 2.9857831857221075e-05, "loss": 0.22200479507446289, "memory(GiB)": 70.5, "step": 73755, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.45073 }, { "epoch": 3.1601045370806733, "grad_norm": 3.4131007194519043, "learning_rate": 2.9851672487471728e-05, "loss": 0.437656831741333, "memory(GiB)": 70.5, "step": 73760, "token_acc": 0.912, "train_speed(iter/s)": 1.450736 }, { "epoch": 3.1603187524099225, "grad_norm": 1.745124101638794, "learning_rate": 2.9845513482728217e-05, "loss": 0.39933362007141116, "memory(GiB)": 70.5, "step": 73765, "token_acc": 0.9131832797427653, "train_speed(iter/s)": 1.45074 }, { "epoch": 3.1605329677391714, "grad_norm": 0.37855005264282227, "learning_rate": 2.983935484310215e-05, "loss": 0.07022379636764527, "memory(GiB)": 70.5, "step": 73770, "token_acc": 0.983739837398374, "train_speed(iter/s)": 1.450741 }, { "epoch": 3.16074718306842, "grad_norm": 4.82307767868042, "learning_rate": 2.983319656870508e-05, "loss": 0.3043058395385742, "memory(GiB)": 70.5, "step": 73775, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.450744 }, { "epoch": 3.1609613983976694, "grad_norm": 4.199720859527588, "learning_rate": 2.9827038659648566e-05, "loss": 0.3113786458969116, "memory(GiB)": 70.5, "step": 73780, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.450748 }, { "epoch": 3.1611756137269182, "grad_norm": 3.9056410789489746, "learning_rate": 2.982088111604418e-05, "loss": 0.5143939971923828, "memory(GiB)": 70.5, "step": 73785, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.450751 }, { "epoch": 3.161389829056167, "grad_norm": 3.0293538570404053, "learning_rate": 2.9814723938003463e-05, "loss": 0.30783026218414306, "memory(GiB)": 70.5, "step": 73790, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.450764 }, { "epoch": 3.1616040443854163, "grad_norm": 2.8279664516448975, "learning_rate": 2.980856712563794e-05, "loss": 0.3129492521286011, "memory(GiB)": 70.5, "step": 73795, "token_acc": 0.930327868852459, "train_speed(iter/s)": 1.450781 }, { "epoch": 3.161818259714665, "grad_norm": 2.4701969623565674, "learning_rate": 2.9802410679059163e-05, "loss": 0.5710129261016845, "memory(GiB)": 70.5, "step": 73800, "token_acc": 0.8944281524926686, "train_speed(iter/s)": 1.450785 }, { "epoch": 3.162032475043914, "grad_norm": 6.4426774978637695, "learning_rate": 2.9796254598378683e-05, "loss": 0.4479838371276855, "memory(GiB)": 70.5, "step": 73805, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.450784 }, { "epoch": 3.162246690373163, "grad_norm": 3.375001907348633, "learning_rate": 2.9790098883707996e-05, "loss": 0.45462684631347655, "memory(GiB)": 70.5, "step": 73810, "token_acc": 0.8604651162790697, "train_speed(iter/s)": 1.450784 }, { "epoch": 3.162460905702412, "grad_norm": 2.8101425170898438, "learning_rate": 2.978394353515862e-05, "loss": 0.16532402038574218, "memory(GiB)": 70.5, "step": 73815, "token_acc": 0.9435483870967742, "train_speed(iter/s)": 1.45079 }, { "epoch": 3.162675121031661, "grad_norm": 3.363626718521118, "learning_rate": 2.977778855284208e-05, "loss": 0.3641984462738037, "memory(GiB)": 70.5, "step": 73820, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.450799 }, { "epoch": 3.16288933636091, "grad_norm": 2.8561513423919678, "learning_rate": 2.9771633936869863e-05, "loss": 0.2787427186965942, "memory(GiB)": 70.5, "step": 73825, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.450814 }, { "epoch": 3.163103551690159, "grad_norm": 2.7178382873535156, "learning_rate": 2.976547968735348e-05, "loss": 0.09896411895751953, "memory(GiB)": 70.5, "step": 73830, "token_acc": 0.9750889679715302, "train_speed(iter/s)": 1.450821 }, { "epoch": 3.163317767019408, "grad_norm": 0.36398953199386597, "learning_rate": 2.9759325804404418e-05, "loss": 0.2761090755462646, "memory(GiB)": 70.5, "step": 73835, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.450819 }, { "epoch": 3.163531982348657, "grad_norm": 2.685086488723755, "learning_rate": 2.9753172288134146e-05, "loss": 0.2926935195922852, "memory(GiB)": 70.5, "step": 73840, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.450834 }, { "epoch": 3.1637461976779058, "grad_norm": 2.706902265548706, "learning_rate": 2.9747019138654157e-05, "loss": 0.5007126331329346, "memory(GiB)": 70.5, "step": 73845, "token_acc": 0.8895522388059701, "train_speed(iter/s)": 1.450838 }, { "epoch": 3.163960413007155, "grad_norm": 5.32720422744751, "learning_rate": 2.974086635607592e-05, "loss": 0.42036027908325196, "memory(GiB)": 70.5, "step": 73850, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.450847 }, { "epoch": 3.164174628336404, "grad_norm": 2.1561789512634277, "learning_rate": 2.9734713940510884e-05, "loss": 0.2865294456481934, "memory(GiB)": 70.5, "step": 73855, "token_acc": 0.9382716049382716, "train_speed(iter/s)": 1.450852 }, { "epoch": 3.1643888436656527, "grad_norm": 2.730875015258789, "learning_rate": 2.9728561892070518e-05, "loss": 0.44747328758239746, "memory(GiB)": 70.5, "step": 73860, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.450865 }, { "epoch": 3.164603058994902, "grad_norm": 1.7887673377990723, "learning_rate": 2.972241021086627e-05, "loss": 0.20036771297454833, "memory(GiB)": 70.5, "step": 73865, "token_acc": 0.9551724137931035, "train_speed(iter/s)": 1.450879 }, { "epoch": 3.1648172743241507, "grad_norm": 4.0305609703063965, "learning_rate": 2.9716258897009586e-05, "loss": 0.3108454704284668, "memory(GiB)": 70.5, "step": 73870, "token_acc": 0.9452887537993921, "train_speed(iter/s)": 1.450887 }, { "epoch": 3.1650314896533995, "grad_norm": 3.4208807945251465, "learning_rate": 2.9710107950611878e-05, "loss": 0.2152768611907959, "memory(GiB)": 70.5, "step": 73875, "token_acc": 0.9619377162629758, "train_speed(iter/s)": 1.450889 }, { "epoch": 3.165245704982649, "grad_norm": 4.154252529144287, "learning_rate": 2.970395737178462e-05, "loss": 0.5478622913360596, "memory(GiB)": 70.5, "step": 73880, "token_acc": 0.8767605633802817, "train_speed(iter/s)": 1.450893 }, { "epoch": 3.1654599203118976, "grad_norm": 1.9658557176589966, "learning_rate": 2.96978071606392e-05, "loss": 0.21710073947906494, "memory(GiB)": 70.5, "step": 73885, "token_acc": 0.9678362573099415, "train_speed(iter/s)": 1.450893 }, { "epoch": 3.1656741356411464, "grad_norm": 3.892930030822754, "learning_rate": 2.9691657317287068e-05, "loss": 0.26632041931152345, "memory(GiB)": 70.5, "step": 73890, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.450894 }, { "epoch": 3.1658883509703957, "grad_norm": 2.456718921661377, "learning_rate": 2.968550784183961e-05, "loss": 0.2304760217666626, "memory(GiB)": 70.5, "step": 73895, "token_acc": 0.9603174603174603, "train_speed(iter/s)": 1.450898 }, { "epoch": 3.1661025662996445, "grad_norm": 2.8494770526885986, "learning_rate": 2.967935873440822e-05, "loss": 0.3206746816635132, "memory(GiB)": 70.5, "step": 73900, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.4509 }, { "epoch": 3.1663167816288933, "grad_norm": 2.1755025386810303, "learning_rate": 2.9673209995104322e-05, "loss": 0.297525691986084, "memory(GiB)": 70.5, "step": 73905, "token_acc": 0.92, "train_speed(iter/s)": 1.450901 }, { "epoch": 3.1665309969581426, "grad_norm": 1.727205753326416, "learning_rate": 2.9667061624039282e-05, "loss": 0.36710872650146487, "memory(GiB)": 70.5, "step": 73910, "token_acc": 0.9188311688311688, "train_speed(iter/s)": 1.450905 }, { "epoch": 3.1667452122873914, "grad_norm": 4.356240749359131, "learning_rate": 2.9660913621324503e-05, "loss": 0.38805341720581055, "memory(GiB)": 70.5, "step": 73915, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.450918 }, { "epoch": 3.16695942761664, "grad_norm": 2.4198412895202637, "learning_rate": 2.965476598707136e-05, "loss": 0.276625657081604, "memory(GiB)": 70.5, "step": 73920, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.450916 }, { "epoch": 3.1671736429458894, "grad_norm": 5.161329746246338, "learning_rate": 2.9648618721391197e-05, "loss": 0.6913477897644043, "memory(GiB)": 70.5, "step": 73925, "token_acc": 0.8274647887323944, "train_speed(iter/s)": 1.450939 }, { "epoch": 3.1673878582751382, "grad_norm": 5.139573574066162, "learning_rate": 2.9642471824395413e-05, "loss": 0.12344775199890137, "memory(GiB)": 70.5, "step": 73930, "token_acc": 0.9690265486725663, "train_speed(iter/s)": 1.450942 }, { "epoch": 3.167602073604387, "grad_norm": 2.9933383464813232, "learning_rate": 2.963632529619535e-05, "loss": 0.27944011688232423, "memory(GiB)": 70.5, "step": 73935, "token_acc": 0.9361022364217252, "train_speed(iter/s)": 1.450942 }, { "epoch": 3.1678162889336363, "grad_norm": 2.698333978652954, "learning_rate": 2.9630179136902346e-05, "loss": 0.3282012462615967, "memory(GiB)": 70.5, "step": 73940, "token_acc": 0.9283018867924528, "train_speed(iter/s)": 1.45096 }, { "epoch": 3.168030504262885, "grad_norm": 2.232055187225342, "learning_rate": 2.9624033346627766e-05, "loss": 0.2640849113464355, "memory(GiB)": 70.5, "step": 73945, "token_acc": 0.9435215946843853, "train_speed(iter/s)": 1.450964 }, { "epoch": 3.168244719592134, "grad_norm": 3.09303617477417, "learning_rate": 2.9617887925482914e-05, "loss": 0.620563268661499, "memory(GiB)": 70.5, "step": 73950, "token_acc": 0.8481848184818482, "train_speed(iter/s)": 1.45096 }, { "epoch": 3.168458934921383, "grad_norm": 5.422499656677246, "learning_rate": 2.961174287357916e-05, "loss": 0.4584686756134033, "memory(GiB)": 70.5, "step": 73955, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.450961 }, { "epoch": 3.168673150250632, "grad_norm": 1.876177430152893, "learning_rate": 2.9605598191027805e-05, "loss": 0.28288955688476564, "memory(GiB)": 70.5, "step": 73960, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.450967 }, { "epoch": 3.168887365579881, "grad_norm": 1.9717637300491333, "learning_rate": 2.9599453877940175e-05, "loss": 0.21593430042266845, "memory(GiB)": 70.5, "step": 73965, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.450963 }, { "epoch": 3.16910158090913, "grad_norm": 4.760017395019531, "learning_rate": 2.9593309934427565e-05, "loss": 0.232094407081604, "memory(GiB)": 70.5, "step": 73970, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.450968 }, { "epoch": 3.169315796238379, "grad_norm": 4.222288131713867, "learning_rate": 2.9587166360601303e-05, "loss": 0.4077012062072754, "memory(GiB)": 70.5, "step": 73975, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.450984 }, { "epoch": 3.1695300115676277, "grad_norm": 1.7237732410430908, "learning_rate": 2.9581023156572668e-05, "loss": 0.22245285511016846, "memory(GiB)": 70.5, "step": 73980, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.450991 }, { "epoch": 3.169744226896877, "grad_norm": 2.502964735031128, "learning_rate": 2.9574880322452947e-05, "loss": 0.4351677417755127, "memory(GiB)": 70.5, "step": 73985, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.450994 }, { "epoch": 3.1699584422261258, "grad_norm": 2.2134838104248047, "learning_rate": 2.9568737858353436e-05, "loss": 0.15552339553833008, "memory(GiB)": 70.5, "step": 73990, "token_acc": 0.9563758389261745, "train_speed(iter/s)": 1.450992 }, { "epoch": 3.1701726575553746, "grad_norm": 15.443849563598633, "learning_rate": 2.9562595764385405e-05, "loss": 0.22272725105285646, "memory(GiB)": 70.5, "step": 73995, "token_acc": 0.9568627450980393, "train_speed(iter/s)": 1.450997 }, { "epoch": 3.170386872884624, "grad_norm": 5.000420570373535, "learning_rate": 2.9556454040660114e-05, "loss": 0.24307940006256104, "memory(GiB)": 70.5, "step": 74000, "token_acc": 0.9511400651465798, "train_speed(iter/s)": 1.450998 }, { "epoch": 3.170386872884624, "eval_loss": 2.4051995277404785, "eval_runtime": 11.8145, "eval_samples_per_second": 8.464, "eval_steps_per_second": 8.464, "eval_token_acc": 0.4448441247002398, "step": 74000 }, { "epoch": 3.1706010882138727, "grad_norm": 4.298523902893066, "learning_rate": 2.9550312687288846e-05, "loss": 0.45514378547668455, "memory(GiB)": 70.5, "step": 74005, "token_acc": 0.5646743978590544, "train_speed(iter/s)": 1.450631 }, { "epoch": 3.1708153035431215, "grad_norm": 1.9985061883926392, "learning_rate": 2.954417170438284e-05, "loss": 0.17940104007720947, "memory(GiB)": 70.5, "step": 74010, "token_acc": 0.9653846153846154, "train_speed(iter/s)": 1.450634 }, { "epoch": 3.1710295188723707, "grad_norm": 2.6135401725769043, "learning_rate": 2.9538031092053354e-05, "loss": 0.21646199226379395, "memory(GiB)": 70.5, "step": 74015, "token_acc": 0.9566666666666667, "train_speed(iter/s)": 1.450636 }, { "epoch": 3.1712437342016195, "grad_norm": 4.022128582000732, "learning_rate": 2.9531890850411637e-05, "loss": 0.37653532028198244, "memory(GiB)": 70.5, "step": 74020, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.450657 }, { "epoch": 3.1714579495308683, "grad_norm": 0.4954313039779663, "learning_rate": 2.9525750979568895e-05, "loss": 0.25681371688842775, "memory(GiB)": 70.5, "step": 74025, "token_acc": 0.9462809917355371, "train_speed(iter/s)": 1.450658 }, { "epoch": 3.1716721648601176, "grad_norm": 1.3919023275375366, "learning_rate": 2.951961147963639e-05, "loss": 0.28982067108154297, "memory(GiB)": 70.5, "step": 74030, "token_acc": 0.9022082018927445, "train_speed(iter/s)": 1.450661 }, { "epoch": 3.1718863801893664, "grad_norm": 3.784055471420288, "learning_rate": 2.9513472350725347e-05, "loss": 0.4577195167541504, "memory(GiB)": 70.5, "step": 74035, "token_acc": 0.9137254901960784, "train_speed(iter/s)": 1.450665 }, { "epoch": 3.1721005955186152, "grad_norm": 2.6568596363067627, "learning_rate": 2.9507333592946975e-05, "loss": 0.17545132637023925, "memory(GiB)": 70.5, "step": 74040, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.450665 }, { "epoch": 3.1723148108478645, "grad_norm": 1.563262939453125, "learning_rate": 2.9501195206412457e-05, "loss": 0.2233975648880005, "memory(GiB)": 70.5, "step": 74045, "token_acc": 0.954954954954955, "train_speed(iter/s)": 1.450666 }, { "epoch": 3.1725290261771133, "grad_norm": 8.384830474853516, "learning_rate": 2.9495057191233033e-05, "loss": 0.38671844005584716, "memory(GiB)": 70.5, "step": 74050, "token_acc": 0.9325513196480938, "train_speed(iter/s)": 1.450666 }, { "epoch": 3.172743241506362, "grad_norm": 2.737292766571045, "learning_rate": 2.9488919547519878e-05, "loss": 0.15792441368103027, "memory(GiB)": 70.5, "step": 74055, "token_acc": 0.9541666666666667, "train_speed(iter/s)": 1.450669 }, { "epoch": 3.1729574568356114, "grad_norm": 3.4511430263519287, "learning_rate": 2.948278227538419e-05, "loss": 0.4553673267364502, "memory(GiB)": 70.5, "step": 74060, "token_acc": 0.9149659863945578, "train_speed(iter/s)": 1.450682 }, { "epoch": 3.17317167216486, "grad_norm": 1.4609752893447876, "learning_rate": 2.947664537493715e-05, "loss": 0.22149477005004883, "memory(GiB)": 70.5, "step": 74065, "token_acc": 0.9587628865979382, "train_speed(iter/s)": 1.450685 }, { "epoch": 3.173385887494109, "grad_norm": 3.3233325481414795, "learning_rate": 2.9470508846289924e-05, "loss": 0.3015456199645996, "memory(GiB)": 70.5, "step": 74070, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.45069 }, { "epoch": 3.1736001028233582, "grad_norm": 4.396424770355225, "learning_rate": 2.9464372689553692e-05, "loss": 0.390878701210022, "memory(GiB)": 70.5, "step": 74075, "token_acc": 0.9151515151515152, "train_speed(iter/s)": 1.45069 }, { "epoch": 3.173814318152607, "grad_norm": 0.29145383834838867, "learning_rate": 2.945823690483962e-05, "loss": 0.44698362350463866, "memory(GiB)": 70.5, "step": 74080, "token_acc": 0.89375, "train_speed(iter/s)": 1.450691 }, { "epoch": 3.174028533481856, "grad_norm": 2.183135986328125, "learning_rate": 2.9452101492258843e-05, "loss": 0.46283206939697263, "memory(GiB)": 70.5, "step": 74085, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.450695 }, { "epoch": 3.174242748811105, "grad_norm": 2.0563695430755615, "learning_rate": 2.9445966451922536e-05, "loss": 0.2753352880477905, "memory(GiB)": 70.5, "step": 74090, "token_acc": 0.9379562043795621, "train_speed(iter/s)": 1.450699 }, { "epoch": 3.174456964140354, "grad_norm": 0.1136130839586258, "learning_rate": 2.9439831783941807e-05, "loss": 0.12681785821914673, "memory(GiB)": 70.5, "step": 74095, "token_acc": 0.9671052631578947, "train_speed(iter/s)": 1.45071 }, { "epoch": 3.1746711794696028, "grad_norm": 4.588412761688232, "learning_rate": 2.9433697488427825e-05, "loss": 0.23513879776000976, "memory(GiB)": 70.5, "step": 74100, "token_acc": 0.9457627118644067, "train_speed(iter/s)": 1.450728 }, { "epoch": 3.174885394798852, "grad_norm": 4.253533363342285, "learning_rate": 2.94275635654917e-05, "loss": 0.2777801513671875, "memory(GiB)": 70.5, "step": 74105, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.450725 }, { "epoch": 3.175099610128101, "grad_norm": 3.5397145748138428, "learning_rate": 2.9421430015244565e-05, "loss": 0.2956386089324951, "memory(GiB)": 70.5, "step": 74110, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.45073 }, { "epoch": 3.1753138254573496, "grad_norm": 4.455765247344971, "learning_rate": 2.941529683779753e-05, "loss": 0.14858289957046508, "memory(GiB)": 70.5, "step": 74115, "token_acc": 0.9635258358662614, "train_speed(iter/s)": 1.450726 }, { "epoch": 3.175528040786599, "grad_norm": 0.6761698126792908, "learning_rate": 2.9409164033261704e-05, "loss": 0.4678169250488281, "memory(GiB)": 70.5, "step": 74120, "token_acc": 0.8952380952380953, "train_speed(iter/s)": 1.450725 }, { "epoch": 3.1757422561158477, "grad_norm": 1.9660050868988037, "learning_rate": 2.9403031601748192e-05, "loss": 0.6424643993377686, "memory(GiB)": 70.5, "step": 74125, "token_acc": 0.8816793893129771, "train_speed(iter/s)": 1.450727 }, { "epoch": 3.1759564714450965, "grad_norm": 2.754026412963867, "learning_rate": 2.939689954336807e-05, "loss": 0.6753204822540283, "memory(GiB)": 70.5, "step": 74130, "token_acc": 0.878698224852071, "train_speed(iter/s)": 1.450728 }, { "epoch": 3.1761706867743458, "grad_norm": 6.175771236419678, "learning_rate": 2.939076785823246e-05, "loss": 0.7241111278533936, "memory(GiB)": 70.5, "step": 74135, "token_acc": 0.8534201954397395, "train_speed(iter/s)": 1.450735 }, { "epoch": 3.1763849021035946, "grad_norm": 4.878801345825195, "learning_rate": 2.9384636546452415e-05, "loss": 0.71885085105896, "memory(GiB)": 70.5, "step": 74140, "token_acc": 0.8553459119496856, "train_speed(iter/s)": 1.450737 }, { "epoch": 3.1765991174328434, "grad_norm": 5.781327724456787, "learning_rate": 2.9378505608139007e-05, "loss": 0.5100934028625488, "memory(GiB)": 70.5, "step": 74145, "token_acc": 0.8769230769230769, "train_speed(iter/s)": 1.450733 }, { "epoch": 3.1768133327620927, "grad_norm": 1.7705647945404053, "learning_rate": 2.937237504340333e-05, "loss": 0.29520742893218993, "memory(GiB)": 70.5, "step": 74150, "token_acc": 0.9390681003584229, "train_speed(iter/s)": 1.450749 }, { "epoch": 3.1770275480913415, "grad_norm": 3.483248233795166, "learning_rate": 2.9366244852356418e-05, "loss": 0.5521910190582275, "memory(GiB)": 70.5, "step": 74155, "token_acc": 0.8742138364779874, "train_speed(iter/s)": 1.450757 }, { "epoch": 3.1772417634205903, "grad_norm": 2.962245464324951, "learning_rate": 2.9360115035109337e-05, "loss": 0.246460223197937, "memory(GiB)": 70.5, "step": 74160, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.450777 }, { "epoch": 3.1774559787498395, "grad_norm": 3.4776883125305176, "learning_rate": 2.9353985591773148e-05, "loss": 0.38447568416595457, "memory(GiB)": 70.5, "step": 74165, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.450779 }, { "epoch": 3.1776701940790884, "grad_norm": 4.373709678649902, "learning_rate": 2.934785652245885e-05, "loss": 0.40436697006225586, "memory(GiB)": 70.5, "step": 74170, "token_acc": 0.9075907590759076, "train_speed(iter/s)": 1.450784 }, { "epoch": 3.177884409408337, "grad_norm": 2.3353054523468018, "learning_rate": 2.934172782727751e-05, "loss": 0.4174190521240234, "memory(GiB)": 70.5, "step": 74175, "token_acc": 0.8961038961038961, "train_speed(iter/s)": 1.450789 }, { "epoch": 3.1780986247375864, "grad_norm": 3.249697208404541, "learning_rate": 2.9335599506340166e-05, "loss": 0.3121920585632324, "memory(GiB)": 70.5, "step": 74180, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.450792 }, { "epoch": 3.1783128400668352, "grad_norm": 1.5569493770599365, "learning_rate": 2.9329471559757822e-05, "loss": 0.2546222686767578, "memory(GiB)": 70.5, "step": 74185, "token_acc": 0.95, "train_speed(iter/s)": 1.450796 }, { "epoch": 3.178527055396084, "grad_norm": 2.095327138900757, "learning_rate": 2.932334398764148e-05, "loss": 0.20984148979187012, "memory(GiB)": 70.5, "step": 74190, "token_acc": 0.9536423841059603, "train_speed(iter/s)": 1.45081 }, { "epoch": 3.1787412707253333, "grad_norm": 3.536846876144409, "learning_rate": 2.9317216790102164e-05, "loss": 0.17366336584091185, "memory(GiB)": 70.5, "step": 74195, "token_acc": 0.9622093023255814, "train_speed(iter/s)": 1.450815 }, { "epoch": 3.178955486054582, "grad_norm": 0.49624982476234436, "learning_rate": 2.931108996725086e-05, "loss": 0.29935531616210936, "memory(GiB)": 70.5, "step": 74200, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.450815 }, { "epoch": 3.179169701383831, "grad_norm": 5.480079650878906, "learning_rate": 2.930496351919858e-05, "loss": 0.34906535148620604, "memory(GiB)": 70.5, "step": 74205, "token_acc": 0.8930817610062893, "train_speed(iter/s)": 1.450821 }, { "epoch": 3.17938391671308, "grad_norm": 3.1332273483276367, "learning_rate": 2.9298837446056304e-05, "loss": 0.3838343143463135, "memory(GiB)": 70.5, "step": 74210, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.450832 }, { "epoch": 3.179598132042329, "grad_norm": 1.9202078580856323, "learning_rate": 2.9292711747934993e-05, "loss": 0.29856505393981936, "memory(GiB)": 70.5, "step": 74215, "token_acc": 0.9425981873111783, "train_speed(iter/s)": 1.450842 }, { "epoch": 3.179812347371578, "grad_norm": 2.0958921909332275, "learning_rate": 2.928658642494564e-05, "loss": 0.24149980545043945, "memory(GiB)": 70.5, "step": 74220, "token_acc": 0.95, "train_speed(iter/s)": 1.450847 }, { "epoch": 3.180026562700827, "grad_norm": 2.6942193508148193, "learning_rate": 2.9280461477199205e-05, "loss": 0.3512250900268555, "memory(GiB)": 70.5, "step": 74225, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.450853 }, { "epoch": 3.180240778030076, "grad_norm": 4.783834457397461, "learning_rate": 2.9274336904806642e-05, "loss": 0.3673590660095215, "memory(GiB)": 70.5, "step": 74230, "token_acc": 0.9319727891156463, "train_speed(iter/s)": 1.45087 }, { "epoch": 3.1804549933593247, "grad_norm": 3.4693288803100586, "learning_rate": 2.9268212707878917e-05, "loss": 0.3092702150344849, "memory(GiB)": 70.5, "step": 74235, "token_acc": 0.9388489208633094, "train_speed(iter/s)": 1.450871 }, { "epoch": 3.180669208688574, "grad_norm": 4.398706436157227, "learning_rate": 2.9262088886526962e-05, "loss": 0.46518712043762206, "memory(GiB)": 70.5, "step": 74240, "token_acc": 0.9003690036900369, "train_speed(iter/s)": 1.450875 }, { "epoch": 3.1808834240178228, "grad_norm": 3.777771234512329, "learning_rate": 2.925596544086171e-05, "loss": 0.20002622604370118, "memory(GiB)": 70.5, "step": 74245, "token_acc": 0.9504132231404959, "train_speed(iter/s)": 1.450874 }, { "epoch": 3.1810976393470716, "grad_norm": 5.894285202026367, "learning_rate": 2.9249842370994106e-05, "loss": 0.30640623569488523, "memory(GiB)": 70.5, "step": 74250, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.450875 }, { "epoch": 3.181311854676321, "grad_norm": 3.4102110862731934, "learning_rate": 2.9243719677035086e-05, "loss": 0.43627128601074217, "memory(GiB)": 70.5, "step": 74255, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.450876 }, { "epoch": 3.1815260700055696, "grad_norm": 2.4246256351470947, "learning_rate": 2.9237597359095547e-05, "loss": 0.20536515712738038, "memory(GiB)": 70.5, "step": 74260, "token_acc": 0.9573643410852714, "train_speed(iter/s)": 1.450893 }, { "epoch": 3.1817402853348185, "grad_norm": 5.295307159423828, "learning_rate": 2.9231475417286412e-05, "loss": 0.365095329284668, "memory(GiB)": 70.5, "step": 74265, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.450911 }, { "epoch": 3.1819545006640677, "grad_norm": 2.4074020385742188, "learning_rate": 2.9225353851718586e-05, "loss": 0.36978898048400877, "memory(GiB)": 70.5, "step": 74270, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.450923 }, { "epoch": 3.1821687159933165, "grad_norm": 4.467382907867432, "learning_rate": 2.9219232662502948e-05, "loss": 0.2133394241333008, "memory(GiB)": 70.5, "step": 74275, "token_acc": 0.953416149068323, "train_speed(iter/s)": 1.450926 }, { "epoch": 3.1823829313225653, "grad_norm": 3.612732410430908, "learning_rate": 2.9213111849750425e-05, "loss": 0.32138011455535886, "memory(GiB)": 70.5, "step": 74280, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.450931 }, { "epoch": 3.1825971466518146, "grad_norm": 5.854883193969727, "learning_rate": 2.9206991413571872e-05, "loss": 0.6115692138671875, "memory(GiB)": 70.5, "step": 74285, "token_acc": 0.8640483383685801, "train_speed(iter/s)": 1.450945 }, { "epoch": 3.1828113619810634, "grad_norm": 2.1564717292785645, "learning_rate": 2.9200871354078167e-05, "loss": 0.28852708339691163, "memory(GiB)": 70.5, "step": 74290, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.450947 }, { "epoch": 3.183025577310312, "grad_norm": 2.090170383453369, "learning_rate": 2.91947516713802e-05, "loss": 0.3860557317733765, "memory(GiB)": 70.5, "step": 74295, "token_acc": 0.9071428571428571, "train_speed(iter/s)": 1.45095 }, { "epoch": 3.1832397926395615, "grad_norm": 1.7657554149627686, "learning_rate": 2.918863236558882e-05, "loss": 0.22537581920623778, "memory(GiB)": 70.5, "step": 74300, "token_acc": 0.9542682926829268, "train_speed(iter/s)": 1.450954 }, { "epoch": 3.1834540079688103, "grad_norm": 3.2145636081695557, "learning_rate": 2.9182513436814897e-05, "loss": 0.25955705642700194, "memory(GiB)": 70.5, "step": 74305, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.450959 }, { "epoch": 3.183668223298059, "grad_norm": 0.5910168290138245, "learning_rate": 2.9176394885169238e-05, "loss": 0.138176429271698, "memory(GiB)": 70.5, "step": 74310, "token_acc": 0.9647435897435898, "train_speed(iter/s)": 1.450968 }, { "epoch": 3.1838824386273084, "grad_norm": 1.1756232976913452, "learning_rate": 2.9170276710762746e-05, "loss": 0.1989973783493042, "memory(GiB)": 70.5, "step": 74315, "token_acc": 0.9640718562874252, "train_speed(iter/s)": 1.450969 }, { "epoch": 3.184096653956557, "grad_norm": 2.4997475147247314, "learning_rate": 2.9164158913706207e-05, "loss": 0.1652943253517151, "memory(GiB)": 70.5, "step": 74320, "token_acc": 0.964968152866242, "train_speed(iter/s)": 1.450972 }, { "epoch": 3.184310869285806, "grad_norm": 1.7179347276687622, "learning_rate": 2.91580414941105e-05, "loss": 0.354105281829834, "memory(GiB)": 70.5, "step": 74325, "token_acc": 0.9141104294478528, "train_speed(iter/s)": 1.450972 }, { "epoch": 3.1845250846150552, "grad_norm": 2.413621664047241, "learning_rate": 2.9151924452086428e-05, "loss": 0.331425142288208, "memory(GiB)": 70.5, "step": 74330, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.450976 }, { "epoch": 3.184739299944304, "grad_norm": 5.260150909423828, "learning_rate": 2.9145807787744795e-05, "loss": 0.22783117294311522, "memory(GiB)": 70.5, "step": 74335, "token_acc": 0.941908713692946, "train_speed(iter/s)": 1.450973 }, { "epoch": 3.184953515273553, "grad_norm": 3.2386672496795654, "learning_rate": 2.91396915011964e-05, "loss": 0.3945464611053467, "memory(GiB)": 70.5, "step": 74340, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.450971 }, { "epoch": 3.185167730602802, "grad_norm": 1.7354106903076172, "learning_rate": 2.9133575592552076e-05, "loss": 0.3688825607299805, "memory(GiB)": 70.5, "step": 74345, "token_acc": 0.9126760563380282, "train_speed(iter/s)": 1.450976 }, { "epoch": 3.185381945932051, "grad_norm": 0.6970155835151672, "learning_rate": 2.9127460061922607e-05, "loss": 0.4552006244659424, "memory(GiB)": 70.5, "step": 74350, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.45098 }, { "epoch": 3.1855961612612997, "grad_norm": 4.4111104011535645, "learning_rate": 2.912134490941878e-05, "loss": 0.45589332580566405, "memory(GiB)": 70.5, "step": 74355, "token_acc": 0.9016949152542373, "train_speed(iter/s)": 1.450992 }, { "epoch": 3.185810376590549, "grad_norm": 5.689512252807617, "learning_rate": 2.911523013515137e-05, "loss": 0.3462419033050537, "memory(GiB)": 70.5, "step": 74360, "token_acc": 0.927797833935018, "train_speed(iter/s)": 1.451 }, { "epoch": 3.186024591919798, "grad_norm": 2.29459547996521, "learning_rate": 2.9109115739231145e-05, "loss": 0.2614685297012329, "memory(GiB)": 70.5, "step": 74365, "token_acc": 0.9425287356321839, "train_speed(iter/s)": 1.451007 }, { "epoch": 3.1862388072490466, "grad_norm": 2.2864596843719482, "learning_rate": 2.910300172176891e-05, "loss": 0.2590022563934326, "memory(GiB)": 70.5, "step": 74370, "token_acc": 0.956081081081081, "train_speed(iter/s)": 1.451017 }, { "epoch": 3.186453022578296, "grad_norm": 4.620111465454102, "learning_rate": 2.909688808287539e-05, "loss": 0.18715690374374389, "memory(GiB)": 70.5, "step": 74375, "token_acc": 0.9566787003610109, "train_speed(iter/s)": 1.451021 }, { "epoch": 3.1866672379075447, "grad_norm": 2.824869155883789, "learning_rate": 2.909077482266136e-05, "loss": 0.3120777130126953, "memory(GiB)": 70.5, "step": 74380, "token_acc": 0.9174311926605505, "train_speed(iter/s)": 1.451018 }, { "epoch": 3.1868814532367935, "grad_norm": 1.683587670326233, "learning_rate": 2.9084661941237552e-05, "loss": 0.5073543071746827, "memory(GiB)": 70.5, "step": 74385, "token_acc": 0.8766233766233766, "train_speed(iter/s)": 1.451021 }, { "epoch": 3.1870956685660428, "grad_norm": 2.4090986251831055, "learning_rate": 2.9078549438714715e-05, "loss": 0.4756914615631104, "memory(GiB)": 70.5, "step": 74390, "token_acc": 0.8885017421602788, "train_speed(iter/s)": 1.451029 }, { "epoch": 3.1873098838952916, "grad_norm": 7.548488616943359, "learning_rate": 2.907243731520356e-05, "loss": 0.3236818790435791, "memory(GiB)": 70.5, "step": 74395, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.45103 }, { "epoch": 3.1875240992245404, "grad_norm": 3.299933910369873, "learning_rate": 2.9066325570814834e-05, "loss": 0.33263483047485354, "memory(GiB)": 70.5, "step": 74400, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.451046 }, { "epoch": 3.1877383145537896, "grad_norm": 4.559695243835449, "learning_rate": 2.9060214205659286e-05, "loss": 0.17305960655212402, "memory(GiB)": 70.5, "step": 74405, "token_acc": 0.9785714285714285, "train_speed(iter/s)": 1.451052 }, { "epoch": 3.1879525298830385, "grad_norm": 4.0877485275268555, "learning_rate": 2.9054103219847596e-05, "loss": 0.32170872688293456, "memory(GiB)": 70.5, "step": 74410, "token_acc": 0.9180887372013652, "train_speed(iter/s)": 1.451052 }, { "epoch": 3.1881667452122873, "grad_norm": 5.773012638092041, "learning_rate": 2.904799261349047e-05, "loss": 0.8243888854980469, "memory(GiB)": 70.5, "step": 74415, "token_acc": 0.8480662983425414, "train_speed(iter/s)": 1.451063 }, { "epoch": 3.1883809605415365, "grad_norm": 2.434576988220215, "learning_rate": 2.9041882386698616e-05, "loss": 0.3908622026443481, "memory(GiB)": 70.5, "step": 74420, "token_acc": 0.9115853658536586, "train_speed(iter/s)": 1.451065 }, { "epoch": 3.1885951758707853, "grad_norm": 2.990117073059082, "learning_rate": 2.9035772539582706e-05, "loss": 0.32243025302886963, "memory(GiB)": 70.5, "step": 74425, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.45107 }, { "epoch": 3.188809391200034, "grad_norm": 3.3220949172973633, "learning_rate": 2.9029663072253455e-05, "loss": 0.3661219596862793, "memory(GiB)": 70.5, "step": 74430, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.451086 }, { "epoch": 3.1890236065292834, "grad_norm": 4.612640380859375, "learning_rate": 2.9023553984821537e-05, "loss": 0.3145071744918823, "memory(GiB)": 70.5, "step": 74435, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.451087 }, { "epoch": 3.189237821858532, "grad_norm": 3.0210089683532715, "learning_rate": 2.9017445277397616e-05, "loss": 0.38254556655883787, "memory(GiB)": 70.5, "step": 74440, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.451095 }, { "epoch": 3.189452037187781, "grad_norm": 4.60438871383667, "learning_rate": 2.901133695009235e-05, "loss": 0.3549495697021484, "memory(GiB)": 70.5, "step": 74445, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.451106 }, { "epoch": 3.1896662525170303, "grad_norm": 2.495847225189209, "learning_rate": 2.900522900301641e-05, "loss": 0.23902301788330077, "memory(GiB)": 70.5, "step": 74450, "token_acc": 0.9335347432024169, "train_speed(iter/s)": 1.451114 }, { "epoch": 3.189880467846279, "grad_norm": 0.8752538561820984, "learning_rate": 2.899912143628042e-05, "loss": 0.17869194746017455, "memory(GiB)": 70.5, "step": 74455, "token_acc": 0.9623188405797102, "train_speed(iter/s)": 1.451114 }, { "epoch": 3.190094683175528, "grad_norm": 3.9221692085266113, "learning_rate": 2.8993014249995066e-05, "loss": 0.3283254623413086, "memory(GiB)": 70.5, "step": 74460, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.451109 }, { "epoch": 3.190308898504777, "grad_norm": 1.6910640001296997, "learning_rate": 2.898690744427096e-05, "loss": 0.4313383102416992, "memory(GiB)": 70.5, "step": 74465, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.451124 }, { "epoch": 3.190523113834026, "grad_norm": 0.7399028539657593, "learning_rate": 2.8980801019218718e-05, "loss": 0.41312503814697266, "memory(GiB)": 70.5, "step": 74470, "token_acc": 0.9198813056379822, "train_speed(iter/s)": 1.451133 }, { "epoch": 3.190737329163275, "grad_norm": 4.323029041290283, "learning_rate": 2.8974694974949013e-05, "loss": 0.374501895904541, "memory(GiB)": 70.5, "step": 74475, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.451141 }, { "epoch": 3.190951544492524, "grad_norm": 1.4087481498718262, "learning_rate": 2.8968589311572426e-05, "loss": 0.1979094624519348, "memory(GiB)": 70.5, "step": 74480, "token_acc": 0.9573770491803278, "train_speed(iter/s)": 1.451141 }, { "epoch": 3.191165759821773, "grad_norm": 2.3574330806732178, "learning_rate": 2.896248402919955e-05, "loss": 0.44668169021606446, "memory(GiB)": 70.5, "step": 74485, "token_acc": 0.8971631205673759, "train_speed(iter/s)": 1.451143 }, { "epoch": 3.1913799751510217, "grad_norm": 3.5676798820495605, "learning_rate": 2.895637912794103e-05, "loss": 0.32168126106262207, "memory(GiB)": 70.5, "step": 74490, "token_acc": 0.9321533923303835, "train_speed(iter/s)": 1.451143 }, { "epoch": 3.191594190480271, "grad_norm": 2.6364123821258545, "learning_rate": 2.8950274607907447e-05, "loss": 0.752114200592041, "memory(GiB)": 70.5, "step": 74495, "token_acc": 0.8540925266903915, "train_speed(iter/s)": 1.451144 }, { "epoch": 3.1918084058095197, "grad_norm": 2.6159439086914062, "learning_rate": 2.8944170469209386e-05, "loss": 0.38425984382629397, "memory(GiB)": 70.5, "step": 74500, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.451144 }, { "epoch": 3.1918084058095197, "eval_loss": 2.3852274417877197, "eval_runtime": 10.9418, "eval_samples_per_second": 9.139, "eval_steps_per_second": 9.139, "eval_token_acc": 0.47214076246334313, "step": 74500 }, { "epoch": 3.1920226211387686, "grad_norm": 3.2331414222717285, "learning_rate": 2.8938066711957423e-05, "loss": 0.37349414825439453, "memory(GiB)": 70.5, "step": 74505, "token_acc": 0.6047966631908238, "train_speed(iter/s)": 1.450818 }, { "epoch": 3.192236836468018, "grad_norm": 3.703043222427368, "learning_rate": 2.8931963336262125e-05, "loss": 0.64370698928833, "memory(GiB)": 70.5, "step": 74510, "token_acc": 0.8619631901840491, "train_speed(iter/s)": 1.450832 }, { "epoch": 3.1924510517972666, "grad_norm": 1.7531580924987793, "learning_rate": 2.8925860342234092e-05, "loss": 0.4230493068695068, "memory(GiB)": 70.5, "step": 74515, "token_acc": 0.9204545454545454, "train_speed(iter/s)": 1.450837 }, { "epoch": 3.1926652671265154, "grad_norm": 2.5547115802764893, "learning_rate": 2.8919757729983866e-05, "loss": 0.23397254943847656, "memory(GiB)": 70.5, "step": 74520, "token_acc": 0.95703125, "train_speed(iter/s)": 1.450833 }, { "epoch": 3.1928794824557647, "grad_norm": 5.1927490234375, "learning_rate": 2.8913655499622012e-05, "loss": 0.5408868789672852, "memory(GiB)": 70.5, "step": 74525, "token_acc": 0.8851963746223565, "train_speed(iter/s)": 1.450837 }, { "epoch": 3.1930936977850135, "grad_norm": 0.77693772315979, "learning_rate": 2.890755365125906e-05, "loss": 0.3664238929748535, "memory(GiB)": 70.5, "step": 74530, "token_acc": 0.9367088607594937, "train_speed(iter/s)": 1.45084 }, { "epoch": 3.1933079131142623, "grad_norm": 4.544182777404785, "learning_rate": 2.890145218500556e-05, "loss": 0.23871736526489257, "memory(GiB)": 70.5, "step": 74535, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.450853 }, { "epoch": 3.1935221284435116, "grad_norm": 2.206307888031006, "learning_rate": 2.8895351100972033e-05, "loss": 0.3919790267944336, "memory(GiB)": 70.5, "step": 74540, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.45085 }, { "epoch": 3.1937363437727604, "grad_norm": 5.788297176361084, "learning_rate": 2.888925039926902e-05, "loss": 0.2807874917984009, "memory(GiB)": 70.5, "step": 74545, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.450854 }, { "epoch": 3.193950559102009, "grad_norm": 3.8964695930480957, "learning_rate": 2.8883150080007047e-05, "loss": 0.3525510311126709, "memory(GiB)": 70.5, "step": 74550, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.450856 }, { "epoch": 3.1941647744312585, "grad_norm": 6.740449905395508, "learning_rate": 2.887705014329663e-05, "loss": 0.2470475673675537, "memory(GiB)": 70.5, "step": 74555, "token_acc": 0.9451476793248945, "train_speed(iter/s)": 1.450852 }, { "epoch": 3.1943789897605073, "grad_norm": 3.132669448852539, "learning_rate": 2.8870950589248256e-05, "loss": 0.20751371383666992, "memory(GiB)": 70.5, "step": 74560, "token_acc": 0.9438943894389439, "train_speed(iter/s)": 1.450852 }, { "epoch": 3.194593205089756, "grad_norm": 2.3317618370056152, "learning_rate": 2.8864851417972444e-05, "loss": 0.27766263484954834, "memory(GiB)": 70.5, "step": 74565, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.450855 }, { "epoch": 3.1948074204190053, "grad_norm": 4.058060169219971, "learning_rate": 2.8858752629579648e-05, "loss": 0.49500579833984376, "memory(GiB)": 70.5, "step": 74570, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.450858 }, { "epoch": 3.195021635748254, "grad_norm": 6.206117630004883, "learning_rate": 2.88526542241804e-05, "loss": 0.5567138671875, "memory(GiB)": 70.5, "step": 74575, "token_acc": 0.888, "train_speed(iter/s)": 1.450857 }, { "epoch": 3.195235851077503, "grad_norm": 1.9788141250610352, "learning_rate": 2.8846556201885162e-05, "loss": 0.38670284748077394, "memory(GiB)": 70.5, "step": 74580, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.450858 }, { "epoch": 3.195450066406752, "grad_norm": 2.460484027862549, "learning_rate": 2.8840458562804396e-05, "loss": 0.31816122531890867, "memory(GiB)": 70.5, "step": 74585, "token_acc": 0.9292307692307692, "train_speed(iter/s)": 1.450862 }, { "epoch": 3.195664281736001, "grad_norm": 5.379992485046387, "learning_rate": 2.883436130704858e-05, "loss": 0.4551846981048584, "memory(GiB)": 70.5, "step": 74590, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.450874 }, { "epoch": 3.19587849706525, "grad_norm": 0.7483944892883301, "learning_rate": 2.882826443472816e-05, "loss": 0.3863327741622925, "memory(GiB)": 70.5, "step": 74595, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.450881 }, { "epoch": 3.196092712394499, "grad_norm": 3.2194089889526367, "learning_rate": 2.882216794595357e-05, "loss": 0.5476034641265869, "memory(GiB)": 70.5, "step": 74600, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.450877 }, { "epoch": 3.196306927723748, "grad_norm": 6.238986492156982, "learning_rate": 2.881607184083529e-05, "loss": 0.4485603332519531, "memory(GiB)": 70.5, "step": 74605, "token_acc": 0.8871473354231975, "train_speed(iter/s)": 1.450886 }, { "epoch": 3.1965211430529967, "grad_norm": 3.8706138134002686, "learning_rate": 2.880997611948375e-05, "loss": 0.41308345794677737, "memory(GiB)": 70.5, "step": 74610, "token_acc": 0.905511811023622, "train_speed(iter/s)": 1.450913 }, { "epoch": 3.196735358382246, "grad_norm": 1.8656872510910034, "learning_rate": 2.8803880782009347e-05, "loss": 0.4728723049163818, "memory(GiB)": 70.5, "step": 74615, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.450915 }, { "epoch": 3.196949573711495, "grad_norm": 1.1564931869506836, "learning_rate": 2.8797785828522543e-05, "loss": 0.3891255855560303, "memory(GiB)": 70.5, "step": 74620, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.450921 }, { "epoch": 3.1971637890407436, "grad_norm": 3.106747627258301, "learning_rate": 2.8791691259133745e-05, "loss": 0.1755396842956543, "memory(GiB)": 70.5, "step": 74625, "token_acc": 0.9496644295302014, "train_speed(iter/s)": 1.450925 }, { "epoch": 3.197378004369993, "grad_norm": 2.8863251209259033, "learning_rate": 2.878559707395333e-05, "loss": 0.5964037418365479, "memory(GiB)": 70.5, "step": 74630, "token_acc": 0.9158878504672897, "train_speed(iter/s)": 1.45093 }, { "epoch": 3.1975922196992417, "grad_norm": 3.223210096359253, "learning_rate": 2.8779503273091746e-05, "loss": 0.3493373870849609, "memory(GiB)": 70.5, "step": 74635, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.450947 }, { "epoch": 3.1978064350284905, "grad_norm": 1.5800269842147827, "learning_rate": 2.877340985665936e-05, "loss": 0.2065277576446533, "memory(GiB)": 70.5, "step": 74640, "token_acc": 0.948339483394834, "train_speed(iter/s)": 1.450959 }, { "epoch": 3.1980206503577397, "grad_norm": 5.586320877075195, "learning_rate": 2.8767316824766577e-05, "loss": 0.24634575843811035, "memory(GiB)": 70.5, "step": 74645, "token_acc": 0.9433333333333334, "train_speed(iter/s)": 1.450972 }, { "epoch": 3.1982348656869886, "grad_norm": 2.771001100540161, "learning_rate": 2.876122417752376e-05, "loss": 0.1886073112487793, "memory(GiB)": 70.5, "step": 74650, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.450979 }, { "epoch": 3.1984490810162374, "grad_norm": 0.2792518138885498, "learning_rate": 2.8755131915041277e-05, "loss": 0.2844519138336182, "memory(GiB)": 70.5, "step": 74655, "token_acc": 0.9385964912280702, "train_speed(iter/s)": 1.450978 }, { "epoch": 3.1986632963454866, "grad_norm": 3.2281744480133057, "learning_rate": 2.8749040037429527e-05, "loss": 0.4405482292175293, "memory(GiB)": 70.5, "step": 74660, "token_acc": 0.9006849315068494, "train_speed(iter/s)": 1.45098 }, { "epoch": 3.1988775116747354, "grad_norm": 1.4222691059112549, "learning_rate": 2.874294854479885e-05, "loss": 0.48550825119018554, "memory(GiB)": 70.5, "step": 74665, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.450992 }, { "epoch": 3.1990917270039843, "grad_norm": 3.0362775325775146, "learning_rate": 2.8736857437259602e-05, "loss": 0.4233431816101074, "memory(GiB)": 70.5, "step": 74670, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.450995 }, { "epoch": 3.1993059423332335, "grad_norm": 4.348502159118652, "learning_rate": 2.8730766714922126e-05, "loss": 0.18212172985076905, "memory(GiB)": 70.5, "step": 74675, "token_acc": 0.9578544061302682, "train_speed(iter/s)": 1.450996 }, { "epoch": 3.1995201576624823, "grad_norm": 4.216501712799072, "learning_rate": 2.8724676377896764e-05, "loss": 0.4834470748901367, "memory(GiB)": 70.5, "step": 74680, "token_acc": 0.9064516129032258, "train_speed(iter/s)": 1.451 }, { "epoch": 3.199734372991731, "grad_norm": 3.010514974594116, "learning_rate": 2.8718586426293825e-05, "loss": 0.22386369705200196, "memory(GiB)": 70.5, "step": 74685, "token_acc": 0.9467680608365019, "train_speed(iter/s)": 1.451 }, { "epoch": 3.1999485883209804, "grad_norm": 3.241507053375244, "learning_rate": 2.8712496860223658e-05, "loss": 0.33199028968811034, "memory(GiB)": 70.5, "step": 74690, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.451012 }, { "epoch": 3.200162803650229, "grad_norm": 2.62104868888855, "learning_rate": 2.8706407679796605e-05, "loss": 0.3933661460876465, "memory(GiB)": 70.5, "step": 74695, "token_acc": 0.916010498687664, "train_speed(iter/s)": 1.451019 }, { "epoch": 3.200377018979478, "grad_norm": 2.7846102714538574, "learning_rate": 2.8700318885122944e-05, "loss": 0.40830144882202146, "memory(GiB)": 70.5, "step": 74700, "token_acc": 0.9085545722713865, "train_speed(iter/s)": 1.451027 }, { "epoch": 3.2005912343087273, "grad_norm": 0.9502204060554504, "learning_rate": 2.869423047631299e-05, "loss": 0.09289953708648682, "memory(GiB)": 70.5, "step": 74705, "token_acc": 0.9817518248175182, "train_speed(iter/s)": 1.451029 }, { "epoch": 3.200805449637976, "grad_norm": 3.6787915229797363, "learning_rate": 2.868814245347704e-05, "loss": 0.49650158882141116, "memory(GiB)": 70.5, "step": 74710, "token_acc": 0.868421052631579, "train_speed(iter/s)": 1.451035 }, { "epoch": 3.201019664967225, "grad_norm": 2.8312156200408936, "learning_rate": 2.8682054816725358e-05, "loss": 0.2595843315124512, "memory(GiB)": 70.5, "step": 74715, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.451036 }, { "epoch": 3.201233880296474, "grad_norm": 1.2558432817459106, "learning_rate": 2.867596756616827e-05, "loss": 0.2673351287841797, "memory(GiB)": 70.5, "step": 74720, "token_acc": 0.9282868525896414, "train_speed(iter/s)": 1.451041 }, { "epoch": 3.201448095625723, "grad_norm": 4.193010330200195, "learning_rate": 2.866988070191604e-05, "loss": 0.29291043281555174, "memory(GiB)": 70.5, "step": 74725, "token_acc": 0.9120879120879121, "train_speed(iter/s)": 1.45105 }, { "epoch": 3.201662310954972, "grad_norm": 2.3710291385650635, "learning_rate": 2.8663794224078926e-05, "loss": 0.14268531799316406, "memory(GiB)": 70.5, "step": 74730, "token_acc": 0.9696969696969697, "train_speed(iter/s)": 1.451053 }, { "epoch": 3.201876526284221, "grad_norm": 6.409684181213379, "learning_rate": 2.865770813276719e-05, "loss": 0.4857994556427002, "memory(GiB)": 70.5, "step": 74735, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.451061 }, { "epoch": 3.20209074161347, "grad_norm": 3.085012197494507, "learning_rate": 2.86516224280911e-05, "loss": 0.3116565227508545, "memory(GiB)": 70.5, "step": 74740, "token_acc": 0.940809968847352, "train_speed(iter/s)": 1.451072 }, { "epoch": 3.2023049569427187, "grad_norm": 1.7737325429916382, "learning_rate": 2.8645537110160873e-05, "loss": 0.2795441150665283, "memory(GiB)": 70.5, "step": 74745, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.451074 }, { "epoch": 3.202519172271968, "grad_norm": 3.1545627117156982, "learning_rate": 2.8639452179086788e-05, "loss": 0.17405085563659667, "memory(GiB)": 70.5, "step": 74750, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.451073 }, { "epoch": 3.2027333876012167, "grad_norm": 6.457001209259033, "learning_rate": 2.8633367634979075e-05, "loss": 0.23979551792144777, "memory(GiB)": 70.5, "step": 74755, "token_acc": 0.9391304347826087, "train_speed(iter/s)": 1.451086 }, { "epoch": 3.2029476029304655, "grad_norm": 0.0725308284163475, "learning_rate": 2.8627283477947942e-05, "loss": 0.2566894292831421, "memory(GiB)": 70.5, "step": 74760, "token_acc": 0.9385245901639344, "train_speed(iter/s)": 1.451092 }, { "epoch": 3.203161818259715, "grad_norm": 0.21118973195552826, "learning_rate": 2.8621199708103606e-05, "loss": 0.18740378618240355, "memory(GiB)": 70.5, "step": 74765, "token_acc": 0.9605734767025089, "train_speed(iter/s)": 1.451093 }, { "epoch": 3.2033760335889636, "grad_norm": 4.262868404388428, "learning_rate": 2.8615116325556307e-05, "loss": 0.5697237014770508, "memory(GiB)": 70.5, "step": 74770, "token_acc": 0.8867924528301887, "train_speed(iter/s)": 1.451096 }, { "epoch": 3.2035902489182124, "grad_norm": 5.457889080047607, "learning_rate": 2.8609033330416213e-05, "loss": 0.32398009300231934, "memory(GiB)": 70.5, "step": 74775, "token_acc": 0.9138461538461539, "train_speed(iter/s)": 1.451101 }, { "epoch": 3.2038044642474617, "grad_norm": 3.431885004043579, "learning_rate": 2.8602950722793577e-05, "loss": 0.20732333660125732, "memory(GiB)": 70.5, "step": 74780, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.451109 }, { "epoch": 3.2040186795767105, "grad_norm": 2.617335081100464, "learning_rate": 2.859686850279855e-05, "loss": 0.633833360671997, "memory(GiB)": 70.5, "step": 74785, "token_acc": 0.8794788273615635, "train_speed(iter/s)": 1.451113 }, { "epoch": 3.2042328949059593, "grad_norm": 3.4763591289520264, "learning_rate": 2.8590786670541336e-05, "loss": 0.3447331666946411, "memory(GiB)": 70.5, "step": 74790, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.451116 }, { "epoch": 3.2044471102352086, "grad_norm": 3.0618982315063477, "learning_rate": 2.8584705226132096e-05, "loss": 0.334951114654541, "memory(GiB)": 70.5, "step": 74795, "token_acc": 0.896, "train_speed(iter/s)": 1.451115 }, { "epoch": 3.2046613255644574, "grad_norm": 2.5622918605804443, "learning_rate": 2.8578624169680994e-05, "loss": 0.24486086368560792, "memory(GiB)": 70.5, "step": 74800, "token_acc": 0.9467084639498433, "train_speed(iter/s)": 1.451118 }, { "epoch": 3.204875540893706, "grad_norm": 3.023548126220703, "learning_rate": 2.857254350129823e-05, "loss": 0.2905434608459473, "memory(GiB)": 70.5, "step": 74805, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.451119 }, { "epoch": 3.2050897562229554, "grad_norm": 3.919049024581909, "learning_rate": 2.8566463221093943e-05, "loss": 0.3422385692596436, "memory(GiB)": 70.5, "step": 74810, "token_acc": 0.9381107491856677, "train_speed(iter/s)": 1.451124 }, { "epoch": 3.2053039715522043, "grad_norm": 4.772867202758789, "learning_rate": 2.8560383329178286e-05, "loss": 0.4639772891998291, "memory(GiB)": 70.5, "step": 74815, "token_acc": 0.8920634920634921, "train_speed(iter/s)": 1.451136 }, { "epoch": 3.205518186881453, "grad_norm": 3.7500522136688232, "learning_rate": 2.8554303825661388e-05, "loss": 0.27893290519714353, "memory(GiB)": 70.5, "step": 74820, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.451142 }, { "epoch": 3.2057324022107023, "grad_norm": 6.414547920227051, "learning_rate": 2.8548224710653393e-05, "loss": 0.5302448749542237, "memory(GiB)": 70.5, "step": 74825, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.451146 }, { "epoch": 3.205946617539951, "grad_norm": 2.755232572555542, "learning_rate": 2.8542145984264416e-05, "loss": 0.2945180654525757, "memory(GiB)": 70.5, "step": 74830, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.451145 }, { "epoch": 3.2061608328692, "grad_norm": 7.008225440979004, "learning_rate": 2.8536067646604592e-05, "loss": 0.4174013137817383, "memory(GiB)": 70.5, "step": 74835, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.451144 }, { "epoch": 3.206375048198449, "grad_norm": 3.5102412700653076, "learning_rate": 2.852998969778406e-05, "loss": 0.2565053701400757, "memory(GiB)": 70.5, "step": 74840, "token_acc": 0.9349112426035503, "train_speed(iter/s)": 1.451141 }, { "epoch": 3.206589263527698, "grad_norm": 1.2101764678955078, "learning_rate": 2.85239121379129e-05, "loss": 0.10628035068511962, "memory(GiB)": 70.5, "step": 74845, "token_acc": 0.9795918367346939, "train_speed(iter/s)": 1.451143 }, { "epoch": 3.206803478856947, "grad_norm": 4.111387252807617, "learning_rate": 2.851783496710122e-05, "loss": 0.369500994682312, "memory(GiB)": 70.5, "step": 74850, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.451144 }, { "epoch": 3.207017694186196, "grad_norm": 4.228940963745117, "learning_rate": 2.8511758185459113e-05, "loss": 0.524386215209961, "memory(GiB)": 70.5, "step": 74855, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.451149 }, { "epoch": 3.207231909515445, "grad_norm": 2.3416714668273926, "learning_rate": 2.850568179309665e-05, "loss": 0.37448046207427976, "memory(GiB)": 70.5, "step": 74860, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.451156 }, { "epoch": 3.2074461248446937, "grad_norm": 4.06040620803833, "learning_rate": 2.8499605790123935e-05, "loss": 0.39342048168182375, "memory(GiB)": 70.5, "step": 74865, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.451181 }, { "epoch": 3.207660340173943, "grad_norm": 1.8311493396759033, "learning_rate": 2.8493530176651028e-05, "loss": 0.23276927471160888, "memory(GiB)": 70.5, "step": 74870, "token_acc": 0.9517684887459807, "train_speed(iter/s)": 1.451194 }, { "epoch": 3.207874555503192, "grad_norm": 3.0016162395477295, "learning_rate": 2.8487454952788008e-05, "loss": 0.26689815521240234, "memory(GiB)": 70.5, "step": 74875, "token_acc": 0.9484848484848485, "train_speed(iter/s)": 1.451209 }, { "epoch": 3.2080887708324406, "grad_norm": 0.3620639145374298, "learning_rate": 2.848138011864491e-05, "loss": 0.13766499757766723, "memory(GiB)": 70.5, "step": 74880, "token_acc": 0.9808429118773946, "train_speed(iter/s)": 1.451212 }, { "epoch": 3.20830298616169, "grad_norm": 0.9329511523246765, "learning_rate": 2.8475305674331808e-05, "loss": 0.13462188243865966, "memory(GiB)": 70.5, "step": 74885, "token_acc": 0.9566666666666667, "train_speed(iter/s)": 1.45123 }, { "epoch": 3.2085172014909387, "grad_norm": 4.190216541290283, "learning_rate": 2.8469231619958708e-05, "loss": 0.3240990161895752, "memory(GiB)": 70.5, "step": 74890, "token_acc": 0.948339483394834, "train_speed(iter/s)": 1.451234 }, { "epoch": 3.2087314168201875, "grad_norm": 4.215682506561279, "learning_rate": 2.84631579556357e-05, "loss": 0.3963596343994141, "memory(GiB)": 70.5, "step": 74895, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.451234 }, { "epoch": 3.2089456321494367, "grad_norm": 2.899749279022217, "learning_rate": 2.845708468147278e-05, "loss": 0.4709893226623535, "memory(GiB)": 70.5, "step": 74900, "token_acc": 0.9084507042253521, "train_speed(iter/s)": 1.451235 }, { "epoch": 3.2091598474786855, "grad_norm": 3.525829315185547, "learning_rate": 2.8451011797579985e-05, "loss": 0.7196298122406006, "memory(GiB)": 70.5, "step": 74905, "token_acc": 0.8803088803088803, "train_speed(iter/s)": 1.451239 }, { "epoch": 3.2093740628079344, "grad_norm": 0.9102205038070679, "learning_rate": 2.8444939304067307e-05, "loss": 0.23952717781066896, "memory(GiB)": 70.5, "step": 74910, "token_acc": 0.9429530201342282, "train_speed(iter/s)": 1.45124 }, { "epoch": 3.2095882781371836, "grad_norm": 2.5091781616210938, "learning_rate": 2.8438867201044794e-05, "loss": 0.38771681785583495, "memory(GiB)": 70.5, "step": 74915, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.451257 }, { "epoch": 3.2098024934664324, "grad_norm": 3.191617012023926, "learning_rate": 2.843279548862241e-05, "loss": 0.17954868078231812, "memory(GiB)": 70.5, "step": 74920, "token_acc": 0.95, "train_speed(iter/s)": 1.451273 }, { "epoch": 3.2100167087956812, "grad_norm": 3.1461360454559326, "learning_rate": 2.842672416691018e-05, "loss": 0.3406796455383301, "memory(GiB)": 70.5, "step": 74925, "token_acc": 0.9252336448598131, "train_speed(iter/s)": 1.451275 }, { "epoch": 3.2102309241249305, "grad_norm": 6.077735424041748, "learning_rate": 2.8420653236018086e-05, "loss": 0.4902037620544434, "memory(GiB)": 70.5, "step": 74930, "token_acc": 0.9093333333333333, "train_speed(iter/s)": 1.451274 }, { "epoch": 3.2104451394541793, "grad_norm": 2.213900566101074, "learning_rate": 2.8414582696056102e-05, "loss": 0.492689323425293, "memory(GiB)": 70.5, "step": 74935, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.451287 }, { "epoch": 3.210659354783428, "grad_norm": 4.714061260223389, "learning_rate": 2.840851254713419e-05, "loss": 0.33952343463897705, "memory(GiB)": 70.5, "step": 74940, "token_acc": 0.9079754601226994, "train_speed(iter/s)": 1.451302 }, { "epoch": 3.2108735701126774, "grad_norm": 3.3455677032470703, "learning_rate": 2.8402442789362322e-05, "loss": 0.5150367259979248, "memory(GiB)": 70.5, "step": 74945, "token_acc": 0.9112627986348123, "train_speed(iter/s)": 1.451313 }, { "epoch": 3.211087785441926, "grad_norm": 2.919095754623413, "learning_rate": 2.8396373422850476e-05, "loss": 0.18101009130477905, "memory(GiB)": 70.5, "step": 74950, "token_acc": 0.9519230769230769, "train_speed(iter/s)": 1.451319 }, { "epoch": 3.211302000771175, "grad_norm": 7.547571182250977, "learning_rate": 2.8390304447708594e-05, "loss": 0.4328585624694824, "memory(GiB)": 70.5, "step": 74955, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.451326 }, { "epoch": 3.2115162161004243, "grad_norm": 4.2828545570373535, "learning_rate": 2.8384235864046614e-05, "loss": 0.5414325714111328, "memory(GiB)": 70.5, "step": 74960, "token_acc": 0.8875, "train_speed(iter/s)": 1.451339 }, { "epoch": 3.211730431429673, "grad_norm": 1.1317256689071655, "learning_rate": 2.837816767197448e-05, "loss": 1.0742083549499513, "memory(GiB)": 70.5, "step": 74965, "token_acc": 0.8132183908045977, "train_speed(iter/s)": 1.45134 }, { "epoch": 3.211944646758922, "grad_norm": 0.9500356316566467, "learning_rate": 2.8372099871602127e-05, "loss": 0.5595292091369629, "memory(GiB)": 70.5, "step": 74970, "token_acc": 0.8754208754208754, "train_speed(iter/s)": 1.451357 }, { "epoch": 3.212158862088171, "grad_norm": 1.2352492809295654, "learning_rate": 2.8366032463039448e-05, "loss": 0.33001861572265623, "memory(GiB)": 70.5, "step": 74975, "token_acc": 0.9045092838196287, "train_speed(iter/s)": 1.451352 }, { "epoch": 3.21237307741742, "grad_norm": 6.323176860809326, "learning_rate": 2.835996544639641e-05, "loss": 0.7594161987304687, "memory(GiB)": 70.5, "step": 74980, "token_acc": 0.8703071672354948, "train_speed(iter/s)": 1.451357 }, { "epoch": 3.2125872927466688, "grad_norm": 5.21635627746582, "learning_rate": 2.835389882178287e-05, "loss": 0.3733957767486572, "memory(GiB)": 70.5, "step": 74985, "token_acc": 0.9119718309859155, "train_speed(iter/s)": 1.451384 }, { "epoch": 3.212801508075918, "grad_norm": 3.630369186401367, "learning_rate": 2.834783258930878e-05, "loss": 0.29427213668823243, "memory(GiB)": 70.5, "step": 74990, "token_acc": 0.9431818181818182, "train_speed(iter/s)": 1.451383 }, { "epoch": 3.213015723405167, "grad_norm": 4.234401702880859, "learning_rate": 2.834176674908402e-05, "loss": 0.6786663055419921, "memory(GiB)": 70.5, "step": 74995, "token_acc": 0.8856209150326797, "train_speed(iter/s)": 1.451385 }, { "epoch": 3.2132299387344156, "grad_norm": 0.5088023543357849, "learning_rate": 2.8335701301218472e-05, "loss": 0.5303323268890381, "memory(GiB)": 70.5, "step": 75000, "token_acc": 0.8843537414965986, "train_speed(iter/s)": 1.451385 }, { "epoch": 3.2132299387344156, "eval_loss": 2.483795404434204, "eval_runtime": 11.8288, "eval_samples_per_second": 8.454, "eval_steps_per_second": 8.454, "eval_token_acc": 0.4474034620505992, "step": 75000 }, { "epoch": 3.213444154063665, "grad_norm": 2.9869415760040283, "learning_rate": 2.8329636245821988e-05, "loss": 0.3561123847961426, "memory(GiB)": 70.5, "step": 75005, "token_acc": 0.5801158301158301, "train_speed(iter/s)": 1.45103 }, { "epoch": 3.2136583693929137, "grad_norm": 3.38653302192688, "learning_rate": 2.8323571583004505e-05, "loss": 0.37024452686309817, "memory(GiB)": 70.5, "step": 75010, "token_acc": 0.9102167182662538, "train_speed(iter/s)": 1.451032 }, { "epoch": 3.2138725847221625, "grad_norm": 3.07083797454834, "learning_rate": 2.8317507312875845e-05, "loss": 0.24221265316009521, "memory(GiB)": 70.5, "step": 75015, "token_acc": 0.9381818181818182, "train_speed(iter/s)": 1.451037 }, { "epoch": 3.214086800051412, "grad_norm": 3.9762001037597656, "learning_rate": 2.8311443435545892e-05, "loss": 0.24602828025817872, "memory(GiB)": 70.5, "step": 75020, "token_acc": 0.9409448818897638, "train_speed(iter/s)": 1.451038 }, { "epoch": 3.2143010153806606, "grad_norm": 2.3741447925567627, "learning_rate": 2.8305379951124478e-05, "loss": 0.4387653827667236, "memory(GiB)": 70.5, "step": 75025, "token_acc": 0.9026845637583892, "train_speed(iter/s)": 1.451037 }, { "epoch": 3.2145152307099094, "grad_norm": 2.9164981842041016, "learning_rate": 2.8299316859721435e-05, "loss": 0.07935194969177246, "memory(GiB)": 70.5, "step": 75030, "token_acc": 0.9814126394052045, "train_speed(iter/s)": 1.451051 }, { "epoch": 3.2147294460391587, "grad_norm": 5.527071952819824, "learning_rate": 2.829325416144665e-05, "loss": 0.448537015914917, "memory(GiB)": 70.5, "step": 75035, "token_acc": 0.8932806324110671, "train_speed(iter/s)": 1.45107 }, { "epoch": 3.2149436613684075, "grad_norm": 5.278393268585205, "learning_rate": 2.8287191856409935e-05, "loss": 0.4672675132751465, "memory(GiB)": 70.5, "step": 75040, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.451069 }, { "epoch": 3.2151578766976563, "grad_norm": 2.751208543777466, "learning_rate": 2.8281129944721096e-05, "loss": 0.33105058670043946, "memory(GiB)": 70.5, "step": 75045, "token_acc": 0.9390681003584229, "train_speed(iter/s)": 1.451068 }, { "epoch": 3.2153720920269055, "grad_norm": 1.6802996397018433, "learning_rate": 2.8275068426489965e-05, "loss": 0.28593029975891116, "memory(GiB)": 70.5, "step": 75050, "token_acc": 0.93, "train_speed(iter/s)": 1.451076 }, { "epoch": 3.2155863073561544, "grad_norm": 2.0401666164398193, "learning_rate": 2.826900730182635e-05, "loss": 0.2696421146392822, "memory(GiB)": 70.5, "step": 75055, "token_acc": 0.9486301369863014, "train_speed(iter/s)": 1.451077 }, { "epoch": 3.215800522685403, "grad_norm": 2.247663974761963, "learning_rate": 2.8262946570840032e-05, "loss": 0.3327061653137207, "memory(GiB)": 70.5, "step": 75060, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.451081 }, { "epoch": 3.2160147380146524, "grad_norm": 0.2210189402103424, "learning_rate": 2.8256886233640834e-05, "loss": 0.41120638847351076, "memory(GiB)": 70.5, "step": 75065, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.45109 }, { "epoch": 3.2162289533439012, "grad_norm": 0.23943491280078888, "learning_rate": 2.825082629033855e-05, "loss": 0.46292738914489745, "memory(GiB)": 70.5, "step": 75070, "token_acc": 0.9082278481012658, "train_speed(iter/s)": 1.451099 }, { "epoch": 3.21644316867315, "grad_norm": 0.08721946179866791, "learning_rate": 2.824476674104295e-05, "loss": 0.2721043825149536, "memory(GiB)": 70.5, "step": 75075, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.451101 }, { "epoch": 3.2166573840023993, "grad_norm": 3.0071284770965576, "learning_rate": 2.823870758586381e-05, "loss": 0.3187652587890625, "memory(GiB)": 70.5, "step": 75080, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.451107 }, { "epoch": 3.216871599331648, "grad_norm": 3.3371622562408447, "learning_rate": 2.8232648824910902e-05, "loss": 0.4174026966094971, "memory(GiB)": 70.5, "step": 75085, "token_acc": 0.92, "train_speed(iter/s)": 1.451109 }, { "epoch": 3.217085814660897, "grad_norm": 3.373142957687378, "learning_rate": 2.8226590458293957e-05, "loss": 0.3506643295288086, "memory(GiB)": 70.5, "step": 75090, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.451108 }, { "epoch": 3.217300029990146, "grad_norm": 3.462688446044922, "learning_rate": 2.822053248612277e-05, "loss": 0.23957560062408448, "memory(GiB)": 70.5, "step": 75095, "token_acc": 0.9442231075697212, "train_speed(iter/s)": 1.451123 }, { "epoch": 3.217514245319395, "grad_norm": 4.762001991271973, "learning_rate": 2.8214474908507066e-05, "loss": 0.5819993019104004, "memory(GiB)": 70.5, "step": 75100, "token_acc": 0.8847583643122676, "train_speed(iter/s)": 1.451128 }, { "epoch": 3.217728460648644, "grad_norm": 5.1093363761901855, "learning_rate": 2.820841772555659e-05, "loss": 0.5557600498199463, "memory(GiB)": 70.5, "step": 75105, "token_acc": 0.8553191489361702, "train_speed(iter/s)": 1.45114 }, { "epoch": 3.217942675977893, "grad_norm": 0.6501772999763489, "learning_rate": 2.8202360937381066e-05, "loss": 0.3486659049987793, "memory(GiB)": 70.5, "step": 75110, "token_acc": 0.9083665338645418, "train_speed(iter/s)": 1.451141 }, { "epoch": 3.218156891307142, "grad_norm": 2.242772102355957, "learning_rate": 2.8196304544090225e-05, "loss": 0.2080383539199829, "memory(GiB)": 70.5, "step": 75115, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.451149 }, { "epoch": 3.2183711066363907, "grad_norm": 2.046865940093994, "learning_rate": 2.819024854579375e-05, "loss": 0.47169198989868166, "memory(GiB)": 70.5, "step": 75120, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.451151 }, { "epoch": 3.21858532196564, "grad_norm": 3.5784850120544434, "learning_rate": 2.8184192942601413e-05, "loss": 0.2644627571105957, "memory(GiB)": 70.5, "step": 75125, "token_acc": 0.950530035335689, "train_speed(iter/s)": 1.451153 }, { "epoch": 3.2187995372948888, "grad_norm": 6.444850444793701, "learning_rate": 2.8178137734622878e-05, "loss": 0.32673397064208987, "memory(GiB)": 70.5, "step": 75130, "token_acc": 0.9336099585062241, "train_speed(iter/s)": 1.451154 }, { "epoch": 3.2190137526241376, "grad_norm": 1.873722791671753, "learning_rate": 2.8172082921967836e-05, "loss": 0.28410000801086427, "memory(GiB)": 70.5, "step": 75135, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.451169 }, { "epoch": 3.219227967953387, "grad_norm": 1.8157676458358765, "learning_rate": 2.8166028504746e-05, "loss": 0.28237831592559814, "memory(GiB)": 70.5, "step": 75140, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.451186 }, { "epoch": 3.2194421832826356, "grad_norm": 1.5531004667282104, "learning_rate": 2.8159974483067047e-05, "loss": 0.31177046298980715, "memory(GiB)": 70.5, "step": 75145, "token_acc": 0.9356913183279743, "train_speed(iter/s)": 1.451192 }, { "epoch": 3.2196563986118845, "grad_norm": 2.0258474349975586, "learning_rate": 2.815392085704062e-05, "loss": 0.3340648651123047, "memory(GiB)": 70.5, "step": 75150, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.451196 }, { "epoch": 3.2198706139411337, "grad_norm": 2.4843032360076904, "learning_rate": 2.8147867626776437e-05, "loss": 0.23456413745880128, "memory(GiB)": 70.5, "step": 75155, "token_acc": 0.9519230769230769, "train_speed(iter/s)": 1.451199 }, { "epoch": 3.2200848292703825, "grad_norm": 4.019682884216309, "learning_rate": 2.8141814792384123e-05, "loss": 0.367003345489502, "memory(GiB)": 70.5, "step": 75160, "token_acc": 0.9305019305019305, "train_speed(iter/s)": 1.451196 }, { "epoch": 3.2202990445996313, "grad_norm": 3.8893539905548096, "learning_rate": 2.813576235397334e-05, "loss": 0.27210249900817873, "memory(GiB)": 70.5, "step": 75165, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.451195 }, { "epoch": 3.2205132599288806, "grad_norm": 4.626502513885498, "learning_rate": 2.8129710311653735e-05, "loss": 0.7522389411926269, "memory(GiB)": 70.5, "step": 75170, "token_acc": 0.849112426035503, "train_speed(iter/s)": 1.451211 }, { "epoch": 3.2207274752581294, "grad_norm": 0.6970006823539734, "learning_rate": 2.8123658665534913e-05, "loss": 0.3350327730178833, "memory(GiB)": 70.5, "step": 75175, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.451213 }, { "epoch": 3.2209416905873782, "grad_norm": 2.458442211151123, "learning_rate": 2.8117607415726565e-05, "loss": 0.21803910732269288, "memory(GiB)": 70.5, "step": 75180, "token_acc": 0.946949602122016, "train_speed(iter/s)": 1.451219 }, { "epoch": 3.2211559059166275, "grad_norm": 4.819361686706543, "learning_rate": 2.8111556562338292e-05, "loss": 0.3233463287353516, "memory(GiB)": 70.5, "step": 75185, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.451224 }, { "epoch": 3.2213701212458763, "grad_norm": 7.006316661834717, "learning_rate": 2.8105506105479695e-05, "loss": 0.5230603218078613, "memory(GiB)": 70.5, "step": 75190, "token_acc": 0.8686567164179104, "train_speed(iter/s)": 1.451233 }, { "epoch": 3.221584336575125, "grad_norm": 0.8458946943283081, "learning_rate": 2.8099456045260396e-05, "loss": 0.17894960641860963, "memory(GiB)": 70.5, "step": 75195, "token_acc": 0.956953642384106, "train_speed(iter/s)": 1.451241 }, { "epoch": 3.2217985519043744, "grad_norm": 8.06461238861084, "learning_rate": 2.809340638179e-05, "loss": 0.39478373527526855, "memory(GiB)": 70.5, "step": 75200, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.451246 }, { "epoch": 3.222012767233623, "grad_norm": 2.1162710189819336, "learning_rate": 2.808735711517807e-05, "loss": 0.5972624301910401, "memory(GiB)": 70.5, "step": 75205, "token_acc": 0.8767605633802817, "train_speed(iter/s)": 1.45126 }, { "epoch": 3.222226982562872, "grad_norm": 2.335866689682007, "learning_rate": 2.8081308245534228e-05, "loss": 0.2164177417755127, "memory(GiB)": 70.5, "step": 75210, "token_acc": 0.9449275362318841, "train_speed(iter/s)": 1.451268 }, { "epoch": 3.2224411978921212, "grad_norm": 4.466354846954346, "learning_rate": 2.8075259772968064e-05, "loss": 0.8529810905456543, "memory(GiB)": 70.5, "step": 75215, "token_acc": 0.8058510638297872, "train_speed(iter/s)": 1.451289 }, { "epoch": 3.22265541322137, "grad_norm": 1.1739505529403687, "learning_rate": 2.8069211697589138e-05, "loss": 0.27584741115570066, "memory(GiB)": 70.5, "step": 75220, "token_acc": 0.9482071713147411, "train_speed(iter/s)": 1.451296 }, { "epoch": 3.222869628550619, "grad_norm": 2.7857651710510254, "learning_rate": 2.8063164019507017e-05, "loss": 0.08486671447753906, "memory(GiB)": 70.5, "step": 75225, "token_acc": 0.9817629179331308, "train_speed(iter/s)": 1.451297 }, { "epoch": 3.223083843879868, "grad_norm": 3.360102415084839, "learning_rate": 2.805711673883125e-05, "loss": 0.4099071502685547, "memory(GiB)": 70.5, "step": 75230, "token_acc": 0.9173228346456693, "train_speed(iter/s)": 1.451299 }, { "epoch": 3.223298059209117, "grad_norm": 2.15248441696167, "learning_rate": 2.8051069855671387e-05, "loss": 0.09934760332107544, "memory(GiB)": 70.5, "step": 75235, "token_acc": 0.9705882352941176, "train_speed(iter/s)": 1.451306 }, { "epoch": 3.2235122745383658, "grad_norm": 6.499687194824219, "learning_rate": 2.8045023370136996e-05, "loss": 0.29233403205871583, "memory(GiB)": 70.5, "step": 75240, "token_acc": 0.9368770764119602, "train_speed(iter/s)": 1.451309 }, { "epoch": 3.223726489867615, "grad_norm": 0.4233017861843109, "learning_rate": 2.8038977282337607e-05, "loss": 0.1976815342903137, "memory(GiB)": 70.5, "step": 75245, "token_acc": 0.945619335347432, "train_speed(iter/s)": 1.45132 }, { "epoch": 3.223940705196864, "grad_norm": 0.4835679829120636, "learning_rate": 2.8032931592382743e-05, "loss": 0.3668114900588989, "memory(GiB)": 70.5, "step": 75250, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.451326 }, { "epoch": 3.2241549205261126, "grad_norm": 0.6242585778236389, "learning_rate": 2.802688630038193e-05, "loss": 0.14200583696365357, "memory(GiB)": 70.5, "step": 75255, "token_acc": 0.9724770642201835, "train_speed(iter/s)": 1.451328 }, { "epoch": 3.224369135855362, "grad_norm": 3.3395237922668457, "learning_rate": 2.8020841406444687e-05, "loss": 0.31669204235076903, "memory(GiB)": 70.5, "step": 75260, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.451327 }, { "epoch": 3.2245833511846107, "grad_norm": 0.6448824405670166, "learning_rate": 2.8014796910680498e-05, "loss": 0.267722487449646, "memory(GiB)": 70.5, "step": 75265, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.451334 }, { "epoch": 3.2247975665138595, "grad_norm": 5.412004470825195, "learning_rate": 2.800875281319891e-05, "loss": 0.6098781585693359, "memory(GiB)": 70.5, "step": 75270, "token_acc": 0.8439490445859873, "train_speed(iter/s)": 1.451335 }, { "epoch": 3.2250117818431088, "grad_norm": 0.9521579146385193, "learning_rate": 2.8002709114109392e-05, "loss": 0.5081943035125732, "memory(GiB)": 70.5, "step": 75275, "token_acc": 0.907563025210084, "train_speed(iter/s)": 1.451352 }, { "epoch": 3.2252259971723576, "grad_norm": 2.976186752319336, "learning_rate": 2.799666581352141e-05, "loss": 0.5647845268249512, "memory(GiB)": 70.5, "step": 75280, "token_acc": 0.8867313915857605, "train_speed(iter/s)": 1.451378 }, { "epoch": 3.2254402125016064, "grad_norm": 1.1419010162353516, "learning_rate": 2.7990622911544494e-05, "loss": 0.33280770778656005, "memory(GiB)": 70.5, "step": 75285, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.45138 }, { "epoch": 3.2256544278308557, "grad_norm": 5.527945518493652, "learning_rate": 2.7984580408288087e-05, "loss": 0.5013064861297607, "memory(GiB)": 70.5, "step": 75290, "token_acc": 0.9194444444444444, "train_speed(iter/s)": 1.451394 }, { "epoch": 3.2258686431601045, "grad_norm": 4.872481822967529, "learning_rate": 2.7978538303861635e-05, "loss": 0.4866637229919434, "memory(GiB)": 70.5, "step": 75295, "token_acc": 0.8904593639575972, "train_speed(iter/s)": 1.451394 }, { "epoch": 3.2260828584893533, "grad_norm": 2.8966660499572754, "learning_rate": 2.797249659837464e-05, "loss": 0.38243014812469484, "memory(GiB)": 70.5, "step": 75300, "token_acc": 0.8863636363636364, "train_speed(iter/s)": 1.451394 }, { "epoch": 3.2262970738186025, "grad_norm": 1.3555552959442139, "learning_rate": 2.796645529193654e-05, "loss": 0.35007495880126954, "memory(GiB)": 70.5, "step": 75305, "token_acc": 0.9273504273504274, "train_speed(iter/s)": 1.451397 }, { "epoch": 3.2265112891478513, "grad_norm": 0.3856491446495056, "learning_rate": 2.7960414384656763e-05, "loss": 0.16999369859695435, "memory(GiB)": 70.5, "step": 75310, "token_acc": 0.9554794520547946, "train_speed(iter/s)": 1.4514 }, { "epoch": 3.2267255044771, "grad_norm": 4.894618988037109, "learning_rate": 2.7954373876644756e-05, "loss": 0.2877516269683838, "memory(GiB)": 70.5, "step": 75315, "token_acc": 0.937984496124031, "train_speed(iter/s)": 1.451402 }, { "epoch": 3.2269397198063494, "grad_norm": 1.727864146232605, "learning_rate": 2.7948333768009932e-05, "loss": 0.34389486312866213, "memory(GiB)": 70.5, "step": 75320, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.451427 }, { "epoch": 3.2271539351355982, "grad_norm": 5.307719707489014, "learning_rate": 2.7942294058861745e-05, "loss": 0.3993887424468994, "memory(GiB)": 70.5, "step": 75325, "token_acc": 0.9221311475409836, "train_speed(iter/s)": 1.451426 }, { "epoch": 3.227368150464847, "grad_norm": 4.208259105682373, "learning_rate": 2.7936254749309594e-05, "loss": 0.3057635068893433, "memory(GiB)": 70.5, "step": 75330, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.451421 }, { "epoch": 3.2275823657940963, "grad_norm": 0.03324522078037262, "learning_rate": 2.7930215839462882e-05, "loss": 0.42290534973144533, "memory(GiB)": 70.5, "step": 75335, "token_acc": 0.8987730061349694, "train_speed(iter/s)": 1.451417 }, { "epoch": 3.227796581123345, "grad_norm": 1.1550246477127075, "learning_rate": 2.792417732943101e-05, "loss": 0.36349413394927976, "memory(GiB)": 70.5, "step": 75340, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.451424 }, { "epoch": 3.228010796452594, "grad_norm": 1.9986131191253662, "learning_rate": 2.7918139219323386e-05, "loss": 0.4894977569580078, "memory(GiB)": 70.5, "step": 75345, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.451438 }, { "epoch": 3.228225011781843, "grad_norm": 2.802640914916992, "learning_rate": 2.7912101509249357e-05, "loss": 0.39223551750183105, "memory(GiB)": 70.5, "step": 75350, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.451438 }, { "epoch": 3.228439227111092, "grad_norm": 1.6427676677703857, "learning_rate": 2.7906064199318327e-05, "loss": 0.22025647163391113, "memory(GiB)": 70.5, "step": 75355, "token_acc": 0.9541984732824428, "train_speed(iter/s)": 1.451454 }, { "epoch": 3.228653442440341, "grad_norm": 4.654759407043457, "learning_rate": 2.7900027289639695e-05, "loss": 0.7171690940856934, "memory(GiB)": 70.5, "step": 75360, "token_acc": 0.8787878787878788, "train_speed(iter/s)": 1.451468 }, { "epoch": 3.22886765776959, "grad_norm": 2.379079818725586, "learning_rate": 2.7893990780322798e-05, "loss": 0.21998331546783448, "memory(GiB)": 70.5, "step": 75365, "token_acc": 0.948051948051948, "train_speed(iter/s)": 1.451471 }, { "epoch": 3.229081873098839, "grad_norm": 4.243980407714844, "learning_rate": 2.7887954671477002e-05, "loss": 0.3770092248916626, "memory(GiB)": 70.5, "step": 75370, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.451471 }, { "epoch": 3.2292960884280877, "grad_norm": 5.017594814300537, "learning_rate": 2.7881918963211638e-05, "loss": 0.29110767841339114, "memory(GiB)": 70.5, "step": 75375, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.451478 }, { "epoch": 3.229510303757337, "grad_norm": 4.905587673187256, "learning_rate": 2.787588365563605e-05, "loss": 0.2870715379714966, "memory(GiB)": 70.5, "step": 75380, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.451478 }, { "epoch": 3.2297245190865858, "grad_norm": 1.927597165107727, "learning_rate": 2.7869848748859596e-05, "loss": 0.10694575309753418, "memory(GiB)": 70.5, "step": 75385, "token_acc": 0.9786324786324786, "train_speed(iter/s)": 1.451482 }, { "epoch": 3.2299387344158346, "grad_norm": 0.9936454892158508, "learning_rate": 2.786381424299159e-05, "loss": 0.15108145475387574, "memory(GiB)": 70.5, "step": 75390, "token_acc": 0.9700374531835206, "train_speed(iter/s)": 1.451487 }, { "epoch": 3.230152949745084, "grad_norm": 1.8133394718170166, "learning_rate": 2.7857780138141365e-05, "loss": 0.4697120189666748, "memory(GiB)": 70.5, "step": 75395, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.451487 }, { "epoch": 3.2303671650743326, "grad_norm": 2.3702163696289062, "learning_rate": 2.785174643441822e-05, "loss": 0.15678516626358033, "memory(GiB)": 70.5, "step": 75400, "token_acc": 0.9525316455696202, "train_speed(iter/s)": 1.451494 }, { "epoch": 3.2305813804035814, "grad_norm": 4.752873420715332, "learning_rate": 2.7845713131931462e-05, "loss": 0.17932963371276855, "memory(GiB)": 70.5, "step": 75405, "token_acc": 0.9566787003610109, "train_speed(iter/s)": 1.451496 }, { "epoch": 3.2307955957328307, "grad_norm": 2.7616519927978516, "learning_rate": 2.783968023079038e-05, "loss": 0.34952218532562257, "memory(GiB)": 70.5, "step": 75410, "token_acc": 0.9015873015873016, "train_speed(iter/s)": 1.451501 }, { "epoch": 3.2310098110620795, "grad_norm": 3.089284658432007, "learning_rate": 2.7833647731104284e-05, "loss": 0.33015496730804444, "memory(GiB)": 70.5, "step": 75415, "token_acc": 0.9188311688311688, "train_speed(iter/s)": 1.451503 }, { "epoch": 3.2312240263913283, "grad_norm": 3.248432159423828, "learning_rate": 2.7827615632982462e-05, "loss": 0.28604879379272463, "memory(GiB)": 70.5, "step": 75420, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.451511 }, { "epoch": 3.2314382417205776, "grad_norm": 2.0939388275146484, "learning_rate": 2.782158393653419e-05, "loss": 0.251406192779541, "memory(GiB)": 70.5, "step": 75425, "token_acc": 0.9568345323741008, "train_speed(iter/s)": 1.45151 }, { "epoch": 3.2316524570498264, "grad_norm": 4.207325458526611, "learning_rate": 2.7815552641868703e-05, "loss": 0.35194573402404783, "memory(GiB)": 70.5, "step": 75430, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.451526 }, { "epoch": 3.231866672379075, "grad_norm": 2.057314157485962, "learning_rate": 2.78095217490953e-05, "loss": 0.5321663856506348, "memory(GiB)": 70.5, "step": 75435, "token_acc": 0.8868613138686131, "train_speed(iter/s)": 1.451536 }, { "epoch": 3.2320808877083245, "grad_norm": 1.7218208312988281, "learning_rate": 2.7803491258323224e-05, "loss": 0.5015110969543457, "memory(GiB)": 70.5, "step": 75440, "token_acc": 0.8827361563517915, "train_speed(iter/s)": 1.451543 }, { "epoch": 3.2322951030375733, "grad_norm": 2.0915684700012207, "learning_rate": 2.7797461169661735e-05, "loss": 0.18941234350204467, "memory(GiB)": 70.5, "step": 75445, "token_acc": 0.9575289575289575, "train_speed(iter/s)": 1.451552 }, { "epoch": 3.2325093183668225, "grad_norm": 6.1857829093933105, "learning_rate": 2.779143148322007e-05, "loss": 0.2995100498199463, "memory(GiB)": 70.5, "step": 75450, "token_acc": 0.9331103678929766, "train_speed(iter/s)": 1.451564 }, { "epoch": 3.2327235336960713, "grad_norm": 2.7394487857818604, "learning_rate": 2.7785402199107456e-05, "loss": 0.25883395671844484, "memory(GiB)": 70.5, "step": 75455, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.451562 }, { "epoch": 3.23293774902532, "grad_norm": 5.137842178344727, "learning_rate": 2.7779373317433122e-05, "loss": 0.3596503734588623, "memory(GiB)": 70.5, "step": 75460, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.451563 }, { "epoch": 3.2331519643545694, "grad_norm": 2.9010369777679443, "learning_rate": 2.777334483830626e-05, "loss": 0.2417844295501709, "memory(GiB)": 70.5, "step": 75465, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.451577 }, { "epoch": 3.2333661796838182, "grad_norm": 2.670588731765747, "learning_rate": 2.7767316761836125e-05, "loss": 0.2514598846435547, "memory(GiB)": 70.5, "step": 75470, "token_acc": 0.942652329749104, "train_speed(iter/s)": 1.451581 }, { "epoch": 3.233580395013067, "grad_norm": 4.466032028198242, "learning_rate": 2.7761289088131916e-05, "loss": 0.6576560020446778, "memory(GiB)": 70.5, "step": 75475, "token_acc": 0.8588588588588588, "train_speed(iter/s)": 1.45158 }, { "epoch": 3.2337946103423163, "grad_norm": 1.6880244016647339, "learning_rate": 2.77552618173028e-05, "loss": 0.5644524097442627, "memory(GiB)": 70.5, "step": 75480, "token_acc": 0.88, "train_speed(iter/s)": 1.451583 }, { "epoch": 3.234008825671565, "grad_norm": 5.0584716796875, "learning_rate": 2.7749234949457992e-05, "loss": 0.21197071075439453, "memory(GiB)": 70.5, "step": 75485, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.451596 }, { "epoch": 3.234223041000814, "grad_norm": 1.7449092864990234, "learning_rate": 2.7743208484706667e-05, "loss": 0.2744066953659058, "memory(GiB)": 70.5, "step": 75490, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.451599 }, { "epoch": 3.234437256330063, "grad_norm": 2.2986185550689697, "learning_rate": 2.7737182423157982e-05, "loss": 0.24439573287963867, "memory(GiB)": 70.5, "step": 75495, "token_acc": 0.9535714285714286, "train_speed(iter/s)": 1.451603 }, { "epoch": 3.234651471659312, "grad_norm": 4.945136547088623, "learning_rate": 2.7731156764921112e-05, "loss": 0.4335947513580322, "memory(GiB)": 70.5, "step": 75500, "token_acc": 0.898876404494382, "train_speed(iter/s)": 1.451611 }, { "epoch": 3.234651471659312, "eval_loss": 2.529121160507202, "eval_runtime": 10.9573, "eval_samples_per_second": 9.126, "eval_steps_per_second": 9.126, "eval_token_acc": 0.45594405594405596, "step": 75500 }, { "epoch": 3.234865686988561, "grad_norm": 3.0232255458831787, "learning_rate": 2.772513151010525e-05, "loss": 0.2197362184524536, "memory(GiB)": 70.5, "step": 75505, "token_acc": 0.58004158004158, "train_speed(iter/s)": 1.451276 }, { "epoch": 3.23507990231781, "grad_norm": 3.3081634044647217, "learning_rate": 2.7719106658819526e-05, "loss": 0.44962825775146487, "memory(GiB)": 70.5, "step": 75510, "token_acc": 0.9069069069069069, "train_speed(iter/s)": 1.451273 }, { "epoch": 3.235294117647059, "grad_norm": 2.545828342437744, "learning_rate": 2.771308221117309e-05, "loss": 0.20519671440124512, "memory(GiB)": 70.5, "step": 75515, "token_acc": 0.954225352112676, "train_speed(iter/s)": 1.451273 }, { "epoch": 3.2355083329763077, "grad_norm": 3.8612217903137207, "learning_rate": 2.7707058167275075e-05, "loss": 0.5690186977386474, "memory(GiB)": 70.5, "step": 75520, "token_acc": 0.8675496688741722, "train_speed(iter/s)": 1.451284 }, { "epoch": 3.235722548305557, "grad_norm": 2.6745824813842773, "learning_rate": 2.770103452723459e-05, "loss": 0.4755393981933594, "memory(GiB)": 70.5, "step": 75525, "token_acc": 0.9009009009009009, "train_speed(iter/s)": 1.451283 }, { "epoch": 3.2359367636348058, "grad_norm": 2.7501933574676514, "learning_rate": 2.769501129116081e-05, "loss": 0.4048442840576172, "memory(GiB)": 70.5, "step": 75530, "token_acc": 0.8930817610062893, "train_speed(iter/s)": 1.451287 }, { "epoch": 3.2361509789640546, "grad_norm": 2.588372230529785, "learning_rate": 2.7688988459162818e-05, "loss": 0.3342895030975342, "memory(GiB)": 70.5, "step": 75535, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.451293 }, { "epoch": 3.236365194293304, "grad_norm": 2.1585960388183594, "learning_rate": 2.7682966031349732e-05, "loss": 0.24490184783935548, "memory(GiB)": 70.5, "step": 75540, "token_acc": 0.9387096774193548, "train_speed(iter/s)": 1.451293 }, { "epoch": 3.2365794096225526, "grad_norm": 3.4170517921447754, "learning_rate": 2.7676944007830653e-05, "loss": 0.26085171699523924, "memory(GiB)": 70.5, "step": 75545, "token_acc": 0.9280575539568345, "train_speed(iter/s)": 1.451297 }, { "epoch": 3.2367936249518015, "grad_norm": 2.5120437145233154, "learning_rate": 2.767092238871467e-05, "loss": 0.4636964797973633, "memory(GiB)": 70.5, "step": 75550, "token_acc": 0.8948863636363636, "train_speed(iter/s)": 1.45131 }, { "epoch": 3.2370078402810507, "grad_norm": 0.147556334733963, "learning_rate": 2.7664901174110857e-05, "loss": 0.21711885929107666, "memory(GiB)": 70.5, "step": 75555, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.451318 }, { "epoch": 3.2372220556102995, "grad_norm": 0.35938239097595215, "learning_rate": 2.7658880364128326e-05, "loss": 0.3110125780105591, "memory(GiB)": 70.5, "step": 75560, "token_acc": 0.9404761904761905, "train_speed(iter/s)": 1.451328 }, { "epoch": 3.2374362709395483, "grad_norm": 1.356909155845642, "learning_rate": 2.7652859958876142e-05, "loss": 0.3047926902770996, "memory(GiB)": 70.5, "step": 75565, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.451349 }, { "epoch": 3.2376504862687976, "grad_norm": 5.754091739654541, "learning_rate": 2.764683995846336e-05, "loss": 0.36484925746917723, "memory(GiB)": 70.5, "step": 75570, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.45135 }, { "epoch": 3.2378647015980464, "grad_norm": 0.9414657354354858, "learning_rate": 2.764082036299902e-05, "loss": 0.25726709365844724, "memory(GiB)": 70.5, "step": 75575, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.451368 }, { "epoch": 3.238078916927295, "grad_norm": 2.9723637104034424, "learning_rate": 2.7634801172592217e-05, "loss": 0.5155729293823242, "memory(GiB)": 70.5, "step": 75580, "token_acc": 0.8807339449541285, "train_speed(iter/s)": 1.451372 }, { "epoch": 3.2382931322565445, "grad_norm": 1.8751178979873657, "learning_rate": 2.7628782387351948e-05, "loss": 0.1503794550895691, "memory(GiB)": 70.5, "step": 75585, "token_acc": 0.9631449631449631, "train_speed(iter/s)": 1.451385 }, { "epoch": 3.2385073475857933, "grad_norm": 1.5258194208145142, "learning_rate": 2.762276400738729e-05, "loss": 0.1860993981361389, "memory(GiB)": 70.5, "step": 75590, "token_acc": 0.9458333333333333, "train_speed(iter/s)": 1.451386 }, { "epoch": 3.238721562915042, "grad_norm": 4.962723255157471, "learning_rate": 2.7616746032807262e-05, "loss": 0.35512843132019045, "memory(GiB)": 70.5, "step": 75595, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.451388 }, { "epoch": 3.2389357782442914, "grad_norm": 5.507379531860352, "learning_rate": 2.761072846372087e-05, "loss": 0.2599926948547363, "memory(GiB)": 70.5, "step": 75600, "token_acc": 0.9368770764119602, "train_speed(iter/s)": 1.451404 }, { "epoch": 3.23914999357354, "grad_norm": 2.2790119647979736, "learning_rate": 2.760471130023714e-05, "loss": 0.3100806713104248, "memory(GiB)": 70.5, "step": 75605, "token_acc": 0.9492063492063492, "train_speed(iter/s)": 1.451421 }, { "epoch": 3.239364208902789, "grad_norm": 3.9354288578033447, "learning_rate": 2.7598694542465047e-05, "loss": 0.2228926658630371, "memory(GiB)": 70.5, "step": 75610, "token_acc": 0.9553264604810997, "train_speed(iter/s)": 1.451425 }, { "epoch": 3.2395784242320382, "grad_norm": 3.2070999145507812, "learning_rate": 2.7592678190513638e-05, "loss": 0.15209512710571288, "memory(GiB)": 70.5, "step": 75615, "token_acc": 0.9649122807017544, "train_speed(iter/s)": 1.451426 }, { "epoch": 3.239792639561287, "grad_norm": 5.589025974273682, "learning_rate": 2.7586662244491883e-05, "loss": 0.23962876796722413, "memory(GiB)": 70.5, "step": 75620, "token_acc": 0.9664179104477612, "train_speed(iter/s)": 1.451427 }, { "epoch": 3.240006854890536, "grad_norm": 2.8562707901000977, "learning_rate": 2.7580646704508772e-05, "loss": 0.16425036191940307, "memory(GiB)": 70.5, "step": 75625, "token_acc": 0.9584775086505191, "train_speed(iter/s)": 1.451439 }, { "epoch": 3.240221070219785, "grad_norm": 4.050625801086426, "learning_rate": 2.757463157067327e-05, "loss": 0.4966247081756592, "memory(GiB)": 70.5, "step": 75630, "token_acc": 0.882943143812709, "train_speed(iter/s)": 1.451448 }, { "epoch": 3.240435285549034, "grad_norm": 5.4198503494262695, "learning_rate": 2.756861684309436e-05, "loss": 0.3323625087738037, "memory(GiB)": 70.5, "step": 75635, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.451464 }, { "epoch": 3.2406495008782827, "grad_norm": 0.30669835209846497, "learning_rate": 2.7562602521880975e-05, "loss": 0.3383155584335327, "memory(GiB)": 70.5, "step": 75640, "token_acc": 0.924187725631769, "train_speed(iter/s)": 1.451463 }, { "epoch": 3.240863716207532, "grad_norm": 4.150321006774902, "learning_rate": 2.7556588607142107e-05, "loss": 0.374970531463623, "memory(GiB)": 70.5, "step": 75645, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.451479 }, { "epoch": 3.241077931536781, "grad_norm": 0.22038471698760986, "learning_rate": 2.7550575098986688e-05, "loss": 0.45179238319396975, "memory(GiB)": 70.5, "step": 75650, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.45148 }, { "epoch": 3.2412921468660296, "grad_norm": 0.9197000861167908, "learning_rate": 2.7544561997523665e-05, "loss": 0.18192824125289916, "memory(GiB)": 70.5, "step": 75655, "token_acc": 0.9572649572649573, "train_speed(iter/s)": 1.451488 }, { "epoch": 3.241506362195279, "grad_norm": 3.5915534496307373, "learning_rate": 2.753854930286197e-05, "loss": 0.48256716728210447, "memory(GiB)": 70.5, "step": 75660, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.4515 }, { "epoch": 3.2417205775245277, "grad_norm": 3.391273021697998, "learning_rate": 2.7532537015110528e-05, "loss": 0.35688629150390627, "memory(GiB)": 70.5, "step": 75665, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.4515 }, { "epoch": 3.2419347928537765, "grad_norm": 2.7456202507019043, "learning_rate": 2.752652513437823e-05, "loss": 0.23683226108551025, "memory(GiB)": 70.5, "step": 75670, "token_acc": 0.9471544715447154, "train_speed(iter/s)": 1.451502 }, { "epoch": 3.2421490081830258, "grad_norm": 3.3202965259552, "learning_rate": 2.7520513660774028e-05, "loss": 0.3414145469665527, "memory(GiB)": 70.5, "step": 75675, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.451506 }, { "epoch": 3.2423632235122746, "grad_norm": 3.100064277648926, "learning_rate": 2.7514502594406812e-05, "loss": 0.34651961326599123, "memory(GiB)": 70.5, "step": 75680, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.451504 }, { "epoch": 3.2425774388415234, "grad_norm": 3.7665271759033203, "learning_rate": 2.7508491935385478e-05, "loss": 0.32422442436218263, "memory(GiB)": 70.5, "step": 75685, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.451512 }, { "epoch": 3.2427916541707726, "grad_norm": 0.2175547480583191, "learning_rate": 2.750248168381891e-05, "loss": 0.2849047422409058, "memory(GiB)": 70.5, "step": 75690, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.451518 }, { "epoch": 3.2430058695000215, "grad_norm": 2.6292855739593506, "learning_rate": 2.7496471839815983e-05, "loss": 0.3163318634033203, "memory(GiB)": 70.5, "step": 75695, "token_acc": 0.9380281690140845, "train_speed(iter/s)": 1.451518 }, { "epoch": 3.2432200848292703, "grad_norm": 2.769347906112671, "learning_rate": 2.7490462403485574e-05, "loss": 0.54081130027771, "memory(GiB)": 70.5, "step": 75700, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.451541 }, { "epoch": 3.2434343001585195, "grad_norm": 1.7073606252670288, "learning_rate": 2.7484453374936558e-05, "loss": 0.3987556219100952, "memory(GiB)": 70.5, "step": 75705, "token_acc": 0.9373040752351097, "train_speed(iter/s)": 1.451557 }, { "epoch": 3.2436485154877683, "grad_norm": 1.9477957487106323, "learning_rate": 2.74784447542778e-05, "loss": 0.43057589530944823, "memory(GiB)": 70.5, "step": 75710, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.451558 }, { "epoch": 3.243862730817017, "grad_norm": 2.522026300430298, "learning_rate": 2.7472436541618142e-05, "loss": 0.2138500213623047, "memory(GiB)": 70.5, "step": 75715, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.451564 }, { "epoch": 3.2440769461462664, "grad_norm": 3.1785237789154053, "learning_rate": 2.7466428737066426e-05, "loss": 0.23321690559387206, "memory(GiB)": 70.5, "step": 75720, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.451572 }, { "epoch": 3.244291161475515, "grad_norm": 4.973398685455322, "learning_rate": 2.7460421340731468e-05, "loss": 0.4850743293762207, "memory(GiB)": 70.5, "step": 75725, "token_acc": 0.8992537313432836, "train_speed(iter/s)": 1.451581 }, { "epoch": 3.244505376804764, "grad_norm": 3.492300033569336, "learning_rate": 2.7454414352722128e-05, "loss": 0.38610448837280276, "memory(GiB)": 70.5, "step": 75730, "token_acc": 0.9081967213114754, "train_speed(iter/s)": 1.451592 }, { "epoch": 3.2447195921340133, "grad_norm": 2.026979684829712, "learning_rate": 2.7448407773147238e-05, "loss": 0.4490792751312256, "memory(GiB)": 70.5, "step": 75735, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.451592 }, { "epoch": 3.244933807463262, "grad_norm": 4.928522109985352, "learning_rate": 2.7442401602115596e-05, "loss": 0.45957112312316895, "memory(GiB)": 70.5, "step": 75740, "token_acc": 0.9158576051779935, "train_speed(iter/s)": 1.451595 }, { "epoch": 3.245148022792511, "grad_norm": 3.610081672668457, "learning_rate": 2.7436395839736016e-05, "loss": 0.6258733749389649, "memory(GiB)": 70.5, "step": 75745, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.45162 }, { "epoch": 3.24536223812176, "grad_norm": 0.3534820079803467, "learning_rate": 2.7430390486117284e-05, "loss": 0.17650516033172609, "memory(GiB)": 70.5, "step": 75750, "token_acc": 0.9671532846715328, "train_speed(iter/s)": 1.451619 }, { "epoch": 3.245576453451009, "grad_norm": 2.3934712409973145, "learning_rate": 2.742438554136818e-05, "loss": 0.1969342350959778, "memory(GiB)": 70.5, "step": 75755, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.45162 }, { "epoch": 3.245790668780258, "grad_norm": 2.1072094440460205, "learning_rate": 2.741838100559754e-05, "loss": 0.28892228603363035, "memory(GiB)": 70.5, "step": 75760, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.451626 }, { "epoch": 3.246004884109507, "grad_norm": 2.076842784881592, "learning_rate": 2.7412376878914105e-05, "loss": 0.21032447814941407, "memory(GiB)": 70.5, "step": 75765, "token_acc": 0.9547038327526133, "train_speed(iter/s)": 1.451634 }, { "epoch": 3.246219099438756, "grad_norm": 2.749467372894287, "learning_rate": 2.7406373161426647e-05, "loss": 0.5677361488342285, "memory(GiB)": 70.5, "step": 75770, "token_acc": 0.8761609907120743, "train_speed(iter/s)": 1.451642 }, { "epoch": 3.2464333147680047, "grad_norm": 2.143228769302368, "learning_rate": 2.7400369853243936e-05, "loss": 0.3563957691192627, "memory(GiB)": 70.5, "step": 75775, "token_acc": 0.9149659863945578, "train_speed(iter/s)": 1.451638 }, { "epoch": 3.246647530097254, "grad_norm": 0.052310552448034286, "learning_rate": 2.7394366954474726e-05, "loss": 0.47988409996032716, "memory(GiB)": 70.5, "step": 75780, "token_acc": 0.9108910891089109, "train_speed(iter/s)": 1.45164 }, { "epoch": 3.2468617454265027, "grad_norm": 4.674689769744873, "learning_rate": 2.7388364465227746e-05, "loss": 0.6528712272644043, "memory(GiB)": 70.5, "step": 75785, "token_acc": 0.8862068965517241, "train_speed(iter/s)": 1.451645 }, { "epoch": 3.2470759607557516, "grad_norm": 2.469425678253174, "learning_rate": 2.738236238561177e-05, "loss": 0.27686712741851804, "memory(GiB)": 70.5, "step": 75790, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.451647 }, { "epoch": 3.247290176085001, "grad_norm": 0.48147332668304443, "learning_rate": 2.7376360715735527e-05, "loss": 0.3273857831954956, "memory(GiB)": 70.5, "step": 75795, "token_acc": 0.9383116883116883, "train_speed(iter/s)": 1.451653 }, { "epoch": 3.2475043914142496, "grad_norm": 5.789987087249756, "learning_rate": 2.7370359455707706e-05, "loss": 0.48539342880249026, "memory(GiB)": 70.5, "step": 75800, "token_acc": 0.9006849315068494, "train_speed(iter/s)": 1.451658 }, { "epoch": 3.2477186067434984, "grad_norm": 2.4188015460968018, "learning_rate": 2.736435860563707e-05, "loss": 0.20534048080444336, "memory(GiB)": 70.5, "step": 75805, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.451666 }, { "epoch": 3.2479328220727477, "grad_norm": 5.161648750305176, "learning_rate": 2.7358358165632313e-05, "loss": 0.1888630509376526, "memory(GiB)": 70.5, "step": 75810, "token_acc": 0.959409594095941, "train_speed(iter/s)": 1.451674 }, { "epoch": 3.2481470374019965, "grad_norm": 3.1224586963653564, "learning_rate": 2.7352358135802113e-05, "loss": 0.5582102298736572, "memory(GiB)": 70.5, "step": 75815, "token_acc": 0.8757961783439491, "train_speed(iter/s)": 1.451678 }, { "epoch": 3.2483612527312453, "grad_norm": 4.197760105133057, "learning_rate": 2.7346358516255222e-05, "loss": 0.4212174415588379, "memory(GiB)": 70.5, "step": 75820, "token_acc": 0.9088145896656535, "train_speed(iter/s)": 1.451692 }, { "epoch": 3.2485754680604946, "grad_norm": 1.9539834260940552, "learning_rate": 2.7340359307100287e-05, "loss": 0.3794163703918457, "memory(GiB)": 70.5, "step": 75825, "token_acc": 0.90625, "train_speed(iter/s)": 1.451706 }, { "epoch": 3.2487896833897434, "grad_norm": 2.360140323638916, "learning_rate": 2.7334360508446005e-05, "loss": 0.40594944953918455, "memory(GiB)": 70.5, "step": 75830, "token_acc": 0.9058823529411765, "train_speed(iter/s)": 1.451713 }, { "epoch": 3.249003898718992, "grad_norm": 2.3468973636627197, "learning_rate": 2.732836212040104e-05, "loss": 0.3227132797241211, "memory(GiB)": 70.5, "step": 75835, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.451714 }, { "epoch": 3.2492181140482415, "grad_norm": 2.5965211391448975, "learning_rate": 2.732236414307407e-05, "loss": 0.5598494529724121, "memory(GiB)": 70.5, "step": 75840, "token_acc": 0.8813559322033898, "train_speed(iter/s)": 1.451719 }, { "epoch": 3.2494323293774903, "grad_norm": 4.661868095397949, "learning_rate": 2.731636657657372e-05, "loss": 0.5643983840942383, "memory(GiB)": 70.5, "step": 75845, "token_acc": 0.916923076923077, "train_speed(iter/s)": 1.451726 }, { "epoch": 3.249646544706739, "grad_norm": 2.0008387565612793, "learning_rate": 2.731036942100869e-05, "loss": 0.25427539348602296, "memory(GiB)": 70.5, "step": 75850, "token_acc": 0.9325153374233128, "train_speed(iter/s)": 1.451733 }, { "epoch": 3.2498607600359883, "grad_norm": 0.5089155435562134, "learning_rate": 2.7304372676487606e-05, "loss": 0.37240622043609617, "memory(GiB)": 70.5, "step": 75855, "token_acc": 0.9191176470588235, "train_speed(iter/s)": 1.451735 }, { "epoch": 3.250074975365237, "grad_norm": 0.3083784282207489, "learning_rate": 2.72983763431191e-05, "loss": 0.42049803733825686, "memory(GiB)": 70.5, "step": 75860, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.451742 }, { "epoch": 3.250289190694486, "grad_norm": 3.144010066986084, "learning_rate": 2.72923804210118e-05, "loss": 0.427443790435791, "memory(GiB)": 70.5, "step": 75865, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.451746 }, { "epoch": 3.250503406023735, "grad_norm": 3.2281579971313477, "learning_rate": 2.728638491027431e-05, "loss": 0.54694504737854, "memory(GiB)": 70.5, "step": 75870, "token_acc": 0.8858131487889274, "train_speed(iter/s)": 1.451746 }, { "epoch": 3.250717621352984, "grad_norm": 3.2166032791137695, "learning_rate": 2.7280389811015267e-05, "loss": 0.36889848709106443, "memory(GiB)": 70.5, "step": 75875, "token_acc": 0.937007874015748, "train_speed(iter/s)": 1.45176 }, { "epoch": 3.250931836682233, "grad_norm": 2.0017123222351074, "learning_rate": 2.7274395123343295e-05, "loss": 0.17859561443328859, "memory(GiB)": 70.5, "step": 75880, "token_acc": 0.9622093023255814, "train_speed(iter/s)": 1.451764 }, { "epoch": 3.251146052011482, "grad_norm": 3.093571424484253, "learning_rate": 2.726840084736697e-05, "loss": 0.2922536849975586, "memory(GiB)": 70.5, "step": 75885, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.45177 }, { "epoch": 3.251360267340731, "grad_norm": 5.360564231872559, "learning_rate": 2.7262406983194894e-05, "loss": 0.4428237438201904, "memory(GiB)": 70.5, "step": 75890, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.451779 }, { "epoch": 3.2515744826699797, "grad_norm": 3.2918221950531006, "learning_rate": 2.725641353093564e-05, "loss": 0.3866992235183716, "memory(GiB)": 70.5, "step": 75895, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.45179 }, { "epoch": 3.251788697999229, "grad_norm": 4.3132758140563965, "learning_rate": 2.725042049069777e-05, "loss": 0.38684544563293455, "memory(GiB)": 70.5, "step": 75900, "token_acc": 0.9059561128526645, "train_speed(iter/s)": 1.451788 }, { "epoch": 3.252002913328478, "grad_norm": 8.421517372131348, "learning_rate": 2.7244427862589893e-05, "loss": 0.32124838829040525, "memory(GiB)": 70.5, "step": 75905, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.451791 }, { "epoch": 3.2522171286577266, "grad_norm": 1.5827997922897339, "learning_rate": 2.723843564672056e-05, "loss": 0.15559942722320558, "memory(GiB)": 70.5, "step": 75910, "token_acc": 0.9576547231270358, "train_speed(iter/s)": 1.451804 }, { "epoch": 3.252431343986976, "grad_norm": 2.703334331512451, "learning_rate": 2.723244384319831e-05, "loss": 0.46973366737365724, "memory(GiB)": 70.5, "step": 75915, "token_acc": 0.8983739837398373, "train_speed(iter/s)": 1.451818 }, { "epoch": 3.2526455593162247, "grad_norm": 2.061962127685547, "learning_rate": 2.7226452452131703e-05, "loss": 0.31015868186950685, "memory(GiB)": 70.5, "step": 75920, "token_acc": 0.9298245614035088, "train_speed(iter/s)": 1.451823 }, { "epoch": 3.2528597746454735, "grad_norm": 4.537320613861084, "learning_rate": 2.7220461473629265e-05, "loss": 0.29909043312072753, "memory(GiB)": 70.5, "step": 75925, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.451838 }, { "epoch": 3.2530739899747227, "grad_norm": 3.905756950378418, "learning_rate": 2.7214470907799517e-05, "loss": 0.2790122747421265, "memory(GiB)": 70.5, "step": 75930, "token_acc": 0.9378531073446328, "train_speed(iter/s)": 1.451852 }, { "epoch": 3.2532882053039716, "grad_norm": 4.332085609436035, "learning_rate": 2.7208480754751033e-05, "loss": 0.2520740985870361, "memory(GiB)": 70.5, "step": 75935, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.451848 }, { "epoch": 3.2535024206332204, "grad_norm": 4.2112579345703125, "learning_rate": 2.7202491014592292e-05, "loss": 0.18306574821472169, "memory(GiB)": 70.5, "step": 75940, "token_acc": 0.9547038327526133, "train_speed(iter/s)": 1.45185 }, { "epoch": 3.2537166359624696, "grad_norm": 4.026095390319824, "learning_rate": 2.719650168743179e-05, "loss": 0.3673809051513672, "memory(GiB)": 70.5, "step": 75945, "token_acc": 0.9291338582677166, "train_speed(iter/s)": 1.451858 }, { "epoch": 3.2539308512917184, "grad_norm": 2.262763738632202, "learning_rate": 2.7190512773378075e-05, "loss": 0.31252250671386717, "memory(GiB)": 70.5, "step": 75950, "token_acc": 0.9433333333333334, "train_speed(iter/s)": 1.451862 }, { "epoch": 3.2541450666209673, "grad_norm": 3.34683895111084, "learning_rate": 2.718452427253962e-05, "loss": 0.5140592098236084, "memory(GiB)": 70.5, "step": 75955, "token_acc": 0.8962655601659751, "train_speed(iter/s)": 1.451862 }, { "epoch": 3.2543592819502165, "grad_norm": 5.129993438720703, "learning_rate": 2.717853618502489e-05, "loss": 0.3451760053634644, "memory(GiB)": 70.5, "step": 75960, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.451875 }, { "epoch": 3.2545734972794653, "grad_norm": 4.899639129638672, "learning_rate": 2.7172548510942402e-05, "loss": 0.26637156009674073, "memory(GiB)": 70.5, "step": 75965, "token_acc": 0.9586776859504132, "train_speed(iter/s)": 1.451896 }, { "epoch": 3.254787712608714, "grad_norm": 2.37540864944458, "learning_rate": 2.7166561250400613e-05, "loss": 0.25063228607177734, "memory(GiB)": 70.5, "step": 75970, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.451898 }, { "epoch": 3.2550019279379634, "grad_norm": 0.6421253681182861, "learning_rate": 2.7160574403507988e-05, "loss": 0.7282151222229004, "memory(GiB)": 70.5, "step": 75975, "token_acc": 0.8641509433962264, "train_speed(iter/s)": 1.451904 }, { "epoch": 3.255216143267212, "grad_norm": 0.4150952994823456, "learning_rate": 2.715458797037298e-05, "loss": 0.23336663246154785, "memory(GiB)": 70.5, "step": 75980, "token_acc": 0.9623287671232876, "train_speed(iter/s)": 1.451906 }, { "epoch": 3.255430358596461, "grad_norm": 2.741637945175171, "learning_rate": 2.7148601951104046e-05, "loss": 0.4006908416748047, "memory(GiB)": 70.5, "step": 75985, "token_acc": 0.9, "train_speed(iter/s)": 1.451915 }, { "epoch": 3.2556445739257103, "grad_norm": 3.649942636489868, "learning_rate": 2.71426163458096e-05, "loss": 0.2945849418640137, "memory(GiB)": 70.5, "step": 75990, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.451918 }, { "epoch": 3.255858789254959, "grad_norm": 4.436227321624756, "learning_rate": 2.7136631154598124e-05, "loss": 0.21772444248199463, "memory(GiB)": 70.5, "step": 75995, "token_acc": 0.930327868852459, "train_speed(iter/s)": 1.45192 }, { "epoch": 3.256073004584208, "grad_norm": 2.4202494621276855, "learning_rate": 2.7130646377578018e-05, "loss": 0.13818323612213135, "memory(GiB)": 70.5, "step": 76000, "token_acc": 0.9652173913043478, "train_speed(iter/s)": 1.451924 }, { "epoch": 3.256073004584208, "eval_loss": 2.7528741359710693, "eval_runtime": 10.7677, "eval_samples_per_second": 9.287, "eval_steps_per_second": 9.287, "eval_token_acc": 0.4204685573366215, "step": 76000 }, { "epoch": 3.256287219913457, "grad_norm": 2.8344476222991943, "learning_rate": 2.712466201485771e-05, "loss": 0.31244349479675293, "memory(GiB)": 70.5, "step": 76005, "token_acc": 0.5276967930029155, "train_speed(iter/s)": 1.451598 }, { "epoch": 3.256501435242706, "grad_norm": 4.319677352905273, "learning_rate": 2.7118678066545607e-05, "loss": 0.37259368896484374, "memory(GiB)": 70.5, "step": 76010, "token_acc": 0.9376947040498442, "train_speed(iter/s)": 1.451612 }, { "epoch": 3.256715650571955, "grad_norm": 1.321218490600586, "learning_rate": 2.7112694532750092e-05, "loss": 0.121258544921875, "memory(GiB)": 70.5, "step": 76015, "token_acc": 0.9652509652509652, "train_speed(iter/s)": 1.451611 }, { "epoch": 3.256929865901204, "grad_norm": 3.2012507915496826, "learning_rate": 2.7106711413579587e-05, "loss": 0.37652790546417236, "memory(GiB)": 70.5, "step": 76020, "token_acc": 0.9262295081967213, "train_speed(iter/s)": 1.451622 }, { "epoch": 3.257144081230453, "grad_norm": 3.975170135498047, "learning_rate": 2.710072870914251e-05, "loss": 0.16690773963928224, "memory(GiB)": 70.5, "step": 76025, "token_acc": 0.9560439560439561, "train_speed(iter/s)": 1.451621 }, { "epoch": 3.2573582965597017, "grad_norm": 4.332321643829346, "learning_rate": 2.7094746419547202e-05, "loss": 0.40117650032043456, "memory(GiB)": 70.5, "step": 76030, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.45162 }, { "epoch": 3.257572511888951, "grad_norm": 7.016833305358887, "learning_rate": 2.7088764544902057e-05, "loss": 0.6588433265686036, "memory(GiB)": 70.5, "step": 76035, "token_acc": 0.865625, "train_speed(iter/s)": 1.451617 }, { "epoch": 3.2577867272181997, "grad_norm": 5.606475353240967, "learning_rate": 2.7082783085315426e-05, "loss": 0.3907916069030762, "memory(GiB)": 70.5, "step": 76040, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.451616 }, { "epoch": 3.2580009425474485, "grad_norm": 1.588653326034546, "learning_rate": 2.7076802040895666e-05, "loss": 0.24637329578399658, "memory(GiB)": 70.5, "step": 76045, "token_acc": 0.9574468085106383, "train_speed(iter/s)": 1.451618 }, { "epoch": 3.258215157876698, "grad_norm": 2.572835683822632, "learning_rate": 2.7070821411751158e-05, "loss": 0.49542951583862305, "memory(GiB)": 70.5, "step": 76050, "token_acc": 0.8914473684210527, "train_speed(iter/s)": 1.451624 }, { "epoch": 3.2584293732059466, "grad_norm": 1.0195342302322388, "learning_rate": 2.7064841197990225e-05, "loss": 0.508385705947876, "memory(GiB)": 70.5, "step": 76055, "token_acc": 0.8915254237288136, "train_speed(iter/s)": 1.451642 }, { "epoch": 3.2586435885351954, "grad_norm": 4.122791290283203, "learning_rate": 2.7058861399721213e-05, "loss": 0.4218748092651367, "memory(GiB)": 70.5, "step": 76060, "token_acc": 0.9096045197740112, "train_speed(iter/s)": 1.45165 }, { "epoch": 3.2588578038644447, "grad_norm": 1.5667649507522583, "learning_rate": 2.705288201705245e-05, "loss": 0.5531343460083008, "memory(GiB)": 70.5, "step": 76065, "token_acc": 0.8730650154798761, "train_speed(iter/s)": 1.451656 }, { "epoch": 3.2590720191936935, "grad_norm": 2.693119764328003, "learning_rate": 2.704690305009225e-05, "loss": 0.2880615949630737, "memory(GiB)": 70.5, "step": 76070, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.451654 }, { "epoch": 3.2592862345229423, "grad_norm": 2.768310070037842, "learning_rate": 2.7040924498948918e-05, "loss": 0.47836718559265134, "memory(GiB)": 70.5, "step": 76075, "token_acc": 0.9008498583569405, "train_speed(iter/s)": 1.451663 }, { "epoch": 3.2595004498521916, "grad_norm": 0.47821715474128723, "learning_rate": 2.703494636373079e-05, "loss": 0.21651859283447267, "memory(GiB)": 70.5, "step": 76080, "token_acc": 0.951310861423221, "train_speed(iter/s)": 1.451669 }, { "epoch": 3.2597146651814404, "grad_norm": 5.336543083190918, "learning_rate": 2.7028968644546154e-05, "loss": 0.5504088401794434, "memory(GiB)": 70.5, "step": 76085, "token_acc": 0.897887323943662, "train_speed(iter/s)": 1.451682 }, { "epoch": 3.259928880510689, "grad_norm": 7.244588851928711, "learning_rate": 2.7022991341503302e-05, "loss": 0.27094545364379885, "memory(GiB)": 70.5, "step": 76090, "token_acc": 0.9426751592356688, "train_speed(iter/s)": 1.451687 }, { "epoch": 3.2601430958399384, "grad_norm": 1.690824270248413, "learning_rate": 2.7017014454710492e-05, "loss": 0.3042689323425293, "memory(GiB)": 70.5, "step": 76095, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.45169 }, { "epoch": 3.2603573111691873, "grad_norm": 5.757138729095459, "learning_rate": 2.7011037984276043e-05, "loss": 0.3144578456878662, "memory(GiB)": 70.5, "step": 76100, "token_acc": 0.9233716475095786, "train_speed(iter/s)": 1.451694 }, { "epoch": 3.260571526498436, "grad_norm": 3.4832613468170166, "learning_rate": 2.7005061930308195e-05, "loss": 0.3834530353546143, "memory(GiB)": 70.5, "step": 76105, "token_acc": 0.9008264462809917, "train_speed(iter/s)": 1.451695 }, { "epoch": 3.2607857418276853, "grad_norm": 0.08475606143474579, "learning_rate": 2.6999086292915234e-05, "loss": 0.20022280216217042, "memory(GiB)": 70.5, "step": 76110, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.451692 }, { "epoch": 3.260999957156934, "grad_norm": 1.38406240940094, "learning_rate": 2.6993111072205406e-05, "loss": 0.3430408239364624, "memory(GiB)": 70.5, "step": 76115, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.451695 }, { "epoch": 3.261214172486183, "grad_norm": 0.2587694227695465, "learning_rate": 2.698713626828695e-05, "loss": 0.04526030123233795, "memory(GiB)": 70.5, "step": 76120, "token_acc": 0.9897610921501706, "train_speed(iter/s)": 1.451707 }, { "epoch": 3.261428387815432, "grad_norm": 4.082998752593994, "learning_rate": 2.6981161881268113e-05, "loss": 0.35506424903869627, "memory(GiB)": 70.5, "step": 76125, "token_acc": 0.9212598425196851, "train_speed(iter/s)": 1.45172 }, { "epoch": 3.261642603144681, "grad_norm": 4.38040018081665, "learning_rate": 2.6975187911257116e-05, "loss": 0.28040204048156736, "memory(GiB)": 70.5, "step": 76130, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.451737 }, { "epoch": 3.26185681847393, "grad_norm": 2.644629716873169, "learning_rate": 2.696921435836217e-05, "loss": 0.2507516384124756, "memory(GiB)": 70.5, "step": 76135, "token_acc": 0.934375, "train_speed(iter/s)": 1.451734 }, { "epoch": 3.262071033803179, "grad_norm": 1.5814002752304077, "learning_rate": 2.6963241222691533e-05, "loss": 0.19630749225616456, "memory(GiB)": 70.5, "step": 76140, "token_acc": 0.9453376205787781, "train_speed(iter/s)": 1.451731 }, { "epoch": 3.262285249132428, "grad_norm": 5.04544734954834, "learning_rate": 2.6957268504353394e-05, "loss": 0.4517465114593506, "memory(GiB)": 70.5, "step": 76145, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.451739 }, { "epoch": 3.2624994644616767, "grad_norm": 2.4770517349243164, "learning_rate": 2.6951296203455945e-05, "loss": 0.43717098236083984, "memory(GiB)": 70.5, "step": 76150, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.451741 }, { "epoch": 3.262713679790926, "grad_norm": 3.619467258453369, "learning_rate": 2.694532432010739e-05, "loss": 0.4201968669891357, "memory(GiB)": 70.5, "step": 76155, "token_acc": 0.9042553191489362, "train_speed(iter/s)": 1.451742 }, { "epoch": 3.262927895120175, "grad_norm": 0.24301782250404358, "learning_rate": 2.693935285441589e-05, "loss": 0.10318347215652465, "memory(GiB)": 70.5, "step": 76160, "token_acc": 0.981549815498155, "train_speed(iter/s)": 1.451739 }, { "epoch": 3.2631421104494236, "grad_norm": 3.2053658962249756, "learning_rate": 2.693338180648965e-05, "loss": 0.30606684684753416, "memory(GiB)": 70.5, "step": 76165, "token_acc": 0.921311475409836, "train_speed(iter/s)": 1.451744 }, { "epoch": 3.263356325778673, "grad_norm": 2.063267230987549, "learning_rate": 2.6927411176436857e-05, "loss": 0.4004330635070801, "memory(GiB)": 70.5, "step": 76170, "token_acc": 0.9223880597014925, "train_speed(iter/s)": 1.451744 }, { "epoch": 3.2635705411079217, "grad_norm": 5.491952896118164, "learning_rate": 2.692144096436565e-05, "loss": 0.6407248497009277, "memory(GiB)": 70.5, "step": 76175, "token_acc": 0.8610271903323263, "train_speed(iter/s)": 1.451743 }, { "epoch": 3.2637847564371705, "grad_norm": 2.151737928390503, "learning_rate": 2.6915471170384188e-05, "loss": 0.18359043598175048, "memory(GiB)": 70.5, "step": 76180, "token_acc": 0.955719557195572, "train_speed(iter/s)": 1.451748 }, { "epoch": 3.2639989717664197, "grad_norm": 2.48724365234375, "learning_rate": 2.6909501794600622e-05, "loss": 0.2843721866607666, "memory(GiB)": 70.5, "step": 76185, "token_acc": 0.9429530201342282, "train_speed(iter/s)": 1.451756 }, { "epoch": 3.2642131870956685, "grad_norm": 2.741041421890259, "learning_rate": 2.690353283712308e-05, "loss": 0.29929242134094236, "memory(GiB)": 70.5, "step": 76190, "token_acc": 0.9223300970873787, "train_speed(iter/s)": 1.451762 }, { "epoch": 3.2644274024249174, "grad_norm": 0.19481870532035828, "learning_rate": 2.689756429805972e-05, "loss": 0.2940845489501953, "memory(GiB)": 70.5, "step": 76195, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.451768 }, { "epoch": 3.2646416177541666, "grad_norm": 3.515092134475708, "learning_rate": 2.6891596177518652e-05, "loss": 0.37049615383148193, "memory(GiB)": 70.5, "step": 76200, "token_acc": 0.9011857707509882, "train_speed(iter/s)": 1.451787 }, { "epoch": 3.2648558330834154, "grad_norm": 2.0847530364990234, "learning_rate": 2.6885628475608006e-05, "loss": 0.29238803386688234, "memory(GiB)": 70.5, "step": 76205, "token_acc": 0.9368770764119602, "train_speed(iter/s)": 1.451792 }, { "epoch": 3.2650700484126642, "grad_norm": 6.533871173858643, "learning_rate": 2.6879661192435866e-05, "loss": 0.35365309715271, "memory(GiB)": 70.5, "step": 76210, "token_acc": 0.9211356466876972, "train_speed(iter/s)": 1.451792 }, { "epoch": 3.2652842637419135, "grad_norm": 0.716683030128479, "learning_rate": 2.6873694328110365e-05, "loss": 0.44804606437683103, "memory(GiB)": 70.5, "step": 76215, "token_acc": 0.9507042253521126, "train_speed(iter/s)": 1.4518 }, { "epoch": 3.2654984790711623, "grad_norm": 6.531966686248779, "learning_rate": 2.6867727882739568e-05, "loss": 0.38264830112457277, "memory(GiB)": 70.5, "step": 76220, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.451799 }, { "epoch": 3.265712694400411, "grad_norm": 4.732046127319336, "learning_rate": 2.6861761856431595e-05, "loss": 0.14369401931762696, "memory(GiB)": 70.5, "step": 76225, "token_acc": 0.968503937007874, "train_speed(iter/s)": 1.451796 }, { "epoch": 3.2659269097296604, "grad_norm": 4.1569623947143555, "learning_rate": 2.6855796249294512e-05, "loss": 0.2958106279373169, "memory(GiB)": 70.5, "step": 76230, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.451797 }, { "epoch": 3.266141125058909, "grad_norm": 2.3334619998931885, "learning_rate": 2.6849831061436394e-05, "loss": 0.3202864408493042, "memory(GiB)": 70.5, "step": 76235, "token_acc": 0.925, "train_speed(iter/s)": 1.451806 }, { "epoch": 3.266355340388158, "grad_norm": 3.5411336421966553, "learning_rate": 2.684386629296528e-05, "loss": 0.5692065715789795, "memory(GiB)": 70.5, "step": 76240, "token_acc": 0.8753799392097265, "train_speed(iter/s)": 1.451828 }, { "epoch": 3.2665695557174073, "grad_norm": 4.243320465087891, "learning_rate": 2.683790194398927e-05, "loss": 0.30570416450500487, "memory(GiB)": 70.5, "step": 76245, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.451833 }, { "epoch": 3.266783771046656, "grad_norm": 5.010836124420166, "learning_rate": 2.6831938014616377e-05, "loss": 0.3511080503463745, "memory(GiB)": 70.5, "step": 76250, "token_acc": 0.9388489208633094, "train_speed(iter/s)": 1.451832 }, { "epoch": 3.266997986375905, "grad_norm": 2.6087164878845215, "learning_rate": 2.6825974504954676e-05, "loss": 0.37510342597961427, "memory(GiB)": 70.5, "step": 76255, "token_acc": 0.9241379310344827, "train_speed(iter/s)": 1.451841 }, { "epoch": 3.267212201705154, "grad_norm": 1.9517488479614258, "learning_rate": 2.6820011415112183e-05, "loss": 0.3994414806365967, "memory(GiB)": 70.5, "step": 76260, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.451859 }, { "epoch": 3.267426417034403, "grad_norm": 3.945502758026123, "learning_rate": 2.6814048745196933e-05, "loss": 0.5241445541381836, "memory(GiB)": 70.5, "step": 76265, "token_acc": 0.8848684210526315, "train_speed(iter/s)": 1.451862 }, { "epoch": 3.2676406323636518, "grad_norm": 2.4876866340637207, "learning_rate": 2.6808086495316943e-05, "loss": 0.4586804389953613, "memory(GiB)": 70.5, "step": 76270, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.451864 }, { "epoch": 3.267854847692901, "grad_norm": 5.508580207824707, "learning_rate": 2.68021246655802e-05, "loss": 0.4147289276123047, "memory(GiB)": 70.5, "step": 76275, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.451865 }, { "epoch": 3.26806906302215, "grad_norm": 3.642164945602417, "learning_rate": 2.679616325609474e-05, "loss": 0.20950989723205565, "memory(GiB)": 70.5, "step": 76280, "token_acc": 0.9496644295302014, "train_speed(iter/s)": 1.451867 }, { "epoch": 3.2682832783513986, "grad_norm": 3.44710373878479, "learning_rate": 2.679020226696856e-05, "loss": 0.32474002838134763, "memory(GiB)": 70.5, "step": 76285, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.451868 }, { "epoch": 3.268497493680648, "grad_norm": 3.669088840484619, "learning_rate": 2.678424169830964e-05, "loss": 0.34071803092956543, "memory(GiB)": 70.5, "step": 76290, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.451883 }, { "epoch": 3.2687117090098967, "grad_norm": 9.639293670654297, "learning_rate": 2.6778281550225957e-05, "loss": 0.5248431205749512, "memory(GiB)": 70.5, "step": 76295, "token_acc": 0.8829113924050633, "train_speed(iter/s)": 1.451887 }, { "epoch": 3.2689259243391455, "grad_norm": 0.9854648113250732, "learning_rate": 2.6772321822825486e-05, "loss": 0.22946412563323976, "memory(GiB)": 70.5, "step": 76300, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.451889 }, { "epoch": 3.269140139668395, "grad_norm": 2.6481409072875977, "learning_rate": 2.6766362516216177e-05, "loss": 0.1619524598121643, "memory(GiB)": 70.5, "step": 76305, "token_acc": 0.9685314685314685, "train_speed(iter/s)": 1.451893 }, { "epoch": 3.2693543549976436, "grad_norm": 2.569620132446289, "learning_rate": 2.6760403630506027e-05, "loss": 0.6142435073852539, "memory(GiB)": 70.5, "step": 76310, "token_acc": 0.8851351351351351, "train_speed(iter/s)": 1.451891 }, { "epoch": 3.2695685703268924, "grad_norm": 2.0251073837280273, "learning_rate": 2.6754445165802938e-05, "loss": 0.5830944538116455, "memory(GiB)": 70.5, "step": 76315, "token_acc": 0.8838709677419355, "train_speed(iter/s)": 1.451895 }, { "epoch": 3.2697827856561417, "grad_norm": 4.288359642028809, "learning_rate": 2.6748487122214905e-05, "loss": 0.36306066513061525, "memory(GiB)": 70.5, "step": 76320, "token_acc": 0.905511811023622, "train_speed(iter/s)": 1.451907 }, { "epoch": 3.2699970009853905, "grad_norm": 2.472191572189331, "learning_rate": 2.6742529499849846e-05, "loss": 0.517972755432129, "memory(GiB)": 70.5, "step": 76325, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.451911 }, { "epoch": 3.2702112163146393, "grad_norm": 3.6178696155548096, "learning_rate": 2.673657229881567e-05, "loss": 0.424453067779541, "memory(GiB)": 70.5, "step": 76330, "token_acc": 0.9064327485380117, "train_speed(iter/s)": 1.451932 }, { "epoch": 3.2704254316438885, "grad_norm": 1.9958457946777344, "learning_rate": 2.67306155192203e-05, "loss": 0.41380987167358396, "memory(GiB)": 70.5, "step": 76335, "token_acc": 0.9125, "train_speed(iter/s)": 1.451937 }, { "epoch": 3.2706396469731374, "grad_norm": 3.182771921157837, "learning_rate": 2.6724659161171674e-05, "loss": 0.2921538591384888, "memory(GiB)": 70.5, "step": 76340, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.451941 }, { "epoch": 3.270853862302386, "grad_norm": 8.254858016967773, "learning_rate": 2.671870322477768e-05, "loss": 0.43375911712646487, "memory(GiB)": 70.5, "step": 76345, "token_acc": 0.896875, "train_speed(iter/s)": 1.451943 }, { "epoch": 3.2710680776316354, "grad_norm": 10.355396270751953, "learning_rate": 2.6712747710146224e-05, "loss": 0.6105273246765137, "memory(GiB)": 70.5, "step": 76350, "token_acc": 0.8895705521472392, "train_speed(iter/s)": 1.451941 }, { "epoch": 3.2712822929608842, "grad_norm": 3.891303300857544, "learning_rate": 2.670679261738518e-05, "loss": 0.14992096424102783, "memory(GiB)": 70.5, "step": 76355, "token_acc": 0.973384030418251, "train_speed(iter/s)": 1.451941 }, { "epoch": 3.271496508290133, "grad_norm": 2.4190688133239746, "learning_rate": 2.670083794660244e-05, "loss": 0.535509443283081, "memory(GiB)": 70.5, "step": 76360, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.451957 }, { "epoch": 3.2717107236193823, "grad_norm": 3.6338918209075928, "learning_rate": 2.669488369790586e-05, "loss": 0.4157379627227783, "memory(GiB)": 70.5, "step": 76365, "token_acc": 0.8958990536277602, "train_speed(iter/s)": 1.451958 }, { "epoch": 3.271924938948631, "grad_norm": 0.24921394884586334, "learning_rate": 2.6688929871403346e-05, "loss": 0.39438045024871826, "memory(GiB)": 70.5, "step": 76370, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.451965 }, { "epoch": 3.27213915427788, "grad_norm": 1.541229248046875, "learning_rate": 2.6682976467202726e-05, "loss": 0.16443777084350586, "memory(GiB)": 70.5, "step": 76375, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.45198 }, { "epoch": 3.272353369607129, "grad_norm": 0.059872761368751526, "learning_rate": 2.6677023485411866e-05, "loss": 0.21179277896881105, "memory(GiB)": 70.5, "step": 76380, "token_acc": 0.959409594095941, "train_speed(iter/s)": 1.451984 }, { "epoch": 3.272567584936378, "grad_norm": 4.766190052032471, "learning_rate": 2.667107092613861e-05, "loss": 0.2856011867523193, "memory(GiB)": 70.5, "step": 76385, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.451975 }, { "epoch": 3.272781800265627, "grad_norm": 5.7105231285095215, "learning_rate": 2.6665118789490763e-05, "loss": 0.5275274753570557, "memory(GiB)": 70.5, "step": 76390, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.451971 }, { "epoch": 3.272996015594876, "grad_norm": 5.018089771270752, "learning_rate": 2.6659167075576176e-05, "loss": 0.3310786247253418, "memory(GiB)": 70.5, "step": 76395, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.451982 }, { "epoch": 3.273210230924125, "grad_norm": 4.217703342437744, "learning_rate": 2.6653215784502693e-05, "loss": 0.3090616226196289, "memory(GiB)": 70.5, "step": 76400, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.451996 }, { "epoch": 3.2734244462533737, "grad_norm": 8.256301879882812, "learning_rate": 2.664726491637811e-05, "loss": 0.4097901344299316, "memory(GiB)": 70.5, "step": 76405, "token_acc": 0.9, "train_speed(iter/s)": 1.451997 }, { "epoch": 3.273638661582623, "grad_norm": 4.529444217681885, "learning_rate": 2.664131447131023e-05, "loss": 0.31313748359680177, "memory(GiB)": 70.5, "step": 76410, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.451991 }, { "epoch": 3.2738528769118718, "grad_norm": 4.802251815795898, "learning_rate": 2.6635364449406853e-05, "loss": 0.3521687984466553, "memory(GiB)": 70.5, "step": 76415, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.451994 }, { "epoch": 3.274067092241121, "grad_norm": 3.0835447311401367, "learning_rate": 2.662941485077574e-05, "loss": 0.4410865306854248, "memory(GiB)": 70.5, "step": 76420, "token_acc": 0.9107142857142857, "train_speed(iter/s)": 1.451996 }, { "epoch": 3.27428130757037, "grad_norm": 2.945716142654419, "learning_rate": 2.6623465675524728e-05, "loss": 0.533375597000122, "memory(GiB)": 70.5, "step": 76425, "token_acc": 0.8991596638655462, "train_speed(iter/s)": 1.452002 }, { "epoch": 3.2744955228996186, "grad_norm": 0.5498405694961548, "learning_rate": 2.6617516923761553e-05, "loss": 0.21074562072753905, "memory(GiB)": 70.5, "step": 76430, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.452002 }, { "epoch": 3.274709738228868, "grad_norm": 4.22842264175415, "learning_rate": 2.6611568595594006e-05, "loss": 0.21281490325927735, "memory(GiB)": 70.5, "step": 76435, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.45202 }, { "epoch": 3.2749239535581167, "grad_norm": 4.649778842926025, "learning_rate": 2.6605620691129828e-05, "loss": 0.27886433601379396, "memory(GiB)": 70.5, "step": 76440, "token_acc": 0.9418604651162791, "train_speed(iter/s)": 1.452022 }, { "epoch": 3.2751381688873655, "grad_norm": 4.489305019378662, "learning_rate": 2.659967321047678e-05, "loss": 0.42623534202575686, "memory(GiB)": 70.5, "step": 76445, "token_acc": 0.9003690036900369, "train_speed(iter/s)": 1.452038 }, { "epoch": 3.275352384216615, "grad_norm": 6.451976776123047, "learning_rate": 2.6593726153742575e-05, "loss": 0.4903857231140137, "memory(GiB)": 70.5, "step": 76450, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.452037 }, { "epoch": 3.2755665995458636, "grad_norm": 0.18716081976890564, "learning_rate": 2.6587779521035007e-05, "loss": 0.33517019748687743, "memory(GiB)": 70.5, "step": 76455, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.452049 }, { "epoch": 3.2757808148751124, "grad_norm": 0.4254010021686554, "learning_rate": 2.6581833312461768e-05, "loss": 0.47787837982177733, "memory(GiB)": 70.5, "step": 76460, "token_acc": 0.8980891719745223, "train_speed(iter/s)": 1.452062 }, { "epoch": 3.2759950302043617, "grad_norm": 4.7963948249816895, "learning_rate": 2.657588752813057e-05, "loss": 0.4076075077056885, "memory(GiB)": 70.5, "step": 76465, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.45206 }, { "epoch": 3.2762092455336105, "grad_norm": 5.772552490234375, "learning_rate": 2.6569942168149165e-05, "loss": 0.1987290859222412, "memory(GiB)": 70.5, "step": 76470, "token_acc": 0.9522184300341296, "train_speed(iter/s)": 1.452074 }, { "epoch": 3.2764234608628593, "grad_norm": 0.7666959762573242, "learning_rate": 2.6563997232625236e-05, "loss": 0.7133056163787842, "memory(GiB)": 70.5, "step": 76475, "token_acc": 0.8697068403908795, "train_speed(iter/s)": 1.452098 }, { "epoch": 3.2766376761921086, "grad_norm": 5.8526811599731445, "learning_rate": 2.655805272166646e-05, "loss": 0.30223860740661623, "memory(GiB)": 70.5, "step": 76480, "token_acc": 0.9274193548387096, "train_speed(iter/s)": 1.452113 }, { "epoch": 3.2768518915213574, "grad_norm": 3.8514111042022705, "learning_rate": 2.6552108635380567e-05, "loss": 0.2495481252670288, "memory(GiB)": 70.5, "step": 76485, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.452114 }, { "epoch": 3.277066106850606, "grad_norm": 3.3813552856445312, "learning_rate": 2.654616497387523e-05, "loss": 0.36240808963775634, "memory(GiB)": 70.5, "step": 76490, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.45211 }, { "epoch": 3.2772803221798554, "grad_norm": 2.458725929260254, "learning_rate": 2.654022173725811e-05, "loss": 0.1408071756362915, "memory(GiB)": 70.5, "step": 76495, "token_acc": 0.9577922077922078, "train_speed(iter/s)": 1.452111 }, { "epoch": 3.2774945375091042, "grad_norm": 7.479637145996094, "learning_rate": 2.653427892563688e-05, "loss": 0.7017494201660156, "memory(GiB)": 70.5, "step": 76500, "token_acc": 0.8717948717948718, "train_speed(iter/s)": 1.452111 }, { "epoch": 3.2774945375091042, "eval_loss": 2.387399673461914, "eval_runtime": 11.1039, "eval_samples_per_second": 9.006, "eval_steps_per_second": 9.006, "eval_token_acc": 0.45614035087719296, "step": 76500 }, { "epoch": 3.277708752838353, "grad_norm": 5.025294780731201, "learning_rate": 2.65283365391192e-05, "loss": 0.3781342744827271, "memory(GiB)": 70.5, "step": 76505, "token_acc": 0.5981220657276995, "train_speed(iter/s)": 1.451788 }, { "epoch": 3.2779229681676023, "grad_norm": 6.616147994995117, "learning_rate": 2.6522394577812702e-05, "loss": 0.6299065589904785, "memory(GiB)": 70.5, "step": 76510, "token_acc": 0.8616352201257862, "train_speed(iter/s)": 1.451803 }, { "epoch": 3.278137183496851, "grad_norm": 0.2163451611995697, "learning_rate": 2.6516453041825075e-05, "loss": 0.2646117448806763, "memory(GiB)": 70.5, "step": 76515, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.451804 }, { "epoch": 3.2783513988261, "grad_norm": 3.065812587738037, "learning_rate": 2.6510511931263926e-05, "loss": 0.553142786026001, "memory(GiB)": 70.5, "step": 76520, "token_acc": 0.8975069252077562, "train_speed(iter/s)": 1.451819 }, { "epoch": 3.278565614155349, "grad_norm": 2.4777002334594727, "learning_rate": 2.6504571246236893e-05, "loss": 0.4359261989593506, "memory(GiB)": 70.5, "step": 76525, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.451833 }, { "epoch": 3.278779829484598, "grad_norm": 2.9784586429595947, "learning_rate": 2.6498630986851584e-05, "loss": 0.3177149772644043, "memory(GiB)": 70.5, "step": 76530, "token_acc": 0.8956521739130435, "train_speed(iter/s)": 1.451848 }, { "epoch": 3.278994044813847, "grad_norm": 1.1861964464187622, "learning_rate": 2.649269115321561e-05, "loss": 0.2966787576675415, "memory(GiB)": 70.5, "step": 76535, "token_acc": 0.9477124183006536, "train_speed(iter/s)": 1.451851 }, { "epoch": 3.279208260143096, "grad_norm": 4.319001197814941, "learning_rate": 2.6486751745436578e-05, "loss": 0.7464372634887695, "memory(GiB)": 70.5, "step": 76540, "token_acc": 0.8413284132841329, "train_speed(iter/s)": 1.451868 }, { "epoch": 3.279422475472345, "grad_norm": 4.873758792877197, "learning_rate": 2.648081276362212e-05, "loss": 0.4464472770690918, "memory(GiB)": 70.5, "step": 76545, "token_acc": 0.9, "train_speed(iter/s)": 1.45187 }, { "epoch": 3.2796366908015937, "grad_norm": 4.635267734527588, "learning_rate": 2.6474874207879807e-05, "loss": 0.17822663784027098, "memory(GiB)": 70.5, "step": 76550, "token_acc": 0.9598393574297188, "train_speed(iter/s)": 1.451869 }, { "epoch": 3.279850906130843, "grad_norm": 7.223546504974365, "learning_rate": 2.6468936078317207e-05, "loss": 0.5652945518493653, "memory(GiB)": 70.5, "step": 76555, "token_acc": 0.875, "train_speed(iter/s)": 1.451871 }, { "epoch": 3.2800651214600918, "grad_norm": 2.4301297664642334, "learning_rate": 2.6462998375041908e-05, "loss": 0.29809749126434326, "memory(GiB)": 70.5, "step": 76560, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.451875 }, { "epoch": 3.2802793367893406, "grad_norm": 2.853715658187866, "learning_rate": 2.645706109816145e-05, "loss": 0.31579885482788084, "memory(GiB)": 70.5, "step": 76565, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.451878 }, { "epoch": 3.28049355211859, "grad_norm": 4.7010498046875, "learning_rate": 2.6451124247783443e-05, "loss": 0.35991334915161133, "memory(GiB)": 70.5, "step": 76570, "token_acc": 0.9262295081967213, "train_speed(iter/s)": 1.451881 }, { "epoch": 3.2807077674478387, "grad_norm": 2.2927815914154053, "learning_rate": 2.6445187824015406e-05, "loss": 0.28517200946807864, "memory(GiB)": 70.5, "step": 76575, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.451882 }, { "epoch": 3.2809219827770875, "grad_norm": 1.7922147512435913, "learning_rate": 2.6439251826964882e-05, "loss": 0.43421158790588377, "memory(GiB)": 70.5, "step": 76580, "token_acc": 0.9129129129129129, "train_speed(iter/s)": 1.451887 }, { "epoch": 3.2811361981063367, "grad_norm": 3.6542603969573975, "learning_rate": 2.6433316256739417e-05, "loss": 0.3685925960540771, "memory(GiB)": 70.5, "step": 76585, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.451896 }, { "epoch": 3.2813504134355855, "grad_norm": 2.498288154602051, "learning_rate": 2.6427381113446536e-05, "loss": 0.52473464012146, "memory(GiB)": 70.5, "step": 76590, "token_acc": 0.8885793871866295, "train_speed(iter/s)": 1.45191 }, { "epoch": 3.2815646287648343, "grad_norm": 3.175996780395508, "learning_rate": 2.642144639719374e-05, "loss": 0.34852075576782227, "memory(GiB)": 70.5, "step": 76595, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.451912 }, { "epoch": 3.2817788440940836, "grad_norm": 2.172394037246704, "learning_rate": 2.641551210808858e-05, "loss": 0.5031616687774658, "memory(GiB)": 70.5, "step": 76600, "token_acc": 0.8988326848249028, "train_speed(iter/s)": 1.451924 }, { "epoch": 3.2819930594233324, "grad_norm": 3.5148324966430664, "learning_rate": 2.640957824623854e-05, "loss": 0.35151722431182864, "memory(GiB)": 70.5, "step": 76605, "token_acc": 0.9060402684563759, "train_speed(iter/s)": 1.451929 }, { "epoch": 3.2822072747525812, "grad_norm": 2.3919434547424316, "learning_rate": 2.6403644811751104e-05, "loss": 0.48991875648498534, "memory(GiB)": 70.5, "step": 76610, "token_acc": 0.8802588996763754, "train_speed(iter/s)": 1.451938 }, { "epoch": 3.2824214900818305, "grad_norm": 5.275424957275391, "learning_rate": 2.63977118047338e-05, "loss": 0.5485272884368897, "memory(GiB)": 70.5, "step": 76615, "token_acc": 0.9039735099337748, "train_speed(iter/s)": 1.45195 }, { "epoch": 3.2826357054110793, "grad_norm": 2.9855806827545166, "learning_rate": 2.6391779225294084e-05, "loss": 0.36270482540130616, "memory(GiB)": 70.5, "step": 76620, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.451968 }, { "epoch": 3.282849920740328, "grad_norm": 0.06336402148008347, "learning_rate": 2.638584707353941e-05, "loss": 0.24460160732269287, "memory(GiB)": 70.5, "step": 76625, "token_acc": 0.9515570934256056, "train_speed(iter/s)": 1.451977 }, { "epoch": 3.2830641360695774, "grad_norm": 3.13659405708313, "learning_rate": 2.637991534957729e-05, "loss": 0.4573357582092285, "memory(GiB)": 70.5, "step": 76630, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.451986 }, { "epoch": 3.283278351398826, "grad_norm": 1.8359402418136597, "learning_rate": 2.6373984053515167e-05, "loss": 0.2951441764831543, "memory(GiB)": 70.5, "step": 76635, "token_acc": 0.9372822299651568, "train_speed(iter/s)": 1.45201 }, { "epoch": 3.283492566728075, "grad_norm": 5.8910322189331055, "learning_rate": 2.636805318546048e-05, "loss": 0.3293306350708008, "memory(GiB)": 70.5, "step": 76640, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.452011 }, { "epoch": 3.2837067820573242, "grad_norm": 10.332794189453125, "learning_rate": 2.636212274552068e-05, "loss": 0.3114823579788208, "memory(GiB)": 70.5, "step": 76645, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.452011 }, { "epoch": 3.283920997386573, "grad_norm": 3.0490899085998535, "learning_rate": 2.6356192733803197e-05, "loss": 0.23594608306884765, "memory(GiB)": 70.5, "step": 76650, "token_acc": 0.9571865443425076, "train_speed(iter/s)": 1.452018 }, { "epoch": 3.284135212715822, "grad_norm": 0.11844010651111603, "learning_rate": 2.6350263150415443e-05, "loss": 0.3598337650299072, "memory(GiB)": 70.5, "step": 76655, "token_acc": 0.9312714776632303, "train_speed(iter/s)": 1.45202 }, { "epoch": 3.284349428045071, "grad_norm": 0.9906735420227051, "learning_rate": 2.6344333995464875e-05, "loss": 0.2674895763397217, "memory(GiB)": 70.5, "step": 76660, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.452046 }, { "epoch": 3.28456364337432, "grad_norm": 3.0822699069976807, "learning_rate": 2.6338405269058896e-05, "loss": 0.2819631576538086, "memory(GiB)": 70.5, "step": 76665, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.452048 }, { "epoch": 3.2847778587035688, "grad_norm": 4.072748184204102, "learning_rate": 2.6332476971304898e-05, "loss": 0.4179238319396973, "memory(GiB)": 70.5, "step": 76670, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.452051 }, { "epoch": 3.284992074032818, "grad_norm": 4.245687007904053, "learning_rate": 2.6326549102310284e-05, "loss": 0.3083707094192505, "memory(GiB)": 70.5, "step": 76675, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.452049 }, { "epoch": 3.285206289362067, "grad_norm": 5.346494674682617, "learning_rate": 2.632062166218241e-05, "loss": 0.2953944683074951, "memory(GiB)": 70.5, "step": 76680, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.452055 }, { "epoch": 3.2854205046913156, "grad_norm": 4.618851661682129, "learning_rate": 2.6314694651028697e-05, "loss": 0.34159815311431885, "memory(GiB)": 70.5, "step": 76685, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.452063 }, { "epoch": 3.285634720020565, "grad_norm": 3.797084331512451, "learning_rate": 2.630876806895653e-05, "loss": 0.4620035171508789, "memory(GiB)": 70.5, "step": 76690, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.452074 }, { "epoch": 3.2858489353498137, "grad_norm": 3.868077516555786, "learning_rate": 2.630284191607325e-05, "loss": 0.1991954565048218, "memory(GiB)": 70.5, "step": 76695, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.45208 }, { "epoch": 3.2860631506790625, "grad_norm": 5.628716945648193, "learning_rate": 2.629691619248622e-05, "loss": 0.4369189262390137, "memory(GiB)": 70.5, "step": 76700, "token_acc": 0.9237668161434978, "train_speed(iter/s)": 1.452079 }, { "epoch": 3.2862773660083118, "grad_norm": 1.3731839656829834, "learning_rate": 2.6290990898302786e-05, "loss": 0.3892019987106323, "memory(GiB)": 70.5, "step": 76705, "token_acc": 0.9118773946360154, "train_speed(iter/s)": 1.452083 }, { "epoch": 3.2864915813375606, "grad_norm": 2.9734721183776855, "learning_rate": 2.6285066033630278e-05, "loss": 0.2454594612121582, "memory(GiB)": 70.5, "step": 76710, "token_acc": 0.959866220735786, "train_speed(iter/s)": 1.452091 }, { "epoch": 3.2867057966668094, "grad_norm": 2.747163772583008, "learning_rate": 2.6279141598576062e-05, "loss": 0.17326005697250366, "memory(GiB)": 70.5, "step": 76715, "token_acc": 0.9593220338983051, "train_speed(iter/s)": 1.452091 }, { "epoch": 3.2869200119960587, "grad_norm": 3.8622357845306396, "learning_rate": 2.6273217593247447e-05, "loss": 0.2732563495635986, "memory(GiB)": 70.5, "step": 76720, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.452099 }, { "epoch": 3.2871342273253075, "grad_norm": 2.2098424434661865, "learning_rate": 2.6267294017751753e-05, "loss": 0.330877947807312, "memory(GiB)": 70.5, "step": 76725, "token_acc": 0.9195046439628483, "train_speed(iter/s)": 1.452102 }, { "epoch": 3.2873484426545563, "grad_norm": 5.051963806152344, "learning_rate": 2.626137087219629e-05, "loss": 0.38532190322875975, "memory(GiB)": 70.5, "step": 76730, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.452104 }, { "epoch": 3.2875626579838055, "grad_norm": 3.3704986572265625, "learning_rate": 2.625544815668836e-05, "loss": 0.2078942060470581, "memory(GiB)": 70.5, "step": 76735, "token_acc": 0.9510703363914373, "train_speed(iter/s)": 1.452112 }, { "epoch": 3.2877768733130543, "grad_norm": 8.728253364562988, "learning_rate": 2.6249525871335246e-05, "loss": 0.29122438430786135, "memory(GiB)": 70.5, "step": 76740, "token_acc": 0.937007874015748, "train_speed(iter/s)": 1.452113 }, { "epoch": 3.287991088642303, "grad_norm": 3.5367636680603027, "learning_rate": 2.624360401624427e-05, "loss": 0.2793485403060913, "memory(GiB)": 70.5, "step": 76745, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.452115 }, { "epoch": 3.2882053039715524, "grad_norm": 3.3246164321899414, "learning_rate": 2.6237682591522693e-05, "loss": 0.30452260971069334, "memory(GiB)": 70.5, "step": 76750, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.452119 }, { "epoch": 3.2884195193008012, "grad_norm": 6.45590877532959, "learning_rate": 2.6231761597277783e-05, "loss": 0.5126235008239746, "memory(GiB)": 70.5, "step": 76755, "token_acc": 0.8670886075949367, "train_speed(iter/s)": 1.452117 }, { "epoch": 3.28863373463005, "grad_norm": 2.5275983810424805, "learning_rate": 2.6225841033616794e-05, "loss": 0.23394570350646973, "memory(GiB)": 70.5, "step": 76760, "token_acc": 0.9480968858131488, "train_speed(iter/s)": 1.452117 }, { "epoch": 3.2888479499592993, "grad_norm": 2.6395649909973145, "learning_rate": 2.621992090064701e-05, "loss": 0.23291027545928955, "memory(GiB)": 70.5, "step": 76765, "token_acc": 0.9530201342281879, "train_speed(iter/s)": 1.452119 }, { "epoch": 3.289062165288548, "grad_norm": 0.506517767906189, "learning_rate": 2.6214001198475645e-05, "loss": 0.04757034182548523, "memory(GiB)": 70.5, "step": 76770, "token_acc": 0.9823321554770318, "train_speed(iter/s)": 1.452119 }, { "epoch": 3.289276380617797, "grad_norm": 2.748030662536621, "learning_rate": 2.6208081927209988e-05, "loss": 0.6995439052581787, "memory(GiB)": 70.5, "step": 76775, "token_acc": 0.8647798742138365, "train_speed(iter/s)": 1.452122 }, { "epoch": 3.289490595947046, "grad_norm": 3.674467086791992, "learning_rate": 2.6202163086957243e-05, "loss": 0.35117411613464355, "memory(GiB)": 70.5, "step": 76780, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.452123 }, { "epoch": 3.289704811276295, "grad_norm": 4.410989761352539, "learning_rate": 2.6196244677824637e-05, "loss": 0.16702357530593873, "memory(GiB)": 70.5, "step": 76785, "token_acc": 0.9587628865979382, "train_speed(iter/s)": 1.452124 }, { "epoch": 3.289919026605544, "grad_norm": 0.5464633703231812, "learning_rate": 2.6190326699919388e-05, "loss": 0.3933538913726807, "memory(GiB)": 70.5, "step": 76790, "token_acc": 0.92578125, "train_speed(iter/s)": 1.452125 }, { "epoch": 3.290133241934793, "grad_norm": 1.6414282321929932, "learning_rate": 2.618440915334871e-05, "loss": 0.2762465000152588, "memory(GiB)": 70.5, "step": 76795, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.452124 }, { "epoch": 3.290347457264042, "grad_norm": 0.6464276909828186, "learning_rate": 2.617849203821978e-05, "loss": 0.2959075927734375, "memory(GiB)": 70.5, "step": 76800, "token_acc": 0.9578544061302682, "train_speed(iter/s)": 1.452141 }, { "epoch": 3.2905616725932907, "grad_norm": 6.258229732513428, "learning_rate": 2.617257535463983e-05, "loss": 0.547246789932251, "memory(GiB)": 70.5, "step": 76805, "token_acc": 0.9058823529411765, "train_speed(iter/s)": 1.452146 }, { "epoch": 3.29077588792254, "grad_norm": 3.300493001937866, "learning_rate": 2.616665910271603e-05, "loss": 0.26830918788909913, "memory(GiB)": 70.5, "step": 76810, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.452156 }, { "epoch": 3.2909901032517888, "grad_norm": 5.869393348693848, "learning_rate": 2.616074328255556e-05, "loss": 0.46480398178100585, "memory(GiB)": 70.5, "step": 76815, "token_acc": 0.9003690036900369, "train_speed(iter/s)": 1.452171 }, { "epoch": 3.2912043185810376, "grad_norm": 1.129615068435669, "learning_rate": 2.6154827894265588e-05, "loss": 0.2867327451705933, "memory(GiB)": 70.5, "step": 76820, "token_acc": 0.9446064139941691, "train_speed(iter/s)": 1.452173 }, { "epoch": 3.291418533910287, "grad_norm": 2.5732228755950928, "learning_rate": 2.6148912937953256e-05, "loss": 0.25867576599121095, "memory(GiB)": 70.5, "step": 76825, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.452176 }, { "epoch": 3.2916327492395356, "grad_norm": 3.1211678981781006, "learning_rate": 2.614299841372576e-05, "loss": 0.611177921295166, "memory(GiB)": 70.5, "step": 76830, "token_acc": 0.8801261829652997, "train_speed(iter/s)": 1.452178 }, { "epoch": 3.2918469645687845, "grad_norm": 2.4550585746765137, "learning_rate": 2.613708432169021e-05, "loss": 0.5895524978637695, "memory(GiB)": 70.5, "step": 76835, "token_acc": 0.875, "train_speed(iter/s)": 1.45219 }, { "epoch": 3.2920611798980337, "grad_norm": 3.2068796157836914, "learning_rate": 2.613117066195378e-05, "loss": 0.30170016288757323, "memory(GiB)": 70.5, "step": 76840, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.452188 }, { "epoch": 3.2922753952272825, "grad_norm": 6.669992923736572, "learning_rate": 2.6125257434623584e-05, "loss": 0.35001840591430666, "memory(GiB)": 70.5, "step": 76845, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.452197 }, { "epoch": 3.2924896105565313, "grad_norm": 1.1124095916748047, "learning_rate": 2.6119344639806753e-05, "loss": 0.5108280658721924, "memory(GiB)": 70.5, "step": 76850, "token_acc": 0.9054054054054054, "train_speed(iter/s)": 1.452196 }, { "epoch": 3.2927038258857806, "grad_norm": 3.59405779838562, "learning_rate": 2.6113432277610367e-05, "loss": 0.40539045333862306, "memory(GiB)": 70.5, "step": 76855, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.45219 }, { "epoch": 3.2929180412150294, "grad_norm": 4.316535949707031, "learning_rate": 2.6107520348141585e-05, "loss": 0.29770073890686033, "memory(GiB)": 70.5, "step": 76860, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.452211 }, { "epoch": 3.293132256544278, "grad_norm": 2.8732967376708984, "learning_rate": 2.6101608851507486e-05, "loss": 0.4522103309631348, "memory(GiB)": 70.5, "step": 76865, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.452211 }, { "epoch": 3.2933464718735275, "grad_norm": 5.229853630065918, "learning_rate": 2.609569778781516e-05, "loss": 0.3962454557418823, "memory(GiB)": 70.5, "step": 76870, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.452212 }, { "epoch": 3.2935606872027763, "grad_norm": 2.200648546218872, "learning_rate": 2.6089787157171697e-05, "loss": 0.3615330696105957, "memory(GiB)": 70.5, "step": 76875, "token_acc": 0.9238095238095239, "train_speed(iter/s)": 1.45222 }, { "epoch": 3.293774902532025, "grad_norm": 1.4156920909881592, "learning_rate": 2.6083876959684162e-05, "loss": 0.380196738243103, "memory(GiB)": 70.5, "step": 76880, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.452238 }, { "epoch": 3.2939891178612744, "grad_norm": 4.291169166564941, "learning_rate": 2.607796719545962e-05, "loss": 0.7108460903167725, "memory(GiB)": 70.5, "step": 76885, "token_acc": 0.8614864864864865, "train_speed(iter/s)": 1.452247 }, { "epoch": 3.294203333190523, "grad_norm": 3.8340587615966797, "learning_rate": 2.6072057864605147e-05, "loss": 0.31327099800109864, "memory(GiB)": 70.5, "step": 76890, "token_acc": 0.9243027888446215, "train_speed(iter/s)": 1.452257 }, { "epoch": 3.294417548519772, "grad_norm": 4.1326117515563965, "learning_rate": 2.606614896722781e-05, "loss": 0.5003059387207032, "memory(GiB)": 70.5, "step": 76895, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.452261 }, { "epoch": 3.2946317638490212, "grad_norm": 4.446069240570068, "learning_rate": 2.6060240503434623e-05, "loss": 0.4409774303436279, "memory(GiB)": 70.5, "step": 76900, "token_acc": 0.8893129770992366, "train_speed(iter/s)": 1.452271 }, { "epoch": 3.29484597917827, "grad_norm": 4.424976348876953, "learning_rate": 2.6054332473332622e-05, "loss": 0.4880673885345459, "memory(GiB)": 70.5, "step": 76905, "token_acc": 0.8819188191881919, "train_speed(iter/s)": 1.452272 }, { "epoch": 3.295060194507519, "grad_norm": 4.416528224945068, "learning_rate": 2.6048424877028876e-05, "loss": 0.31078195571899414, "memory(GiB)": 70.5, "step": 76910, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.452275 }, { "epoch": 3.295274409836768, "grad_norm": 4.617395401000977, "learning_rate": 2.6042517714630354e-05, "loss": 0.25372674465179446, "memory(GiB)": 70.5, "step": 76915, "token_acc": 0.9479166666666666, "train_speed(iter/s)": 1.452281 }, { "epoch": 3.295488625166017, "grad_norm": 2.935662269592285, "learning_rate": 2.6036610986244125e-05, "loss": 0.2922112464904785, "memory(GiB)": 70.5, "step": 76920, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.452295 }, { "epoch": 3.2957028404952657, "grad_norm": 3.857929229736328, "learning_rate": 2.6030704691977158e-05, "loss": 0.36679415702819823, "memory(GiB)": 70.5, "step": 76925, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.452309 }, { "epoch": 3.295917055824515, "grad_norm": 3.5675017833709717, "learning_rate": 2.602479883193647e-05, "loss": 0.6777045249938964, "memory(GiB)": 70.5, "step": 76930, "token_acc": 0.8606271777003485, "train_speed(iter/s)": 1.45231 }, { "epoch": 3.296131271153764, "grad_norm": 2.9051756858825684, "learning_rate": 2.6018893406229033e-05, "loss": 0.2318629026412964, "memory(GiB)": 70.5, "step": 76935, "token_acc": 0.9299363057324841, "train_speed(iter/s)": 1.452318 }, { "epoch": 3.2963454864830126, "grad_norm": 3.0396170616149902, "learning_rate": 2.6012988414961848e-05, "loss": 0.3649153709411621, "memory(GiB)": 70.5, "step": 76940, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.452321 }, { "epoch": 3.296559701812262, "grad_norm": 3.3418056964874268, "learning_rate": 2.6007083858241853e-05, "loss": 0.24148693084716796, "memory(GiB)": 70.5, "step": 76945, "token_acc": 0.952, "train_speed(iter/s)": 1.452326 }, { "epoch": 3.2967739171415107, "grad_norm": 3.745088815689087, "learning_rate": 2.6001179736176064e-05, "loss": 0.4761608123779297, "memory(GiB)": 70.5, "step": 76950, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.452333 }, { "epoch": 3.2969881324707595, "grad_norm": 5.574545860290527, "learning_rate": 2.599527604887141e-05, "loss": 0.41780743598937986, "memory(GiB)": 70.5, "step": 76955, "token_acc": 0.8932806324110671, "train_speed(iter/s)": 1.452343 }, { "epoch": 3.2972023478000088, "grad_norm": 2.1674506664276123, "learning_rate": 2.5989372796434854e-05, "loss": 0.5570093154907226, "memory(GiB)": 70.5, "step": 76960, "token_acc": 0.8875502008032129, "train_speed(iter/s)": 1.452357 }, { "epoch": 3.2974165631292576, "grad_norm": 2.0786712169647217, "learning_rate": 2.598346997897333e-05, "loss": 0.14633880853652953, "memory(GiB)": 70.5, "step": 76965, "token_acc": 0.9558823529411765, "train_speed(iter/s)": 1.452353 }, { "epoch": 3.2976307784585064, "grad_norm": 3.953307867050171, "learning_rate": 2.597756759659376e-05, "loss": 0.2768892288208008, "memory(GiB)": 70.5, "step": 76970, "token_acc": 0.9287833827893175, "train_speed(iter/s)": 1.452354 }, { "epoch": 3.2978449937877556, "grad_norm": 2.696359634399414, "learning_rate": 2.597166564940311e-05, "loss": 0.3215658664703369, "memory(GiB)": 70.5, "step": 76975, "token_acc": 0.947075208913649, "train_speed(iter/s)": 1.452367 }, { "epoch": 3.2980592091170045, "grad_norm": 0.7035775780677795, "learning_rate": 2.5965764137508253e-05, "loss": 0.1351390838623047, "memory(GiB)": 70.5, "step": 76980, "token_acc": 0.9627118644067797, "train_speed(iter/s)": 1.452373 }, { "epoch": 3.2982734244462533, "grad_norm": 2.4156672954559326, "learning_rate": 2.5959863061016144e-05, "loss": 0.3432953357696533, "memory(GiB)": 70.5, "step": 76985, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.452366 }, { "epoch": 3.2984876397755025, "grad_norm": 2.967818021774292, "learning_rate": 2.5953962420033673e-05, "loss": 0.40259556770324706, "memory(GiB)": 70.5, "step": 76990, "token_acc": 0.923728813559322, "train_speed(iter/s)": 1.452368 }, { "epoch": 3.2987018551047513, "grad_norm": 2.3434271812438965, "learning_rate": 2.5948062214667723e-05, "loss": 0.31866071224212644, "memory(GiB)": 70.5, "step": 76995, "token_acc": 0.9221556886227545, "train_speed(iter/s)": 1.452373 }, { "epoch": 3.298916070434, "grad_norm": 3.5670974254608154, "learning_rate": 2.5942162445025174e-05, "loss": 0.23844079971313475, "memory(GiB)": 70.5, "step": 77000, "token_acc": 0.9421487603305785, "train_speed(iter/s)": 1.45239 }, { "epoch": 3.298916070434, "eval_loss": 2.3783631324768066, "eval_runtime": 11.5419, "eval_samples_per_second": 8.664, "eval_steps_per_second": 8.664, "eval_token_acc": 0.450070323488045, "step": 77000 }, { "epoch": 3.2991302857632494, "grad_norm": 2.474433422088623, "learning_rate": 2.593626311121294e-05, "loss": 0.1481081247329712, "memory(GiB)": 70.5, "step": 77005, "token_acc": 0.6145733461169702, "train_speed(iter/s)": 1.452053 }, { "epoch": 3.299344501092498, "grad_norm": 2.587158203125, "learning_rate": 2.5930364213337865e-05, "loss": 0.5920118808746337, "memory(GiB)": 70.5, "step": 77010, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.452054 }, { "epoch": 3.299558716421747, "grad_norm": 6.768134117126465, "learning_rate": 2.592446575150683e-05, "loss": 0.4049673557281494, "memory(GiB)": 70.5, "step": 77015, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.452057 }, { "epoch": 3.2997729317509963, "grad_norm": 1.1538375616073608, "learning_rate": 2.5918567725826682e-05, "loss": 0.22710347175598145, "memory(GiB)": 70.5, "step": 77020, "token_acc": 0.9402390438247012, "train_speed(iter/s)": 1.452058 }, { "epoch": 3.299987147080245, "grad_norm": 2.184762477874756, "learning_rate": 2.591267013640427e-05, "loss": 0.3710577964782715, "memory(GiB)": 70.5, "step": 77025, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.452053 }, { "epoch": 3.300201362409494, "grad_norm": 1.7545151710510254, "learning_rate": 2.590677298334641e-05, "loss": 0.3122023820877075, "memory(GiB)": 70.5, "step": 77030, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.452055 }, { "epoch": 3.300415577738743, "grad_norm": 1.550054669380188, "learning_rate": 2.590087626675998e-05, "loss": 0.3167469263076782, "memory(GiB)": 70.5, "step": 77035, "token_acc": 0.9153605015673981, "train_speed(iter/s)": 1.452058 }, { "epoch": 3.300629793067992, "grad_norm": 5.710585594177246, "learning_rate": 2.589497998675179e-05, "loss": 0.31827120780944823, "memory(GiB)": 70.5, "step": 77040, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.452058 }, { "epoch": 3.300844008397241, "grad_norm": 0.5469594597816467, "learning_rate": 2.5889084143428643e-05, "loss": 0.2775467872619629, "memory(GiB)": 70.5, "step": 77045, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.452063 }, { "epoch": 3.30105822372649, "grad_norm": 2.9800915718078613, "learning_rate": 2.5883188736897356e-05, "loss": 0.24135997295379638, "memory(GiB)": 70.5, "step": 77050, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.452071 }, { "epoch": 3.301272439055739, "grad_norm": 4.771019458770752, "learning_rate": 2.587729376726471e-05, "loss": 0.3298182010650635, "memory(GiB)": 70.5, "step": 77055, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.452081 }, { "epoch": 3.3014866543849877, "grad_norm": 1.5624454021453857, "learning_rate": 2.587139923463751e-05, "loss": 0.2901656150817871, "memory(GiB)": 70.5, "step": 77060, "token_acc": 0.9375, "train_speed(iter/s)": 1.452084 }, { "epoch": 3.301700869714237, "grad_norm": 3.2429683208465576, "learning_rate": 2.586550513912257e-05, "loss": 0.43256301879882814, "memory(GiB)": 70.5, "step": 77065, "token_acc": 0.8896321070234113, "train_speed(iter/s)": 1.452102 }, { "epoch": 3.3019150850434857, "grad_norm": 4.336655139923096, "learning_rate": 2.585961148082665e-05, "loss": 0.4001461029052734, "memory(GiB)": 70.5, "step": 77070, "token_acc": 0.9012738853503185, "train_speed(iter/s)": 1.452104 }, { "epoch": 3.3021293003727346, "grad_norm": 4.892737865447998, "learning_rate": 2.5853718259856507e-05, "loss": 0.49856033325195315, "memory(GiB)": 70.5, "step": 77075, "token_acc": 0.8895705521472392, "train_speed(iter/s)": 1.4521 }, { "epoch": 3.302343515701984, "grad_norm": 3.0388927459716797, "learning_rate": 2.584782547631891e-05, "loss": 0.5360273838043212, "memory(GiB)": 70.5, "step": 77080, "token_acc": 0.878419452887538, "train_speed(iter/s)": 1.452103 }, { "epoch": 3.3025577310312326, "grad_norm": 3.964317560195923, "learning_rate": 2.5841933130320618e-05, "loss": 0.42726407051086424, "memory(GiB)": 70.5, "step": 77085, "token_acc": 0.886435331230284, "train_speed(iter/s)": 1.452116 }, { "epoch": 3.3027719463604814, "grad_norm": 5.260479927062988, "learning_rate": 2.5836041221968345e-05, "loss": 0.34369540214538574, "memory(GiB)": 70.5, "step": 77090, "token_acc": 0.9196675900277008, "train_speed(iter/s)": 1.452117 }, { "epoch": 3.3029861616897307, "grad_norm": 1.97653067111969, "learning_rate": 2.583014975136887e-05, "loss": 0.27677268981933595, "memory(GiB)": 70.5, "step": 77095, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.452122 }, { "epoch": 3.3032003770189795, "grad_norm": 1.978279948234558, "learning_rate": 2.5824258718628906e-05, "loss": 0.243666934967041, "memory(GiB)": 70.5, "step": 77100, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.45214 }, { "epoch": 3.3034145923482283, "grad_norm": 2.440368175506592, "learning_rate": 2.5818368123855176e-05, "loss": 0.5059046745300293, "memory(GiB)": 70.5, "step": 77105, "token_acc": 0.9015384615384615, "train_speed(iter/s)": 1.45214 }, { "epoch": 3.3036288076774776, "grad_norm": 5.485935211181641, "learning_rate": 2.581247796715439e-05, "loss": 0.48399858474731444, "memory(GiB)": 70.5, "step": 77110, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.452145 }, { "epoch": 3.3038430230067264, "grad_norm": 3.0370798110961914, "learning_rate": 2.5806588248633235e-05, "loss": 0.31580295562744143, "memory(GiB)": 70.5, "step": 77115, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.452149 }, { "epoch": 3.304057238335975, "grad_norm": 0.6982353329658508, "learning_rate": 2.580069896839845e-05, "loss": 0.41789703369140624, "memory(GiB)": 70.5, "step": 77120, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.452159 }, { "epoch": 3.3042714536652245, "grad_norm": 3.92846941947937, "learning_rate": 2.5794810126556707e-05, "loss": 0.4866915702819824, "memory(GiB)": 70.5, "step": 77125, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.452174 }, { "epoch": 3.3044856689944733, "grad_norm": 3.5864956378936768, "learning_rate": 2.5788921723214664e-05, "loss": 0.5444355010986328, "memory(GiB)": 70.5, "step": 77130, "token_acc": 0.8859649122807017, "train_speed(iter/s)": 1.452173 }, { "epoch": 3.304699884323722, "grad_norm": 3.8228840827941895, "learning_rate": 2.5783033758479035e-05, "loss": 0.435744047164917, "memory(GiB)": 70.5, "step": 77135, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.452179 }, { "epoch": 3.3049140996529713, "grad_norm": 4.838926792144775, "learning_rate": 2.5777146232456463e-05, "loss": 0.284867262840271, "memory(GiB)": 70.5, "step": 77140, "token_acc": 0.9315068493150684, "train_speed(iter/s)": 1.452181 }, { "epoch": 3.30512831498222, "grad_norm": 2.503333330154419, "learning_rate": 2.577125914525359e-05, "loss": 0.33821754455566405, "memory(GiB)": 70.5, "step": 77145, "token_acc": 0.9228571428571428, "train_speed(iter/s)": 1.452182 }, { "epoch": 3.305342530311469, "grad_norm": 2.8914456367492676, "learning_rate": 2.5765372496977113e-05, "loss": 0.24369351863861083, "memory(GiB)": 70.5, "step": 77150, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.452189 }, { "epoch": 3.305556745640718, "grad_norm": 2.9796035289764404, "learning_rate": 2.575948628773364e-05, "loss": 0.3653004884719849, "memory(GiB)": 70.5, "step": 77155, "token_acc": 0.9176829268292683, "train_speed(iter/s)": 1.452186 }, { "epoch": 3.305770960969967, "grad_norm": 2.828022003173828, "learning_rate": 2.5753600517629817e-05, "loss": 0.19531185626983644, "memory(GiB)": 70.5, "step": 77160, "token_acc": 0.9492063492063492, "train_speed(iter/s)": 1.452188 }, { "epoch": 3.305985176299216, "grad_norm": 2.4584479331970215, "learning_rate": 2.5747715186772264e-05, "loss": 0.17680854797363282, "memory(GiB)": 70.5, "step": 77165, "token_acc": 0.966789667896679, "train_speed(iter/s)": 1.452193 }, { "epoch": 3.306199391628465, "grad_norm": 4.247645854949951, "learning_rate": 2.5741830295267598e-05, "loss": 0.18952689170837403, "memory(GiB)": 70.5, "step": 77170, "token_acc": 0.9548872180451128, "train_speed(iter/s)": 1.452204 }, { "epoch": 3.306413606957714, "grad_norm": 5.772783279418945, "learning_rate": 2.573594584322242e-05, "loss": 0.34143548011779784, "memory(GiB)": 70.5, "step": 77175, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.452206 }, { "epoch": 3.3066278222869627, "grad_norm": 2.0898985862731934, "learning_rate": 2.5730061830743358e-05, "loss": 0.1653245210647583, "memory(GiB)": 70.5, "step": 77180, "token_acc": 0.9580152671755725, "train_speed(iter/s)": 1.452209 }, { "epoch": 3.306842037616212, "grad_norm": 2.9417989253997803, "learning_rate": 2.5724178257936992e-05, "loss": 0.34453089237213136, "memory(GiB)": 70.5, "step": 77185, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.452213 }, { "epoch": 3.307056252945461, "grad_norm": 6.46550178527832, "learning_rate": 2.5718295124909913e-05, "loss": 0.3173671245574951, "memory(GiB)": 70.5, "step": 77190, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.452224 }, { "epoch": 3.3072704682747096, "grad_norm": 1.334376573562622, "learning_rate": 2.5712412431768696e-05, "loss": 0.33625593185424807, "memory(GiB)": 70.5, "step": 77195, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.45223 }, { "epoch": 3.307484683603959, "grad_norm": 2.2443113327026367, "learning_rate": 2.5706530178619893e-05, "loss": 0.2027369976043701, "memory(GiB)": 70.5, "step": 77200, "token_acc": 0.9480968858131488, "train_speed(iter/s)": 1.452242 }, { "epoch": 3.3076988989332077, "grad_norm": 4.591334342956543, "learning_rate": 2.570064836557008e-05, "loss": 0.3058155536651611, "memory(GiB)": 70.5, "step": 77205, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.452246 }, { "epoch": 3.3079131142624565, "grad_norm": 3.273210048675537, "learning_rate": 2.5694766992725837e-05, "loss": 0.3939868450164795, "memory(GiB)": 70.5, "step": 77210, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.452243 }, { "epoch": 3.3081273295917057, "grad_norm": 3.3610966205596924, "learning_rate": 2.5688886060193694e-05, "loss": 0.44031782150268556, "memory(GiB)": 70.5, "step": 77215, "token_acc": 0.90625, "train_speed(iter/s)": 1.452249 }, { "epoch": 3.3083415449209546, "grad_norm": 4.294234752655029, "learning_rate": 2.5683005568080188e-05, "loss": 0.5030562877655029, "memory(GiB)": 70.5, "step": 77220, "token_acc": 0.8838028169014085, "train_speed(iter/s)": 1.452249 }, { "epoch": 3.3085557602502034, "grad_norm": 1.894299030303955, "learning_rate": 2.567712551649184e-05, "loss": 0.16843465566635132, "memory(GiB)": 70.5, "step": 77225, "token_acc": 0.9572368421052632, "train_speed(iter/s)": 1.452252 }, { "epoch": 3.3087699755794526, "grad_norm": 1.0773066282272339, "learning_rate": 2.567124590553518e-05, "loss": 0.5772853851318359, "memory(GiB)": 70.5, "step": 77230, "token_acc": 0.8961038961038961, "train_speed(iter/s)": 1.452256 }, { "epoch": 3.3089841909087014, "grad_norm": 3.472975254058838, "learning_rate": 2.5665366735316708e-05, "loss": 0.49607133865356445, "memory(GiB)": 70.5, "step": 77235, "token_acc": 0.8764044943820225, "train_speed(iter/s)": 1.45228 }, { "epoch": 3.3091984062379503, "grad_norm": 6.163510322570801, "learning_rate": 2.565948800594296e-05, "loss": 0.24466986656188966, "memory(GiB)": 70.5, "step": 77240, "token_acc": 0.94, "train_speed(iter/s)": 1.452282 }, { "epoch": 3.3094126215671995, "grad_norm": 5.961765766143799, "learning_rate": 2.5653609717520423e-05, "loss": 0.17700474262237548, "memory(GiB)": 70.5, "step": 77245, "token_acc": 0.9704918032786886, "train_speed(iter/s)": 1.452285 }, { "epoch": 3.3096268368964483, "grad_norm": 0.4393741190433502, "learning_rate": 2.5647731870155585e-05, "loss": 0.22028462886810302, "memory(GiB)": 70.5, "step": 77250, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.452287 }, { "epoch": 3.309841052225697, "grad_norm": 0.6925139427185059, "learning_rate": 2.5641854463954928e-05, "loss": 0.46458120346069337, "memory(GiB)": 70.5, "step": 77255, "token_acc": 0.9007352941176471, "train_speed(iter/s)": 1.452301 }, { "epoch": 3.3100552675549464, "grad_norm": 2.5225603580474854, "learning_rate": 2.563597749902491e-05, "loss": 0.22338290214538575, "memory(GiB)": 70.5, "step": 77260, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.452304 }, { "epoch": 3.310269482884195, "grad_norm": 3.9728097915649414, "learning_rate": 2.5630100975472026e-05, "loss": 0.456798791885376, "memory(GiB)": 70.5, "step": 77265, "token_acc": 0.8950617283950617, "train_speed(iter/s)": 1.452314 }, { "epoch": 3.310483698213444, "grad_norm": 3.9645442962646484, "learning_rate": 2.5624224893402733e-05, "loss": 0.3399303436279297, "memory(GiB)": 70.5, "step": 77270, "token_acc": 0.9202453987730062, "train_speed(iter/s)": 1.452319 }, { "epoch": 3.3106979135426933, "grad_norm": 1.9632794857025146, "learning_rate": 2.5618349252923448e-05, "loss": 0.2510066032409668, "memory(GiB)": 70.5, "step": 77275, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.452319 }, { "epoch": 3.310912128871942, "grad_norm": 5.084936618804932, "learning_rate": 2.5612474054140657e-05, "loss": 0.506840705871582, "memory(GiB)": 70.5, "step": 77280, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.452319 }, { "epoch": 3.311126344201191, "grad_norm": 3.0524044036865234, "learning_rate": 2.560659929716078e-05, "loss": 0.3791138410568237, "memory(GiB)": 70.5, "step": 77285, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.45232 }, { "epoch": 3.31134055953044, "grad_norm": 3.7285985946655273, "learning_rate": 2.560072498209022e-05, "loss": 0.4130814552307129, "memory(GiB)": 70.5, "step": 77290, "token_acc": 0.9037037037037037, "train_speed(iter/s)": 1.452331 }, { "epoch": 3.311554774859689, "grad_norm": 3.6025588512420654, "learning_rate": 2.5594851109035435e-05, "loss": 0.3365223169326782, "memory(GiB)": 70.5, "step": 77295, "token_acc": 0.93993993993994, "train_speed(iter/s)": 1.452334 }, { "epoch": 3.3117689901889378, "grad_norm": 3.5365846157073975, "learning_rate": 2.558897767810281e-05, "loss": 0.35763001441955566, "memory(GiB)": 70.5, "step": 77300, "token_acc": 0.9233870967741935, "train_speed(iter/s)": 1.452339 }, { "epoch": 3.311983205518187, "grad_norm": 3.2835440635681152, "learning_rate": 2.5583104689398757e-05, "loss": 0.22645974159240723, "memory(GiB)": 70.5, "step": 77305, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.452336 }, { "epoch": 3.312197420847436, "grad_norm": 0.14120757579803467, "learning_rate": 2.5577232143029672e-05, "loss": 0.2994734048843384, "memory(GiB)": 70.5, "step": 77310, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.452339 }, { "epoch": 3.3124116361766847, "grad_norm": 0.8975096344947815, "learning_rate": 2.5571360039101932e-05, "loss": 0.2729493141174316, "memory(GiB)": 70.5, "step": 77315, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.452345 }, { "epoch": 3.312625851505934, "grad_norm": 4.36777925491333, "learning_rate": 2.5565488377721903e-05, "loss": 0.26666827201843263, "memory(GiB)": 70.5, "step": 77320, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.452352 }, { "epoch": 3.3128400668351827, "grad_norm": 2.3762032985687256, "learning_rate": 2.555961715899599e-05, "loss": 0.2443293809890747, "memory(GiB)": 70.5, "step": 77325, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.452358 }, { "epoch": 3.3130542821644315, "grad_norm": 3.1454970836639404, "learning_rate": 2.555374638303054e-05, "loss": 0.28769233226776125, "memory(GiB)": 70.5, "step": 77330, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.45237 }, { "epoch": 3.313268497493681, "grad_norm": 1.3094725608825684, "learning_rate": 2.554787604993191e-05, "loss": 0.2975149154663086, "memory(GiB)": 70.5, "step": 77335, "token_acc": 0.9425675675675675, "train_speed(iter/s)": 1.452372 }, { "epoch": 3.3134827128229296, "grad_norm": 3.125610828399658, "learning_rate": 2.5542006159806444e-05, "loss": 0.41517934799194334, "memory(GiB)": 70.5, "step": 77340, "token_acc": 0.9161073825503355, "train_speed(iter/s)": 1.452374 }, { "epoch": 3.3136969281521784, "grad_norm": 1.0462902784347534, "learning_rate": 2.5536136712760457e-05, "loss": 0.2665847301483154, "memory(GiB)": 70.5, "step": 77345, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.452374 }, { "epoch": 3.3139111434814277, "grad_norm": 2.5068857669830322, "learning_rate": 2.5530267708900303e-05, "loss": 0.2529139518737793, "memory(GiB)": 70.5, "step": 77350, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.452384 }, { "epoch": 3.3141253588106765, "grad_norm": 7.15622615814209, "learning_rate": 2.5524399148332325e-05, "loss": 0.3422494888305664, "memory(GiB)": 70.5, "step": 77355, "token_acc": 0.9049429657794676, "train_speed(iter/s)": 1.45239 }, { "epoch": 3.3143395741399253, "grad_norm": 0.3171575367450714, "learning_rate": 2.551853103116281e-05, "loss": 0.28219056129455566, "memory(GiB)": 70.5, "step": 77360, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.452406 }, { "epoch": 3.3145537894691746, "grad_norm": 5.301991939544678, "learning_rate": 2.5512663357498078e-05, "loss": 0.30156383514404295, "memory(GiB)": 70.5, "step": 77365, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.452407 }, { "epoch": 3.3147680047984234, "grad_norm": 8.848540306091309, "learning_rate": 2.550679612744442e-05, "loss": 0.4016864776611328, "memory(GiB)": 70.5, "step": 77370, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.452436 }, { "epoch": 3.314982220127672, "grad_norm": 0.7569389939308167, "learning_rate": 2.5500929341108104e-05, "loss": 0.20277187824249268, "memory(GiB)": 70.5, "step": 77375, "token_acc": 0.950381679389313, "train_speed(iter/s)": 1.452442 }, { "epoch": 3.3151964354569214, "grad_norm": 2.5333054065704346, "learning_rate": 2.5495062998595455e-05, "loss": 0.40318760871887205, "memory(GiB)": 70.5, "step": 77380, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.452438 }, { "epoch": 3.3154106507861703, "grad_norm": 3.2230966091156006, "learning_rate": 2.548919710001273e-05, "loss": 0.4001627922058105, "memory(GiB)": 70.5, "step": 77385, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.45244 }, { "epoch": 3.315624866115419, "grad_norm": 0.4828869104385376, "learning_rate": 2.5483331645466192e-05, "loss": 0.2285315752029419, "memory(GiB)": 70.5, "step": 77390, "token_acc": 0.951417004048583, "train_speed(iter/s)": 1.452441 }, { "epoch": 3.3158390814446683, "grad_norm": 4.062793254852295, "learning_rate": 2.5477466635062097e-05, "loss": 0.4768829822540283, "memory(GiB)": 70.5, "step": 77395, "token_acc": 0.8830409356725146, "train_speed(iter/s)": 1.452456 }, { "epoch": 3.316053296773917, "grad_norm": 4.564026832580566, "learning_rate": 2.5471602068906697e-05, "loss": 0.29398605823516843, "memory(GiB)": 70.5, "step": 77400, "token_acc": 0.9418960244648318, "train_speed(iter/s)": 1.45247 }, { "epoch": 3.316267512103166, "grad_norm": 1.6099045276641846, "learning_rate": 2.5465737947106218e-05, "loss": 0.3630005121231079, "memory(GiB)": 70.5, "step": 77405, "token_acc": 0.906060606060606, "train_speed(iter/s)": 1.452479 }, { "epoch": 3.316481727432415, "grad_norm": 1.7982620000839233, "learning_rate": 2.545987426976693e-05, "loss": 0.57140212059021, "memory(GiB)": 70.5, "step": 77410, "token_acc": 0.8753623188405797, "train_speed(iter/s)": 1.452491 }, { "epoch": 3.316695942761664, "grad_norm": 4.229897499084473, "learning_rate": 2.545401103699504e-05, "loss": 0.5148476600646973, "memory(GiB)": 70.5, "step": 77415, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.452497 }, { "epoch": 3.316910158090913, "grad_norm": 3.775531530380249, "learning_rate": 2.5448148248896768e-05, "loss": 0.497381591796875, "memory(GiB)": 70.5, "step": 77420, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.4525 }, { "epoch": 3.317124373420162, "grad_norm": 2.1402130126953125, "learning_rate": 2.54422859055783e-05, "loss": 0.3375077486038208, "memory(GiB)": 70.5, "step": 77425, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.452501 }, { "epoch": 3.317338588749411, "grad_norm": 1.054521083831787, "learning_rate": 2.543642400714588e-05, "loss": 0.6214736461639404, "memory(GiB)": 70.5, "step": 77430, "token_acc": 0.8723404255319149, "train_speed(iter/s)": 1.452501 }, { "epoch": 3.3175528040786597, "grad_norm": 2.9014110565185547, "learning_rate": 2.543056255370566e-05, "loss": 0.24893298149108886, "memory(GiB)": 70.5, "step": 77435, "token_acc": 0.9388489208633094, "train_speed(iter/s)": 1.4525 }, { "epoch": 3.317767019407909, "grad_norm": 6.5759100914001465, "learning_rate": 2.542470154536387e-05, "loss": 0.30449671745300294, "memory(GiB)": 70.5, "step": 77440, "token_acc": 0.9535714285714286, "train_speed(iter/s)": 1.452514 }, { "epoch": 3.317981234737158, "grad_norm": 3.5020949840545654, "learning_rate": 2.5418840982226667e-05, "loss": 0.46390738487243655, "memory(GiB)": 70.5, "step": 77445, "token_acc": 0.8919753086419753, "train_speed(iter/s)": 1.452514 }, { "epoch": 3.3181954500664066, "grad_norm": 2.5936119556427, "learning_rate": 2.5412980864400217e-05, "loss": 0.39694643020629883, "memory(GiB)": 70.5, "step": 77450, "token_acc": 0.9343629343629344, "train_speed(iter/s)": 1.452521 }, { "epoch": 3.318409665395656, "grad_norm": 4.520538330078125, "learning_rate": 2.540712119199068e-05, "loss": 0.25325465202331543, "memory(GiB)": 70.5, "step": 77455, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.45253 }, { "epoch": 3.3186238807249047, "grad_norm": 3.9510340690612793, "learning_rate": 2.5401261965104217e-05, "loss": 0.3288752317428589, "memory(GiB)": 70.5, "step": 77460, "token_acc": 0.9296636085626911, "train_speed(iter/s)": 1.452542 }, { "epoch": 3.3188380960541535, "grad_norm": 4.246988773345947, "learning_rate": 2.5395403183846945e-05, "loss": 0.39505300521850584, "memory(GiB)": 70.5, "step": 77465, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.452555 }, { "epoch": 3.3190523113834027, "grad_norm": 2.5110344886779785, "learning_rate": 2.5389544848325054e-05, "loss": 0.18043875694274902, "memory(GiB)": 70.5, "step": 77470, "token_acc": 0.9709677419354839, "train_speed(iter/s)": 1.452558 }, { "epoch": 3.3192665267126515, "grad_norm": 5.081747055053711, "learning_rate": 2.5383686958644636e-05, "loss": 0.5505040645599365, "memory(GiB)": 70.5, "step": 77475, "token_acc": 0.8768768768768769, "train_speed(iter/s)": 1.452564 }, { "epoch": 3.3194807420419004, "grad_norm": 5.138447284698486, "learning_rate": 2.5377829514911822e-05, "loss": 0.5222280025482178, "memory(GiB)": 70.5, "step": 77480, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.452577 }, { "epoch": 3.3196949573711496, "grad_norm": 1.4028955698013306, "learning_rate": 2.537197251723272e-05, "loss": 0.25616753101348877, "memory(GiB)": 70.5, "step": 77485, "token_acc": 0.9467084639498433, "train_speed(iter/s)": 1.45259 }, { "epoch": 3.3199091727003984, "grad_norm": 3.4164931774139404, "learning_rate": 2.5366115965713422e-05, "loss": 0.21673815250396727, "memory(GiB)": 70.5, "step": 77490, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.452593 }, { "epoch": 3.3201233880296472, "grad_norm": 3.3582985401153564, "learning_rate": 2.5360259860460066e-05, "loss": 0.15215089321136474, "memory(GiB)": 70.5, "step": 77495, "token_acc": 0.9717868338557993, "train_speed(iter/s)": 1.452596 }, { "epoch": 3.3203376033588965, "grad_norm": 4.761602401733398, "learning_rate": 2.5354404201578696e-05, "loss": 0.5689416408538819, "memory(GiB)": 70.5, "step": 77500, "token_acc": 0.8841463414634146, "train_speed(iter/s)": 1.452598 }, { "epoch": 3.3203376033588965, "eval_loss": 2.375746965408325, "eval_runtime": 11.7345, "eval_samples_per_second": 8.522, "eval_steps_per_second": 8.522, "eval_token_acc": 0.44686648501362397, "step": 77500 }, { "epoch": 3.3205518186881453, "grad_norm": 5.188143730163574, "learning_rate": 2.5348548989175424e-05, "loss": 0.5294296264648437, "memory(GiB)": 70.5, "step": 77505, "token_acc": 0.5765503875968992, "train_speed(iter/s)": 1.452254 }, { "epoch": 3.320766034017394, "grad_norm": 1.880135178565979, "learning_rate": 2.534269422335632e-05, "loss": 0.5255216121673584, "memory(GiB)": 70.5, "step": 77510, "token_acc": 0.8990825688073395, "train_speed(iter/s)": 1.452267 }, { "epoch": 3.3209802493466434, "grad_norm": 2.5287203788757324, "learning_rate": 2.533683990422744e-05, "loss": 0.30752580165863036, "memory(GiB)": 70.5, "step": 77515, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.452279 }, { "epoch": 3.321194464675892, "grad_norm": 4.497685432434082, "learning_rate": 2.5330986031894823e-05, "loss": 0.48399782180786133, "memory(GiB)": 70.5, "step": 77520, "token_acc": 0.8996282527881041, "train_speed(iter/s)": 1.45229 }, { "epoch": 3.321408680005141, "grad_norm": 1.050594449043274, "learning_rate": 2.5325132606464552e-05, "loss": 0.2748749017715454, "memory(GiB)": 70.5, "step": 77525, "token_acc": 0.9446254071661238, "train_speed(iter/s)": 1.452304 }, { "epoch": 3.3216228953343903, "grad_norm": 4.657716751098633, "learning_rate": 2.5319279628042647e-05, "loss": 0.45987257957458494, "memory(GiB)": 70.5, "step": 77530, "token_acc": 0.9161290322580645, "train_speed(iter/s)": 1.452317 }, { "epoch": 3.321837110663639, "grad_norm": 1.5794154405593872, "learning_rate": 2.5313427096735155e-05, "loss": 0.3844373941421509, "memory(GiB)": 70.5, "step": 77535, "token_acc": 0.92, "train_speed(iter/s)": 1.452321 }, { "epoch": 3.322051325992888, "grad_norm": 2.6889891624450684, "learning_rate": 2.530757501264808e-05, "loss": 0.335671067237854, "memory(GiB)": 70.5, "step": 77540, "token_acc": 0.92, "train_speed(iter/s)": 1.452322 }, { "epoch": 3.322265541322137, "grad_norm": 1.1630446910858154, "learning_rate": 2.5301723375887447e-05, "loss": 0.17518677711486816, "memory(GiB)": 70.5, "step": 77545, "token_acc": 0.9702970297029703, "train_speed(iter/s)": 1.452322 }, { "epoch": 3.322479756651386, "grad_norm": 2.362600564956665, "learning_rate": 2.5295872186559243e-05, "loss": 0.3000542163848877, "memory(GiB)": 70.5, "step": 77550, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.452325 }, { "epoch": 3.3226939719806348, "grad_norm": 3.0244109630584717, "learning_rate": 2.5290021444769517e-05, "loss": 0.24012014865875245, "memory(GiB)": 70.5, "step": 77555, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.45235 }, { "epoch": 3.322908187309884, "grad_norm": 1.1568515300750732, "learning_rate": 2.5284171150624225e-05, "loss": 0.20124950408935546, "memory(GiB)": 70.5, "step": 77560, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.452349 }, { "epoch": 3.323122402639133, "grad_norm": 0.4936404228210449, "learning_rate": 2.5278321304229358e-05, "loss": 0.3653169870376587, "memory(GiB)": 70.5, "step": 77565, "token_acc": 0.9137380191693291, "train_speed(iter/s)": 1.452355 }, { "epoch": 3.3233366179683816, "grad_norm": 6.05821418762207, "learning_rate": 2.5272471905690876e-05, "loss": 0.27275335788726807, "memory(GiB)": 70.5, "step": 77570, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.452358 }, { "epoch": 3.323550833297631, "grad_norm": 4.135031700134277, "learning_rate": 2.526662295511478e-05, "loss": 0.6209010124206543, "memory(GiB)": 70.5, "step": 77575, "token_acc": 0.8896551724137931, "train_speed(iter/s)": 1.452357 }, { "epoch": 3.3237650486268797, "grad_norm": 2.9161324501037598, "learning_rate": 2.5260774452606993e-05, "loss": 0.45498132705688477, "memory(GiB)": 70.5, "step": 77580, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.452349 }, { "epoch": 3.3239792639561285, "grad_norm": 2.743109703063965, "learning_rate": 2.5254926398273498e-05, "loss": 0.592125940322876, "memory(GiB)": 70.5, "step": 77585, "token_acc": 0.8614457831325302, "train_speed(iter/s)": 1.45235 }, { "epoch": 3.324193479285378, "grad_norm": 4.628730297088623, "learning_rate": 2.5249078792220227e-05, "loss": 0.22437186241149903, "memory(GiB)": 70.5, "step": 77590, "token_acc": 0.9537366548042705, "train_speed(iter/s)": 1.452351 }, { "epoch": 3.3244076946146266, "grad_norm": 0.39912480115890503, "learning_rate": 2.524323163455311e-05, "loss": 0.4210383892059326, "memory(GiB)": 70.5, "step": 77595, "token_acc": 0.92, "train_speed(iter/s)": 1.452357 }, { "epoch": 3.3246219099438754, "grad_norm": 3.8397789001464844, "learning_rate": 2.5237384925378084e-05, "loss": 0.10915524959564209, "memory(GiB)": 70.5, "step": 77600, "token_acc": 0.973293768545994, "train_speed(iter/s)": 1.45236 }, { "epoch": 3.3248361252731247, "grad_norm": 8.80649471282959, "learning_rate": 2.5231538664801047e-05, "loss": 0.25123560428619385, "memory(GiB)": 70.5, "step": 77605, "token_acc": 0.9425287356321839, "train_speed(iter/s)": 1.45236 }, { "epoch": 3.3250503406023735, "grad_norm": 4.049310684204102, "learning_rate": 2.5225692852927913e-05, "loss": 0.27088565826416017, "memory(GiB)": 70.5, "step": 77610, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.452362 }, { "epoch": 3.3252645559316223, "grad_norm": 1.4954571723937988, "learning_rate": 2.5219847489864606e-05, "loss": 0.4110464572906494, "memory(GiB)": 70.5, "step": 77615, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.452361 }, { "epoch": 3.3254787712608715, "grad_norm": 3.7086544036865234, "learning_rate": 2.521400257571701e-05, "loss": 0.25428051948547364, "memory(GiB)": 70.5, "step": 77620, "token_acc": 0.937888198757764, "train_speed(iter/s)": 1.452363 }, { "epoch": 3.3256929865901204, "grad_norm": 3.5001273155212402, "learning_rate": 2.5208158110591006e-05, "loss": 0.3255154609680176, "memory(GiB)": 70.5, "step": 77625, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.452366 }, { "epoch": 3.325907201919369, "grad_norm": 6.6854634284973145, "learning_rate": 2.5202314094592478e-05, "loss": 0.4919573783874512, "memory(GiB)": 70.5, "step": 77630, "token_acc": 0.8930041152263375, "train_speed(iter/s)": 1.452373 }, { "epoch": 3.3261214172486184, "grad_norm": 3.5587406158447266, "learning_rate": 2.519647052782727e-05, "loss": 0.19458725452423095, "memory(GiB)": 70.5, "step": 77635, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.452374 }, { "epoch": 3.3263356325778672, "grad_norm": 4.830319404602051, "learning_rate": 2.519062741040129e-05, "loss": 0.5512583255767822, "memory(GiB)": 70.5, "step": 77640, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.452377 }, { "epoch": 3.326549847907116, "grad_norm": 0.6359609961509705, "learning_rate": 2.5184784742420342e-05, "loss": 0.29691014289855955, "memory(GiB)": 70.5, "step": 77645, "token_acc": 0.9516129032258065, "train_speed(iter/s)": 1.452401 }, { "epoch": 3.3267640632363653, "grad_norm": 6.276535987854004, "learning_rate": 2.5178942523990324e-05, "loss": 0.409560489654541, "memory(GiB)": 70.5, "step": 77650, "token_acc": 0.8896103896103896, "train_speed(iter/s)": 1.452407 }, { "epoch": 3.326978278565614, "grad_norm": 1.197473406791687, "learning_rate": 2.5173100755217037e-05, "loss": 0.1283421754837036, "memory(GiB)": 70.5, "step": 77655, "token_acc": 0.9713375796178344, "train_speed(iter/s)": 1.452411 }, { "epoch": 3.327192493894863, "grad_norm": 1.8907541036605835, "learning_rate": 2.5167259436206325e-05, "loss": 0.23925845623016356, "memory(GiB)": 70.5, "step": 77660, "token_acc": 0.94140625, "train_speed(iter/s)": 1.45242 }, { "epoch": 3.327406709224112, "grad_norm": 4.175384044647217, "learning_rate": 2.5161418567063983e-05, "loss": 0.33089661598205566, "memory(GiB)": 70.5, "step": 77665, "token_acc": 0.9239543726235742, "train_speed(iter/s)": 1.452414 }, { "epoch": 3.327620924553361, "grad_norm": 4.637680530548096, "learning_rate": 2.5155578147895862e-05, "loss": 0.2093562126159668, "memory(GiB)": 70.5, "step": 77670, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.452429 }, { "epoch": 3.32783513988261, "grad_norm": 4.360167980194092, "learning_rate": 2.5149738178807745e-05, "loss": 0.5067473888397217, "memory(GiB)": 70.5, "step": 77675, "token_acc": 0.8958990536277602, "train_speed(iter/s)": 1.45243 }, { "epoch": 3.328049355211859, "grad_norm": 4.023850440979004, "learning_rate": 2.5143898659905442e-05, "loss": 0.6189085006713867, "memory(GiB)": 70.5, "step": 77680, "token_acc": 0.875968992248062, "train_speed(iter/s)": 1.452432 }, { "epoch": 3.328263570541108, "grad_norm": 6.420633316040039, "learning_rate": 2.5138059591294727e-05, "loss": 0.7702089786529541, "memory(GiB)": 70.5, "step": 77685, "token_acc": 0.8455882352941176, "train_speed(iter/s)": 1.452442 }, { "epoch": 3.3284777858703567, "grad_norm": 2.072373151779175, "learning_rate": 2.513222097308138e-05, "loss": 0.20767827033996583, "memory(GiB)": 70.5, "step": 77690, "token_acc": 0.9546925566343042, "train_speed(iter/s)": 1.452447 }, { "epoch": 3.328692001199606, "grad_norm": 2.5408294200897217, "learning_rate": 2.512638280537117e-05, "loss": 0.20053648948669434, "memory(GiB)": 70.5, "step": 77695, "token_acc": 0.9577922077922078, "train_speed(iter/s)": 1.452444 }, { "epoch": 3.3289062165288548, "grad_norm": 3.137625217437744, "learning_rate": 2.5120545088269877e-05, "loss": 0.2705045700073242, "memory(GiB)": 70.5, "step": 77700, "token_acc": 0.9431137724550899, "train_speed(iter/s)": 1.452444 }, { "epoch": 3.3291204318581036, "grad_norm": 1.4595495462417603, "learning_rate": 2.5114707821883253e-05, "loss": 0.162355375289917, "memory(GiB)": 70.5, "step": 77705, "token_acc": 0.9556451612903226, "train_speed(iter/s)": 1.452453 }, { "epoch": 3.329334647187353, "grad_norm": 2.197077751159668, "learning_rate": 2.5108871006317046e-05, "loss": 0.3908404350280762, "memory(GiB)": 70.5, "step": 77710, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.452458 }, { "epoch": 3.3295488625166016, "grad_norm": 3.688750982284546, "learning_rate": 2.510303464167698e-05, "loss": 0.43451700210571287, "memory(GiB)": 70.5, "step": 77715, "token_acc": 0.912751677852349, "train_speed(iter/s)": 1.452465 }, { "epoch": 3.3297630778458505, "grad_norm": 0.09747504442930222, "learning_rate": 2.509719872806878e-05, "loss": 0.2544887065887451, "memory(GiB)": 70.5, "step": 77720, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.452468 }, { "epoch": 3.3299772931750997, "grad_norm": 4.585292339324951, "learning_rate": 2.5091363265598196e-05, "loss": 0.33856825828552245, "memory(GiB)": 70.5, "step": 77725, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.452468 }, { "epoch": 3.3301915085043485, "grad_norm": 3.459836006164551, "learning_rate": 2.5085528254370942e-05, "loss": 0.30467543601989744, "memory(GiB)": 70.5, "step": 77730, "token_acc": 0.916, "train_speed(iter/s)": 1.452466 }, { "epoch": 3.3304057238335973, "grad_norm": 2.6346468925476074, "learning_rate": 2.5079693694492722e-05, "loss": 0.2334094762802124, "memory(GiB)": 70.5, "step": 77735, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.452468 }, { "epoch": 3.3306199391628466, "grad_norm": 0.2543889582157135, "learning_rate": 2.507385958606922e-05, "loss": 0.11762173175811767, "memory(GiB)": 70.5, "step": 77740, "token_acc": 0.967391304347826, "train_speed(iter/s)": 1.452483 }, { "epoch": 3.3308341544920954, "grad_norm": 2.889033794403076, "learning_rate": 2.506802592920614e-05, "loss": 0.47585673332214357, "memory(GiB)": 70.5, "step": 77745, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.452478 }, { "epoch": 3.3310483698213442, "grad_norm": 4.663197994232178, "learning_rate": 2.506219272400916e-05, "loss": 0.26875545978546145, "memory(GiB)": 70.5, "step": 77750, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.452495 }, { "epoch": 3.3312625851505935, "grad_norm": 5.1400556564331055, "learning_rate": 2.5056359970583925e-05, "loss": 0.5285397052764893, "memory(GiB)": 70.5, "step": 77755, "token_acc": 0.8726591760299626, "train_speed(iter/s)": 1.452496 }, { "epoch": 3.3314768004798423, "grad_norm": 5.155440330505371, "learning_rate": 2.505052766903615e-05, "loss": 0.4853353023529053, "memory(GiB)": 70.5, "step": 77760, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.452495 }, { "epoch": 3.331691015809091, "grad_norm": 2.797515869140625, "learning_rate": 2.5044695819471475e-05, "loss": 0.4628288745880127, "memory(GiB)": 70.5, "step": 77765, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.452504 }, { "epoch": 3.3319052311383404, "grad_norm": 0.7431730031967163, "learning_rate": 2.5038864421995545e-05, "loss": 0.46004343032836914, "memory(GiB)": 70.5, "step": 77770, "token_acc": 0.9078947368421053, "train_speed(iter/s)": 1.452506 }, { "epoch": 3.332119446467589, "grad_norm": 0.5414047241210938, "learning_rate": 2.5033033476713998e-05, "loss": 0.16576920747756957, "memory(GiB)": 70.5, "step": 77775, "token_acc": 0.9471544715447154, "train_speed(iter/s)": 1.452507 }, { "epoch": 3.332333661796838, "grad_norm": 5.3219895362854, "learning_rate": 2.5027202983732456e-05, "loss": 0.3809701204299927, "memory(GiB)": 70.5, "step": 77780, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.452512 }, { "epoch": 3.3325478771260872, "grad_norm": 3.239652156829834, "learning_rate": 2.5021372943156575e-05, "loss": 0.3057529449462891, "memory(GiB)": 70.5, "step": 77785, "token_acc": 0.9292307692307692, "train_speed(iter/s)": 1.452516 }, { "epoch": 3.332762092455336, "grad_norm": 0.6794885396957397, "learning_rate": 2.501554335509195e-05, "loss": 0.34268057346343994, "memory(GiB)": 70.5, "step": 77790, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.452515 }, { "epoch": 3.332976307784585, "grad_norm": 3.9347782135009766, "learning_rate": 2.5009714219644175e-05, "loss": 0.24181170463562013, "memory(GiB)": 70.5, "step": 77795, "token_acc": 0.9469387755102041, "train_speed(iter/s)": 1.452518 }, { "epoch": 3.333190523113834, "grad_norm": 0.6669278144836426, "learning_rate": 2.5003885536918897e-05, "loss": 0.22285475730895996, "memory(GiB)": 70.5, "step": 77800, "token_acc": 0.9469387755102041, "train_speed(iter/s)": 1.452519 }, { "epoch": 3.333404738443083, "grad_norm": 3.7969393730163574, "learning_rate": 2.4998057307021677e-05, "loss": 0.27949769496917726, "memory(GiB)": 70.5, "step": 77805, "token_acc": 0.94921875, "train_speed(iter/s)": 1.452529 }, { "epoch": 3.3336189537723317, "grad_norm": 3.711181163787842, "learning_rate": 2.4992229530058087e-05, "loss": 0.377128005027771, "memory(GiB)": 70.5, "step": 77810, "token_acc": 0.9159420289855073, "train_speed(iter/s)": 1.45253 }, { "epoch": 3.333833169101581, "grad_norm": 6.748050689697266, "learning_rate": 2.498640220613373e-05, "loss": 0.49857940673828127, "memory(GiB)": 70.5, "step": 77815, "token_acc": 0.9051987767584098, "train_speed(iter/s)": 1.452534 }, { "epoch": 3.33404738443083, "grad_norm": 0.5122215151786804, "learning_rate": 2.498057533535417e-05, "loss": 0.19049018621444702, "memory(GiB)": 70.5, "step": 77820, "token_acc": 0.9736842105263158, "train_speed(iter/s)": 1.452549 }, { "epoch": 3.3342615997600786, "grad_norm": 2.641037702560425, "learning_rate": 2.4974748917824954e-05, "loss": 0.34295408725738524, "memory(GiB)": 70.5, "step": 77825, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.452553 }, { "epoch": 3.334475815089328, "grad_norm": 4.764046669006348, "learning_rate": 2.4968922953651635e-05, "loss": 0.4471895694732666, "memory(GiB)": 70.5, "step": 77830, "token_acc": 0.8920634920634921, "train_speed(iter/s)": 1.452555 }, { "epoch": 3.3346900304185767, "grad_norm": 2.605194091796875, "learning_rate": 2.496309744293976e-05, "loss": 0.47827773094177245, "memory(GiB)": 70.5, "step": 77835, "token_acc": 0.9034267912772586, "train_speed(iter/s)": 1.452563 }, { "epoch": 3.3349042457478255, "grad_norm": 2.7751996517181396, "learning_rate": 2.495727238579484e-05, "loss": 0.5233914375305175, "memory(GiB)": 70.5, "step": 77840, "token_acc": 0.9107692307692308, "train_speed(iter/s)": 1.452565 }, { "epoch": 3.3351184610770748, "grad_norm": 0.7207188606262207, "learning_rate": 2.495144778232244e-05, "loss": 0.6705893993377685, "memory(GiB)": 70.5, "step": 77845, "token_acc": 0.8778877887788779, "train_speed(iter/s)": 1.452569 }, { "epoch": 3.3353326764063236, "grad_norm": 5.314512729644775, "learning_rate": 2.4945623632628068e-05, "loss": 0.28265676498413084, "memory(GiB)": 70.5, "step": 77850, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.452571 }, { "epoch": 3.3355468917355724, "grad_norm": 2.97049880027771, "learning_rate": 2.4939799936817222e-05, "loss": 0.31410770416259765, "memory(GiB)": 70.5, "step": 77855, "token_acc": 0.9158249158249159, "train_speed(iter/s)": 1.452571 }, { "epoch": 3.3357611070648217, "grad_norm": 6.075656890869141, "learning_rate": 2.493397669499541e-05, "loss": 0.47241578102111814, "memory(GiB)": 70.5, "step": 77860, "token_acc": 0.8951048951048951, "train_speed(iter/s)": 1.452568 }, { "epoch": 3.3359753223940705, "grad_norm": 1.8383910655975342, "learning_rate": 2.4928153907268104e-05, "loss": 0.3222075939178467, "memory(GiB)": 70.5, "step": 77865, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.452569 }, { "epoch": 3.3361895377233193, "grad_norm": 3.1219491958618164, "learning_rate": 2.4922331573740808e-05, "loss": 0.27536778450012206, "memory(GiB)": 70.5, "step": 77870, "token_acc": 0.9388489208633094, "train_speed(iter/s)": 1.452571 }, { "epoch": 3.3364037530525685, "grad_norm": 5.1829023361206055, "learning_rate": 2.491650969451902e-05, "loss": 0.25839920043945314, "memory(GiB)": 70.5, "step": 77875, "token_acc": 0.9537366548042705, "train_speed(iter/s)": 1.452577 }, { "epoch": 3.3366179683818173, "grad_norm": 2.71146559715271, "learning_rate": 2.4910688269708195e-05, "loss": 0.3806927680969238, "memory(GiB)": 70.5, "step": 77880, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.452576 }, { "epoch": 3.336832183711066, "grad_norm": 0.6322245001792908, "learning_rate": 2.4904867299413783e-05, "loss": 0.1383345365524292, "memory(GiB)": 70.5, "step": 77885, "token_acc": 0.9706959706959707, "train_speed(iter/s)": 1.452581 }, { "epoch": 3.3370463990403154, "grad_norm": 0.6190029978752136, "learning_rate": 2.4899046783741235e-05, "loss": 0.1819856882095337, "memory(GiB)": 70.5, "step": 77890, "token_acc": 0.9451612903225807, "train_speed(iter/s)": 1.45258 }, { "epoch": 3.3372606143695642, "grad_norm": 0.19511382281780243, "learning_rate": 2.4893226722796008e-05, "loss": 0.32038905620574953, "memory(GiB)": 70.5, "step": 77895, "token_acc": 0.9292307692307692, "train_speed(iter/s)": 1.452592 }, { "epoch": 3.337474829698813, "grad_norm": 4.911189556121826, "learning_rate": 2.4887407116683516e-05, "loss": 0.535123348236084, "memory(GiB)": 70.5, "step": 77900, "token_acc": 0.8970588235294118, "train_speed(iter/s)": 1.452604 }, { "epoch": 3.3376890450280623, "grad_norm": 2.8479387760162354, "learning_rate": 2.488158796550921e-05, "loss": 0.2864459276199341, "memory(GiB)": 70.5, "step": 77905, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.452605 }, { "epoch": 3.337903260357311, "grad_norm": 0.23886089026927948, "learning_rate": 2.487576926937851e-05, "loss": 0.24432833194732667, "memory(GiB)": 70.5, "step": 77910, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.452609 }, { "epoch": 3.33811747568656, "grad_norm": 6.825482368469238, "learning_rate": 2.4869951028396813e-05, "loss": 0.33167119026184083, "memory(GiB)": 70.5, "step": 77915, "token_acc": 0.92578125, "train_speed(iter/s)": 1.45261 }, { "epoch": 3.338331691015809, "grad_norm": 5.0975422859191895, "learning_rate": 2.486413324266953e-05, "loss": 0.38232429027557374, "memory(GiB)": 70.5, "step": 77920, "token_acc": 0.9122257053291536, "train_speed(iter/s)": 1.452615 }, { "epoch": 3.338545906345058, "grad_norm": 3.162856101989746, "learning_rate": 2.4858315912302033e-05, "loss": 0.20814692974090576, "memory(GiB)": 70.5, "step": 77925, "token_acc": 0.9486404833836858, "train_speed(iter/s)": 1.45262 }, { "epoch": 3.338760121674307, "grad_norm": 2.0097570419311523, "learning_rate": 2.4852499037399747e-05, "loss": 0.42900891304016114, "memory(GiB)": 70.5, "step": 77930, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.452624 }, { "epoch": 3.338974337003556, "grad_norm": 2.299622058868408, "learning_rate": 2.484668261806803e-05, "loss": 0.34510538578033445, "memory(GiB)": 70.5, "step": 77935, "token_acc": 0.9295774647887324, "train_speed(iter/s)": 1.452632 }, { "epoch": 3.339188552332805, "grad_norm": 0.32974496483802795, "learning_rate": 2.4840866654412232e-05, "loss": 0.11496576070785522, "memory(GiB)": 70.5, "step": 77940, "token_acc": 0.9683098591549296, "train_speed(iter/s)": 1.452637 }, { "epoch": 3.3394027676620537, "grad_norm": 1.9547072649002075, "learning_rate": 2.483505114653776e-05, "loss": 0.29766552448272704, "memory(GiB)": 70.5, "step": 77945, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.45264 }, { "epoch": 3.339616982991303, "grad_norm": 5.55319881439209, "learning_rate": 2.482923609454994e-05, "loss": 0.2964426755905151, "memory(GiB)": 70.5, "step": 77950, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.452647 }, { "epoch": 3.3398311983205518, "grad_norm": 2.103694200515747, "learning_rate": 2.48234214985541e-05, "loss": 0.28181073665618894, "memory(GiB)": 70.5, "step": 77955, "token_acc": 0.9398496240601504, "train_speed(iter/s)": 1.452659 }, { "epoch": 3.3400454136498006, "grad_norm": 0.27719423174858093, "learning_rate": 2.4817607358655614e-05, "loss": 0.4664303302764893, "memory(GiB)": 70.5, "step": 77960, "token_acc": 0.8991935483870968, "train_speed(iter/s)": 1.452661 }, { "epoch": 3.34025962897905, "grad_norm": 4.159386157989502, "learning_rate": 2.48117936749598e-05, "loss": 0.2670133590698242, "memory(GiB)": 70.5, "step": 77965, "token_acc": 0.95578231292517, "train_speed(iter/s)": 1.452665 }, { "epoch": 3.3404738443082986, "grad_norm": 1.6092922687530518, "learning_rate": 2.480598044757197e-05, "loss": 0.38036112785339354, "memory(GiB)": 70.5, "step": 77970, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.45267 }, { "epoch": 3.3406880596375474, "grad_norm": 1.2379223108291626, "learning_rate": 2.4800167676597436e-05, "loss": 0.1949336051940918, "memory(GiB)": 70.5, "step": 77975, "token_acc": 0.9456066945606695, "train_speed(iter/s)": 1.452672 }, { "epoch": 3.3409022749667967, "grad_norm": 1.3192789554595947, "learning_rate": 2.4794355362141508e-05, "loss": 0.17384363412857057, "memory(GiB)": 70.5, "step": 77980, "token_acc": 0.9603399433427762, "train_speed(iter/s)": 1.452676 }, { "epoch": 3.3411164902960455, "grad_norm": 1.7729939222335815, "learning_rate": 2.4788543504309454e-05, "loss": 0.5069224834442139, "memory(GiB)": 70.5, "step": 77985, "token_acc": 0.9174917491749175, "train_speed(iter/s)": 1.452681 }, { "epoch": 3.3413307056252943, "grad_norm": 1.9604384899139404, "learning_rate": 2.4782732103206607e-05, "loss": 0.2587468385696411, "memory(GiB)": 70.5, "step": 77990, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.452684 }, { "epoch": 3.3415449209545436, "grad_norm": 0.6111294031143188, "learning_rate": 2.4776921158938222e-05, "loss": 0.18797047138214112, "memory(GiB)": 70.5, "step": 77995, "token_acc": 0.9539007092198581, "train_speed(iter/s)": 1.452694 }, { "epoch": 3.3417591362837924, "grad_norm": 0.9023430943489075, "learning_rate": 2.4771110671609573e-05, "loss": 0.26372947692871096, "memory(GiB)": 70.5, "step": 78000, "token_acc": 0.9638989169675091, "train_speed(iter/s)": 1.452709 }, { "epoch": 3.3417591362837924, "eval_loss": 2.433716058731079, "eval_runtime": 11.6563, "eval_samples_per_second": 8.579, "eval_steps_per_second": 8.579, "eval_token_acc": 0.4562091503267974, "step": 78000 }, { "epoch": 3.341973351613041, "grad_norm": 6.116693496704102, "learning_rate": 2.4765300641325915e-05, "loss": 0.8011038780212403, "memory(GiB)": 70.5, "step": 78005, "token_acc": 0.5609981515711645, "train_speed(iter/s)": 1.452363 }, { "epoch": 3.3421875669422905, "grad_norm": 1.4440550804138184, "learning_rate": 2.4759491068192496e-05, "loss": 0.2519721508026123, "memory(GiB)": 70.5, "step": 78010, "token_acc": 0.944078947368421, "train_speed(iter/s)": 1.452377 }, { "epoch": 3.3424017822715393, "grad_norm": 3.9233944416046143, "learning_rate": 2.4753681952314573e-05, "loss": 0.3481459617614746, "memory(GiB)": 70.5, "step": 78015, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.45238 }, { "epoch": 3.342615997600788, "grad_norm": 5.283749580383301, "learning_rate": 2.4747873293797396e-05, "loss": 0.5328766822814941, "memory(GiB)": 70.5, "step": 78020, "token_acc": 0.8842105263157894, "train_speed(iter/s)": 1.452386 }, { "epoch": 3.3428302129300373, "grad_norm": 2.7426490783691406, "learning_rate": 2.474206509274619e-05, "loss": 0.2705348253250122, "memory(GiB)": 70.5, "step": 78025, "token_acc": 0.9442815249266863, "train_speed(iter/s)": 1.452384 }, { "epoch": 3.343044428259286, "grad_norm": 2.5596625804901123, "learning_rate": 2.4736257349266167e-05, "loss": 0.11749334335327148, "memory(GiB)": 70.5, "step": 78030, "token_acc": 0.9822222222222222, "train_speed(iter/s)": 1.452388 }, { "epoch": 3.343258643588535, "grad_norm": 2.8401167392730713, "learning_rate": 2.473045006346254e-05, "loss": 0.3036047458648682, "memory(GiB)": 70.5, "step": 78035, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.452391 }, { "epoch": 3.3434728589177842, "grad_norm": 1.3005129098892212, "learning_rate": 2.472464323544052e-05, "loss": 0.41666088104248045, "memory(GiB)": 70.5, "step": 78040, "token_acc": 0.9359756097560976, "train_speed(iter/s)": 1.452393 }, { "epoch": 3.343687074247033, "grad_norm": 1.7357302904129028, "learning_rate": 2.4718836865305274e-05, "loss": 0.39350042343139646, "memory(GiB)": 70.5, "step": 78045, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.452396 }, { "epoch": 3.343901289576282, "grad_norm": 3.6577913761138916, "learning_rate": 2.4713030953162036e-05, "loss": 0.40500526428222655, "memory(GiB)": 70.5, "step": 78050, "token_acc": 0.9078498293515358, "train_speed(iter/s)": 1.4524 }, { "epoch": 3.344115504905531, "grad_norm": 4.752927780151367, "learning_rate": 2.470722549911596e-05, "loss": 0.5801928520202637, "memory(GiB)": 70.5, "step": 78055, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.452404 }, { "epoch": 3.34432972023478, "grad_norm": 2.0166056156158447, "learning_rate": 2.470142050327222e-05, "loss": 0.35126571655273436, "memory(GiB)": 70.5, "step": 78060, "token_acc": 0.907185628742515, "train_speed(iter/s)": 1.452409 }, { "epoch": 3.3445439355640287, "grad_norm": 2.12361216545105, "learning_rate": 2.4695615965735984e-05, "loss": 0.20585370063781738, "memory(GiB)": 70.5, "step": 78065, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.452423 }, { "epoch": 3.344758150893278, "grad_norm": 3.3613810539245605, "learning_rate": 2.468981188661238e-05, "loss": 0.3201587438583374, "memory(GiB)": 70.5, "step": 78070, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.452428 }, { "epoch": 3.344972366222527, "grad_norm": 3.48781681060791, "learning_rate": 2.46840082660066e-05, "loss": 0.5351302623748779, "memory(GiB)": 70.5, "step": 78075, "token_acc": 0.9119496855345912, "train_speed(iter/s)": 1.452443 }, { "epoch": 3.3451865815517756, "grad_norm": 0.05939272418618202, "learning_rate": 2.467820510402375e-05, "loss": 0.3325479030609131, "memory(GiB)": 70.5, "step": 78080, "token_acc": 0.9348958333333334, "train_speed(iter/s)": 1.452458 }, { "epoch": 3.345400796881025, "grad_norm": 1.2039744853973389, "learning_rate": 2.4672402400768973e-05, "loss": 0.24162616729736328, "memory(GiB)": 70.5, "step": 78085, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.452457 }, { "epoch": 3.3456150122102737, "grad_norm": 0.3632568120956421, "learning_rate": 2.466660015634737e-05, "loss": 0.471040678024292, "memory(GiB)": 70.5, "step": 78090, "token_acc": 0.9203539823008849, "train_speed(iter/s)": 1.452472 }, { "epoch": 3.3458292275395225, "grad_norm": 5.894540786743164, "learning_rate": 2.4660798370864086e-05, "loss": 0.4551403045654297, "memory(GiB)": 70.5, "step": 78095, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.452482 }, { "epoch": 3.3460434428687718, "grad_norm": 1.6339294910430908, "learning_rate": 2.465499704442419e-05, "loss": 0.23656392097473145, "memory(GiB)": 70.5, "step": 78100, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.452486 }, { "epoch": 3.3462576581980206, "grad_norm": 6.407426357269287, "learning_rate": 2.4649196177132818e-05, "loss": 0.5282564163208008, "memory(GiB)": 70.5, "step": 78105, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.452486 }, { "epoch": 3.3464718735272694, "grad_norm": 6.032691955566406, "learning_rate": 2.4643395769095035e-05, "loss": 0.23117365837097167, "memory(GiB)": 70.5, "step": 78110, "token_acc": 0.9462025316455697, "train_speed(iter/s)": 1.452486 }, { "epoch": 3.3466860888565186, "grad_norm": 4.333324909210205, "learning_rate": 2.4637595820415925e-05, "loss": 0.3095167875289917, "memory(GiB)": 70.5, "step": 78115, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.452486 }, { "epoch": 3.3469003041857674, "grad_norm": 2.1950817108154297, "learning_rate": 2.4631796331200564e-05, "loss": 0.2693670988082886, "memory(GiB)": 70.5, "step": 78120, "token_acc": 0.9173553719008265, "train_speed(iter/s)": 1.452486 }, { "epoch": 3.3471145195150163, "grad_norm": 2.7948265075683594, "learning_rate": 2.4625997301554005e-05, "loss": 0.3643715143203735, "memory(GiB)": 70.5, "step": 78125, "token_acc": 0.9003115264797508, "train_speed(iter/s)": 1.452486 }, { "epoch": 3.3473287348442655, "grad_norm": 4.436467170715332, "learning_rate": 2.462019873158129e-05, "loss": 0.33903253078460693, "memory(GiB)": 70.5, "step": 78130, "token_acc": 0.9192200557103064, "train_speed(iter/s)": 1.452485 }, { "epoch": 3.3475429501735143, "grad_norm": 4.000362396240234, "learning_rate": 2.46144006213875e-05, "loss": 0.23951292037963867, "memory(GiB)": 70.5, "step": 78135, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.452496 }, { "epoch": 3.347757165502763, "grad_norm": 4.348373889923096, "learning_rate": 2.460860297107766e-05, "loss": 0.5379643440246582, "memory(GiB)": 70.5, "step": 78140, "token_acc": 0.9009584664536742, "train_speed(iter/s)": 1.452492 }, { "epoch": 3.3479713808320124, "grad_norm": 0.8678717613220215, "learning_rate": 2.4602805780756795e-05, "loss": 0.27115716934204104, "memory(GiB)": 70.5, "step": 78145, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.452492 }, { "epoch": 3.348185596161261, "grad_norm": 1.490216612815857, "learning_rate": 2.459700905052993e-05, "loss": 0.22916417121887206, "memory(GiB)": 70.5, "step": 78150, "token_acc": 0.9307228915662651, "train_speed(iter/s)": 1.452496 }, { "epoch": 3.34839981149051, "grad_norm": 4.67518424987793, "learning_rate": 2.459121278050205e-05, "loss": 0.2994534015655518, "memory(GiB)": 70.5, "step": 78155, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.452507 }, { "epoch": 3.3486140268197593, "grad_norm": 3.191516160964966, "learning_rate": 2.4585416970778207e-05, "loss": 0.29970440864562986, "memory(GiB)": 70.5, "step": 78160, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.452507 }, { "epoch": 3.348828242149008, "grad_norm": 5.77773380279541, "learning_rate": 2.4579621621463362e-05, "loss": 0.40645761489868165, "memory(GiB)": 70.5, "step": 78165, "token_acc": 0.9067524115755627, "train_speed(iter/s)": 1.452509 }, { "epoch": 3.349042457478257, "grad_norm": 1.8525331020355225, "learning_rate": 2.4573826732662537e-05, "loss": 0.21349709033966063, "memory(GiB)": 70.5, "step": 78170, "token_acc": 0.9438596491228071, "train_speed(iter/s)": 1.452519 }, { "epoch": 3.349256672807506, "grad_norm": 1.9154529571533203, "learning_rate": 2.4568032304480687e-05, "loss": 0.2971083164215088, "memory(GiB)": 70.5, "step": 78175, "token_acc": 0.9553903345724907, "train_speed(iter/s)": 1.452524 }, { "epoch": 3.349470888136755, "grad_norm": 4.027176380157471, "learning_rate": 2.4562238337022793e-05, "loss": 0.44158315658569336, "memory(GiB)": 70.5, "step": 78180, "token_acc": 0.891213389121339, "train_speed(iter/s)": 1.452524 }, { "epoch": 3.349685103466004, "grad_norm": 1.168401837348938, "learning_rate": 2.455644483039381e-05, "loss": 0.3112322807312012, "memory(GiB)": 70.5, "step": 78185, "token_acc": 0.9234972677595629, "train_speed(iter/s)": 1.452539 }, { "epoch": 3.349899318795253, "grad_norm": 5.187276363372803, "learning_rate": 2.455065178469868e-05, "loss": 0.5445015907287598, "memory(GiB)": 70.5, "step": 78190, "token_acc": 0.8901515151515151, "train_speed(iter/s)": 1.452556 }, { "epoch": 3.350113534124502, "grad_norm": 2.122426986694336, "learning_rate": 2.4544859200042386e-05, "loss": 0.2708376407623291, "memory(GiB)": 70.5, "step": 78195, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.452565 }, { "epoch": 3.3503277494537507, "grad_norm": 2.6421351432800293, "learning_rate": 2.4539067076529847e-05, "loss": 0.47429676055908204, "memory(GiB)": 70.5, "step": 78200, "token_acc": 0.8955223880597015, "train_speed(iter/s)": 1.452577 }, { "epoch": 3.350541964783, "grad_norm": 1.709932804107666, "learning_rate": 2.4533275414265992e-05, "loss": 0.3435033082962036, "memory(GiB)": 70.5, "step": 78205, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.452576 }, { "epoch": 3.3507561801122487, "grad_norm": 4.136984825134277, "learning_rate": 2.452748421335574e-05, "loss": 0.41091341972351075, "memory(GiB)": 70.5, "step": 78210, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.452578 }, { "epoch": 3.3509703954414976, "grad_norm": 4.484251976013184, "learning_rate": 2.452169347390399e-05, "loss": 0.2551116943359375, "memory(GiB)": 70.5, "step": 78215, "token_acc": 0.9554794520547946, "train_speed(iter/s)": 1.452572 }, { "epoch": 3.351184610770747, "grad_norm": 2.0876801013946533, "learning_rate": 2.4515903196015684e-05, "loss": 0.3137326240539551, "memory(GiB)": 70.5, "step": 78220, "token_acc": 0.9309462915601023, "train_speed(iter/s)": 1.452575 }, { "epoch": 3.3513988260999956, "grad_norm": 5.77435302734375, "learning_rate": 2.4510113379795696e-05, "loss": 0.38836097717285156, "memory(GiB)": 70.5, "step": 78225, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.452578 }, { "epoch": 3.3516130414292444, "grad_norm": 2.6573729515075684, "learning_rate": 2.4504324025348912e-05, "loss": 0.2866321086883545, "memory(GiB)": 70.5, "step": 78230, "token_acc": 0.9337539432176656, "train_speed(iter/s)": 1.452585 }, { "epoch": 3.3518272567584937, "grad_norm": 3.025238037109375, "learning_rate": 2.44985351327802e-05, "loss": 0.347208571434021, "memory(GiB)": 70.5, "step": 78235, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.452592 }, { "epoch": 3.3520414720877425, "grad_norm": 2.464386463165283, "learning_rate": 2.4492746702194463e-05, "loss": 0.1984414577484131, "memory(GiB)": 70.5, "step": 78240, "token_acc": 0.9478260869565217, "train_speed(iter/s)": 1.4526 }, { "epoch": 3.3522556874169913, "grad_norm": 5.099549770355225, "learning_rate": 2.448695873369653e-05, "loss": 0.5052221298217774, "memory(GiB)": 70.5, "step": 78245, "token_acc": 0.8700361010830325, "train_speed(iter/s)": 1.452614 }, { "epoch": 3.3524699027462406, "grad_norm": 3.55637264251709, "learning_rate": 2.4481171227391293e-05, "loss": 0.4804656982421875, "memory(GiB)": 70.5, "step": 78250, "token_acc": 0.8969465648854962, "train_speed(iter/s)": 1.452633 }, { "epoch": 3.3526841180754894, "grad_norm": 1.3110451698303223, "learning_rate": 2.4475384183383577e-05, "loss": 0.28634445667266845, "memory(GiB)": 70.5, "step": 78255, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.452635 }, { "epoch": 3.352898333404738, "grad_norm": 6.210657596588135, "learning_rate": 2.4469597601778222e-05, "loss": 0.25720510482788084, "memory(GiB)": 70.5, "step": 78260, "token_acc": 0.9425675675675675, "train_speed(iter/s)": 1.45264 }, { "epoch": 3.3531125487339875, "grad_norm": 3.3236641883850098, "learning_rate": 2.446381148268005e-05, "loss": 0.20095763206481934, "memory(GiB)": 70.5, "step": 78265, "token_acc": 0.9569230769230769, "train_speed(iter/s)": 1.452642 }, { "epoch": 3.3533267640632363, "grad_norm": 1.869219183921814, "learning_rate": 2.445802582619389e-05, "loss": 0.17851028442382813, "memory(GiB)": 70.5, "step": 78270, "token_acc": 0.9552845528455285, "train_speed(iter/s)": 1.452661 }, { "epoch": 3.353540979392485, "grad_norm": 3.8665449619293213, "learning_rate": 2.4452240632424538e-05, "loss": 0.335982084274292, "memory(GiB)": 70.5, "step": 78275, "token_acc": 0.934375, "train_speed(iter/s)": 1.452672 }, { "epoch": 3.3537551947217343, "grad_norm": 1.3316243886947632, "learning_rate": 2.4446455901476828e-05, "loss": 0.26202433109283446, "memory(GiB)": 70.5, "step": 78280, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.452668 }, { "epoch": 3.353969410050983, "grad_norm": 2.2789146900177, "learning_rate": 2.4440671633455543e-05, "loss": 0.32268719673156737, "memory(GiB)": 70.5, "step": 78285, "token_acc": 0.9314516129032258, "train_speed(iter/s)": 1.452672 }, { "epoch": 3.3541836253802324, "grad_norm": 4.523410797119141, "learning_rate": 2.4434887828465463e-05, "loss": 0.3102756977081299, "memory(GiB)": 70.5, "step": 78290, "token_acc": 0.9519230769230769, "train_speed(iter/s)": 1.452674 }, { "epoch": 3.354397840709481, "grad_norm": 5.096879005432129, "learning_rate": 2.4429104486611376e-05, "loss": 0.33536500930786134, "memory(GiB)": 70.5, "step": 78295, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.452677 }, { "epoch": 3.35461205603873, "grad_norm": 2.634005546569824, "learning_rate": 2.4423321607998028e-05, "loss": 0.4404607772827148, "memory(GiB)": 70.5, "step": 78300, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.452673 }, { "epoch": 3.3548262713679793, "grad_norm": 5.265409469604492, "learning_rate": 2.4417539192730226e-05, "loss": 0.43228793144226074, "memory(GiB)": 70.5, "step": 78305, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.452679 }, { "epoch": 3.355040486697228, "grad_norm": 0.7847948670387268, "learning_rate": 2.4411757240912675e-05, "loss": 0.1457539439201355, "memory(GiB)": 70.5, "step": 78310, "token_acc": 0.9791666666666666, "train_speed(iter/s)": 1.452689 }, { "epoch": 3.355254702026477, "grad_norm": 0.502834677696228, "learning_rate": 2.4405975752650168e-05, "loss": 0.49944357872009276, "memory(GiB)": 70.5, "step": 78315, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.452695 }, { "epoch": 3.355468917355726, "grad_norm": 3.3414478302001953, "learning_rate": 2.4400194728047414e-05, "loss": 0.4325685977935791, "memory(GiB)": 70.5, "step": 78320, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.452698 }, { "epoch": 3.355683132684975, "grad_norm": 5.31052827835083, "learning_rate": 2.4394414167209152e-05, "loss": 0.40782904624938965, "memory(GiB)": 70.5, "step": 78325, "token_acc": 0.9132947976878613, "train_speed(iter/s)": 1.452699 }, { "epoch": 3.355897348014224, "grad_norm": 2.888258457183838, "learning_rate": 2.4388634070240097e-05, "loss": 0.47646398544311525, "memory(GiB)": 70.5, "step": 78330, "token_acc": 0.8782894736842105, "train_speed(iter/s)": 1.452696 }, { "epoch": 3.356111563343473, "grad_norm": 2.188753843307495, "learning_rate": 2.438285443724494e-05, "loss": 0.23748676776885985, "memory(GiB)": 70.5, "step": 78335, "token_acc": 0.9512820512820512, "train_speed(iter/s)": 1.452706 }, { "epoch": 3.356325778672722, "grad_norm": 4.877483367919922, "learning_rate": 2.4377075268328426e-05, "loss": 0.3567486763000488, "memory(GiB)": 70.5, "step": 78340, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.452709 }, { "epoch": 3.3565399940019707, "grad_norm": 1.937904953956604, "learning_rate": 2.437129656359523e-05, "loss": 0.6107395648956299, "memory(GiB)": 70.5, "step": 78345, "token_acc": 0.8518518518518519, "train_speed(iter/s)": 1.45272 }, { "epoch": 3.35675420933122, "grad_norm": 2.0989990234375, "learning_rate": 2.4365518323150037e-05, "loss": 0.21655261516571045, "memory(GiB)": 70.5, "step": 78350, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.452721 }, { "epoch": 3.3569684246604687, "grad_norm": 3.8378446102142334, "learning_rate": 2.4359740547097526e-05, "loss": 0.37326822280883787, "memory(GiB)": 70.5, "step": 78355, "token_acc": 0.9202898550724637, "train_speed(iter/s)": 1.452724 }, { "epoch": 3.3571826399897176, "grad_norm": 0.6654361486434937, "learning_rate": 2.435396323554235e-05, "loss": 0.2761711120605469, "memory(GiB)": 70.5, "step": 78360, "token_acc": 0.937125748502994, "train_speed(iter/s)": 1.452728 }, { "epoch": 3.357396855318967, "grad_norm": 2.8888680934906006, "learning_rate": 2.4348186388589206e-05, "loss": 0.36757113933563235, "memory(GiB)": 70.5, "step": 78365, "token_acc": 0.9372384937238494, "train_speed(iter/s)": 1.452731 }, { "epoch": 3.3576110706482156, "grad_norm": 2.510707378387451, "learning_rate": 2.4342410006342732e-05, "loss": 0.41069726943969725, "memory(GiB)": 70.5, "step": 78370, "token_acc": 0.9135802469135802, "train_speed(iter/s)": 1.452746 }, { "epoch": 3.3578252859774644, "grad_norm": 3.7046456336975098, "learning_rate": 2.4336634088907566e-05, "loss": 0.6009008884429932, "memory(GiB)": 70.5, "step": 78375, "token_acc": 0.8581081081081081, "train_speed(iter/s)": 1.452748 }, { "epoch": 3.3580395013067137, "grad_norm": 2.0989553928375244, "learning_rate": 2.4330858636388348e-05, "loss": 0.16226078271865846, "memory(GiB)": 70.5, "step": 78380, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.45275 }, { "epoch": 3.3582537166359625, "grad_norm": 6.0914835929870605, "learning_rate": 2.432508364888969e-05, "loss": 0.746791934967041, "memory(GiB)": 70.5, "step": 78385, "token_acc": 0.8456140350877193, "train_speed(iter/s)": 1.452753 }, { "epoch": 3.3584679319652113, "grad_norm": 2.4480953216552734, "learning_rate": 2.431930912651622e-05, "loss": 0.3260363578796387, "memory(GiB)": 70.5, "step": 78390, "token_acc": 0.932258064516129, "train_speed(iter/s)": 1.452763 }, { "epoch": 3.3586821472944606, "grad_norm": 1.9927682876586914, "learning_rate": 2.4313535069372584e-05, "loss": 0.3284327030181885, "memory(GiB)": 70.5, "step": 78395, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.452766 }, { "epoch": 3.3588963626237094, "grad_norm": 2.8612029552459717, "learning_rate": 2.430776147756335e-05, "loss": 0.35557971000671384, "memory(GiB)": 70.5, "step": 78400, "token_acc": 0.915625, "train_speed(iter/s)": 1.452764 }, { "epoch": 3.359110577952958, "grad_norm": 1.1914727687835693, "learning_rate": 2.4301988351193117e-05, "loss": 0.21464135646820068, "memory(GiB)": 70.5, "step": 78405, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.452768 }, { "epoch": 3.3593247932822075, "grad_norm": 3.1413068771362305, "learning_rate": 2.4296215690366476e-05, "loss": 0.3348440647125244, "memory(GiB)": 70.5, "step": 78410, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.452775 }, { "epoch": 3.3595390086114563, "grad_norm": 3.0133585929870605, "learning_rate": 2.4290443495188e-05, "loss": 0.2541869878768921, "memory(GiB)": 70.5, "step": 78415, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.452778 }, { "epoch": 3.359753223940705, "grad_norm": 2.1982357501983643, "learning_rate": 2.4284671765762235e-05, "loss": 0.4835477352142334, "memory(GiB)": 70.5, "step": 78420, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.452779 }, { "epoch": 3.3599674392699543, "grad_norm": 4.070955276489258, "learning_rate": 2.427890050219378e-05, "loss": 0.28475260734558105, "memory(GiB)": 70.5, "step": 78425, "token_acc": 0.9576547231270358, "train_speed(iter/s)": 1.452782 }, { "epoch": 3.360181654599203, "grad_norm": 3.125042200088501, "learning_rate": 2.427312970458718e-05, "loss": 0.2918938159942627, "memory(GiB)": 70.5, "step": 78430, "token_acc": 0.94921875, "train_speed(iter/s)": 1.452784 }, { "epoch": 3.360395869928452, "grad_norm": 4.184458255767822, "learning_rate": 2.426735937304696e-05, "loss": 0.3311163902282715, "memory(GiB)": 70.5, "step": 78435, "token_acc": 0.9351032448377581, "train_speed(iter/s)": 1.45279 }, { "epoch": 3.360610085257701, "grad_norm": 2.6020021438598633, "learning_rate": 2.426158950767767e-05, "loss": 0.3484587430953979, "memory(GiB)": 70.5, "step": 78440, "token_acc": 0.9304347826086956, "train_speed(iter/s)": 1.452792 }, { "epoch": 3.36082430058695, "grad_norm": 3.7155516147613525, "learning_rate": 2.425582010858381e-05, "loss": 0.5624904632568359, "memory(GiB)": 70.5, "step": 78445, "token_acc": 0.9006622516556292, "train_speed(iter/s)": 1.452793 }, { "epoch": 3.361038515916199, "grad_norm": 5.714865684509277, "learning_rate": 2.4250051175869938e-05, "loss": 0.5890889167785645, "memory(GiB)": 70.5, "step": 78450, "token_acc": 0.8929765886287625, "train_speed(iter/s)": 1.452801 }, { "epoch": 3.361252731245448, "grad_norm": 3.3580756187438965, "learning_rate": 2.4244282709640542e-05, "loss": 0.3946933031082153, "memory(GiB)": 70.5, "step": 78455, "token_acc": 0.9096209912536443, "train_speed(iter/s)": 1.452816 }, { "epoch": 3.361466946574697, "grad_norm": 3.553385019302368, "learning_rate": 2.4238514710000103e-05, "loss": 0.3979171276092529, "memory(GiB)": 70.5, "step": 78460, "token_acc": 0.9182389937106918, "train_speed(iter/s)": 1.452813 }, { "epoch": 3.3616811619039457, "grad_norm": 5.568390369415283, "learning_rate": 2.4232747177053167e-05, "loss": 0.54788818359375, "memory(GiB)": 70.5, "step": 78465, "token_acc": 0.8770764119601329, "train_speed(iter/s)": 1.452815 }, { "epoch": 3.361895377233195, "grad_norm": 3.634706497192383, "learning_rate": 2.422698011090418e-05, "loss": 0.19232583045959473, "memory(GiB)": 70.5, "step": 78470, "token_acc": 0.9540636042402827, "train_speed(iter/s)": 1.452823 }, { "epoch": 3.362109592562444, "grad_norm": 3.213806390762329, "learning_rate": 2.4221213511657624e-05, "loss": 0.5510274410247803, "memory(GiB)": 70.5, "step": 78475, "token_acc": 0.8612903225806452, "train_speed(iter/s)": 1.452828 }, { "epoch": 3.3623238078916926, "grad_norm": 3.5728933811187744, "learning_rate": 2.421544737941795e-05, "loss": 0.22204489707946778, "memory(GiB)": 70.5, "step": 78480, "token_acc": 0.9442379182156134, "train_speed(iter/s)": 1.452823 }, { "epoch": 3.362538023220942, "grad_norm": 3.4354004859924316, "learning_rate": 2.4209681714289655e-05, "loss": 0.45642337799072263, "memory(GiB)": 70.5, "step": 78485, "token_acc": 0.9079365079365079, "train_speed(iter/s)": 1.452825 }, { "epoch": 3.3627522385501907, "grad_norm": 0.535491406917572, "learning_rate": 2.4203916516377167e-05, "loss": 0.13571768999099731, "memory(GiB)": 70.5, "step": 78490, "token_acc": 0.954248366013072, "train_speed(iter/s)": 1.452826 }, { "epoch": 3.3629664538794395, "grad_norm": 1.0854538679122925, "learning_rate": 2.4198151785784934e-05, "loss": 0.1785615563392639, "memory(GiB)": 70.5, "step": 78495, "token_acc": 0.9461538461538461, "train_speed(iter/s)": 1.45283 }, { "epoch": 3.3631806692086887, "grad_norm": 5.704626560211182, "learning_rate": 2.4192387522617384e-05, "loss": 0.5393951892852783, "memory(GiB)": 70.5, "step": 78500, "token_acc": 0.9027237354085603, "train_speed(iter/s)": 1.452846 }, { "epoch": 3.3631806692086887, "eval_loss": 2.5548012256622314, "eval_runtime": 11.5491, "eval_samples_per_second": 8.659, "eval_steps_per_second": 8.659, "eval_token_acc": 0.45515394912985274, "step": 78500 }, { "epoch": 3.3633948845379376, "grad_norm": 3.1386213302612305, "learning_rate": 2.4186623726978925e-05, "loss": 0.22570929527282715, "memory(GiB)": 70.5, "step": 78505, "token_acc": 0.5877192982456141, "train_speed(iter/s)": 1.452525 }, { "epoch": 3.3636090998671864, "grad_norm": 2.810786485671997, "learning_rate": 2.418086039897401e-05, "loss": 0.5628177642822265, "memory(GiB)": 70.5, "step": 78510, "token_acc": 0.8585526315789473, "train_speed(iter/s)": 1.452525 }, { "epoch": 3.3638233151964356, "grad_norm": 3.392580986022949, "learning_rate": 2.4175097538707025e-05, "loss": 0.43055362701416017, "memory(GiB)": 70.5, "step": 78515, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.452528 }, { "epoch": 3.3640375305256844, "grad_norm": 4.887880802154541, "learning_rate": 2.4169335146282378e-05, "loss": 0.5123721122741699, "memory(GiB)": 70.5, "step": 78520, "token_acc": 0.8778625954198473, "train_speed(iter/s)": 1.452542 }, { "epoch": 3.3642517458549333, "grad_norm": 2.435434103012085, "learning_rate": 2.4163573221804457e-05, "loss": 0.5270310878753662, "memory(GiB)": 70.5, "step": 78525, "token_acc": 0.9071428571428571, "train_speed(iter/s)": 1.452542 }, { "epoch": 3.3644659611841825, "grad_norm": 4.457792282104492, "learning_rate": 2.4157811765377624e-05, "loss": 0.5445519924163819, "memory(GiB)": 70.5, "step": 78530, "token_acc": 0.8931750741839762, "train_speed(iter/s)": 1.452544 }, { "epoch": 3.3646801765134313, "grad_norm": 2.4414138793945312, "learning_rate": 2.4152050777106273e-05, "loss": 0.5350643157958984, "memory(GiB)": 70.5, "step": 78535, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.452545 }, { "epoch": 3.36489439184268, "grad_norm": 2.6395492553710938, "learning_rate": 2.414629025709479e-05, "loss": 0.5015573501586914, "memory(GiB)": 70.5, "step": 78540, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.45255 }, { "epoch": 3.3651086071719294, "grad_norm": 2.2441916465759277, "learning_rate": 2.414053020544751e-05, "loss": 0.24782259464263917, "memory(GiB)": 70.5, "step": 78545, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.452546 }, { "epoch": 3.365322822501178, "grad_norm": 0.6916298270225525, "learning_rate": 2.4134770622268783e-05, "loss": 0.46539783477783203, "memory(GiB)": 70.5, "step": 78550, "token_acc": 0.875, "train_speed(iter/s)": 1.452543 }, { "epoch": 3.365537037830427, "grad_norm": 4.1259918212890625, "learning_rate": 2.4129011507662945e-05, "loss": 0.3567992925643921, "memory(GiB)": 70.5, "step": 78555, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.452544 }, { "epoch": 3.3657512531596763, "grad_norm": 0.17121843993663788, "learning_rate": 2.4123252861734334e-05, "loss": 0.10485645532608032, "memory(GiB)": 70.5, "step": 78560, "token_acc": 0.976027397260274, "train_speed(iter/s)": 1.452548 }, { "epoch": 3.365965468488925, "grad_norm": 3.285890579223633, "learning_rate": 2.4117494684587262e-05, "loss": 0.4274646282196045, "memory(GiB)": 70.5, "step": 78565, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.452554 }, { "epoch": 3.366179683818174, "grad_norm": 3.7479324340820312, "learning_rate": 2.4111736976326066e-05, "loss": 0.2276614189147949, "memory(GiB)": 70.5, "step": 78570, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.452555 }, { "epoch": 3.366393899147423, "grad_norm": 5.198769569396973, "learning_rate": 2.410597973705504e-05, "loss": 0.5006923198699951, "memory(GiB)": 70.5, "step": 78575, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.452571 }, { "epoch": 3.366608114476672, "grad_norm": 3.6906707286834717, "learning_rate": 2.4100222966878484e-05, "loss": 0.698622179031372, "memory(GiB)": 70.5, "step": 78580, "token_acc": 0.8687943262411347, "train_speed(iter/s)": 1.452572 }, { "epoch": 3.3668223298059208, "grad_norm": 1.0113763809204102, "learning_rate": 2.409446666590068e-05, "loss": 0.23407416343688964, "memory(GiB)": 70.5, "step": 78585, "token_acc": 0.9629629629629629, "train_speed(iter/s)": 1.452585 }, { "epoch": 3.36703654513517, "grad_norm": 2.060856342315674, "learning_rate": 2.4088710834225896e-05, "loss": 0.3316983222961426, "memory(GiB)": 70.5, "step": 78590, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.452587 }, { "epoch": 3.367250760464419, "grad_norm": 4.134859561920166, "learning_rate": 2.408295547195844e-05, "loss": 0.4849548816680908, "memory(GiB)": 70.5, "step": 78595, "token_acc": 0.9026548672566371, "train_speed(iter/s)": 1.45259 }, { "epoch": 3.3674649757936677, "grad_norm": 1.4068779945373535, "learning_rate": 2.4077200579202563e-05, "loss": 0.15294694900512695, "memory(GiB)": 70.5, "step": 78600, "token_acc": 0.9627329192546584, "train_speed(iter/s)": 1.452588 }, { "epoch": 3.367679191122917, "grad_norm": 5.05280876159668, "learning_rate": 2.4071446156062494e-05, "loss": 0.13581908941268922, "memory(GiB)": 70.5, "step": 78605, "token_acc": 0.96, "train_speed(iter/s)": 1.452589 }, { "epoch": 3.3678934064521657, "grad_norm": 5.089373588562012, "learning_rate": 2.406569220264252e-05, "loss": 0.31252126693725585, "memory(GiB)": 70.5, "step": 78610, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.452603 }, { "epoch": 3.3681076217814145, "grad_norm": 2.8637442588806152, "learning_rate": 2.405993871904686e-05, "loss": 0.4491405010223389, "memory(GiB)": 70.5, "step": 78615, "token_acc": 0.9093851132686084, "train_speed(iter/s)": 1.452605 }, { "epoch": 3.368321837110664, "grad_norm": 3.4507694244384766, "learning_rate": 2.4054185705379724e-05, "loss": 0.5758914470672607, "memory(GiB)": 70.5, "step": 78620, "token_acc": 0.8378378378378378, "train_speed(iter/s)": 1.452611 }, { "epoch": 3.3685360524399126, "grad_norm": 4.219239711761475, "learning_rate": 2.404843316174537e-05, "loss": 0.2849971532821655, "memory(GiB)": 70.5, "step": 78625, "token_acc": 0.95, "train_speed(iter/s)": 1.452623 }, { "epoch": 3.3687502677691614, "grad_norm": 3.239497661590576, "learning_rate": 2.4042681088248e-05, "loss": 0.6901252269744873, "memory(GiB)": 70.5, "step": 78630, "token_acc": 0.8592592592592593, "train_speed(iter/s)": 1.452638 }, { "epoch": 3.3689644830984107, "grad_norm": 4.728368759155273, "learning_rate": 2.4036929484991804e-05, "loss": 0.29308271408081055, "memory(GiB)": 70.5, "step": 78635, "token_acc": 0.9283018867924528, "train_speed(iter/s)": 1.452641 }, { "epoch": 3.3691786984276595, "grad_norm": 1.9639208316802979, "learning_rate": 2.4031178352080992e-05, "loss": 0.43001999855041506, "memory(GiB)": 70.5, "step": 78640, "token_acc": 0.9202453987730062, "train_speed(iter/s)": 1.452653 }, { "epoch": 3.3693929137569083, "grad_norm": 3.716834783554077, "learning_rate": 2.402542768961974e-05, "loss": 0.46588888168334963, "memory(GiB)": 70.5, "step": 78645, "token_acc": 0.8879310344827587, "train_speed(iter/s)": 1.452664 }, { "epoch": 3.3696071290861576, "grad_norm": 1.7863279581069946, "learning_rate": 2.4019677497712216e-05, "loss": 0.12158517837524414, "memory(GiB)": 70.5, "step": 78650, "token_acc": 0.966542750929368, "train_speed(iter/s)": 1.452674 }, { "epoch": 3.3698213444154064, "grad_norm": 4.186831951141357, "learning_rate": 2.4013927776462625e-05, "loss": 0.3428311824798584, "memory(GiB)": 70.5, "step": 78655, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.45268 }, { "epoch": 3.370035559744655, "grad_norm": 1.771992802619934, "learning_rate": 2.4008178525975105e-05, "loss": 0.38280837535858153, "memory(GiB)": 70.5, "step": 78660, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.452681 }, { "epoch": 3.3702497750739044, "grad_norm": 5.226715087890625, "learning_rate": 2.4002429746353817e-05, "loss": 0.583591365814209, "memory(GiB)": 70.5, "step": 78665, "token_acc": 0.8629629629629629, "train_speed(iter/s)": 1.452692 }, { "epoch": 3.3704639904031533, "grad_norm": 5.0699567794799805, "learning_rate": 2.39966814377029e-05, "loss": 0.6065030097961426, "memory(GiB)": 70.5, "step": 78670, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.452703 }, { "epoch": 3.370678205732402, "grad_norm": 0.5214705467224121, "learning_rate": 2.3990933600126476e-05, "loss": 0.3885796308517456, "memory(GiB)": 70.5, "step": 78675, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.452717 }, { "epoch": 3.3708924210616513, "grad_norm": 3.8473868370056152, "learning_rate": 2.3985186233728686e-05, "loss": 0.19067974090576173, "memory(GiB)": 70.5, "step": 78680, "token_acc": 0.9597523219814241, "train_speed(iter/s)": 1.452719 }, { "epoch": 3.3711066363909, "grad_norm": 3.807373523712158, "learning_rate": 2.3979439338613668e-05, "loss": 0.4728085994720459, "memory(GiB)": 70.5, "step": 78685, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.452744 }, { "epoch": 3.371320851720149, "grad_norm": 2.0356907844543457, "learning_rate": 2.397369291488552e-05, "loss": 0.2948775291442871, "memory(GiB)": 70.5, "step": 78690, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.452761 }, { "epoch": 3.371535067049398, "grad_norm": 6.1090497970581055, "learning_rate": 2.3967946962648334e-05, "loss": 0.3616189002990723, "memory(GiB)": 70.5, "step": 78695, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.452756 }, { "epoch": 3.371749282378647, "grad_norm": 3.007840871810913, "learning_rate": 2.3962201482006215e-05, "loss": 0.2684352874755859, "memory(GiB)": 70.5, "step": 78700, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.452756 }, { "epoch": 3.371963497707896, "grad_norm": 3.851588726043701, "learning_rate": 2.395645647306324e-05, "loss": 0.4447641372680664, "memory(GiB)": 70.5, "step": 78705, "token_acc": 0.8893280632411067, "train_speed(iter/s)": 1.45276 }, { "epoch": 3.372177713037145, "grad_norm": 2.9217162132263184, "learning_rate": 2.3950711935923466e-05, "loss": 0.19317418336868286, "memory(GiB)": 70.5, "step": 78710, "token_acc": 0.9644128113879004, "train_speed(iter/s)": 1.452766 }, { "epoch": 3.372391928366394, "grad_norm": 0.25048765540122986, "learning_rate": 2.3944967870691003e-05, "loss": 0.20002763271331786, "memory(GiB)": 70.5, "step": 78715, "token_acc": 0.9616519174041298, "train_speed(iter/s)": 1.452771 }, { "epoch": 3.3726061436956427, "grad_norm": 0.2672334313392639, "learning_rate": 2.3939224277469886e-05, "loss": 0.378288197517395, "memory(GiB)": 70.5, "step": 78720, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.452772 }, { "epoch": 3.372820359024892, "grad_norm": 0.4226509928703308, "learning_rate": 2.3933481156364168e-05, "loss": 0.09232691526412964, "memory(GiB)": 70.5, "step": 78725, "token_acc": 0.9776119402985075, "train_speed(iter/s)": 1.452769 }, { "epoch": 3.373034574354141, "grad_norm": 3.1943657398223877, "learning_rate": 2.392773850747789e-05, "loss": 0.2649980068206787, "memory(GiB)": 70.5, "step": 78730, "token_acc": 0.9438596491228071, "train_speed(iter/s)": 1.452769 }, { "epoch": 3.3732487896833896, "grad_norm": 0.38678884506225586, "learning_rate": 2.3921996330915076e-05, "loss": 0.2872999429702759, "memory(GiB)": 70.5, "step": 78735, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.452769 }, { "epoch": 3.373463005012639, "grad_norm": 3.3316001892089844, "learning_rate": 2.391625462677977e-05, "loss": 0.37246153354644773, "memory(GiB)": 70.5, "step": 78740, "token_acc": 0.8835227272727273, "train_speed(iter/s)": 1.452771 }, { "epoch": 3.3736772203418877, "grad_norm": 2.167222738265991, "learning_rate": 2.3910513395175988e-05, "loss": 0.20870823860168458, "memory(GiB)": 70.5, "step": 78745, "token_acc": 0.9543568464730291, "train_speed(iter/s)": 1.452777 }, { "epoch": 3.3738914356711365, "grad_norm": 6.065853595733643, "learning_rate": 2.3904772636207723e-05, "loss": 0.3328468561172485, "memory(GiB)": 70.5, "step": 78750, "token_acc": 0.9085173501577287, "train_speed(iter/s)": 1.45279 }, { "epoch": 3.3741056510003857, "grad_norm": 3.301539182662964, "learning_rate": 2.3899032349978967e-05, "loss": 0.24842898845672606, "memory(GiB)": 70.5, "step": 78755, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.452787 }, { "epoch": 3.3743198663296345, "grad_norm": 3.299285650253296, "learning_rate": 2.389329253659374e-05, "loss": 0.44006872177124023, "memory(GiB)": 70.5, "step": 78760, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.452791 }, { "epoch": 3.3745340816588834, "grad_norm": 4.061077117919922, "learning_rate": 2.3887553196155995e-05, "loss": 0.3744380474090576, "memory(GiB)": 70.5, "step": 78765, "token_acc": 0.8993506493506493, "train_speed(iter/s)": 1.452791 }, { "epoch": 3.3747482969881326, "grad_norm": 4.451459884643555, "learning_rate": 2.3881814328769737e-05, "loss": 0.18556900024414064, "memory(GiB)": 70.5, "step": 78770, "token_acc": 0.9627118644067797, "train_speed(iter/s)": 1.452791 }, { "epoch": 3.3749625123173814, "grad_norm": 7.1551055908203125, "learning_rate": 2.387607593453891e-05, "loss": 0.22322065830230714, "memory(GiB)": 70.5, "step": 78775, "token_acc": 0.9620253164556962, "train_speed(iter/s)": 1.452794 }, { "epoch": 3.3751767276466302, "grad_norm": 4.760532379150391, "learning_rate": 2.3870338013567474e-05, "loss": 0.21448419094085694, "memory(GiB)": 70.5, "step": 78780, "token_acc": 0.9509433962264151, "train_speed(iter/s)": 1.452796 }, { "epoch": 3.3753909429758795, "grad_norm": 1.773089051246643, "learning_rate": 2.3864600565959377e-05, "loss": 0.2505277395248413, "memory(GiB)": 70.5, "step": 78785, "token_acc": 0.9546925566343042, "train_speed(iter/s)": 1.452799 }, { "epoch": 3.3756051583051283, "grad_norm": 3.243138551712036, "learning_rate": 2.3858863591818558e-05, "loss": 0.5259219169616699, "memory(GiB)": 70.5, "step": 78790, "token_acc": 0.8875, "train_speed(iter/s)": 1.452798 }, { "epoch": 3.375819373634377, "grad_norm": 4.885090351104736, "learning_rate": 2.385312709124893e-05, "loss": 0.5123190879821777, "memory(GiB)": 70.5, "step": 78795, "token_acc": 0.8927335640138409, "train_speed(iter/s)": 1.452804 }, { "epoch": 3.3760335889636264, "grad_norm": 5.642121315002441, "learning_rate": 2.3847391064354453e-05, "loss": 0.43967428207397463, "memory(GiB)": 70.5, "step": 78800, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.452804 }, { "epoch": 3.376247804292875, "grad_norm": 0.6879695653915405, "learning_rate": 2.3841655511239013e-05, "loss": 0.28861174583435056, "memory(GiB)": 70.5, "step": 78805, "token_acc": 0.9426229508196722, "train_speed(iter/s)": 1.452814 }, { "epoch": 3.376462019622124, "grad_norm": 3.5620057582855225, "learning_rate": 2.3835920432006527e-05, "loss": 0.5505706787109375, "memory(GiB)": 70.5, "step": 78810, "token_acc": 0.8975155279503105, "train_speed(iter/s)": 1.452828 }, { "epoch": 3.3766762349513733, "grad_norm": 1.5606770515441895, "learning_rate": 2.3830185826760887e-05, "loss": 0.16945838928222656, "memory(GiB)": 70.5, "step": 78815, "token_acc": 0.948905109489051, "train_speed(iter/s)": 1.452833 }, { "epoch": 3.376890450280622, "grad_norm": 0.5344064831733704, "learning_rate": 2.3824451695605958e-05, "loss": 0.20311329364776612, "memory(GiB)": 70.5, "step": 78820, "token_acc": 0.9529411764705882, "train_speed(iter/s)": 1.452836 }, { "epoch": 3.377104665609871, "grad_norm": 3.643660545349121, "learning_rate": 2.381871803864566e-05, "loss": 0.3634533166885376, "memory(GiB)": 70.5, "step": 78825, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.452844 }, { "epoch": 3.37731888093912, "grad_norm": 2.0988497734069824, "learning_rate": 2.381298485598383e-05, "loss": 0.2601193904876709, "memory(GiB)": 70.5, "step": 78830, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.452846 }, { "epoch": 3.377533096268369, "grad_norm": 5.719571590423584, "learning_rate": 2.3807252147724362e-05, "loss": 0.3226768493652344, "memory(GiB)": 70.5, "step": 78835, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.452848 }, { "epoch": 3.3777473115976178, "grad_norm": 2.2969396114349365, "learning_rate": 2.380151991397109e-05, "loss": 0.27203660011291503, "memory(GiB)": 70.5, "step": 78840, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.452848 }, { "epoch": 3.377961526926867, "grad_norm": 2.949617862701416, "learning_rate": 2.3795788154827862e-05, "loss": 0.3108374118804932, "memory(GiB)": 70.5, "step": 78845, "token_acc": 0.9312714776632303, "train_speed(iter/s)": 1.452853 }, { "epoch": 3.378175742256116, "grad_norm": 2.2293808460235596, "learning_rate": 2.3790056870398515e-05, "loss": 0.3626237869262695, "memory(GiB)": 70.5, "step": 78850, "token_acc": 0.9159159159159159, "train_speed(iter/s)": 1.452855 }, { "epoch": 3.3783899575853646, "grad_norm": 0.2666146755218506, "learning_rate": 2.3784326060786855e-05, "loss": 0.4182257652282715, "memory(GiB)": 70.5, "step": 78855, "token_acc": 0.932258064516129, "train_speed(iter/s)": 1.452857 }, { "epoch": 3.378604172914614, "grad_norm": 0.3005029559135437, "learning_rate": 2.3778595726096737e-05, "loss": 0.13591084480285645, "memory(GiB)": 70.5, "step": 78860, "token_acc": 0.9671641791044776, "train_speed(iter/s)": 1.452859 }, { "epoch": 3.3788183882438627, "grad_norm": 4.068772315979004, "learning_rate": 2.3772865866431955e-05, "loss": 0.30746870040893554, "memory(GiB)": 70.5, "step": 78865, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.452859 }, { "epoch": 3.3790326035731115, "grad_norm": 2.4213595390319824, "learning_rate": 2.3767136481896312e-05, "loss": 0.28689401149749755, "memory(GiB)": 70.5, "step": 78870, "token_acc": 0.9494047619047619, "train_speed(iter/s)": 1.452863 }, { "epoch": 3.379246818902361, "grad_norm": 4.548860549926758, "learning_rate": 2.3761407572593603e-05, "loss": 0.36079773902893064, "memory(GiB)": 70.5, "step": 78875, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.452864 }, { "epoch": 3.3794610342316096, "grad_norm": 8.68009090423584, "learning_rate": 2.375567913862759e-05, "loss": 0.3419347763061523, "memory(GiB)": 70.5, "step": 78880, "token_acc": 0.9297520661157025, "train_speed(iter/s)": 1.452862 }, { "epoch": 3.3796752495608584, "grad_norm": 3.9226291179656982, "learning_rate": 2.3749951180102082e-05, "loss": 0.40648632049560546, "memory(GiB)": 70.5, "step": 78885, "token_acc": 0.9116719242902208, "train_speed(iter/s)": 1.452874 }, { "epoch": 3.3798894648901077, "grad_norm": 3.430513620376587, "learning_rate": 2.3744223697120836e-05, "loss": 0.13671023845672609, "memory(GiB)": 70.5, "step": 78890, "token_acc": 0.9656488549618321, "train_speed(iter/s)": 1.45287 }, { "epoch": 3.3801036802193565, "grad_norm": 3.384965658187866, "learning_rate": 2.373849668978761e-05, "loss": 0.27232339382171633, "memory(GiB)": 70.5, "step": 78895, "token_acc": 0.9399293286219081, "train_speed(iter/s)": 1.452878 }, { "epoch": 3.3803178955486053, "grad_norm": 6.928366184234619, "learning_rate": 2.373277015820613e-05, "loss": 0.3910300016403198, "memory(GiB)": 70.5, "step": 78900, "token_acc": 0.9068825910931174, "train_speed(iter/s)": 1.452884 }, { "epoch": 3.3805321108778545, "grad_norm": 1.9804586172103882, "learning_rate": 2.3727044102480184e-05, "loss": 0.5623263835906982, "memory(GiB)": 70.5, "step": 78905, "token_acc": 0.8582677165354331, "train_speed(iter/s)": 1.452881 }, { "epoch": 3.3807463262071034, "grad_norm": 3.616950035095215, "learning_rate": 2.3721318522713453e-05, "loss": 0.26888513565063477, "memory(GiB)": 70.5, "step": 78910, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.452884 }, { "epoch": 3.380960541536352, "grad_norm": 5.480851173400879, "learning_rate": 2.3715593419009714e-05, "loss": 0.46216440200805664, "memory(GiB)": 70.5, "step": 78915, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.452886 }, { "epoch": 3.3811747568656014, "grad_norm": 1.7701053619384766, "learning_rate": 2.3709868791472652e-05, "loss": 0.34623396396636963, "memory(GiB)": 70.5, "step": 78920, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.45289 }, { "epoch": 3.3813889721948502, "grad_norm": 3.7268660068511963, "learning_rate": 2.3704144640205983e-05, "loss": 0.3118525266647339, "memory(GiB)": 70.5, "step": 78925, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.452901 }, { "epoch": 3.381603187524099, "grad_norm": 11.126194953918457, "learning_rate": 2.3698420965313395e-05, "loss": 0.35961313247680665, "memory(GiB)": 70.5, "step": 78930, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.452903 }, { "epoch": 3.3818174028533483, "grad_norm": 5.51720666885376, "learning_rate": 2.3692697766898592e-05, "loss": 0.43100895881652834, "memory(GiB)": 70.5, "step": 78935, "token_acc": 0.93, "train_speed(iter/s)": 1.452909 }, { "epoch": 3.382031618182597, "grad_norm": 2.409587860107422, "learning_rate": 2.3686975045065223e-05, "loss": 0.4764093399047852, "memory(GiB)": 70.5, "step": 78940, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.45292 }, { "epoch": 3.382245833511846, "grad_norm": 5.114251613616943, "learning_rate": 2.3681252799917002e-05, "loss": 0.3175379514694214, "memory(GiB)": 70.5, "step": 78945, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.452933 }, { "epoch": 3.382460048841095, "grad_norm": 2.500772714614868, "learning_rate": 2.367553103155758e-05, "loss": 0.16132690906524658, "memory(GiB)": 70.5, "step": 78950, "token_acc": 0.9602888086642599, "train_speed(iter/s)": 1.452944 }, { "epoch": 3.382674264170344, "grad_norm": 0.31106752157211304, "learning_rate": 2.366980974009061e-05, "loss": 0.30916500091552734, "memory(GiB)": 70.5, "step": 78955, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.452945 }, { "epoch": 3.382888479499593, "grad_norm": 4.698276996612549, "learning_rate": 2.3664088925619732e-05, "loss": 0.5753740310668946, "memory(GiB)": 70.5, "step": 78960, "token_acc": 0.8732394366197183, "train_speed(iter/s)": 1.45295 }, { "epoch": 3.383102694828842, "grad_norm": 5.601687431335449, "learning_rate": 2.365836858824857e-05, "loss": 0.377410888671875, "memory(GiB)": 70.5, "step": 78965, "token_acc": 0.92, "train_speed(iter/s)": 1.452957 }, { "epoch": 3.383316910158091, "grad_norm": 3.402315616607666, "learning_rate": 2.365264872808079e-05, "loss": 0.39947071075439455, "memory(GiB)": 70.5, "step": 78970, "token_acc": 0.9224489795918367, "train_speed(iter/s)": 1.452961 }, { "epoch": 3.3835311254873397, "grad_norm": 5.141327381134033, "learning_rate": 2.3646929345219975e-05, "loss": 0.6470312118530274, "memory(GiB)": 70.5, "step": 78975, "token_acc": 0.8647798742138365, "train_speed(iter/s)": 1.452961 }, { "epoch": 3.383745340816589, "grad_norm": 2.936906099319458, "learning_rate": 2.3641210439769773e-05, "loss": 0.24275248050689696, "memory(GiB)": 70.5, "step": 78980, "token_acc": 0.9433333333333334, "train_speed(iter/s)": 1.452964 }, { "epoch": 3.3839595561458378, "grad_norm": 3.097946882247925, "learning_rate": 2.3635492011833778e-05, "loss": 0.38442130088806153, "memory(GiB)": 70.5, "step": 78985, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.452968 }, { "epoch": 3.3841737714750866, "grad_norm": 0.45016252994537354, "learning_rate": 2.362977406151557e-05, "loss": 0.30955770015716555, "memory(GiB)": 70.5, "step": 78990, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.45297 }, { "epoch": 3.384387986804336, "grad_norm": 7.179278373718262, "learning_rate": 2.362405658891874e-05, "loss": 0.44128990173339844, "memory(GiB)": 70.5, "step": 78995, "token_acc": 0.9256965944272446, "train_speed(iter/s)": 1.452969 }, { "epoch": 3.3846022021335846, "grad_norm": 2.825849771499634, "learning_rate": 2.3618339594146853e-05, "loss": 0.15058984756469726, "memory(GiB)": 70.5, "step": 79000, "token_acc": 0.9598540145985401, "train_speed(iter/s)": 1.452972 }, { "epoch": 3.3846022021335846, "eval_loss": 2.502291440963745, "eval_runtime": 11.8399, "eval_samples_per_second": 8.446, "eval_steps_per_second": 8.446, "eval_token_acc": 0.45269461077844314, "step": 79000 }, { "epoch": 3.3848164174628335, "grad_norm": 3.6579337120056152, "learning_rate": 2.3612623077303514e-05, "loss": 0.4652080535888672, "memory(GiB)": 70.5, "step": 79005, "token_acc": 0.5813356164383562, "train_speed(iter/s)": 1.452626 }, { "epoch": 3.3850306327920827, "grad_norm": 3.554410934448242, "learning_rate": 2.360690703849226e-05, "loss": 0.5583171844482422, "memory(GiB)": 70.5, "step": 79010, "token_acc": 0.8875, "train_speed(iter/s)": 1.452626 }, { "epoch": 3.3852448481213315, "grad_norm": 1.911500096321106, "learning_rate": 2.360119147781664e-05, "loss": 0.26251299381256105, "memory(GiB)": 70.5, "step": 79015, "token_acc": 0.9547038327526133, "train_speed(iter/s)": 1.452632 }, { "epoch": 3.3854590634505803, "grad_norm": 2.2827775478363037, "learning_rate": 2.35954763953802e-05, "loss": 0.2825585603713989, "memory(GiB)": 70.5, "step": 79020, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.452637 }, { "epoch": 3.3856732787798296, "grad_norm": 3.114319324493408, "learning_rate": 2.3589761791286462e-05, "loss": 0.3267849922180176, "memory(GiB)": 70.5, "step": 79025, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.452647 }, { "epoch": 3.3858874941090784, "grad_norm": 2.547780990600586, "learning_rate": 2.3584047665638977e-05, "loss": 0.5302196502685547, "memory(GiB)": 70.5, "step": 79030, "token_acc": 0.8754098360655738, "train_speed(iter/s)": 1.452653 }, { "epoch": 3.3861017094383272, "grad_norm": 7.3905158042907715, "learning_rate": 2.3578334018541254e-05, "loss": 0.7593622207641602, "memory(GiB)": 70.5, "step": 79035, "token_acc": 0.8566775244299675, "train_speed(iter/s)": 1.452665 }, { "epoch": 3.3863159247675765, "grad_norm": 0.4999566078186035, "learning_rate": 2.3572620850096787e-05, "loss": 0.2679693937301636, "memory(GiB)": 70.5, "step": 79040, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.452667 }, { "epoch": 3.3865301400968253, "grad_norm": 5.322432518005371, "learning_rate": 2.356690816040909e-05, "loss": 0.6801563262939453, "memory(GiB)": 70.5, "step": 79045, "token_acc": 0.85, "train_speed(iter/s)": 1.452664 }, { "epoch": 3.386744355426074, "grad_norm": 4.882505893707275, "learning_rate": 2.3561195949581634e-05, "loss": 0.4080522060394287, "memory(GiB)": 70.5, "step": 79050, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.452671 }, { "epoch": 3.3869585707553234, "grad_norm": 5.0974955558776855, "learning_rate": 2.3555484217717904e-05, "loss": 0.33201026916503906, "memory(GiB)": 70.5, "step": 79055, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.452677 }, { "epoch": 3.387172786084572, "grad_norm": 2.9933106899261475, "learning_rate": 2.3549772964921413e-05, "loss": 0.24079580307006837, "memory(GiB)": 70.5, "step": 79060, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.45268 }, { "epoch": 3.387387001413821, "grad_norm": 2.761547565460205, "learning_rate": 2.354406219129559e-05, "loss": 0.3925196647644043, "memory(GiB)": 70.5, "step": 79065, "token_acc": 0.8954703832752613, "train_speed(iter/s)": 1.452692 }, { "epoch": 3.3876012167430702, "grad_norm": 3.356717824935913, "learning_rate": 2.35383518969439e-05, "loss": 0.2993541479110718, "memory(GiB)": 70.5, "step": 79070, "token_acc": 0.9233038348082596, "train_speed(iter/s)": 1.452695 }, { "epoch": 3.387815432072319, "grad_norm": 3.478573799133301, "learning_rate": 2.353264208196979e-05, "loss": 0.1663053274154663, "memory(GiB)": 70.5, "step": 79075, "token_acc": 0.9644268774703557, "train_speed(iter/s)": 1.452692 }, { "epoch": 3.388029647401568, "grad_norm": 7.844701766967773, "learning_rate": 2.35269327464767e-05, "loss": 0.5118204593658447, "memory(GiB)": 70.5, "step": 79080, "token_acc": 0.891640866873065, "train_speed(iter/s)": 1.452689 }, { "epoch": 3.388243862730817, "grad_norm": 3.137420654296875, "learning_rate": 2.3521223890568032e-05, "loss": 0.2921419620513916, "memory(GiB)": 70.5, "step": 79085, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.452702 }, { "epoch": 3.388458078060066, "grad_norm": 3.0160293579101562, "learning_rate": 2.3515515514347252e-05, "loss": 0.5615220546722413, "memory(GiB)": 70.5, "step": 79090, "token_acc": 0.8753709198813057, "train_speed(iter/s)": 1.4527 }, { "epoch": 3.3886722933893147, "grad_norm": 2.4000086784362793, "learning_rate": 2.3509807617917757e-05, "loss": 0.39062483310699464, "memory(GiB)": 70.5, "step": 79095, "token_acc": 0.9233870967741935, "train_speed(iter/s)": 1.452705 }, { "epoch": 3.388886508718564, "grad_norm": 2.409170627593994, "learning_rate": 2.3504100201382945e-05, "loss": 0.6927772998809815, "memory(GiB)": 70.5, "step": 79100, "token_acc": 0.8673469387755102, "train_speed(iter/s)": 1.452706 }, { "epoch": 3.389100724047813, "grad_norm": 5.576630592346191, "learning_rate": 2.3498393264846212e-05, "loss": 0.4842530250549316, "memory(GiB)": 70.5, "step": 79105, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.452711 }, { "epoch": 3.3893149393770616, "grad_norm": 4.776660442352295, "learning_rate": 2.349268680841093e-05, "loss": 0.38490703105926516, "memory(GiB)": 70.5, "step": 79110, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.452727 }, { "epoch": 3.389529154706311, "grad_norm": 0.5906184911727905, "learning_rate": 2.3486980832180505e-05, "loss": 0.3786055088043213, "memory(GiB)": 70.5, "step": 79115, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.452742 }, { "epoch": 3.3897433700355597, "grad_norm": 5.35623836517334, "learning_rate": 2.34812753362583e-05, "loss": 0.33508806228637694, "memory(GiB)": 70.5, "step": 79120, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.452745 }, { "epoch": 3.3899575853648085, "grad_norm": 5.119927883148193, "learning_rate": 2.3475570320747647e-05, "loss": 0.3483626365661621, "memory(GiB)": 70.5, "step": 79125, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.452743 }, { "epoch": 3.3901718006940578, "grad_norm": 6.288876056671143, "learning_rate": 2.3469865785751938e-05, "loss": 0.3931364297866821, "memory(GiB)": 70.5, "step": 79130, "token_acc": 0.9157088122605364, "train_speed(iter/s)": 1.452747 }, { "epoch": 3.3903860160233066, "grad_norm": 3.0077881813049316, "learning_rate": 2.3464161731374496e-05, "loss": 0.48528246879577636, "memory(GiB)": 70.5, "step": 79135, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.452758 }, { "epoch": 3.390600231352556, "grad_norm": 2.114576816558838, "learning_rate": 2.345845815771866e-05, "loss": 0.23717358112335205, "memory(GiB)": 70.5, "step": 79140, "token_acc": 0.9507575757575758, "train_speed(iter/s)": 1.452769 }, { "epoch": 3.3908144466818047, "grad_norm": 0.34763485193252563, "learning_rate": 2.3452755064887732e-05, "loss": 0.17309553623199464, "memory(GiB)": 70.5, "step": 79145, "token_acc": 0.962457337883959, "train_speed(iter/s)": 1.45278 }, { "epoch": 3.3910286620110535, "grad_norm": 6.249352931976318, "learning_rate": 2.3447052452985068e-05, "loss": 0.5964873790740967, "memory(GiB)": 70.5, "step": 79150, "token_acc": 0.88125, "train_speed(iter/s)": 1.452784 }, { "epoch": 3.3912428773403027, "grad_norm": 2.34047794342041, "learning_rate": 2.3441350322113957e-05, "loss": 0.5917922019958496, "memory(GiB)": 70.5, "step": 79155, "token_acc": 0.8881578947368421, "train_speed(iter/s)": 1.452784 }, { "epoch": 3.3914570926695515, "grad_norm": 2.353882312774658, "learning_rate": 2.3435648672377702e-05, "loss": 0.19732013940811158, "memory(GiB)": 70.5, "step": 79160, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.452801 }, { "epoch": 3.3916713079988003, "grad_norm": 6.5087995529174805, "learning_rate": 2.342994750387959e-05, "loss": 0.473177433013916, "memory(GiB)": 70.5, "step": 79165, "token_acc": 0.8982035928143712, "train_speed(iter/s)": 1.452804 }, { "epoch": 3.3918855233280496, "grad_norm": 0.4054802358150482, "learning_rate": 2.3424246816722884e-05, "loss": 0.1568283200263977, "memory(GiB)": 70.5, "step": 79170, "token_acc": 0.9647887323943662, "train_speed(iter/s)": 1.452805 }, { "epoch": 3.3920997386572984, "grad_norm": 3.586491584777832, "learning_rate": 2.3418546611010895e-05, "loss": 0.4633010387420654, "memory(GiB)": 70.5, "step": 79175, "token_acc": 0.9171597633136095, "train_speed(iter/s)": 1.452815 }, { "epoch": 3.3923139539865472, "grad_norm": 6.741687774658203, "learning_rate": 2.3412846886846867e-05, "loss": 0.27527453899383547, "memory(GiB)": 70.5, "step": 79180, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.452827 }, { "epoch": 3.3925281693157965, "grad_norm": 4.874637603759766, "learning_rate": 2.3407147644334067e-05, "loss": 0.3243346452713013, "memory(GiB)": 70.5, "step": 79185, "token_acc": 0.9395973154362416, "train_speed(iter/s)": 1.452824 }, { "epoch": 3.3927423846450453, "grad_norm": 4.154573440551758, "learning_rate": 2.340144888357572e-05, "loss": 0.335373854637146, "memory(GiB)": 70.5, "step": 79190, "token_acc": 0.9240121580547113, "train_speed(iter/s)": 1.452825 }, { "epoch": 3.392956599974294, "grad_norm": 1.9809952974319458, "learning_rate": 2.339575060467507e-05, "loss": 0.2637688636779785, "memory(GiB)": 70.5, "step": 79195, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.452824 }, { "epoch": 3.3931708153035434, "grad_norm": 1.5639753341674805, "learning_rate": 2.3390052807735352e-05, "loss": 0.4759495258331299, "memory(GiB)": 70.5, "step": 79200, "token_acc": 0.8997289972899729, "train_speed(iter/s)": 1.452824 }, { "epoch": 3.393385030632792, "grad_norm": 11.070868492126465, "learning_rate": 2.338435549285981e-05, "loss": 0.7134777545928955, "memory(GiB)": 70.5, "step": 79205, "token_acc": 0.8548812664907651, "train_speed(iter/s)": 1.452827 }, { "epoch": 3.393599245962041, "grad_norm": 5.657430171966553, "learning_rate": 2.337865866015163e-05, "loss": 0.5790190696716309, "memory(GiB)": 70.5, "step": 79210, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.452852 }, { "epoch": 3.3938134612912902, "grad_norm": 4.42686653137207, "learning_rate": 2.3372962309714023e-05, "loss": 0.23923258781433104, "memory(GiB)": 70.5, "step": 79215, "token_acc": 0.9398496240601504, "train_speed(iter/s)": 1.452854 }, { "epoch": 3.394027676620539, "grad_norm": 3.6175074577331543, "learning_rate": 2.3367266441650188e-05, "loss": 0.3856826305389404, "memory(GiB)": 70.5, "step": 79220, "token_acc": 0.924791086350975, "train_speed(iter/s)": 1.452865 }, { "epoch": 3.394241891949788, "grad_norm": 1.5739353895187378, "learning_rate": 2.3361571056063302e-05, "loss": 0.39290571212768555, "memory(GiB)": 70.5, "step": 79225, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.452867 }, { "epoch": 3.394456107279037, "grad_norm": 3.3416450023651123, "learning_rate": 2.335587615305652e-05, "loss": 0.3688590288162231, "memory(GiB)": 70.5, "step": 79230, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.45287 }, { "epoch": 3.394670322608286, "grad_norm": 2.0455379486083984, "learning_rate": 2.335018173273306e-05, "loss": 0.20198473930358887, "memory(GiB)": 70.5, "step": 79235, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.452882 }, { "epoch": 3.3948845379375348, "grad_norm": 1.6649935245513916, "learning_rate": 2.3344487795196063e-05, "loss": 0.223266077041626, "memory(GiB)": 70.5, "step": 79240, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.452887 }, { "epoch": 3.395098753266784, "grad_norm": 5.327976226806641, "learning_rate": 2.3338794340548666e-05, "loss": 0.8140510559082031, "memory(GiB)": 70.5, "step": 79245, "token_acc": 0.8496932515337423, "train_speed(iter/s)": 1.452885 }, { "epoch": 3.395312968596033, "grad_norm": 5.439789772033691, "learning_rate": 2.3333101368894024e-05, "loss": 0.5795151233673096, "memory(GiB)": 70.5, "step": 79250, "token_acc": 0.8884462151394422, "train_speed(iter/s)": 1.452883 }, { "epoch": 3.3955271839252816, "grad_norm": 3.3713791370391846, "learning_rate": 2.3327408880335245e-05, "loss": 0.893238639831543, "memory(GiB)": 70.5, "step": 79255, "token_acc": 0.8131147540983606, "train_speed(iter/s)": 1.45289 }, { "epoch": 3.395741399254531, "grad_norm": 1.3969827890396118, "learning_rate": 2.3321716874975498e-05, "loss": 0.28191843032836916, "memory(GiB)": 70.5, "step": 79260, "token_acc": 0.9536423841059603, "train_speed(iter/s)": 1.452893 }, { "epoch": 3.3959556145837797, "grad_norm": 2.0735862255096436, "learning_rate": 2.331602535291787e-05, "loss": 0.347678017616272, "memory(GiB)": 70.5, "step": 79265, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.452895 }, { "epoch": 3.3961698299130285, "grad_norm": 2.616770029067993, "learning_rate": 2.331033431426546e-05, "loss": 0.23721938133239745, "memory(GiB)": 70.5, "step": 79270, "token_acc": 0.9375, "train_speed(iter/s)": 1.452903 }, { "epoch": 3.3963840452422778, "grad_norm": 2.959458351135254, "learning_rate": 2.3304643759121398e-05, "loss": 0.17467098236083983, "memory(GiB)": 70.5, "step": 79275, "token_acc": 0.9551724137931035, "train_speed(iter/s)": 1.452909 }, { "epoch": 3.3965982605715266, "grad_norm": 2.528858184814453, "learning_rate": 2.3298953687588753e-05, "loss": 0.5174966335296631, "memory(GiB)": 70.5, "step": 79280, "token_acc": 0.9272151898734177, "train_speed(iter/s)": 1.452915 }, { "epoch": 3.3968124759007754, "grad_norm": 2.2934470176696777, "learning_rate": 2.3293264099770613e-05, "loss": 0.34972386360168456, "memory(GiB)": 70.5, "step": 79285, "token_acc": 0.9293286219081273, "train_speed(iter/s)": 1.452917 }, { "epoch": 3.3970266912300247, "grad_norm": 3.487302303314209, "learning_rate": 2.3287574995770028e-05, "loss": 0.4129924774169922, "memory(GiB)": 70.5, "step": 79290, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.452926 }, { "epoch": 3.3972409065592735, "grad_norm": 4.73537015914917, "learning_rate": 2.32818863756901e-05, "loss": 0.4287053108215332, "memory(GiB)": 70.5, "step": 79295, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.452943 }, { "epoch": 3.3974551218885223, "grad_norm": 3.69347882270813, "learning_rate": 2.327619823963386e-05, "loss": 0.5705577850341796, "memory(GiB)": 70.5, "step": 79300, "token_acc": 0.8817567567567568, "train_speed(iter/s)": 1.452952 }, { "epoch": 3.3976693372177715, "grad_norm": 3.569279432296753, "learning_rate": 2.3270510587704365e-05, "loss": 0.3989548206329346, "memory(GiB)": 70.5, "step": 79305, "token_acc": 0.9099378881987578, "train_speed(iter/s)": 1.452986 }, { "epoch": 3.3978835525470203, "grad_norm": 3.7879281044006348, "learning_rate": 2.326482342000464e-05, "loss": 0.34657213687896726, "memory(GiB)": 70.5, "step": 79310, "token_acc": 0.9029126213592233, "train_speed(iter/s)": 1.452994 }, { "epoch": 3.398097767876269, "grad_norm": 3.0124869346618652, "learning_rate": 2.3259136736637697e-05, "loss": 0.21854448318481445, "memory(GiB)": 70.5, "step": 79315, "token_acc": 0.9504643962848297, "train_speed(iter/s)": 1.452998 }, { "epoch": 3.3983119832055184, "grad_norm": 3.271376609802246, "learning_rate": 2.32534505377066e-05, "loss": 0.6189711570739747, "memory(GiB)": 70.5, "step": 79320, "token_acc": 0.8862745098039215, "train_speed(iter/s)": 1.452998 }, { "epoch": 3.3985261985347672, "grad_norm": 3.485388994216919, "learning_rate": 2.324776482331434e-05, "loss": 0.42650699615478516, "memory(GiB)": 70.5, "step": 79325, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.453008 }, { "epoch": 3.398740413864016, "grad_norm": 1.1662896871566772, "learning_rate": 2.324207959356391e-05, "loss": 0.2527564525604248, "memory(GiB)": 70.5, "step": 79330, "token_acc": 0.9461538461538461, "train_speed(iter/s)": 1.453012 }, { "epoch": 3.3989546291932653, "grad_norm": 4.830506801605225, "learning_rate": 2.323639484855831e-05, "loss": 0.37065250873565675, "memory(GiB)": 70.5, "step": 79335, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.453012 }, { "epoch": 3.399168844522514, "grad_norm": 0.8448554873466492, "learning_rate": 2.3230710588400505e-05, "loss": 0.15156248807907105, "memory(GiB)": 70.5, "step": 79340, "token_acc": 0.9550561797752809, "train_speed(iter/s)": 1.453015 }, { "epoch": 3.399383059851763, "grad_norm": 3.238293409347534, "learning_rate": 2.322502681319349e-05, "loss": 0.474755859375, "memory(GiB)": 70.5, "step": 79345, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.453027 }, { "epoch": 3.399597275181012, "grad_norm": 3.2572808265686035, "learning_rate": 2.321934352304025e-05, "loss": 0.3055521726608276, "memory(GiB)": 70.5, "step": 79350, "token_acc": 0.9297124600638977, "train_speed(iter/s)": 1.453028 }, { "epoch": 3.399811490510261, "grad_norm": 0.11388809233903885, "learning_rate": 2.321366071804373e-05, "loss": 0.32190167903900146, "memory(GiB)": 70.5, "step": 79355, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.453031 }, { "epoch": 3.40002570583951, "grad_norm": 7.1540751457214355, "learning_rate": 2.320797839830686e-05, "loss": 0.5331521034240723, "memory(GiB)": 70.5, "step": 79360, "token_acc": 0.9033613445378151, "train_speed(iter/s)": 1.453033 }, { "epoch": 3.400239921168759, "grad_norm": 0.7892248630523682, "learning_rate": 2.3202296563932607e-05, "loss": 0.6468306541442871, "memory(GiB)": 70.5, "step": 79365, "token_acc": 0.8621908127208481, "train_speed(iter/s)": 1.453035 }, { "epoch": 3.400454136498008, "grad_norm": 1.9559593200683594, "learning_rate": 2.3196615215023886e-05, "loss": 0.18868815898895264, "memory(GiB)": 70.5, "step": 79370, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.453038 }, { "epoch": 3.4006683518272567, "grad_norm": 2.0898451805114746, "learning_rate": 2.3190934351683602e-05, "loss": 0.3451821804046631, "memory(GiB)": 70.5, "step": 79375, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.453042 }, { "epoch": 3.400882567156506, "grad_norm": 2.952850818634033, "learning_rate": 2.3185253974014714e-05, "loss": 0.6411798477172852, "memory(GiB)": 70.5, "step": 79380, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.453041 }, { "epoch": 3.4010967824857548, "grad_norm": 0.7391749024391174, "learning_rate": 2.3179574082120105e-05, "loss": 0.1257768988609314, "memory(GiB)": 70.5, "step": 79385, "token_acc": 0.9692307692307692, "train_speed(iter/s)": 1.453045 }, { "epoch": 3.4013109978150036, "grad_norm": 4.748368263244629, "learning_rate": 2.317389467610267e-05, "loss": 0.4937911033630371, "memory(GiB)": 70.5, "step": 79390, "token_acc": 0.8850931677018633, "train_speed(iter/s)": 1.453049 }, { "epoch": 3.401525213144253, "grad_norm": 3.453850269317627, "learning_rate": 2.3168215756065292e-05, "loss": 0.08712942600250244, "memory(GiB)": 70.5, "step": 79395, "token_acc": 0.9795918367346939, "train_speed(iter/s)": 1.453051 }, { "epoch": 3.4017394284735016, "grad_norm": 4.652786731719971, "learning_rate": 2.3162537322110843e-05, "loss": 0.18933123350143433, "memory(GiB)": 70.5, "step": 79400, "token_acc": 0.9602888086642599, "train_speed(iter/s)": 1.453053 }, { "epoch": 3.4019536438027504, "grad_norm": 3.64693021774292, "learning_rate": 2.3156859374342226e-05, "loss": 0.4901641845703125, "memory(GiB)": 70.5, "step": 79405, "token_acc": 0.8968481375358166, "train_speed(iter/s)": 1.453056 }, { "epoch": 3.4021678591319997, "grad_norm": 4.269679546356201, "learning_rate": 2.315118191286228e-05, "loss": 0.607763147354126, "memory(GiB)": 70.5, "step": 79410, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.453065 }, { "epoch": 3.4023820744612485, "grad_norm": 5.335304260253906, "learning_rate": 2.314550493777386e-05, "loss": 0.43931093215942385, "memory(GiB)": 70.5, "step": 79415, "token_acc": 0.891566265060241, "train_speed(iter/s)": 1.453066 }, { "epoch": 3.4025962897904973, "grad_norm": 5.449860095977783, "learning_rate": 2.313982844917979e-05, "loss": 0.41167821884155276, "memory(GiB)": 70.5, "step": 79420, "token_acc": 0.9202898550724637, "train_speed(iter/s)": 1.45307 }, { "epoch": 3.4028105051197466, "grad_norm": 4.872257709503174, "learning_rate": 2.3134152447182945e-05, "loss": 0.2512674808502197, "memory(GiB)": 70.5, "step": 79425, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.453074 }, { "epoch": 3.4030247204489954, "grad_norm": 1.422683835029602, "learning_rate": 2.3128476931886128e-05, "loss": 0.13867095708847046, "memory(GiB)": 70.5, "step": 79430, "token_acc": 0.962457337883959, "train_speed(iter/s)": 1.453079 }, { "epoch": 3.403238935778244, "grad_norm": 2.011566400527954, "learning_rate": 2.3122801903392145e-05, "loss": 0.25821623802185056, "memory(GiB)": 70.5, "step": 79435, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.453075 }, { "epoch": 3.4034531511074935, "grad_norm": 3.727705717086792, "learning_rate": 2.311712736180383e-05, "loss": 0.36882433891296384, "memory(GiB)": 70.5, "step": 79440, "token_acc": 0.9094488188976378, "train_speed(iter/s)": 1.45308 }, { "epoch": 3.4036673664367423, "grad_norm": 1.514695405960083, "learning_rate": 2.3111453307223978e-05, "loss": 0.4805056095123291, "memory(GiB)": 70.5, "step": 79445, "token_acc": 0.8949579831932774, "train_speed(iter/s)": 1.453084 }, { "epoch": 3.403881581765991, "grad_norm": 2.4059486389160156, "learning_rate": 2.3105779739755368e-05, "loss": 0.2046415090560913, "memory(GiB)": 70.5, "step": 79450, "token_acc": 0.9585987261146497, "train_speed(iter/s)": 1.453096 }, { "epoch": 3.4040957970952403, "grad_norm": 2.5557873249053955, "learning_rate": 2.3100106659500794e-05, "loss": 0.35009589195251467, "memory(GiB)": 70.5, "step": 79455, "token_acc": 0.9240924092409241, "train_speed(iter/s)": 1.453107 }, { "epoch": 3.404310012424489, "grad_norm": 0.8783413171768188, "learning_rate": 2.3094434066562993e-05, "loss": 0.4509144306182861, "memory(GiB)": 70.5, "step": 79460, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.453123 }, { "epoch": 3.404524227753738, "grad_norm": 1.4278956651687622, "learning_rate": 2.3088761961044786e-05, "loss": 0.4167478084564209, "memory(GiB)": 70.5, "step": 79465, "token_acc": 0.907051282051282, "train_speed(iter/s)": 1.453125 }, { "epoch": 3.4047384430829872, "grad_norm": 5.299797534942627, "learning_rate": 2.3083090343048898e-05, "loss": 0.3096369504928589, "memory(GiB)": 70.5, "step": 79470, "token_acc": 0.9363057324840764, "train_speed(iter/s)": 1.453131 }, { "epoch": 3.404952658412236, "grad_norm": 1.6972854137420654, "learning_rate": 2.3077419212678076e-05, "loss": 0.4122015476226807, "memory(GiB)": 70.5, "step": 79475, "token_acc": 0.9080118694362018, "train_speed(iter/s)": 1.45314 }, { "epoch": 3.405166873741485, "grad_norm": 3.555048704147339, "learning_rate": 2.3071748570035063e-05, "loss": 0.2821906805038452, "memory(GiB)": 70.5, "step": 79480, "token_acc": 0.9316546762589928, "train_speed(iter/s)": 1.453146 }, { "epoch": 3.405381089070734, "grad_norm": 1.5091917514801025, "learning_rate": 2.3066078415222563e-05, "loss": 0.3870696544647217, "memory(GiB)": 70.5, "step": 79485, "token_acc": 0.9177631578947368, "train_speed(iter/s)": 1.453146 }, { "epoch": 3.405595304399983, "grad_norm": 5.526209831237793, "learning_rate": 2.306040874834334e-05, "loss": 0.36778979301452636, "memory(GiB)": 70.5, "step": 79490, "token_acc": 0.9207317073170732, "train_speed(iter/s)": 1.453149 }, { "epoch": 3.4058095197292317, "grad_norm": 1.9980103969573975, "learning_rate": 2.3054739569500066e-05, "loss": 0.17416614294052124, "memory(GiB)": 70.5, "step": 79495, "token_acc": 0.9554794520547946, "train_speed(iter/s)": 1.453155 }, { "epoch": 3.406023735058481, "grad_norm": 2.795689582824707, "learning_rate": 2.3049070878795477e-05, "loss": 0.22520799636840821, "memory(GiB)": 70.5, "step": 79500, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.453161 }, { "epoch": 3.406023735058481, "eval_loss": 2.570753574371338, "eval_runtime": 11.4222, "eval_samples_per_second": 8.755, "eval_steps_per_second": 8.755, "eval_token_acc": 0.41228070175438597, "step": 79500 }, { "epoch": 3.40623795038773, "grad_norm": 2.8963935375213623, "learning_rate": 2.3043402676332253e-05, "loss": 0.4535993576049805, "memory(GiB)": 70.5, "step": 79505, "token_acc": 0.529245283018868, "train_speed(iter/s)": 1.452825 }, { "epoch": 3.4064521657169786, "grad_norm": 2.4440386295318604, "learning_rate": 2.303773496221308e-05, "loss": 0.18721747398376465, "memory(GiB)": 70.5, "step": 79510, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.452828 }, { "epoch": 3.406666381046228, "grad_norm": 2.0388219356536865, "learning_rate": 2.3032067736540626e-05, "loss": 0.4263473987579346, "memory(GiB)": 70.5, "step": 79515, "token_acc": 0.9135802469135802, "train_speed(iter/s)": 1.452825 }, { "epoch": 3.4068805963754767, "grad_norm": 4.422087669372559, "learning_rate": 2.3026400999417546e-05, "loss": 0.31681480407714846, "memory(GiB)": 70.5, "step": 79520, "token_acc": 0.9221789883268483, "train_speed(iter/s)": 1.452826 }, { "epoch": 3.4070948117047255, "grad_norm": 6.456435203552246, "learning_rate": 2.3020734750946534e-05, "loss": 0.4679572105407715, "memory(GiB)": 70.5, "step": 79525, "token_acc": 0.9045643153526971, "train_speed(iter/s)": 1.45283 }, { "epoch": 3.4073090270339748, "grad_norm": 4.099552631378174, "learning_rate": 2.3015068991230222e-05, "loss": 0.5541656494140625, "memory(GiB)": 70.5, "step": 79530, "token_acc": 0.8904109589041096, "train_speed(iter/s)": 1.452838 }, { "epoch": 3.4075232423632236, "grad_norm": 2.253422737121582, "learning_rate": 2.3009403720371247e-05, "loss": 0.29284181594848635, "memory(GiB)": 70.5, "step": 79535, "token_acc": 0.9266862170087976, "train_speed(iter/s)": 1.452842 }, { "epoch": 3.4077374576924724, "grad_norm": 0.05013202503323555, "learning_rate": 2.300373893847224e-05, "loss": 0.31674442291259763, "memory(GiB)": 70.5, "step": 79540, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.452843 }, { "epoch": 3.4079516730217216, "grad_norm": 1.4698325395584106, "learning_rate": 2.2998074645635815e-05, "loss": 0.23315911293029784, "memory(GiB)": 70.5, "step": 79545, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.452857 }, { "epoch": 3.4081658883509705, "grad_norm": 4.734478950500488, "learning_rate": 2.299241084196461e-05, "loss": 0.421690034866333, "memory(GiB)": 70.5, "step": 79550, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.452855 }, { "epoch": 3.4083801036802193, "grad_norm": 1.0528535842895508, "learning_rate": 2.2986747527561227e-05, "loss": 0.2023686408996582, "memory(GiB)": 70.5, "step": 79555, "token_acc": 0.9547169811320755, "train_speed(iter/s)": 1.45286 }, { "epoch": 3.4085943190094685, "grad_norm": 2.3297479152679443, "learning_rate": 2.2981084702528244e-05, "loss": 0.34743380546569824, "memory(GiB)": 70.5, "step": 79560, "token_acc": 0.93125, "train_speed(iter/s)": 1.452863 }, { "epoch": 3.4088085343387173, "grad_norm": 3.2138233184814453, "learning_rate": 2.297542236696824e-05, "loss": 0.43732528686523436, "memory(GiB)": 70.5, "step": 79565, "token_acc": 0.8884892086330936, "train_speed(iter/s)": 1.452865 }, { "epoch": 3.409022749667966, "grad_norm": 4.172484874725342, "learning_rate": 2.296976052098383e-05, "loss": 0.4538902282714844, "memory(GiB)": 70.5, "step": 79570, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.452886 }, { "epoch": 3.4092369649972154, "grad_norm": 4.174638271331787, "learning_rate": 2.2964099164677567e-05, "loss": 0.35164480209350585, "memory(GiB)": 70.5, "step": 79575, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.452909 }, { "epoch": 3.409451180326464, "grad_norm": 1.2611064910888672, "learning_rate": 2.295843829815199e-05, "loss": 0.1876812219619751, "memory(GiB)": 70.5, "step": 79580, "token_acc": 0.9739776951672863, "train_speed(iter/s)": 1.452924 }, { "epoch": 3.409665395655713, "grad_norm": 4.5616912841796875, "learning_rate": 2.295277792150969e-05, "loss": 0.3113088130950928, "memory(GiB)": 70.5, "step": 79585, "token_acc": 0.9279661016949152, "train_speed(iter/s)": 1.452933 }, { "epoch": 3.4098796109849623, "grad_norm": 5.586533546447754, "learning_rate": 2.294711803485319e-05, "loss": 0.3423830270767212, "memory(GiB)": 70.5, "step": 79590, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.452937 }, { "epoch": 3.410093826314211, "grad_norm": 1.6541253328323364, "learning_rate": 2.2941458638285017e-05, "loss": 0.13465598821640015, "memory(GiB)": 70.5, "step": 79595, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.452944 }, { "epoch": 3.41030804164346, "grad_norm": 2.7444491386413574, "learning_rate": 2.2935799731907707e-05, "loss": 0.6652851581573487, "memory(GiB)": 70.5, "step": 79600, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.452949 }, { "epoch": 3.410522256972709, "grad_norm": 2.332505702972412, "learning_rate": 2.293014131582376e-05, "loss": 0.40000624656677247, "memory(GiB)": 70.5, "step": 79605, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.45295 }, { "epoch": 3.410736472301958, "grad_norm": 3.119729518890381, "learning_rate": 2.2924483390135716e-05, "loss": 0.16918137073516845, "memory(GiB)": 70.5, "step": 79610, "token_acc": 0.959731543624161, "train_speed(iter/s)": 1.452954 }, { "epoch": 3.410950687631207, "grad_norm": 4.154654026031494, "learning_rate": 2.291882595494605e-05, "loss": 0.3364307165145874, "memory(GiB)": 70.5, "step": 79615, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.452953 }, { "epoch": 3.411164902960456, "grad_norm": 9.815827369689941, "learning_rate": 2.2913169010357256e-05, "loss": 0.24074606895446776, "memory(GiB)": 70.5, "step": 79620, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.452953 }, { "epoch": 3.411379118289705, "grad_norm": 11.885319709777832, "learning_rate": 2.2907512556471817e-05, "loss": 0.15232818126678466, "memory(GiB)": 70.5, "step": 79625, "token_acc": 0.9568627450980393, "train_speed(iter/s)": 1.452953 }, { "epoch": 3.4115933336189537, "grad_norm": 1.1811139583587646, "learning_rate": 2.290185659339218e-05, "loss": 0.5307595252990722, "memory(GiB)": 70.5, "step": 79630, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.452957 }, { "epoch": 3.411807548948203, "grad_norm": 0.3957265317440033, "learning_rate": 2.2896201121220856e-05, "loss": 0.2852344274520874, "memory(GiB)": 70.5, "step": 79635, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.452962 }, { "epoch": 3.4120217642774517, "grad_norm": 4.036625385284424, "learning_rate": 2.289054614006025e-05, "loss": 0.2901793956756592, "memory(GiB)": 70.5, "step": 79640, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.452966 }, { "epoch": 3.4122359796067006, "grad_norm": 0.1589118242263794, "learning_rate": 2.288489165001285e-05, "loss": 0.09322633147239685, "memory(GiB)": 70.5, "step": 79645, "token_acc": 0.9777777777777777, "train_speed(iter/s)": 1.452977 }, { "epoch": 3.41245019493595, "grad_norm": 3.564002275466919, "learning_rate": 2.287923765118108e-05, "loss": 0.34879963397979735, "memory(GiB)": 70.5, "step": 79650, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.452983 }, { "epoch": 3.4126644102651986, "grad_norm": 2.409087657928467, "learning_rate": 2.2873584143667352e-05, "loss": 0.5516327857971192, "memory(GiB)": 70.5, "step": 79655, "token_acc": 0.8987730061349694, "train_speed(iter/s)": 1.452985 }, { "epoch": 3.4128786255944474, "grad_norm": 2.123304843902588, "learning_rate": 2.28679311275741e-05, "loss": 0.45308985710144045, "memory(GiB)": 70.5, "step": 79660, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.452997 }, { "epoch": 3.4130928409236967, "grad_norm": 3.8082451820373535, "learning_rate": 2.2862278603003707e-05, "loss": 0.5164053916931153, "memory(GiB)": 70.5, "step": 79665, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.452994 }, { "epoch": 3.4133070562529455, "grad_norm": 2.4317991733551025, "learning_rate": 2.2856626570058613e-05, "loss": 0.4767920970916748, "memory(GiB)": 70.5, "step": 79670, "token_acc": 0.9028213166144201, "train_speed(iter/s)": 1.452997 }, { "epoch": 3.4135212715821943, "grad_norm": 3.3930013179779053, "learning_rate": 2.2850975028841194e-05, "loss": 0.2460331916809082, "memory(GiB)": 70.5, "step": 79675, "token_acc": 0.9440559440559441, "train_speed(iter/s)": 1.452999 }, { "epoch": 3.4137354869114436, "grad_norm": 5.667302131652832, "learning_rate": 2.2845323979453832e-05, "loss": 0.38921935558319093, "memory(GiB)": 70.5, "step": 79680, "token_acc": 0.9249146757679181, "train_speed(iter/s)": 1.452998 }, { "epoch": 3.4139497022406924, "grad_norm": 2.1068289279937744, "learning_rate": 2.2839673421998892e-05, "loss": 0.30207765102386475, "memory(GiB)": 70.5, "step": 79685, "token_acc": 0.924, "train_speed(iter/s)": 1.452999 }, { "epoch": 3.414163917569941, "grad_norm": 3.1554369926452637, "learning_rate": 2.283402335657873e-05, "loss": 0.35673339366912843, "memory(GiB)": 70.5, "step": 79690, "token_acc": 0.927038626609442, "train_speed(iter/s)": 1.452995 }, { "epoch": 3.4143781328991905, "grad_norm": 3.774549722671509, "learning_rate": 2.2828373783295743e-05, "loss": 0.44616079330444336, "memory(GiB)": 70.5, "step": 79695, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.452997 }, { "epoch": 3.4145923482284393, "grad_norm": 4.1973371505737305, "learning_rate": 2.2822724702252245e-05, "loss": 0.26577417850494384, "memory(GiB)": 70.5, "step": 79700, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.452997 }, { "epoch": 3.414806563557688, "grad_norm": 5.054286956787109, "learning_rate": 2.281707611355059e-05, "loss": 0.3360737800598145, "memory(GiB)": 70.5, "step": 79705, "token_acc": 0.9250814332247557, "train_speed(iter/s)": 1.453012 }, { "epoch": 3.4150207788869373, "grad_norm": 3.9113588333129883, "learning_rate": 2.2811428017293096e-05, "loss": 0.329095721244812, "memory(GiB)": 70.5, "step": 79710, "token_acc": 0.9235474006116208, "train_speed(iter/s)": 1.453024 }, { "epoch": 3.415234994216186, "grad_norm": 5.817533493041992, "learning_rate": 2.2805780413582075e-05, "loss": 0.37959506511688235, "memory(GiB)": 70.5, "step": 79715, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.453024 }, { "epoch": 3.415449209545435, "grad_norm": 3.0334110260009766, "learning_rate": 2.2800133302519866e-05, "loss": 0.6534608840942383, "memory(GiB)": 70.5, "step": 79720, "token_acc": 0.8611111111111112, "train_speed(iter/s)": 1.453027 }, { "epoch": 3.415663424874684, "grad_norm": 1.445177674293518, "learning_rate": 2.2794486684208744e-05, "loss": 0.39212908744812014, "memory(GiB)": 70.5, "step": 79725, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.453034 }, { "epoch": 3.415877640203933, "grad_norm": 4.011203765869141, "learning_rate": 2.278884055875103e-05, "loss": 0.4626944541931152, "memory(GiB)": 70.5, "step": 79730, "token_acc": 0.8955823293172691, "train_speed(iter/s)": 1.453032 }, { "epoch": 3.416091855533182, "grad_norm": 5.161853313446045, "learning_rate": 2.2783194926248996e-05, "loss": 0.464077091217041, "memory(GiB)": 70.5, "step": 79735, "token_acc": 0.8909090909090909, "train_speed(iter/s)": 1.453033 }, { "epoch": 3.416306070862431, "grad_norm": 3.2100367546081543, "learning_rate": 2.2777549786804918e-05, "loss": 0.07216816544532775, "memory(GiB)": 70.5, "step": 79740, "token_acc": 0.9820143884892086, "train_speed(iter/s)": 1.453048 }, { "epoch": 3.41652028619168, "grad_norm": 2.2682416439056396, "learning_rate": 2.2771905140521067e-05, "loss": 0.42665896415710447, "memory(GiB)": 70.5, "step": 79745, "token_acc": 0.9197530864197531, "train_speed(iter/s)": 1.45305 }, { "epoch": 3.4167345015209287, "grad_norm": 2.8817262649536133, "learning_rate": 2.2766260987499677e-05, "loss": 0.3796292066574097, "memory(GiB)": 70.5, "step": 79750, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.45306 }, { "epoch": 3.416948716850178, "grad_norm": 1.6347676515579224, "learning_rate": 2.276061732784303e-05, "loss": 0.49863104820251464, "memory(GiB)": 70.5, "step": 79755, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.453068 }, { "epoch": 3.417162932179427, "grad_norm": 5.2497782707214355, "learning_rate": 2.275497416165335e-05, "loss": 0.4398489475250244, "memory(GiB)": 70.5, "step": 79760, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.45307 }, { "epoch": 3.4173771475086756, "grad_norm": 1.878904938697815, "learning_rate": 2.2749331489032884e-05, "loss": 0.19116381406784058, "memory(GiB)": 70.5, "step": 79765, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.453071 }, { "epoch": 3.417591362837925, "grad_norm": 4.17919397354126, "learning_rate": 2.274368931008383e-05, "loss": 0.49515624046325685, "memory(GiB)": 70.5, "step": 79770, "token_acc": 0.8858131487889274, "train_speed(iter/s)": 1.453087 }, { "epoch": 3.4178055781671737, "grad_norm": 2.3639140129089355, "learning_rate": 2.27380476249084e-05, "loss": 0.263809061050415, "memory(GiB)": 70.5, "step": 79775, "token_acc": 0.9316546762589928, "train_speed(iter/s)": 1.453092 }, { "epoch": 3.4180197934964225, "grad_norm": 4.189029693603516, "learning_rate": 2.2732406433608826e-05, "loss": 0.3072323799133301, "memory(GiB)": 70.5, "step": 79780, "token_acc": 0.9255952380952381, "train_speed(iter/s)": 1.453094 }, { "epoch": 3.4182340088256717, "grad_norm": 4.690891265869141, "learning_rate": 2.2726765736287287e-05, "loss": 0.5670672416687011, "memory(GiB)": 70.5, "step": 79785, "token_acc": 0.8761904761904762, "train_speed(iter/s)": 1.453095 }, { "epoch": 3.4184482241549206, "grad_norm": 2.373112201690674, "learning_rate": 2.2721125533045955e-05, "loss": 0.35882112979888914, "memory(GiB)": 70.5, "step": 79790, "token_acc": 0.9307692307692308, "train_speed(iter/s)": 1.453105 }, { "epoch": 3.4186624394841694, "grad_norm": 2.6309592723846436, "learning_rate": 2.2715485823987043e-05, "loss": 0.45547900199890134, "memory(GiB)": 70.5, "step": 79795, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.453104 }, { "epoch": 3.4188766548134186, "grad_norm": 3.9537243843078613, "learning_rate": 2.27098466092127e-05, "loss": 0.2713521957397461, "memory(GiB)": 70.5, "step": 79800, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.45311 }, { "epoch": 3.4190908701426674, "grad_norm": 1.8863908052444458, "learning_rate": 2.2704207888825086e-05, "loss": 0.21465983390808105, "memory(GiB)": 70.5, "step": 79805, "token_acc": 0.946843853820598, "train_speed(iter/s)": 1.453113 }, { "epoch": 3.4193050854719163, "grad_norm": 0.8265047073364258, "learning_rate": 2.2698569662926333e-05, "loss": 0.2431049346923828, "memory(GiB)": 70.5, "step": 79810, "token_acc": 0.9465648854961832, "train_speed(iter/s)": 1.453113 }, { "epoch": 3.4195193008011655, "grad_norm": 6.12999963760376, "learning_rate": 2.2692931931618622e-05, "loss": 0.41582636833190917, "memory(GiB)": 70.5, "step": 79815, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.453116 }, { "epoch": 3.4197335161304143, "grad_norm": 2.576352119445801, "learning_rate": 2.2687294695004058e-05, "loss": 0.19512515068054198, "memory(GiB)": 70.5, "step": 79820, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.453121 }, { "epoch": 3.419947731459663, "grad_norm": 3.620027542114258, "learning_rate": 2.2681657953184775e-05, "loss": 0.4287726402282715, "memory(GiB)": 70.5, "step": 79825, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.453125 }, { "epoch": 3.4201619467889124, "grad_norm": 4.8779377937316895, "learning_rate": 2.267602170626289e-05, "loss": 0.48172917366027834, "memory(GiB)": 70.5, "step": 79830, "token_acc": 0.8844621513944223, "train_speed(iter/s)": 1.453139 }, { "epoch": 3.420376162118161, "grad_norm": 5.661881446838379, "learning_rate": 2.267038595434048e-05, "loss": 0.23629193305969237, "memory(GiB)": 70.5, "step": 79835, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.453145 }, { "epoch": 3.42059037744741, "grad_norm": 1.5707281827926636, "learning_rate": 2.266475069751969e-05, "loss": 0.12095733880996704, "memory(GiB)": 70.5, "step": 79840, "token_acc": 0.9735099337748344, "train_speed(iter/s)": 1.45315 }, { "epoch": 3.4208045927766593, "grad_norm": 3.683227300643921, "learning_rate": 2.2659115935902576e-05, "loss": 0.2356748104095459, "memory(GiB)": 70.5, "step": 79845, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.453151 }, { "epoch": 3.421018808105908, "grad_norm": 2.1809213161468506, "learning_rate": 2.265348166959123e-05, "loss": 0.47263784408569337, "memory(GiB)": 70.5, "step": 79850, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.453156 }, { "epoch": 3.421233023435157, "grad_norm": 2.4596424102783203, "learning_rate": 2.2647847898687712e-05, "loss": 0.21412291526794433, "memory(GiB)": 70.5, "step": 79855, "token_acc": 0.9448051948051948, "train_speed(iter/s)": 1.453164 }, { "epoch": 3.421447238764406, "grad_norm": 4.551463603973389, "learning_rate": 2.2642214623294073e-05, "loss": 0.43079700469970705, "memory(GiB)": 70.5, "step": 79860, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.453164 }, { "epoch": 3.421661454093655, "grad_norm": 0.18867598474025726, "learning_rate": 2.2636581843512378e-05, "loss": 0.22078940868377686, "memory(GiB)": 70.5, "step": 79865, "token_acc": 0.9550898203592815, "train_speed(iter/s)": 1.453162 }, { "epoch": 3.4218756694229038, "grad_norm": 2.8402061462402344, "learning_rate": 2.2630949559444693e-05, "loss": 0.48152914047241213, "memory(GiB)": 70.5, "step": 79870, "token_acc": 0.9105058365758755, "train_speed(iter/s)": 1.453176 }, { "epoch": 3.422089884752153, "grad_norm": 2.602130174636841, "learning_rate": 2.2625317771193027e-05, "loss": 0.31981887817382815, "memory(GiB)": 70.5, "step": 79875, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.45319 }, { "epoch": 3.422304100081402, "grad_norm": 1.9054824113845825, "learning_rate": 2.2619686478859416e-05, "loss": 0.32015151977539064, "memory(GiB)": 70.5, "step": 79880, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.453207 }, { "epoch": 3.4225183154106507, "grad_norm": 3.023322105407715, "learning_rate": 2.261405568254586e-05, "loss": 0.7845259666442871, "memory(GiB)": 70.5, "step": 79885, "token_acc": 0.8533834586466166, "train_speed(iter/s)": 1.453225 }, { "epoch": 3.4227325307399, "grad_norm": 3.0458953380584717, "learning_rate": 2.2608425382354383e-05, "loss": 0.4896393299102783, "memory(GiB)": 70.5, "step": 79890, "token_acc": 0.8963210702341137, "train_speed(iter/s)": 1.453238 }, { "epoch": 3.4229467460691487, "grad_norm": 5.87508487701416, "learning_rate": 2.2602795578386954e-05, "loss": 0.3195977210998535, "memory(GiB)": 70.5, "step": 79895, "token_acc": 0.9309309309309309, "train_speed(iter/s)": 1.453241 }, { "epoch": 3.4231609613983975, "grad_norm": 2.5789601802825928, "learning_rate": 2.2597166270745603e-05, "loss": 0.1714937210083008, "memory(GiB)": 70.5, "step": 79900, "token_acc": 0.9587155963302753, "train_speed(iter/s)": 1.453247 }, { "epoch": 3.423375176727647, "grad_norm": 5.318449974060059, "learning_rate": 2.2591537459532287e-05, "loss": 0.37885165214538574, "memory(GiB)": 70.5, "step": 79905, "token_acc": 0.926605504587156, "train_speed(iter/s)": 1.453247 }, { "epoch": 3.4235893920568956, "grad_norm": 3.4986443519592285, "learning_rate": 2.258590914484898e-05, "loss": 0.42702765464782716, "memory(GiB)": 70.5, "step": 79910, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.453249 }, { "epoch": 3.4238036073861444, "grad_norm": 3.9583494663238525, "learning_rate": 2.2580281326797653e-05, "loss": 0.5507444381713867, "memory(GiB)": 70.5, "step": 79915, "token_acc": 0.8597560975609756, "train_speed(iter/s)": 1.453259 }, { "epoch": 3.4240178227153937, "grad_norm": 1.7343939542770386, "learning_rate": 2.2574654005480232e-05, "loss": 0.37800087928771975, "memory(GiB)": 70.5, "step": 79920, "token_acc": 0.915129151291513, "train_speed(iter/s)": 1.453277 }, { "epoch": 3.4242320380446425, "grad_norm": 1.6424907445907593, "learning_rate": 2.2569027180998697e-05, "loss": 0.18270022869110109, "memory(GiB)": 70.5, "step": 79925, "token_acc": 0.95703125, "train_speed(iter/s)": 1.453279 }, { "epoch": 3.4244462533738913, "grad_norm": 3.0326859951019287, "learning_rate": 2.2563400853454963e-05, "loss": 0.3633050203323364, "memory(GiB)": 70.5, "step": 79930, "token_acc": 0.9308943089430894, "train_speed(iter/s)": 1.453283 }, { "epoch": 3.4246604687031406, "grad_norm": 2.7008755207061768, "learning_rate": 2.2557775022950948e-05, "loss": 0.4402029991149902, "memory(GiB)": 70.5, "step": 79935, "token_acc": 0.9075342465753424, "train_speed(iter/s)": 1.453283 }, { "epoch": 3.4248746840323894, "grad_norm": 1.0268765687942505, "learning_rate": 2.25521496895886e-05, "loss": 0.23307764530181885, "memory(GiB)": 70.5, "step": 79940, "token_acc": 0.941358024691358, "train_speed(iter/s)": 1.453292 }, { "epoch": 3.425088899361638, "grad_norm": 0.41006579995155334, "learning_rate": 2.2546524853469802e-05, "loss": 0.10655752420425416, "memory(GiB)": 70.5, "step": 79945, "token_acc": 0.9777777777777777, "train_speed(iter/s)": 1.453298 }, { "epoch": 3.4253031146908874, "grad_norm": 1.9244821071624756, "learning_rate": 2.2540900514696466e-05, "loss": 0.2764756202697754, "memory(GiB)": 70.5, "step": 79950, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.453304 }, { "epoch": 3.4255173300201363, "grad_norm": 4.945650100708008, "learning_rate": 2.2535276673370453e-05, "loss": 0.3271637916564941, "memory(GiB)": 70.5, "step": 79955, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.453303 }, { "epoch": 3.425731545349385, "grad_norm": 1.5599305629730225, "learning_rate": 2.2529653329593686e-05, "loss": 0.27095348834991456, "memory(GiB)": 70.5, "step": 79960, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.453304 }, { "epoch": 3.4259457606786343, "grad_norm": 2.027907609939575, "learning_rate": 2.2524030483468024e-05, "loss": 0.21167280673980712, "memory(GiB)": 70.5, "step": 79965, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.453309 }, { "epoch": 3.426159976007883, "grad_norm": 2.070970058441162, "learning_rate": 2.251840813509532e-05, "loss": 0.2758042812347412, "memory(GiB)": 70.5, "step": 79970, "token_acc": 0.939297124600639, "train_speed(iter/s)": 1.453322 }, { "epoch": 3.426374191337132, "grad_norm": 7.873590469360352, "learning_rate": 2.251278628457743e-05, "loss": 0.7399838447570801, "memory(GiB)": 70.5, "step": 79975, "token_acc": 0.8793774319066148, "train_speed(iter/s)": 1.453329 }, { "epoch": 3.426588406666381, "grad_norm": 0.6727350950241089, "learning_rate": 2.2507164932016188e-05, "loss": 0.10313514471054078, "memory(GiB)": 70.5, "step": 79980, "token_acc": 0.9776119402985075, "train_speed(iter/s)": 1.453345 }, { "epoch": 3.42680262199563, "grad_norm": 2.289975881576538, "learning_rate": 2.250154407751346e-05, "loss": 0.21591682434082032, "memory(GiB)": 70.5, "step": 79985, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.453348 }, { "epoch": 3.427016837324879, "grad_norm": 0.6300206184387207, "learning_rate": 2.2495923721171053e-05, "loss": 0.2141669750213623, "memory(GiB)": 70.5, "step": 79990, "token_acc": 0.9498207885304659, "train_speed(iter/s)": 1.453355 }, { "epoch": 3.427231052654128, "grad_norm": 6.205417633056641, "learning_rate": 2.2490303863090793e-05, "loss": 0.3760800838470459, "memory(GiB)": 70.5, "step": 79995, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.453363 }, { "epoch": 3.427445267983377, "grad_norm": 3.5800585746765137, "learning_rate": 2.2484684503374487e-05, "loss": 0.5040995121002197, "memory(GiB)": 70.5, "step": 80000, "token_acc": 0.8942307692307693, "train_speed(iter/s)": 1.453363 }, { "epoch": 3.427445267983377, "eval_loss": 2.3836863040924072, "eval_runtime": 11.3938, "eval_samples_per_second": 8.777, "eval_steps_per_second": 8.777, "eval_token_acc": 0.4594972067039106, "step": 80000 }, { "epoch": 3.4276594833126257, "grad_norm": 2.620495557785034, "learning_rate": 2.2479065642123908e-05, "loss": 0.6588260173797608, "memory(GiB)": 70.5, "step": 80005, "token_acc": 0.5784797630799605, "train_speed(iter/s)": 1.453051 }, { "epoch": 3.427873698641875, "grad_norm": 5.263635158538818, "learning_rate": 2.2473447279440875e-05, "loss": 0.49750442504882814, "memory(GiB)": 70.5, "step": 80010, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.453052 }, { "epoch": 3.428087913971124, "grad_norm": 2.0383169651031494, "learning_rate": 2.246782941542718e-05, "loss": 0.5286773204803467, "memory(GiB)": 70.5, "step": 80015, "token_acc": 0.8873239436619719, "train_speed(iter/s)": 1.453052 }, { "epoch": 3.4283021293003726, "grad_norm": 2.721139669418335, "learning_rate": 2.2462212050184594e-05, "loss": 0.29412221908569336, "memory(GiB)": 70.5, "step": 80020, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.453056 }, { "epoch": 3.428516344629622, "grad_norm": 0.7917177677154541, "learning_rate": 2.245659518381486e-05, "loss": 0.16541765928268432, "memory(GiB)": 70.5, "step": 80025, "token_acc": 0.956081081081081, "train_speed(iter/s)": 1.453057 }, { "epoch": 3.4287305599588707, "grad_norm": 3.3182363510131836, "learning_rate": 2.2450978816419742e-05, "loss": 0.25096745491027833, "memory(GiB)": 70.5, "step": 80030, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.45306 }, { "epoch": 3.4289447752881195, "grad_norm": 2.8694469928741455, "learning_rate": 2.244536294810099e-05, "loss": 0.22281115055084227, "memory(GiB)": 70.5, "step": 80035, "token_acc": 0.9573170731707317, "train_speed(iter/s)": 1.453063 }, { "epoch": 3.4291589906173687, "grad_norm": 4.315533638000488, "learning_rate": 2.2439747578960318e-05, "loss": 0.2405646562576294, "memory(GiB)": 70.5, "step": 80040, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.453065 }, { "epoch": 3.4293732059466175, "grad_norm": 3.410693883895874, "learning_rate": 2.2434132709099487e-05, "loss": 0.36632490158081055, "memory(GiB)": 70.5, "step": 80045, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.453074 }, { "epoch": 3.4295874212758664, "grad_norm": 2.6700870990753174, "learning_rate": 2.24285183386202e-05, "loss": 0.4751246929168701, "memory(GiB)": 70.5, "step": 80050, "token_acc": 0.9093959731543624, "train_speed(iter/s)": 1.453074 }, { "epoch": 3.4298016366051156, "grad_norm": 2.7412891387939453, "learning_rate": 2.2422904467624172e-05, "loss": 0.342789363861084, "memory(GiB)": 70.5, "step": 80055, "token_acc": 0.908695652173913, "train_speed(iter/s)": 1.453078 }, { "epoch": 3.4300158519343644, "grad_norm": 4.799342632293701, "learning_rate": 2.24172910962131e-05, "loss": 0.2690619707107544, "memory(GiB)": 70.5, "step": 80060, "token_acc": 0.9404255319148936, "train_speed(iter/s)": 1.453075 }, { "epoch": 3.4302300672636132, "grad_norm": 5.362198829650879, "learning_rate": 2.241167822448865e-05, "loss": 0.5476122856140136, "memory(GiB)": 70.5, "step": 80065, "token_acc": 0.8880866425992779, "train_speed(iter/s)": 1.45309 }, { "epoch": 3.4304442825928625, "grad_norm": 0.9829422235488892, "learning_rate": 2.2406065852552556e-05, "loss": 0.4863574504852295, "memory(GiB)": 70.5, "step": 80070, "token_acc": 0.9045801526717557, "train_speed(iter/s)": 1.453095 }, { "epoch": 3.4306584979221113, "grad_norm": 3.866316080093384, "learning_rate": 2.240045398050646e-05, "loss": 0.2732864856719971, "memory(GiB)": 70.5, "step": 80075, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.453097 }, { "epoch": 3.43087271325136, "grad_norm": 4.147935390472412, "learning_rate": 2.2394842608452026e-05, "loss": 0.23529500961303712, "memory(GiB)": 70.5, "step": 80080, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.453104 }, { "epoch": 3.4310869285806094, "grad_norm": 2.523693323135376, "learning_rate": 2.2389231736490896e-05, "loss": 0.32791666984558104, "memory(GiB)": 70.5, "step": 80085, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.453107 }, { "epoch": 3.431301143909858, "grad_norm": 4.87439489364624, "learning_rate": 2.2383621364724755e-05, "loss": 0.3238863945007324, "memory(GiB)": 70.5, "step": 80090, "token_acc": 0.934375, "train_speed(iter/s)": 1.453104 }, { "epoch": 3.431515359239107, "grad_norm": 2.920933246612549, "learning_rate": 2.2378011493255225e-05, "loss": 0.30217196941375735, "memory(GiB)": 70.5, "step": 80095, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.453107 }, { "epoch": 3.4317295745683563, "grad_norm": 4.098968029022217, "learning_rate": 2.23724021221839e-05, "loss": 0.31686880588531496, "memory(GiB)": 70.5, "step": 80100, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.45311 }, { "epoch": 3.431943789897605, "grad_norm": 0.09224782139062881, "learning_rate": 2.236679325161245e-05, "loss": 0.19648391008377075, "memory(GiB)": 70.5, "step": 80105, "token_acc": 0.9593220338983051, "train_speed(iter/s)": 1.453111 }, { "epoch": 3.432158005226854, "grad_norm": 2.733078956604004, "learning_rate": 2.2361184881642468e-05, "loss": 0.5052392959594727, "memory(GiB)": 70.5, "step": 80110, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.453112 }, { "epoch": 3.432372220556103, "grad_norm": 4.244035720825195, "learning_rate": 2.2355577012375538e-05, "loss": 0.28272438049316406, "memory(GiB)": 70.5, "step": 80115, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.453116 }, { "epoch": 3.432586435885352, "grad_norm": 4.13874626159668, "learning_rate": 2.234996964391327e-05, "loss": 0.2509037971496582, "memory(GiB)": 70.5, "step": 80120, "token_acc": 0.9440559440559441, "train_speed(iter/s)": 1.453123 }, { "epoch": 3.4328006512146008, "grad_norm": 3.038557767868042, "learning_rate": 2.2344362776357218e-05, "loss": 0.42783164978027344, "memory(GiB)": 70.5, "step": 80125, "token_acc": 0.9184952978056427, "train_speed(iter/s)": 1.453126 }, { "epoch": 3.43301486654385, "grad_norm": 5.627216339111328, "learning_rate": 2.2338756409809004e-05, "loss": 0.33856844902038574, "memory(GiB)": 70.5, "step": 80130, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.45313 }, { "epoch": 3.433229081873099, "grad_norm": 2.2770981788635254, "learning_rate": 2.2333150544370157e-05, "loss": 0.535616397857666, "memory(GiB)": 70.5, "step": 80135, "token_acc": 0.879245283018868, "train_speed(iter/s)": 1.453133 }, { "epoch": 3.4334432972023476, "grad_norm": 3.2379884719848633, "learning_rate": 2.2327545180142247e-05, "loss": 0.39782602787017823, "memory(GiB)": 70.5, "step": 80140, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.453131 }, { "epoch": 3.433657512531597, "grad_norm": 1.2357394695281982, "learning_rate": 2.232194031722681e-05, "loss": 0.34079809188842775, "memory(GiB)": 70.5, "step": 80145, "token_acc": 0.9470198675496688, "train_speed(iter/s)": 1.453148 }, { "epoch": 3.4338717278608457, "grad_norm": 4.112829685211182, "learning_rate": 2.231633595572537e-05, "loss": 0.5077475547790528, "memory(GiB)": 70.5, "step": 80150, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.453162 }, { "epoch": 3.4340859431900945, "grad_norm": 3.3366942405700684, "learning_rate": 2.2310732095739494e-05, "loss": 0.1538116455078125, "memory(GiB)": 70.5, "step": 80155, "token_acc": 0.9695817490494296, "train_speed(iter/s)": 1.453167 }, { "epoch": 3.434300158519344, "grad_norm": 3.821031332015991, "learning_rate": 2.2305128737370662e-05, "loss": 0.2895024299621582, "memory(GiB)": 70.5, "step": 80160, "token_acc": 0.927797833935018, "train_speed(iter/s)": 1.453173 }, { "epoch": 3.4345143738485926, "grad_norm": 6.257387161254883, "learning_rate": 2.2299525880720424e-05, "loss": 0.42113122940063474, "memory(GiB)": 70.5, "step": 80165, "token_acc": 0.8980891719745223, "train_speed(iter/s)": 1.453178 }, { "epoch": 3.4347285891778414, "grad_norm": 1.7211272716522217, "learning_rate": 2.2293923525890258e-05, "loss": 0.274519157409668, "memory(GiB)": 70.5, "step": 80170, "token_acc": 0.9342105263157895, "train_speed(iter/s)": 1.45318 }, { "epoch": 3.4349428045070907, "grad_norm": 7.183842658996582, "learning_rate": 2.2288321672981655e-05, "loss": 0.3720468282699585, "memory(GiB)": 70.5, "step": 80175, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.453182 }, { "epoch": 3.4351570198363395, "grad_norm": 1.9720033407211304, "learning_rate": 2.22827203220961e-05, "loss": 0.273167085647583, "memory(GiB)": 70.5, "step": 80180, "token_acc": 0.9442622950819672, "train_speed(iter/s)": 1.453205 }, { "epoch": 3.4353712351655883, "grad_norm": 2.8251805305480957, "learning_rate": 2.2277119473335058e-05, "loss": 0.41399059295654295, "memory(GiB)": 70.5, "step": 80185, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.45321 }, { "epoch": 3.4355854504948375, "grad_norm": 2.083016872406006, "learning_rate": 2.2271519126800013e-05, "loss": 0.2354429006576538, "memory(GiB)": 70.5, "step": 80190, "token_acc": 0.9453376205787781, "train_speed(iter/s)": 1.453209 }, { "epoch": 3.4357996658240864, "grad_norm": 3.506251335144043, "learning_rate": 2.2265919282592412e-05, "loss": 0.41162781715393065, "memory(GiB)": 70.5, "step": 80195, "token_acc": 0.9208860759493671, "train_speed(iter/s)": 1.453208 }, { "epoch": 3.436013881153335, "grad_norm": 1.9567266702651978, "learning_rate": 2.22603199408137e-05, "loss": 0.24565744400024414, "memory(GiB)": 70.5, "step": 80200, "token_acc": 0.9568627450980393, "train_speed(iter/s)": 1.453211 }, { "epoch": 3.4362280964825844, "grad_norm": 4.601008892059326, "learning_rate": 2.225472110156531e-05, "loss": 0.4772765636444092, "memory(GiB)": 70.5, "step": 80205, "token_acc": 0.8977635782747604, "train_speed(iter/s)": 1.453223 }, { "epoch": 3.4364423118118332, "grad_norm": 3.8507230281829834, "learning_rate": 2.2249122764948665e-05, "loss": 0.3261376142501831, "memory(GiB)": 70.5, "step": 80210, "token_acc": 0.9203821656050956, "train_speed(iter/s)": 1.453235 }, { "epoch": 3.436656527141082, "grad_norm": 4.011268615722656, "learning_rate": 2.2243524931065206e-05, "loss": 0.4175966739654541, "memory(GiB)": 70.5, "step": 80215, "token_acc": 0.91, "train_speed(iter/s)": 1.453248 }, { "epoch": 3.4368707424703313, "grad_norm": 5.301118850708008, "learning_rate": 2.2237927600016334e-05, "loss": 0.5838340759277344, "memory(GiB)": 70.5, "step": 80220, "token_acc": 0.8971428571428571, "train_speed(iter/s)": 1.453259 }, { "epoch": 3.43708495779958, "grad_norm": 2.388693332672119, "learning_rate": 2.223233077190344e-05, "loss": 0.4924635887145996, "memory(GiB)": 70.5, "step": 80225, "token_acc": 0.8758169934640523, "train_speed(iter/s)": 1.453264 }, { "epoch": 3.437299173128829, "grad_norm": 4.34173583984375, "learning_rate": 2.222673444682793e-05, "loss": 0.4740150928497314, "memory(GiB)": 70.5, "step": 80230, "token_acc": 0.8913857677902621, "train_speed(iter/s)": 1.453274 }, { "epoch": 3.437513388458078, "grad_norm": 3.2687060832977295, "learning_rate": 2.2221138624891158e-05, "loss": 0.2650440216064453, "memory(GiB)": 70.5, "step": 80235, "token_acc": 0.940625, "train_speed(iter/s)": 1.45328 }, { "epoch": 3.437727603787327, "grad_norm": 1.9811162948608398, "learning_rate": 2.2215543306194535e-05, "loss": 0.2757431983947754, "memory(GiB)": 70.5, "step": 80240, "token_acc": 0.9573643410852714, "train_speed(iter/s)": 1.453286 }, { "epoch": 3.437941819116576, "grad_norm": 3.1304357051849365, "learning_rate": 2.2209948490839388e-05, "loss": 0.6641081809997559, "memory(GiB)": 70.5, "step": 80245, "token_acc": 0.856140350877193, "train_speed(iter/s)": 1.453295 }, { "epoch": 3.438156034445825, "grad_norm": 2.071887254714966, "learning_rate": 2.2204354178927116e-05, "loss": 0.40590720176696776, "memory(GiB)": 70.5, "step": 80250, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.453293 }, { "epoch": 3.438370249775074, "grad_norm": 5.696352005004883, "learning_rate": 2.2198760370559048e-05, "loss": 0.3541254997253418, "memory(GiB)": 70.5, "step": 80255, "token_acc": 0.9186991869918699, "train_speed(iter/s)": 1.453297 }, { "epoch": 3.4385844651043227, "grad_norm": 1.532353162765503, "learning_rate": 2.2193167065836506e-05, "loss": 0.23120813369750975, "memory(GiB)": 70.5, "step": 80260, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.453302 }, { "epoch": 3.438798680433572, "grad_norm": 3.914254665374756, "learning_rate": 2.218757426486083e-05, "loss": 0.4034672737121582, "memory(GiB)": 70.5, "step": 80265, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.453312 }, { "epoch": 3.4390128957628208, "grad_norm": 1.665992021560669, "learning_rate": 2.2181981967733316e-05, "loss": 0.25437312126159667, "memory(GiB)": 70.5, "step": 80270, "token_acc": 0.9389067524115756, "train_speed(iter/s)": 1.453308 }, { "epoch": 3.4392271110920696, "grad_norm": 2.7504687309265137, "learning_rate": 2.2176390174555316e-05, "loss": 0.2998585939407349, "memory(GiB)": 70.5, "step": 80275, "token_acc": 0.9345794392523364, "train_speed(iter/s)": 1.453316 }, { "epoch": 3.439441326421319, "grad_norm": 2.9342172145843506, "learning_rate": 2.2170798885428113e-05, "loss": 0.2234325408935547, "memory(GiB)": 70.5, "step": 80280, "token_acc": 0.9601226993865031, "train_speed(iter/s)": 1.453318 }, { "epoch": 3.4396555417505676, "grad_norm": 5.644237995147705, "learning_rate": 2.2165208100452982e-05, "loss": 0.38924713134765626, "memory(GiB)": 70.5, "step": 80285, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.453338 }, { "epoch": 3.4398697570798165, "grad_norm": 4.641189098358154, "learning_rate": 2.2159617819731226e-05, "loss": 0.46694126129150393, "memory(GiB)": 70.5, "step": 80290, "token_acc": 0.8932806324110671, "train_speed(iter/s)": 1.453359 }, { "epoch": 3.4400839724090657, "grad_norm": 2.6358399391174316, "learning_rate": 2.215402804336409e-05, "loss": 0.4069619655609131, "memory(GiB)": 70.5, "step": 80295, "token_acc": 0.9025270758122743, "train_speed(iter/s)": 1.45336 }, { "epoch": 3.4402981877383145, "grad_norm": 0.3433573544025421, "learning_rate": 2.214843877145288e-05, "loss": 0.3808490037918091, "memory(GiB)": 70.5, "step": 80300, "token_acc": 0.9120879120879121, "train_speed(iter/s)": 1.453365 }, { "epoch": 3.4405124030675633, "grad_norm": 3.168592929840088, "learning_rate": 2.2142850004098804e-05, "loss": 0.2889427661895752, "memory(GiB)": 70.5, "step": 80305, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.453372 }, { "epoch": 3.4407266183968126, "grad_norm": 3.6631357669830322, "learning_rate": 2.2137261741403154e-05, "loss": 0.4771984577178955, "memory(GiB)": 70.5, "step": 80310, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.453377 }, { "epoch": 3.4409408337260614, "grad_norm": 3.5181753635406494, "learning_rate": 2.213167398346714e-05, "loss": 0.32999966144561765, "memory(GiB)": 70.5, "step": 80315, "token_acc": 0.9294871794871795, "train_speed(iter/s)": 1.45338 }, { "epoch": 3.4411550490553102, "grad_norm": 3.475588083267212, "learning_rate": 2.2126086730391997e-05, "loss": 0.3549479961395264, "memory(GiB)": 70.5, "step": 80320, "token_acc": 0.9176029962546817, "train_speed(iter/s)": 1.453401 }, { "epoch": 3.4413692643845595, "grad_norm": 9.618620872497559, "learning_rate": 2.2120499982278942e-05, "loss": 0.4938480377197266, "memory(GiB)": 70.5, "step": 80325, "token_acc": 0.9316546762589928, "train_speed(iter/s)": 1.45341 }, { "epoch": 3.4415834797138083, "grad_norm": 2.0896005630493164, "learning_rate": 2.2114913739229166e-05, "loss": 0.5153651237487793, "memory(GiB)": 70.5, "step": 80330, "token_acc": 0.888135593220339, "train_speed(iter/s)": 1.453415 }, { "epoch": 3.441797695043057, "grad_norm": 3.652217149734497, "learning_rate": 2.2109328001343898e-05, "loss": 0.6106052875518799, "memory(GiB)": 70.5, "step": 80335, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.453417 }, { "epoch": 3.4420119103723064, "grad_norm": 0.8799142241477966, "learning_rate": 2.2103742768724322e-05, "loss": 0.194494891166687, "memory(GiB)": 70.5, "step": 80340, "token_acc": 0.9546925566343042, "train_speed(iter/s)": 1.453414 }, { "epoch": 3.442226125701555, "grad_norm": 1.86473548412323, "learning_rate": 2.2098158041471617e-05, "loss": 0.37816448211669923, "memory(GiB)": 70.5, "step": 80345, "token_acc": 0.9227467811158798, "train_speed(iter/s)": 1.453425 }, { "epoch": 3.442440341030804, "grad_norm": 3.329303026199341, "learning_rate": 2.2092573819686945e-05, "loss": 0.3359161138534546, "memory(GiB)": 70.5, "step": 80350, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.453428 }, { "epoch": 3.4426545563600532, "grad_norm": 4.435821056365967, "learning_rate": 2.2086990103471468e-05, "loss": 0.42406888008117677, "memory(GiB)": 70.5, "step": 80355, "token_acc": 0.9044117647058824, "train_speed(iter/s)": 1.453438 }, { "epoch": 3.442868771689302, "grad_norm": 3.5517680644989014, "learning_rate": 2.2081406892926366e-05, "loss": 0.8112746238708496, "memory(GiB)": 70.5, "step": 80360, "token_acc": 0.8316831683168316, "train_speed(iter/s)": 1.453456 }, { "epoch": 3.443082987018551, "grad_norm": 2.4735779762268066, "learning_rate": 2.2075824188152762e-05, "loss": 0.2329643726348877, "memory(GiB)": 70.5, "step": 80365, "token_acc": 0.9316770186335404, "train_speed(iter/s)": 1.453472 }, { "epoch": 3.4432972023478, "grad_norm": 2.12776780128479, "learning_rate": 2.2070241989251805e-05, "loss": 0.4707451343536377, "memory(GiB)": 70.5, "step": 80370, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.453476 }, { "epoch": 3.443511417677049, "grad_norm": 6.0229363441467285, "learning_rate": 2.2064660296324607e-05, "loss": 0.2871856212615967, "memory(GiB)": 70.5, "step": 80375, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.453475 }, { "epoch": 3.4437256330062977, "grad_norm": 3.0248255729675293, "learning_rate": 2.205907910947228e-05, "loss": 0.42510509490966797, "memory(GiB)": 70.5, "step": 80380, "token_acc": 0.890625, "train_speed(iter/s)": 1.453479 }, { "epoch": 3.443939848335547, "grad_norm": 3.34686017036438, "learning_rate": 2.2053498428795965e-05, "loss": 0.526761531829834, "memory(GiB)": 70.5, "step": 80385, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.453484 }, { "epoch": 3.444154063664796, "grad_norm": 1.908753752708435, "learning_rate": 2.2047918254396722e-05, "loss": 0.2946446180343628, "memory(GiB)": 70.5, "step": 80390, "token_acc": 0.934640522875817, "train_speed(iter/s)": 1.45349 }, { "epoch": 3.4443682789940446, "grad_norm": 1.347379207611084, "learning_rate": 2.2042338586375676e-05, "loss": 0.4960760593414307, "memory(GiB)": 70.5, "step": 80395, "token_acc": 0.8986928104575164, "train_speed(iter/s)": 1.453495 }, { "epoch": 3.444582494323294, "grad_norm": 4.1317548751831055, "learning_rate": 2.2036759424833898e-05, "loss": 0.22181310653686523, "memory(GiB)": 70.5, "step": 80400, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.453509 }, { "epoch": 3.4447967096525427, "grad_norm": 3.026085376739502, "learning_rate": 2.2031180769872456e-05, "loss": 0.22056725025177001, "memory(GiB)": 70.5, "step": 80405, "token_acc": 0.9541666666666667, "train_speed(iter/s)": 1.453514 }, { "epoch": 3.4450109249817915, "grad_norm": 4.507383823394775, "learning_rate": 2.2025602621592404e-05, "loss": 0.3022098779678345, "memory(GiB)": 70.5, "step": 80410, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.453523 }, { "epoch": 3.4452251403110408, "grad_norm": 2.9686696529388428, "learning_rate": 2.2020024980094788e-05, "loss": 0.29084908962249756, "memory(GiB)": 70.5, "step": 80415, "token_acc": 0.9392857142857143, "train_speed(iter/s)": 1.45353 }, { "epoch": 3.4454393556402896, "grad_norm": 2.3245935440063477, "learning_rate": 2.2014447845480684e-05, "loss": 0.2776059865951538, "memory(GiB)": 70.5, "step": 80420, "token_acc": 0.9375, "train_speed(iter/s)": 1.45354 }, { "epoch": 3.4456535709695384, "grad_norm": 5.672439098358154, "learning_rate": 2.2008871217851114e-05, "loss": 0.6522374629974366, "memory(GiB)": 70.5, "step": 80425, "token_acc": 0.872852233676976, "train_speed(iter/s)": 1.453542 }, { "epoch": 3.4458677862987876, "grad_norm": 3.123239755630493, "learning_rate": 2.2003295097307097e-05, "loss": 0.49366068840026855, "memory(GiB)": 70.5, "step": 80430, "token_acc": 0.8619246861924686, "train_speed(iter/s)": 1.453543 }, { "epoch": 3.4460820016280365, "grad_norm": 5.421292781829834, "learning_rate": 2.1997719483949652e-05, "loss": 0.4109602451324463, "memory(GiB)": 70.5, "step": 80435, "token_acc": 0.8921568627450981, "train_speed(iter/s)": 1.453546 }, { "epoch": 3.4462962169572853, "grad_norm": 0.3337084949016571, "learning_rate": 2.199214437787977e-05, "loss": 0.3564543008804321, "memory(GiB)": 70.5, "step": 80440, "token_acc": 0.916030534351145, "train_speed(iter/s)": 1.45356 }, { "epoch": 3.4465104322865345, "grad_norm": 4.647666931152344, "learning_rate": 2.1986569779198486e-05, "loss": 0.6504140853881836, "memory(GiB)": 70.5, "step": 80445, "token_acc": 0.873015873015873, "train_speed(iter/s)": 1.453582 }, { "epoch": 3.4467246476157833, "grad_norm": 2.1657774448394775, "learning_rate": 2.198099568800676e-05, "loss": 0.24093432426452638, "memory(GiB)": 70.5, "step": 80450, "token_acc": 0.9532374100719424, "train_speed(iter/s)": 1.453584 }, { "epoch": 3.446938862945032, "grad_norm": 1.0916528701782227, "learning_rate": 2.1975422104405576e-05, "loss": 0.11400125026702881, "memory(GiB)": 70.5, "step": 80455, "token_acc": 0.9700374531835206, "train_speed(iter/s)": 1.453585 }, { "epoch": 3.4471530782742814, "grad_norm": 2.776909589767456, "learning_rate": 2.1969849028495915e-05, "loss": 0.4370891094207764, "memory(GiB)": 70.5, "step": 80460, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.453586 }, { "epoch": 3.4473672936035302, "grad_norm": 3.8486440181732178, "learning_rate": 2.196427646037874e-05, "loss": 0.3496169328689575, "memory(GiB)": 70.5, "step": 80465, "token_acc": 0.9187279151943463, "train_speed(iter/s)": 1.453597 }, { "epoch": 3.447581508932779, "grad_norm": 3.0836315155029297, "learning_rate": 2.1958704400154995e-05, "loss": 0.3596588850021362, "memory(GiB)": 70.5, "step": 80470, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.453611 }, { "epoch": 3.4477957242620283, "grad_norm": 5.985394477844238, "learning_rate": 2.195313284792561e-05, "loss": 0.37122421264648436, "memory(GiB)": 70.5, "step": 80475, "token_acc": 0.9343283582089552, "train_speed(iter/s)": 1.453613 }, { "epoch": 3.448009939591277, "grad_norm": 3.009371280670166, "learning_rate": 2.194756180379155e-05, "loss": 0.2347771167755127, "memory(GiB)": 70.5, "step": 80480, "token_acc": 0.9540229885057471, "train_speed(iter/s)": 1.453619 }, { "epoch": 3.448224154920526, "grad_norm": 1.2070766687393188, "learning_rate": 2.1941991267853722e-05, "loss": 0.5108346939086914, "memory(GiB)": 70.5, "step": 80485, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.453634 }, { "epoch": 3.448438370249775, "grad_norm": 2.175841808319092, "learning_rate": 2.193642124021304e-05, "loss": 0.1824570417404175, "memory(GiB)": 70.5, "step": 80490, "token_acc": 0.9570957095709571, "train_speed(iter/s)": 1.45364 }, { "epoch": 3.448652585579024, "grad_norm": 4.224128723144531, "learning_rate": 2.193085172097042e-05, "loss": 0.5333437919616699, "memory(GiB)": 70.5, "step": 80495, "token_acc": 0.8978723404255319, "train_speed(iter/s)": 1.453656 }, { "epoch": 3.448866800908273, "grad_norm": 1.7470221519470215, "learning_rate": 2.192528271022673e-05, "loss": 0.2835529088973999, "memory(GiB)": 70.5, "step": 80500, "token_acc": 0.9441176470588235, "train_speed(iter/s)": 1.453659 }, { "epoch": 3.448866800908273, "eval_loss": 2.3567068576812744, "eval_runtime": 11.4157, "eval_samples_per_second": 8.76, "eval_steps_per_second": 8.76, "eval_token_acc": 0.4474431818181818, "step": 80500 }, { "epoch": 3.449081016237522, "grad_norm": 4.882108688354492, "learning_rate": 2.1919714208082898e-05, "loss": 0.18098353147506713, "memory(GiB)": 70.5, "step": 80505, "token_acc": 0.5895598771750256, "train_speed(iter/s)": 1.453346 }, { "epoch": 3.449295231566771, "grad_norm": 0.275188684463501, "learning_rate": 2.1914146214639782e-05, "loss": 0.3243074417114258, "memory(GiB)": 70.5, "step": 80510, "token_acc": 0.9078341013824884, "train_speed(iter/s)": 1.453349 }, { "epoch": 3.4495094468960197, "grad_norm": 3.6658318042755127, "learning_rate": 2.1908578729998258e-05, "loss": 0.2892108201980591, "memory(GiB)": 70.5, "step": 80515, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.453356 }, { "epoch": 3.449723662225269, "grad_norm": 3.3533222675323486, "learning_rate": 2.1903011754259177e-05, "loss": 0.3163425445556641, "memory(GiB)": 70.5, "step": 80520, "token_acc": 0.9122807017543859, "train_speed(iter/s)": 1.453356 }, { "epoch": 3.4499378775545178, "grad_norm": 0.7878171801567078, "learning_rate": 2.1897445287523384e-05, "loss": 0.18544386625289916, "memory(GiB)": 70.5, "step": 80525, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.453361 }, { "epoch": 3.4501520928837666, "grad_norm": 3.680673360824585, "learning_rate": 2.1891879329891746e-05, "loss": 0.24040665626525878, "memory(GiB)": 70.5, "step": 80530, "token_acc": 0.9506578947368421, "train_speed(iter/s)": 1.453369 }, { "epoch": 3.450366308213016, "grad_norm": 2.296499490737915, "learning_rate": 2.1886313881465063e-05, "loss": 0.5415105819702148, "memory(GiB)": 70.5, "step": 80535, "token_acc": 0.8715277777777778, "train_speed(iter/s)": 1.453375 }, { "epoch": 3.4505805235422646, "grad_norm": 0.9512338638305664, "learning_rate": 2.1880748942344198e-05, "loss": 0.2840474843978882, "memory(GiB)": 70.5, "step": 80540, "token_acc": 0.9359756097560976, "train_speed(iter/s)": 1.453378 }, { "epoch": 3.4507947388715134, "grad_norm": 2.6140811443328857, "learning_rate": 2.1875184512629943e-05, "loss": 0.4585977077484131, "memory(GiB)": 70.5, "step": 80545, "token_acc": 0.8982558139534884, "train_speed(iter/s)": 1.453385 }, { "epoch": 3.4510089542007627, "grad_norm": 6.132997512817383, "learning_rate": 2.18696205924231e-05, "loss": 0.41057634353637695, "memory(GiB)": 70.5, "step": 80550, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.453399 }, { "epoch": 3.4512231695300115, "grad_norm": 1.6368706226348877, "learning_rate": 2.186405718182447e-05, "loss": 0.46308274269104005, "memory(GiB)": 70.5, "step": 80555, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.453408 }, { "epoch": 3.4514373848592603, "grad_norm": 0.9493449330329895, "learning_rate": 2.1858494280934826e-05, "loss": 0.45900940895080566, "memory(GiB)": 70.5, "step": 80560, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.453411 }, { "epoch": 3.4516516001885096, "grad_norm": 0.9176276326179504, "learning_rate": 2.1852931889854966e-05, "loss": 0.10013262033462525, "memory(GiB)": 70.5, "step": 80565, "token_acc": 0.9802371541501976, "train_speed(iter/s)": 1.453416 }, { "epoch": 3.4518658155177584, "grad_norm": 6.7246994972229, "learning_rate": 2.1847370008685657e-05, "loss": 0.5073727130889892, "memory(GiB)": 70.5, "step": 80570, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.453423 }, { "epoch": 3.452080030847007, "grad_norm": 4.034964084625244, "learning_rate": 2.1841808637527645e-05, "loss": 0.6356029033660888, "memory(GiB)": 70.5, "step": 80575, "token_acc": 0.86, "train_speed(iter/s)": 1.453427 }, { "epoch": 3.4522942461762565, "grad_norm": 2.3496925830841064, "learning_rate": 2.1836247776481684e-05, "loss": 0.22360715866088868, "memory(GiB)": 70.5, "step": 80580, "token_acc": 0.9495548961424333, "train_speed(iter/s)": 1.453435 }, { "epoch": 3.4525084615055053, "grad_norm": 0.115418441593647, "learning_rate": 2.1830687425648495e-05, "loss": 0.2870137691497803, "memory(GiB)": 70.5, "step": 80585, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.453454 }, { "epoch": 3.452722676834754, "grad_norm": 2.942112922668457, "learning_rate": 2.182512758512885e-05, "loss": 0.31833348274230955, "memory(GiB)": 70.5, "step": 80590, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.453452 }, { "epoch": 3.4529368921640033, "grad_norm": 6.060787677764893, "learning_rate": 2.181956825502345e-05, "loss": 0.48612537384033205, "memory(GiB)": 70.5, "step": 80595, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.453456 }, { "epoch": 3.453151107493252, "grad_norm": 4.366207122802734, "learning_rate": 2.181400943543298e-05, "loss": 0.4491166114807129, "memory(GiB)": 70.5, "step": 80600, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.45346 }, { "epoch": 3.453365322822501, "grad_norm": 4.692354679107666, "learning_rate": 2.1808451126458196e-05, "loss": 0.5981409072875976, "memory(GiB)": 70.5, "step": 80605, "token_acc": 0.8885542168674698, "train_speed(iter/s)": 1.453462 }, { "epoch": 3.4535795381517502, "grad_norm": 3.5910582542419434, "learning_rate": 2.1802893328199763e-05, "loss": 0.28061695098876954, "memory(GiB)": 70.5, "step": 80610, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.45348 }, { "epoch": 3.453793753480999, "grad_norm": 7.4011759757995605, "learning_rate": 2.1797336040758365e-05, "loss": 0.43217668533325193, "memory(GiB)": 70.5, "step": 80615, "token_acc": 0.9088050314465409, "train_speed(iter/s)": 1.45348 }, { "epoch": 3.454007968810248, "grad_norm": 5.500823020935059, "learning_rate": 2.179177926423467e-05, "loss": 0.3734651327133179, "memory(GiB)": 70.5, "step": 80620, "token_acc": 0.9223300970873787, "train_speed(iter/s)": 1.453486 }, { "epoch": 3.454222184139497, "grad_norm": 4.568827152252197, "learning_rate": 2.1786222998729367e-05, "loss": 0.5885896205902099, "memory(GiB)": 70.5, "step": 80625, "token_acc": 0.8760806916426513, "train_speed(iter/s)": 1.453501 }, { "epoch": 3.454436399468746, "grad_norm": 3.932589292526245, "learning_rate": 2.1780667244343106e-05, "loss": 0.34556097984313966, "memory(GiB)": 70.5, "step": 80630, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.453508 }, { "epoch": 3.4546506147979947, "grad_norm": 0.25241318345069885, "learning_rate": 2.1775112001176522e-05, "loss": 0.3947384595870972, "memory(GiB)": 70.5, "step": 80635, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.453514 }, { "epoch": 3.454864830127244, "grad_norm": 2.6836631298065186, "learning_rate": 2.176955726933027e-05, "loss": 0.3049488067626953, "memory(GiB)": 70.5, "step": 80640, "token_acc": 0.9250814332247557, "train_speed(iter/s)": 1.453511 }, { "epoch": 3.455079045456493, "grad_norm": 6.592354774475098, "learning_rate": 2.1764003048904947e-05, "loss": 0.4316534519195557, "memory(GiB)": 70.5, "step": 80645, "token_acc": 0.9020100502512562, "train_speed(iter/s)": 1.453514 }, { "epoch": 3.4552932607857416, "grad_norm": 2.384061336517334, "learning_rate": 2.1758449340001213e-05, "loss": 0.3353961706161499, "memory(GiB)": 70.5, "step": 80650, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.453511 }, { "epoch": 3.455507476114991, "grad_norm": 5.725975513458252, "learning_rate": 2.175289614271967e-05, "loss": 0.3668787479400635, "memory(GiB)": 70.5, "step": 80655, "token_acc": 0.9404761904761905, "train_speed(iter/s)": 1.453513 }, { "epoch": 3.4557216914442397, "grad_norm": 2.55399751663208, "learning_rate": 2.1747343457160902e-05, "loss": 0.2175389528274536, "memory(GiB)": 70.5, "step": 80660, "token_acc": 0.956989247311828, "train_speed(iter/s)": 1.45353 }, { "epoch": 3.4559359067734885, "grad_norm": 4.223352432250977, "learning_rate": 2.1741791283425516e-05, "loss": 0.4090309143066406, "memory(GiB)": 70.5, "step": 80665, "token_acc": 0.9046052631578947, "train_speed(iter/s)": 1.453541 }, { "epoch": 3.4561501221027378, "grad_norm": 1.1677024364471436, "learning_rate": 2.173623962161407e-05, "loss": 0.4552435398101807, "memory(GiB)": 70.5, "step": 80670, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.453546 }, { "epoch": 3.4563643374319866, "grad_norm": 0.3426019847393036, "learning_rate": 2.1730688471827176e-05, "loss": 0.18311145305633544, "memory(GiB)": 70.5, "step": 80675, "token_acc": 0.9527272727272728, "train_speed(iter/s)": 1.453558 }, { "epoch": 3.4565785527612354, "grad_norm": 3.1140971183776855, "learning_rate": 2.1725137834165356e-05, "loss": 0.2809462547302246, "memory(GiB)": 70.5, "step": 80680, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.45357 }, { "epoch": 3.4567927680904846, "grad_norm": 1.3502174615859985, "learning_rate": 2.171958770872921e-05, "loss": 0.5613816261291504, "memory(GiB)": 70.5, "step": 80685, "token_acc": 0.8789808917197452, "train_speed(iter/s)": 1.453574 }, { "epoch": 3.4570069834197334, "grad_norm": 1.7909374237060547, "learning_rate": 2.171403809561926e-05, "loss": 0.6041789531707764, "memory(GiB)": 70.5, "step": 80690, "token_acc": 0.8701754385964913, "train_speed(iter/s)": 1.453578 }, { "epoch": 3.4572211987489823, "grad_norm": 3.7126011848449707, "learning_rate": 2.1708488994936048e-05, "loss": 0.25853097438812256, "memory(GiB)": 70.5, "step": 80695, "token_acc": 0.9217687074829932, "train_speed(iter/s)": 1.453601 }, { "epoch": 3.4574354140782315, "grad_norm": 3.6645641326904297, "learning_rate": 2.1702940406780097e-05, "loss": 0.4041797161102295, "memory(GiB)": 70.5, "step": 80700, "token_acc": 0.9119718309859155, "train_speed(iter/s)": 1.453601 }, { "epoch": 3.4576496294074803, "grad_norm": 3.547271251678467, "learning_rate": 2.1697392331251904e-05, "loss": 0.3007791996002197, "memory(GiB)": 70.5, "step": 80705, "token_acc": 0.934931506849315, "train_speed(iter/s)": 1.453607 }, { "epoch": 3.457863844736729, "grad_norm": 4.495199680328369, "learning_rate": 2.169184476845202e-05, "loss": 0.26369123458862304, "memory(GiB)": 70.5, "step": 80710, "token_acc": 0.9323076923076923, "train_speed(iter/s)": 1.45362 }, { "epoch": 3.4580780600659784, "grad_norm": 2.249826669692993, "learning_rate": 2.168629771848092e-05, "loss": 0.5223316192626953, "memory(GiB)": 70.5, "step": 80715, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.453636 }, { "epoch": 3.458292275395227, "grad_norm": 4.8733367919921875, "learning_rate": 2.1680751181439096e-05, "loss": 0.5998940467834473, "memory(GiB)": 70.5, "step": 80720, "token_acc": 0.8666666666666667, "train_speed(iter/s)": 1.453636 }, { "epoch": 3.458506490724476, "grad_norm": 1.6934782266616821, "learning_rate": 2.1675205157427035e-05, "loss": 0.17016186714172363, "memory(GiB)": 70.5, "step": 80725, "token_acc": 0.9614147909967846, "train_speed(iter/s)": 1.453644 }, { "epoch": 3.4587207060537253, "grad_norm": 3.668933391571045, "learning_rate": 2.1669659646545183e-05, "loss": 0.5004278659820557, "memory(GiB)": 70.5, "step": 80730, "token_acc": 0.8753894080996885, "train_speed(iter/s)": 1.453658 }, { "epoch": 3.458934921382974, "grad_norm": 2.7239773273468018, "learning_rate": 2.166411464889404e-05, "loss": 0.1802880048751831, "memory(GiB)": 70.5, "step": 80735, "token_acc": 0.9496644295302014, "train_speed(iter/s)": 1.453655 }, { "epoch": 3.459149136712223, "grad_norm": 1.9456429481506348, "learning_rate": 2.1658570164574044e-05, "loss": 0.16006437540054322, "memory(GiB)": 70.5, "step": 80740, "token_acc": 0.9586206896551724, "train_speed(iter/s)": 1.453675 }, { "epoch": 3.459363352041472, "grad_norm": 2.1227986812591553, "learning_rate": 2.1653026193685632e-05, "loss": 0.23038778305053711, "memory(GiB)": 70.5, "step": 80745, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.453686 }, { "epoch": 3.459577567370721, "grad_norm": 1.8659569025039673, "learning_rate": 2.1647482736329227e-05, "loss": 0.3019408702850342, "memory(GiB)": 70.5, "step": 80750, "token_acc": 0.9214876033057852, "train_speed(iter/s)": 1.453692 }, { "epoch": 3.45979178269997, "grad_norm": 2.4263482093811035, "learning_rate": 2.1641939792605287e-05, "loss": 0.28381354808807374, "memory(GiB)": 70.5, "step": 80755, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.453696 }, { "epoch": 3.460005998029219, "grad_norm": 7.534095764160156, "learning_rate": 2.163639736261421e-05, "loss": 0.3248116493225098, "memory(GiB)": 70.5, "step": 80760, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.453697 }, { "epoch": 3.460220213358468, "grad_norm": 5.125756740570068, "learning_rate": 2.1630855446456382e-05, "loss": 0.43561453819274903, "memory(GiB)": 70.5, "step": 80765, "token_acc": 0.9037037037037037, "train_speed(iter/s)": 1.453702 }, { "epoch": 3.4604344286877167, "grad_norm": 2.921372652053833, "learning_rate": 2.1625314044232236e-05, "loss": 0.37186081409454347, "memory(GiB)": 70.5, "step": 80770, "token_acc": 0.9137254901960784, "train_speed(iter/s)": 1.453702 }, { "epoch": 3.460648644016966, "grad_norm": 1.8095355033874512, "learning_rate": 2.161977315604215e-05, "loss": 0.4873004913330078, "memory(GiB)": 70.5, "step": 80775, "token_acc": 0.8941605839416058, "train_speed(iter/s)": 1.453716 }, { "epoch": 3.4608628593462147, "grad_norm": 1.961319088935852, "learning_rate": 2.1614232781986493e-05, "loss": 0.5036880970001221, "memory(GiB)": 70.5, "step": 80780, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.453721 }, { "epoch": 3.4610770746754635, "grad_norm": 4.536710262298584, "learning_rate": 2.1608692922165634e-05, "loss": 0.6592637062072754, "memory(GiB)": 70.5, "step": 80785, "token_acc": 0.8427299703264095, "train_speed(iter/s)": 1.453722 }, { "epoch": 3.461291290004713, "grad_norm": 1.8273299932479858, "learning_rate": 2.1603153576679917e-05, "loss": 0.27084147930145264, "memory(GiB)": 70.5, "step": 80790, "token_acc": 0.9367088607594937, "train_speed(iter/s)": 1.453724 }, { "epoch": 3.4615055053339616, "grad_norm": 1.0511797666549683, "learning_rate": 2.1597614745629734e-05, "loss": 0.2507464408874512, "memory(GiB)": 70.5, "step": 80795, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.453732 }, { "epoch": 3.4617197206632104, "grad_norm": 0.29472965002059937, "learning_rate": 2.1592076429115406e-05, "loss": 0.17510796785354615, "memory(GiB)": 70.5, "step": 80800, "token_acc": 0.9636871508379888, "train_speed(iter/s)": 1.453733 }, { "epoch": 3.4619339359924597, "grad_norm": 1.6357502937316895, "learning_rate": 2.1586538627237257e-05, "loss": 0.4597630500793457, "memory(GiB)": 70.5, "step": 80805, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.45374 }, { "epoch": 3.4621481513217085, "grad_norm": 2.057133436203003, "learning_rate": 2.158100134009562e-05, "loss": 0.3461956262588501, "memory(GiB)": 70.5, "step": 80810, "token_acc": 0.9304347826086956, "train_speed(iter/s)": 1.453738 }, { "epoch": 3.4623623666509573, "grad_norm": 2.565314531326294, "learning_rate": 2.1575464567790778e-05, "loss": 0.6603796482086182, "memory(GiB)": 70.5, "step": 80815, "token_acc": 0.8825301204819277, "train_speed(iter/s)": 1.45375 }, { "epoch": 3.4625765819802066, "grad_norm": 1.3514645099639893, "learning_rate": 2.1569928310423082e-05, "loss": 0.29069950580596926, "memory(GiB)": 70.5, "step": 80820, "token_acc": 0.9446808510638298, "train_speed(iter/s)": 1.453756 }, { "epoch": 3.4627907973094554, "grad_norm": 2.773369312286377, "learning_rate": 2.1564392568092785e-05, "loss": 0.4608603477478027, "memory(GiB)": 70.5, "step": 80825, "token_acc": 0.8847352024922118, "train_speed(iter/s)": 1.453764 }, { "epoch": 3.463005012638704, "grad_norm": 4.537888050079346, "learning_rate": 2.1558857340900207e-05, "loss": 0.34101076126098634, "memory(GiB)": 70.5, "step": 80830, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.453768 }, { "epoch": 3.4632192279679535, "grad_norm": 3.3066799640655518, "learning_rate": 2.1553322628945615e-05, "loss": 0.39176321029663086, "memory(GiB)": 70.5, "step": 80835, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.453777 }, { "epoch": 3.4634334432972023, "grad_norm": 5.504043102264404, "learning_rate": 2.1547788432329263e-05, "loss": 0.28571605682373047, "memory(GiB)": 70.5, "step": 80840, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.453781 }, { "epoch": 3.463647658626451, "grad_norm": 6.3899126052856445, "learning_rate": 2.1542254751151412e-05, "loss": 0.46151556968688967, "memory(GiB)": 70.5, "step": 80845, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.453785 }, { "epoch": 3.4638618739557003, "grad_norm": 3.417525291442871, "learning_rate": 2.1536721585512298e-05, "loss": 0.5499919414520263, "memory(GiB)": 70.5, "step": 80850, "token_acc": 0.8953846153846153, "train_speed(iter/s)": 1.453798 }, { "epoch": 3.464076089284949, "grad_norm": 1.9535330533981323, "learning_rate": 2.153118893551219e-05, "loss": 0.13932641744613647, "memory(GiB)": 70.5, "step": 80855, "token_acc": 0.9657320872274143, "train_speed(iter/s)": 1.453803 }, { "epoch": 3.464290304614198, "grad_norm": 7.464634895324707, "learning_rate": 2.1525656801251304e-05, "loss": 0.5645918846130371, "memory(GiB)": 70.5, "step": 80860, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.453808 }, { "epoch": 3.464504519943447, "grad_norm": 5.171570301055908, "learning_rate": 2.1520125182829852e-05, "loss": 0.35607175827026366, "memory(GiB)": 70.5, "step": 80865, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.453819 }, { "epoch": 3.464718735272696, "grad_norm": 2.1700804233551025, "learning_rate": 2.151459408034805e-05, "loss": 0.40825204849243163, "memory(GiB)": 70.5, "step": 80870, "token_acc": 0.9198717948717948, "train_speed(iter/s)": 1.453821 }, { "epoch": 3.464932950601945, "grad_norm": 1.6185543537139893, "learning_rate": 2.150906349390608e-05, "loss": 0.1627745270729065, "memory(GiB)": 70.5, "step": 80875, "token_acc": 0.9535603715170279, "train_speed(iter/s)": 1.453819 }, { "epoch": 3.465147165931194, "grad_norm": 3.8526575565338135, "learning_rate": 2.1503533423604167e-05, "loss": 0.22167720794677734, "memory(GiB)": 70.5, "step": 80880, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.45384 }, { "epoch": 3.465361381260443, "grad_norm": 8.273395538330078, "learning_rate": 2.1498003869542483e-05, "loss": 0.18465831279754638, "memory(GiB)": 70.5, "step": 80885, "token_acc": 0.96875, "train_speed(iter/s)": 1.453866 }, { "epoch": 3.4655755965896917, "grad_norm": 2.7676279544830322, "learning_rate": 2.149247483182119e-05, "loss": 0.2376502513885498, "memory(GiB)": 70.5, "step": 80890, "token_acc": 0.9493243243243243, "train_speed(iter/s)": 1.45387 }, { "epoch": 3.465789811918941, "grad_norm": 3.2287204265594482, "learning_rate": 2.148694631054046e-05, "loss": 0.3325405836105347, "memory(GiB)": 70.5, "step": 80895, "token_acc": 0.9221789883268483, "train_speed(iter/s)": 1.45387 }, { "epoch": 3.46600402724819, "grad_norm": 1.3376410007476807, "learning_rate": 2.1481418305800425e-05, "loss": 0.5002545356750489, "memory(GiB)": 70.5, "step": 80900, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.453872 }, { "epoch": 3.4662182425774386, "grad_norm": 4.151279449462891, "learning_rate": 2.1475890817701272e-05, "loss": 0.8327085494995117, "memory(GiB)": 70.5, "step": 80905, "token_acc": 0.8267716535433071, "train_speed(iter/s)": 1.453897 }, { "epoch": 3.466432457906688, "grad_norm": 1.775893211364746, "learning_rate": 2.1470363846343096e-05, "loss": 0.3114492416381836, "memory(GiB)": 70.5, "step": 80910, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.453905 }, { "epoch": 3.4666466732359367, "grad_norm": 1.9644277095794678, "learning_rate": 2.146483739182606e-05, "loss": 0.31747725009918215, "memory(GiB)": 70.5, "step": 80915, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.453905 }, { "epoch": 3.4668608885651855, "grad_norm": 1.335756778717041, "learning_rate": 2.1459311454250258e-05, "loss": 0.34374589920043946, "memory(GiB)": 70.5, "step": 80920, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.453901 }, { "epoch": 3.4670751038944347, "grad_norm": 4.107268333435059, "learning_rate": 2.1453786033715807e-05, "loss": 0.37493581771850587, "memory(GiB)": 70.5, "step": 80925, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.453917 }, { "epoch": 3.4672893192236836, "grad_norm": 2.678293228149414, "learning_rate": 2.14482611303228e-05, "loss": 0.5426869869232178, "memory(GiB)": 70.5, "step": 80930, "token_acc": 0.8945454545454545, "train_speed(iter/s)": 1.453917 }, { "epoch": 3.4675035345529324, "grad_norm": 5.260917663574219, "learning_rate": 2.1442736744171305e-05, "loss": 0.49528846740722654, "memory(GiB)": 70.5, "step": 80935, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.453922 }, { "epoch": 3.4677177498821816, "grad_norm": 4.2814459800720215, "learning_rate": 2.1437212875361444e-05, "loss": 0.4579918384552002, "memory(GiB)": 70.5, "step": 80940, "token_acc": 0.9179331306990881, "train_speed(iter/s)": 1.453929 }, { "epoch": 3.4679319652114304, "grad_norm": 3.9924564361572266, "learning_rate": 2.1431689523993263e-05, "loss": 0.23610024452209472, "memory(GiB)": 70.5, "step": 80945, "token_acc": 0.9516129032258065, "train_speed(iter/s)": 1.453935 }, { "epoch": 3.4681461805406792, "grad_norm": 1.6449795961380005, "learning_rate": 2.142616669016683e-05, "loss": 0.38866453170776366, "memory(GiB)": 70.5, "step": 80950, "token_acc": 0.9127906976744186, "train_speed(iter/s)": 1.45394 }, { "epoch": 3.4683603958699285, "grad_norm": 5.52529764175415, "learning_rate": 2.142064437398219e-05, "loss": 0.31546754837036134, "memory(GiB)": 70.5, "step": 80955, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.453958 }, { "epoch": 3.4685746111991773, "grad_norm": 4.8324480056762695, "learning_rate": 2.141512257553937e-05, "loss": 0.2835926294326782, "memory(GiB)": 70.5, "step": 80960, "token_acc": 0.9292035398230089, "train_speed(iter/s)": 1.453966 }, { "epoch": 3.468788826528426, "grad_norm": 3.8626418113708496, "learning_rate": 2.1409601294938436e-05, "loss": 0.37064156532287595, "memory(GiB)": 70.5, "step": 80965, "token_acc": 0.9064516129032258, "train_speed(iter/s)": 1.453968 }, { "epoch": 3.4690030418576754, "grad_norm": 3.9215714931488037, "learning_rate": 2.1404080532279376e-05, "loss": 0.5761750698089599, "memory(GiB)": 70.5, "step": 80970, "token_acc": 0.8613138686131386, "train_speed(iter/s)": 1.453976 }, { "epoch": 3.469217257186924, "grad_norm": 1.2037063837051392, "learning_rate": 2.139856028766224e-05, "loss": 0.11285953521728516, "memory(GiB)": 70.5, "step": 80975, "token_acc": 0.9806949806949807, "train_speed(iter/s)": 1.453978 }, { "epoch": 3.469431472516173, "grad_norm": 4.493349552154541, "learning_rate": 2.139304056118701e-05, "loss": 0.5560451030731202, "memory(GiB)": 70.5, "step": 80980, "token_acc": 0.888135593220339, "train_speed(iter/s)": 1.453992 }, { "epoch": 3.4696456878454223, "grad_norm": 0.9983137845993042, "learning_rate": 2.1387521352953695e-05, "loss": 0.24523017406463624, "memory(GiB)": 70.5, "step": 80985, "token_acc": 0.9501779359430605, "train_speed(iter/s)": 1.454002 }, { "epoch": 3.469859903174671, "grad_norm": 3.3983418941497803, "learning_rate": 2.138200266306227e-05, "loss": 0.35065150260925293, "memory(GiB)": 70.5, "step": 80990, "token_acc": 0.9567099567099567, "train_speed(iter/s)": 1.454006 }, { "epoch": 3.47007411850392, "grad_norm": 2.3701138496398926, "learning_rate": 2.1376484491612685e-05, "loss": 0.3071516275405884, "memory(GiB)": 70.5, "step": 80995, "token_acc": 0.9446254071661238, "train_speed(iter/s)": 1.45401 }, { "epoch": 3.470288333833169, "grad_norm": 3.835613250732422, "learning_rate": 2.1370966838704963e-05, "loss": 0.25717217922210694, "memory(GiB)": 70.5, "step": 81000, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.454009 }, { "epoch": 3.470288333833169, "eval_loss": 2.4831223487854004, "eval_runtime": 11.7139, "eval_samples_per_second": 8.537, "eval_steps_per_second": 8.537, "eval_token_acc": 0.4863013698630137, "step": 81000 }, { "epoch": 3.470502549162418, "grad_norm": 5.53778600692749, "learning_rate": 2.1365449704439028e-05, "loss": 0.31853384971618653, "memory(GiB)": 70.5, "step": 81005, "token_acc": 0.6200578592092575, "train_speed(iter/s)": 1.453672 }, { "epoch": 3.470716764491667, "grad_norm": 5.190446376800537, "learning_rate": 2.1359933088914836e-05, "loss": 0.2967466592788696, "memory(GiB)": 70.5, "step": 81010, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.453676 }, { "epoch": 3.470930979820916, "grad_norm": 3.859654188156128, "learning_rate": 2.1354416992232318e-05, "loss": 0.3033421993255615, "memory(GiB)": 70.5, "step": 81015, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.453681 }, { "epoch": 3.471145195150165, "grad_norm": 1.6542445421218872, "learning_rate": 2.1348901414491396e-05, "loss": 0.3388563871383667, "memory(GiB)": 70.5, "step": 81020, "token_acc": 0.9218106995884774, "train_speed(iter/s)": 1.453695 }, { "epoch": 3.471359410479414, "grad_norm": 3.2318334579467773, "learning_rate": 2.1343386355792017e-05, "loss": 0.32026913166046145, "memory(GiB)": 70.5, "step": 81025, "token_acc": 0.9287925696594427, "train_speed(iter/s)": 1.453695 }, { "epoch": 3.471573625808663, "grad_norm": 0.7787281274795532, "learning_rate": 2.1337871816234072e-05, "loss": 0.15349316596984863, "memory(GiB)": 70.5, "step": 81030, "token_acc": 0.9750889679715302, "train_speed(iter/s)": 1.453696 }, { "epoch": 3.4717878411379117, "grad_norm": 3.114506244659424, "learning_rate": 2.1332357795917475e-05, "loss": 0.28525662422180176, "memory(GiB)": 70.5, "step": 81035, "token_acc": 0.9271255060728745, "train_speed(iter/s)": 1.4537 }, { "epoch": 3.472002056467161, "grad_norm": 3.9250550270080566, "learning_rate": 2.13268442949421e-05, "loss": 0.4411940097808838, "memory(GiB)": 70.5, "step": 81040, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.453706 }, { "epoch": 3.47221627179641, "grad_norm": 7.268063068389893, "learning_rate": 2.132133131340782e-05, "loss": 0.5594324588775634, "memory(GiB)": 70.5, "step": 81045, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.453713 }, { "epoch": 3.4724304871256586, "grad_norm": 3.2087178230285645, "learning_rate": 2.1315818851414553e-05, "loss": 0.5189381599426269, "memory(GiB)": 70.5, "step": 81050, "token_acc": 0.8881118881118881, "train_speed(iter/s)": 1.45373 }, { "epoch": 3.472644702454908, "grad_norm": 4.606488227844238, "learning_rate": 2.131030690906211e-05, "loss": 0.3513856649398804, "memory(GiB)": 70.5, "step": 81055, "token_acc": 0.9022082018927445, "train_speed(iter/s)": 1.453741 }, { "epoch": 3.4728589177841567, "grad_norm": 2.116501808166504, "learning_rate": 2.1304795486450398e-05, "loss": 0.46817879676818847, "memory(GiB)": 70.5, "step": 81060, "token_acc": 0.9003215434083601, "train_speed(iter/s)": 1.453745 }, { "epoch": 3.4730731331134055, "grad_norm": 5.836980819702148, "learning_rate": 2.1299284583679224e-05, "loss": 0.6184920310974121, "memory(GiB)": 70.5, "step": 81065, "token_acc": 0.8661202185792349, "train_speed(iter/s)": 1.453759 }, { "epoch": 3.4732873484426547, "grad_norm": 2.9514963626861572, "learning_rate": 2.1293774200848437e-05, "loss": 0.44359679222106935, "memory(GiB)": 70.5, "step": 81070, "token_acc": 0.8950819672131147, "train_speed(iter/s)": 1.45376 }, { "epoch": 3.4735015637719036, "grad_norm": 3.8578906059265137, "learning_rate": 2.1288264338057855e-05, "loss": 0.3494542598724365, "memory(GiB)": 70.5, "step": 81075, "token_acc": 0.9186991869918699, "train_speed(iter/s)": 1.453772 }, { "epoch": 3.4737157791011524, "grad_norm": 1.9940303564071655, "learning_rate": 2.1282754995407285e-05, "loss": 0.347849440574646, "memory(GiB)": 70.5, "step": 81080, "token_acc": 0.9156118143459916, "train_speed(iter/s)": 1.453778 }, { "epoch": 3.4739299944304016, "grad_norm": 5.591125965118408, "learning_rate": 2.1277246172996562e-05, "loss": 0.4578393459320068, "memory(GiB)": 70.5, "step": 81085, "token_acc": 0.9009287925696594, "train_speed(iter/s)": 1.453792 }, { "epoch": 3.4741442097596504, "grad_norm": 5.541318893432617, "learning_rate": 2.1271737870925473e-05, "loss": 0.5238632202148438, "memory(GiB)": 70.5, "step": 81090, "token_acc": 0.8696969696969697, "train_speed(iter/s)": 1.453814 }, { "epoch": 3.4743584250888992, "grad_norm": 3.104017734527588, "learning_rate": 2.12662300892938e-05, "loss": 0.55813570022583, "memory(GiB)": 70.5, "step": 81095, "token_acc": 0.8862876254180602, "train_speed(iter/s)": 1.453817 }, { "epoch": 3.4745726404181485, "grad_norm": 5.101730823516846, "learning_rate": 2.1260722828201323e-05, "loss": 0.4531412124633789, "memory(GiB)": 70.5, "step": 81100, "token_acc": 0.9094827586206896, "train_speed(iter/s)": 1.453817 }, { "epoch": 3.4747868557473973, "grad_norm": 1.830749750137329, "learning_rate": 2.1255216087747792e-05, "loss": 0.10131175518035888, "memory(GiB)": 70.5, "step": 81105, "token_acc": 0.9678362573099415, "train_speed(iter/s)": 1.453815 }, { "epoch": 3.475001071076646, "grad_norm": 3.3344974517822266, "learning_rate": 2.1249709868033007e-05, "loss": 0.3339547634124756, "memory(GiB)": 70.5, "step": 81110, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.453818 }, { "epoch": 3.4752152864058954, "grad_norm": 0.5239583253860474, "learning_rate": 2.1244204169156694e-05, "loss": 0.5347932338714599, "memory(GiB)": 70.5, "step": 81115, "token_acc": 0.8934707903780069, "train_speed(iter/s)": 1.453824 }, { "epoch": 3.475429501735144, "grad_norm": 6.5349626541137695, "learning_rate": 2.123869899121858e-05, "loss": 0.2270221710205078, "memory(GiB)": 70.5, "step": 81120, "token_acc": 0.9392857142857143, "train_speed(iter/s)": 1.453826 }, { "epoch": 3.475643717064393, "grad_norm": 3.499279499053955, "learning_rate": 2.1233194334318435e-05, "loss": 0.44556636810302735, "memory(GiB)": 70.5, "step": 81125, "token_acc": 0.9196141479099679, "train_speed(iter/s)": 1.453827 }, { "epoch": 3.4758579323936423, "grad_norm": 2.7478461265563965, "learning_rate": 2.122769019855596e-05, "loss": 0.5206759452819825, "memory(GiB)": 70.5, "step": 81130, "token_acc": 0.9036144578313253, "train_speed(iter/s)": 1.453832 }, { "epoch": 3.476072147722891, "grad_norm": 2.3472537994384766, "learning_rate": 2.1222186584030867e-05, "loss": 0.20355396270751952, "memory(GiB)": 70.5, "step": 81135, "token_acc": 0.9609375, "train_speed(iter/s)": 1.453831 }, { "epoch": 3.47628636305214, "grad_norm": 1.0713778734207153, "learning_rate": 2.1216683490842843e-05, "loss": 0.41494183540344237, "memory(GiB)": 70.5, "step": 81140, "token_acc": 0.9120234604105572, "train_speed(iter/s)": 1.453837 }, { "epoch": 3.476500578381389, "grad_norm": 4.83389139175415, "learning_rate": 2.1211180919091615e-05, "loss": 0.34470138549804685, "memory(GiB)": 70.5, "step": 81145, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.45384 }, { "epoch": 3.476714793710638, "grad_norm": 1.0492347478866577, "learning_rate": 2.120567886887685e-05, "loss": 0.29687180519104006, "memory(GiB)": 70.5, "step": 81150, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.453842 }, { "epoch": 3.4769290090398868, "grad_norm": 2.8925788402557373, "learning_rate": 2.1200177340298222e-05, "loss": 0.3381908893585205, "memory(GiB)": 70.5, "step": 81155, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.45385 }, { "epoch": 3.477143224369136, "grad_norm": 2.296764850616455, "learning_rate": 2.1194676333455404e-05, "loss": 0.24847793579101562, "memory(GiB)": 70.5, "step": 81160, "token_acc": 0.9630996309963099, "train_speed(iter/s)": 1.453855 }, { "epoch": 3.477357439698385, "grad_norm": 3.3911404609680176, "learning_rate": 2.1189175848448023e-05, "loss": 0.226577091217041, "memory(GiB)": 70.5, "step": 81165, "token_acc": 0.9466192170818505, "train_speed(iter/s)": 1.453851 }, { "epoch": 3.4775716550276337, "grad_norm": 5.955121994018555, "learning_rate": 2.1183675885375764e-05, "loss": 0.5482751846313476, "memory(GiB)": 70.5, "step": 81170, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.45386 }, { "epoch": 3.477785870356883, "grad_norm": 2.2076492309570312, "learning_rate": 2.1178176444338255e-05, "loss": 0.41240906715393066, "memory(GiB)": 70.5, "step": 81175, "token_acc": 0.9067055393586005, "train_speed(iter/s)": 1.453866 }, { "epoch": 3.4780000856861317, "grad_norm": 5.904462814331055, "learning_rate": 2.1172677525435115e-05, "loss": 0.2816878318786621, "memory(GiB)": 70.5, "step": 81180, "token_acc": 0.933933933933934, "train_speed(iter/s)": 1.453871 }, { "epoch": 3.4782143010153805, "grad_norm": 1.8302979469299316, "learning_rate": 2.1167179128765958e-05, "loss": 0.4102791309356689, "memory(GiB)": 70.5, "step": 81185, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.453888 }, { "epoch": 3.47842851634463, "grad_norm": 5.340857982635498, "learning_rate": 2.1161681254430387e-05, "loss": 0.3004926681518555, "memory(GiB)": 70.5, "step": 81190, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.453891 }, { "epoch": 3.4786427316738786, "grad_norm": 1.7267982959747314, "learning_rate": 2.1156183902528024e-05, "loss": 0.16555048227310182, "memory(GiB)": 70.5, "step": 81195, "token_acc": 0.9561752988047809, "train_speed(iter/s)": 1.453901 }, { "epoch": 3.4788569470031274, "grad_norm": 2.9400787353515625, "learning_rate": 2.1150687073158437e-05, "loss": 0.2953826427459717, "memory(GiB)": 70.5, "step": 81200, "token_acc": 0.9331210191082803, "train_speed(iter/s)": 1.453916 }, { "epoch": 3.4790711623323767, "grad_norm": 0.2162265181541443, "learning_rate": 2.114519076642123e-05, "loss": 0.5257091045379638, "memory(GiB)": 70.5, "step": 81205, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.453926 }, { "epoch": 3.4792853776616255, "grad_norm": 5.7889862060546875, "learning_rate": 2.1139694982415965e-05, "loss": 0.4471411228179932, "memory(GiB)": 70.5, "step": 81210, "token_acc": 0.8932038834951457, "train_speed(iter/s)": 1.453928 }, { "epoch": 3.4794995929908743, "grad_norm": 3.475250482559204, "learning_rate": 2.1134199721242194e-05, "loss": 0.355495810508728, "memory(GiB)": 70.5, "step": 81215, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.453929 }, { "epoch": 3.4797138083201236, "grad_norm": 1.4599528312683105, "learning_rate": 2.112870498299947e-05, "loss": 0.3235763072967529, "memory(GiB)": 70.5, "step": 81220, "token_acc": 0.9359756097560976, "train_speed(iter/s)": 1.453933 }, { "epoch": 3.4799280236493724, "grad_norm": 3.494663715362549, "learning_rate": 2.112321076778733e-05, "loss": 0.44307880401611327, "memory(GiB)": 70.5, "step": 81225, "token_acc": 0.9084507042253521, "train_speed(iter/s)": 1.453956 }, { "epoch": 3.480142238978621, "grad_norm": 5.115729808807373, "learning_rate": 2.1117717075705324e-05, "loss": 0.20812859535217285, "memory(GiB)": 70.5, "step": 81230, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.453964 }, { "epoch": 3.4803564543078704, "grad_norm": 4.420717239379883, "learning_rate": 2.1112223906852973e-05, "loss": 0.36073110103607176, "memory(GiB)": 70.5, "step": 81235, "token_acc": 0.9316239316239316, "train_speed(iter/s)": 1.45397 }, { "epoch": 3.4805706696371193, "grad_norm": 2.7155182361602783, "learning_rate": 2.1106731261329783e-05, "loss": 0.24417662620544434, "memory(GiB)": 70.5, "step": 81240, "token_acc": 0.9625468164794008, "train_speed(iter/s)": 1.453977 }, { "epoch": 3.480784884966368, "grad_norm": 4.406661033630371, "learning_rate": 2.1101239139235263e-05, "loss": 0.4428106307983398, "memory(GiB)": 70.5, "step": 81245, "token_acc": 0.9045801526717557, "train_speed(iter/s)": 1.453973 }, { "epoch": 3.4809991002956173, "grad_norm": 3.352377414703369, "learning_rate": 2.109574754066888e-05, "loss": 0.5660656929016114, "memory(GiB)": 70.5, "step": 81250, "token_acc": 0.8794788273615635, "train_speed(iter/s)": 1.453978 }, { "epoch": 3.481213315624866, "grad_norm": 0.22840501368045807, "learning_rate": 2.109025646573017e-05, "loss": 0.2955115079879761, "memory(GiB)": 70.5, "step": 81255, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.453987 }, { "epoch": 3.481427530954115, "grad_norm": 4.173431873321533, "learning_rate": 2.108476591451858e-05, "loss": 0.2445124864578247, "memory(GiB)": 70.5, "step": 81260, "token_acc": 0.9531772575250836, "train_speed(iter/s)": 1.453987 }, { "epoch": 3.481641746283364, "grad_norm": 2.138188362121582, "learning_rate": 2.1079275887133565e-05, "loss": 0.290667986869812, "memory(GiB)": 70.5, "step": 81265, "token_acc": 0.9281437125748503, "train_speed(iter/s)": 1.453991 }, { "epoch": 3.481855961612613, "grad_norm": 1.639840841293335, "learning_rate": 2.107378638367461e-05, "loss": 0.3683326244354248, "memory(GiB)": 73.83, "step": 81270, "token_acc": 0.9295774647887324, "train_speed(iter/s)": 1.453985 }, { "epoch": 3.482070176941862, "grad_norm": 3.351896286010742, "learning_rate": 2.1068297404241156e-05, "loss": 0.3359540462493896, "memory(GiB)": 73.83, "step": 81275, "token_acc": 0.9387096774193548, "train_speed(iter/s)": 1.453987 }, { "epoch": 3.482284392271111, "grad_norm": 2.543724536895752, "learning_rate": 2.1062808948932633e-05, "loss": 0.554437255859375, "memory(GiB)": 73.83, "step": 81280, "token_acc": 0.884514435695538, "train_speed(iter/s)": 1.453996 }, { "epoch": 3.48249860760036, "grad_norm": 7.193424224853516, "learning_rate": 2.105732101784845e-05, "loss": 0.5075149536132812, "memory(GiB)": 73.83, "step": 81285, "token_acc": 0.875, "train_speed(iter/s)": 1.453998 }, { "epoch": 3.4827128229296087, "grad_norm": 0.8996047973632812, "learning_rate": 2.105183361108807e-05, "loss": 0.30447514057159425, "memory(GiB)": 73.83, "step": 81290, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.453997 }, { "epoch": 3.482927038258858, "grad_norm": 1.9967526197433472, "learning_rate": 2.1046346728750872e-05, "loss": 0.4100306034088135, "memory(GiB)": 73.83, "step": 81295, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.454004 }, { "epoch": 3.483141253588107, "grad_norm": 6.192001819610596, "learning_rate": 2.104086037093627e-05, "loss": 0.42238407135009765, "memory(GiB)": 73.83, "step": 81300, "token_acc": 0.909967845659164, "train_speed(iter/s)": 1.454017 }, { "epoch": 3.4833554689173556, "grad_norm": 2.7722439765930176, "learning_rate": 2.1035374537743645e-05, "loss": 0.47970876693725584, "memory(GiB)": 73.83, "step": 81305, "token_acc": 0.9026217228464419, "train_speed(iter/s)": 1.45403 }, { "epoch": 3.483569684246605, "grad_norm": 2.9687280654907227, "learning_rate": 2.1029889229272364e-05, "loss": 0.17924811840057372, "memory(GiB)": 73.83, "step": 81310, "token_acc": 0.9530685920577617, "train_speed(iter/s)": 1.454034 }, { "epoch": 3.4837838995758537, "grad_norm": 3.3324544429779053, "learning_rate": 2.1024404445621837e-05, "loss": 0.4005227565765381, "memory(GiB)": 73.83, "step": 81315, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.454043 }, { "epoch": 3.4839981149051025, "grad_norm": 2.951585054397583, "learning_rate": 2.1018920186891407e-05, "loss": 0.3267778158187866, "memory(GiB)": 73.83, "step": 81320, "token_acc": 0.9205776173285198, "train_speed(iter/s)": 1.454057 }, { "epoch": 3.4842123302343517, "grad_norm": 3.4580771923065186, "learning_rate": 2.1013436453180417e-05, "loss": 0.22062366008758544, "memory(GiB)": 73.83, "step": 81325, "token_acc": 0.9484848484848485, "train_speed(iter/s)": 1.45407 }, { "epoch": 3.4844265455636005, "grad_norm": 0.526546061038971, "learning_rate": 2.1007953244588218e-05, "loss": 0.08401111364364625, "memory(GiB)": 73.83, "step": 81330, "token_acc": 0.975975975975976, "train_speed(iter/s)": 1.454072 }, { "epoch": 3.4846407608928494, "grad_norm": 8.579249382019043, "learning_rate": 2.1002470561214126e-05, "loss": 0.32668142318725585, "memory(GiB)": 73.83, "step": 81335, "token_acc": 0.9240924092409241, "train_speed(iter/s)": 1.454071 }, { "epoch": 3.4848549762220986, "grad_norm": 2.8653552532196045, "learning_rate": 2.09969884031575e-05, "loss": 0.5591494560241699, "memory(GiB)": 73.83, "step": 81340, "token_acc": 0.8855218855218855, "train_speed(iter/s)": 1.454083 }, { "epoch": 3.4850691915513474, "grad_norm": 0.13778500258922577, "learning_rate": 2.0991506770517622e-05, "loss": 0.25391278266906736, "memory(GiB)": 73.83, "step": 81345, "token_acc": 0.947945205479452, "train_speed(iter/s)": 1.454095 }, { "epoch": 3.4852834068805962, "grad_norm": 3.3516197204589844, "learning_rate": 2.0986025663393823e-05, "loss": 0.2920116662979126, "memory(GiB)": 73.83, "step": 81350, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.454096 }, { "epoch": 3.4854976222098455, "grad_norm": 2.2969586849212646, "learning_rate": 2.098054508188539e-05, "loss": 0.26838092803955077, "memory(GiB)": 73.83, "step": 81355, "token_acc": 0.9453924914675768, "train_speed(iter/s)": 1.454098 }, { "epoch": 3.4857118375390943, "grad_norm": 2.619990110397339, "learning_rate": 2.0975065026091606e-05, "loss": 0.30835537910461425, "memory(GiB)": 73.83, "step": 81360, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.454107 }, { "epoch": 3.485926052868343, "grad_norm": 5.119794845581055, "learning_rate": 2.0969585496111743e-05, "loss": 0.5656827926635742, "memory(GiB)": 73.83, "step": 81365, "token_acc": 0.8766666666666667, "train_speed(iter/s)": 1.454113 }, { "epoch": 3.4861402681975924, "grad_norm": 3.841442346572876, "learning_rate": 2.0964106492045055e-05, "loss": 0.2703425168991089, "memory(GiB)": 73.83, "step": 81370, "token_acc": 0.946969696969697, "train_speed(iter/s)": 1.454126 }, { "epoch": 3.486354483526841, "grad_norm": 5.267829418182373, "learning_rate": 2.0958628013990832e-05, "loss": 0.34229331016540526, "memory(GiB)": 73.83, "step": 81375, "token_acc": 0.9235127478753541, "train_speed(iter/s)": 1.454125 }, { "epoch": 3.48656869885609, "grad_norm": 2.1666481494903564, "learning_rate": 2.095315006204831e-05, "loss": 0.6334119319915772, "memory(GiB)": 73.83, "step": 81380, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.454141 }, { "epoch": 3.4867829141853393, "grad_norm": 1.9954233169555664, "learning_rate": 2.0947672636316723e-05, "loss": 0.1607840895652771, "memory(GiB)": 73.83, "step": 81385, "token_acc": 0.9625468164794008, "train_speed(iter/s)": 1.454143 }, { "epoch": 3.486997129514588, "grad_norm": 5.530221939086914, "learning_rate": 2.0942195736895293e-05, "loss": 0.21646676063537598, "memory(GiB)": 73.83, "step": 81390, "token_acc": 0.9704918032786886, "train_speed(iter/s)": 1.454139 }, { "epoch": 3.487211344843837, "grad_norm": 2.327008008956909, "learning_rate": 2.0936719363883234e-05, "loss": 0.24822566509246827, "memory(GiB)": 73.83, "step": 81395, "token_acc": 0.9338842975206612, "train_speed(iter/s)": 1.45414 }, { "epoch": 3.487425560173086, "grad_norm": 4.287769794464111, "learning_rate": 2.0931243517379784e-05, "loss": 0.222160005569458, "memory(GiB)": 73.83, "step": 81400, "token_acc": 0.9452554744525548, "train_speed(iter/s)": 1.454143 }, { "epoch": 3.487639775502335, "grad_norm": 3.5168135166168213, "learning_rate": 2.0925768197484124e-05, "loss": 0.48912601470947265, "memory(GiB)": 73.83, "step": 81405, "token_acc": 0.903954802259887, "train_speed(iter/s)": 1.454143 }, { "epoch": 3.4878539908315838, "grad_norm": 3.3779821395874023, "learning_rate": 2.092029340429545e-05, "loss": 0.3721599102020264, "memory(GiB)": 73.83, "step": 81410, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.454142 }, { "epoch": 3.488068206160833, "grad_norm": 2.246291399002075, "learning_rate": 2.0914819137912916e-05, "loss": 0.273052978515625, "memory(GiB)": 73.83, "step": 81415, "token_acc": 0.9246987951807228, "train_speed(iter/s)": 1.454145 }, { "epoch": 3.488282421490082, "grad_norm": 6.340610980987549, "learning_rate": 2.0909345398435743e-05, "loss": 0.6555489063262939, "memory(GiB)": 73.83, "step": 81420, "token_acc": 0.8585209003215434, "train_speed(iter/s)": 1.454144 }, { "epoch": 3.4884966368193306, "grad_norm": 5.439996719360352, "learning_rate": 2.0903872185963063e-05, "loss": 0.22186524868011476, "memory(GiB)": 73.83, "step": 81425, "token_acc": 0.9566666666666667, "train_speed(iter/s)": 1.454141 }, { "epoch": 3.48871085214858, "grad_norm": 3.1997392177581787, "learning_rate": 2.0898399500594012e-05, "loss": 0.2734357833862305, "memory(GiB)": 73.83, "step": 81430, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.454145 }, { "epoch": 3.4889250674778287, "grad_norm": 0.9897499680519104, "learning_rate": 2.0892927342427775e-05, "loss": 0.3347002029418945, "memory(GiB)": 73.83, "step": 81435, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.454144 }, { "epoch": 3.4891392828070775, "grad_norm": 3.0403223037719727, "learning_rate": 2.0887455711563468e-05, "loss": 0.17552192211151124, "memory(GiB)": 73.83, "step": 81440, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.454149 }, { "epoch": 3.489353498136327, "grad_norm": 1.0780224800109863, "learning_rate": 2.0881984608100207e-05, "loss": 0.21754255294799804, "memory(GiB)": 73.83, "step": 81445, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.454159 }, { "epoch": 3.4895677134655756, "grad_norm": 1.1902390718460083, "learning_rate": 2.0876514032137105e-05, "loss": 0.35248570442199706, "memory(GiB)": 73.83, "step": 81450, "token_acc": 0.9320754716981132, "train_speed(iter/s)": 1.454164 }, { "epoch": 3.4897819287948244, "grad_norm": 3.6916005611419678, "learning_rate": 2.087104398377326e-05, "loss": 0.444822359085083, "memory(GiB)": 73.83, "step": 81455, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.454185 }, { "epoch": 3.4899961441240737, "grad_norm": 3.3820979595184326, "learning_rate": 2.0865574463107794e-05, "loss": 0.32882022857666016, "memory(GiB)": 73.83, "step": 81460, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.4542 }, { "epoch": 3.4902103594533225, "grad_norm": 2.400461435317993, "learning_rate": 2.0860105470239777e-05, "loss": 0.25287694931030275, "memory(GiB)": 73.83, "step": 81465, "token_acc": 0.9277566539923955, "train_speed(iter/s)": 1.454203 }, { "epoch": 3.4904245747825713, "grad_norm": 3.8483450412750244, "learning_rate": 2.0854637005268284e-05, "loss": 0.3925706624984741, "memory(GiB)": 73.83, "step": 81470, "token_acc": 0.9044368600682594, "train_speed(iter/s)": 1.454206 }, { "epoch": 3.4906387901118205, "grad_norm": 4.435704231262207, "learning_rate": 2.0849169068292386e-05, "loss": 0.35428428649902344, "memory(GiB)": 73.83, "step": 81475, "token_acc": 0.8975741239892183, "train_speed(iter/s)": 1.45421 }, { "epoch": 3.4908530054410694, "grad_norm": 0.13594317436218262, "learning_rate": 2.0843701659411112e-05, "loss": 0.3365787029266357, "memory(GiB)": 73.83, "step": 81480, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.454214 }, { "epoch": 3.491067220770318, "grad_norm": 2.9756669998168945, "learning_rate": 2.0838234778723555e-05, "loss": 0.4056069850921631, "memory(GiB)": 73.83, "step": 81485, "token_acc": 0.9262295081967213, "train_speed(iter/s)": 1.454219 }, { "epoch": 3.4912814360995674, "grad_norm": 5.4531331062316895, "learning_rate": 2.0832768426328715e-05, "loss": 0.652430248260498, "memory(GiB)": 73.83, "step": 81490, "token_acc": 0.8611111111111112, "train_speed(iter/s)": 1.45424 }, { "epoch": 3.4914956514288162, "grad_norm": 8.029071807861328, "learning_rate": 2.082730260232565e-05, "loss": 0.34937219619750975, "memory(GiB)": 73.83, "step": 81495, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.454243 }, { "epoch": 3.491709866758065, "grad_norm": 0.8243170976638794, "learning_rate": 2.0821837306813364e-05, "loss": 0.24186296463012696, "memory(GiB)": 73.83, "step": 81500, "token_acc": 0.937007874015748, "train_speed(iter/s)": 1.454267 }, { "epoch": 3.491709866758065, "eval_loss": 2.4165773391723633, "eval_runtime": 11.3639, "eval_samples_per_second": 8.8, "eval_steps_per_second": 8.8, "eval_token_acc": 0.473463687150838, "step": 81500 }, { "epoch": 3.4919240820873143, "grad_norm": 1.4784773588180542, "learning_rate": 2.0816372539890872e-05, "loss": 0.17802565097808837, "memory(GiB)": 73.83, "step": 81505, "token_acc": 0.6219864995178399, "train_speed(iter/s)": 1.453945 }, { "epoch": 3.492138297416563, "grad_norm": 4.486754894256592, "learning_rate": 2.081090830165716e-05, "loss": 0.1539832592010498, "memory(GiB)": 73.83, "step": 81510, "token_acc": 0.9631901840490797, "train_speed(iter/s)": 1.453952 }, { "epoch": 3.492352512745812, "grad_norm": 3.6938397884368896, "learning_rate": 2.0805444592211216e-05, "loss": 0.6674226284027099, "memory(GiB)": 73.83, "step": 81515, "token_acc": 0.8772455089820359, "train_speed(iter/s)": 1.453965 }, { "epoch": 3.492566728075061, "grad_norm": 2.675342082977295, "learning_rate": 2.079998141165204e-05, "loss": 0.39171395301818845, "memory(GiB)": 73.83, "step": 81520, "token_acc": 0.9226006191950464, "train_speed(iter/s)": 1.453963 }, { "epoch": 3.49278094340431, "grad_norm": 1.8528356552124023, "learning_rate": 2.0794518760078596e-05, "loss": 0.28821940422058107, "memory(GiB)": 73.83, "step": 81525, "token_acc": 0.9462809917355371, "train_speed(iter/s)": 1.453969 }, { "epoch": 3.492995158733559, "grad_norm": 2.9294748306274414, "learning_rate": 2.0789056637589845e-05, "loss": 0.4489725112915039, "memory(GiB)": 73.83, "step": 81530, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.453968 }, { "epoch": 3.493209374062808, "grad_norm": 1.3576107025146484, "learning_rate": 2.078359504428473e-05, "loss": 0.4909515857696533, "memory(GiB)": 73.83, "step": 81535, "token_acc": 0.9058441558441559, "train_speed(iter/s)": 1.453975 }, { "epoch": 3.493423589392057, "grad_norm": 0.5862759351730347, "learning_rate": 2.077813398026218e-05, "loss": 0.4220736980438232, "memory(GiB)": 73.83, "step": 81540, "token_acc": 0.9186746987951807, "train_speed(iter/s)": 1.453977 }, { "epoch": 3.4936378047213057, "grad_norm": 3.7071986198425293, "learning_rate": 2.0772673445621165e-05, "loss": 0.39397602081298827, "memory(GiB)": 73.83, "step": 81545, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.453982 }, { "epoch": 3.493852020050555, "grad_norm": 0.3064061105251312, "learning_rate": 2.076721344046059e-05, "loss": 0.3419880628585815, "memory(GiB)": 73.83, "step": 81550, "token_acc": 0.9239543726235742, "train_speed(iter/s)": 1.453988 }, { "epoch": 3.4940662353798038, "grad_norm": 0.3064539134502411, "learning_rate": 2.0761753964879367e-05, "loss": 0.13955488204956054, "memory(GiB)": 73.83, "step": 81555, "token_acc": 0.9537037037037037, "train_speed(iter/s)": 1.453994 }, { "epoch": 3.4942804507090526, "grad_norm": 2.7633211612701416, "learning_rate": 2.0756295018976397e-05, "loss": 0.4463229179382324, "memory(GiB)": 73.83, "step": 81560, "token_acc": 0.9052287581699346, "train_speed(iter/s)": 1.454 }, { "epoch": 3.494494666038302, "grad_norm": 6.787531852722168, "learning_rate": 2.0750836602850553e-05, "loss": 0.36295645236968993, "memory(GiB)": 73.83, "step": 81565, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.454003 }, { "epoch": 3.4947088813675506, "grad_norm": 3.035984516143799, "learning_rate": 2.0745378716600766e-05, "loss": 0.1993985056877136, "memory(GiB)": 73.83, "step": 81570, "token_acc": 0.9511278195488722, "train_speed(iter/s)": 1.454009 }, { "epoch": 3.4949230966967995, "grad_norm": 2.428264856338501, "learning_rate": 2.0739921360325866e-05, "loss": 0.24046759605407714, "memory(GiB)": 73.83, "step": 81575, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.454015 }, { "epoch": 3.4951373120260487, "grad_norm": 1.5297752618789673, "learning_rate": 2.0734464534124754e-05, "loss": 0.15398337841033935, "memory(GiB)": 73.83, "step": 81580, "token_acc": 0.959349593495935, "train_speed(iter/s)": 1.45403 }, { "epoch": 3.4953515273552975, "grad_norm": 3.590397357940674, "learning_rate": 2.0729008238096275e-05, "loss": 0.6444928646087646, "memory(GiB)": 73.83, "step": 81585, "token_acc": 0.8602941176470589, "train_speed(iter/s)": 1.454034 }, { "epoch": 3.4955657426845463, "grad_norm": 0.09495767205953598, "learning_rate": 2.0723552472339268e-05, "loss": 0.147231388092041, "memory(GiB)": 73.83, "step": 81590, "token_acc": 0.9653846153846154, "train_speed(iter/s)": 1.454032 }, { "epoch": 3.4957799580137956, "grad_norm": 4.503708362579346, "learning_rate": 2.071809723695257e-05, "loss": 0.3987578392028809, "memory(GiB)": 73.83, "step": 81595, "token_acc": 0.93, "train_speed(iter/s)": 1.454035 }, { "epoch": 3.4959941733430444, "grad_norm": 3.09845232963562, "learning_rate": 2.071264253203499e-05, "loss": 0.4430992603302002, "memory(GiB)": 73.83, "step": 81600, "token_acc": 0.9093851132686084, "train_speed(iter/s)": 1.454034 }, { "epoch": 3.496208388672293, "grad_norm": 3.848686933517456, "learning_rate": 2.0707188357685393e-05, "loss": 0.6113576889038086, "memory(GiB)": 73.83, "step": 81605, "token_acc": 0.8538205980066446, "train_speed(iter/s)": 1.454036 }, { "epoch": 3.4964226040015425, "grad_norm": 0.2765128016471863, "learning_rate": 2.0701734714002545e-05, "loss": 0.41216106414794923, "memory(GiB)": 73.83, "step": 81610, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.454039 }, { "epoch": 3.4966368193307913, "grad_norm": 0.16251805424690247, "learning_rate": 2.0696281601085264e-05, "loss": 0.15041977167129517, "memory(GiB)": 73.83, "step": 81615, "token_acc": 0.9605734767025089, "train_speed(iter/s)": 1.454043 }, { "epoch": 3.49685103466004, "grad_norm": 3.607287883758545, "learning_rate": 2.0690829019032327e-05, "loss": 0.31801567077636717, "memory(GiB)": 73.83, "step": 81620, "token_acc": 0.9276595744680851, "train_speed(iter/s)": 1.454049 }, { "epoch": 3.4970652499892894, "grad_norm": 2.955057382583618, "learning_rate": 2.06853769679425e-05, "loss": 0.126358699798584, "memory(GiB)": 73.83, "step": 81625, "token_acc": 0.9700996677740864, "train_speed(iter/s)": 1.454054 }, { "epoch": 3.497279465318538, "grad_norm": 1.4989509582519531, "learning_rate": 2.067992544791459e-05, "loss": 0.11614234447479248, "memory(GiB)": 73.83, "step": 81630, "token_acc": 0.9694915254237289, "train_speed(iter/s)": 1.454061 }, { "epoch": 3.497493680647787, "grad_norm": 4.079159259796143, "learning_rate": 2.067447445904731e-05, "loss": 0.41420583724975585, "memory(GiB)": 73.83, "step": 81635, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.454066 }, { "epoch": 3.4977078959770362, "grad_norm": 1.4491477012634277, "learning_rate": 2.066902400143946e-05, "loss": 0.4548398494720459, "memory(GiB)": 73.83, "step": 81640, "token_acc": 0.9105960264900662, "train_speed(iter/s)": 1.454067 }, { "epoch": 3.497922111306285, "grad_norm": 2.4018893241882324, "learning_rate": 2.066357407518975e-05, "loss": 0.18503435850143432, "memory(GiB)": 73.83, "step": 81645, "token_acc": 0.9607843137254902, "train_speed(iter/s)": 1.454071 }, { "epoch": 3.498136326635534, "grad_norm": 3.7331793308258057, "learning_rate": 2.0658124680396918e-05, "loss": 0.5661936283111573, "memory(GiB)": 73.83, "step": 81650, "token_acc": 0.8754448398576512, "train_speed(iter/s)": 1.454071 }, { "epoch": 3.498350541964783, "grad_norm": 4.861818313598633, "learning_rate": 2.065267581715968e-05, "loss": 0.36453607082366946, "memory(GiB)": 73.83, "step": 81655, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.454077 }, { "epoch": 3.498564757294032, "grad_norm": 0.12053276598453522, "learning_rate": 2.0647227485576733e-05, "loss": 0.34164891242980955, "memory(GiB)": 73.83, "step": 81660, "token_acc": 0.9375, "train_speed(iter/s)": 1.454087 }, { "epoch": 3.4987789726232807, "grad_norm": 4.004881381988525, "learning_rate": 2.064177968574681e-05, "loss": 0.5140721321105957, "memory(GiB)": 73.83, "step": 81665, "token_acc": 0.8941605839416058, "train_speed(iter/s)": 1.454089 }, { "epoch": 3.49899318795253, "grad_norm": 1.2734955549240112, "learning_rate": 2.0636332417768595e-05, "loss": 0.11377791166305543, "memory(GiB)": 73.83, "step": 81670, "token_acc": 0.9713375796178344, "train_speed(iter/s)": 1.454096 }, { "epoch": 3.499207403281779, "grad_norm": 3.854006290435791, "learning_rate": 2.0630885681740758e-05, "loss": 0.3429248332977295, "memory(GiB)": 73.83, "step": 81675, "token_acc": 0.9331210191082803, "train_speed(iter/s)": 1.454097 }, { "epoch": 3.4994216186110276, "grad_norm": 3.8860301971435547, "learning_rate": 2.0625439477761975e-05, "loss": 0.3210040807723999, "memory(GiB)": 73.83, "step": 81680, "token_acc": 0.9224806201550387, "train_speed(iter/s)": 1.454106 }, { "epoch": 3.499635833940277, "grad_norm": 1.0992770195007324, "learning_rate": 2.0619993805930904e-05, "loss": 0.2852896451950073, "memory(GiB)": 73.83, "step": 81685, "token_acc": 0.9271137026239067, "train_speed(iter/s)": 1.454103 }, { "epoch": 3.4998500492695257, "grad_norm": 3.846190929412842, "learning_rate": 2.0614548666346212e-05, "loss": 0.636055850982666, "memory(GiB)": 73.83, "step": 81690, "token_acc": 0.8657243816254417, "train_speed(iter/s)": 1.454123 }, { "epoch": 3.500064264598775, "grad_norm": 2.98571515083313, "learning_rate": 2.060910405910654e-05, "loss": 0.3953345537185669, "memory(GiB)": 73.83, "step": 81695, "token_acc": 0.9343629343629344, "train_speed(iter/s)": 1.454124 }, { "epoch": 3.5002784799280238, "grad_norm": 3.214301824569702, "learning_rate": 2.0603659984310525e-05, "loss": 0.3240651607513428, "memory(GiB)": 73.83, "step": 81700, "token_acc": 0.912, "train_speed(iter/s)": 1.454127 }, { "epoch": 3.5004926952572726, "grad_norm": 5.418041706085205, "learning_rate": 2.0598216442056784e-05, "loss": 0.3577093601226807, "memory(GiB)": 73.83, "step": 81705, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.454132 }, { "epoch": 3.500706910586522, "grad_norm": 0.8732283711433411, "learning_rate": 2.0592773432443913e-05, "loss": 0.3739889144897461, "memory(GiB)": 73.83, "step": 81710, "token_acc": 0.9203821656050956, "train_speed(iter/s)": 1.45413 }, { "epoch": 3.5009211259157706, "grad_norm": 6.330348491668701, "learning_rate": 2.058733095557055e-05, "loss": 0.6027040958404541, "memory(GiB)": 73.83, "step": 81715, "token_acc": 0.8916967509025271, "train_speed(iter/s)": 1.454131 }, { "epoch": 3.5011353412450195, "grad_norm": 4.9842424392700195, "learning_rate": 2.0581889011535264e-05, "loss": 0.5100002765655518, "memory(GiB)": 73.83, "step": 81720, "token_acc": 0.9097744360902256, "train_speed(iter/s)": 1.454132 }, { "epoch": 3.5013495565742687, "grad_norm": 4.75447416305542, "learning_rate": 2.0576447600436676e-05, "loss": 0.3727993965148926, "memory(GiB)": 73.83, "step": 81725, "token_acc": 0.9224489795918367, "train_speed(iter/s)": 1.454133 }, { "epoch": 3.5015637719035175, "grad_norm": 0.8651868104934692, "learning_rate": 2.0571006722373343e-05, "loss": 0.2459939956665039, "memory(GiB)": 73.83, "step": 81730, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.454135 }, { "epoch": 3.5017779872327663, "grad_norm": 7.843503475189209, "learning_rate": 2.0565566377443823e-05, "loss": 0.6126383781433106, "memory(GiB)": 73.83, "step": 81735, "token_acc": 0.8833922261484098, "train_speed(iter/s)": 1.454135 }, { "epoch": 3.5019922025620156, "grad_norm": 2.9296247959136963, "learning_rate": 2.0560126565746686e-05, "loss": 0.2596792221069336, "memory(GiB)": 73.83, "step": 81740, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.454138 }, { "epoch": 3.5022064178912644, "grad_norm": 3.1936352252960205, "learning_rate": 2.0554687287380447e-05, "loss": 0.24991979598999023, "memory(GiB)": 73.83, "step": 81745, "token_acc": 0.9453924914675768, "train_speed(iter/s)": 1.454141 }, { "epoch": 3.5024206332205132, "grad_norm": 1.4658533334732056, "learning_rate": 2.0549248542443695e-05, "loss": 0.14710804224014282, "memory(GiB)": 73.83, "step": 81750, "token_acc": 0.9544072948328267, "train_speed(iter/s)": 1.454147 }, { "epoch": 3.5026348485497625, "grad_norm": 5.25615930557251, "learning_rate": 2.0543810331034928e-05, "loss": 0.6023241996765136, "memory(GiB)": 73.83, "step": 81755, "token_acc": 0.874251497005988, "train_speed(iter/s)": 1.454158 }, { "epoch": 3.5028490638790113, "grad_norm": 3.6785850524902344, "learning_rate": 2.0538372653252673e-05, "loss": 0.23158721923828124, "memory(GiB)": 73.83, "step": 81760, "token_acc": 0.9518900343642611, "train_speed(iter/s)": 1.454162 }, { "epoch": 3.50306327920826, "grad_norm": 2.376525402069092, "learning_rate": 2.0532935509195432e-05, "loss": 0.5134202480316162, "memory(GiB)": 73.83, "step": 81765, "token_acc": 0.887719298245614, "train_speed(iter/s)": 1.45417 }, { "epoch": 3.5032774945375094, "grad_norm": 2.3110852241516113, "learning_rate": 2.0527498898961695e-05, "loss": 0.2085875988006592, "memory(GiB)": 73.83, "step": 81770, "token_acc": 0.948905109489051, "train_speed(iter/s)": 1.454175 }, { "epoch": 3.503491709866758, "grad_norm": 4.164665699005127, "learning_rate": 2.0522062822649973e-05, "loss": 0.3810727119445801, "memory(GiB)": 73.83, "step": 81775, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.45419 }, { "epoch": 3.503705925196007, "grad_norm": 1.6350557804107666, "learning_rate": 2.0516627280358737e-05, "loss": 0.4254435062408447, "memory(GiB)": 73.83, "step": 81780, "token_acc": 0.9049295774647887, "train_speed(iter/s)": 1.454198 }, { "epoch": 3.5039201405252562, "grad_norm": 3.5741782188415527, "learning_rate": 2.051119227218644e-05, "loss": 0.2668855428695679, "memory(GiB)": 73.83, "step": 81785, "token_acc": 0.9529780564263323, "train_speed(iter/s)": 1.454209 }, { "epoch": 3.504134355854505, "grad_norm": 2.2737207412719727, "learning_rate": 2.0505757798231573e-05, "loss": 0.2978676319122314, "memory(GiB)": 73.83, "step": 81790, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.454212 }, { "epoch": 3.504348571183754, "grad_norm": 2.254509210586548, "learning_rate": 2.0500323858592572e-05, "loss": 0.3546449184417725, "memory(GiB)": 73.83, "step": 81795, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.454225 }, { "epoch": 3.504562786513003, "grad_norm": 4.304095268249512, "learning_rate": 2.0494890453367875e-05, "loss": 0.6187854290008545, "memory(GiB)": 73.83, "step": 81800, "token_acc": 0.867816091954023, "train_speed(iter/s)": 1.454225 }, { "epoch": 3.504777001842252, "grad_norm": 3.937211036682129, "learning_rate": 2.0489457582655902e-05, "loss": 0.3319952964782715, "memory(GiB)": 73.83, "step": 81805, "token_acc": 0.9233038348082596, "train_speed(iter/s)": 1.454235 }, { "epoch": 3.5049912171715008, "grad_norm": 4.278759956359863, "learning_rate": 2.0484025246555107e-05, "loss": 0.2771162986755371, "memory(GiB)": 73.83, "step": 81810, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.454236 }, { "epoch": 3.50520543250075, "grad_norm": 1.3828049898147583, "learning_rate": 2.047859344516388e-05, "loss": 0.17799665927886962, "memory(GiB)": 73.83, "step": 81815, "token_acc": 0.9536423841059603, "train_speed(iter/s)": 1.454243 }, { "epoch": 3.505419647829999, "grad_norm": 2.1136093139648438, "learning_rate": 2.0473162178580622e-05, "loss": 0.41824731826782224, "memory(GiB)": 73.83, "step": 81820, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.454254 }, { "epoch": 3.5056338631592476, "grad_norm": 2.7391369342803955, "learning_rate": 2.0467731446903736e-05, "loss": 0.3240753173828125, "memory(GiB)": 73.83, "step": 81825, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.454255 }, { "epoch": 3.505848078488497, "grad_norm": 2.2051892280578613, "learning_rate": 2.046230125023158e-05, "loss": 0.5487382411956787, "memory(GiB)": 73.83, "step": 81830, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.454267 }, { "epoch": 3.5060622938177457, "grad_norm": 4.248746871948242, "learning_rate": 2.0456871588662563e-05, "loss": 0.4222099781036377, "memory(GiB)": 73.83, "step": 81835, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.454271 }, { "epoch": 3.5062765091469945, "grad_norm": 2.302772045135498, "learning_rate": 2.0451442462295028e-05, "loss": 0.7407795429229737, "memory(GiB)": 73.83, "step": 81840, "token_acc": 0.8652482269503546, "train_speed(iter/s)": 1.454293 }, { "epoch": 3.5064907244762438, "grad_norm": 5.744873523712158, "learning_rate": 2.044601387122733e-05, "loss": 0.4040919303894043, "memory(GiB)": 73.83, "step": 81845, "token_acc": 0.9246376811594202, "train_speed(iter/s)": 1.454304 }, { "epoch": 3.5067049398054926, "grad_norm": 6.215102672576904, "learning_rate": 2.044058581555782e-05, "loss": 0.44893360137939453, "memory(GiB)": 73.83, "step": 81850, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.454314 }, { "epoch": 3.5069191551347414, "grad_norm": 5.77133321762085, "learning_rate": 2.0435158295384826e-05, "loss": 0.3475500583648682, "memory(GiB)": 73.83, "step": 81855, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.454313 }, { "epoch": 3.5071333704639907, "grad_norm": 7.1499834060668945, "learning_rate": 2.0429731310806655e-05, "loss": 0.4663414001464844, "memory(GiB)": 73.83, "step": 81860, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.454319 }, { "epoch": 3.5073475857932395, "grad_norm": 3.2363228797912598, "learning_rate": 2.042430486192164e-05, "loss": 0.2010741949081421, "memory(GiB)": 73.83, "step": 81865, "token_acc": 0.9589552238805971, "train_speed(iter/s)": 1.454326 }, { "epoch": 3.5075618011224883, "grad_norm": 3.3855934143066406, "learning_rate": 2.04188789488281e-05, "loss": 0.21447358131408692, "memory(GiB)": 73.83, "step": 81870, "token_acc": 0.954225352112676, "train_speed(iter/s)": 1.454327 }, { "epoch": 3.5077760164517375, "grad_norm": 2.757859706878662, "learning_rate": 2.0413453571624323e-05, "loss": 0.28073883056640625, "memory(GiB)": 73.83, "step": 81875, "token_acc": 0.95, "train_speed(iter/s)": 1.454331 }, { "epoch": 3.5079902317809863, "grad_norm": 5.877250671386719, "learning_rate": 2.0408028730408584e-05, "loss": 0.5159302711486816, "memory(GiB)": 73.83, "step": 81880, "token_acc": 0.8984615384615384, "train_speed(iter/s)": 1.454332 }, { "epoch": 3.508204447110235, "grad_norm": 2.705023765563965, "learning_rate": 2.040260442527917e-05, "loss": 0.3082413196563721, "memory(GiB)": 73.83, "step": 81885, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.454338 }, { "epoch": 3.5084186624394844, "grad_norm": 3.604522943496704, "learning_rate": 2.0397180656334318e-05, "loss": 0.22171368598937988, "memory(GiB)": 73.83, "step": 81890, "token_acc": 0.9427480916030534, "train_speed(iter/s)": 1.45434 }, { "epoch": 3.5086328777687332, "grad_norm": 0.5309501886367798, "learning_rate": 2.039175742367233e-05, "loss": 0.20624353885650634, "memory(GiB)": 73.83, "step": 81895, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.454338 }, { "epoch": 3.508847093097982, "grad_norm": 3.566711664199829, "learning_rate": 2.0386334727391432e-05, "loss": 0.33571529388427734, "memory(GiB)": 73.83, "step": 81900, "token_acc": 0.9304635761589404, "train_speed(iter/s)": 1.454352 }, { "epoch": 3.5090613084272313, "grad_norm": 2.7978405952453613, "learning_rate": 2.038091256758986e-05, "loss": 0.26230266094207766, "memory(GiB)": 73.83, "step": 81905, "token_acc": 0.9379562043795621, "train_speed(iter/s)": 1.454357 }, { "epoch": 3.50927552375648, "grad_norm": 0.9230490326881409, "learning_rate": 2.037549094436584e-05, "loss": 0.30072526931762694, "memory(GiB)": 73.83, "step": 81910, "token_acc": 0.9388489208633094, "train_speed(iter/s)": 1.454361 }, { "epoch": 3.509489739085729, "grad_norm": 0.42022451758384705, "learning_rate": 2.0370069857817576e-05, "loss": 0.41652717590332033, "memory(GiB)": 73.83, "step": 81915, "token_acc": 0.9042553191489362, "train_speed(iter/s)": 1.454366 }, { "epoch": 3.509703954414978, "grad_norm": 7.795313358306885, "learning_rate": 2.0364649308043303e-05, "loss": 0.439239501953125, "memory(GiB)": 73.83, "step": 81920, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.454366 }, { "epoch": 3.509918169744227, "grad_norm": 1.257448434829712, "learning_rate": 2.0359229295141213e-05, "loss": 0.297123646736145, "memory(GiB)": 73.83, "step": 81925, "token_acc": 0.932258064516129, "train_speed(iter/s)": 1.454379 }, { "epoch": 3.510132385073476, "grad_norm": 4.485296249389648, "learning_rate": 2.035380981920949e-05, "loss": 0.5895516872406006, "memory(GiB)": 73.83, "step": 81930, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.454375 }, { "epoch": 3.510346600402725, "grad_norm": 4.107789993286133, "learning_rate": 2.034839088034629e-05, "loss": 0.2921851634979248, "memory(GiB)": 73.83, "step": 81935, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.454381 }, { "epoch": 3.510560815731974, "grad_norm": 2.3485350608825684, "learning_rate": 2.034297247864983e-05, "loss": 0.3015226602554321, "memory(GiB)": 73.83, "step": 81940, "token_acc": 0.9594202898550724, "train_speed(iter/s)": 1.454398 }, { "epoch": 3.5107750310612227, "grad_norm": 5.436483383178711, "learning_rate": 2.033755461421824e-05, "loss": 0.3990980386734009, "memory(GiB)": 73.83, "step": 81945, "token_acc": 0.9054054054054054, "train_speed(iter/s)": 1.454406 }, { "epoch": 3.510989246390472, "grad_norm": 3.2077314853668213, "learning_rate": 2.0332137287149654e-05, "loss": 0.24229609966278076, "memory(GiB)": 73.83, "step": 81950, "token_acc": 0.950207468879668, "train_speed(iter/s)": 1.454424 }, { "epoch": 3.5112034617197208, "grad_norm": 2.38175106048584, "learning_rate": 2.0326720497542252e-05, "loss": 0.3785180330276489, "memory(GiB)": 73.83, "step": 81955, "token_acc": 0.9346153846153846, "train_speed(iter/s)": 1.454431 }, { "epoch": 3.5114176770489696, "grad_norm": 0.561284065246582, "learning_rate": 2.0321304245494145e-05, "loss": 0.5114459991455078, "memory(GiB)": 73.83, "step": 81960, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.454431 }, { "epoch": 3.511631892378219, "grad_norm": 2.8476462364196777, "learning_rate": 2.031588853110345e-05, "loss": 0.46484670639038084, "memory(GiB)": 73.83, "step": 81965, "token_acc": 0.90625, "train_speed(iter/s)": 1.454448 }, { "epoch": 3.5118461077074676, "grad_norm": 1.2434003353118896, "learning_rate": 2.0310473354468283e-05, "loss": 0.18286147117614746, "memory(GiB)": 73.83, "step": 81970, "token_acc": 0.9606557377049181, "train_speed(iter/s)": 1.45446 }, { "epoch": 3.5120603230367164, "grad_norm": 2.599616050720215, "learning_rate": 2.030505871568672e-05, "loss": 0.49138641357421875, "memory(GiB)": 73.83, "step": 81975, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.454464 }, { "epoch": 3.5122745383659657, "grad_norm": 4.120331287384033, "learning_rate": 2.0299644614856895e-05, "loss": 0.29954659938812256, "memory(GiB)": 73.83, "step": 81980, "token_acc": 0.919831223628692, "train_speed(iter/s)": 1.454488 }, { "epoch": 3.5124887536952145, "grad_norm": 5.337890625, "learning_rate": 2.0294231052076874e-05, "loss": 0.3554528713226318, "memory(GiB)": 73.83, "step": 81985, "token_acc": 0.9424242424242424, "train_speed(iter/s)": 1.454492 }, { "epoch": 3.5127029690244633, "grad_norm": 3.652970314025879, "learning_rate": 2.028881802744472e-05, "loss": 0.4754783630371094, "memory(GiB)": 73.83, "step": 81990, "token_acc": 0.8859060402684564, "train_speed(iter/s)": 1.454493 }, { "epoch": 3.5129171843537126, "grad_norm": 2.5963385105133057, "learning_rate": 2.0283405541058494e-05, "loss": 0.41719746589660645, "memory(GiB)": 73.83, "step": 81995, "token_acc": 0.901840490797546, "train_speed(iter/s)": 1.454497 }, { "epoch": 3.5131313996829614, "grad_norm": 1.5668424367904663, "learning_rate": 2.0277993593016252e-05, "loss": 0.1477912187576294, "memory(GiB)": 73.83, "step": 82000, "token_acc": 0.9650793650793651, "train_speed(iter/s)": 1.454501 }, { "epoch": 3.5131313996829614, "eval_loss": 2.39411997795105, "eval_runtime": 11.8602, "eval_samples_per_second": 8.432, "eval_steps_per_second": 8.432, "eval_token_acc": 0.4610951008645533, "step": 82000 }, { "epoch": 3.51334561501221, "grad_norm": 4.1569318771362305, "learning_rate": 2.0272582183416027e-05, "loss": 0.3343372344970703, "memory(GiB)": 73.83, "step": 82005, "token_acc": 0.5938461538461538, "train_speed(iter/s)": 1.45417 }, { "epoch": 3.5135598303414595, "grad_norm": 1.011996865272522, "learning_rate": 2.026717131235586e-05, "loss": 0.3292783975601196, "memory(GiB)": 73.83, "step": 82010, "token_acc": 0.9432624113475178, "train_speed(iter/s)": 1.454182 }, { "epoch": 3.5137740456707083, "grad_norm": 1.4700415134429932, "learning_rate": 2.0261760979933796e-05, "loss": 0.31898088455200196, "memory(GiB)": 73.83, "step": 82015, "token_acc": 0.939209726443769, "train_speed(iter/s)": 1.454189 }, { "epoch": 3.513988260999957, "grad_norm": 6.035800933837891, "learning_rate": 2.0256351186247824e-05, "loss": 0.36334905624389646, "memory(GiB)": 73.83, "step": 82020, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.454193 }, { "epoch": 3.5142024763292063, "grad_norm": 4.010992050170898, "learning_rate": 2.0250941931395957e-05, "loss": 0.35979082584381106, "memory(GiB)": 73.83, "step": 82025, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.454202 }, { "epoch": 3.514416691658455, "grad_norm": 3.9002628326416016, "learning_rate": 2.024553321547618e-05, "loss": 0.4889513969421387, "memory(GiB)": 73.83, "step": 82030, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.454208 }, { "epoch": 3.514630906987704, "grad_norm": 4.003895282745361, "learning_rate": 2.0240125038586465e-05, "loss": 0.6733206748962403, "memory(GiB)": 73.83, "step": 82035, "token_acc": 0.8529411764705882, "train_speed(iter/s)": 1.454207 }, { "epoch": 3.5148451223169532, "grad_norm": 3.1325817108154297, "learning_rate": 2.0234717400824814e-05, "loss": 0.40572199821472166, "memory(GiB)": 73.83, "step": 82040, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.45421 }, { "epoch": 3.515059337646202, "grad_norm": 3.001085042953491, "learning_rate": 2.022931030228919e-05, "loss": 0.6564332962036132, "memory(GiB)": 73.83, "step": 82045, "token_acc": 0.8525641025641025, "train_speed(iter/s)": 1.45423 }, { "epoch": 3.515273552975451, "grad_norm": 2.231724500656128, "learning_rate": 2.0223903743077528e-05, "loss": 0.3553037166595459, "memory(GiB)": 73.83, "step": 82050, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.454245 }, { "epoch": 3.5154877683047, "grad_norm": 0.7295597195625305, "learning_rate": 2.0218497723287788e-05, "loss": 0.2846890687942505, "memory(GiB)": 73.83, "step": 82055, "token_acc": 0.9319526627218935, "train_speed(iter/s)": 1.454244 }, { "epoch": 3.515701983633949, "grad_norm": 0.37895461916923523, "learning_rate": 2.021309224301788e-05, "loss": 0.34370200634002684, "memory(GiB)": 73.83, "step": 82060, "token_acc": 0.923943661971831, "train_speed(iter/s)": 1.454251 }, { "epoch": 3.5159161989631977, "grad_norm": 2.9099369049072266, "learning_rate": 2.0207687302365762e-05, "loss": 0.3181877851486206, "memory(GiB)": 73.83, "step": 82065, "token_acc": 0.925, "train_speed(iter/s)": 1.454255 }, { "epoch": 3.516130414292447, "grad_norm": 7.337696075439453, "learning_rate": 2.0202282901429338e-05, "loss": 0.5173470497131347, "memory(GiB)": 73.83, "step": 82070, "token_acc": 0.8943396226415095, "train_speed(iter/s)": 1.454252 }, { "epoch": 3.516344629621696, "grad_norm": 3.004106283187866, "learning_rate": 2.019687904030651e-05, "loss": 0.3376582384109497, "memory(GiB)": 73.83, "step": 82075, "token_acc": 0.921146953405018, "train_speed(iter/s)": 1.454261 }, { "epoch": 3.5165588449509446, "grad_norm": 3.7901041507720947, "learning_rate": 2.0191475719095166e-05, "loss": 0.45209307670593263, "memory(GiB)": 73.83, "step": 82080, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.454265 }, { "epoch": 3.516773060280194, "grad_norm": 4.233193397521973, "learning_rate": 2.0186072937893212e-05, "loss": 0.5218348979949952, "memory(GiB)": 73.83, "step": 82085, "token_acc": 0.8655172413793103, "train_speed(iter/s)": 1.454282 }, { "epoch": 3.5169872756094427, "grad_norm": 3.2257206439971924, "learning_rate": 2.0180670696798514e-05, "loss": 0.528790807723999, "memory(GiB)": 73.83, "step": 82090, "token_acc": 0.8934169278996865, "train_speed(iter/s)": 1.454292 }, { "epoch": 3.5172014909386915, "grad_norm": 2.964360475540161, "learning_rate": 2.0175268995908924e-05, "loss": 0.349299955368042, "memory(GiB)": 73.83, "step": 82095, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.454292 }, { "epoch": 3.5174157062679408, "grad_norm": 5.0487775802612305, "learning_rate": 2.0169867835322332e-05, "loss": 0.24162416458129882, "memory(GiB)": 73.83, "step": 82100, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.454299 }, { "epoch": 3.5176299215971896, "grad_norm": 1.6863470077514648, "learning_rate": 2.0164467215136566e-05, "loss": 0.42829103469848634, "memory(GiB)": 73.83, "step": 82105, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.454302 }, { "epoch": 3.5178441369264384, "grad_norm": 1.2964589595794678, "learning_rate": 2.0159067135449467e-05, "loss": 0.130772864818573, "memory(GiB)": 73.83, "step": 82110, "token_acc": 0.9636871508379888, "train_speed(iter/s)": 1.454308 }, { "epoch": 3.5180583522556876, "grad_norm": 1.0234650373458862, "learning_rate": 2.0153667596358862e-05, "loss": 0.1410611391067505, "memory(GiB)": 73.83, "step": 82115, "token_acc": 0.9694323144104804, "train_speed(iter/s)": 1.454309 }, { "epoch": 3.5182725675849364, "grad_norm": 3.5213849544525146, "learning_rate": 2.0148268597962544e-05, "loss": 0.31611328125, "memory(GiB)": 73.83, "step": 82120, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.454314 }, { "epoch": 3.5184867829141853, "grad_norm": 2.45074200630188, "learning_rate": 2.014287014035836e-05, "loss": 0.4213207721710205, "memory(GiB)": 73.83, "step": 82125, "token_acc": 0.8984615384615384, "train_speed(iter/s)": 1.454322 }, { "epoch": 3.5187009982434345, "grad_norm": 2.4297828674316406, "learning_rate": 2.0137472223644093e-05, "loss": 0.5156672477722168, "memory(GiB)": 73.83, "step": 82130, "token_acc": 0.887719298245614, "train_speed(iter/s)": 1.454321 }, { "epoch": 3.5189152135726833, "grad_norm": 3.758256435394287, "learning_rate": 2.0132074847917533e-05, "loss": 0.4463181018829346, "memory(GiB)": 73.83, "step": 82135, "token_acc": 0.8976897689768977, "train_speed(iter/s)": 1.454319 }, { "epoch": 3.519129428901932, "grad_norm": 0.19992172718048096, "learning_rate": 2.012667801327645e-05, "loss": 0.34397616386413576, "memory(GiB)": 73.83, "step": 82140, "token_acc": 0.9159420289855073, "train_speed(iter/s)": 1.454317 }, { "epoch": 3.5193436442311814, "grad_norm": 2.8609206676483154, "learning_rate": 2.0121281719818625e-05, "loss": 0.18394804000854492, "memory(GiB)": 73.83, "step": 82145, "token_acc": 0.9525547445255474, "train_speed(iter/s)": 1.45432 }, { "epoch": 3.51955785956043, "grad_norm": 1.6796661615371704, "learning_rate": 2.011588596764179e-05, "loss": 0.13468751907348633, "memory(GiB)": 73.83, "step": 82150, "token_acc": 0.9736842105263158, "train_speed(iter/s)": 1.454323 }, { "epoch": 3.519772074889679, "grad_norm": 3.948446750640869, "learning_rate": 2.0110490756843714e-05, "loss": 0.37701902389526365, "memory(GiB)": 73.83, "step": 82155, "token_acc": 0.9078498293515358, "train_speed(iter/s)": 1.454332 }, { "epoch": 3.5199862902189283, "grad_norm": 4.6521830558776855, "learning_rate": 2.0105096087522153e-05, "loss": 0.29623286724090575, "memory(GiB)": 73.83, "step": 82160, "token_acc": 0.9331103678929766, "train_speed(iter/s)": 1.454335 }, { "epoch": 3.520200505548177, "grad_norm": 0.03154844790697098, "learning_rate": 2.009970195977482e-05, "loss": 0.3195100545883179, "memory(GiB)": 73.83, "step": 82165, "token_acc": 0.9303135888501742, "train_speed(iter/s)": 1.454337 }, { "epoch": 3.520414720877426, "grad_norm": 2.693312883377075, "learning_rate": 2.0094308373699434e-05, "loss": 0.33431472778320315, "memory(GiB)": 73.83, "step": 82170, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.454337 }, { "epoch": 3.520628936206675, "grad_norm": 6.508995056152344, "learning_rate": 2.008891532939371e-05, "loss": 0.6221721649169922, "memory(GiB)": 73.83, "step": 82175, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.454344 }, { "epoch": 3.520843151535924, "grad_norm": 3.7755191326141357, "learning_rate": 2.008352282695532e-05, "loss": 0.47798857688903806, "memory(GiB)": 73.83, "step": 82180, "token_acc": 0.8993506493506493, "train_speed(iter/s)": 1.454343 }, { "epoch": 3.521057366865173, "grad_norm": 3.4144976139068604, "learning_rate": 2.0078130866481998e-05, "loss": 0.3666471242904663, "memory(GiB)": 73.83, "step": 82185, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.454339 }, { "epoch": 3.521271582194422, "grad_norm": 1.104866623878479, "learning_rate": 2.0072739448071405e-05, "loss": 0.3251136302947998, "memory(GiB)": 73.83, "step": 82190, "token_acc": 0.927038626609442, "train_speed(iter/s)": 1.454338 }, { "epoch": 3.521485797523671, "grad_norm": 4.349210262298584, "learning_rate": 2.0067348571821214e-05, "loss": 0.22411050796508789, "memory(GiB)": 73.83, "step": 82195, "token_acc": 0.928125, "train_speed(iter/s)": 1.454338 }, { "epoch": 3.5217000128529197, "grad_norm": 2.6343040466308594, "learning_rate": 2.0061958237829075e-05, "loss": 0.3123350143432617, "memory(GiB)": 73.83, "step": 82200, "token_acc": 0.9361111111111111, "train_speed(iter/s)": 1.454351 }, { "epoch": 3.521914228182169, "grad_norm": 1.866194248199463, "learning_rate": 2.0056568446192636e-05, "loss": 0.1596990704536438, "memory(GiB)": 73.83, "step": 82205, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.454346 }, { "epoch": 3.5221284435114177, "grad_norm": 3.7026336193084717, "learning_rate": 2.0051179197009568e-05, "loss": 0.3351642847061157, "memory(GiB)": 73.83, "step": 82210, "token_acc": 0.8923076923076924, "train_speed(iter/s)": 1.454346 }, { "epoch": 3.5223426588406666, "grad_norm": 3.932443618774414, "learning_rate": 2.0045790490377476e-05, "loss": 0.382962703704834, "memory(GiB)": 73.83, "step": 82215, "token_acc": 0.9033232628398792, "train_speed(iter/s)": 1.454343 }, { "epoch": 3.522556874169916, "grad_norm": 2.5704336166381836, "learning_rate": 2.0040402326393993e-05, "loss": 0.19385374784469606, "memory(GiB)": 73.83, "step": 82220, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.454359 }, { "epoch": 3.5227710894991646, "grad_norm": 2.39162278175354, "learning_rate": 2.0035014705156723e-05, "loss": 0.1329616665840149, "memory(GiB)": 73.83, "step": 82225, "token_acc": 0.966789667896679, "train_speed(iter/s)": 1.45438 }, { "epoch": 3.5229853048284134, "grad_norm": 2.670222520828247, "learning_rate": 2.002962762676326e-05, "loss": 0.218384051322937, "memory(GiB)": 73.83, "step": 82230, "token_acc": 0.9465020576131687, "train_speed(iter/s)": 1.454383 }, { "epoch": 3.5231995201576627, "grad_norm": 2.5512523651123047, "learning_rate": 2.0024241091311218e-05, "loss": 0.2942268133163452, "memory(GiB)": 73.83, "step": 82235, "token_acc": 0.9282868525896414, "train_speed(iter/s)": 1.454392 }, { "epoch": 3.5234137354869115, "grad_norm": 2.785784959793091, "learning_rate": 2.0018855098898153e-05, "loss": 0.24587514400482177, "memory(GiB)": 73.83, "step": 82240, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.454388 }, { "epoch": 3.5236279508161603, "grad_norm": 3.2760109901428223, "learning_rate": 2.001346964962167e-05, "loss": 0.42155656814575193, "memory(GiB)": 73.83, "step": 82245, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.454395 }, { "epoch": 3.5238421661454096, "grad_norm": 2.1966803073883057, "learning_rate": 2.0008084743579313e-05, "loss": 0.46341261863708494, "memory(GiB)": 73.83, "step": 82250, "token_acc": 0.87890625, "train_speed(iter/s)": 1.454394 }, { "epoch": 3.5240563814746584, "grad_norm": 0.6726401448249817, "learning_rate": 2.0002700380868633e-05, "loss": 0.19932737350463867, "memory(GiB)": 73.83, "step": 82255, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.454397 }, { "epoch": 3.524270596803907, "grad_norm": 6.563268184661865, "learning_rate": 1.999731656158718e-05, "loss": 0.2745058059692383, "memory(GiB)": 73.83, "step": 82260, "token_acc": 0.9307958477508651, "train_speed(iter/s)": 1.454408 }, { "epoch": 3.5244848121331565, "grad_norm": 1.7359968423843384, "learning_rate": 1.9991933285832465e-05, "loss": 0.27789485454559326, "memory(GiB)": 73.83, "step": 82265, "token_acc": 0.943217665615142, "train_speed(iter/s)": 1.45441 }, { "epoch": 3.5246990274624053, "grad_norm": 4.119586944580078, "learning_rate": 1.9986550553702045e-05, "loss": 0.22395100593566894, "memory(GiB)": 73.83, "step": 82270, "token_acc": 0.9448051948051948, "train_speed(iter/s)": 1.454417 }, { "epoch": 3.524913242791654, "grad_norm": 3.7875266075134277, "learning_rate": 1.998116836529341e-05, "loss": 0.2637214183807373, "memory(GiB)": 73.83, "step": 82275, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.454422 }, { "epoch": 3.5251274581209033, "grad_norm": 1.5528676509857178, "learning_rate": 1.9975786720704083e-05, "loss": 0.2718015193939209, "memory(GiB)": 73.83, "step": 82280, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.454426 }, { "epoch": 3.525341673450152, "grad_norm": 1.3916367292404175, "learning_rate": 1.9970405620031534e-05, "loss": 0.46739954948425294, "memory(GiB)": 73.83, "step": 82285, "token_acc": 0.8909090909090909, "train_speed(iter/s)": 1.454425 }, { "epoch": 3.525555888779401, "grad_norm": 2.3963780403137207, "learning_rate": 1.996502506337326e-05, "loss": 0.6622530937194824, "memory(GiB)": 73.83, "step": 82290, "token_acc": 0.869281045751634, "train_speed(iter/s)": 1.45444 }, { "epoch": 3.52577010410865, "grad_norm": 4.706532001495361, "learning_rate": 1.9959645050826714e-05, "loss": 0.2541104078292847, "memory(GiB)": 73.83, "step": 82295, "token_acc": 0.9518518518518518, "train_speed(iter/s)": 1.454443 }, { "epoch": 3.525984319437899, "grad_norm": 2.0859999656677246, "learning_rate": 1.995426558248938e-05, "loss": 0.4784048080444336, "memory(GiB)": 73.83, "step": 82300, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.454445 }, { "epoch": 3.526198534767148, "grad_norm": 3.169126033782959, "learning_rate": 1.9948886658458727e-05, "loss": 0.30265939235687256, "memory(GiB)": 73.83, "step": 82305, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.454449 }, { "epoch": 3.526412750096397, "grad_norm": 3.186474323272705, "learning_rate": 1.9943508278832186e-05, "loss": 0.2386218547821045, "memory(GiB)": 73.83, "step": 82310, "token_acc": 0.9558823529411765, "train_speed(iter/s)": 1.454457 }, { "epoch": 3.526626965425646, "grad_norm": 3.0825819969177246, "learning_rate": 1.993813044370718e-05, "loss": 0.6129711627960205, "memory(GiB)": 73.83, "step": 82315, "token_acc": 0.8717948717948718, "train_speed(iter/s)": 1.454457 }, { "epoch": 3.5268411807548947, "grad_norm": 2.9334146976470947, "learning_rate": 1.993275315318115e-05, "loss": 0.6878533840179444, "memory(GiB)": 73.83, "step": 82320, "token_acc": 0.8565573770491803, "train_speed(iter/s)": 1.454467 }, { "epoch": 3.527055396084144, "grad_norm": 6.68211555480957, "learning_rate": 1.992737640735148e-05, "loss": 0.6533576011657715, "memory(GiB)": 73.83, "step": 82325, "token_acc": 0.8348623853211009, "train_speed(iter/s)": 1.454469 }, { "epoch": 3.527269611413393, "grad_norm": 0.2237248569726944, "learning_rate": 1.992200020631561e-05, "loss": 0.188856303691864, "memory(GiB)": 73.83, "step": 82330, "token_acc": 0.9641791044776119, "train_speed(iter/s)": 1.45448 }, { "epoch": 3.5274838267426416, "grad_norm": 3.3940865993499756, "learning_rate": 1.9916624550170926e-05, "loss": 0.4009819507598877, "memory(GiB)": 73.83, "step": 82335, "token_acc": 0.9096385542168675, "train_speed(iter/s)": 1.454491 }, { "epoch": 3.527698042071891, "grad_norm": 2.807896614074707, "learning_rate": 1.9911249439014812e-05, "loss": 0.3260331153869629, "memory(GiB)": 73.83, "step": 82340, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.454492 }, { "epoch": 3.5279122574011397, "grad_norm": 2.8024706840515137, "learning_rate": 1.9905874872944635e-05, "loss": 0.2421409845352173, "memory(GiB)": 73.83, "step": 82345, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.45449 }, { "epoch": 3.5281264727303885, "grad_norm": 0.8328789472579956, "learning_rate": 1.9900500852057747e-05, "loss": 0.1778336763381958, "memory(GiB)": 73.83, "step": 82350, "token_acc": 0.96, "train_speed(iter/s)": 1.454492 }, { "epoch": 3.5283406880596377, "grad_norm": 0.28612324595451355, "learning_rate": 1.9895127376451544e-05, "loss": 0.14297913312911986, "memory(GiB)": 73.83, "step": 82355, "token_acc": 0.9711191335740073, "train_speed(iter/s)": 1.454492 }, { "epoch": 3.5285549033888866, "grad_norm": 4.520959854125977, "learning_rate": 1.9889754446223348e-05, "loss": 0.48426237106323244, "memory(GiB)": 73.83, "step": 82360, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.454497 }, { "epoch": 3.5287691187181354, "grad_norm": 2.0157699584960938, "learning_rate": 1.9884382061470492e-05, "loss": 0.2771573781967163, "memory(GiB)": 73.83, "step": 82365, "token_acc": 0.9409282700421941, "train_speed(iter/s)": 1.454511 }, { "epoch": 3.5289833340473846, "grad_norm": 2.596770763397217, "learning_rate": 1.9879010222290314e-05, "loss": 0.45852932929992674, "memory(GiB)": 73.83, "step": 82370, "token_acc": 0.9115646258503401, "train_speed(iter/s)": 1.454513 }, { "epoch": 3.5291975493766334, "grad_norm": 4.5718512535095215, "learning_rate": 1.9873638928780093e-05, "loss": 0.38181052207946775, "memory(GiB)": 73.83, "step": 82375, "token_acc": 0.9249146757679181, "train_speed(iter/s)": 1.454518 }, { "epoch": 3.5294117647058822, "grad_norm": 0.34332212805747986, "learning_rate": 1.9868268181037185e-05, "loss": 0.27155656814575196, "memory(GiB)": 73.83, "step": 82380, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.454526 }, { "epoch": 3.5296259800351315, "grad_norm": 1.7735415697097778, "learning_rate": 1.986289797915885e-05, "loss": 0.20013442039489746, "memory(GiB)": 73.83, "step": 82385, "token_acc": 0.956, "train_speed(iter/s)": 1.454529 }, { "epoch": 3.5298401953643803, "grad_norm": 5.445033550262451, "learning_rate": 1.9857528323242407e-05, "loss": 0.2673558235168457, "memory(GiB)": 73.83, "step": 82390, "token_acc": 0.9381107491856677, "train_speed(iter/s)": 1.454533 }, { "epoch": 3.530054410693629, "grad_norm": 2.5721893310546875, "learning_rate": 1.9852159213385113e-05, "loss": 0.3562350034713745, "memory(GiB)": 73.83, "step": 82395, "token_acc": 0.912751677852349, "train_speed(iter/s)": 1.454535 }, { "epoch": 3.5302686260228784, "grad_norm": 3.926835298538208, "learning_rate": 1.9846790649684232e-05, "loss": 0.230469274520874, "memory(GiB)": 73.83, "step": 82400, "token_acc": 0.948905109489051, "train_speed(iter/s)": 1.454548 }, { "epoch": 3.530482841352127, "grad_norm": 2.8885767459869385, "learning_rate": 1.9841422632237026e-05, "loss": 0.4327532768249512, "memory(GiB)": 73.83, "step": 82405, "token_acc": 0.9070422535211268, "train_speed(iter/s)": 1.45455 }, { "epoch": 3.530697056681376, "grad_norm": 4.523051738739014, "learning_rate": 1.9836055161140725e-05, "loss": 0.3707016944885254, "memory(GiB)": 73.83, "step": 82410, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.454562 }, { "epoch": 3.5309112720106253, "grad_norm": 0.08148772269487381, "learning_rate": 1.9830688236492602e-05, "loss": 0.28222663402557374, "memory(GiB)": 73.83, "step": 82415, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.454571 }, { "epoch": 3.531125487339874, "grad_norm": 20.77364730834961, "learning_rate": 1.9825321858389855e-05, "loss": 0.4390868186950684, "memory(GiB)": 73.83, "step": 82420, "token_acc": 0.9016949152542373, "train_speed(iter/s)": 1.454593 }, { "epoch": 3.531339702669123, "grad_norm": 2.9099950790405273, "learning_rate": 1.9819956026929715e-05, "loss": 0.22618741989135743, "memory(GiB)": 73.83, "step": 82425, "token_acc": 0.950530035335689, "train_speed(iter/s)": 1.454592 }, { "epoch": 3.531553917998372, "grad_norm": 4.9377312660217285, "learning_rate": 1.9814590742209382e-05, "loss": 0.33415923118591306, "memory(GiB)": 73.83, "step": 82430, "token_acc": 0.9138888888888889, "train_speed(iter/s)": 1.454592 }, { "epoch": 3.531768133327621, "grad_norm": 5.307802200317383, "learning_rate": 1.9809226004326032e-05, "loss": 0.3641620397567749, "memory(GiB)": 73.83, "step": 82435, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.454607 }, { "epoch": 3.5319823486568698, "grad_norm": 3.591377019882202, "learning_rate": 1.9803861813376895e-05, "loss": 0.30177271366119385, "memory(GiB)": 73.83, "step": 82440, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.454609 }, { "epoch": 3.532196563986119, "grad_norm": 2.857503652572632, "learning_rate": 1.9798498169459124e-05, "loss": 0.2678952693939209, "memory(GiB)": 73.83, "step": 82445, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.454611 }, { "epoch": 3.532410779315368, "grad_norm": 4.231253623962402, "learning_rate": 1.9793135072669878e-05, "loss": 0.34844317436218264, "memory(GiB)": 73.83, "step": 82450, "token_acc": 0.9331103678929766, "train_speed(iter/s)": 1.454624 }, { "epoch": 3.5326249946446167, "grad_norm": 0.05932183191180229, "learning_rate": 1.978777252310634e-05, "loss": 0.21596174240112304, "memory(GiB)": 73.83, "step": 82455, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.454625 }, { "epoch": 3.532839209973866, "grad_norm": 0.9688424468040466, "learning_rate": 1.9782410520865648e-05, "loss": 0.4393308639526367, "memory(GiB)": 73.83, "step": 82460, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.454628 }, { "epoch": 3.5330534253031147, "grad_norm": 4.535739898681641, "learning_rate": 1.977704906604493e-05, "loss": 0.3548715591430664, "memory(GiB)": 73.83, "step": 82465, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.454632 }, { "epoch": 3.5332676406323635, "grad_norm": 0.187859445810318, "learning_rate": 1.9771688158741302e-05, "loss": 0.4465212821960449, "memory(GiB)": 73.83, "step": 82470, "token_acc": 0.8613138686131386, "train_speed(iter/s)": 1.454629 }, { "epoch": 3.533481855961613, "grad_norm": 1.099095106124878, "learning_rate": 1.976632779905192e-05, "loss": 0.11815659999847412, "memory(GiB)": 73.83, "step": 82475, "token_acc": 0.9845679012345679, "train_speed(iter/s)": 1.454636 }, { "epoch": 3.5336960712908616, "grad_norm": 4.529973030090332, "learning_rate": 1.9760967987073876e-05, "loss": 0.3436870813369751, "memory(GiB)": 73.83, "step": 82480, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.454641 }, { "epoch": 3.5339102866201104, "grad_norm": 2.262990713119507, "learning_rate": 1.9755608722904256e-05, "loss": 0.11489747762680054, "memory(GiB)": 73.83, "step": 82485, "token_acc": 0.9586466165413534, "train_speed(iter/s)": 1.454648 }, { "epoch": 3.5341245019493597, "grad_norm": 1.5356703996658325, "learning_rate": 1.975025000664016e-05, "loss": 0.20530059337615966, "memory(GiB)": 73.83, "step": 82490, "token_acc": 0.9558823529411765, "train_speed(iter/s)": 1.454656 }, { "epoch": 3.5343387172786085, "grad_norm": 1.9315327405929565, "learning_rate": 1.974489183837864e-05, "loss": 0.21419048309326172, "memory(GiB)": 73.83, "step": 82495, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.454664 }, { "epoch": 3.5345529326078573, "grad_norm": 2.939486265182495, "learning_rate": 1.9739534218216805e-05, "loss": 0.2525035858154297, "memory(GiB)": 73.83, "step": 82500, "token_acc": 0.9394812680115274, "train_speed(iter/s)": 1.454659 }, { "epoch": 3.5345529326078573, "eval_loss": 2.35441517829895, "eval_runtime": 11.3925, "eval_samples_per_second": 8.778, "eval_steps_per_second": 8.778, "eval_token_acc": 0.45758661887694146, "step": 82500 }, { "epoch": 3.5347671479371066, "grad_norm": 4.2282538414001465, "learning_rate": 1.97341771462517e-05, "loss": 0.5876214027404785, "memory(GiB)": 73.83, "step": 82505, "token_acc": 0.5652557319223986, "train_speed(iter/s)": 1.454343 }, { "epoch": 3.5349813632663554, "grad_norm": 8.17991828918457, "learning_rate": 1.9728820622580358e-05, "loss": 0.2664613723754883, "memory(GiB)": 73.83, "step": 82510, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.454347 }, { "epoch": 3.535195578595604, "grad_norm": 4.789729595184326, "learning_rate": 1.9723464647299834e-05, "loss": 0.6136574745178223, "memory(GiB)": 73.83, "step": 82515, "token_acc": 0.8821656050955414, "train_speed(iter/s)": 1.454346 }, { "epoch": 3.5354097939248534, "grad_norm": 2.8816606998443604, "learning_rate": 1.9718109220507147e-05, "loss": 0.36656005382537843, "memory(GiB)": 73.83, "step": 82520, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.454346 }, { "epoch": 3.5356240092541023, "grad_norm": 3.6177852153778076, "learning_rate": 1.97127543422993e-05, "loss": 0.5167116165161133, "memory(GiB)": 73.83, "step": 82525, "token_acc": 0.8903654485049833, "train_speed(iter/s)": 1.454356 }, { "epoch": 3.535838224583351, "grad_norm": 10.40047836303711, "learning_rate": 1.9707400012773318e-05, "loss": 0.18457144498825073, "memory(GiB)": 73.83, "step": 82530, "token_acc": 0.9567901234567902, "train_speed(iter/s)": 1.454367 }, { "epoch": 3.5360524399126003, "grad_norm": 3.5050692558288574, "learning_rate": 1.9702046232026222e-05, "loss": 0.4814561367034912, "memory(GiB)": 73.83, "step": 82535, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.454379 }, { "epoch": 3.536266655241849, "grad_norm": 4.54662561416626, "learning_rate": 1.969669300015498e-05, "loss": 0.7454927444458008, "memory(GiB)": 73.83, "step": 82540, "token_acc": 0.8588235294117647, "train_speed(iter/s)": 1.454382 }, { "epoch": 3.536480870571098, "grad_norm": 6.319517135620117, "learning_rate": 1.9691340317256575e-05, "loss": 0.5823107719421386, "memory(GiB)": 73.83, "step": 82545, "token_acc": 0.8664596273291926, "train_speed(iter/s)": 1.454398 }, { "epoch": 3.536695085900347, "grad_norm": 2.806483030319214, "learning_rate": 1.9685988183427968e-05, "loss": 0.3139964580535889, "memory(GiB)": 73.83, "step": 82550, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.454399 }, { "epoch": 3.536909301229596, "grad_norm": 4.204104900360107, "learning_rate": 1.9680636598766104e-05, "loss": 0.3377241611480713, "memory(GiB)": 73.83, "step": 82555, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.454407 }, { "epoch": 3.537123516558845, "grad_norm": 4.173004627227783, "learning_rate": 1.967528556336797e-05, "loss": 0.3321798324584961, "memory(GiB)": 73.83, "step": 82560, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.454412 }, { "epoch": 3.537337731888094, "grad_norm": 6.577325820922852, "learning_rate": 1.966993507733048e-05, "loss": 0.4091049671173096, "memory(GiB)": 73.83, "step": 82565, "token_acc": 0.9076433121019108, "train_speed(iter/s)": 1.454415 }, { "epoch": 3.537551947217343, "grad_norm": 2.5336148738861084, "learning_rate": 1.9664585140750574e-05, "loss": 0.3280740261077881, "memory(GiB)": 73.83, "step": 82570, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.454415 }, { "epoch": 3.5377661625465917, "grad_norm": 4.711719512939453, "learning_rate": 1.965923575372516e-05, "loss": 0.19516866207122802, "memory(GiB)": 73.83, "step": 82575, "token_acc": 0.9451476793248945, "train_speed(iter/s)": 1.454422 }, { "epoch": 3.537980377875841, "grad_norm": 5.178194999694824, "learning_rate": 1.965388691635114e-05, "loss": 0.5930856227874756, "memory(GiB)": 73.83, "step": 82580, "token_acc": 0.87890625, "train_speed(iter/s)": 1.454423 }, { "epoch": 3.5381945932050898, "grad_norm": 5.566661357879639, "learning_rate": 1.964853862872544e-05, "loss": 0.3426445245742798, "memory(GiB)": 73.83, "step": 82585, "token_acc": 0.913946587537092, "train_speed(iter/s)": 1.454426 }, { "epoch": 3.5384088085343386, "grad_norm": 3.2036776542663574, "learning_rate": 1.964319089094493e-05, "loss": 0.23098621368408204, "memory(GiB)": 73.83, "step": 82590, "token_acc": 0.9502923976608187, "train_speed(iter/s)": 1.454427 }, { "epoch": 3.538623023863588, "grad_norm": 0.20904022455215454, "learning_rate": 1.9637843703106503e-05, "loss": 0.2971553564071655, "memory(GiB)": 73.83, "step": 82595, "token_acc": 0.9475138121546961, "train_speed(iter/s)": 1.45443 }, { "epoch": 3.5388372391928367, "grad_norm": 2.556765079498291, "learning_rate": 1.9632497065306993e-05, "loss": 0.3328325986862183, "memory(GiB)": 73.83, "step": 82600, "token_acc": 0.9343629343629344, "train_speed(iter/s)": 1.454432 }, { "epoch": 3.5390514545220855, "grad_norm": 2.9842782020568848, "learning_rate": 1.962715097764331e-05, "loss": 0.284726619720459, "memory(GiB)": 73.83, "step": 82605, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.454435 }, { "epoch": 3.5392656698513347, "grad_norm": 0.7353065609931946, "learning_rate": 1.9621805440212275e-05, "loss": 0.4976183891296387, "memory(GiB)": 73.83, "step": 82610, "token_acc": 0.8920308483290489, "train_speed(iter/s)": 1.454446 }, { "epoch": 3.5394798851805835, "grad_norm": 5.545009136199951, "learning_rate": 1.9616460453110712e-05, "loss": 0.46688551902770997, "memory(GiB)": 73.83, "step": 82615, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.454457 }, { "epoch": 3.5396941005098324, "grad_norm": 2.3774805068969727, "learning_rate": 1.9611116016435495e-05, "loss": 0.26785621643066404, "memory(GiB)": 73.83, "step": 82620, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.454463 }, { "epoch": 3.5399083158390816, "grad_norm": 1.4531787633895874, "learning_rate": 1.9605772130283413e-05, "loss": 0.3961701154708862, "memory(GiB)": 73.83, "step": 82625, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.454474 }, { "epoch": 3.5401225311683304, "grad_norm": 3.822765588760376, "learning_rate": 1.9600428794751285e-05, "loss": 0.18073736429214476, "memory(GiB)": 73.83, "step": 82630, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.454488 }, { "epoch": 3.5403367464975792, "grad_norm": 6.461400032043457, "learning_rate": 1.9595086009935903e-05, "loss": 0.4524085521697998, "memory(GiB)": 73.83, "step": 82635, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.454487 }, { "epoch": 3.5405509618268285, "grad_norm": 4.98022985458374, "learning_rate": 1.9589743775934043e-05, "loss": 0.44606585502624513, "memory(GiB)": 73.83, "step": 82640, "token_acc": 0.9112627986348123, "train_speed(iter/s)": 1.454498 }, { "epoch": 3.5407651771560773, "grad_norm": 3.981722593307495, "learning_rate": 1.958440209284252e-05, "loss": 0.4202610492706299, "memory(GiB)": 73.83, "step": 82645, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.454498 }, { "epoch": 3.540979392485326, "grad_norm": 2.983908176422119, "learning_rate": 1.9579060960758082e-05, "loss": 0.3347383975982666, "memory(GiB)": 73.83, "step": 82650, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.454509 }, { "epoch": 3.5411936078145754, "grad_norm": 5.143003940582275, "learning_rate": 1.9573720379777498e-05, "loss": 0.2620274782180786, "memory(GiB)": 73.83, "step": 82655, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.454518 }, { "epoch": 3.541407823143824, "grad_norm": 3.428283452987671, "learning_rate": 1.9568380349997506e-05, "loss": 0.2807153224945068, "memory(GiB)": 73.83, "step": 82660, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.454517 }, { "epoch": 3.541622038473073, "grad_norm": 2.423985004425049, "learning_rate": 1.9563040871514854e-05, "loss": 0.2712676525115967, "memory(GiB)": 73.83, "step": 82665, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.454521 }, { "epoch": 3.5418362538023223, "grad_norm": 4.127657413482666, "learning_rate": 1.955770194442625e-05, "loss": 0.17274234294891358, "memory(GiB)": 73.83, "step": 82670, "token_acc": 0.9566929133858267, "train_speed(iter/s)": 1.45452 }, { "epoch": 3.542050469131571, "grad_norm": 2.2236850261688232, "learning_rate": 1.9552363568828437e-05, "loss": 0.39564085006713867, "memory(GiB)": 73.83, "step": 82675, "token_acc": 0.9153605015673981, "train_speed(iter/s)": 1.454527 }, { "epoch": 3.54226468446082, "grad_norm": 2.5940864086151123, "learning_rate": 1.954702574481813e-05, "loss": 0.22694807052612304, "memory(GiB)": 73.83, "step": 82680, "token_acc": 0.9602272727272727, "train_speed(iter/s)": 1.454528 }, { "epoch": 3.542478899790069, "grad_norm": 4.146087169647217, "learning_rate": 1.954168847249202e-05, "loss": 0.31277573108673096, "memory(GiB)": 73.83, "step": 82685, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.45453 }, { "epoch": 3.542693115119318, "grad_norm": 0.17823390662670135, "learning_rate": 1.95363517519468e-05, "loss": 0.38398215770721433, "memory(GiB)": 73.83, "step": 82690, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.454533 }, { "epoch": 3.5429073304485668, "grad_norm": 1.5300134420394897, "learning_rate": 1.9531015583279143e-05, "loss": 0.19076429605484008, "memory(GiB)": 73.83, "step": 82695, "token_acc": 0.9648093841642229, "train_speed(iter/s)": 1.45454 }, { "epoch": 3.543121545777816, "grad_norm": 7.345556259155273, "learning_rate": 1.9525679966585703e-05, "loss": 0.403255558013916, "memory(GiB)": 73.83, "step": 82700, "token_acc": 0.928125, "train_speed(iter/s)": 1.454542 }, { "epoch": 3.543335761107065, "grad_norm": 3.318521738052368, "learning_rate": 1.9520344901963177e-05, "loss": 0.31938719749450684, "memory(GiB)": 73.83, "step": 82705, "token_acc": 0.9287749287749287, "train_speed(iter/s)": 1.454552 }, { "epoch": 3.5435499764363136, "grad_norm": 2.394274950027466, "learning_rate": 1.951501038950819e-05, "loss": 0.3487586975097656, "memory(GiB)": 73.83, "step": 82710, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.454552 }, { "epoch": 3.543764191765563, "grad_norm": 1.366603970527649, "learning_rate": 1.9509676429317397e-05, "loss": 0.2715928077697754, "memory(GiB)": 73.83, "step": 82715, "token_acc": 0.9367088607594937, "train_speed(iter/s)": 1.454551 }, { "epoch": 3.5439784070948117, "grad_norm": 2.5646867752075195, "learning_rate": 1.950434302148741e-05, "loss": 0.4729043960571289, "memory(GiB)": 73.83, "step": 82720, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.454565 }, { "epoch": 3.5441926224240605, "grad_norm": 0.23017120361328125, "learning_rate": 1.9499010166114838e-05, "loss": 0.08979672193527222, "memory(GiB)": 73.83, "step": 82725, "token_acc": 0.9791666666666666, "train_speed(iter/s)": 1.454573 }, { "epoch": 3.54440683775331, "grad_norm": 4.1108317375183105, "learning_rate": 1.9493677863296332e-05, "loss": 0.4562382698059082, "memory(GiB)": 73.83, "step": 82730, "token_acc": 0.9006024096385542, "train_speed(iter/s)": 1.454583 }, { "epoch": 3.5446210530825586, "grad_norm": 5.117690563201904, "learning_rate": 1.948834611312847e-05, "loss": 0.47574405670166015, "memory(GiB)": 73.83, "step": 82735, "token_acc": 0.8811188811188811, "train_speed(iter/s)": 1.454593 }, { "epoch": 3.5448352684118074, "grad_norm": 1.8388948440551758, "learning_rate": 1.9483014915707832e-05, "loss": 0.2162003993988037, "memory(GiB)": 73.83, "step": 82740, "token_acc": 0.9476190476190476, "train_speed(iter/s)": 1.45459 }, { "epoch": 3.5450494837410567, "grad_norm": 0.16069647669792175, "learning_rate": 1.9477684271130995e-05, "loss": 0.10869067907333374, "memory(GiB)": 73.83, "step": 82745, "token_acc": 0.9633699633699634, "train_speed(iter/s)": 1.454589 }, { "epoch": 3.5452636990703055, "grad_norm": 2.6425490379333496, "learning_rate": 1.9472354179494556e-05, "loss": 0.6520006656646729, "memory(GiB)": 73.83, "step": 82750, "token_acc": 0.8634686346863468, "train_speed(iter/s)": 1.45459 }, { "epoch": 3.5454779143995543, "grad_norm": 0.19709831476211548, "learning_rate": 1.9467024640895053e-05, "loss": 0.28069376945495605, "memory(GiB)": 73.83, "step": 82755, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.454594 }, { "epoch": 3.5456921297288035, "grad_norm": 5.297754764556885, "learning_rate": 1.946169565542903e-05, "loss": 0.4118873119354248, "memory(GiB)": 73.83, "step": 82760, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.454592 }, { "epoch": 3.5459063450580524, "grad_norm": 4.119978427886963, "learning_rate": 1.9456367223193055e-05, "loss": 0.3134270191192627, "memory(GiB)": 73.83, "step": 82765, "token_acc": 0.9247648902821317, "train_speed(iter/s)": 1.454597 }, { "epoch": 3.546120560387301, "grad_norm": 4.2476372718811035, "learning_rate": 1.945103934428364e-05, "loss": 0.3530858278274536, "memory(GiB)": 73.83, "step": 82770, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.454597 }, { "epoch": 3.5463347757165504, "grad_norm": 2.5471556186676025, "learning_rate": 1.9445712018797296e-05, "loss": 0.20343732833862305, "memory(GiB)": 73.83, "step": 82775, "token_acc": 0.9509433962264151, "train_speed(iter/s)": 1.454594 }, { "epoch": 3.5465489910457992, "grad_norm": 1.0826584100723267, "learning_rate": 1.944038524683055e-05, "loss": 0.47513432502746583, "memory(GiB)": 73.83, "step": 82780, "token_acc": 0.8688046647230321, "train_speed(iter/s)": 1.454597 }, { "epoch": 3.546763206375048, "grad_norm": 0.47294384241104126, "learning_rate": 1.9435059028479874e-05, "loss": 0.3533350944519043, "memory(GiB)": 73.83, "step": 82785, "token_acc": 0.9156118143459916, "train_speed(iter/s)": 1.454598 }, { "epoch": 3.5469774217042973, "grad_norm": 1.9201053380966187, "learning_rate": 1.942973336384179e-05, "loss": 0.5789191246032714, "memory(GiB)": 73.83, "step": 82790, "token_acc": 0.8986486486486487, "train_speed(iter/s)": 1.454603 }, { "epoch": 3.547191637033546, "grad_norm": 1.910318374633789, "learning_rate": 1.9424408253012765e-05, "loss": 0.28914432525634765, "memory(GiB)": 73.83, "step": 82795, "token_acc": 0.9025270758122743, "train_speed(iter/s)": 1.454603 }, { "epoch": 3.547405852362795, "grad_norm": 8.661239624023438, "learning_rate": 1.941908369608927e-05, "loss": 0.25398266315460205, "memory(GiB)": 73.83, "step": 82800, "token_acc": 0.9360902255639098, "train_speed(iter/s)": 1.454619 }, { "epoch": 3.547620067692044, "grad_norm": 4.250175952911377, "learning_rate": 1.9413759693167754e-05, "loss": 0.29757013320922854, "memory(GiB)": 73.83, "step": 82805, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.454621 }, { "epoch": 3.547834283021293, "grad_norm": 2.251918315887451, "learning_rate": 1.940843624434468e-05, "loss": 0.2865839242935181, "memory(GiB)": 73.83, "step": 82810, "token_acc": 0.939209726443769, "train_speed(iter/s)": 1.454623 }, { "epoch": 3.548048498350542, "grad_norm": 12.606675148010254, "learning_rate": 1.940311334971646e-05, "loss": 0.3350926160812378, "memory(GiB)": 73.83, "step": 82815, "token_acc": 0.9375, "train_speed(iter/s)": 1.454633 }, { "epoch": 3.548262713679791, "grad_norm": 1.340903878211975, "learning_rate": 1.9397791009379546e-05, "loss": 0.2777761936187744, "memory(GiB)": 73.83, "step": 82820, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.454646 }, { "epoch": 3.54847692900904, "grad_norm": 1.7114394903182983, "learning_rate": 1.939246922343037e-05, "loss": 0.24206264019012452, "memory(GiB)": 73.83, "step": 82825, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.454646 }, { "epoch": 3.5486911443382887, "grad_norm": 2.013950824737549, "learning_rate": 1.9387147991965327e-05, "loss": 0.29332928657531737, "memory(GiB)": 73.83, "step": 82830, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.454652 }, { "epoch": 3.548905359667538, "grad_norm": 3.9970319271087646, "learning_rate": 1.9381827315080815e-05, "loss": 0.3624089002609253, "memory(GiB)": 73.83, "step": 82835, "token_acc": 0.9250720461095101, "train_speed(iter/s)": 1.454652 }, { "epoch": 3.5491195749967868, "grad_norm": 0.9062240719795227, "learning_rate": 1.9376507192873223e-05, "loss": 0.19256811141967772, "memory(GiB)": 73.83, "step": 82840, "token_acc": 0.946969696969697, "train_speed(iter/s)": 1.454657 }, { "epoch": 3.5493337903260356, "grad_norm": 4.381115436553955, "learning_rate": 1.9371187625438913e-05, "loss": 0.328546404838562, "memory(GiB)": 73.83, "step": 82845, "token_acc": 0.9298780487804879, "train_speed(iter/s)": 1.454666 }, { "epoch": 3.549548005655285, "grad_norm": 1.9053192138671875, "learning_rate": 1.9365868612874287e-05, "loss": 0.5042897701263428, "memory(GiB)": 73.83, "step": 82850, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.45467 }, { "epoch": 3.5497622209845336, "grad_norm": 5.121315002441406, "learning_rate": 1.9360550155275693e-05, "loss": 0.6331744194030762, "memory(GiB)": 73.83, "step": 82855, "token_acc": 0.8703703703703703, "train_speed(iter/s)": 1.454685 }, { "epoch": 3.5499764363137825, "grad_norm": 5.05047082901001, "learning_rate": 1.9355232252739474e-05, "loss": 0.6306487083435058, "memory(GiB)": 73.83, "step": 82860, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.454695 }, { "epoch": 3.5501906516430317, "grad_norm": 3.516031503677368, "learning_rate": 1.9349914905361964e-05, "loss": 0.26174657344818114, "memory(GiB)": 73.83, "step": 82865, "token_acc": 0.9389312977099237, "train_speed(iter/s)": 1.454697 }, { "epoch": 3.5504048669722805, "grad_norm": 1.3436590433120728, "learning_rate": 1.9344598113239488e-05, "loss": 0.32530989646911623, "memory(GiB)": 73.83, "step": 82870, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.4547 }, { "epoch": 3.5506190823015293, "grad_norm": 4.4822893142700195, "learning_rate": 1.933928187646839e-05, "loss": 1.1711520195007323, "memory(GiB)": 73.83, "step": 82875, "token_acc": 0.8364779874213837, "train_speed(iter/s)": 1.454698 }, { "epoch": 3.5508332976307786, "grad_norm": 5.5832319259643555, "learning_rate": 1.9333966195144958e-05, "loss": 0.4131014347076416, "memory(GiB)": 73.83, "step": 82880, "token_acc": 0.926530612244898, "train_speed(iter/s)": 1.454702 }, { "epoch": 3.5510475129600274, "grad_norm": 5.0093584060668945, "learning_rate": 1.93286510693655e-05, "loss": 0.3403113603591919, "memory(GiB)": 73.83, "step": 82885, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.454706 }, { "epoch": 3.551261728289276, "grad_norm": 1.2177108526229858, "learning_rate": 1.93233364992263e-05, "loss": 0.5297760009765625, "memory(GiB)": 73.83, "step": 82890, "token_acc": 0.9040247678018576, "train_speed(iter/s)": 1.454698 }, { "epoch": 3.5514759436185255, "grad_norm": 3.495889902114868, "learning_rate": 1.9318022484823618e-05, "loss": 0.34728469848632815, "memory(GiB)": 73.83, "step": 82895, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.454703 }, { "epoch": 3.5516901589477743, "grad_norm": 4.941993236541748, "learning_rate": 1.9312709026253756e-05, "loss": 0.33500194549560547, "memory(GiB)": 73.83, "step": 82900, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.454706 }, { "epoch": 3.551904374277023, "grad_norm": 2.5807418823242188, "learning_rate": 1.9307396123612942e-05, "loss": 0.23053793907165526, "memory(GiB)": 73.83, "step": 82905, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.454717 }, { "epoch": 3.5521185896062724, "grad_norm": 4.91776704788208, "learning_rate": 1.9302083776997454e-05, "loss": 0.36643271446228026, "memory(GiB)": 73.83, "step": 82910, "token_acc": 0.9227941176470589, "train_speed(iter/s)": 1.45473 }, { "epoch": 3.552332804935521, "grad_norm": 4.1877007484436035, "learning_rate": 1.929677198650352e-05, "loss": 0.28546738624572754, "memory(GiB)": 73.83, "step": 82915, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.454739 }, { "epoch": 3.55254702026477, "grad_norm": 2.09858775138855, "learning_rate": 1.929146075222736e-05, "loss": 0.2902543306350708, "memory(GiB)": 73.83, "step": 82920, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.454743 }, { "epoch": 3.5527612355940192, "grad_norm": 2.036283493041992, "learning_rate": 1.9286150074265198e-05, "loss": 0.17223151922225952, "memory(GiB)": 73.83, "step": 82925, "token_acc": 0.9556313993174061, "train_speed(iter/s)": 1.454751 }, { "epoch": 3.552975450923268, "grad_norm": 3.284486770629883, "learning_rate": 1.928083995271322e-05, "loss": 0.2189486026763916, "memory(GiB)": 73.83, "step": 82930, "token_acc": 0.9539007092198581, "train_speed(iter/s)": 1.454756 }, { "epoch": 3.553189666252517, "grad_norm": 2.149376153945923, "learning_rate": 1.9275530387667655e-05, "loss": 0.2705267667770386, "memory(GiB)": 73.83, "step": 82935, "token_acc": 0.943342776203966, "train_speed(iter/s)": 1.454758 }, { "epoch": 3.553403881581766, "grad_norm": 2.4914627075195312, "learning_rate": 1.927022137922469e-05, "loss": 0.3102184534072876, "memory(GiB)": 73.83, "step": 82940, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.454772 }, { "epoch": 3.553618096911015, "grad_norm": 3.8765177726745605, "learning_rate": 1.9264912927480483e-05, "loss": 0.31335935592651365, "memory(GiB)": 73.83, "step": 82945, "token_acc": 0.9335260115606936, "train_speed(iter/s)": 1.454784 }, { "epoch": 3.5538323122402637, "grad_norm": 4.67002010345459, "learning_rate": 1.9259605032531213e-05, "loss": 0.4019003391265869, "memory(GiB)": 73.83, "step": 82950, "token_acc": 0.8952095808383234, "train_speed(iter/s)": 1.454795 }, { "epoch": 3.554046527569513, "grad_norm": 7.0305047035217285, "learning_rate": 1.9254297694473038e-05, "loss": 0.36602885723114015, "memory(GiB)": 73.83, "step": 82955, "token_acc": 0.9087301587301587, "train_speed(iter/s)": 1.454808 }, { "epoch": 3.554260742898762, "grad_norm": 4.073988437652588, "learning_rate": 1.9248990913402083e-05, "loss": 0.3614218235015869, "memory(GiB)": 73.83, "step": 82960, "token_acc": 0.9173553719008265, "train_speed(iter/s)": 1.454821 }, { "epoch": 3.5544749582280106, "grad_norm": 3.0530757904052734, "learning_rate": 1.9243684689414498e-05, "loss": 0.1022728443145752, "memory(GiB)": 73.83, "step": 82965, "token_acc": 0.975177304964539, "train_speed(iter/s)": 1.454821 }, { "epoch": 3.55468917355726, "grad_norm": 2.7092041969299316, "learning_rate": 1.923837902260644e-05, "loss": 0.3971680164337158, "memory(GiB)": 73.83, "step": 82970, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.454827 }, { "epoch": 3.5549033888865087, "grad_norm": 2.406215190887451, "learning_rate": 1.9233073913074002e-05, "loss": 0.296903395652771, "memory(GiB)": 73.83, "step": 82975, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.454831 }, { "epoch": 3.5551176042157575, "grad_norm": 2.089139938354492, "learning_rate": 1.9227769360913296e-05, "loss": 0.18735543489456177, "memory(GiB)": 73.83, "step": 82980, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.454834 }, { "epoch": 3.5553318195450068, "grad_norm": 1.3975841999053955, "learning_rate": 1.9222465366220406e-05, "loss": 0.2675034046173096, "memory(GiB)": 73.83, "step": 82985, "token_acc": 0.9404255319148936, "train_speed(iter/s)": 1.454837 }, { "epoch": 3.5555460348742556, "grad_norm": 5.6203413009643555, "learning_rate": 1.9217161929091415e-05, "loss": 0.41572394371032717, "memory(GiB)": 73.83, "step": 82990, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.454844 }, { "epoch": 3.5557602502035044, "grad_norm": 4.204762935638428, "learning_rate": 1.921185904962243e-05, "loss": 0.840580940246582, "memory(GiB)": 73.83, "step": 82995, "token_acc": 0.8346456692913385, "train_speed(iter/s)": 1.454842 }, { "epoch": 3.5559744655327536, "grad_norm": 2.863253593444824, "learning_rate": 1.9206556727909496e-05, "loss": 0.19037904739379882, "memory(GiB)": 73.83, "step": 83000, "token_acc": 0.9594594594594594, "train_speed(iter/s)": 1.454844 }, { "epoch": 3.5559744655327536, "eval_loss": 2.273937463760376, "eval_runtime": 10.8737, "eval_samples_per_second": 9.197, "eval_steps_per_second": 9.197, "eval_token_acc": 0.4667609618104668, "step": 83000 }, { "epoch": 3.5561886808620025, "grad_norm": 4.325973033905029, "learning_rate": 1.9201254964048675e-05, "loss": 0.6652445316314697, "memory(GiB)": 73.83, "step": 83005, "token_acc": 0.597, "train_speed(iter/s)": 1.454551 }, { "epoch": 3.5564028961912513, "grad_norm": 0.7960842251777649, "learning_rate": 1.9195953758136015e-05, "loss": 0.30016422271728516, "memory(GiB)": 73.83, "step": 83010, "token_acc": 0.9182879377431906, "train_speed(iter/s)": 1.454565 }, { "epoch": 3.5566171115205005, "grad_norm": 1.838743805885315, "learning_rate": 1.9190653110267532e-05, "loss": 0.25514869689941405, "memory(GiB)": 73.83, "step": 83015, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.454569 }, { "epoch": 3.5568313268497493, "grad_norm": 1.0332485437393188, "learning_rate": 1.9185353020539288e-05, "loss": 0.48401570320129395, "memory(GiB)": 73.83, "step": 83020, "token_acc": 0.9128440366972477, "train_speed(iter/s)": 1.454574 }, { "epoch": 3.557045542178998, "grad_norm": 4.520545959472656, "learning_rate": 1.9180053489047273e-05, "loss": 0.6448479652404785, "memory(GiB)": 73.83, "step": 83025, "token_acc": 0.8598382749326146, "train_speed(iter/s)": 1.454597 }, { "epoch": 3.5572597575082474, "grad_norm": 2.200951337814331, "learning_rate": 1.9174754515887506e-05, "loss": 0.4420962333679199, "memory(GiB)": 73.83, "step": 83030, "token_acc": 0.9005681818181818, "train_speed(iter/s)": 1.454611 }, { "epoch": 3.5574739728374962, "grad_norm": 3.6328885555267334, "learning_rate": 1.9169456101155974e-05, "loss": 0.40580196380615235, "memory(GiB)": 73.83, "step": 83035, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.454612 }, { "epoch": 3.557688188166745, "grad_norm": 0.32057109475135803, "learning_rate": 1.9164158244948654e-05, "loss": 0.2165534019470215, "memory(GiB)": 73.83, "step": 83040, "token_acc": 0.9594594594594594, "train_speed(iter/s)": 1.454612 }, { "epoch": 3.5579024034959943, "grad_norm": 5.127159595489502, "learning_rate": 1.9158860947361545e-05, "loss": 0.5256817817687989, "memory(GiB)": 73.83, "step": 83045, "token_acc": 0.8843537414965986, "train_speed(iter/s)": 1.454614 }, { "epoch": 3.558116618825243, "grad_norm": 0.6617876887321472, "learning_rate": 1.9153564208490588e-05, "loss": 0.3177895784378052, "memory(GiB)": 73.83, "step": 83050, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.454629 }, { "epoch": 3.558330834154492, "grad_norm": 0.13228046894073486, "learning_rate": 1.9148268028431772e-05, "loss": 0.20580461025238037, "memory(GiB)": 73.83, "step": 83055, "token_acc": 0.9418604651162791, "train_speed(iter/s)": 1.454632 }, { "epoch": 3.558545049483741, "grad_norm": 3.1403517723083496, "learning_rate": 1.9142972407281017e-05, "loss": 0.18877851963043213, "memory(GiB)": 73.83, "step": 83060, "token_acc": 0.9611307420494699, "train_speed(iter/s)": 1.454635 }, { "epoch": 3.55875926481299, "grad_norm": 1.4625861644744873, "learning_rate": 1.9137677345134262e-05, "loss": 0.18824415206909179, "memory(GiB)": 73.83, "step": 83065, "token_acc": 0.955719557195572, "train_speed(iter/s)": 1.454641 }, { "epoch": 3.558973480142239, "grad_norm": 4.848532676696777, "learning_rate": 1.9132382842087438e-05, "loss": 0.7871330261230469, "memory(GiB)": 73.83, "step": 83070, "token_acc": 0.8527397260273972, "train_speed(iter/s)": 1.454638 }, { "epoch": 3.559187695471488, "grad_norm": 7.5791335105896, "learning_rate": 1.9127088898236434e-05, "loss": 0.47654151916503906, "memory(GiB)": 73.83, "step": 83075, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.454638 }, { "epoch": 3.559401910800737, "grad_norm": 5.2572760581970215, "learning_rate": 1.912179551367719e-05, "loss": 0.2736239433288574, "memory(GiB)": 73.83, "step": 83080, "token_acc": 0.936, "train_speed(iter/s)": 1.454637 }, { "epoch": 3.5596161261299857, "grad_norm": 1.441311001777649, "learning_rate": 1.9116502688505593e-05, "loss": 0.2443466901779175, "memory(GiB)": 73.83, "step": 83085, "token_acc": 0.9540983606557377, "train_speed(iter/s)": 1.454642 }, { "epoch": 3.559830341459235, "grad_norm": 6.587320327758789, "learning_rate": 1.911121042281752e-05, "loss": 0.5177589416503906, "memory(GiB)": 73.83, "step": 83090, "token_acc": 0.9144736842105263, "train_speed(iter/s)": 1.454647 }, { "epoch": 3.5600445567884837, "grad_norm": 3.483332872390747, "learning_rate": 1.9105918716708843e-05, "loss": 0.27240495681762694, "memory(GiB)": 73.83, "step": 83095, "token_acc": 0.918429003021148, "train_speed(iter/s)": 1.454659 }, { "epoch": 3.5602587721177326, "grad_norm": 1.44533109664917, "learning_rate": 1.910062757027543e-05, "loss": 0.30533835887908933, "memory(GiB)": 73.83, "step": 83100, "token_acc": 0.9331210191082803, "train_speed(iter/s)": 1.454665 }, { "epoch": 3.560472987446982, "grad_norm": 1.771960735321045, "learning_rate": 1.9095336983613122e-05, "loss": 0.13405508995056153, "memory(GiB)": 73.83, "step": 83105, "token_acc": 0.968503937007874, "train_speed(iter/s)": 1.45467 }, { "epoch": 3.5606872027762306, "grad_norm": 4.657987594604492, "learning_rate": 1.9090046956817786e-05, "loss": 0.11787590980529786, "memory(GiB)": 73.83, "step": 83110, "token_acc": 0.9672727272727273, "train_speed(iter/s)": 1.454674 }, { "epoch": 3.5609014181054794, "grad_norm": 3.872509002685547, "learning_rate": 1.908475748998523e-05, "loss": 0.36898603439331057, "memory(GiB)": 73.83, "step": 83115, "token_acc": 0.951310861423221, "train_speed(iter/s)": 1.454673 }, { "epoch": 3.5611156334347287, "grad_norm": 2.3780064582824707, "learning_rate": 1.9079468583211314e-05, "loss": 0.32010855674743655, "memory(GiB)": 73.83, "step": 83120, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.454678 }, { "epoch": 3.5613298487639775, "grad_norm": 2.199603796005249, "learning_rate": 1.9074180236591826e-05, "loss": 0.18745445013046264, "memory(GiB)": 73.83, "step": 83125, "token_acc": 0.9580838323353293, "train_speed(iter/s)": 1.454685 }, { "epoch": 3.5615440640932263, "grad_norm": 4.034177303314209, "learning_rate": 1.9068892450222577e-05, "loss": 0.6678765773773193, "memory(GiB)": 73.83, "step": 83130, "token_acc": 0.850415512465374, "train_speed(iter/s)": 1.454705 }, { "epoch": 3.5617582794224756, "grad_norm": 6.844472408294678, "learning_rate": 1.9063605224199333e-05, "loss": 0.7936118125915528, "memory(GiB)": 73.83, "step": 83135, "token_acc": 0.8184143222506394, "train_speed(iter/s)": 1.45471 }, { "epoch": 3.5619724947517244, "grad_norm": 0.7377251386642456, "learning_rate": 1.9058318558617927e-05, "loss": 0.204569673538208, "memory(GiB)": 73.83, "step": 83140, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.454711 }, { "epoch": 3.562186710080973, "grad_norm": 3.5014166831970215, "learning_rate": 1.9053032453574098e-05, "loss": 0.3721733808517456, "memory(GiB)": 73.83, "step": 83145, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.454713 }, { "epoch": 3.5624009254102225, "grad_norm": 6.5488152503967285, "learning_rate": 1.9047746909163623e-05, "loss": 0.3844463348388672, "memory(GiB)": 73.83, "step": 83150, "token_acc": 0.9266666666666666, "train_speed(iter/s)": 1.454715 }, { "epoch": 3.5626151407394713, "grad_norm": 4.683045387268066, "learning_rate": 1.9042461925482248e-05, "loss": 0.45168204307556153, "memory(GiB)": 73.83, "step": 83155, "token_acc": 0.87890625, "train_speed(iter/s)": 1.45472 }, { "epoch": 3.56282935606872, "grad_norm": 2.0080182552337646, "learning_rate": 1.9037177502625697e-05, "loss": 0.4480120658874512, "memory(GiB)": 73.83, "step": 83160, "token_acc": 0.9125475285171103, "train_speed(iter/s)": 1.454723 }, { "epoch": 3.5630435713979693, "grad_norm": 3.441713333129883, "learning_rate": 1.9031893640689735e-05, "loss": 0.31622045040130614, "memory(GiB)": 73.83, "step": 83165, "token_acc": 0.9262295081967213, "train_speed(iter/s)": 1.454729 }, { "epoch": 3.563257786727218, "grad_norm": 2.5885865688323975, "learning_rate": 1.9026610339770074e-05, "loss": 0.32485575675964357, "memory(GiB)": 73.83, "step": 83170, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.454735 }, { "epoch": 3.563472002056467, "grad_norm": 3.747483253479004, "learning_rate": 1.902132759996242e-05, "loss": 0.29209222793579104, "memory(GiB)": 73.83, "step": 83175, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.454747 }, { "epoch": 3.5636862173857162, "grad_norm": 5.113152980804443, "learning_rate": 1.9016045421362478e-05, "loss": 0.5367871284484863, "memory(GiB)": 73.83, "step": 83180, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.454749 }, { "epoch": 3.563900432714965, "grad_norm": 1.6357795000076294, "learning_rate": 1.901076380406594e-05, "loss": 0.5205078125, "memory(GiB)": 73.83, "step": 83185, "token_acc": 0.8746081504702194, "train_speed(iter/s)": 1.454751 }, { "epoch": 3.564114648044214, "grad_norm": 1.8484914302825928, "learning_rate": 1.9005482748168468e-05, "loss": 0.3461295127868652, "memory(GiB)": 73.83, "step": 83190, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.454756 }, { "epoch": 3.564328863373463, "grad_norm": 1.899789810180664, "learning_rate": 1.900020225376575e-05, "loss": 0.4665538787841797, "memory(GiB)": 73.83, "step": 83195, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.454759 }, { "epoch": 3.564543078702712, "grad_norm": 1.2041634321212769, "learning_rate": 1.8994922320953462e-05, "loss": 0.1408810019493103, "memory(GiB)": 73.83, "step": 83200, "token_acc": 0.9516728624535316, "train_speed(iter/s)": 1.45477 }, { "epoch": 3.5647572940319607, "grad_norm": 1.863242506980896, "learning_rate": 1.8989642949827246e-05, "loss": 0.2961249351501465, "memory(GiB)": 73.83, "step": 83205, "token_acc": 0.9321533923303835, "train_speed(iter/s)": 1.45477 }, { "epoch": 3.56497150936121, "grad_norm": 2.7906134128570557, "learning_rate": 1.898436414048274e-05, "loss": 0.46633281707763674, "memory(GiB)": 73.83, "step": 83210, "token_acc": 0.9138755980861244, "train_speed(iter/s)": 1.454775 }, { "epoch": 3.565185724690459, "grad_norm": 2.6256091594696045, "learning_rate": 1.897908589301557e-05, "loss": 0.29205572605133057, "memory(GiB)": 73.83, "step": 83215, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.454786 }, { "epoch": 3.5653999400197076, "grad_norm": 3.1572585105895996, "learning_rate": 1.8973808207521344e-05, "loss": 0.5493988037109375, "memory(GiB)": 73.83, "step": 83220, "token_acc": 0.8821917808219178, "train_speed(iter/s)": 1.454794 }, { "epoch": 3.565614155348957, "grad_norm": 3.4405524730682373, "learning_rate": 1.89685310840957e-05, "loss": 0.2133331298828125, "memory(GiB)": 73.83, "step": 83225, "token_acc": 0.9373040752351097, "train_speed(iter/s)": 1.454807 }, { "epoch": 3.5658283706782057, "grad_norm": 6.040633201599121, "learning_rate": 1.8963254522834227e-05, "loss": 0.19840580224990845, "memory(GiB)": 73.83, "step": 83230, "token_acc": 0.9603174603174603, "train_speed(iter/s)": 1.454824 }, { "epoch": 3.5660425860074545, "grad_norm": 1.533953070640564, "learning_rate": 1.8957978523832514e-05, "loss": 0.42151904106140137, "memory(GiB)": 73.83, "step": 83235, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.454826 }, { "epoch": 3.5662568013367038, "grad_norm": 2.2695865631103516, "learning_rate": 1.8952703087186142e-05, "loss": 0.22348875999450685, "memory(GiB)": 73.83, "step": 83240, "token_acc": 0.9507692307692308, "train_speed(iter/s)": 1.454831 }, { "epoch": 3.5664710166659526, "grad_norm": 4.656896591186523, "learning_rate": 1.894742821299067e-05, "loss": 0.3124478816986084, "memory(GiB)": 73.83, "step": 83245, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.454837 }, { "epoch": 3.5666852319952014, "grad_norm": 1.2086801528930664, "learning_rate": 1.894215390134166e-05, "loss": 0.20404586791992188, "memory(GiB)": 73.83, "step": 83250, "token_acc": 0.9465648854961832, "train_speed(iter/s)": 1.454839 }, { "epoch": 3.5668994473244506, "grad_norm": 13.151522636413574, "learning_rate": 1.8936880152334674e-05, "loss": 0.4190983295440674, "memory(GiB)": 73.83, "step": 83255, "token_acc": 0.9014598540145985, "train_speed(iter/s)": 1.454842 }, { "epoch": 3.5671136626536994, "grad_norm": 2.7549970149993896, "learning_rate": 1.8931606966065247e-05, "loss": 0.3210273742675781, "memory(GiB)": 73.83, "step": 83260, "token_acc": 0.9398496240601504, "train_speed(iter/s)": 1.454848 }, { "epoch": 3.5673278779829483, "grad_norm": 4.723351955413818, "learning_rate": 1.8926334342628888e-05, "loss": 0.4464094161987305, "memory(GiB)": 73.83, "step": 83265, "token_acc": 0.9049586776859504, "train_speed(iter/s)": 1.45485 }, { "epoch": 3.5675420933121975, "grad_norm": 0.14952020347118378, "learning_rate": 1.892106228212115e-05, "loss": 0.1290210485458374, "memory(GiB)": 73.83, "step": 83270, "token_acc": 0.9668874172185431, "train_speed(iter/s)": 1.45485 }, { "epoch": 3.5677563086414463, "grad_norm": 1.6283111572265625, "learning_rate": 1.891579078463752e-05, "loss": 0.4963671207427979, "memory(GiB)": 73.83, "step": 83275, "token_acc": 0.9057377049180327, "train_speed(iter/s)": 1.454861 }, { "epoch": 3.567970523970695, "grad_norm": 1.8061492443084717, "learning_rate": 1.8910519850273485e-05, "loss": 0.22953457832336427, "memory(GiB)": 73.83, "step": 83280, "token_acc": 0.9584775086505191, "train_speed(iter/s)": 1.454874 }, { "epoch": 3.5681847392999444, "grad_norm": 0.691479504108429, "learning_rate": 1.8905249479124563e-05, "loss": 0.239646577835083, "memory(GiB)": 73.83, "step": 83285, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.454877 }, { "epoch": 3.568398954629193, "grad_norm": 2.9299333095550537, "learning_rate": 1.8899979671286216e-05, "loss": 0.36429145336151125, "memory(GiB)": 73.83, "step": 83290, "token_acc": 0.9114285714285715, "train_speed(iter/s)": 1.454878 }, { "epoch": 3.568613169958442, "grad_norm": 0.4936175048351288, "learning_rate": 1.8894710426853913e-05, "loss": 0.20470848083496093, "memory(GiB)": 73.83, "step": 83295, "token_acc": 0.9342105263157895, "train_speed(iter/s)": 1.45488 }, { "epoch": 3.5688273852876913, "grad_norm": 3.9269139766693115, "learning_rate": 1.888944174592311e-05, "loss": 0.6389199256896972, "memory(GiB)": 73.83, "step": 83300, "token_acc": 0.8375796178343949, "train_speed(iter/s)": 1.454889 }, { "epoch": 3.56904160061694, "grad_norm": 2.703742742538452, "learning_rate": 1.8884173628589236e-05, "loss": 0.4914207935333252, "memory(GiB)": 73.83, "step": 83305, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.454894 }, { "epoch": 3.569255815946189, "grad_norm": 3.8393967151641846, "learning_rate": 1.887890607494776e-05, "loss": 0.7071762084960938, "memory(GiB)": 73.83, "step": 83310, "token_acc": 0.852233676975945, "train_speed(iter/s)": 1.454898 }, { "epoch": 3.569470031275438, "grad_norm": 3.129276752471924, "learning_rate": 1.8873639085094097e-05, "loss": 0.21298296451568605, "memory(GiB)": 73.83, "step": 83315, "token_acc": 0.9525691699604744, "train_speed(iter/s)": 1.454902 }, { "epoch": 3.569684246604687, "grad_norm": 1.7277249097824097, "learning_rate": 1.8868372659123655e-05, "loss": 0.18405275344848632, "memory(GiB)": 73.83, "step": 83320, "token_acc": 0.9594594594594594, "train_speed(iter/s)": 1.454906 }, { "epoch": 3.569898461933936, "grad_norm": 0.7340644598007202, "learning_rate": 1.8863106797131857e-05, "loss": 0.3669852018356323, "memory(GiB)": 73.83, "step": 83325, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.454907 }, { "epoch": 3.570112677263185, "grad_norm": 4.982223987579346, "learning_rate": 1.885784149921408e-05, "loss": 0.22461464405059814, "memory(GiB)": 73.83, "step": 83330, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.454914 }, { "epoch": 3.570326892592434, "grad_norm": 2.8463690280914307, "learning_rate": 1.88525767654657e-05, "loss": 0.16648612022399903, "memory(GiB)": 73.83, "step": 83335, "token_acc": 0.9594594594594594, "train_speed(iter/s)": 1.45492 }, { "epoch": 3.5705411079216827, "grad_norm": 1.6854814291000366, "learning_rate": 1.884731259598212e-05, "loss": 0.5353155136108398, "memory(GiB)": 73.83, "step": 83340, "token_acc": 0.8938053097345132, "train_speed(iter/s)": 1.454925 }, { "epoch": 3.570755323250932, "grad_norm": 1.9441086053848267, "learning_rate": 1.8842048990858706e-05, "loss": 0.28264384269714354, "memory(GiB)": 73.83, "step": 83345, "token_acc": 0.9305019305019305, "train_speed(iter/s)": 1.454928 }, { "epoch": 3.5709695385801807, "grad_norm": 3.461916446685791, "learning_rate": 1.8836785950190804e-05, "loss": 0.18782401084899902, "memory(GiB)": 73.83, "step": 83350, "token_acc": 0.9664179104477612, "train_speed(iter/s)": 1.454937 }, { "epoch": 3.5711837539094295, "grad_norm": 4.252659797668457, "learning_rate": 1.883152347407376e-05, "loss": 0.2700216293334961, "memory(GiB)": 73.83, "step": 83355, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.454941 }, { "epoch": 3.571397969238679, "grad_norm": 2.41085147857666, "learning_rate": 1.8826261562602905e-05, "loss": 0.35955564975738524, "memory(GiB)": 73.83, "step": 83360, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.454941 }, { "epoch": 3.5716121845679276, "grad_norm": 1.3237805366516113, "learning_rate": 1.8821000215873548e-05, "loss": 0.37464520931243894, "memory(GiB)": 73.83, "step": 83365, "token_acc": 0.9154078549848943, "train_speed(iter/s)": 1.454943 }, { "epoch": 3.5718263998971764, "grad_norm": 4.138921737670898, "learning_rate": 1.8815739433981034e-05, "loss": 0.2973820447921753, "memory(GiB)": 73.83, "step": 83370, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.454951 }, { "epoch": 3.5720406152264257, "grad_norm": 5.130585670471191, "learning_rate": 1.8810479217020654e-05, "loss": 0.5985065460205078, "memory(GiB)": 73.83, "step": 83375, "token_acc": 0.8870431893687708, "train_speed(iter/s)": 1.454952 }, { "epoch": 3.5722548305556745, "grad_norm": 2.258355140686035, "learning_rate": 1.8805219565087707e-05, "loss": 0.35796661376953126, "memory(GiB)": 73.83, "step": 83380, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.454963 }, { "epoch": 3.5724690458849233, "grad_norm": 3.809922695159912, "learning_rate": 1.879996047827747e-05, "loss": 0.2661355257034302, "memory(GiB)": 73.83, "step": 83385, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.454966 }, { "epoch": 3.5726832612141726, "grad_norm": 1.8173246383666992, "learning_rate": 1.8794701956685212e-05, "loss": 0.14454957246780395, "memory(GiB)": 73.83, "step": 83390, "token_acc": 0.9702380952380952, "train_speed(iter/s)": 1.454974 }, { "epoch": 3.5728974765434214, "grad_norm": 4.507790565490723, "learning_rate": 1.878944400040618e-05, "loss": 0.25590713024139405, "memory(GiB)": 73.83, "step": 83395, "token_acc": 0.9418960244648318, "train_speed(iter/s)": 1.454977 }, { "epoch": 3.57311169187267, "grad_norm": 0.7590718865394592, "learning_rate": 1.8784186609535677e-05, "loss": 0.1475422501564026, "memory(GiB)": 73.83, "step": 83400, "token_acc": 0.9652777777777778, "train_speed(iter/s)": 1.454979 }, { "epoch": 3.5733259072019194, "grad_norm": 3.1602745056152344, "learning_rate": 1.877892978416891e-05, "loss": 0.17734873294830322, "memory(GiB)": 73.83, "step": 83405, "token_acc": 0.9576547231270358, "train_speed(iter/s)": 1.454979 }, { "epoch": 3.5735401225311683, "grad_norm": 1.4825173616409302, "learning_rate": 1.87736735244011e-05, "loss": 0.3649128437042236, "memory(GiB)": 73.83, "step": 83410, "token_acc": 0.92578125, "train_speed(iter/s)": 1.454986 }, { "epoch": 3.573754337860417, "grad_norm": 2.781432867050171, "learning_rate": 1.8768417830327507e-05, "loss": 0.43600778579711913, "memory(GiB)": 73.83, "step": 83415, "token_acc": 0.8962264150943396, "train_speed(iter/s)": 1.454988 }, { "epoch": 3.5739685531896663, "grad_norm": 5.880857944488525, "learning_rate": 1.876316270204332e-05, "loss": 0.4411581516265869, "memory(GiB)": 73.83, "step": 83420, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.455006 }, { "epoch": 3.574182768518915, "grad_norm": 8.978337287902832, "learning_rate": 1.875790813964372e-05, "loss": 0.4469489097595215, "memory(GiB)": 73.83, "step": 83425, "token_acc": 0.8972602739726028, "train_speed(iter/s)": 1.455005 }, { "epoch": 3.574396983848164, "grad_norm": 4.5266032218933105, "learning_rate": 1.8752654143223946e-05, "loss": 0.6063688278198243, "memory(GiB)": 73.83, "step": 83430, "token_acc": 0.8725099601593626, "train_speed(iter/s)": 1.454999 }, { "epoch": 3.574611199177413, "grad_norm": 1.8259265422821045, "learning_rate": 1.8747400712879154e-05, "loss": 0.324949049949646, "memory(GiB)": 73.83, "step": 83435, "token_acc": 0.9381107491856677, "train_speed(iter/s)": 1.454997 }, { "epoch": 3.574825414506662, "grad_norm": 1.7260884046554565, "learning_rate": 1.8742147848704516e-05, "loss": 0.26282343864440916, "memory(GiB)": 73.83, "step": 83440, "token_acc": 0.9514285714285714, "train_speed(iter/s)": 1.454997 }, { "epoch": 3.575039629835911, "grad_norm": 0.6976860761642456, "learning_rate": 1.873689555079519e-05, "loss": 0.30485105514526367, "memory(GiB)": 73.83, "step": 83445, "token_acc": 0.9246987951807228, "train_speed(iter/s)": 1.455004 }, { "epoch": 3.57525384516516, "grad_norm": 6.593593120574951, "learning_rate": 1.8731643819246312e-05, "loss": 0.4526062965393066, "memory(GiB)": 73.83, "step": 83450, "token_acc": 0.9051094890510949, "train_speed(iter/s)": 1.455007 }, { "epoch": 3.575468060494409, "grad_norm": 2.6501033306121826, "learning_rate": 1.8726392654153058e-05, "loss": 0.4441520690917969, "memory(GiB)": 73.83, "step": 83455, "token_acc": 0.8994252873563219, "train_speed(iter/s)": 1.455017 }, { "epoch": 3.5756822758236577, "grad_norm": 3.656818151473999, "learning_rate": 1.8721142055610536e-05, "loss": 0.3176210880279541, "memory(GiB)": 73.83, "step": 83460, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.45503 }, { "epoch": 3.575896491152907, "grad_norm": 5.11185884475708, "learning_rate": 1.8715892023713866e-05, "loss": 0.5747550010681153, "memory(GiB)": 73.83, "step": 83465, "token_acc": 0.8706293706293706, "train_speed(iter/s)": 1.455044 }, { "epoch": 3.576110706482156, "grad_norm": 2.327676296234131, "learning_rate": 1.8710642558558162e-05, "loss": 0.3058695554733276, "memory(GiB)": 73.83, "step": 83470, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.455049 }, { "epoch": 3.5763249218114046, "grad_norm": 1.1566280126571655, "learning_rate": 1.870539366023852e-05, "loss": 0.2435901403427124, "memory(GiB)": 73.83, "step": 83475, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.455049 }, { "epoch": 3.576539137140654, "grad_norm": 4.9418721199035645, "learning_rate": 1.8700145328850005e-05, "loss": 0.46068811416625977, "memory(GiB)": 73.83, "step": 83480, "token_acc": 0.9145569620253164, "train_speed(iter/s)": 1.455047 }, { "epoch": 3.5767533524699027, "grad_norm": 0.16073213517665863, "learning_rate": 1.8694897564487725e-05, "loss": 0.259464693069458, "memory(GiB)": 73.83, "step": 83485, "token_acc": 0.9466192170818505, "train_speed(iter/s)": 1.455052 }, { "epoch": 3.5769675677991515, "grad_norm": 0.6190118193626404, "learning_rate": 1.868965036724676e-05, "loss": 0.29434933662414553, "memory(GiB)": 73.83, "step": 83490, "token_acc": 0.9506172839506173, "train_speed(iter/s)": 1.455054 }, { "epoch": 3.5771817831284007, "grad_norm": 1.359618067741394, "learning_rate": 1.8684403737222144e-05, "loss": 0.31998133659362793, "memory(GiB)": 73.83, "step": 83495, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.455053 }, { "epoch": 3.5773959984576496, "grad_norm": 1.4282809495925903, "learning_rate": 1.8679157674508936e-05, "loss": 0.29971842765808104, "memory(GiB)": 73.83, "step": 83500, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.455054 }, { "epoch": 3.5773959984576496, "eval_loss": 2.3384077548980713, "eval_runtime": 11.5986, "eval_samples_per_second": 8.622, "eval_steps_per_second": 8.622, "eval_token_acc": 0.456, "step": 83500 }, { "epoch": 3.5776102137868984, "grad_norm": 3.308218002319336, "learning_rate": 1.8673912179202163e-05, "loss": 0.5179628372192383, "memory(GiB)": 73.83, "step": 83505, "token_acc": 0.5868544600938967, "train_speed(iter/s)": 1.454737 }, { "epoch": 3.5778244291161476, "grad_norm": 2.993112802505493, "learning_rate": 1.8668667251396847e-05, "loss": 0.3608139753341675, "memory(GiB)": 73.83, "step": 83510, "token_acc": 0.9127725856697819, "train_speed(iter/s)": 1.454744 }, { "epoch": 3.5780386444453964, "grad_norm": 4.1140522956848145, "learning_rate": 1.8663422891188024e-05, "loss": 0.8149113655090332, "memory(GiB)": 73.83, "step": 83515, "token_acc": 0.8540372670807453, "train_speed(iter/s)": 1.454742 }, { "epoch": 3.5782528597746452, "grad_norm": 1.3410913944244385, "learning_rate": 1.8658179098670698e-05, "loss": 0.11841206550598145, "memory(GiB)": 73.83, "step": 83520, "token_acc": 0.9681274900398407, "train_speed(iter/s)": 1.454746 }, { "epoch": 3.5784670751038945, "grad_norm": 7.435318946838379, "learning_rate": 1.8652935873939858e-05, "loss": 0.7217827796936035, "memory(GiB)": 73.83, "step": 83525, "token_acc": 0.8588709677419355, "train_speed(iter/s)": 1.454747 }, { "epoch": 3.5786812904331433, "grad_norm": 2.4547533988952637, "learning_rate": 1.864769321709049e-05, "loss": 0.552239179611206, "memory(GiB)": 73.83, "step": 83530, "token_acc": 0.8621794871794872, "train_speed(iter/s)": 1.454762 }, { "epoch": 3.578895505762392, "grad_norm": 2.090229034423828, "learning_rate": 1.864245112821757e-05, "loss": 0.3075311899185181, "memory(GiB)": 73.83, "step": 83535, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.454765 }, { "epoch": 3.5791097210916414, "grad_norm": 3.2673285007476807, "learning_rate": 1.8637209607416044e-05, "loss": 0.23045220375061035, "memory(GiB)": 73.83, "step": 83540, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.454766 }, { "epoch": 3.57932393642089, "grad_norm": 4.15147066116333, "learning_rate": 1.8631968654780906e-05, "loss": 0.3188222169876099, "memory(GiB)": 73.83, "step": 83545, "token_acc": 0.934984520123839, "train_speed(iter/s)": 1.454772 }, { "epoch": 3.579538151750139, "grad_norm": 4.070400238037109, "learning_rate": 1.8626728270407078e-05, "loss": 0.43737058639526366, "memory(GiB)": 73.83, "step": 83550, "token_acc": 0.8985074626865671, "train_speed(iter/s)": 1.454778 }, { "epoch": 3.5797523670793883, "grad_norm": 0.48707789182662964, "learning_rate": 1.86214884543895e-05, "loss": 0.2727456569671631, "memory(GiB)": 73.83, "step": 83555, "token_acc": 0.9539007092198581, "train_speed(iter/s)": 1.454779 }, { "epoch": 3.579966582408637, "grad_norm": 3.516005516052246, "learning_rate": 1.8616249206823085e-05, "loss": 0.16080327033996583, "memory(GiB)": 73.83, "step": 83560, "token_acc": 0.9575289575289575, "train_speed(iter/s)": 1.454782 }, { "epoch": 3.580180797737886, "grad_norm": 4.173983573913574, "learning_rate": 1.8611010527802763e-05, "loss": 0.2500807285308838, "memory(GiB)": 73.83, "step": 83565, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.454777 }, { "epoch": 3.580395013067135, "grad_norm": 5.255550861358643, "learning_rate": 1.8605772417423417e-05, "loss": 0.5850184440612793, "memory(GiB)": 73.83, "step": 83570, "token_acc": 0.8862068965517241, "train_speed(iter/s)": 1.454779 }, { "epoch": 3.580609228396384, "grad_norm": 0.9731529951095581, "learning_rate": 1.8600534875779967e-05, "loss": 0.17501707077026368, "memory(GiB)": 73.83, "step": 83575, "token_acc": 0.9520958083832335, "train_speed(iter/s)": 1.45478 }, { "epoch": 3.5808234437256328, "grad_norm": 2.460339069366455, "learning_rate": 1.859529790296729e-05, "loss": 0.43626961708068845, "memory(GiB)": 73.83, "step": 83580, "token_acc": 0.9016949152542373, "train_speed(iter/s)": 1.454781 }, { "epoch": 3.581037659054882, "grad_norm": 5.151998996734619, "learning_rate": 1.8590061499080246e-05, "loss": 0.5948426723480225, "memory(GiB)": 73.83, "step": 83585, "token_acc": 0.8736462093862816, "train_speed(iter/s)": 1.454783 }, { "epoch": 3.581251874384131, "grad_norm": 9.977481842041016, "learning_rate": 1.8584825664213707e-05, "loss": 0.4972344398498535, "memory(GiB)": 73.83, "step": 83590, "token_acc": 0.9145299145299145, "train_speed(iter/s)": 1.454797 }, { "epoch": 3.5814660897133797, "grad_norm": 5.1662068367004395, "learning_rate": 1.85795903984625e-05, "loss": 0.37358109951019286, "memory(GiB)": 73.83, "step": 83595, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.4548 }, { "epoch": 3.581680305042629, "grad_norm": 5.404575347900391, "learning_rate": 1.8574355701921502e-05, "loss": 0.32264080047607424, "memory(GiB)": 73.83, "step": 83600, "token_acc": 0.9291784702549575, "train_speed(iter/s)": 1.454813 }, { "epoch": 3.5818945203718777, "grad_norm": 4.094618797302246, "learning_rate": 1.8569121574685538e-05, "loss": 0.49597735404968263, "memory(GiB)": 73.83, "step": 83605, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.454827 }, { "epoch": 3.5821087357011265, "grad_norm": 2.692103385925293, "learning_rate": 1.8563888016849417e-05, "loss": 0.6652783870697021, "memory(GiB)": 73.83, "step": 83610, "token_acc": 0.8717948717948718, "train_speed(iter/s)": 1.454848 }, { "epoch": 3.582322951030376, "grad_norm": 0.24769043922424316, "learning_rate": 1.8558655028507954e-05, "loss": 0.2245013952255249, "memory(GiB)": 73.83, "step": 83615, "token_acc": 0.9342465753424658, "train_speed(iter/s)": 1.454855 }, { "epoch": 3.5825371663596246, "grad_norm": 3.0825791358947754, "learning_rate": 1.855342260975595e-05, "loss": 0.5007987022399902, "memory(GiB)": 73.83, "step": 83620, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.454859 }, { "epoch": 3.5827513816888734, "grad_norm": 3.236578941345215, "learning_rate": 1.8548190760688176e-05, "loss": 0.23279197216033937, "memory(GiB)": 73.83, "step": 83625, "token_acc": 0.9438943894389439, "train_speed(iter/s)": 1.454872 }, { "epoch": 3.5829655970181227, "grad_norm": 10.316884994506836, "learning_rate": 1.8542959481399435e-05, "loss": 0.37217006683349607, "memory(GiB)": 73.83, "step": 83630, "token_acc": 0.9386281588447654, "train_speed(iter/s)": 1.454873 }, { "epoch": 3.5831798123473715, "grad_norm": 4.722620487213135, "learning_rate": 1.8537728771984504e-05, "loss": 0.5569898128509522, "memory(GiB)": 73.83, "step": 83635, "token_acc": 0.8678414096916299, "train_speed(iter/s)": 1.454872 }, { "epoch": 3.5833940276766203, "grad_norm": 4.977457046508789, "learning_rate": 1.853249863253813e-05, "loss": 0.3246821165084839, "memory(GiB)": 73.83, "step": 83640, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.454873 }, { "epoch": 3.5836082430058696, "grad_norm": 5.414421558380127, "learning_rate": 1.8527269063155068e-05, "loss": 0.5046092987060546, "memory(GiB)": 73.83, "step": 83645, "token_acc": 0.8664688427299704, "train_speed(iter/s)": 1.454878 }, { "epoch": 3.5838224583351184, "grad_norm": 2.2027649879455566, "learning_rate": 1.8522040063930045e-05, "loss": 0.32155771255493165, "memory(GiB)": 73.83, "step": 83650, "token_acc": 0.9436201780415431, "train_speed(iter/s)": 1.454878 }, { "epoch": 3.584036673664367, "grad_norm": 3.881291151046753, "learning_rate": 1.851681163495778e-05, "loss": 0.38850693702697753, "memory(GiB)": 73.83, "step": 83655, "token_acc": 0.9155844155844156, "train_speed(iter/s)": 1.45488 }, { "epoch": 3.5842508889936164, "grad_norm": 3.1986396312713623, "learning_rate": 1.8511583776333026e-05, "loss": 0.30535039901733396, "memory(GiB)": 73.83, "step": 83660, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.454892 }, { "epoch": 3.5844651043228652, "grad_norm": 5.400013446807861, "learning_rate": 1.8506356488150467e-05, "loss": 0.28010756969451905, "memory(GiB)": 73.83, "step": 83665, "token_acc": 0.937888198757764, "train_speed(iter/s)": 1.454906 }, { "epoch": 3.584679319652114, "grad_norm": 3.4327588081359863, "learning_rate": 1.8501129770504805e-05, "loss": 0.4064549446105957, "memory(GiB)": 73.83, "step": 83670, "token_acc": 0.896414342629482, "train_speed(iter/s)": 1.454912 }, { "epoch": 3.5848935349813633, "grad_norm": 2.240985155105591, "learning_rate": 1.8495903623490722e-05, "loss": 0.20125257968902588, "memory(GiB)": 73.83, "step": 83675, "token_acc": 0.9607843137254902, "train_speed(iter/s)": 1.454918 }, { "epoch": 3.585107750310612, "grad_norm": 4.970316410064697, "learning_rate": 1.8490678047202885e-05, "loss": 0.535600996017456, "memory(GiB)": 73.83, "step": 83680, "token_acc": 0.8814102564102564, "train_speed(iter/s)": 1.454922 }, { "epoch": 3.585321965639861, "grad_norm": 2.544646978378296, "learning_rate": 1.848545304173599e-05, "loss": 0.1666674017906189, "memory(GiB)": 73.83, "step": 83685, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.454925 }, { "epoch": 3.58553618096911, "grad_norm": 2.412808895111084, "learning_rate": 1.8480228607184675e-05, "loss": 0.45598907470703126, "memory(GiB)": 73.83, "step": 83690, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.454928 }, { "epoch": 3.585750396298359, "grad_norm": 2.860337972640991, "learning_rate": 1.8475004743643587e-05, "loss": 0.2688626766204834, "memory(GiB)": 73.83, "step": 83695, "token_acc": 0.9426229508196722, "train_speed(iter/s)": 1.454929 }, { "epoch": 3.585964611627608, "grad_norm": 3.730309247970581, "learning_rate": 1.846978145120736e-05, "loss": 0.15022494792938232, "memory(GiB)": 73.83, "step": 83700, "token_acc": 0.9666666666666667, "train_speed(iter/s)": 1.454941 }, { "epoch": 3.586178826956857, "grad_norm": 3.70564866065979, "learning_rate": 1.8464558729970603e-05, "loss": 0.5120251178741455, "memory(GiB)": 73.83, "step": 83705, "token_acc": 0.869198312236287, "train_speed(iter/s)": 1.454946 }, { "epoch": 3.586393042286106, "grad_norm": 3.9638783931732178, "learning_rate": 1.845933658002797e-05, "loss": 0.5247711181640625, "memory(GiB)": 73.83, "step": 83710, "token_acc": 0.8931750741839762, "train_speed(iter/s)": 1.454966 }, { "epoch": 3.5866072576153547, "grad_norm": 4.3615641593933105, "learning_rate": 1.845411500147402e-05, "loss": 0.40490293502807617, "memory(GiB)": 73.83, "step": 83715, "token_acc": 0.9178470254957507, "train_speed(iter/s)": 1.454968 }, { "epoch": 3.586821472944604, "grad_norm": 3.4243288040161133, "learning_rate": 1.844889399440338e-05, "loss": 0.4278581142425537, "memory(GiB)": 73.83, "step": 83720, "token_acc": 0.9120879120879121, "train_speed(iter/s)": 1.45497 }, { "epoch": 3.5870356882738528, "grad_norm": 2.5954065322875977, "learning_rate": 1.8443673558910635e-05, "loss": 0.5436376571655274, "memory(GiB)": 73.83, "step": 83725, "token_acc": 0.8927444794952681, "train_speed(iter/s)": 1.454965 }, { "epoch": 3.5872499036031016, "grad_norm": 4.132730007171631, "learning_rate": 1.8438453695090342e-05, "loss": 0.3531831741333008, "memory(GiB)": 73.83, "step": 83730, "token_acc": 0.93, "train_speed(iter/s)": 1.454968 }, { "epoch": 3.587464118932351, "grad_norm": 3.5181641578674316, "learning_rate": 1.8433234403037065e-05, "loss": 0.40205774307250974, "memory(GiB)": 73.83, "step": 83735, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.45497 }, { "epoch": 3.5876783342615997, "grad_norm": 6.839803695678711, "learning_rate": 1.842801568284534e-05, "loss": 0.43111767768859866, "memory(GiB)": 73.83, "step": 83740, "token_acc": 0.9212328767123288, "train_speed(iter/s)": 1.454972 }, { "epoch": 3.5878925495908485, "grad_norm": 3.6611015796661377, "learning_rate": 1.8422797534609748e-05, "loss": 0.22689337730407716, "memory(GiB)": 73.83, "step": 83745, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.454971 }, { "epoch": 3.5881067649200977, "grad_norm": 4.713680744171143, "learning_rate": 1.8417579958424797e-05, "loss": 0.28430914878845215, "memory(GiB)": 73.83, "step": 83750, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.454971 }, { "epoch": 3.5883209802493465, "grad_norm": 3.721791982650757, "learning_rate": 1.8412362954385015e-05, "loss": 0.3749727487564087, "memory(GiB)": 73.83, "step": 83755, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.454978 }, { "epoch": 3.5885351955785953, "grad_norm": 0.7684990763664246, "learning_rate": 1.8407146522584906e-05, "loss": 0.15600347518920898, "memory(GiB)": 73.83, "step": 83760, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.454979 }, { "epoch": 3.5887494109078446, "grad_norm": 3.795145273208618, "learning_rate": 1.8401930663118983e-05, "loss": 0.39545931816101076, "memory(GiB)": 73.83, "step": 83765, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.454982 }, { "epoch": 3.5889636262370934, "grad_norm": 2.1206538677215576, "learning_rate": 1.83967153760817e-05, "loss": 0.19584736824035645, "memory(GiB)": 73.83, "step": 83770, "token_acc": 0.9456066945606695, "train_speed(iter/s)": 1.454985 }, { "epoch": 3.5891778415663422, "grad_norm": 3.023594617843628, "learning_rate": 1.8391500661567584e-05, "loss": 0.2550068378448486, "memory(GiB)": 73.83, "step": 83775, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.454988 }, { "epoch": 3.5893920568955915, "grad_norm": 4.155740261077881, "learning_rate": 1.838628651967107e-05, "loss": 0.6118322372436523, "memory(GiB)": 73.83, "step": 83780, "token_acc": 0.8651315789473685, "train_speed(iter/s)": 1.455018 }, { "epoch": 3.5896062722248403, "grad_norm": 2.649761199951172, "learning_rate": 1.838107295048665e-05, "loss": 0.19266440868377685, "memory(GiB)": 73.83, "step": 83785, "token_acc": 0.9630996309963099, "train_speed(iter/s)": 1.455021 }, { "epoch": 3.589820487554089, "grad_norm": 3.2899978160858154, "learning_rate": 1.8375859954108753e-05, "loss": 0.281846284866333, "memory(GiB)": 73.83, "step": 83790, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.455027 }, { "epoch": 3.5900347028833384, "grad_norm": 1.8669207096099854, "learning_rate": 1.8370647530631818e-05, "loss": 0.2779437780380249, "memory(GiB)": 73.83, "step": 83795, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.455038 }, { "epoch": 3.590248918212587, "grad_norm": 3.877760171890259, "learning_rate": 1.8365435680150256e-05, "loss": 0.180114483833313, "memory(GiB)": 73.83, "step": 83800, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.455037 }, { "epoch": 3.590463133541836, "grad_norm": 3.305105686187744, "learning_rate": 1.8360224402758525e-05, "loss": 0.48349785804748535, "memory(GiB)": 73.83, "step": 83805, "token_acc": 0.9068100358422939, "train_speed(iter/s)": 1.455035 }, { "epoch": 3.5906773488710853, "grad_norm": 2.419466257095337, "learning_rate": 1.8355013698551004e-05, "loss": 0.6338447570800781, "memory(GiB)": 73.83, "step": 83810, "token_acc": 0.8754208754208754, "train_speed(iter/s)": 1.455047 }, { "epoch": 3.590891564200334, "grad_norm": 3.5607404708862305, "learning_rate": 1.8349803567622094e-05, "loss": 0.24837181568145753, "memory(GiB)": 73.83, "step": 83815, "token_acc": 0.9461077844311377, "train_speed(iter/s)": 1.455062 }, { "epoch": 3.591105779529583, "grad_norm": 4.527656078338623, "learning_rate": 1.8344594010066184e-05, "loss": 0.40319271087646485, "memory(GiB)": 73.83, "step": 83820, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.455075 }, { "epoch": 3.591319994858832, "grad_norm": 3.0153157711029053, "learning_rate": 1.8339385025977635e-05, "loss": 0.4990804195404053, "memory(GiB)": 73.83, "step": 83825, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.455072 }, { "epoch": 3.591534210188081, "grad_norm": 4.788932800292969, "learning_rate": 1.833417661545084e-05, "loss": 0.48853492736816406, "memory(GiB)": 73.83, "step": 83830, "token_acc": 0.9013157894736842, "train_speed(iter/s)": 1.45508 }, { "epoch": 3.5917484255173298, "grad_norm": 2.0129573345184326, "learning_rate": 1.832896877858014e-05, "loss": 0.4999572277069092, "memory(GiB)": 73.83, "step": 83835, "token_acc": 0.8909090909090909, "train_speed(iter/s)": 1.455091 }, { "epoch": 3.591962640846579, "grad_norm": 3.7990715503692627, "learning_rate": 1.832376151545988e-05, "loss": 0.37760334014892577, "memory(GiB)": 73.83, "step": 83840, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.455088 }, { "epoch": 3.592176856175828, "grad_norm": 2.248725652694702, "learning_rate": 1.83185548261844e-05, "loss": 0.3722368717193604, "memory(GiB)": 73.83, "step": 83845, "token_acc": 0.9013157894736842, "train_speed(iter/s)": 1.455088 }, { "epoch": 3.5923910715050766, "grad_norm": 3.130156993865967, "learning_rate": 1.8313348710848018e-05, "loss": 0.3320658206939697, "memory(GiB)": 73.83, "step": 83850, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.455093 }, { "epoch": 3.592605286834326, "grad_norm": 3.2610011100769043, "learning_rate": 1.8308143169545028e-05, "loss": 0.48497848510742186, "memory(GiB)": 73.83, "step": 83855, "token_acc": 0.8767123287671232, "train_speed(iter/s)": 1.455112 }, { "epoch": 3.5928195021635747, "grad_norm": 2.5953991413116455, "learning_rate": 1.8302938202369752e-05, "loss": 0.15958253145217896, "memory(GiB)": 73.83, "step": 83860, "token_acc": 0.946875, "train_speed(iter/s)": 1.455117 }, { "epoch": 3.5930337174928235, "grad_norm": 2.5142154693603516, "learning_rate": 1.8297733809416505e-05, "loss": 0.2394260883331299, "memory(GiB)": 73.83, "step": 83865, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.455128 }, { "epoch": 3.5932479328220728, "grad_norm": 1.418036699295044, "learning_rate": 1.829252999077955e-05, "loss": 0.1978306531906128, "memory(GiB)": 73.83, "step": 83870, "token_acc": 0.9552631578947368, "train_speed(iter/s)": 1.455135 }, { "epoch": 3.5934621481513216, "grad_norm": 2.3894522190093994, "learning_rate": 1.8287326746553157e-05, "loss": 0.39829206466674805, "memory(GiB)": 73.83, "step": 83875, "token_acc": 0.9244712990936556, "train_speed(iter/s)": 1.455154 }, { "epoch": 3.5936763634805704, "grad_norm": 4.444431304931641, "learning_rate": 1.8282124076831592e-05, "loss": 0.4021416664123535, "memory(GiB)": 73.83, "step": 83880, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.455163 }, { "epoch": 3.5938905788098197, "grad_norm": 3.906766414642334, "learning_rate": 1.8276921981709077e-05, "loss": 0.21227715015411378, "memory(GiB)": 73.83, "step": 83885, "token_acc": 0.940677966101695, "train_speed(iter/s)": 1.455168 }, { "epoch": 3.5941047941390685, "grad_norm": 3.2344326972961426, "learning_rate": 1.8271720461279907e-05, "loss": 0.48596863746643065, "memory(GiB)": 73.83, "step": 83890, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.455179 }, { "epoch": 3.5943190094683173, "grad_norm": 1.0713523626327515, "learning_rate": 1.8266519515638277e-05, "loss": 0.2736168146133423, "memory(GiB)": 73.83, "step": 83895, "token_acc": 0.958904109589041, "train_speed(iter/s)": 1.455177 }, { "epoch": 3.5945332247975665, "grad_norm": 2.6600849628448486, "learning_rate": 1.826131914487842e-05, "loss": 0.25623693466186526, "memory(GiB)": 73.83, "step": 83900, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.455182 }, { "epoch": 3.5947474401268154, "grad_norm": 0.09106186777353287, "learning_rate": 1.8256119349094538e-05, "loss": 0.21782009601593016, "memory(GiB)": 73.83, "step": 83905, "token_acc": 0.948051948051948, "train_speed(iter/s)": 1.455185 }, { "epoch": 3.594961655456064, "grad_norm": 1.533737063407898, "learning_rate": 1.8250920128380826e-05, "loss": 0.11724026203155517, "memory(GiB)": 73.83, "step": 83910, "token_acc": 0.968503937007874, "train_speed(iter/s)": 1.455182 }, { "epoch": 3.5951758707853134, "grad_norm": 1.4775067567825317, "learning_rate": 1.8245721482831468e-05, "loss": 0.41606483459472654, "memory(GiB)": 73.83, "step": 83915, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.455193 }, { "epoch": 3.5953900861145622, "grad_norm": 3.3315887451171875, "learning_rate": 1.8240523412540667e-05, "loss": 0.5974349498748779, "memory(GiB)": 73.83, "step": 83920, "token_acc": 0.8664596273291926, "train_speed(iter/s)": 1.455206 }, { "epoch": 3.595604301443811, "grad_norm": 5.6490044593811035, "learning_rate": 1.823532591760258e-05, "loss": 0.49692187309265134, "memory(GiB)": 73.83, "step": 83925, "token_acc": 0.8903654485049833, "train_speed(iter/s)": 1.455211 }, { "epoch": 3.5958185167730603, "grad_norm": 2.940509080886841, "learning_rate": 1.8230128998111346e-05, "loss": 0.33269944190979006, "memory(GiB)": 73.83, "step": 83930, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.45521 }, { "epoch": 3.596032732102309, "grad_norm": 0.5419237613677979, "learning_rate": 1.8224932654161143e-05, "loss": 0.3651425838470459, "memory(GiB)": 73.83, "step": 83935, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.455224 }, { "epoch": 3.596246947431558, "grad_norm": 7.007791042327881, "learning_rate": 1.821973688584609e-05, "loss": 0.17122186422348024, "memory(GiB)": 73.83, "step": 83940, "token_acc": 0.963265306122449, "train_speed(iter/s)": 1.455226 }, { "epoch": 3.596461162760807, "grad_norm": 2.6602556705474854, "learning_rate": 1.82145416932603e-05, "loss": 0.5245487689971924, "memory(GiB)": 73.83, "step": 83945, "token_acc": 0.8860759493670886, "train_speed(iter/s)": 1.455237 }, { "epoch": 3.596675378090056, "grad_norm": 5.152325630187988, "learning_rate": 1.8209347076497924e-05, "loss": 0.6444392204284668, "memory(GiB)": 73.83, "step": 83950, "token_acc": 0.8668831168831169, "train_speed(iter/s)": 1.455232 }, { "epoch": 3.596889593419305, "grad_norm": 3.653844118118286, "learning_rate": 1.820415303565305e-05, "loss": 0.27729430198669436, "memory(GiB)": 73.83, "step": 83955, "token_acc": 0.9280205655526992, "train_speed(iter/s)": 1.455228 }, { "epoch": 3.597103808748554, "grad_norm": 2.600645065307617, "learning_rate": 1.8198959570819763e-05, "loss": 0.35165834426879883, "memory(GiB)": 73.83, "step": 83960, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.455234 }, { "epoch": 3.597318024077803, "grad_norm": 5.178638935089111, "learning_rate": 1.8193766682092155e-05, "loss": 0.46962995529174806, "memory(GiB)": 73.83, "step": 83965, "token_acc": 0.8903654485049833, "train_speed(iter/s)": 1.455244 }, { "epoch": 3.597532239407052, "grad_norm": 2.933964729309082, "learning_rate": 1.8188574369564293e-05, "loss": 0.40755791664123536, "memory(GiB)": 73.83, "step": 83970, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.45525 }, { "epoch": 3.597746454736301, "grad_norm": 3.322059154510498, "learning_rate": 1.8183382633330255e-05, "loss": 0.3220839023590088, "memory(GiB)": 73.83, "step": 83975, "token_acc": 0.9329268292682927, "train_speed(iter/s)": 1.455252 }, { "epoch": 3.5979606700655498, "grad_norm": 3.960167169570923, "learning_rate": 1.817819147348409e-05, "loss": 0.29753024578094484, "memory(GiB)": 73.83, "step": 83980, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.455255 }, { "epoch": 3.598174885394799, "grad_norm": 0.09577272832393646, "learning_rate": 1.817300089011984e-05, "loss": 0.3367322444915771, "memory(GiB)": 73.83, "step": 83985, "token_acc": 0.9385964912280702, "train_speed(iter/s)": 1.455263 }, { "epoch": 3.598389100724048, "grad_norm": 1.7502124309539795, "learning_rate": 1.816781088333153e-05, "loss": 0.28187549114227295, "memory(GiB)": 73.83, "step": 83990, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.455263 }, { "epoch": 3.5986033160532966, "grad_norm": 2.918959617614746, "learning_rate": 1.8162621453213197e-05, "loss": 0.3637899875640869, "memory(GiB)": 73.83, "step": 83995, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.455265 }, { "epoch": 3.598817531382546, "grad_norm": 3.084956645965576, "learning_rate": 1.8157432599858814e-05, "loss": 0.34767067432403564, "memory(GiB)": 73.83, "step": 84000, "token_acc": 0.9157509157509157, "train_speed(iter/s)": 1.455274 }, { "epoch": 3.598817531382546, "eval_loss": 2.3531312942504883, "eval_runtime": 11.7244, "eval_samples_per_second": 8.529, "eval_steps_per_second": 8.529, "eval_token_acc": 0.4518950437317784, "step": 84000 }, { "epoch": 3.5990317467117947, "grad_norm": 4.862483501434326, "learning_rate": 1.8152244323362416e-05, "loss": 0.25082058906555177, "memory(GiB)": 73.83, "step": 84005, "token_acc": 0.5852090032154341, "train_speed(iter/s)": 1.454966 }, { "epoch": 3.5992459620410435, "grad_norm": 2.9174506664276123, "learning_rate": 1.8147056623818004e-05, "loss": 0.41420464515686034, "memory(GiB)": 73.83, "step": 84010, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.454971 }, { "epoch": 3.599460177370293, "grad_norm": 0.3214283287525177, "learning_rate": 1.814186950131954e-05, "loss": 0.2997806310653687, "memory(GiB)": 73.83, "step": 84015, "token_acc": 0.9620853080568721, "train_speed(iter/s)": 1.454976 }, { "epoch": 3.5996743926995416, "grad_norm": 1.6118556261062622, "learning_rate": 1.8136682955960997e-05, "loss": 0.13780838251113892, "memory(GiB)": 73.83, "step": 84020, "token_acc": 0.9616858237547893, "train_speed(iter/s)": 1.454964 }, { "epoch": 3.5998886080287904, "grad_norm": 4.181699275970459, "learning_rate": 1.813149698783633e-05, "loss": 0.4362928867340088, "memory(GiB)": 73.83, "step": 84025, "token_acc": 0.9004149377593361, "train_speed(iter/s)": 1.454965 }, { "epoch": 3.6001028233580397, "grad_norm": 2.7917661666870117, "learning_rate": 1.812631159703948e-05, "loss": 0.24245057106018067, "memory(GiB)": 73.83, "step": 84030, "token_acc": 0.924924924924925, "train_speed(iter/s)": 1.454966 }, { "epoch": 3.6003170386872885, "grad_norm": 3.00593900680542, "learning_rate": 1.8121126783664406e-05, "loss": 0.31756227016448973, "memory(GiB)": 73.83, "step": 84035, "token_acc": 0.9203821656050956, "train_speed(iter/s)": 1.454966 }, { "epoch": 3.6005312540165373, "grad_norm": 0.12913979589939117, "learning_rate": 1.8115942547805025e-05, "loss": 0.26597068309783933, "memory(GiB)": 73.83, "step": 84040, "token_acc": 0.9543859649122807, "train_speed(iter/s)": 1.454966 }, { "epoch": 3.6007454693457865, "grad_norm": 5.272270202636719, "learning_rate": 1.8110758889555256e-05, "loss": 0.3613454818725586, "memory(GiB)": 73.83, "step": 84045, "token_acc": 0.8921568627450981, "train_speed(iter/s)": 1.454966 }, { "epoch": 3.6009596846750354, "grad_norm": 3.2140979766845703, "learning_rate": 1.8105575809009008e-05, "loss": 0.30244901180267336, "memory(GiB)": 73.83, "step": 84050, "token_acc": 0.9107981220657277, "train_speed(iter/s)": 1.454965 }, { "epoch": 3.601173900004284, "grad_norm": 5.081482410430908, "learning_rate": 1.8100393306260166e-05, "loss": 0.24523451328277587, "memory(GiB)": 73.83, "step": 84055, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.454963 }, { "epoch": 3.6013881153335334, "grad_norm": 3.0181641578674316, "learning_rate": 1.809521138140261e-05, "loss": 0.25851755142211913, "memory(GiB)": 73.83, "step": 84060, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.454965 }, { "epoch": 3.6016023306627822, "grad_norm": 2.1120572090148926, "learning_rate": 1.8090030034530247e-05, "loss": 0.28589372634887694, "memory(GiB)": 73.83, "step": 84065, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.454974 }, { "epoch": 3.601816545992031, "grad_norm": 4.210361003875732, "learning_rate": 1.8084849265736927e-05, "loss": 0.4907377243041992, "memory(GiB)": 73.83, "step": 84070, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.454976 }, { "epoch": 3.6020307613212803, "grad_norm": 4.671712875366211, "learning_rate": 1.807966907511648e-05, "loss": 0.23020679950714112, "memory(GiB)": 73.83, "step": 84075, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.454983 }, { "epoch": 3.602244976650529, "grad_norm": 3.0572946071624756, "learning_rate": 1.807448946276279e-05, "loss": 0.32141215801239015, "memory(GiB)": 73.83, "step": 84080, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.454999 }, { "epoch": 3.602459191979778, "grad_norm": 2.7339980602264404, "learning_rate": 1.806931042876967e-05, "loss": 0.35782608985900877, "memory(GiB)": 73.83, "step": 84085, "token_acc": 0.9186991869918699, "train_speed(iter/s)": 1.455008 }, { "epoch": 3.602673407309027, "grad_norm": 1.4496005773544312, "learning_rate": 1.806413197323093e-05, "loss": 0.3886574745178223, "memory(GiB)": 73.83, "step": 84090, "token_acc": 0.9172185430463576, "train_speed(iter/s)": 1.45501 }, { "epoch": 3.602887622638276, "grad_norm": 4.592532634735107, "learning_rate": 1.805895409624041e-05, "loss": 0.3122557640075684, "memory(GiB)": 73.83, "step": 84095, "token_acc": 0.9434523809523809, "train_speed(iter/s)": 1.455014 }, { "epoch": 3.603101837967525, "grad_norm": 3.890413761138916, "learning_rate": 1.80537767978919e-05, "loss": 0.586841344833374, "memory(GiB)": 73.83, "step": 84100, "token_acc": 0.8913738019169329, "train_speed(iter/s)": 1.455022 }, { "epoch": 3.603316053296774, "grad_norm": 2.0631563663482666, "learning_rate": 1.80486000782792e-05, "loss": 0.18741388320922853, "memory(GiB)": 73.83, "step": 84105, "token_acc": 0.9484126984126984, "train_speed(iter/s)": 1.455022 }, { "epoch": 3.603530268626023, "grad_norm": 2.6025807857513428, "learning_rate": 1.804342393749608e-05, "loss": 0.2340993881225586, "memory(GiB)": 73.83, "step": 84110, "token_acc": 0.9501779359430605, "train_speed(iter/s)": 1.455025 }, { "epoch": 3.6037444839552717, "grad_norm": 0.5295717716217041, "learning_rate": 1.803824837563629e-05, "loss": 0.2798517942428589, "memory(GiB)": 73.83, "step": 84115, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.455023 }, { "epoch": 3.603958699284521, "grad_norm": 4.162315845489502, "learning_rate": 1.803307339279363e-05, "loss": 0.2141578435897827, "memory(GiB)": 73.83, "step": 84120, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.45503 }, { "epoch": 3.6041729146137698, "grad_norm": 0.15541864931583405, "learning_rate": 1.8027898989061838e-05, "loss": 0.333713960647583, "memory(GiB)": 73.83, "step": 84125, "token_acc": 0.9236947791164659, "train_speed(iter/s)": 1.455033 }, { "epoch": 3.6043871299430186, "grad_norm": 2.355768918991089, "learning_rate": 1.8022725164534647e-05, "loss": 0.27733640670776366, "memory(GiB)": 73.83, "step": 84130, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.455032 }, { "epoch": 3.604601345272268, "grad_norm": 2.697409152984619, "learning_rate": 1.8017551919305787e-05, "loss": 0.3305061817169189, "memory(GiB)": 73.83, "step": 84135, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.455038 }, { "epoch": 3.6048155606015166, "grad_norm": 2.994265556335449, "learning_rate": 1.8012379253468975e-05, "loss": 0.33380439281463625, "memory(GiB)": 73.83, "step": 84140, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.455046 }, { "epoch": 3.6050297759307655, "grad_norm": 3.0136072635650635, "learning_rate": 1.80072071671179e-05, "loss": 0.39999823570251464, "memory(GiB)": 73.83, "step": 84145, "token_acc": 0.8972602739726028, "train_speed(iter/s)": 1.455049 }, { "epoch": 3.6052439912600147, "grad_norm": 2.614556074142456, "learning_rate": 1.8002035660346283e-05, "loss": 0.3853168964385986, "memory(GiB)": 73.83, "step": 84150, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.455047 }, { "epoch": 3.6054582065892635, "grad_norm": 1.0659188032150269, "learning_rate": 1.7996864733247822e-05, "loss": 0.2703362464904785, "memory(GiB)": 73.83, "step": 84155, "token_acc": 0.9455782312925171, "train_speed(iter/s)": 1.45505 }, { "epoch": 3.6056724219185123, "grad_norm": 3.6102335453033447, "learning_rate": 1.7991694385916176e-05, "loss": 0.38863515853881836, "memory(GiB)": 73.83, "step": 84160, "token_acc": 0.9101123595505618, "train_speed(iter/s)": 1.455046 }, { "epoch": 3.6058866372477616, "grad_norm": 2.2605018615722656, "learning_rate": 1.7986524618445017e-05, "loss": 0.24661026000976563, "memory(GiB)": 73.83, "step": 84165, "token_acc": 0.9451219512195121, "train_speed(iter/s)": 1.455051 }, { "epoch": 3.6061008525770104, "grad_norm": 5.156017780303955, "learning_rate": 1.7981355430927992e-05, "loss": 0.4328977108001709, "memory(GiB)": 73.83, "step": 84170, "token_acc": 0.9077380952380952, "train_speed(iter/s)": 1.45505 }, { "epoch": 3.606315067906259, "grad_norm": 1.7021595239639282, "learning_rate": 1.797618682345873e-05, "loss": 0.12054787874221802, "memory(GiB)": 73.83, "step": 84175, "token_acc": 0.9586206896551724, "train_speed(iter/s)": 1.455055 }, { "epoch": 3.6065292832355085, "grad_norm": 0.10237816721200943, "learning_rate": 1.7971018796130906e-05, "loss": 0.4718155384063721, "memory(GiB)": 73.83, "step": 84180, "token_acc": 0.8976377952755905, "train_speed(iter/s)": 1.455062 }, { "epoch": 3.6067434985647573, "grad_norm": 3.531283378601074, "learning_rate": 1.796585134903812e-05, "loss": 0.2466805934906006, "memory(GiB)": 73.83, "step": 84185, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.455068 }, { "epoch": 3.606957713894006, "grad_norm": 3.4534084796905518, "learning_rate": 1.7960684482273986e-05, "loss": 0.5181027889251709, "memory(GiB)": 73.83, "step": 84190, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.455068 }, { "epoch": 3.6071719292232554, "grad_norm": 7.89737606048584, "learning_rate": 1.7955518195932114e-05, "loss": 0.6232117652893067, "memory(GiB)": 73.83, "step": 84195, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.455071 }, { "epoch": 3.607386144552504, "grad_norm": 2.775392770767212, "learning_rate": 1.7950352490106083e-05, "loss": 0.539932107925415, "memory(GiB)": 73.83, "step": 84200, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.455094 }, { "epoch": 3.607600359881753, "grad_norm": 3.079920530319214, "learning_rate": 1.794518736488947e-05, "loss": 0.44122610092163084, "memory(GiB)": 73.83, "step": 84205, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.455092 }, { "epoch": 3.6078145752110022, "grad_norm": 2.0752761363983154, "learning_rate": 1.794002282037587e-05, "loss": 0.2060330867767334, "memory(GiB)": 73.83, "step": 84210, "token_acc": 0.9591078066914498, "train_speed(iter/s)": 1.455095 }, { "epoch": 3.608028790540251, "grad_norm": 3.435302734375, "learning_rate": 1.793485885665883e-05, "loss": 0.3577293634414673, "memory(GiB)": 73.83, "step": 84215, "token_acc": 0.9083665338645418, "train_speed(iter/s)": 1.455107 }, { "epoch": 3.6082430058695, "grad_norm": 1.5544615983963013, "learning_rate": 1.792969547383191e-05, "loss": 0.27386162281036375, "memory(GiB)": 73.83, "step": 84220, "token_acc": 0.9493243243243243, "train_speed(iter/s)": 1.455122 }, { "epoch": 3.608457221198749, "grad_norm": 3.081909656524658, "learning_rate": 1.792453267198862e-05, "loss": 0.2732645750045776, "memory(GiB)": 73.83, "step": 84225, "token_acc": 0.952054794520548, "train_speed(iter/s)": 1.45512 }, { "epoch": 3.608671436527998, "grad_norm": 6.09186315536499, "learning_rate": 1.7919370451222523e-05, "loss": 0.656531286239624, "memory(GiB)": 73.83, "step": 84230, "token_acc": 0.8807947019867549, "train_speed(iter/s)": 1.45514 }, { "epoch": 3.6088856518572467, "grad_norm": 3.5534493923187256, "learning_rate": 1.791420881162711e-05, "loss": 0.44430222511291506, "memory(GiB)": 73.83, "step": 84235, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.455152 }, { "epoch": 3.609099867186496, "grad_norm": 5.168217658996582, "learning_rate": 1.790904775329592e-05, "loss": 0.39095230102539064, "memory(GiB)": 73.83, "step": 84240, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.455153 }, { "epoch": 3.609314082515745, "grad_norm": 3.8494389057159424, "learning_rate": 1.7903887276322433e-05, "loss": 0.3009265184402466, "memory(GiB)": 73.83, "step": 84245, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.455152 }, { "epoch": 3.6095282978449936, "grad_norm": 3.633589029312134, "learning_rate": 1.789872738080014e-05, "loss": 0.204177188873291, "memory(GiB)": 73.83, "step": 84250, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.455154 }, { "epoch": 3.609742513174243, "grad_norm": 5.139145851135254, "learning_rate": 1.7893568066822507e-05, "loss": 0.5844810485839844, "memory(GiB)": 73.83, "step": 84255, "token_acc": 0.8804347826086957, "train_speed(iter/s)": 1.455154 }, { "epoch": 3.6099567285034917, "grad_norm": 1.3206192255020142, "learning_rate": 1.7888409334483e-05, "loss": 0.12763750553131104, "memory(GiB)": 73.83, "step": 84260, "token_acc": 0.9683098591549296, "train_speed(iter/s)": 1.455167 }, { "epoch": 3.6101709438327405, "grad_norm": 2.060940742492676, "learning_rate": 1.788325118387509e-05, "loss": 0.3594426393508911, "memory(GiB)": 73.83, "step": 84265, "token_acc": 0.9110320284697508, "train_speed(iter/s)": 1.455172 }, { "epoch": 3.6103851591619898, "grad_norm": 1.84982168674469, "learning_rate": 1.787809361509221e-05, "loss": 0.2207127809524536, "memory(GiB)": 73.83, "step": 84270, "token_acc": 0.928030303030303, "train_speed(iter/s)": 1.455182 }, { "epoch": 3.6105993744912386, "grad_norm": 1.6860064268112183, "learning_rate": 1.7872936628227804e-05, "loss": 0.4120178699493408, "memory(GiB)": 73.83, "step": 84275, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.455196 }, { "epoch": 3.6108135898204874, "grad_norm": 0.10394991189241409, "learning_rate": 1.7867780223375284e-05, "loss": 0.22043933868408203, "memory(GiB)": 73.83, "step": 84280, "token_acc": 0.9559228650137741, "train_speed(iter/s)": 1.455195 }, { "epoch": 3.6110278051497366, "grad_norm": 0.07354453951120377, "learning_rate": 1.7862624400628074e-05, "loss": 0.2711911916732788, "memory(GiB)": 73.83, "step": 84285, "token_acc": 0.915625, "train_speed(iter/s)": 1.455195 }, { "epoch": 3.6112420204789855, "grad_norm": 2.304638147354126, "learning_rate": 1.785746916007955e-05, "loss": 0.3315845251083374, "memory(GiB)": 73.83, "step": 84290, "token_acc": 0.9342105263157895, "train_speed(iter/s)": 1.455197 }, { "epoch": 3.6114562358082343, "grad_norm": 2.1741621494293213, "learning_rate": 1.785231450182314e-05, "loss": 0.19761362075805664, "memory(GiB)": 73.83, "step": 84295, "token_acc": 0.9681159420289855, "train_speed(iter/s)": 1.455197 }, { "epoch": 3.6116704511374835, "grad_norm": 5.261202335357666, "learning_rate": 1.7847160425952193e-05, "loss": 0.4102391242980957, "memory(GiB)": 73.83, "step": 84300, "token_acc": 0.9172932330827067, "train_speed(iter/s)": 1.4552 }, { "epoch": 3.6118846664667323, "grad_norm": 6.212505340576172, "learning_rate": 1.7842006932560113e-05, "loss": 0.7748819828033447, "memory(GiB)": 73.83, "step": 84305, "token_acc": 0.8403041825095057, "train_speed(iter/s)": 1.455207 }, { "epoch": 3.6120988817959816, "grad_norm": 4.025906085968018, "learning_rate": 1.7836854021740245e-05, "loss": 0.3390922784805298, "memory(GiB)": 73.83, "step": 84310, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.455211 }, { "epoch": 3.6123130971252304, "grad_norm": 0.14971007406711578, "learning_rate": 1.783170169358594e-05, "loss": 0.2111055612564087, "memory(GiB)": 73.83, "step": 84315, "token_acc": 0.9616613418530351, "train_speed(iter/s)": 1.455224 }, { "epoch": 3.6125273124544792, "grad_norm": 0.1410369873046875, "learning_rate": 1.7826549948190514e-05, "loss": 0.1648914098739624, "memory(GiB)": 73.83, "step": 84320, "token_acc": 0.9584775086505191, "train_speed(iter/s)": 1.455226 }, { "epoch": 3.6127415277837285, "grad_norm": 4.032229900360107, "learning_rate": 1.7821398785647338e-05, "loss": 0.4423069953918457, "memory(GiB)": 73.83, "step": 84325, "token_acc": 0.8855218855218855, "train_speed(iter/s)": 1.455225 }, { "epoch": 3.6129557431129773, "grad_norm": 2.270805597305298, "learning_rate": 1.7816248206049708e-05, "loss": 0.44481310844421384, "memory(GiB)": 73.83, "step": 84330, "token_acc": 0.8943661971830986, "train_speed(iter/s)": 1.455225 }, { "epoch": 3.613169958442226, "grad_norm": 0.3732922673225403, "learning_rate": 1.7811098209490927e-05, "loss": 0.280358624458313, "memory(GiB)": 73.83, "step": 84335, "token_acc": 0.9627118644067797, "train_speed(iter/s)": 1.455232 }, { "epoch": 3.6133841737714754, "grad_norm": 5.753911018371582, "learning_rate": 1.7805948796064305e-05, "loss": 0.60927734375, "memory(GiB)": 73.83, "step": 84340, "token_acc": 0.875, "train_speed(iter/s)": 1.455245 }, { "epoch": 3.613598389100724, "grad_norm": 3.927255392074585, "learning_rate": 1.780079996586312e-05, "loss": 0.3377927303314209, "memory(GiB)": 73.83, "step": 84345, "token_acc": 0.92, "train_speed(iter/s)": 1.455256 }, { "epoch": 3.613812604429973, "grad_norm": 2.5210132598876953, "learning_rate": 1.7795651718980626e-05, "loss": 0.3088147878646851, "memory(GiB)": 73.83, "step": 84350, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.455272 }, { "epoch": 3.6140268197592222, "grad_norm": 3.9322967529296875, "learning_rate": 1.779050405551013e-05, "loss": 0.5872735500335693, "memory(GiB)": 73.83, "step": 84355, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.455293 }, { "epoch": 3.614241035088471, "grad_norm": 3.9893009662628174, "learning_rate": 1.7785356975544875e-05, "loss": 0.6364948272705078, "memory(GiB)": 73.83, "step": 84360, "token_acc": 0.851063829787234, "train_speed(iter/s)": 1.455295 }, { "epoch": 3.61445525041772, "grad_norm": 5.873030662536621, "learning_rate": 1.7780210479178095e-05, "loss": 0.3870628595352173, "memory(GiB)": 73.83, "step": 84365, "token_acc": 0.9138461538461539, "train_speed(iter/s)": 1.455293 }, { "epoch": 3.614669465746969, "grad_norm": 3.2750041484832764, "learning_rate": 1.7775064566503008e-05, "loss": 0.3280526399612427, "memory(GiB)": 73.83, "step": 84370, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.455296 }, { "epoch": 3.614883681076218, "grad_norm": 3.6202187538146973, "learning_rate": 1.776991923761287e-05, "loss": 0.4508617401123047, "memory(GiB)": 73.83, "step": 84375, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.455315 }, { "epoch": 3.6150978964054667, "grad_norm": 3.1152145862579346, "learning_rate": 1.7764774492600866e-05, "loss": 0.3024734020233154, "memory(GiB)": 73.83, "step": 84380, "token_acc": 0.9543973941368078, "train_speed(iter/s)": 1.455326 }, { "epoch": 3.615312111734716, "grad_norm": 2.4500908851623535, "learning_rate": 1.775963033156023e-05, "loss": 0.4032568454742432, "memory(GiB)": 73.83, "step": 84385, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.45533 }, { "epoch": 3.615526327063965, "grad_norm": 1.5512195825576782, "learning_rate": 1.7754486754584132e-05, "loss": 0.19552197456359863, "memory(GiB)": 73.83, "step": 84390, "token_acc": 0.9647435897435898, "train_speed(iter/s)": 1.455338 }, { "epoch": 3.6157405423932136, "grad_norm": 3.827230930328369, "learning_rate": 1.7749343761765753e-05, "loss": 0.5017722606658935, "memory(GiB)": 73.83, "step": 84395, "token_acc": 0.899641577060932, "train_speed(iter/s)": 1.455345 }, { "epoch": 3.615954757722463, "grad_norm": 5.409109592437744, "learning_rate": 1.7744201353198263e-05, "loss": 0.567815113067627, "memory(GiB)": 73.83, "step": 84400, "token_acc": 0.8735632183908046, "train_speed(iter/s)": 1.455347 }, { "epoch": 3.6161689730517117, "grad_norm": 2.6805431842803955, "learning_rate": 1.7739059528974815e-05, "loss": 0.25431070327758787, "memory(GiB)": 73.83, "step": 84405, "token_acc": 0.9389067524115756, "train_speed(iter/s)": 1.45535 }, { "epoch": 3.6163831883809605, "grad_norm": 2.1323933601379395, "learning_rate": 1.7733918289188577e-05, "loss": 0.4484558582305908, "memory(GiB)": 73.83, "step": 84410, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.455363 }, { "epoch": 3.6165974037102098, "grad_norm": 4.117961883544922, "learning_rate": 1.7728777633932688e-05, "loss": 0.4076814651489258, "memory(GiB)": 73.83, "step": 84415, "token_acc": 0.9187279151943463, "train_speed(iter/s)": 1.455373 }, { "epoch": 3.6168116190394586, "grad_norm": 5.277315139770508, "learning_rate": 1.7723637563300256e-05, "loss": 0.377270770072937, "memory(GiB)": 73.83, "step": 84420, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.455376 }, { "epoch": 3.6170258343687074, "grad_norm": 5.531269550323486, "learning_rate": 1.7718498077384406e-05, "loss": 0.3211493492126465, "memory(GiB)": 73.83, "step": 84425, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.455381 }, { "epoch": 3.6172400496979566, "grad_norm": 0.13552404940128326, "learning_rate": 1.771335917627825e-05, "loss": 0.19428508281707763, "memory(GiB)": 73.83, "step": 84430, "token_acc": 0.9609929078014184, "train_speed(iter/s)": 1.455385 }, { "epoch": 3.6174542650272055, "grad_norm": 3.419173002243042, "learning_rate": 1.7708220860074858e-05, "loss": 0.3414153575897217, "memory(GiB)": 73.83, "step": 84435, "token_acc": 0.9453924914675768, "train_speed(iter/s)": 1.4554 }, { "epoch": 3.6176684803564543, "grad_norm": 6.4722089767456055, "learning_rate": 1.770308312886736e-05, "loss": 0.3270731925964355, "memory(GiB)": 73.83, "step": 84440, "token_acc": 0.9363957597173145, "train_speed(iter/s)": 1.455413 }, { "epoch": 3.6178826956857035, "grad_norm": 3.8911468982696533, "learning_rate": 1.7697945982748786e-05, "loss": 0.2569157600402832, "memory(GiB)": 73.83, "step": 84445, "token_acc": 0.9433333333333334, "train_speed(iter/s)": 1.455415 }, { "epoch": 3.6180969110149523, "grad_norm": 5.318711280822754, "learning_rate": 1.7692809421812246e-05, "loss": 0.3063901424407959, "memory(GiB)": 73.83, "step": 84450, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.455418 }, { "epoch": 3.618311126344201, "grad_norm": 7.108423233032227, "learning_rate": 1.768767344615076e-05, "loss": 0.650964879989624, "memory(GiB)": 73.83, "step": 84455, "token_acc": 0.8639240506329114, "train_speed(iter/s)": 1.455421 }, { "epoch": 3.6185253416734504, "grad_norm": 5.227898597717285, "learning_rate": 1.768253805585739e-05, "loss": 0.3530306816101074, "memory(GiB)": 73.83, "step": 84460, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.455428 }, { "epoch": 3.6187395570026992, "grad_norm": 3.4455742835998535, "learning_rate": 1.767740325102514e-05, "loss": 0.5094467163085937, "memory(GiB)": 73.83, "step": 84465, "token_acc": 0.8932584269662921, "train_speed(iter/s)": 1.455438 }, { "epoch": 3.618953772331948, "grad_norm": 2.3202943801879883, "learning_rate": 1.7672269031747065e-05, "loss": 0.23366374969482423, "memory(GiB)": 73.83, "step": 84470, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.45545 }, { "epoch": 3.6191679876611973, "grad_norm": 7.409015655517578, "learning_rate": 1.7667135398116163e-05, "loss": 0.4340362071990967, "memory(GiB)": 73.83, "step": 84475, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.455458 }, { "epoch": 3.619382202990446, "grad_norm": 3.7387142181396484, "learning_rate": 1.7662002350225437e-05, "loss": 0.3367814064025879, "memory(GiB)": 73.83, "step": 84480, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.455458 }, { "epoch": 3.619596418319695, "grad_norm": 2.6357638835906982, "learning_rate": 1.765686988816787e-05, "loss": 0.2825544595718384, "memory(GiB)": 73.83, "step": 84485, "token_acc": 0.9455128205128205, "train_speed(iter/s)": 1.455458 }, { "epoch": 3.619810633648944, "grad_norm": 2.401129722595215, "learning_rate": 1.765173801203645e-05, "loss": 0.32756991386413575, "memory(GiB)": 73.83, "step": 84490, "token_acc": 0.9367816091954023, "train_speed(iter/s)": 1.455463 }, { "epoch": 3.620024848978193, "grad_norm": 2.765345811843872, "learning_rate": 1.7646606721924126e-05, "loss": 0.29236438274383547, "memory(GiB)": 73.83, "step": 84495, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.455465 }, { "epoch": 3.620239064307442, "grad_norm": 2.988874673843384, "learning_rate": 1.764147601792388e-05, "loss": 0.3148065090179443, "memory(GiB)": 73.83, "step": 84500, "token_acc": 0.9323308270676691, "train_speed(iter/s)": 1.455479 }, { "epoch": 3.620239064307442, "eval_loss": 2.599130868911743, "eval_runtime": 12.0009, "eval_samples_per_second": 8.333, "eval_steps_per_second": 8.333, "eval_token_acc": 0.4375804375804376, "step": 84500 }, { "epoch": 3.620453279636691, "grad_norm": 2.268540859222412, "learning_rate": 1.7636345900128654e-05, "loss": 0.28068807125091555, "memory(GiB)": 73.83, "step": 84505, "token_acc": 0.5801668211306765, "train_speed(iter/s)": 1.455157 }, { "epoch": 3.62066749496594, "grad_norm": 3.708876371383667, "learning_rate": 1.763121636863138e-05, "loss": 0.46747651100158694, "memory(GiB)": 73.83, "step": 84510, "token_acc": 0.9120879120879121, "train_speed(iter/s)": 1.455168 }, { "epoch": 3.6208817102951887, "grad_norm": 3.2853832244873047, "learning_rate": 1.762608742352499e-05, "loss": 0.23694238662719727, "memory(GiB)": 73.83, "step": 84515, "token_acc": 0.9486301369863014, "train_speed(iter/s)": 1.455174 }, { "epoch": 3.621095925624438, "grad_norm": 0.40233224630355835, "learning_rate": 1.7620959064902376e-05, "loss": 0.4559582233428955, "memory(GiB)": 73.83, "step": 84520, "token_acc": 0.8685015290519877, "train_speed(iter/s)": 1.455186 }, { "epoch": 3.6213101409536868, "grad_norm": 6.409286975860596, "learning_rate": 1.7615831292856468e-05, "loss": 0.6856688499450684, "memory(GiB)": 73.83, "step": 84525, "token_acc": 0.8633333333333333, "train_speed(iter/s)": 1.455189 }, { "epoch": 3.6215243562829356, "grad_norm": 3.9631752967834473, "learning_rate": 1.7610704107480165e-05, "loss": 0.14927912950515748, "memory(GiB)": 73.83, "step": 84530, "token_acc": 0.9719298245614035, "train_speed(iter/s)": 1.455201 }, { "epoch": 3.621738571612185, "grad_norm": 3.220599412918091, "learning_rate": 1.760557750886634e-05, "loss": 0.2813973903656006, "memory(GiB)": 73.83, "step": 84535, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.455199 }, { "epoch": 3.6219527869414336, "grad_norm": 4.972516059875488, "learning_rate": 1.7600451497107873e-05, "loss": 0.36602215766906737, "memory(GiB)": 73.83, "step": 84540, "token_acc": 0.9179331306990881, "train_speed(iter/s)": 1.455201 }, { "epoch": 3.6221670022706824, "grad_norm": 4.122875213623047, "learning_rate": 1.7595326072297615e-05, "loss": 0.4107524871826172, "memory(GiB)": 73.83, "step": 84545, "token_acc": 0.9110320284697508, "train_speed(iter/s)": 1.455201 }, { "epoch": 3.6223812175999317, "grad_norm": 1.5955376625061035, "learning_rate": 1.759020123452841e-05, "loss": 0.3604423522949219, "memory(GiB)": 73.83, "step": 84550, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.455204 }, { "epoch": 3.6225954329291805, "grad_norm": 1.5362006425857544, "learning_rate": 1.758507698389313e-05, "loss": 0.41260762214660646, "memory(GiB)": 73.83, "step": 84555, "token_acc": 0.9217391304347826, "train_speed(iter/s)": 1.455204 }, { "epoch": 3.6228096482584293, "grad_norm": 1.8054417371749878, "learning_rate": 1.757995332048458e-05, "loss": 0.34844722747802737, "memory(GiB)": 73.83, "step": 84560, "token_acc": 0.9184397163120568, "train_speed(iter/s)": 1.455207 }, { "epoch": 3.6230238635876786, "grad_norm": 2.1866321563720703, "learning_rate": 1.7574830244395597e-05, "loss": 0.448900842666626, "memory(GiB)": 73.83, "step": 84565, "token_acc": 0.8933333333333333, "train_speed(iter/s)": 1.455228 }, { "epoch": 3.6232380789169274, "grad_norm": 4.603865146636963, "learning_rate": 1.756970775571898e-05, "loss": 0.3563878297805786, "memory(GiB)": 73.83, "step": 84570, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.455231 }, { "epoch": 3.623452294246176, "grad_norm": 2.96891450881958, "learning_rate": 1.7564585854547522e-05, "loss": 0.5242298603057861, "memory(GiB)": 73.83, "step": 84575, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.455231 }, { "epoch": 3.6236665095754255, "grad_norm": 10.131481170654297, "learning_rate": 1.755946454097401e-05, "loss": 0.5442590236663818, "memory(GiB)": 73.83, "step": 84580, "token_acc": 0.888235294117647, "train_speed(iter/s)": 1.455236 }, { "epoch": 3.6238807249046743, "grad_norm": 1.6663265228271484, "learning_rate": 1.755434381509124e-05, "loss": 0.17388161420822143, "memory(GiB)": 73.83, "step": 84585, "token_acc": 0.9519230769230769, "train_speed(iter/s)": 1.455242 }, { "epoch": 3.624094940233923, "grad_norm": 4.326173305511475, "learning_rate": 1.754922367699197e-05, "loss": 0.1737210750579834, "memory(GiB)": 73.83, "step": 84590, "token_acc": 0.9602649006622517, "train_speed(iter/s)": 1.455243 }, { "epoch": 3.6243091555631723, "grad_norm": 1.7945348024368286, "learning_rate": 1.754410412676894e-05, "loss": 0.2767282485961914, "memory(GiB)": 73.83, "step": 84595, "token_acc": 0.9588477366255144, "train_speed(iter/s)": 1.455242 }, { "epoch": 3.624523370892421, "grad_norm": 4.214382648468018, "learning_rate": 1.7538985164514925e-05, "loss": 0.3401796817779541, "memory(GiB)": 73.83, "step": 84600, "token_acc": 0.9118773946360154, "train_speed(iter/s)": 1.455249 }, { "epoch": 3.62473758622167, "grad_norm": 2.5410680770874023, "learning_rate": 1.7533866790322647e-05, "loss": 0.28688242435455324, "memory(GiB)": 73.83, "step": 84605, "token_acc": 0.9331210191082803, "train_speed(iter/s)": 1.455256 }, { "epoch": 3.6249518015509192, "grad_norm": 1.8872032165527344, "learning_rate": 1.7528749004284812e-05, "loss": 0.7526691913604736, "memory(GiB)": 73.83, "step": 84610, "token_acc": 0.8709677419354839, "train_speed(iter/s)": 1.455259 }, { "epoch": 3.625166016880168, "grad_norm": 3.2456727027893066, "learning_rate": 1.752363180649416e-05, "loss": 0.22249393463134765, "memory(GiB)": 73.83, "step": 84615, "token_acc": 0.9575971731448764, "train_speed(iter/s)": 1.455259 }, { "epoch": 3.625380232209417, "grad_norm": 3.2275168895721436, "learning_rate": 1.7518515197043383e-05, "loss": 0.2632605075836182, "memory(GiB)": 73.83, "step": 84620, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.455263 }, { "epoch": 3.625594447538666, "grad_norm": 3.983342409133911, "learning_rate": 1.751339917602518e-05, "loss": 0.18620814085006715, "memory(GiB)": 73.83, "step": 84625, "token_acc": 0.9635036496350365, "train_speed(iter/s)": 1.455261 }, { "epoch": 3.625808662867915, "grad_norm": 3.919196605682373, "learning_rate": 1.7508283743532223e-05, "loss": 0.2220099687576294, "memory(GiB)": 73.83, "step": 84630, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.455265 }, { "epoch": 3.6260228781971637, "grad_norm": 3.5931615829467773, "learning_rate": 1.7503168899657184e-05, "loss": 0.13175851106643677, "memory(GiB)": 73.83, "step": 84635, "token_acc": 0.9694189602446484, "train_speed(iter/s)": 1.455263 }, { "epoch": 3.626237093526413, "grad_norm": 3.554229497909546, "learning_rate": 1.749805464449271e-05, "loss": 0.3175937652587891, "memory(GiB)": 73.83, "step": 84640, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.455264 }, { "epoch": 3.626451308855662, "grad_norm": 2.942842483520508, "learning_rate": 1.7492940978131478e-05, "loss": 0.47330360412597655, "memory(GiB)": 73.83, "step": 84645, "token_acc": 0.8992537313432836, "train_speed(iter/s)": 1.455268 }, { "epoch": 3.6266655241849106, "grad_norm": 2.6412055492401123, "learning_rate": 1.748782790066612e-05, "loss": 0.36357247829437256, "memory(GiB)": 73.83, "step": 84650, "token_acc": 0.9303135888501742, "train_speed(iter/s)": 1.455279 }, { "epoch": 3.62687973951416, "grad_norm": 0.18503254652023315, "learning_rate": 1.7482715412189253e-05, "loss": 0.2262439250946045, "memory(GiB)": 73.83, "step": 84655, "token_acc": 0.9586206896551724, "train_speed(iter/s)": 1.455284 }, { "epoch": 3.6270939548434087, "grad_norm": 2.152604818344116, "learning_rate": 1.7477603512793504e-05, "loss": 0.47420392036437986, "memory(GiB)": 73.83, "step": 84660, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.455281 }, { "epoch": 3.6273081701726575, "grad_norm": 2.340676784515381, "learning_rate": 1.747249220257146e-05, "loss": 0.18244754076004027, "memory(GiB)": 73.83, "step": 84665, "token_acc": 0.9689655172413794, "train_speed(iter/s)": 1.455288 }, { "epoch": 3.6275223855019068, "grad_norm": 2.2578368186950684, "learning_rate": 1.746738148161573e-05, "loss": 0.29522769451141356, "memory(GiB)": 73.83, "step": 84670, "token_acc": 0.9405940594059405, "train_speed(iter/s)": 1.455292 }, { "epoch": 3.6277366008311556, "grad_norm": 1.2384312152862549, "learning_rate": 1.746227135001892e-05, "loss": 0.29834043979644775, "memory(GiB)": 73.83, "step": 84675, "token_acc": 0.9323308270676691, "train_speed(iter/s)": 1.455299 }, { "epoch": 3.6279508161604044, "grad_norm": 9.88237190246582, "learning_rate": 1.745716180787359e-05, "loss": 0.9224892616271972, "memory(GiB)": 73.83, "step": 84680, "token_acc": 0.8566176470588235, "train_speed(iter/s)": 1.455316 }, { "epoch": 3.6281650314896536, "grad_norm": 2.7513656616210938, "learning_rate": 1.7452052855272298e-05, "loss": 0.27963504791259763, "memory(GiB)": 73.83, "step": 84685, "token_acc": 0.9415584415584416, "train_speed(iter/s)": 1.455319 }, { "epoch": 3.6283792468189024, "grad_norm": 0.46101856231689453, "learning_rate": 1.7446944492307605e-05, "loss": 0.15969583988189698, "memory(GiB)": 73.83, "step": 84690, "token_acc": 0.9572953736654805, "train_speed(iter/s)": 1.45532 }, { "epoch": 3.6285934621481513, "grad_norm": 2.299628734588623, "learning_rate": 1.744183671907203e-05, "loss": 0.6869184494018554, "memory(GiB)": 73.83, "step": 84695, "token_acc": 0.8838709677419355, "train_speed(iter/s)": 1.455319 }, { "epoch": 3.6288076774774005, "grad_norm": 2.1364521980285645, "learning_rate": 1.743672953565814e-05, "loss": 0.17343060970306395, "memory(GiB)": 73.83, "step": 84700, "token_acc": 0.958904109589041, "train_speed(iter/s)": 1.455325 }, { "epoch": 3.6290218928066493, "grad_norm": 1.6832716464996338, "learning_rate": 1.7431622942158445e-05, "loss": 0.41300296783447266, "memory(GiB)": 73.83, "step": 84705, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.455324 }, { "epoch": 3.629236108135898, "grad_norm": 0.0870814248919487, "learning_rate": 1.742651693866545e-05, "loss": 0.28507606983184813, "memory(GiB)": 73.83, "step": 84710, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.455331 }, { "epoch": 3.6294503234651474, "grad_norm": 1.973046064376831, "learning_rate": 1.742141152527166e-05, "loss": 0.5529794692993164, "memory(GiB)": 73.83, "step": 84715, "token_acc": 0.900990099009901, "train_speed(iter/s)": 1.455352 }, { "epoch": 3.629664538794396, "grad_norm": 3.8759260177612305, "learning_rate": 1.7416306702069558e-05, "loss": 0.3662945508956909, "memory(GiB)": 73.83, "step": 84720, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.455354 }, { "epoch": 3.629878754123645, "grad_norm": 1.2035819292068481, "learning_rate": 1.7411202469151606e-05, "loss": 0.3199406385421753, "memory(GiB)": 73.83, "step": 84725, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.455353 }, { "epoch": 3.6300929694528943, "grad_norm": 3.2085936069488525, "learning_rate": 1.7406098826610313e-05, "loss": 0.5073770999908447, "memory(GiB)": 73.83, "step": 84730, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.455359 }, { "epoch": 3.630307184782143, "grad_norm": 4.693083763122559, "learning_rate": 1.7400995774538116e-05, "loss": 0.3149114608764648, "memory(GiB)": 73.83, "step": 84735, "token_acc": 0.9346153846153846, "train_speed(iter/s)": 1.455362 }, { "epoch": 3.630521400111392, "grad_norm": 2.4075093269348145, "learning_rate": 1.739589331302744e-05, "loss": 0.48406238555908204, "memory(GiB)": 73.83, "step": 84740, "token_acc": 0.8865979381443299, "train_speed(iter/s)": 1.455366 }, { "epoch": 3.630735615440641, "grad_norm": 2.5791890621185303, "learning_rate": 1.7390791442170763e-05, "loss": 0.19986716508865357, "memory(GiB)": 73.83, "step": 84745, "token_acc": 0.939873417721519, "train_speed(iter/s)": 1.455365 }, { "epoch": 3.63094983076989, "grad_norm": 1.928654670715332, "learning_rate": 1.7385690162060487e-05, "loss": 0.20792243480682374, "memory(GiB)": 73.83, "step": 84750, "token_acc": 0.9550561797752809, "train_speed(iter/s)": 1.455371 }, { "epoch": 3.631164046099139, "grad_norm": 4.148499011993408, "learning_rate": 1.7380589472789015e-05, "loss": 0.49706258773803713, "memory(GiB)": 73.83, "step": 84755, "token_acc": 0.9026845637583892, "train_speed(iter/s)": 1.455383 }, { "epoch": 3.631378261428388, "grad_norm": 1.7117925882339478, "learning_rate": 1.7375489374448776e-05, "loss": 0.6889981269836426, "memory(GiB)": 73.83, "step": 84760, "token_acc": 0.8666666666666667, "train_speed(iter/s)": 1.455385 }, { "epoch": 3.631592476757637, "grad_norm": 1.9761631488800049, "learning_rate": 1.7370389867132154e-05, "loss": 0.4737450122833252, "memory(GiB)": 73.83, "step": 84765, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.455407 }, { "epoch": 3.6318066920868857, "grad_norm": 0.7478182911872864, "learning_rate": 1.7365290950931527e-05, "loss": 0.24358510971069336, "memory(GiB)": 73.83, "step": 84770, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.45541 }, { "epoch": 3.632020907416135, "grad_norm": 2.429429769515991, "learning_rate": 1.736019262593927e-05, "loss": 0.27790539264678954, "memory(GiB)": 73.83, "step": 84775, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.455424 }, { "epoch": 3.6322351227453837, "grad_norm": 2.5810718536376953, "learning_rate": 1.7355094892247725e-05, "loss": 0.47934589385986326, "memory(GiB)": 73.83, "step": 84780, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.455431 }, { "epoch": 3.6324493380746325, "grad_norm": 1.4737153053283691, "learning_rate": 1.7349997749949277e-05, "loss": 0.366125226020813, "memory(GiB)": 73.83, "step": 84785, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.455439 }, { "epoch": 3.632663553403882, "grad_norm": 3.4532766342163086, "learning_rate": 1.734490119913625e-05, "loss": 0.5486924171447753, "memory(GiB)": 73.83, "step": 84790, "token_acc": 0.88, "train_speed(iter/s)": 1.455441 }, { "epoch": 3.6328777687331306, "grad_norm": 3.79963755607605, "learning_rate": 1.7339805239900964e-05, "loss": 0.39167869091033936, "memory(GiB)": 73.83, "step": 84795, "token_acc": 0.8933717579250721, "train_speed(iter/s)": 1.455445 }, { "epoch": 3.6330919840623794, "grad_norm": 4.724519729614258, "learning_rate": 1.7334709872335746e-05, "loss": 0.5849620819091796, "memory(GiB)": 73.83, "step": 84800, "token_acc": 0.8790560471976401, "train_speed(iter/s)": 1.45545 }, { "epoch": 3.6333061993916287, "grad_norm": 0.6091716885566711, "learning_rate": 1.7329615096532905e-05, "loss": 0.2853246212005615, "memory(GiB)": 73.83, "step": 84805, "token_acc": 0.9431818181818182, "train_speed(iter/s)": 1.455468 }, { "epoch": 3.6335204147208775, "grad_norm": 3.6296536922454834, "learning_rate": 1.7324520912584713e-05, "loss": 0.44779696464538576, "memory(GiB)": 73.83, "step": 84810, "token_acc": 0.9025270758122743, "train_speed(iter/s)": 1.455479 }, { "epoch": 3.6337346300501263, "grad_norm": 0.5422691106796265, "learning_rate": 1.731942732058348e-05, "loss": 0.28625035285949707, "memory(GiB)": 73.83, "step": 84815, "token_acc": 0.9296296296296296, "train_speed(iter/s)": 1.455478 }, { "epoch": 3.6339488453793756, "grad_norm": 1.4898202419281006, "learning_rate": 1.7314334320621494e-05, "loss": 0.2983667850494385, "memory(GiB)": 73.83, "step": 84820, "token_acc": 0.9303030303030303, "train_speed(iter/s)": 1.455485 }, { "epoch": 3.6341630607086244, "grad_norm": 1.488313913345337, "learning_rate": 1.7309241912791e-05, "loss": 0.17286795377731323, "memory(GiB)": 73.83, "step": 84825, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.45549 }, { "epoch": 3.634377276037873, "grad_norm": 3.184971332550049, "learning_rate": 1.7304150097184257e-05, "loss": 0.5806660652160645, "memory(GiB)": 73.83, "step": 84830, "token_acc": 0.8697183098591549, "train_speed(iter/s)": 1.455492 }, { "epoch": 3.6345914913671225, "grad_norm": 3.238234758377075, "learning_rate": 1.7299058873893507e-05, "loss": 0.3849640846252441, "memory(GiB)": 73.83, "step": 84835, "token_acc": 0.9153005464480874, "train_speed(iter/s)": 1.455505 }, { "epoch": 3.6348057066963713, "grad_norm": 0.6076539754867554, "learning_rate": 1.7293968243010967e-05, "loss": 0.3264890432357788, "memory(GiB)": 73.83, "step": 84840, "token_acc": 0.9371428571428572, "train_speed(iter/s)": 1.45552 }, { "epoch": 3.63501992202562, "grad_norm": 4.209481239318848, "learning_rate": 1.728887820462889e-05, "loss": 0.474809455871582, "memory(GiB)": 73.83, "step": 84845, "token_acc": 0.8758169934640523, "train_speed(iter/s)": 1.455522 }, { "epoch": 3.6352341373548693, "grad_norm": 0.05387924984097481, "learning_rate": 1.7283788758839463e-05, "loss": 0.3784687042236328, "memory(GiB)": 73.83, "step": 84850, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.455522 }, { "epoch": 3.635448352684118, "grad_norm": 3.2663023471832275, "learning_rate": 1.7278699905734897e-05, "loss": 0.3462654113769531, "memory(GiB)": 73.83, "step": 84855, "token_acc": 0.9252873563218391, "train_speed(iter/s)": 1.455526 }, { "epoch": 3.635662568013367, "grad_norm": 2.904634714126587, "learning_rate": 1.727361164540738e-05, "loss": 0.332716178894043, "memory(GiB)": 73.83, "step": 84860, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.455528 }, { "epoch": 3.635876783342616, "grad_norm": 2.59281587600708, "learning_rate": 1.7268523977949084e-05, "loss": 0.3801145076751709, "memory(GiB)": 73.83, "step": 84865, "token_acc": 0.8942857142857142, "train_speed(iter/s)": 1.45553 }, { "epoch": 3.636090998671865, "grad_norm": 1.904036045074463, "learning_rate": 1.7263436903452162e-05, "loss": 0.26854207515716555, "memory(GiB)": 73.83, "step": 84870, "token_acc": 0.9303135888501742, "train_speed(iter/s)": 1.455535 }, { "epoch": 3.636305214001114, "grad_norm": 4.159412384033203, "learning_rate": 1.7258350422008813e-05, "loss": 0.2652168035507202, "memory(GiB)": 73.83, "step": 84875, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.455537 }, { "epoch": 3.636519429330363, "grad_norm": 2.1200814247131348, "learning_rate": 1.7253264533711155e-05, "loss": 0.42333507537841797, "memory(GiB)": 73.83, "step": 84880, "token_acc": 0.909433962264151, "train_speed(iter/s)": 1.455541 }, { "epoch": 3.636733644659612, "grad_norm": 3.960738182067871, "learning_rate": 1.7248179238651334e-05, "loss": 0.5000528812408447, "memory(GiB)": 73.83, "step": 84885, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.455535 }, { "epoch": 3.6369478599888607, "grad_norm": 7.796975612640381, "learning_rate": 1.724309453692145e-05, "loss": 0.3123769760131836, "memory(GiB)": 73.83, "step": 84890, "token_acc": 0.930327868852459, "train_speed(iter/s)": 1.45554 }, { "epoch": 3.63716207531811, "grad_norm": 1.746814489364624, "learning_rate": 1.7238010428613654e-05, "loss": 0.2749701738357544, "memory(GiB)": 73.83, "step": 84895, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.455539 }, { "epoch": 3.637376290647359, "grad_norm": 2.9658570289611816, "learning_rate": 1.7232926913820015e-05, "loss": 0.3321285009384155, "memory(GiB)": 73.83, "step": 84900, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.455546 }, { "epoch": 3.6375905059766076, "grad_norm": 1.7014753818511963, "learning_rate": 1.7227843992632658e-05, "loss": 0.15717430114746095, "memory(GiB)": 73.83, "step": 84905, "token_acc": 0.9495268138801262, "train_speed(iter/s)": 1.45555 }, { "epoch": 3.637804721305857, "grad_norm": 2.4631309509277344, "learning_rate": 1.7222761665143654e-05, "loss": 0.17426137924194335, "memory(GiB)": 73.83, "step": 84910, "token_acc": 0.9536231884057971, "train_speed(iter/s)": 1.455564 }, { "epoch": 3.6380189366351057, "grad_norm": 4.583262920379639, "learning_rate": 1.7217679931445068e-05, "loss": 0.5617744445800781, "memory(GiB)": 73.83, "step": 84915, "token_acc": 0.8672839506172839, "train_speed(iter/s)": 1.455566 }, { "epoch": 3.6382331519643545, "grad_norm": 4.975739002227783, "learning_rate": 1.721259879162896e-05, "loss": 0.4750051498413086, "memory(GiB)": 73.83, "step": 84920, "token_acc": 0.9026548672566371, "train_speed(iter/s)": 1.455567 }, { "epoch": 3.6384473672936037, "grad_norm": 3.4463553428649902, "learning_rate": 1.720751824578737e-05, "loss": 0.3932312488555908, "memory(GiB)": 73.83, "step": 84925, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.455575 }, { "epoch": 3.6386615826228526, "grad_norm": 4.216175556182861, "learning_rate": 1.7202438294012363e-05, "loss": 0.4523204803466797, "memory(GiB)": 73.83, "step": 84930, "token_acc": 0.8909090909090909, "train_speed(iter/s)": 1.455581 }, { "epoch": 3.6388757979521014, "grad_norm": 2.0542876720428467, "learning_rate": 1.719735893639595e-05, "loss": 0.3252962350845337, "memory(GiB)": 73.83, "step": 84935, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.455588 }, { "epoch": 3.6390900132813506, "grad_norm": 1.1314418315887451, "learning_rate": 1.7192280173030156e-05, "loss": 0.10671091079711914, "memory(GiB)": 73.83, "step": 84940, "token_acc": 0.976, "train_speed(iter/s)": 1.45559 }, { "epoch": 3.6393042286105994, "grad_norm": 3.301194190979004, "learning_rate": 1.718720200400698e-05, "loss": 0.49462018013000486, "memory(GiB)": 73.83, "step": 84945, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.455589 }, { "epoch": 3.6395184439398482, "grad_norm": 4.0315775871276855, "learning_rate": 1.718212442941842e-05, "loss": 0.2776619911193848, "memory(GiB)": 73.83, "step": 84950, "token_acc": 0.9343065693430657, "train_speed(iter/s)": 1.455602 }, { "epoch": 3.6397326592690975, "grad_norm": 2.3925485610961914, "learning_rate": 1.7177047449356447e-05, "loss": 0.4321001052856445, "memory(GiB)": 73.83, "step": 84955, "token_acc": 0.8876811594202898, "train_speed(iter/s)": 1.455606 }, { "epoch": 3.6399468745983463, "grad_norm": 0.15649345517158508, "learning_rate": 1.7171971063913066e-05, "loss": 0.10088841915130616, "memory(GiB)": 73.83, "step": 84960, "token_acc": 0.9854545454545455, "train_speed(iter/s)": 1.455606 }, { "epoch": 3.640161089927595, "grad_norm": 1.8411109447479248, "learning_rate": 1.716689527318021e-05, "loss": 0.2164064407348633, "memory(GiB)": 73.83, "step": 84965, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.455609 }, { "epoch": 3.6403753052568444, "grad_norm": 3.215235948562622, "learning_rate": 1.7161820077249856e-05, "loss": 0.3384519100189209, "memory(GiB)": 73.83, "step": 84970, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.45562 }, { "epoch": 3.640589520586093, "grad_norm": 0.6410582661628723, "learning_rate": 1.715674547621394e-05, "loss": 0.22209951877593995, "memory(GiB)": 73.83, "step": 84975, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.455629 }, { "epoch": 3.640803735915342, "grad_norm": 3.258997678756714, "learning_rate": 1.7151671470164392e-05, "loss": 0.44397554397583006, "memory(GiB)": 73.83, "step": 84980, "token_acc": 0.9115646258503401, "train_speed(iter/s)": 1.45564 }, { "epoch": 3.6410179512445913, "grad_norm": 3.2499234676361084, "learning_rate": 1.7146598059193113e-05, "loss": 0.4746755599975586, "memory(GiB)": 73.83, "step": 84985, "token_acc": 0.9189944134078212, "train_speed(iter/s)": 1.455649 }, { "epoch": 3.64123216657384, "grad_norm": 10.539017677307129, "learning_rate": 1.714152524339204e-05, "loss": 0.31077051162719727, "memory(GiB)": 73.83, "step": 84990, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.455665 }, { "epoch": 3.641446381903089, "grad_norm": 0.46304407715797424, "learning_rate": 1.7136453022853067e-05, "loss": 0.4647937774658203, "memory(GiB)": 73.83, "step": 84995, "token_acc": 0.9128787878787878, "train_speed(iter/s)": 1.455671 }, { "epoch": 3.641660597232338, "grad_norm": 2.6787619590759277, "learning_rate": 1.7131381397668072e-05, "loss": 0.5713058471679687, "memory(GiB)": 73.83, "step": 85000, "token_acc": 0.8860759493670886, "train_speed(iter/s)": 1.455692 }, { "epoch": 3.641660597232338, "eval_loss": 2.480652093887329, "eval_runtime": 11.6774, "eval_samples_per_second": 8.564, "eval_steps_per_second": 8.564, "eval_token_acc": 0.46984924623115576, "step": 85000 }, { "epoch": 3.641874812561587, "grad_norm": 1.658840298652649, "learning_rate": 1.7126310367928937e-05, "loss": 0.24935743808746338, "memory(GiB)": 73.83, "step": 85005, "token_acc": 0.5942698706099815, "train_speed(iter/s)": 1.455384 }, { "epoch": 3.6420890278908358, "grad_norm": 3.3213610649108887, "learning_rate": 1.712123993372753e-05, "loss": 0.38069634437561034, "memory(GiB)": 73.83, "step": 85010, "token_acc": 0.9176029962546817, "train_speed(iter/s)": 1.455382 }, { "epoch": 3.642303243220085, "grad_norm": 1.8307712078094482, "learning_rate": 1.7116170095155687e-05, "loss": 0.37449705600738525, "memory(GiB)": 73.83, "step": 85015, "token_acc": 0.933933933933934, "train_speed(iter/s)": 1.455388 }, { "epoch": 3.642517458549334, "grad_norm": 3.459676504135132, "learning_rate": 1.7111100852305283e-05, "loss": 0.5411282539367676, "memory(GiB)": 73.83, "step": 85020, "token_acc": 0.869198312236287, "train_speed(iter/s)": 1.455409 }, { "epoch": 3.6427316738785827, "grad_norm": 3.0025839805603027, "learning_rate": 1.710603220526814e-05, "loss": 0.5254673957824707, "memory(GiB)": 73.83, "step": 85025, "token_acc": 0.9005847953216374, "train_speed(iter/s)": 1.455409 }, { "epoch": 3.642945889207832, "grad_norm": 2.4461143016815186, "learning_rate": 1.7100964154136086e-05, "loss": 0.36118690967559813, "memory(GiB)": 73.83, "step": 85030, "token_acc": 0.9385245901639344, "train_speed(iter/s)": 1.455412 }, { "epoch": 3.6431601045370807, "grad_norm": 0.7349397540092468, "learning_rate": 1.7095896699000908e-05, "loss": 0.3303581953048706, "memory(GiB)": 73.83, "step": 85035, "token_acc": 0.9339622641509434, "train_speed(iter/s)": 1.455426 }, { "epoch": 3.6433743198663295, "grad_norm": 2.8592042922973633, "learning_rate": 1.709082983995444e-05, "loss": 0.24393000602722167, "memory(GiB)": 73.83, "step": 85040, "token_acc": 0.9596273291925466, "train_speed(iter/s)": 1.455427 }, { "epoch": 3.643588535195579, "grad_norm": 3.1705336570739746, "learning_rate": 1.7085763577088453e-05, "loss": 0.5330641269683838, "memory(GiB)": 73.83, "step": 85045, "token_acc": 0.8741007194244604, "train_speed(iter/s)": 1.45543 }, { "epoch": 3.6438027505248276, "grad_norm": 2.4354286193847656, "learning_rate": 1.708069791049475e-05, "loss": 0.33804802894592284, "memory(GiB)": 73.83, "step": 85050, "token_acc": 0.912, "train_speed(iter/s)": 1.455433 }, { "epoch": 3.6440169658540764, "grad_norm": 4.236267566680908, "learning_rate": 1.7075632840265083e-05, "loss": 0.19280288219451905, "memory(GiB)": 73.83, "step": 85055, "token_acc": 0.9586776859504132, "train_speed(iter/s)": 1.455439 }, { "epoch": 3.6442311811833257, "grad_norm": 0.6290088295936584, "learning_rate": 1.7070568366491208e-05, "loss": 0.2019174575805664, "memory(GiB)": 73.83, "step": 85060, "token_acc": 0.953125, "train_speed(iter/s)": 1.455448 }, { "epoch": 3.6444453965125745, "grad_norm": 0.40611037611961365, "learning_rate": 1.7065504489264884e-05, "loss": 0.19389145374298095, "memory(GiB)": 73.83, "step": 85065, "token_acc": 0.9658703071672355, "train_speed(iter/s)": 1.45545 }, { "epoch": 3.6446596118418233, "grad_norm": 1.6086479425430298, "learning_rate": 1.7060441208677818e-05, "loss": 0.24945759773254395, "memory(GiB)": 73.83, "step": 85070, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.455462 }, { "epoch": 3.6448738271710726, "grad_norm": 1.7841715812683105, "learning_rate": 1.705537852482178e-05, "loss": 0.40915560722351074, "memory(GiB)": 73.83, "step": 85075, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.455465 }, { "epoch": 3.6450880425003214, "grad_norm": 2.227587938308716, "learning_rate": 1.7050316437788466e-05, "loss": 0.4381927490234375, "memory(GiB)": 73.83, "step": 85080, "token_acc": 0.8844884488448845, "train_speed(iter/s)": 1.455477 }, { "epoch": 3.64530225782957, "grad_norm": 1.1842252016067505, "learning_rate": 1.704525494766958e-05, "loss": 0.32046988010406496, "memory(GiB)": 73.83, "step": 85085, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.455481 }, { "epoch": 3.6455164731588194, "grad_norm": 0.3329131305217743, "learning_rate": 1.7040194054556813e-05, "loss": 0.24882223606109619, "memory(GiB)": 73.83, "step": 85090, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.455486 }, { "epoch": 3.6457306884880682, "grad_norm": 3.4257755279541016, "learning_rate": 1.703513375854185e-05, "loss": 0.4820141315460205, "memory(GiB)": 73.83, "step": 85095, "token_acc": 0.8932038834951457, "train_speed(iter/s)": 1.455506 }, { "epoch": 3.645944903817317, "grad_norm": 0.2610407769680023, "learning_rate": 1.7030074059716338e-05, "loss": 0.2698278665542603, "memory(GiB)": 73.83, "step": 85100, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.45551 }, { "epoch": 3.6461591191465663, "grad_norm": 3.684516191482544, "learning_rate": 1.7025014958171993e-05, "loss": 0.4262853622436523, "memory(GiB)": 73.83, "step": 85105, "token_acc": 0.90625, "train_speed(iter/s)": 1.455509 }, { "epoch": 3.646373334475815, "grad_norm": 1.4784436225891113, "learning_rate": 1.7019956454000406e-05, "loss": 0.32891390323638914, "memory(GiB)": 73.83, "step": 85110, "token_acc": 0.9188311688311688, "train_speed(iter/s)": 1.45551 }, { "epoch": 3.646587549805064, "grad_norm": 3.056138515472412, "learning_rate": 1.7014898547293266e-05, "loss": 0.3020773410797119, "memory(GiB)": 73.83, "step": 85115, "token_acc": 0.9303030303030303, "train_speed(iter/s)": 1.455517 }, { "epoch": 3.646801765134313, "grad_norm": 8.816566467285156, "learning_rate": 1.7009841238142188e-05, "loss": 0.3571493148803711, "memory(GiB)": 73.83, "step": 85120, "token_acc": 0.9433198380566802, "train_speed(iter/s)": 1.455526 }, { "epoch": 3.647015980463562, "grad_norm": 1.7757985591888428, "learning_rate": 1.7004784526638777e-05, "loss": 0.09310899376869201, "memory(GiB)": 73.83, "step": 85125, "token_acc": 0.9785932721712538, "train_speed(iter/s)": 1.455529 }, { "epoch": 3.647230195792811, "grad_norm": 5.146609783172607, "learning_rate": 1.6999728412874632e-05, "loss": 0.37035164833068845, "memory(GiB)": 73.83, "step": 85130, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.455539 }, { "epoch": 3.64744441112206, "grad_norm": 0.42925769090652466, "learning_rate": 1.6994672896941377e-05, "loss": 0.20235414505004884, "memory(GiB)": 73.83, "step": 85135, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.455551 }, { "epoch": 3.647658626451309, "grad_norm": 4.060281753540039, "learning_rate": 1.6989617978930587e-05, "loss": 0.4045952320098877, "memory(GiB)": 73.83, "step": 85140, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.455552 }, { "epoch": 3.6478728417805577, "grad_norm": 3.9016005992889404, "learning_rate": 1.6984563658933834e-05, "loss": 0.3653436183929443, "memory(GiB)": 73.83, "step": 85145, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.455555 }, { "epoch": 3.648087057109807, "grad_norm": 2.847740411758423, "learning_rate": 1.697950993704268e-05, "loss": 0.36018483638763427, "memory(GiB)": 73.83, "step": 85150, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.455561 }, { "epoch": 3.6483012724390558, "grad_norm": 3.4928317070007324, "learning_rate": 1.6974456813348678e-05, "loss": 0.2980233430862427, "memory(GiB)": 73.83, "step": 85155, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.455574 }, { "epoch": 3.6485154877683046, "grad_norm": 1.2492904663085938, "learning_rate": 1.6969404287943356e-05, "loss": 0.5699841499328613, "memory(GiB)": 73.83, "step": 85160, "token_acc": 0.8785714285714286, "train_speed(iter/s)": 1.455579 }, { "epoch": 3.648729703097554, "grad_norm": 1.2042981386184692, "learning_rate": 1.6964352360918277e-05, "loss": 0.23140065670013427, "memory(GiB)": 73.83, "step": 85165, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.455586 }, { "epoch": 3.6489439184268027, "grad_norm": 2.7284605503082275, "learning_rate": 1.6959301032364944e-05, "loss": 0.1567060112953186, "memory(GiB)": 73.83, "step": 85170, "token_acc": 0.9611650485436893, "train_speed(iter/s)": 1.455596 }, { "epoch": 3.6491581337560515, "grad_norm": 2.881941556930542, "learning_rate": 1.695425030237487e-05, "loss": 0.42087488174438475, "memory(GiB)": 73.83, "step": 85175, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.455602 }, { "epoch": 3.6493723490853007, "grad_norm": 2.2493457794189453, "learning_rate": 1.6949200171039547e-05, "loss": 0.28561174869537354, "memory(GiB)": 73.83, "step": 85180, "token_acc": 0.9319526627218935, "train_speed(iter/s)": 1.455619 }, { "epoch": 3.6495865644145495, "grad_norm": 4.29683780670166, "learning_rate": 1.6944150638450457e-05, "loss": 0.6185987949371338, "memory(GiB)": 73.83, "step": 85185, "token_acc": 0.8762886597938144, "train_speed(iter/s)": 1.455637 }, { "epoch": 3.6498007797437984, "grad_norm": 3.1530022621154785, "learning_rate": 1.6939101704699077e-05, "loss": 0.36092052459716795, "memory(GiB)": 73.83, "step": 85190, "token_acc": 0.9249146757679181, "train_speed(iter/s)": 1.455643 }, { "epoch": 3.6500149950730476, "grad_norm": 3.0987744331359863, "learning_rate": 1.6934053369876912e-05, "loss": 0.3608614444732666, "memory(GiB)": 73.83, "step": 85195, "token_acc": 0.9281437125748503, "train_speed(iter/s)": 1.455644 }, { "epoch": 3.6502292104022964, "grad_norm": 6.195904731750488, "learning_rate": 1.692900563407538e-05, "loss": 0.6218371391296387, "memory(GiB)": 73.83, "step": 85200, "token_acc": 0.8616352201257862, "train_speed(iter/s)": 1.455653 }, { "epoch": 3.6504434257315452, "grad_norm": 4.082941055297852, "learning_rate": 1.692395849738594e-05, "loss": 0.1719043731689453, "memory(GiB)": 73.83, "step": 85205, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.455664 }, { "epoch": 3.6506576410607945, "grad_norm": 2.4983794689178467, "learning_rate": 1.6918911959900015e-05, "loss": 0.3838533401489258, "memory(GiB)": 73.83, "step": 85210, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.455666 }, { "epoch": 3.6508718563900433, "grad_norm": 1.4284487962722778, "learning_rate": 1.691386602170902e-05, "loss": 0.35015454292297366, "memory(GiB)": 73.83, "step": 85215, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.455668 }, { "epoch": 3.651086071719292, "grad_norm": 1.9926799535751343, "learning_rate": 1.6908820682904387e-05, "loss": 0.18020226955413818, "memory(GiB)": 73.83, "step": 85220, "token_acc": 0.9578947368421052, "train_speed(iter/s)": 1.455671 }, { "epoch": 3.6513002870485414, "grad_norm": 3.178421974182129, "learning_rate": 1.6903775943577516e-05, "loss": 0.17347185611724852, "memory(GiB)": 73.83, "step": 85225, "token_acc": 0.9442622950819672, "train_speed(iter/s)": 1.455677 }, { "epoch": 3.65151450237779, "grad_norm": 2.6514203548431396, "learning_rate": 1.689873180381979e-05, "loss": 0.31878750324249266, "memory(GiB)": 73.83, "step": 85230, "token_acc": 0.9161290322580645, "train_speed(iter/s)": 1.455682 }, { "epoch": 3.651728717707039, "grad_norm": 5.146410942077637, "learning_rate": 1.689368826372259e-05, "loss": 0.5499526023864746, "memory(GiB)": 73.83, "step": 85235, "token_acc": 0.8677966101694915, "train_speed(iter/s)": 1.455697 }, { "epoch": 3.6519429330362883, "grad_norm": 2.203754186630249, "learning_rate": 1.6888645323377283e-05, "loss": 0.4187911510467529, "memory(GiB)": 73.83, "step": 85240, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.455696 }, { "epoch": 3.652157148365537, "grad_norm": 1.2788670063018799, "learning_rate": 1.6883602982875207e-05, "loss": 0.24135329723358154, "memory(GiB)": 73.83, "step": 85245, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.455699 }, { "epoch": 3.652371363694786, "grad_norm": 2.581453800201416, "learning_rate": 1.687856124230775e-05, "loss": 0.35436215400695803, "memory(GiB)": 73.83, "step": 85250, "token_acc": 0.9042904290429042, "train_speed(iter/s)": 1.455712 }, { "epoch": 3.652585579024035, "grad_norm": 5.25248384475708, "learning_rate": 1.6873520101766223e-05, "loss": 0.17770180702209473, "memory(GiB)": 73.83, "step": 85255, "token_acc": 0.9586466165413534, "train_speed(iter/s)": 1.455712 }, { "epoch": 3.652799794353284, "grad_norm": 4.122638702392578, "learning_rate": 1.6868479561341936e-05, "loss": 0.48606457710266116, "memory(GiB)": 73.83, "step": 85260, "token_acc": 0.9071428571428571, "train_speed(iter/s)": 1.455721 }, { "epoch": 3.6530140096825328, "grad_norm": 3.089265823364258, "learning_rate": 1.6863439621126243e-05, "loss": 0.3240772247314453, "memory(GiB)": 73.83, "step": 85265, "token_acc": 0.9172185430463576, "train_speed(iter/s)": 1.455721 }, { "epoch": 3.653228225011782, "grad_norm": 1.4804751873016357, "learning_rate": 1.6858400281210423e-05, "loss": 0.22364933490753175, "memory(GiB)": 73.83, "step": 85270, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.455724 }, { "epoch": 3.653442440341031, "grad_norm": 4.9647955894470215, "learning_rate": 1.685336154168576e-05, "loss": 0.611812973022461, "memory(GiB)": 73.83, "step": 85275, "token_acc": 0.8650306748466258, "train_speed(iter/s)": 1.455727 }, { "epoch": 3.6536566556702796, "grad_norm": 3.4440724849700928, "learning_rate": 1.6848323402643558e-05, "loss": 0.44167141914367675, "memory(GiB)": 73.83, "step": 85280, "token_acc": 0.8698630136986302, "train_speed(iter/s)": 1.455736 }, { "epoch": 3.653870870999529, "grad_norm": 4.644118309020996, "learning_rate": 1.684328586417508e-05, "loss": 0.5079066276550293, "memory(GiB)": 73.83, "step": 85285, "token_acc": 0.8746268656716418, "train_speed(iter/s)": 1.455736 }, { "epoch": 3.6540850863287777, "grad_norm": 2.421334981918335, "learning_rate": 1.6838248926371575e-05, "loss": 0.42854795455932615, "memory(GiB)": 73.83, "step": 85290, "token_acc": 0.8858131487889274, "train_speed(iter/s)": 1.455746 }, { "epoch": 3.6542993016580265, "grad_norm": 2.8078558444976807, "learning_rate": 1.6833212589324304e-05, "loss": 0.2702193260192871, "memory(GiB)": 73.83, "step": 85295, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.455759 }, { "epoch": 3.654513516987276, "grad_norm": 2.7431845664978027, "learning_rate": 1.68281768531245e-05, "loss": 0.47848973274230955, "memory(GiB)": 73.83, "step": 85300, "token_acc": 0.9018987341772152, "train_speed(iter/s)": 1.45576 }, { "epoch": 3.6547277323165246, "grad_norm": 4.509565830230713, "learning_rate": 1.682314171786337e-05, "loss": 0.2445290803909302, "memory(GiB)": 73.83, "step": 85305, "token_acc": 0.9407407407407408, "train_speed(iter/s)": 1.455772 }, { "epoch": 3.6549419476457734, "grad_norm": 4.395664691925049, "learning_rate": 1.6818107183632176e-05, "loss": 0.618990421295166, "memory(GiB)": 73.83, "step": 85310, "token_acc": 0.8834586466165414, "train_speed(iter/s)": 1.455784 }, { "epoch": 3.6551561629750227, "grad_norm": 2.6855077743530273, "learning_rate": 1.681307325052209e-05, "loss": 0.34689228534698485, "memory(GiB)": 73.83, "step": 85315, "token_acc": 0.9429530201342282, "train_speed(iter/s)": 1.455792 }, { "epoch": 3.6553703783042715, "grad_norm": 0.9291130900382996, "learning_rate": 1.6808039918624313e-05, "loss": 0.30923378467559814, "memory(GiB)": 73.83, "step": 85320, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.455795 }, { "epoch": 3.6555845936335203, "grad_norm": 0.5144397020339966, "learning_rate": 1.6803007188030036e-05, "loss": 0.4605696678161621, "memory(GiB)": 73.83, "step": 85325, "token_acc": 0.8933333333333333, "train_speed(iter/s)": 1.455796 }, { "epoch": 3.6557988089627695, "grad_norm": 4.429498672485352, "learning_rate": 1.67979750588304e-05, "loss": 0.29962069988250734, "memory(GiB)": 73.83, "step": 85330, "token_acc": 0.950530035335689, "train_speed(iter/s)": 1.455812 }, { "epoch": 3.6560130242920184, "grad_norm": 4.912877082824707, "learning_rate": 1.6792943531116594e-05, "loss": 0.5517233848571778, "memory(GiB)": 73.83, "step": 85335, "token_acc": 0.8884758364312267, "train_speed(iter/s)": 1.455813 }, { "epoch": 3.656227239621267, "grad_norm": 4.168037414550781, "learning_rate": 1.678791260497979e-05, "loss": 0.4059295654296875, "memory(GiB)": 73.83, "step": 85340, "token_acc": 0.9442379182156134, "train_speed(iter/s)": 1.455817 }, { "epoch": 3.6564414549505164, "grad_norm": 2.3151729106903076, "learning_rate": 1.6782882280511097e-05, "loss": 0.263493013381958, "memory(GiB)": 73.83, "step": 85345, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.45582 }, { "epoch": 3.6566556702797652, "grad_norm": 4.255663871765137, "learning_rate": 1.6777852557801655e-05, "loss": 0.5080109596252441, "memory(GiB)": 73.83, "step": 85350, "token_acc": 0.8870431893687708, "train_speed(iter/s)": 1.455832 }, { "epoch": 3.656869885609014, "grad_norm": 0.7807896733283997, "learning_rate": 1.6772823436942582e-05, "loss": 0.205572509765625, "memory(GiB)": 73.83, "step": 85355, "token_acc": 0.9638989169675091, "train_speed(iter/s)": 1.455836 }, { "epoch": 3.6570841009382633, "grad_norm": 0.06924213469028473, "learning_rate": 1.676779491802496e-05, "loss": 0.3454721212387085, "memory(GiB)": 73.83, "step": 85360, "token_acc": 0.9296187683284457, "train_speed(iter/s)": 1.455837 }, { "epoch": 3.657298316267512, "grad_norm": 2.061215400695801, "learning_rate": 1.676276700113992e-05, "loss": 0.24504239559173585, "memory(GiB)": 73.83, "step": 85365, "token_acc": 0.9627329192546584, "train_speed(iter/s)": 1.455843 }, { "epoch": 3.657512531596761, "grad_norm": 0.23981158435344696, "learning_rate": 1.6757739686378543e-05, "loss": 0.1617782235145569, "memory(GiB)": 73.83, "step": 85370, "token_acc": 0.9689655172413794, "train_speed(iter/s)": 1.455849 }, { "epoch": 3.65772674692601, "grad_norm": 9.2125825881958, "learning_rate": 1.6752712973831898e-05, "loss": 0.7635993003845215, "memory(GiB)": 73.83, "step": 85375, "token_acc": 0.8544061302681992, "train_speed(iter/s)": 1.455867 }, { "epoch": 3.657940962255259, "grad_norm": 2.733912467956543, "learning_rate": 1.674768686359104e-05, "loss": 0.24520764350891114, "memory(GiB)": 73.83, "step": 85380, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.455875 }, { "epoch": 3.658155177584508, "grad_norm": 3.432100534439087, "learning_rate": 1.6742661355747026e-05, "loss": 0.45276460647583006, "memory(GiB)": 73.83, "step": 85385, "token_acc": 0.8625429553264605, "train_speed(iter/s)": 1.455878 }, { "epoch": 3.658369392913757, "grad_norm": 4.28159236907959, "learning_rate": 1.6737636450390887e-05, "loss": 0.30258448123931886, "memory(GiB)": 73.83, "step": 85390, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.455882 }, { "epoch": 3.658583608243006, "grad_norm": 1.3914330005645752, "learning_rate": 1.673261214761368e-05, "loss": 0.44840850830078127, "memory(GiB)": 73.83, "step": 85395, "token_acc": 0.8771186440677966, "train_speed(iter/s)": 1.455889 }, { "epoch": 3.6587978235722547, "grad_norm": 2.568897008895874, "learning_rate": 1.6727588447506408e-05, "loss": 0.30730490684509276, "memory(GiB)": 73.83, "step": 85400, "token_acc": 0.9195402298850575, "train_speed(iter/s)": 1.455887 }, { "epoch": 3.659012038901504, "grad_norm": 4.025860786437988, "learning_rate": 1.6722565350160075e-05, "loss": 0.4565929889678955, "memory(GiB)": 73.83, "step": 85405, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.455891 }, { "epoch": 3.6592262542307528, "grad_norm": 1.2617261409759521, "learning_rate": 1.6717542855665702e-05, "loss": 0.3724967479705811, "memory(GiB)": 73.83, "step": 85410, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.455896 }, { "epoch": 3.6594404695600016, "grad_norm": 2.8795323371887207, "learning_rate": 1.671252096411426e-05, "loss": 0.2536862850189209, "memory(GiB)": 73.83, "step": 85415, "token_acc": 0.947565543071161, "train_speed(iter/s)": 1.455896 }, { "epoch": 3.659654684889251, "grad_norm": 5.336465358734131, "learning_rate": 1.670749967559671e-05, "loss": 0.34186253547668455, "memory(GiB)": 73.83, "step": 85420, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.455907 }, { "epoch": 3.6598689002184996, "grad_norm": 3.079824924468994, "learning_rate": 1.670247899020405e-05, "loss": 0.4480170249938965, "memory(GiB)": 73.83, "step": 85425, "token_acc": 0.9058441558441559, "train_speed(iter/s)": 1.455915 }, { "epoch": 3.6600831155477485, "grad_norm": 3.547112464904785, "learning_rate": 1.669745890802722e-05, "loss": 0.4339588165283203, "memory(GiB)": 73.83, "step": 85430, "token_acc": 0.88, "train_speed(iter/s)": 1.455937 }, { "epoch": 3.6602973308769977, "grad_norm": 2.7204720973968506, "learning_rate": 1.6692439429157164e-05, "loss": 0.5012048244476318, "memory(GiB)": 73.83, "step": 85435, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.455941 }, { "epoch": 3.6605115462062465, "grad_norm": 5.675271511077881, "learning_rate": 1.6687420553684808e-05, "loss": 0.4012794494628906, "memory(GiB)": 73.83, "step": 85440, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.45594 }, { "epoch": 3.6607257615354953, "grad_norm": 3.7571208477020264, "learning_rate": 1.668240228170108e-05, "loss": 0.1875072479248047, "memory(GiB)": 73.83, "step": 85445, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.455954 }, { "epoch": 3.6609399768647446, "grad_norm": 0.06047994643449783, "learning_rate": 1.6677384613296875e-05, "loss": 0.4474231243133545, "memory(GiB)": 73.83, "step": 85450, "token_acc": 0.8927444794952681, "train_speed(iter/s)": 1.45596 }, { "epoch": 3.6611541921939934, "grad_norm": 3.4464192390441895, "learning_rate": 1.6672367548563124e-05, "loss": 0.3406353950500488, "memory(GiB)": 73.83, "step": 85455, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.455966 }, { "epoch": 3.661368407523242, "grad_norm": 7.788421154022217, "learning_rate": 1.6667351087590693e-05, "loss": 0.5091786861419678, "memory(GiB)": 73.83, "step": 85460, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.455989 }, { "epoch": 3.6615826228524915, "grad_norm": 2.2980434894561768, "learning_rate": 1.6662335230470472e-05, "loss": 0.37836270332336425, "memory(GiB)": 73.83, "step": 85465, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.455993 }, { "epoch": 3.6617968381817403, "grad_norm": 4.589694023132324, "learning_rate": 1.665731997729332e-05, "loss": 0.39293575286865234, "memory(GiB)": 73.83, "step": 85470, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.456008 }, { "epoch": 3.662011053510989, "grad_norm": 0.7374311685562134, "learning_rate": 1.6652305328150074e-05, "loss": 0.23578188419342042, "memory(GiB)": 73.83, "step": 85475, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.456008 }, { "epoch": 3.6622252688402384, "grad_norm": 3.001132011413574, "learning_rate": 1.6647291283131606e-05, "loss": 0.32589921951293943, "memory(GiB)": 73.83, "step": 85480, "token_acc": 0.92, "train_speed(iter/s)": 1.456013 }, { "epoch": 3.662439484169487, "grad_norm": 3.355503797531128, "learning_rate": 1.664227784232876e-05, "loss": 0.22255079746246337, "memory(GiB)": 73.83, "step": 85485, "token_acc": 0.9296875, "train_speed(iter/s)": 1.456017 }, { "epoch": 3.662653699498736, "grad_norm": 6.457998752593994, "learning_rate": 1.6637265005832335e-05, "loss": 0.35281686782836913, "memory(GiB)": 73.83, "step": 85490, "token_acc": 0.9131832797427653, "train_speed(iter/s)": 1.456019 }, { "epoch": 3.6628679148279852, "grad_norm": 1.3810322284698486, "learning_rate": 1.6632252773733154e-05, "loss": 0.23752291202545167, "memory(GiB)": 73.83, "step": 85495, "token_acc": 0.95, "train_speed(iter/s)": 1.456024 }, { "epoch": 3.663082130157234, "grad_norm": 2.4690048694610596, "learning_rate": 1.6627241146122017e-05, "loss": 0.2771263599395752, "memory(GiB)": 73.83, "step": 85500, "token_acc": 0.9490909090909091, "train_speed(iter/s)": 1.456026 }, { "epoch": 3.663082130157234, "eval_loss": 2.480156183242798, "eval_runtime": 11.3211, "eval_samples_per_second": 8.833, "eval_steps_per_second": 8.833, "eval_token_acc": 0.4534412955465587, "step": 85500 }, { "epoch": 3.663296345486483, "grad_norm": 3.3228061199188232, "learning_rate": 1.66222301230897e-05, "loss": 0.24368879795074463, "memory(GiB)": 73.83, "step": 85505, "token_acc": 0.5934489402697495, "train_speed(iter/s)": 1.455723 }, { "epoch": 3.663510560815732, "grad_norm": 2.555943489074707, "learning_rate": 1.6617219704727004e-05, "loss": 0.44446773529052735, "memory(GiB)": 73.83, "step": 85510, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.455728 }, { "epoch": 3.663724776144981, "grad_norm": 0.19601626694202423, "learning_rate": 1.6612209891124696e-05, "loss": 0.20760009288787842, "memory(GiB)": 73.83, "step": 85515, "token_acc": 0.9481481481481482, "train_speed(iter/s)": 1.455732 }, { "epoch": 3.6639389914742297, "grad_norm": 3.1283724308013916, "learning_rate": 1.6607200682373526e-05, "loss": 0.3231987237930298, "memory(GiB)": 73.83, "step": 85520, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.455736 }, { "epoch": 3.664153206803479, "grad_norm": 2.5909717082977295, "learning_rate": 1.6602192078564237e-05, "loss": 0.3547219276428223, "memory(GiB)": 73.83, "step": 85525, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.45574 }, { "epoch": 3.664367422132728, "grad_norm": 1.321881651878357, "learning_rate": 1.659718407978757e-05, "loss": 0.3256812572479248, "memory(GiB)": 73.83, "step": 85530, "token_acc": 0.9094650205761317, "train_speed(iter/s)": 1.455743 }, { "epoch": 3.6645816374619766, "grad_norm": 1.2784698009490967, "learning_rate": 1.6592176686134232e-05, "loss": 0.40720367431640625, "memory(GiB)": 73.83, "step": 85535, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.455747 }, { "epoch": 3.664795852791226, "grad_norm": 3.4416208267211914, "learning_rate": 1.6587169897694965e-05, "loss": 0.3554516315460205, "memory(GiB)": 73.83, "step": 85540, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.455757 }, { "epoch": 3.6650100681204747, "grad_norm": 1.31523859500885, "learning_rate": 1.658216371456046e-05, "loss": 0.08070356845855713, "memory(GiB)": 73.83, "step": 85545, "token_acc": 0.9813432835820896, "train_speed(iter/s)": 1.45576 }, { "epoch": 3.6652242834497235, "grad_norm": 2.863142967224121, "learning_rate": 1.6577158136821408e-05, "loss": 0.30609705448150637, "memory(GiB)": 73.83, "step": 85550, "token_acc": 0.9303030303030303, "train_speed(iter/s)": 1.455756 }, { "epoch": 3.6654384987789728, "grad_norm": 4.716795444488525, "learning_rate": 1.6572153164568478e-05, "loss": 0.3746392011642456, "memory(GiB)": 73.83, "step": 85555, "token_acc": 0.9121338912133892, "train_speed(iter/s)": 1.455761 }, { "epoch": 3.6656527141082216, "grad_norm": 3.56790828704834, "learning_rate": 1.6567148797892362e-05, "loss": 0.27708275318145753, "memory(GiB)": 73.83, "step": 85560, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.455764 }, { "epoch": 3.6658669294374704, "grad_norm": 3.6156623363494873, "learning_rate": 1.6562145036883692e-05, "loss": 0.2397301197052002, "memory(GiB)": 73.83, "step": 85565, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.455766 }, { "epoch": 3.6660811447667196, "grad_norm": 3.6083669662475586, "learning_rate": 1.6557141881633152e-05, "loss": 0.46715970039367677, "memory(GiB)": 73.83, "step": 85570, "token_acc": 0.9093484419263456, "train_speed(iter/s)": 1.455767 }, { "epoch": 3.6662953600959685, "grad_norm": 0.6637826561927795, "learning_rate": 1.6552139332231354e-05, "loss": 0.3384159326553345, "memory(GiB)": 73.83, "step": 85575, "token_acc": 0.9209726443768997, "train_speed(iter/s)": 1.45577 }, { "epoch": 3.6665095754252173, "grad_norm": 0.1318703442811966, "learning_rate": 1.6547137388768937e-05, "loss": 0.12819000482559204, "memory(GiB)": 73.83, "step": 85580, "token_acc": 0.9715189873417721, "train_speed(iter/s)": 1.455785 }, { "epoch": 3.6667237907544665, "grad_norm": 2.758835792541504, "learning_rate": 1.65421360513365e-05, "loss": 0.3514916181564331, "memory(GiB)": 73.83, "step": 85585, "token_acc": 0.9218106995884774, "train_speed(iter/s)": 1.455783 }, { "epoch": 3.6669380060837153, "grad_norm": 3.936845541000366, "learning_rate": 1.6537135320024654e-05, "loss": 0.3210469722747803, "memory(GiB)": 73.83, "step": 85590, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.455786 }, { "epoch": 3.667152221412964, "grad_norm": 8.713886260986328, "learning_rate": 1.6532135194923982e-05, "loss": 0.3991896152496338, "memory(GiB)": 73.83, "step": 85595, "token_acc": 0.9030303030303031, "train_speed(iter/s)": 1.455798 }, { "epoch": 3.6673664367422134, "grad_norm": 4.43655252456665, "learning_rate": 1.652713567612509e-05, "loss": 0.37987775802612306, "memory(GiB)": 73.83, "step": 85600, "token_acc": 0.910394265232975, "train_speed(iter/s)": 1.455797 }, { "epoch": 3.667580652071462, "grad_norm": 3.1140778064727783, "learning_rate": 1.6522136763718533e-05, "loss": 0.5236390590667724, "memory(GiB)": 73.83, "step": 85605, "token_acc": 0.8737541528239202, "train_speed(iter/s)": 1.455801 }, { "epoch": 3.667794867400711, "grad_norm": 2.0063483715057373, "learning_rate": 1.6517138457794873e-05, "loss": 0.48443970680236814, "memory(GiB)": 73.83, "step": 85610, "token_acc": 0.9236947791164659, "train_speed(iter/s)": 1.45582 }, { "epoch": 3.6680090827299603, "grad_norm": 1.2072612047195435, "learning_rate": 1.651214075844466e-05, "loss": 0.13519517183303834, "memory(GiB)": 73.83, "step": 85615, "token_acc": 0.9662576687116564, "train_speed(iter/s)": 1.455819 }, { "epoch": 3.668223298059209, "grad_norm": 4.182684421539307, "learning_rate": 1.6507143665758417e-05, "loss": 0.23797082901000977, "memory(GiB)": 73.83, "step": 85620, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.455822 }, { "epoch": 3.668437513388458, "grad_norm": 1.9104297161102295, "learning_rate": 1.65021471798267e-05, "loss": 0.1816033124923706, "memory(GiB)": 73.83, "step": 85625, "token_acc": 0.9608540925266904, "train_speed(iter/s)": 1.455824 }, { "epoch": 3.668651728717707, "grad_norm": 3.4781603813171387, "learning_rate": 1.6497151300739994e-05, "loss": 0.3202920436859131, "memory(GiB)": 73.83, "step": 85630, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.455826 }, { "epoch": 3.668865944046956, "grad_norm": 4.398792743682861, "learning_rate": 1.6492156028588833e-05, "loss": 0.22837426662445068, "memory(GiB)": 73.83, "step": 85635, "token_acc": 0.9622641509433962, "train_speed(iter/s)": 1.455818 }, { "epoch": 3.669080159376205, "grad_norm": 2.7799346446990967, "learning_rate": 1.6487161363463705e-05, "loss": 0.4384561538696289, "memory(GiB)": 73.83, "step": 85640, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.455816 }, { "epoch": 3.669294374705454, "grad_norm": 2.437143564224243, "learning_rate": 1.6482167305455086e-05, "loss": 0.45282754898071287, "memory(GiB)": 73.83, "step": 85645, "token_acc": 0.9110320284697508, "train_speed(iter/s)": 1.455822 }, { "epoch": 3.669508590034703, "grad_norm": 0.42133015394210815, "learning_rate": 1.6477173854653433e-05, "loss": 0.1953510284423828, "memory(GiB)": 73.83, "step": 85650, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.455827 }, { "epoch": 3.6697228053639517, "grad_norm": 1.3981984853744507, "learning_rate": 1.6472181011149233e-05, "loss": 0.3667210578918457, "memory(GiB)": 73.83, "step": 85655, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.455832 }, { "epoch": 3.669937020693201, "grad_norm": 1.826606035232544, "learning_rate": 1.6467188775032925e-05, "loss": 0.30634613037109376, "memory(GiB)": 73.83, "step": 85660, "token_acc": 0.9191919191919192, "train_speed(iter/s)": 1.455834 }, { "epoch": 3.6701512360224497, "grad_norm": 4.300900459289551, "learning_rate": 1.646219714639496e-05, "loss": 0.2943427085876465, "memory(GiB)": 73.83, "step": 85665, "token_acc": 0.928, "train_speed(iter/s)": 1.455835 }, { "epoch": 3.6703654513516986, "grad_norm": 3.8381662368774414, "learning_rate": 1.6457206125325747e-05, "loss": 0.31317408084869386, "memory(GiB)": 73.83, "step": 85670, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.45583 }, { "epoch": 3.670579666680948, "grad_norm": 3.0502569675445557, "learning_rate": 1.6452215711915708e-05, "loss": 0.46876039505004885, "memory(GiB)": 73.83, "step": 85675, "token_acc": 0.8981132075471698, "train_speed(iter/s)": 1.45583 }, { "epoch": 3.6707938820101966, "grad_norm": 6.733222007751465, "learning_rate": 1.6447225906255237e-05, "loss": 0.26306962966918945, "memory(GiB)": 73.83, "step": 85680, "token_acc": 0.9503311258278145, "train_speed(iter/s)": 1.455831 }, { "epoch": 3.6710080973394454, "grad_norm": 2.704061985015869, "learning_rate": 1.6442236708434767e-05, "loss": 0.35575549602508544, "memory(GiB)": 73.83, "step": 85685, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.455834 }, { "epoch": 3.6712223126686947, "grad_norm": 7.065561294555664, "learning_rate": 1.643724811854465e-05, "loss": 0.5679893016815185, "memory(GiB)": 73.83, "step": 85690, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.455845 }, { "epoch": 3.6714365279979435, "grad_norm": 2.079160213470459, "learning_rate": 1.643226013667527e-05, "loss": 0.21528258323669433, "memory(GiB)": 73.83, "step": 85695, "token_acc": 0.967741935483871, "train_speed(iter/s)": 1.455853 }, { "epoch": 3.6716507433271923, "grad_norm": 1.2198034524917603, "learning_rate": 1.6427272762916973e-05, "loss": 0.39172232151031494, "memory(GiB)": 73.83, "step": 85700, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.455857 }, { "epoch": 3.6718649586564416, "grad_norm": 2.659916877746582, "learning_rate": 1.6422285997360136e-05, "loss": 0.2926163911819458, "memory(GiB)": 73.83, "step": 85705, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.455865 }, { "epoch": 3.6720791739856904, "grad_norm": 2.750915288925171, "learning_rate": 1.6417299840095073e-05, "loss": 0.2426231861114502, "memory(GiB)": 73.83, "step": 85710, "token_acc": 0.9561128526645768, "train_speed(iter/s)": 1.455866 }, { "epoch": 3.672293389314939, "grad_norm": 1.4829301834106445, "learning_rate": 1.6412314291212143e-05, "loss": 0.2907297134399414, "memory(GiB)": 73.83, "step": 85715, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.455863 }, { "epoch": 3.6725076046441885, "grad_norm": 0.6401042938232422, "learning_rate": 1.6407329350801647e-05, "loss": 0.18085321187973022, "memory(GiB)": 73.83, "step": 85720, "token_acc": 0.9668674698795181, "train_speed(iter/s)": 1.455866 }, { "epoch": 3.6727218199734373, "grad_norm": 4.2723469734191895, "learning_rate": 1.640234501895389e-05, "loss": 0.38560757637023924, "memory(GiB)": 73.83, "step": 85725, "token_acc": 0.9136690647482014, "train_speed(iter/s)": 1.45588 }, { "epoch": 3.672936035302686, "grad_norm": 3.8191514015197754, "learning_rate": 1.6397361295759172e-05, "loss": 0.35737996101379393, "memory(GiB)": 73.83, "step": 85730, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.455875 }, { "epoch": 3.6731502506319353, "grad_norm": 5.115346908569336, "learning_rate": 1.6392378181307777e-05, "loss": 0.3188910007476807, "memory(GiB)": 73.83, "step": 85735, "token_acc": 0.9237668161434978, "train_speed(iter/s)": 1.455875 }, { "epoch": 3.673364465961184, "grad_norm": 3.200712203979492, "learning_rate": 1.6387395675689964e-05, "loss": 0.35973300933837893, "memory(GiB)": 73.83, "step": 85740, "token_acc": 0.9155555555555556, "train_speed(iter/s)": 1.455884 }, { "epoch": 3.673578681290433, "grad_norm": 2.7655177116394043, "learning_rate": 1.6382413778996018e-05, "loss": 0.42629122734069824, "memory(GiB)": 73.83, "step": 85745, "token_acc": 0.906060606060606, "train_speed(iter/s)": 1.455901 }, { "epoch": 3.6737928966196822, "grad_norm": 3.917531728744507, "learning_rate": 1.637743249131619e-05, "loss": 0.369297456741333, "memory(GiB)": 73.83, "step": 85750, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.455908 }, { "epoch": 3.674007111948931, "grad_norm": 1.2728103399276733, "learning_rate": 1.637245181274071e-05, "loss": 0.5813206195831299, "memory(GiB)": 73.83, "step": 85755, "token_acc": 0.8603773584905661, "train_speed(iter/s)": 1.455923 }, { "epoch": 3.67422132727818, "grad_norm": 4.151219844818115, "learning_rate": 1.6367471743359814e-05, "loss": 0.4867987632751465, "memory(GiB)": 73.83, "step": 85760, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.455926 }, { "epoch": 3.674435542607429, "grad_norm": 4.367049694061279, "learning_rate": 1.636249228326369e-05, "loss": 0.37424745559692385, "memory(GiB)": 73.83, "step": 85765, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.455928 }, { "epoch": 3.674649757936678, "grad_norm": 3.86497163772583, "learning_rate": 1.63575134325426e-05, "loss": 0.40400142669677735, "memory(GiB)": 73.83, "step": 85770, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.455933 }, { "epoch": 3.6748639732659267, "grad_norm": 4.430786609649658, "learning_rate": 1.6352535191286693e-05, "loss": 0.4475820541381836, "memory(GiB)": 73.83, "step": 85775, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.455935 }, { "epoch": 3.675078188595176, "grad_norm": 5.519897937774658, "learning_rate": 1.6347557559586192e-05, "loss": 0.16984057426452637, "memory(GiB)": 73.83, "step": 85780, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.455951 }, { "epoch": 3.675292403924425, "grad_norm": 3.2697713375091553, "learning_rate": 1.6342580537531256e-05, "loss": 0.39145755767822266, "memory(GiB)": 73.83, "step": 85785, "token_acc": 0.9053708439897699, "train_speed(iter/s)": 1.455961 }, { "epoch": 3.6755066192536736, "grad_norm": 3.1280739307403564, "learning_rate": 1.633760412521204e-05, "loss": 0.29427304267883303, "memory(GiB)": 73.83, "step": 85790, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.455971 }, { "epoch": 3.675720834582923, "grad_norm": 1.483605980873108, "learning_rate": 1.6332628322718696e-05, "loss": 0.2715842962265015, "memory(GiB)": 73.83, "step": 85795, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.455978 }, { "epoch": 3.6759350499121717, "grad_norm": 0.32730188965797424, "learning_rate": 1.6327653130141385e-05, "loss": 0.17031391859054565, "memory(GiB)": 73.83, "step": 85800, "token_acc": 0.9609929078014184, "train_speed(iter/s)": 1.455985 }, { "epoch": 3.6761492652414205, "grad_norm": 2.4048733711242676, "learning_rate": 1.6322678547570224e-05, "loss": 0.3292550563812256, "memory(GiB)": 73.83, "step": 85805, "token_acc": 0.924812030075188, "train_speed(iter/s)": 1.455983 }, { "epoch": 3.6763634805706698, "grad_norm": 3.3971147537231445, "learning_rate": 1.6317704575095333e-05, "loss": 0.446869945526123, "memory(GiB)": 73.83, "step": 85810, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.455983 }, { "epoch": 3.6765776958999186, "grad_norm": 3.836110830307007, "learning_rate": 1.631273121280682e-05, "loss": 0.36555185317993166, "memory(GiB)": 73.83, "step": 85815, "token_acc": 0.9368029739776952, "train_speed(iter/s)": 1.455986 }, { "epoch": 3.6767919112291674, "grad_norm": 2.6939339637756348, "learning_rate": 1.630775846079478e-05, "loss": 0.36034255027770995, "memory(GiB)": 73.83, "step": 85820, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.455994 }, { "epoch": 3.6770061265584166, "grad_norm": 4.361734390258789, "learning_rate": 1.630278631914929e-05, "loss": 0.2187499761581421, "memory(GiB)": 73.83, "step": 85825, "token_acc": 0.9505703422053232, "train_speed(iter/s)": 1.456009 }, { "epoch": 3.6772203418876654, "grad_norm": 1.4638450145721436, "learning_rate": 1.6297814787960453e-05, "loss": 0.35900495052337644, "memory(GiB)": 73.83, "step": 85830, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.45601 }, { "epoch": 3.6774345572169143, "grad_norm": 6.052178859710693, "learning_rate": 1.6292843867318307e-05, "loss": 0.5527513980865478, "memory(GiB)": 73.83, "step": 85835, "token_acc": 0.8979591836734694, "train_speed(iter/s)": 1.456013 }, { "epoch": 3.6776487725461635, "grad_norm": 6.2615742683410645, "learning_rate": 1.6287873557312927e-05, "loss": 0.42367162704467776, "memory(GiB)": 73.83, "step": 85840, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.456015 }, { "epoch": 3.6778629878754123, "grad_norm": 0.14822062849998474, "learning_rate": 1.628290385803433e-05, "loss": 0.3093691825866699, "memory(GiB)": 73.83, "step": 85845, "token_acc": 0.9085173501577287, "train_speed(iter/s)": 1.456028 }, { "epoch": 3.678077203204661, "grad_norm": 3.588545322418213, "learning_rate": 1.6277934769572552e-05, "loss": 0.32029476165771487, "memory(GiB)": 73.83, "step": 85850, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.456028 }, { "epoch": 3.6782914185339104, "grad_norm": 3.5951216220855713, "learning_rate": 1.6272966292017616e-05, "loss": 0.2953484296798706, "memory(GiB)": 73.83, "step": 85855, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.456031 }, { "epoch": 3.678505633863159, "grad_norm": 6.847243309020996, "learning_rate": 1.6267998425459552e-05, "loss": 0.336742639541626, "memory(GiB)": 73.83, "step": 85860, "token_acc": 0.9391891891891891, "train_speed(iter/s)": 1.456036 }, { "epoch": 3.678719849192408, "grad_norm": 1.1270840167999268, "learning_rate": 1.6263031169988337e-05, "loss": 0.23390662670135498, "memory(GiB)": 73.83, "step": 85865, "token_acc": 0.9432835820895522, "train_speed(iter/s)": 1.456047 }, { "epoch": 3.6789340645216573, "grad_norm": 1.5542718172073364, "learning_rate": 1.625806452569396e-05, "loss": 0.2598039865493774, "memory(GiB)": 73.83, "step": 85870, "token_acc": 0.9521276595744681, "train_speed(iter/s)": 1.456055 }, { "epoch": 3.679148279850906, "grad_norm": 0.34121549129486084, "learning_rate": 1.6253098492666397e-05, "loss": 0.3513902425765991, "memory(GiB)": 73.83, "step": 85875, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.456053 }, { "epoch": 3.679362495180155, "grad_norm": 3.192391872406006, "learning_rate": 1.6248133070995613e-05, "loss": 0.5075417518615722, "memory(GiB)": 73.83, "step": 85880, "token_acc": 0.8758620689655172, "train_speed(iter/s)": 1.456052 }, { "epoch": 3.679576710509404, "grad_norm": 0.9737005829811096, "learning_rate": 1.6243168260771547e-05, "loss": 0.3058526039123535, "memory(GiB)": 73.83, "step": 85885, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.456057 }, { "epoch": 3.679790925838653, "grad_norm": 3.6033308506011963, "learning_rate": 1.623820406208417e-05, "loss": 0.4912543296813965, "memory(GiB)": 73.83, "step": 85890, "token_acc": 0.8773006134969326, "train_speed(iter/s)": 1.456058 }, { "epoch": 3.680005141167902, "grad_norm": 1.2774447202682495, "learning_rate": 1.6233240475023394e-05, "loss": 0.42133173942565916, "memory(GiB)": 73.83, "step": 85895, "token_acc": 0.9187279151943463, "train_speed(iter/s)": 1.456071 }, { "epoch": 3.680219356497151, "grad_norm": 2.6917333602905273, "learning_rate": 1.622827749967914e-05, "loss": 0.31435940265655515, "memory(GiB)": 73.83, "step": 85900, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.456073 }, { "epoch": 3.6804335718264, "grad_norm": 2.53239107131958, "learning_rate": 1.6223315136141327e-05, "loss": 0.3651005744934082, "memory(GiB)": 73.83, "step": 85905, "token_acc": 0.923728813559322, "train_speed(iter/s)": 1.456082 }, { "epoch": 3.6806477871556487, "grad_norm": 2.9649109840393066, "learning_rate": 1.6218353384499824e-05, "loss": 0.39882750511169435, "memory(GiB)": 73.83, "step": 85910, "token_acc": 0.9049295774647887, "train_speed(iter/s)": 1.456081 }, { "epoch": 3.680862002484898, "grad_norm": 4.4029340744018555, "learning_rate": 1.6213392244844554e-05, "loss": 0.4204558849334717, "memory(GiB)": 73.83, "step": 85915, "token_acc": 0.9015873015873016, "train_speed(iter/s)": 1.456084 }, { "epoch": 3.6810762178141467, "grad_norm": 2.9976003170013428, "learning_rate": 1.6208431717265382e-05, "loss": 0.5344247341156005, "memory(GiB)": 73.83, "step": 85920, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.456089 }, { "epoch": 3.6812904331433955, "grad_norm": 1.8039182424545288, "learning_rate": 1.6203471801852148e-05, "loss": 0.30608229637145995, "memory(GiB)": 73.83, "step": 85925, "token_acc": 0.9407894736842105, "train_speed(iter/s)": 1.456094 }, { "epoch": 3.681504648472645, "grad_norm": 3.5821192264556885, "learning_rate": 1.6198512498694744e-05, "loss": 0.23120782375335694, "memory(GiB)": 73.83, "step": 85930, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.456105 }, { "epoch": 3.6817188638018936, "grad_norm": 3.245274782180786, "learning_rate": 1.6193553807882998e-05, "loss": 0.38730435371398925, "memory(GiB)": 73.83, "step": 85935, "token_acc": 0.9139072847682119, "train_speed(iter/s)": 1.456108 }, { "epoch": 3.6819330791311424, "grad_norm": 2.749162435531616, "learning_rate": 1.6188595729506722e-05, "loss": 0.3406409502029419, "memory(GiB)": 73.83, "step": 85940, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.456111 }, { "epoch": 3.6821472944603917, "grad_norm": 3.2846837043762207, "learning_rate": 1.6183638263655765e-05, "loss": 0.27967472076416017, "memory(GiB)": 73.83, "step": 85945, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.456119 }, { "epoch": 3.6823615097896405, "grad_norm": 2.678818464279175, "learning_rate": 1.617868141041992e-05, "loss": 0.2652535676956177, "memory(GiB)": 73.83, "step": 85950, "token_acc": 0.958904109589041, "train_speed(iter/s)": 1.456129 }, { "epoch": 3.6825757251188893, "grad_norm": 2.77040433883667, "learning_rate": 1.6173725169888994e-05, "loss": 0.33785419464111327, "memory(GiB)": 73.83, "step": 85955, "token_acc": 0.9342105263157895, "train_speed(iter/s)": 1.456131 }, { "epoch": 3.6827899404481386, "grad_norm": 0.4535753130912781, "learning_rate": 1.6168769542152767e-05, "loss": 0.2504436492919922, "memory(GiB)": 73.83, "step": 85960, "token_acc": 0.9575757575757575, "train_speed(iter/s)": 1.456139 }, { "epoch": 3.6830041557773874, "grad_norm": 5.017334461212158, "learning_rate": 1.6163814527301013e-05, "loss": 0.4826827049255371, "memory(GiB)": 73.83, "step": 85965, "token_acc": 0.8996865203761756, "train_speed(iter/s)": 1.456144 }, { "epoch": 3.683218371106636, "grad_norm": 2.2979044914245605, "learning_rate": 1.6158860125423485e-05, "loss": 0.3173861265182495, "memory(GiB)": 73.83, "step": 85970, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.456147 }, { "epoch": 3.6834325864358854, "grad_norm": 1.7847763299942017, "learning_rate": 1.615390633660997e-05, "loss": 0.16367714405059813, "memory(GiB)": 73.83, "step": 85975, "token_acc": 0.9699453551912568, "train_speed(iter/s)": 1.456148 }, { "epoch": 3.6836468017651343, "grad_norm": 3.0547780990600586, "learning_rate": 1.6148953160950187e-05, "loss": 0.5965682983398437, "memory(GiB)": 73.83, "step": 85980, "token_acc": 0.8711656441717791, "train_speed(iter/s)": 1.456148 }, { "epoch": 3.683861017094383, "grad_norm": 2.2586748600006104, "learning_rate": 1.6144000598533877e-05, "loss": 0.2012610912322998, "memory(GiB)": 73.83, "step": 85985, "token_acc": 0.9601226993865031, "train_speed(iter/s)": 1.456144 }, { "epoch": 3.6840752324236323, "grad_norm": 4.084204196929932, "learning_rate": 1.6139048649450756e-05, "loss": 0.5593907356262207, "memory(GiB)": 73.83, "step": 85990, "token_acc": 0.87890625, "train_speed(iter/s)": 1.456146 }, { "epoch": 3.684289447752881, "grad_norm": 2.4185142517089844, "learning_rate": 1.6134097313790515e-05, "loss": 0.2217728614807129, "memory(GiB)": 73.83, "step": 85995, "token_acc": 0.9515570934256056, "train_speed(iter/s)": 1.456148 }, { "epoch": 3.68450366308213, "grad_norm": 2.715411424636841, "learning_rate": 1.6129146591642868e-05, "loss": 0.39624199867248533, "memory(GiB)": 73.83, "step": 86000, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.456156 }, { "epoch": 3.68450366308213, "eval_loss": 2.520899772644043, "eval_runtime": 11.3049, "eval_samples_per_second": 8.846, "eval_steps_per_second": 8.846, "eval_token_acc": 0.4561933534743202, "step": 86000 }, { "epoch": 3.684717878411379, "grad_norm": 6.514016151428223, "learning_rate": 1.612419648309752e-05, "loss": 0.3164398193359375, "memory(GiB)": 73.83, "step": 86005, "token_acc": 0.6064382139148494, "train_speed(iter/s)": 1.455856 }, { "epoch": 3.684932093740628, "grad_norm": 1.7712620496749878, "learning_rate": 1.6119246988244136e-05, "loss": 0.25573642253875734, "memory(GiB)": 73.83, "step": 86010, "token_acc": 0.9636363636363636, "train_speed(iter/s)": 1.455859 }, { "epoch": 3.685146309069877, "grad_norm": 4.187249183654785, "learning_rate": 1.6114298107172372e-05, "loss": 0.2982205390930176, "memory(GiB)": 73.83, "step": 86015, "token_acc": 0.921875, "train_speed(iter/s)": 1.455867 }, { "epoch": 3.685360524399126, "grad_norm": 4.829846382141113, "learning_rate": 1.6109349839971883e-05, "loss": 0.3440863609313965, "memory(GiB)": 73.83, "step": 86020, "token_acc": 0.948, "train_speed(iter/s)": 1.455868 }, { "epoch": 3.685574739728375, "grad_norm": 3.2263858318328857, "learning_rate": 1.61044021867323e-05, "loss": 0.2412175416946411, "memory(GiB)": 73.83, "step": 86025, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.455864 }, { "epoch": 3.6857889550576237, "grad_norm": 3.001725435256958, "learning_rate": 1.6099455147543284e-05, "loss": 0.47093887329101564, "memory(GiB)": 73.83, "step": 86030, "token_acc": 0.9063670411985019, "train_speed(iter/s)": 1.455863 }, { "epoch": 3.686003170386873, "grad_norm": 1.0416817665100098, "learning_rate": 1.609450872249444e-05, "loss": 0.399900484085083, "memory(GiB)": 73.83, "step": 86035, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.455872 }, { "epoch": 3.686217385716122, "grad_norm": 4.720424175262451, "learning_rate": 1.6089562911675377e-05, "loss": 0.30922350883483884, "memory(GiB)": 73.83, "step": 86040, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.455874 }, { "epoch": 3.6864316010453706, "grad_norm": 2.651855945587158, "learning_rate": 1.6084617715175686e-05, "loss": 0.18607418537139891, "memory(GiB)": 73.83, "step": 86045, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.455881 }, { "epoch": 3.68664581637462, "grad_norm": 3.687436819076538, "learning_rate": 1.6079673133084967e-05, "loss": 0.43903064727783203, "memory(GiB)": 73.83, "step": 86050, "token_acc": 0.905982905982906, "train_speed(iter/s)": 1.455881 }, { "epoch": 3.6868600317038687, "grad_norm": 1.7503939867019653, "learning_rate": 1.607472916549277e-05, "loss": 0.2443093776702881, "memory(GiB)": 73.83, "step": 86055, "token_acc": 0.9348534201954397, "train_speed(iter/s)": 1.455879 }, { "epoch": 3.6870742470331175, "grad_norm": 2.5121026039123535, "learning_rate": 1.6069785812488696e-05, "loss": 0.40393686294555664, "memory(GiB)": 73.83, "step": 86060, "token_acc": 0.916083916083916, "train_speed(iter/s)": 1.455878 }, { "epoch": 3.6872884623623667, "grad_norm": 3.2762131690979004, "learning_rate": 1.6064843074162285e-05, "loss": 0.20856754779815673, "memory(GiB)": 73.83, "step": 86065, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.455881 }, { "epoch": 3.6875026776916155, "grad_norm": 3.852480411529541, "learning_rate": 1.6059900950603052e-05, "loss": 0.1828879952430725, "memory(GiB)": 73.83, "step": 86070, "token_acc": 0.9591836734693877, "train_speed(iter/s)": 1.455879 }, { "epoch": 3.6877168930208644, "grad_norm": 4.963402271270752, "learning_rate": 1.6054959441900574e-05, "loss": 0.1796713352203369, "memory(GiB)": 73.83, "step": 86075, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.455888 }, { "epoch": 3.6879311083501136, "grad_norm": 5.219747066497803, "learning_rate": 1.605001854814434e-05, "loss": 0.15474414825439453, "memory(GiB)": 73.83, "step": 86080, "token_acc": 0.9604743083003953, "train_speed(iter/s)": 1.455891 }, { "epoch": 3.6881453236793624, "grad_norm": 1.0409921407699585, "learning_rate": 1.6045078269423864e-05, "loss": 0.32428631782531736, "memory(GiB)": 73.83, "step": 86085, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.455898 }, { "epoch": 3.6883595390086112, "grad_norm": 3.4910523891448975, "learning_rate": 1.6040138605828653e-05, "loss": 0.30091884136199953, "memory(GiB)": 73.83, "step": 86090, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.455898 }, { "epoch": 3.6885737543378605, "grad_norm": 6.763954162597656, "learning_rate": 1.6035199557448194e-05, "loss": 0.47142753601074217, "memory(GiB)": 73.83, "step": 86095, "token_acc": 0.9, "train_speed(iter/s)": 1.455902 }, { "epoch": 3.6887879696671093, "grad_norm": 0.4360121786594391, "learning_rate": 1.6030261124371953e-05, "loss": 0.22367212772369385, "memory(GiB)": 73.83, "step": 86100, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.455906 }, { "epoch": 3.689002184996358, "grad_norm": 4.936406135559082, "learning_rate": 1.60253233066894e-05, "loss": 0.4561013698577881, "memory(GiB)": 73.83, "step": 86105, "token_acc": 0.9069767441860465, "train_speed(iter/s)": 1.455905 }, { "epoch": 3.6892164003256074, "grad_norm": 0.20614059269428253, "learning_rate": 1.6020386104489986e-05, "loss": 0.2479762315750122, "memory(GiB)": 73.83, "step": 86110, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.455916 }, { "epoch": 3.689430615654856, "grad_norm": 0.48647087812423706, "learning_rate": 1.6015449517863133e-05, "loss": 0.22419435977935792, "memory(GiB)": 73.83, "step": 86115, "token_acc": 0.9460227272727273, "train_speed(iter/s)": 1.455919 }, { "epoch": 3.689644830984105, "grad_norm": 5.665931701660156, "learning_rate": 1.6010513546898316e-05, "loss": 0.2066988468170166, "memory(GiB)": 73.83, "step": 86120, "token_acc": 0.9498432601880877, "train_speed(iter/s)": 1.455929 }, { "epoch": 3.6898590463133543, "grad_norm": 5.039323329925537, "learning_rate": 1.600557819168493e-05, "loss": 0.7267226219177246, "memory(GiB)": 73.83, "step": 86125, "token_acc": 0.8825910931174089, "train_speed(iter/s)": 1.45593 }, { "epoch": 3.690073261642603, "grad_norm": 5.271035671234131, "learning_rate": 1.600064345231238e-05, "loss": 0.2978257656097412, "memory(GiB)": 73.83, "step": 86130, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.455935 }, { "epoch": 3.690287476971852, "grad_norm": 2.2048628330230713, "learning_rate": 1.5995709328870067e-05, "loss": 0.40038042068481444, "memory(GiB)": 73.83, "step": 86135, "token_acc": 0.9219858156028369, "train_speed(iter/s)": 1.455938 }, { "epoch": 3.690501692301101, "grad_norm": 3.7192516326904297, "learning_rate": 1.5990775821447363e-05, "loss": 0.39777650833129885, "memory(GiB)": 73.83, "step": 86140, "token_acc": 0.9131944444444444, "train_speed(iter/s)": 1.455937 }, { "epoch": 3.69071590763035, "grad_norm": 4.611566543579102, "learning_rate": 1.598584293013366e-05, "loss": 0.5543557167053222, "memory(GiB)": 73.83, "step": 86145, "token_acc": 0.881578947368421, "train_speed(iter/s)": 1.455951 }, { "epoch": 3.6909301229595988, "grad_norm": 5.276882648468018, "learning_rate": 1.5980910655018332e-05, "loss": 0.4669938087463379, "memory(GiB)": 73.83, "step": 86150, "token_acc": 0.8884758364312267, "train_speed(iter/s)": 1.455952 }, { "epoch": 3.691144338288848, "grad_norm": 5.369640350341797, "learning_rate": 1.5975978996190727e-05, "loss": 0.7622966766357422, "memory(GiB)": 73.83, "step": 86155, "token_acc": 0.875, "train_speed(iter/s)": 1.455968 }, { "epoch": 3.691358553618097, "grad_norm": 3.379021167755127, "learning_rate": 1.5971047953740174e-05, "loss": 0.37014198303222656, "memory(GiB)": 73.83, "step": 86160, "token_acc": 0.924924924924925, "train_speed(iter/s)": 1.455968 }, { "epoch": 3.6915727689473457, "grad_norm": 2.439462900161743, "learning_rate": 1.5966117527756013e-05, "loss": 0.28227131366729735, "memory(GiB)": 73.83, "step": 86165, "token_acc": 0.939209726443769, "train_speed(iter/s)": 1.45597 }, { "epoch": 3.691786984276595, "grad_norm": 2.7496607303619385, "learning_rate": 1.596118771832754e-05, "loss": 0.3309345722198486, "memory(GiB)": 73.83, "step": 86170, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.455975 }, { "epoch": 3.6920011996058437, "grad_norm": 2.5959527492523193, "learning_rate": 1.59562585255441e-05, "loss": 0.3495686769485474, "memory(GiB)": 73.83, "step": 86175, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.455979 }, { "epoch": 3.6922154149350925, "grad_norm": 1.7547175884246826, "learning_rate": 1.5951329949494976e-05, "loss": 0.280863881111145, "memory(GiB)": 73.83, "step": 86180, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.455984 }, { "epoch": 3.692429630264342, "grad_norm": 3.025142192840576, "learning_rate": 1.5946401990269444e-05, "loss": 0.3376279354095459, "memory(GiB)": 73.83, "step": 86185, "token_acc": 0.9312714776632303, "train_speed(iter/s)": 1.455987 }, { "epoch": 3.6926438455935906, "grad_norm": 2.1948325634002686, "learning_rate": 1.5941474647956788e-05, "loss": 0.2819032669067383, "memory(GiB)": 73.83, "step": 86190, "token_acc": 0.9292307692307692, "train_speed(iter/s)": 1.455994 }, { "epoch": 3.6928580609228394, "grad_norm": 2.798572540283203, "learning_rate": 1.5936547922646268e-05, "loss": 0.5927255153656006, "memory(GiB)": 73.83, "step": 86195, "token_acc": 0.875, "train_speed(iter/s)": 1.455999 }, { "epoch": 3.6930722762520887, "grad_norm": 3.671405553817749, "learning_rate": 1.5931621814427118e-05, "loss": 0.37481091022491453, "memory(GiB)": 73.83, "step": 86200, "token_acc": 0.9173333333333333, "train_speed(iter/s)": 1.456014 }, { "epoch": 3.6932864915813375, "grad_norm": 4.664531707763672, "learning_rate": 1.5926696323388618e-05, "loss": 0.5575116157531739, "memory(GiB)": 73.83, "step": 86205, "token_acc": 0.8803986710963455, "train_speed(iter/s)": 1.456026 }, { "epoch": 3.6935007069105863, "grad_norm": 5.078563690185547, "learning_rate": 1.5921771449619977e-05, "loss": 0.3807117223739624, "memory(GiB)": 73.83, "step": 86210, "token_acc": 0.9375, "train_speed(iter/s)": 1.456028 }, { "epoch": 3.6937149222398356, "grad_norm": 4.747751712799072, "learning_rate": 1.591684719321041e-05, "loss": 0.512553358078003, "memory(GiB)": 73.83, "step": 86215, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.456043 }, { "epoch": 3.6939291375690844, "grad_norm": 4.499035835266113, "learning_rate": 1.5911923554249113e-05, "loss": 0.7828083038330078, "memory(GiB)": 73.83, "step": 86220, "token_acc": 0.8503649635036497, "train_speed(iter/s)": 1.456059 }, { "epoch": 3.694143352898333, "grad_norm": 9.705803871154785, "learning_rate": 1.590700053282532e-05, "loss": 0.5587888717651367, "memory(GiB)": 73.83, "step": 86225, "token_acc": 0.8851351351351351, "train_speed(iter/s)": 1.456071 }, { "epoch": 3.6943575682275824, "grad_norm": 3.2180464267730713, "learning_rate": 1.590207812902817e-05, "loss": 0.458724308013916, "memory(GiB)": 73.83, "step": 86230, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.456071 }, { "epoch": 3.6945717835568312, "grad_norm": 8.176648139953613, "learning_rate": 1.5897156342946883e-05, "loss": 0.4887696743011475, "memory(GiB)": 73.83, "step": 86235, "token_acc": 0.9043824701195219, "train_speed(iter/s)": 1.456086 }, { "epoch": 3.69478599888608, "grad_norm": 4.9323554039001465, "learning_rate": 1.58922351746706e-05, "loss": 0.5633634090423584, "memory(GiB)": 73.83, "step": 86240, "token_acc": 0.8496503496503497, "train_speed(iter/s)": 1.456095 }, { "epoch": 3.6950002142153293, "grad_norm": 1.3778960704803467, "learning_rate": 1.5887314624288467e-05, "loss": 0.3135730981826782, "memory(GiB)": 73.83, "step": 86245, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.456109 }, { "epoch": 3.695214429544578, "grad_norm": 4.179458141326904, "learning_rate": 1.5882394691889636e-05, "loss": 0.7561168670654297, "memory(GiB)": 73.83, "step": 86250, "token_acc": 0.8535911602209945, "train_speed(iter/s)": 1.456109 }, { "epoch": 3.695428644873827, "grad_norm": 1.5537174940109253, "learning_rate": 1.587747537756323e-05, "loss": 0.29591960906982423, "memory(GiB)": 73.83, "step": 86255, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.456111 }, { "epoch": 3.695642860203076, "grad_norm": 2.771462917327881, "learning_rate": 1.5872556681398342e-05, "loss": 0.25510706901550295, "memory(GiB)": 73.83, "step": 86260, "token_acc": 0.9489795918367347, "train_speed(iter/s)": 1.456112 }, { "epoch": 3.695857075532325, "grad_norm": 0.5530331134796143, "learning_rate": 1.5867638603484132e-05, "loss": 0.32990360260009766, "memory(GiB)": 73.83, "step": 86265, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.456125 }, { "epoch": 3.696071290861574, "grad_norm": 3.0807998180389404, "learning_rate": 1.5862721143909658e-05, "loss": 0.35419378280639646, "memory(GiB)": 73.83, "step": 86270, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.456131 }, { "epoch": 3.696285506190823, "grad_norm": 2.784602165222168, "learning_rate": 1.585780430276402e-05, "loss": 0.31907947063446046, "memory(GiB)": 73.83, "step": 86275, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.456134 }, { "epoch": 3.696499721520072, "grad_norm": 2.370708703994751, "learning_rate": 1.585288808013628e-05, "loss": 0.5940621852874756, "memory(GiB)": 73.83, "step": 86280, "token_acc": 0.8872727272727273, "train_speed(iter/s)": 1.456146 }, { "epoch": 3.6967139368493207, "grad_norm": 3.7672665119171143, "learning_rate": 1.584797247611549e-05, "loss": 0.35608444213867185, "memory(GiB)": 73.83, "step": 86285, "token_acc": 0.9218106995884774, "train_speed(iter/s)": 1.456147 }, { "epoch": 3.69692815217857, "grad_norm": 4.789624214172363, "learning_rate": 1.5843057490790737e-05, "loss": 0.6043405532836914, "memory(GiB)": 73.83, "step": 86290, "token_acc": 0.8664596273291926, "train_speed(iter/s)": 1.456149 }, { "epoch": 3.6971423675078188, "grad_norm": 2.8716719150543213, "learning_rate": 1.583814312425102e-05, "loss": 0.49538397789001465, "memory(GiB)": 73.83, "step": 86295, "token_acc": 0.8613138686131386, "train_speed(iter/s)": 1.456152 }, { "epoch": 3.6973565828370676, "grad_norm": 2.2736103534698486, "learning_rate": 1.5833229376585402e-05, "loss": 0.2223264217376709, "memory(GiB)": 73.83, "step": 86300, "token_acc": 0.9442508710801394, "train_speed(iter/s)": 1.456148 }, { "epoch": 3.697570798166317, "grad_norm": 3.163020133972168, "learning_rate": 1.5828316247882884e-05, "loss": 0.39141533374786375, "memory(GiB)": 73.83, "step": 86305, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.456145 }, { "epoch": 3.6977850134955657, "grad_norm": 4.793633460998535, "learning_rate": 1.582340373823248e-05, "loss": 0.5446385860443115, "memory(GiB)": 73.83, "step": 86310, "token_acc": 0.892128279883382, "train_speed(iter/s)": 1.456158 }, { "epoch": 3.6979992288248145, "grad_norm": 0.1738761067390442, "learning_rate": 1.581849184772315e-05, "loss": 0.204237961769104, "memory(GiB)": 73.83, "step": 86315, "token_acc": 0.9625850340136054, "train_speed(iter/s)": 1.456161 }, { "epoch": 3.6982134441540637, "grad_norm": 2.8817601203918457, "learning_rate": 1.581358057644393e-05, "loss": 0.1781486988067627, "memory(GiB)": 73.83, "step": 86320, "token_acc": 0.9522184300341296, "train_speed(iter/s)": 1.456159 }, { "epoch": 3.6984276594833125, "grad_norm": 4.730991363525391, "learning_rate": 1.5808669924483765e-05, "loss": 0.19211474657058716, "memory(GiB)": 73.83, "step": 86325, "token_acc": 0.9584905660377359, "train_speed(iter/s)": 1.456165 }, { "epoch": 3.6986418748125613, "grad_norm": 4.814511299133301, "learning_rate": 1.5803759891931613e-05, "loss": 0.28403291702270506, "memory(GiB)": 73.83, "step": 86330, "token_acc": 0.937888198757764, "train_speed(iter/s)": 1.456171 }, { "epoch": 3.6988560901418106, "grad_norm": 2.8096559047698975, "learning_rate": 1.579885047887644e-05, "loss": 0.3727332830429077, "memory(GiB)": 73.83, "step": 86335, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.456173 }, { "epoch": 3.6990703054710594, "grad_norm": 2.0895299911499023, "learning_rate": 1.5793941685407165e-05, "loss": 0.3703216314315796, "memory(GiB)": 73.83, "step": 86340, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.456175 }, { "epoch": 3.6992845208003082, "grad_norm": 3.9632740020751953, "learning_rate": 1.5789033511612712e-05, "loss": 0.5556241989135742, "memory(GiB)": 73.83, "step": 86345, "token_acc": 0.90625, "train_speed(iter/s)": 1.456187 }, { "epoch": 3.6994987361295575, "grad_norm": 2.1718599796295166, "learning_rate": 1.5784125957582025e-05, "loss": 0.3322957277297974, "memory(GiB)": 73.83, "step": 86350, "token_acc": 0.946360153256705, "train_speed(iter/s)": 1.456191 }, { "epoch": 3.6997129514588063, "grad_norm": 3.3509225845336914, "learning_rate": 1.577921902340399e-05, "loss": 0.41789674758911133, "memory(GiB)": 73.83, "step": 86355, "token_acc": 0.9115853658536586, "train_speed(iter/s)": 1.4562 }, { "epoch": 3.699927166788055, "grad_norm": 1.1213862895965576, "learning_rate": 1.5774312709167505e-05, "loss": 0.09223368167877197, "memory(GiB)": 73.83, "step": 86360, "token_acc": 0.9820143884892086, "train_speed(iter/s)": 1.456203 }, { "epoch": 3.7001413821173044, "grad_norm": 1.4829670190811157, "learning_rate": 1.576940701496144e-05, "loss": 0.10668038129806519, "memory(GiB)": 73.83, "step": 86365, "token_acc": 0.9803921568627451, "train_speed(iter/s)": 1.456212 }, { "epoch": 3.700355597446553, "grad_norm": 3.350041627883911, "learning_rate": 1.5764501940874688e-05, "loss": 0.29639368057250975, "memory(GiB)": 73.83, "step": 86370, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.456217 }, { "epoch": 3.700569812775802, "grad_norm": 2.77520751953125, "learning_rate": 1.5759597486996086e-05, "loss": 0.3453402042388916, "memory(GiB)": 73.83, "step": 86375, "token_acc": 0.9274924471299094, "train_speed(iter/s)": 1.456235 }, { "epoch": 3.7007840281050512, "grad_norm": 1.8816004991531372, "learning_rate": 1.5754693653414515e-05, "loss": 0.23163506984710694, "memory(GiB)": 73.83, "step": 86380, "token_acc": 0.9479166666666666, "train_speed(iter/s)": 1.456247 }, { "epoch": 3.7009982434343, "grad_norm": 3.305325984954834, "learning_rate": 1.5749790440218787e-05, "loss": 0.3815601348876953, "memory(GiB)": 73.83, "step": 86385, "token_acc": 0.928125, "train_speed(iter/s)": 1.456252 }, { "epoch": 3.701212458763549, "grad_norm": 3.1052589416503906, "learning_rate": 1.5744887847497735e-05, "loss": 0.40811829566955565, "memory(GiB)": 73.83, "step": 86390, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.456273 }, { "epoch": 3.701426674092798, "grad_norm": 3.9341483116149902, "learning_rate": 1.573998587534018e-05, "loss": 0.4036720752716064, "memory(GiB)": 73.83, "step": 86395, "token_acc": 0.8905775075987842, "train_speed(iter/s)": 1.456273 }, { "epoch": 3.701640889422047, "grad_norm": 5.993445873260498, "learning_rate": 1.5735084523834913e-05, "loss": 0.34098291397094727, "memory(GiB)": 73.83, "step": 86400, "token_acc": 0.9240121580547113, "train_speed(iter/s)": 1.456274 }, { "epoch": 3.7018551047512958, "grad_norm": 2.8073158264160156, "learning_rate": 1.573018379307072e-05, "loss": 0.2575897932052612, "memory(GiB)": 73.83, "step": 86405, "token_acc": 0.9401993355481728, "train_speed(iter/s)": 1.456274 }, { "epoch": 3.702069320080545, "grad_norm": 4.19777250289917, "learning_rate": 1.572528368313641e-05, "loss": 0.25591697692871096, "memory(GiB)": 73.83, "step": 86410, "token_acc": 0.9372822299651568, "train_speed(iter/s)": 1.456276 }, { "epoch": 3.702283535409794, "grad_norm": 3.5293128490448, "learning_rate": 1.572038419412074e-05, "loss": 0.47513651847839355, "memory(GiB)": 73.83, "step": 86415, "token_acc": 0.9073482428115016, "train_speed(iter/s)": 1.456282 }, { "epoch": 3.7024977507390426, "grad_norm": 2.887653350830078, "learning_rate": 1.5715485326112467e-05, "loss": 0.30338387489318847, "memory(GiB)": 73.83, "step": 86420, "token_acc": 0.9260450160771704, "train_speed(iter/s)": 1.456287 }, { "epoch": 3.702711966068292, "grad_norm": 6.79017972946167, "learning_rate": 1.5710587079200335e-05, "loss": 0.40898609161376953, "memory(GiB)": 73.83, "step": 86425, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.45629 }, { "epoch": 3.7029261813975407, "grad_norm": 0.507209300994873, "learning_rate": 1.5705689453473072e-05, "loss": 0.35961825847625734, "memory(GiB)": 73.83, "step": 86430, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.456291 }, { "epoch": 3.7031403967267895, "grad_norm": 2.2588517665863037, "learning_rate": 1.570079244901943e-05, "loss": 0.5477112293243408, "memory(GiB)": 73.83, "step": 86435, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.456294 }, { "epoch": 3.7033546120560388, "grad_norm": 6.895777702331543, "learning_rate": 1.569589606592809e-05, "loss": 0.35275750160217284, "memory(GiB)": 73.83, "step": 86440, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.456297 }, { "epoch": 3.7035688273852876, "grad_norm": 3.892441749572754, "learning_rate": 1.569100030428779e-05, "loss": 0.5049467563629151, "memory(GiB)": 73.83, "step": 86445, "token_acc": 0.8856088560885609, "train_speed(iter/s)": 1.456297 }, { "epoch": 3.7037830427145364, "grad_norm": 3.3534200191497803, "learning_rate": 1.568610516418721e-05, "loss": 0.24745090007781984, "memory(GiB)": 73.83, "step": 86450, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.45631 }, { "epoch": 3.7039972580437857, "grad_norm": 0.13415198028087616, "learning_rate": 1.5681210645715017e-05, "loss": 0.18126752376556396, "memory(GiB)": 73.83, "step": 86455, "token_acc": 0.9661016949152542, "train_speed(iter/s)": 1.45631 }, { "epoch": 3.7042114733730345, "grad_norm": 0.49916204810142517, "learning_rate": 1.567631674895987e-05, "loss": 0.2538217782974243, "memory(GiB)": 73.83, "step": 86460, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.456315 }, { "epoch": 3.7044256887022833, "grad_norm": 5.125585079193115, "learning_rate": 1.567142347401046e-05, "loss": 0.3713226318359375, "memory(GiB)": 73.83, "step": 86465, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.456313 }, { "epoch": 3.7046399040315325, "grad_norm": 0.13969211280345917, "learning_rate": 1.566653082095542e-05, "loss": 0.297475266456604, "memory(GiB)": 73.83, "step": 86470, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.456314 }, { "epoch": 3.7048541193607814, "grad_norm": 2.605316162109375, "learning_rate": 1.566163878988338e-05, "loss": 0.24478743076324463, "memory(GiB)": 73.83, "step": 86475, "token_acc": 0.9467084639498433, "train_speed(iter/s)": 1.456317 }, { "epoch": 3.70506833469003, "grad_norm": 3.543822765350342, "learning_rate": 1.5656747380882965e-05, "loss": 0.3264608383178711, "memory(GiB)": 73.83, "step": 86480, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.456337 }, { "epoch": 3.7052825500192794, "grad_norm": 2.538923740386963, "learning_rate": 1.565185659404279e-05, "loss": 0.2951693296432495, "memory(GiB)": 73.83, "step": 86485, "token_acc": 0.9358490566037736, "train_speed(iter/s)": 1.456343 }, { "epoch": 3.7054967653485282, "grad_norm": 6.067057132720947, "learning_rate": 1.564696642945143e-05, "loss": 0.3979796409606934, "memory(GiB)": 73.83, "step": 86490, "token_acc": 0.9358490566037736, "train_speed(iter/s)": 1.456346 }, { "epoch": 3.705710980677777, "grad_norm": 3.0073740482330322, "learning_rate": 1.5642076887197527e-05, "loss": 0.4721244812011719, "memory(GiB)": 73.83, "step": 86495, "token_acc": 0.9084967320261438, "train_speed(iter/s)": 1.456345 }, { "epoch": 3.7059251960070263, "grad_norm": 3.4158332347869873, "learning_rate": 1.563718796736962e-05, "loss": 0.27952749729156495, "memory(GiB)": 73.83, "step": 86500, "token_acc": 0.9347181008902077, "train_speed(iter/s)": 1.456353 }, { "epoch": 3.7059251960070263, "eval_loss": 2.3224668502807617, "eval_runtime": 11.0459, "eval_samples_per_second": 9.053, "eval_steps_per_second": 9.053, "eval_token_acc": 0.4894179894179894, "step": 86500 }, { "epoch": 3.706139411336275, "grad_norm": 2.5008304119110107, "learning_rate": 1.5632299670056296e-05, "loss": 0.17160611152648925, "memory(GiB)": 73.83, "step": 86505, "token_acc": 0.6132167152575316, "train_speed(iter/s)": 1.456062 }, { "epoch": 3.706353626665524, "grad_norm": 2.4328625202178955, "learning_rate": 1.5627411995346105e-05, "loss": 0.5757106304168701, "memory(GiB)": 73.83, "step": 86510, "token_acc": 0.8685121107266436, "train_speed(iter/s)": 1.456076 }, { "epoch": 3.706567841994773, "grad_norm": 4.219996452331543, "learning_rate": 1.5622524943327567e-05, "loss": 0.3288883686065674, "memory(GiB)": 73.83, "step": 86515, "token_acc": 0.9317507418397626, "train_speed(iter/s)": 1.456086 }, { "epoch": 3.706782057324022, "grad_norm": 3.865825653076172, "learning_rate": 1.5617638514089237e-05, "loss": 0.397607159614563, "memory(GiB)": 73.83, "step": 86520, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.456086 }, { "epoch": 3.706996272653271, "grad_norm": 3.093689203262329, "learning_rate": 1.561275270771966e-05, "loss": 0.3281337976455688, "memory(GiB)": 73.83, "step": 86525, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.456091 }, { "epoch": 3.70721048798252, "grad_norm": 3.290799379348755, "learning_rate": 1.5607867524307323e-05, "loss": 0.1496854305267334, "memory(GiB)": 73.83, "step": 86530, "token_acc": 0.9742765273311897, "train_speed(iter/s)": 1.456099 }, { "epoch": 3.707424703311769, "grad_norm": 0.9401967525482178, "learning_rate": 1.5602982963940726e-05, "loss": 0.3814944505691528, "memory(GiB)": 73.83, "step": 86535, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.45612 }, { "epoch": 3.7076389186410177, "grad_norm": 2.6221728324890137, "learning_rate": 1.559809902670836e-05, "loss": 0.5684037208557129, "memory(GiB)": 73.83, "step": 86540, "token_acc": 0.8930041152263375, "train_speed(iter/s)": 1.456123 }, { "epoch": 3.707853133970267, "grad_norm": 3.948894500732422, "learning_rate": 1.5593215712698705e-05, "loss": 0.5553199768066406, "memory(GiB)": 73.83, "step": 86545, "token_acc": 0.86, "train_speed(iter/s)": 1.456129 }, { "epoch": 3.7080673492995158, "grad_norm": 3.156867265701294, "learning_rate": 1.5588333022000212e-05, "loss": 0.665732479095459, "memory(GiB)": 73.83, "step": 86550, "token_acc": 0.8586956521739131, "train_speed(iter/s)": 1.456143 }, { "epoch": 3.7082815646287646, "grad_norm": 2.647373676300049, "learning_rate": 1.5583450954701356e-05, "loss": 0.45194687843322756, "memory(GiB)": 73.83, "step": 86555, "token_acc": 0.8945454545454545, "train_speed(iter/s)": 1.456147 }, { "epoch": 3.708495779958014, "grad_norm": 2.694133996963501, "learning_rate": 1.5578569510890574e-05, "loss": 0.3717078685760498, "memory(GiB)": 73.83, "step": 86560, "token_acc": 0.9417989417989417, "train_speed(iter/s)": 1.456151 }, { "epoch": 3.7087099952872626, "grad_norm": 3.8943583965301514, "learning_rate": 1.557368869065629e-05, "loss": 0.22162089347839356, "memory(GiB)": 73.83, "step": 86565, "token_acc": 0.9601226993865031, "train_speed(iter/s)": 1.456153 }, { "epoch": 3.7089242106165115, "grad_norm": 1.316672921180725, "learning_rate": 1.5568808494086933e-05, "loss": 0.10968071222305298, "memory(GiB)": 73.83, "step": 86570, "token_acc": 0.9715189873417721, "train_speed(iter/s)": 1.456152 }, { "epoch": 3.7091384259457607, "grad_norm": 0.4705579876899719, "learning_rate": 1.5563928921270898e-05, "loss": 0.33449602127075195, "memory(GiB)": 73.83, "step": 86575, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.456152 }, { "epoch": 3.7093526412750095, "grad_norm": 1.9868298768997192, "learning_rate": 1.5559049972296606e-05, "loss": 0.29568612575531006, "memory(GiB)": 73.83, "step": 86580, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.456163 }, { "epoch": 3.7095668566042583, "grad_norm": 2.5071280002593994, "learning_rate": 1.555417164725243e-05, "loss": 0.33300812244415284, "memory(GiB)": 73.83, "step": 86585, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.456165 }, { "epoch": 3.7097810719335076, "grad_norm": 3.7040719985961914, "learning_rate": 1.5549293946226734e-05, "loss": 0.18733142614364623, "memory(GiB)": 73.83, "step": 86590, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.456172 }, { "epoch": 3.7099952872627564, "grad_norm": 1.9654927253723145, "learning_rate": 1.5544416869307915e-05, "loss": 0.38363988399505616, "memory(GiB)": 73.83, "step": 86595, "token_acc": 0.9196141479099679, "train_speed(iter/s)": 1.456175 }, { "epoch": 3.710209502592005, "grad_norm": 1.6895934343338013, "learning_rate": 1.5539540416584304e-05, "loss": 0.3300363540649414, "memory(GiB)": 73.83, "step": 86600, "token_acc": 0.9451219512195121, "train_speed(iter/s)": 1.456176 }, { "epoch": 3.7104237179212545, "grad_norm": 1.1763430833816528, "learning_rate": 1.5534664588144226e-05, "loss": 0.26940183639526366, "memory(GiB)": 73.83, "step": 86605, "token_acc": 0.9539007092198581, "train_speed(iter/s)": 1.456181 }, { "epoch": 3.7106379332505033, "grad_norm": 0.6406923532485962, "learning_rate": 1.552978938407605e-05, "loss": 0.1981581449508667, "memory(GiB)": 73.83, "step": 86610, "token_acc": 0.9559322033898305, "train_speed(iter/s)": 1.456183 }, { "epoch": 3.710852148579752, "grad_norm": 6.2750020027160645, "learning_rate": 1.552491480446807e-05, "loss": 0.4411789894104004, "memory(GiB)": 73.83, "step": 86615, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.456186 }, { "epoch": 3.7110663639090014, "grad_norm": 3.9004828929901123, "learning_rate": 1.55200408494086e-05, "loss": 0.21809759140014648, "memory(GiB)": 73.83, "step": 86620, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.456195 }, { "epoch": 3.71128057923825, "grad_norm": 3.414832830429077, "learning_rate": 1.5515167518985933e-05, "loss": 0.35066976547241213, "memory(GiB)": 73.83, "step": 86625, "token_acc": 0.9182879377431906, "train_speed(iter/s)": 1.456198 }, { "epoch": 3.711494794567499, "grad_norm": 2.264183282852173, "learning_rate": 1.551029481328836e-05, "loss": 0.21008901596069335, "memory(GiB)": 73.83, "step": 86630, "token_acc": 0.9451219512195121, "train_speed(iter/s)": 1.456202 }, { "epoch": 3.7117090098967482, "grad_norm": 2.996159076690674, "learning_rate": 1.5505422732404134e-05, "loss": 0.3993529796600342, "memory(GiB)": 73.83, "step": 86635, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.456204 }, { "epoch": 3.711923225225997, "grad_norm": 4.434667587280273, "learning_rate": 1.550055127642155e-05, "loss": 0.5166035175323487, "memory(GiB)": 73.83, "step": 86640, "token_acc": 0.8981132075471698, "train_speed(iter/s)": 1.456209 }, { "epoch": 3.712137440555246, "grad_norm": 6.50139856338501, "learning_rate": 1.549568044542884e-05, "loss": 0.46047163009643555, "memory(GiB)": 73.83, "step": 86645, "token_acc": 0.9129032258064517, "train_speed(iter/s)": 1.456207 }, { "epoch": 3.712351655884495, "grad_norm": 6.109774112701416, "learning_rate": 1.5490810239514254e-05, "loss": 0.4523061752319336, "memory(GiB)": 73.83, "step": 86650, "token_acc": 0.896414342629482, "train_speed(iter/s)": 1.45621 }, { "epoch": 3.712565871213744, "grad_norm": 1.8566553592681885, "learning_rate": 1.5485940658766e-05, "loss": 0.43149394989013673, "memory(GiB)": 73.83, "step": 86655, "token_acc": 0.8881789137380192, "train_speed(iter/s)": 1.456224 }, { "epoch": 3.7127800865429927, "grad_norm": 4.801576137542725, "learning_rate": 1.5481071703272304e-05, "loss": 0.5001307964324951, "memory(GiB)": 73.83, "step": 86660, "token_acc": 0.8987341772151899, "train_speed(iter/s)": 1.456222 }, { "epoch": 3.712994301872242, "grad_norm": 3.267465114593506, "learning_rate": 1.547620337312137e-05, "loss": 0.3090497970581055, "memory(GiB)": 73.83, "step": 86665, "token_acc": 0.9430604982206405, "train_speed(iter/s)": 1.456236 }, { "epoch": 3.713208517201491, "grad_norm": 2.1680099964141846, "learning_rate": 1.547133566840141e-05, "loss": 0.24562594890594483, "memory(GiB)": 73.83, "step": 86670, "token_acc": 0.935672514619883, "train_speed(iter/s)": 1.456236 }, { "epoch": 3.7134227325307396, "grad_norm": 4.225766181945801, "learning_rate": 1.54664685892006e-05, "loss": 0.38193306922912595, "memory(GiB)": 73.83, "step": 86675, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.456242 }, { "epoch": 3.713636947859989, "grad_norm": 2.042895793914795, "learning_rate": 1.5461602135607105e-05, "loss": 0.33429670333862305, "memory(GiB)": 73.83, "step": 86680, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.456244 }, { "epoch": 3.7138511631892377, "grad_norm": 2.486539840698242, "learning_rate": 1.545673630770909e-05, "loss": 0.3747528076171875, "memory(GiB)": 73.83, "step": 86685, "token_acc": 0.9036544850498339, "train_speed(iter/s)": 1.456249 }, { "epoch": 3.714065378518487, "grad_norm": 3.698108196258545, "learning_rate": 1.5451871105594696e-05, "loss": 0.2893351078033447, "memory(GiB)": 73.83, "step": 86690, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.456255 }, { "epoch": 3.7142795938477358, "grad_norm": 5.320209503173828, "learning_rate": 1.5447006529352054e-05, "loss": 0.5064035892486572, "memory(GiB)": 73.83, "step": 86695, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.456259 }, { "epoch": 3.7144938091769846, "grad_norm": 4.384225845336914, "learning_rate": 1.5442142579069315e-05, "loss": 0.19172627925872804, "memory(GiB)": 73.83, "step": 86700, "token_acc": 0.9420849420849421, "train_speed(iter/s)": 1.456258 }, { "epoch": 3.714708024506234, "grad_norm": 3.5765156745910645, "learning_rate": 1.543727925483458e-05, "loss": 0.19160308837890624, "memory(GiB)": 73.83, "step": 86705, "token_acc": 0.9704641350210971, "train_speed(iter/s)": 1.456256 }, { "epoch": 3.7149222398354826, "grad_norm": 2.1006577014923096, "learning_rate": 1.5432416556735957e-05, "loss": 0.25961012840270997, "memory(GiB)": 73.83, "step": 86710, "token_acc": 0.9233038348082596, "train_speed(iter/s)": 1.456257 }, { "epoch": 3.7151364551647315, "grad_norm": 6.88067102432251, "learning_rate": 1.5427554484861532e-05, "loss": 0.45476608276367186, "memory(GiB)": 73.83, "step": 86715, "token_acc": 0.9343629343629344, "train_speed(iter/s)": 1.456259 }, { "epoch": 3.7153506704939807, "grad_norm": 0.7711949348449707, "learning_rate": 1.542269303929938e-05, "loss": 0.2785215377807617, "memory(GiB)": 73.83, "step": 86720, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.45627 }, { "epoch": 3.7155648858232295, "grad_norm": 6.3546061515808105, "learning_rate": 1.541783222013759e-05, "loss": 0.2152684211730957, "memory(GiB)": 73.83, "step": 86725, "token_acc": 0.9503311258278145, "train_speed(iter/s)": 1.45628 }, { "epoch": 3.7157791011524783, "grad_norm": 4.0721845626831055, "learning_rate": 1.5412972027464213e-05, "loss": 0.14652954339981078, "memory(GiB)": 73.83, "step": 86730, "token_acc": 0.9591078066914498, "train_speed(iter/s)": 1.456284 }, { "epoch": 3.7159933164817276, "grad_norm": 4.654726028442383, "learning_rate": 1.5408112461367268e-05, "loss": 0.5859488487243653, "memory(GiB)": 73.83, "step": 86735, "token_acc": 0.8996763754045307, "train_speed(iter/s)": 1.456298 }, { "epoch": 3.7162075318109764, "grad_norm": 3.037738561630249, "learning_rate": 1.540325352193484e-05, "loss": 0.4853825092315674, "memory(GiB)": 73.83, "step": 86740, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.456299 }, { "epoch": 3.716421747140225, "grad_norm": 8.642136573791504, "learning_rate": 1.539839520925493e-05, "loss": 0.7932897090911866, "memory(GiB)": 73.83, "step": 86745, "token_acc": 0.8173652694610778, "train_speed(iter/s)": 1.456297 }, { "epoch": 3.7166359624694745, "grad_norm": 2.207895040512085, "learning_rate": 1.539353752341553e-05, "loss": 0.16908681392669678, "memory(GiB)": 73.83, "step": 86750, "token_acc": 0.9665653495440729, "train_speed(iter/s)": 1.4563 }, { "epoch": 3.7168501777987233, "grad_norm": 4.628236293792725, "learning_rate": 1.538868046450468e-05, "loss": 0.4637720584869385, "memory(GiB)": 73.83, "step": 86755, "token_acc": 0.902027027027027, "train_speed(iter/s)": 1.456298 }, { "epoch": 3.717064393127972, "grad_norm": 3.346383571624756, "learning_rate": 1.538382403261035e-05, "loss": 0.4588298797607422, "memory(GiB)": 73.83, "step": 86760, "token_acc": 0.9027355623100304, "train_speed(iter/s)": 1.456299 }, { "epoch": 3.7172786084572214, "grad_norm": 0.7814935445785522, "learning_rate": 1.5378968227820518e-05, "loss": 0.23215894699096679, "memory(GiB)": 73.83, "step": 86765, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.456309 }, { "epoch": 3.71749282378647, "grad_norm": 1.750460147857666, "learning_rate": 1.5374113050223153e-05, "loss": 0.3060901641845703, "memory(GiB)": 73.83, "step": 86770, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.456311 }, { "epoch": 3.717707039115719, "grad_norm": 3.2228598594665527, "learning_rate": 1.5369258499906215e-05, "loss": 0.4779972076416016, "memory(GiB)": 73.83, "step": 86775, "token_acc": 0.8897338403041825, "train_speed(iter/s)": 1.456325 }, { "epoch": 3.7179212544449682, "grad_norm": 7.056955814361572, "learning_rate": 1.536440457695763e-05, "loss": 0.33585894107818604, "memory(GiB)": 73.83, "step": 86780, "token_acc": 0.9190031152647975, "train_speed(iter/s)": 1.456326 }, { "epoch": 3.718135469774217, "grad_norm": 3.0679073333740234, "learning_rate": 1.5359551281465363e-05, "loss": 0.26288666725158694, "memory(GiB)": 73.83, "step": 86785, "token_acc": 0.9530516431924883, "train_speed(iter/s)": 1.45633 }, { "epoch": 3.718349685103466, "grad_norm": 1.2164638042449951, "learning_rate": 1.5354698613517316e-05, "loss": 0.19422628879547119, "memory(GiB)": 73.83, "step": 86790, "token_acc": 0.9501915708812261, "train_speed(iter/s)": 1.456342 }, { "epoch": 3.718563900432715, "grad_norm": 3.1410534381866455, "learning_rate": 1.534984657320141e-05, "loss": 0.239229154586792, "memory(GiB)": 73.83, "step": 86795, "token_acc": 0.9538043478260869, "train_speed(iter/s)": 1.456347 }, { "epoch": 3.718778115761964, "grad_norm": 3.1731224060058594, "learning_rate": 1.534499516060553e-05, "loss": 0.3527015209197998, "memory(GiB)": 73.83, "step": 86800, "token_acc": 0.9119718309859155, "train_speed(iter/s)": 1.456349 }, { "epoch": 3.7189923310912127, "grad_norm": 2.087923049926758, "learning_rate": 1.534014437581756e-05, "loss": 0.29475719928741456, "memory(GiB)": 73.83, "step": 86805, "token_acc": 0.9391891891891891, "train_speed(iter/s)": 1.456355 }, { "epoch": 3.719206546420462, "grad_norm": 4.8416595458984375, "learning_rate": 1.5335294218925388e-05, "loss": 0.2659485101699829, "memory(GiB)": 73.83, "step": 86810, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.456378 }, { "epoch": 3.719420761749711, "grad_norm": 2.13016414642334, "learning_rate": 1.53304446900169e-05, "loss": 0.5101282119750976, "memory(GiB)": 73.83, "step": 86815, "token_acc": 0.9, "train_speed(iter/s)": 1.456382 }, { "epoch": 3.7196349770789596, "grad_norm": 5.698002815246582, "learning_rate": 1.532559578917992e-05, "loss": 0.3719517707824707, "memory(GiB)": 73.83, "step": 86820, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.456387 }, { "epoch": 3.719849192408209, "grad_norm": 6.162614345550537, "learning_rate": 1.5320747516502303e-05, "loss": 0.34114985466003417, "memory(GiB)": 73.83, "step": 86825, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.456397 }, { "epoch": 3.7200634077374577, "grad_norm": 5.307458877563477, "learning_rate": 1.5315899872071882e-05, "loss": 0.2796562433242798, "memory(GiB)": 73.83, "step": 86830, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.456402 }, { "epoch": 3.7202776230667065, "grad_norm": 2.8511080741882324, "learning_rate": 1.5311052855976465e-05, "loss": 0.3198474645614624, "memory(GiB)": 73.83, "step": 86835, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.456403 }, { "epoch": 3.7204918383959558, "grad_norm": 3.440875768661499, "learning_rate": 1.530620646830385e-05, "loss": 0.3567221164703369, "memory(GiB)": 73.83, "step": 86840, "token_acc": 0.93646408839779, "train_speed(iter/s)": 1.456408 }, { "epoch": 3.7207060537252046, "grad_norm": 2.2378268241882324, "learning_rate": 1.530136070914187e-05, "loss": 0.1804051637649536, "memory(GiB)": 73.83, "step": 86845, "token_acc": 0.9652509652509652, "train_speed(iter/s)": 1.456408 }, { "epoch": 3.7209202690544534, "grad_norm": 3.337488889694214, "learning_rate": 1.5296515578578285e-05, "loss": 0.42589964866638186, "memory(GiB)": 73.83, "step": 86850, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.456411 }, { "epoch": 3.7211344843837026, "grad_norm": 2.234362840652466, "learning_rate": 1.5291671076700882e-05, "loss": 0.4944146633148193, "memory(GiB)": 73.83, "step": 86855, "token_acc": 0.8851963746223565, "train_speed(iter/s)": 1.45642 }, { "epoch": 3.7213486997129515, "grad_norm": 1.3373913764953613, "learning_rate": 1.5286827203597414e-05, "loss": 0.29716331958770753, "memory(GiB)": 73.83, "step": 86860, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.456422 }, { "epoch": 3.7215629150422003, "grad_norm": 3.803710460662842, "learning_rate": 1.5281983959355615e-05, "loss": 0.34403190612792967, "memory(GiB)": 73.83, "step": 86865, "token_acc": 0.9143730886850153, "train_speed(iter/s)": 1.456425 }, { "epoch": 3.7217771303714495, "grad_norm": 4.220864772796631, "learning_rate": 1.5277141344063267e-05, "loss": 0.4485680103302002, "memory(GiB)": 73.83, "step": 86870, "token_acc": 0.9, "train_speed(iter/s)": 1.45643 }, { "epoch": 3.7219913457006983, "grad_norm": 2.3369951248168945, "learning_rate": 1.5272299357808072e-05, "loss": 0.15147160291671752, "memory(GiB)": 73.83, "step": 86875, "token_acc": 0.9577922077922078, "train_speed(iter/s)": 1.456434 }, { "epoch": 3.722205561029947, "grad_norm": 4.0533671379089355, "learning_rate": 1.5267458000677753e-05, "loss": 0.5562352180480957, "memory(GiB)": 73.83, "step": 86880, "token_acc": 0.896414342629482, "train_speed(iter/s)": 1.456455 }, { "epoch": 3.7224197763591964, "grad_norm": 4.939702033996582, "learning_rate": 1.5262617272759993e-05, "loss": 0.4288327217102051, "memory(GiB)": 73.83, "step": 86885, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.456454 }, { "epoch": 3.722633991688445, "grad_norm": 2.0989797115325928, "learning_rate": 1.5257777174142529e-05, "loss": 0.24543309211730957, "memory(GiB)": 73.83, "step": 86890, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.456457 }, { "epoch": 3.722848207017694, "grad_norm": 2.842198610305786, "learning_rate": 1.5252937704913006e-05, "loss": 0.4642052173614502, "memory(GiB)": 73.83, "step": 86895, "token_acc": 0.8898809523809523, "train_speed(iter/s)": 1.456464 }, { "epoch": 3.7230624223469433, "grad_norm": 4.787871837615967, "learning_rate": 1.5248098865159127e-05, "loss": 0.18455131053924562, "memory(GiB)": 73.83, "step": 86900, "token_acc": 0.9670781893004116, "train_speed(iter/s)": 1.456466 }, { "epoch": 3.723276637676192, "grad_norm": 0.5029523968696594, "learning_rate": 1.5243260654968539e-05, "loss": 0.179803204536438, "memory(GiB)": 73.83, "step": 86905, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.456468 }, { "epoch": 3.723490853005441, "grad_norm": 1.3164472579956055, "learning_rate": 1.5238423074428888e-05, "loss": 0.13965196609497071, "memory(GiB)": 73.83, "step": 86910, "token_acc": 0.9595141700404858, "train_speed(iter/s)": 1.456474 }, { "epoch": 3.72370506833469, "grad_norm": 3.5303428173065186, "learning_rate": 1.5233586123627807e-05, "loss": 0.6621017932891846, "memory(GiB)": 73.83, "step": 86915, "token_acc": 0.8389610389610389, "train_speed(iter/s)": 1.456486 }, { "epoch": 3.723919283663939, "grad_norm": 1.9242618083953857, "learning_rate": 1.5228749802652932e-05, "loss": 0.48645291328430174, "memory(GiB)": 73.83, "step": 86920, "token_acc": 0.8876811594202898, "train_speed(iter/s)": 1.456487 }, { "epoch": 3.724133498993188, "grad_norm": 0.11569841206073761, "learning_rate": 1.5223914111591853e-05, "loss": 0.3594342231750488, "memory(GiB)": 73.83, "step": 86925, "token_acc": 0.9322709163346613, "train_speed(iter/s)": 1.456494 }, { "epoch": 3.724347714322437, "grad_norm": 2.152226209640503, "learning_rate": 1.5219079050532209e-05, "loss": 0.2989119291305542, "memory(GiB)": 73.83, "step": 86930, "token_acc": 0.942652329749104, "train_speed(iter/s)": 1.45649 }, { "epoch": 3.724561929651686, "grad_norm": 4.2679948806762695, "learning_rate": 1.521424461956158e-05, "loss": 0.32949802875518797, "memory(GiB)": 73.83, "step": 86935, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.456492 }, { "epoch": 3.7247761449809347, "grad_norm": 0.6844266057014465, "learning_rate": 1.5209410818767539e-05, "loss": 0.330893874168396, "memory(GiB)": 73.83, "step": 86940, "token_acc": 0.9131944444444444, "train_speed(iter/s)": 1.4565 }, { "epoch": 3.724990360310184, "grad_norm": 2.5301647186279297, "learning_rate": 1.5204577648237656e-05, "loss": 0.17922425270080566, "memory(GiB)": 73.83, "step": 86945, "token_acc": 0.9533333333333334, "train_speed(iter/s)": 1.456503 }, { "epoch": 3.7252045756394327, "grad_norm": 5.7215256690979, "learning_rate": 1.519974510805947e-05, "loss": 0.43796358108520506, "memory(GiB)": 73.83, "step": 86950, "token_acc": 0.9204244031830239, "train_speed(iter/s)": 1.456511 }, { "epoch": 3.7254187909686816, "grad_norm": 0.9288546442985535, "learning_rate": 1.519491319832057e-05, "loss": 0.19836182594299318, "memory(GiB)": 73.83, "step": 86955, "token_acc": 0.9591836734693877, "train_speed(iter/s)": 1.456516 }, { "epoch": 3.725633006297931, "grad_norm": 3.9273626804351807, "learning_rate": 1.5190081919108445e-05, "loss": 0.2615997791290283, "memory(GiB)": 73.83, "step": 86960, "token_acc": 0.9463722397476341, "train_speed(iter/s)": 1.456517 }, { "epoch": 3.7258472216271796, "grad_norm": 2.673947334289551, "learning_rate": 1.5185251270510659e-05, "loss": 0.49450030326843264, "memory(GiB)": 73.83, "step": 86965, "token_acc": 0.8881118881118881, "train_speed(iter/s)": 1.456521 }, { "epoch": 3.7260614369564284, "grad_norm": 0.11244572699069977, "learning_rate": 1.5180421252614707e-05, "loss": 0.1952394127845764, "memory(GiB)": 73.83, "step": 86970, "token_acc": 0.9486166007905138, "train_speed(iter/s)": 1.456521 }, { "epoch": 3.7262756522856777, "grad_norm": 1.6918145418167114, "learning_rate": 1.5175591865508082e-05, "loss": 0.2135148286819458, "memory(GiB)": 73.83, "step": 86975, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.456525 }, { "epoch": 3.7264898676149265, "grad_norm": 3.082714796066284, "learning_rate": 1.5170763109278285e-05, "loss": 0.44210500717163087, "memory(GiB)": 73.83, "step": 86980, "token_acc": 0.9083969465648855, "train_speed(iter/s)": 1.456537 }, { "epoch": 3.7267040829441753, "grad_norm": 2.2541584968566895, "learning_rate": 1.5165934984012769e-05, "loss": 0.23516664505004883, "memory(GiB)": 73.83, "step": 86985, "token_acc": 0.9552238805970149, "train_speed(iter/s)": 1.456536 }, { "epoch": 3.7269182982734246, "grad_norm": 5.646673679351807, "learning_rate": 1.5161107489799032e-05, "loss": 0.5169890403747559, "memory(GiB)": 73.83, "step": 86990, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.456539 }, { "epoch": 3.7271325136026734, "grad_norm": 1.6392720937728882, "learning_rate": 1.5156280626724512e-05, "loss": 0.20703017711639404, "memory(GiB)": 73.83, "step": 86995, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.45655 }, { "epoch": 3.727346728931922, "grad_norm": 1.1582248210906982, "learning_rate": 1.5151454394876658e-05, "loss": 0.18393789529800414, "memory(GiB)": 73.83, "step": 87000, "token_acc": 0.9686274509803922, "train_speed(iter/s)": 1.45655 }, { "epoch": 3.727346728931922, "eval_loss": 2.5674145221710205, "eval_runtime": 11.4436, "eval_samples_per_second": 8.738, "eval_steps_per_second": 8.738, "eval_token_acc": 0.45861601085481685, "step": 87000 }, { "epoch": 3.7275609442611715, "grad_norm": 11.885425567626953, "learning_rate": 1.5146628794342898e-05, "loss": 0.185082745552063, "memory(GiB)": 73.83, "step": 87005, "token_acc": 0.6019417475728155, "train_speed(iter/s)": 1.456253 }, { "epoch": 3.7277751595904203, "grad_norm": 1.645896315574646, "learning_rate": 1.5141803825210637e-05, "loss": 0.4032104015350342, "memory(GiB)": 73.83, "step": 87010, "token_acc": 0.9, "train_speed(iter/s)": 1.456255 }, { "epoch": 3.727989374919669, "grad_norm": 9.075528144836426, "learning_rate": 1.5136979487567315e-05, "loss": 0.5876794815063476, "memory(GiB)": 73.83, "step": 87015, "token_acc": 0.89, "train_speed(iter/s)": 1.456275 }, { "epoch": 3.7282035902489183, "grad_norm": 4.053476810455322, "learning_rate": 1.5132155781500313e-05, "loss": 0.33632869720458985, "memory(GiB)": 73.83, "step": 87020, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.456284 }, { "epoch": 3.728417805578167, "grad_norm": 1.9380967617034912, "learning_rate": 1.5127332707097013e-05, "loss": 0.28337917327880857, "memory(GiB)": 73.83, "step": 87025, "token_acc": 0.9409282700421941, "train_speed(iter/s)": 1.456292 }, { "epoch": 3.7286320209074164, "grad_norm": 3.6581881046295166, "learning_rate": 1.5122510264444784e-05, "loss": 0.30681238174438474, "memory(GiB)": 73.83, "step": 87030, "token_acc": 0.9098360655737705, "train_speed(iter/s)": 1.456305 }, { "epoch": 3.7288462362366652, "grad_norm": 4.9294633865356445, "learning_rate": 1.5117688453631019e-05, "loss": 0.34650378227233886, "memory(GiB)": 73.83, "step": 87035, "token_acc": 0.9270516717325228, "train_speed(iter/s)": 1.456308 }, { "epoch": 3.729060451565914, "grad_norm": 3.9521586894989014, "learning_rate": 1.5112867274743026e-05, "loss": 0.2613146066665649, "memory(GiB)": 73.83, "step": 87040, "token_acc": 0.9496644295302014, "train_speed(iter/s)": 1.456318 }, { "epoch": 3.7292746668951633, "grad_norm": 2.7984721660614014, "learning_rate": 1.5108046727868186e-05, "loss": 0.5060186386108398, "memory(GiB)": 73.83, "step": 87045, "token_acc": 0.8907849829351536, "train_speed(iter/s)": 1.456325 }, { "epoch": 3.729488882224412, "grad_norm": 2.644829034805298, "learning_rate": 1.5103226813093813e-05, "loss": 0.38149275779724123, "memory(GiB)": 73.83, "step": 87050, "token_acc": 0.9224376731301939, "train_speed(iter/s)": 1.456323 }, { "epoch": 3.729703097553661, "grad_norm": 3.055943012237549, "learning_rate": 1.5098407530507225e-05, "loss": 0.33494508266448975, "memory(GiB)": 73.83, "step": 87055, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.456324 }, { "epoch": 3.72991731288291, "grad_norm": 5.584092140197754, "learning_rate": 1.509358888019572e-05, "loss": 0.21997852325439454, "memory(GiB)": 73.83, "step": 87060, "token_acc": 0.9537366548042705, "train_speed(iter/s)": 1.456323 }, { "epoch": 3.730131528212159, "grad_norm": 4.721549034118652, "learning_rate": 1.5088770862246599e-05, "loss": 0.7990368366241455, "memory(GiB)": 73.83, "step": 87065, "token_acc": 0.8189189189189189, "train_speed(iter/s)": 1.456325 }, { "epoch": 3.730345743541408, "grad_norm": 7.216846466064453, "learning_rate": 1.508395347674713e-05, "loss": 0.5414923667907715, "memory(GiB)": 73.83, "step": 87070, "token_acc": 0.9352226720647774, "train_speed(iter/s)": 1.456337 }, { "epoch": 3.730559958870657, "grad_norm": 2.8067128658294678, "learning_rate": 1.507913672378461e-05, "loss": 0.6127458572387695, "memory(GiB)": 73.83, "step": 87075, "token_acc": 0.8558282208588958, "train_speed(iter/s)": 1.456342 }, { "epoch": 3.730774174199906, "grad_norm": 3.001377820968628, "learning_rate": 1.507432060344629e-05, "loss": 0.33167123794555664, "memory(GiB)": 73.83, "step": 87080, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.456341 }, { "epoch": 3.7309883895291547, "grad_norm": 1.926758885383606, "learning_rate": 1.506950511581941e-05, "loss": 0.2570651531219482, "memory(GiB)": 73.83, "step": 87085, "token_acc": 0.9391304347826087, "train_speed(iter/s)": 1.456346 }, { "epoch": 3.731202604858404, "grad_norm": 2.334047794342041, "learning_rate": 1.5064690260991215e-05, "loss": 0.49306364059448243, "memory(GiB)": 73.83, "step": 87090, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.456348 }, { "epoch": 3.7314168201876527, "grad_norm": 3.482724905014038, "learning_rate": 1.5059876039048914e-05, "loss": 0.33318586349487306, "memory(GiB)": 73.83, "step": 87095, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.456349 }, { "epoch": 3.7316310355169016, "grad_norm": 4.061036109924316, "learning_rate": 1.5055062450079749e-05, "loss": 0.56529221534729, "memory(GiB)": 73.83, "step": 87100, "token_acc": 0.9023569023569024, "train_speed(iter/s)": 1.456349 }, { "epoch": 3.731845250846151, "grad_norm": 3.7960448265075684, "learning_rate": 1.5050249494170893e-05, "loss": 0.4047727108001709, "memory(GiB)": 73.83, "step": 87105, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.456352 }, { "epoch": 3.7320594661753996, "grad_norm": 4.548225402832031, "learning_rate": 1.5045437171409571e-05, "loss": 0.5714143753051758, "memory(GiB)": 73.83, "step": 87110, "token_acc": 0.8806584362139918, "train_speed(iter/s)": 1.456356 }, { "epoch": 3.7322736815046484, "grad_norm": 3.964718818664551, "learning_rate": 1.5040625481882942e-05, "loss": 0.4020097255706787, "memory(GiB)": 73.83, "step": 87115, "token_acc": 0.915129151291513, "train_speed(iter/s)": 1.456354 }, { "epoch": 3.7324878968338977, "grad_norm": 2.6830694675445557, "learning_rate": 1.5035814425678174e-05, "loss": 0.5679525852203369, "memory(GiB)": 73.83, "step": 87120, "token_acc": 0.8864468864468864, "train_speed(iter/s)": 1.456365 }, { "epoch": 3.7327021121631465, "grad_norm": 2.861301898956299, "learning_rate": 1.5031004002882431e-05, "loss": 0.2845446109771729, "memory(GiB)": 73.83, "step": 87125, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.456369 }, { "epoch": 3.7329163274923953, "grad_norm": 3.646801233291626, "learning_rate": 1.502619421358284e-05, "loss": 0.29165091514587405, "memory(GiB)": 73.83, "step": 87130, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.456375 }, { "epoch": 3.7331305428216446, "grad_norm": 2.69575572013855, "learning_rate": 1.502138505786656e-05, "loss": 0.310504674911499, "memory(GiB)": 73.83, "step": 87135, "token_acc": 0.9417808219178082, "train_speed(iter/s)": 1.456381 }, { "epoch": 3.7333447581508934, "grad_norm": 3.3474133014678955, "learning_rate": 1.5016576535820708e-05, "loss": 0.4947092056274414, "memory(GiB)": 73.83, "step": 87140, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.45639 }, { "epoch": 3.733558973480142, "grad_norm": 3.502133846282959, "learning_rate": 1.5011768647532382e-05, "loss": 0.18060071468353273, "memory(GiB)": 73.83, "step": 87145, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.456399 }, { "epoch": 3.7337731888093915, "grad_norm": 2.918968915939331, "learning_rate": 1.5006961393088692e-05, "loss": 0.45080127716064455, "memory(GiB)": 73.83, "step": 87150, "token_acc": 0.9087719298245615, "train_speed(iter/s)": 1.45641 }, { "epoch": 3.7339874041386403, "grad_norm": 5.782954216003418, "learning_rate": 1.5002154772576709e-05, "loss": 0.6122774124145508, "memory(GiB)": 73.83, "step": 87155, "token_acc": 0.8435114503816794, "train_speed(iter/s)": 1.45641 }, { "epoch": 3.734201619467889, "grad_norm": 3.387338876724243, "learning_rate": 1.4997348786083537e-05, "loss": 0.35100018978118896, "memory(GiB)": 73.83, "step": 87160, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.456412 }, { "epoch": 3.7344158347971383, "grad_norm": 1.893446922302246, "learning_rate": 1.4992543433696228e-05, "loss": 0.33830740451812746, "memory(GiB)": 73.83, "step": 87165, "token_acc": 0.9252336448598131, "train_speed(iter/s)": 1.456421 }, { "epoch": 3.734630050126387, "grad_norm": 3.036210298538208, "learning_rate": 1.4987738715501832e-05, "loss": 0.7033956050872803, "memory(GiB)": 73.83, "step": 87170, "token_acc": 0.8467336683417085, "train_speed(iter/s)": 1.456429 }, { "epoch": 3.734844265455636, "grad_norm": 5.619747161865234, "learning_rate": 1.4982934631587391e-05, "loss": 0.4179225444793701, "memory(GiB)": 73.83, "step": 87175, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.45643 }, { "epoch": 3.7350584807848852, "grad_norm": 4.049317836761475, "learning_rate": 1.4978131182039928e-05, "loss": 0.395086669921875, "memory(GiB)": 73.83, "step": 87180, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.456432 }, { "epoch": 3.735272696114134, "grad_norm": 3.407586097717285, "learning_rate": 1.4973328366946471e-05, "loss": 0.3336267232894897, "memory(GiB)": 73.83, "step": 87185, "token_acc": 0.9283387622149837, "train_speed(iter/s)": 1.456436 }, { "epoch": 3.735486911443383, "grad_norm": 2.577601671218872, "learning_rate": 1.4968526186394039e-05, "loss": 0.3324448108673096, "memory(GiB)": 73.83, "step": 87190, "token_acc": 0.9360902255639098, "train_speed(iter/s)": 1.456439 }, { "epoch": 3.735701126772632, "grad_norm": 5.352195739746094, "learning_rate": 1.4963724640469622e-05, "loss": 0.2483515977859497, "memory(GiB)": 73.83, "step": 87195, "token_acc": 0.9462809917355371, "train_speed(iter/s)": 1.456438 }, { "epoch": 3.735915342101881, "grad_norm": 6.0158281326293945, "learning_rate": 1.4958923729260198e-05, "loss": 0.31373291015625, "memory(GiB)": 73.83, "step": 87200, "token_acc": 0.9532710280373832, "train_speed(iter/s)": 1.456452 }, { "epoch": 3.7361295574311297, "grad_norm": 1.2476273775100708, "learning_rate": 1.4954123452852742e-05, "loss": 0.28197782039642333, "memory(GiB)": 73.83, "step": 87205, "token_acc": 0.9431137724550899, "train_speed(iter/s)": 1.456458 }, { "epoch": 3.736343772760379, "grad_norm": 2.7314748764038086, "learning_rate": 1.4949323811334214e-05, "loss": 0.3397878885269165, "memory(GiB)": 73.83, "step": 87210, "token_acc": 0.9115384615384615, "train_speed(iter/s)": 1.456473 }, { "epoch": 3.736557988089628, "grad_norm": 2.698343515396118, "learning_rate": 1.4944524804791554e-05, "loss": 0.22403936386108397, "memory(GiB)": 73.83, "step": 87215, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.456478 }, { "epoch": 3.7367722034188766, "grad_norm": 4.075242519378662, "learning_rate": 1.4939726433311723e-05, "loss": 0.48752365112304685, "memory(GiB)": 73.83, "step": 87220, "token_acc": 0.8954248366013072, "train_speed(iter/s)": 1.456481 }, { "epoch": 3.736986418748126, "grad_norm": 0.6744279861450195, "learning_rate": 1.4934928696981636e-05, "loss": 0.3445396661758423, "memory(GiB)": 73.83, "step": 87225, "token_acc": 0.9323843416370107, "train_speed(iter/s)": 1.456487 }, { "epoch": 3.7372006340773747, "grad_norm": 4.84022855758667, "learning_rate": 1.4930131595888213e-05, "loss": 0.48425922393798826, "memory(GiB)": 73.83, "step": 87230, "token_acc": 0.903114186851211, "train_speed(iter/s)": 1.456493 }, { "epoch": 3.7374148494066235, "grad_norm": 2.2798514366149902, "learning_rate": 1.4925335130118357e-05, "loss": 0.4832958698272705, "memory(GiB)": 73.83, "step": 87235, "token_acc": 0.8864468864468864, "train_speed(iter/s)": 1.456492 }, { "epoch": 3.7376290647358728, "grad_norm": 1.8397743701934814, "learning_rate": 1.4920539299758935e-05, "loss": 0.1661761522293091, "memory(GiB)": 73.83, "step": 87240, "token_acc": 0.9697802197802198, "train_speed(iter/s)": 1.456502 }, { "epoch": 3.7378432800651216, "grad_norm": 4.958127975463867, "learning_rate": 1.4915744104896872e-05, "loss": 0.32421488761901857, "memory(GiB)": 73.83, "step": 87245, "token_acc": 0.9481132075471698, "train_speed(iter/s)": 1.456513 }, { "epoch": 3.7380574953943704, "grad_norm": 4.178220748901367, "learning_rate": 1.4910949545619013e-05, "loss": 0.33275089263916013, "memory(GiB)": 73.83, "step": 87250, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.456517 }, { "epoch": 3.7382717107236196, "grad_norm": 2.487894058227539, "learning_rate": 1.4906155622012202e-05, "loss": 0.23456521034240724, "memory(GiB)": 73.83, "step": 87255, "token_acc": 0.955719557195572, "train_speed(iter/s)": 1.456513 }, { "epoch": 3.7384859260528684, "grad_norm": 3.865143299102783, "learning_rate": 1.490136233416332e-05, "loss": 0.3065745115280151, "memory(GiB)": 73.83, "step": 87260, "token_acc": 0.936, "train_speed(iter/s)": 1.456527 }, { "epoch": 3.7387001413821173, "grad_norm": 3.905266761779785, "learning_rate": 1.489656968215919e-05, "loss": 0.3347182750701904, "memory(GiB)": 73.83, "step": 87265, "token_acc": 0.9174041297935103, "train_speed(iter/s)": 1.456528 }, { "epoch": 3.7389143567113665, "grad_norm": 2.640137195587158, "learning_rate": 1.4891777666086609e-05, "loss": 0.4570337772369385, "memory(GiB)": 73.83, "step": 87270, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.456538 }, { "epoch": 3.7391285720406153, "grad_norm": 2.532252311706543, "learning_rate": 1.4886986286032423e-05, "loss": 0.3997419118881226, "memory(GiB)": 73.83, "step": 87275, "token_acc": 0.8920634920634921, "train_speed(iter/s)": 1.456542 }, { "epoch": 3.739342787369864, "grad_norm": 4.582937240600586, "learning_rate": 1.4882195542083421e-05, "loss": 0.4792358875274658, "memory(GiB)": 73.83, "step": 87280, "token_acc": 0.9067524115755627, "train_speed(iter/s)": 1.456546 }, { "epoch": 3.7395570026991134, "grad_norm": 3.110398292541504, "learning_rate": 1.487740543432639e-05, "loss": 0.12302658557891846, "memory(GiB)": 73.83, "step": 87285, "token_acc": 0.9631147540983607, "train_speed(iter/s)": 1.456558 }, { "epoch": 3.739771218028362, "grad_norm": 4.628659248352051, "learning_rate": 1.4872615962848113e-05, "loss": 0.28327670097351076, "memory(GiB)": 73.83, "step": 87290, "token_acc": 0.9439252336448598, "train_speed(iter/s)": 1.456565 }, { "epoch": 3.739985433357611, "grad_norm": 4.729551792144775, "learning_rate": 1.4867827127735346e-05, "loss": 0.6145378589630127, "memory(GiB)": 73.83, "step": 87295, "token_acc": 0.8790035587188612, "train_speed(iter/s)": 1.456567 }, { "epoch": 3.7401996486868603, "grad_norm": 0.22413161396980286, "learning_rate": 1.486303892907483e-05, "loss": 0.5575769901275635, "memory(GiB)": 73.83, "step": 87300, "token_acc": 0.8876811594202898, "train_speed(iter/s)": 1.456564 }, { "epoch": 3.740413864016109, "grad_norm": 5.580371856689453, "learning_rate": 1.485825136695334e-05, "loss": 0.3409864902496338, "memory(GiB)": 73.83, "step": 87305, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.456573 }, { "epoch": 3.740628079345358, "grad_norm": 1.6054210662841797, "learning_rate": 1.4853464441457593e-05, "loss": 0.4501772403717041, "memory(GiB)": 73.83, "step": 87310, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.456576 }, { "epoch": 3.740842294674607, "grad_norm": 1.5612330436706543, "learning_rate": 1.4848678152674311e-05, "loss": 0.32308733463287354, "memory(GiB)": 73.83, "step": 87315, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.456577 }, { "epoch": 3.741056510003856, "grad_norm": 7.6802754402160645, "learning_rate": 1.4843892500690193e-05, "loss": 0.3727866172790527, "memory(GiB)": 73.83, "step": 87320, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.456592 }, { "epoch": 3.741270725333105, "grad_norm": 2.3323047161102295, "learning_rate": 1.483910748559193e-05, "loss": 0.3394496440887451, "memory(GiB)": 73.83, "step": 87325, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.456595 }, { "epoch": 3.741484940662354, "grad_norm": 1.8474584817886353, "learning_rate": 1.4834323107466218e-05, "loss": 0.3275322198867798, "memory(GiB)": 73.83, "step": 87330, "token_acc": 0.9442815249266863, "train_speed(iter/s)": 1.45661 }, { "epoch": 3.741699155991603, "grad_norm": 3.070931911468506, "learning_rate": 1.4829539366399747e-05, "loss": 0.23812482357025147, "memory(GiB)": 73.83, "step": 87335, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.456608 }, { "epoch": 3.7419133713208517, "grad_norm": 2.195324420928955, "learning_rate": 1.4824756262479161e-05, "loss": 0.37285771369934084, "memory(GiB)": 73.83, "step": 87340, "token_acc": 0.9155844155844156, "train_speed(iter/s)": 1.456611 }, { "epoch": 3.742127586650101, "grad_norm": 3.356989622116089, "learning_rate": 1.4819973795791115e-05, "loss": 0.5336360454559326, "memory(GiB)": 73.83, "step": 87345, "token_acc": 0.879245283018868, "train_speed(iter/s)": 1.456623 }, { "epoch": 3.7423418019793497, "grad_norm": 1.6469695568084717, "learning_rate": 1.4815191966422243e-05, "loss": 0.4333303451538086, "memory(GiB)": 73.83, "step": 87350, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.456637 }, { "epoch": 3.7425560173085985, "grad_norm": 3.4765286445617676, "learning_rate": 1.4810410774459171e-05, "loss": 0.47931804656982424, "memory(GiB)": 73.83, "step": 87355, "token_acc": 0.8923611111111112, "train_speed(iter/s)": 1.45664 }, { "epoch": 3.742770232637848, "grad_norm": 1.6084353923797607, "learning_rate": 1.4805630219988508e-05, "loss": 0.5836724758148193, "memory(GiB)": 73.83, "step": 87360, "token_acc": 0.8625954198473282, "train_speed(iter/s)": 1.456644 }, { "epoch": 3.7429844479670966, "grad_norm": 2.933058738708496, "learning_rate": 1.4800850303096885e-05, "loss": 0.34665813446044924, "memory(GiB)": 73.83, "step": 87365, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.456645 }, { "epoch": 3.7431986632963454, "grad_norm": 1.6410505771636963, "learning_rate": 1.4796071023870872e-05, "loss": 0.4794106960296631, "memory(GiB)": 73.83, "step": 87370, "token_acc": 0.9052287581699346, "train_speed(iter/s)": 1.456644 }, { "epoch": 3.7434128786255947, "grad_norm": 2.9460997581481934, "learning_rate": 1.4791292382397064e-05, "loss": 0.6299489974975586, "memory(GiB)": 73.83, "step": 87375, "token_acc": 0.8680555555555556, "train_speed(iter/s)": 1.456657 }, { "epoch": 3.7436270939548435, "grad_norm": 4.312121391296387, "learning_rate": 1.4786514378762017e-05, "loss": 0.2077418804168701, "memory(GiB)": 73.83, "step": 87380, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.456658 }, { "epoch": 3.7438413092840923, "grad_norm": 1.4595903158187866, "learning_rate": 1.4781737013052282e-05, "loss": 0.21382105350494385, "memory(GiB)": 73.83, "step": 87385, "token_acc": 0.9409282700421941, "train_speed(iter/s)": 1.456659 }, { "epoch": 3.7440555246133416, "grad_norm": 1.6657493114471436, "learning_rate": 1.4776960285354436e-05, "loss": 0.16623919010162352, "memory(GiB)": 73.83, "step": 87390, "token_acc": 0.9766666666666667, "train_speed(iter/s)": 1.456668 }, { "epoch": 3.7442697399425904, "grad_norm": 6.433486461639404, "learning_rate": 1.4772184195754996e-05, "loss": 0.3977036476135254, "memory(GiB)": 73.83, "step": 87395, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.456667 }, { "epoch": 3.744483955271839, "grad_norm": 3.141331434249878, "learning_rate": 1.4767408744340466e-05, "loss": 0.3634836673736572, "memory(GiB)": 73.83, "step": 87400, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.456665 }, { "epoch": 3.7446981706010884, "grad_norm": 3.0173492431640625, "learning_rate": 1.4762633931197395e-05, "loss": 0.4932608127593994, "memory(GiB)": 73.83, "step": 87405, "token_acc": 0.9013605442176871, "train_speed(iter/s)": 1.456672 }, { "epoch": 3.7449123859303373, "grad_norm": 0.7075473666191101, "learning_rate": 1.4757859756412268e-05, "loss": 0.3787652254104614, "memory(GiB)": 73.83, "step": 87410, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.456683 }, { "epoch": 3.745126601259586, "grad_norm": 2.852567672729492, "learning_rate": 1.475308622007155e-05, "loss": 0.27691450119018557, "memory(GiB)": 73.83, "step": 87415, "token_acc": 0.9501915708812261, "train_speed(iter/s)": 1.456687 }, { "epoch": 3.7453408165888353, "grad_norm": 5.574461936950684, "learning_rate": 1.4748313322261758e-05, "loss": 0.372035551071167, "memory(GiB)": 73.83, "step": 87420, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.456692 }, { "epoch": 3.745555031918084, "grad_norm": 4.580835342407227, "learning_rate": 1.4743541063069339e-05, "loss": 0.4801356792449951, "memory(GiB)": 73.83, "step": 87425, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.456697 }, { "epoch": 3.745769247247333, "grad_norm": 3.8432493209838867, "learning_rate": 1.4738769442580746e-05, "loss": 0.34275381565093993, "memory(GiB)": 73.83, "step": 87430, "token_acc": 0.9382352941176471, "train_speed(iter/s)": 1.456698 }, { "epoch": 3.745983462576582, "grad_norm": 3.273590564727783, "learning_rate": 1.4733998460882425e-05, "loss": 0.3120319366455078, "memory(GiB)": 73.83, "step": 87435, "token_acc": 0.9239766081871345, "train_speed(iter/s)": 1.456699 }, { "epoch": 3.746197677905831, "grad_norm": 4.257087230682373, "learning_rate": 1.4729228118060807e-05, "loss": 0.24862442016601563, "memory(GiB)": 73.83, "step": 87440, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.456706 }, { "epoch": 3.74641189323508, "grad_norm": 3.450883150100708, "learning_rate": 1.4724458414202285e-05, "loss": 0.20722591876983643, "memory(GiB)": 73.83, "step": 87445, "token_acc": 0.9490445859872612, "train_speed(iter/s)": 1.456712 }, { "epoch": 3.746626108564329, "grad_norm": 2.3846495151519775, "learning_rate": 1.4719689349393312e-05, "loss": 0.3172950744628906, "memory(GiB)": 73.83, "step": 87450, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.456714 }, { "epoch": 3.746840323893578, "grad_norm": 4.6452555656433105, "learning_rate": 1.4714920923720254e-05, "loss": 0.3291806221008301, "memory(GiB)": 73.83, "step": 87455, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.456717 }, { "epoch": 3.7470545392228267, "grad_norm": 6.431586265563965, "learning_rate": 1.471015313726951e-05, "loss": 0.31506564617156985, "memory(GiB)": 73.83, "step": 87460, "token_acc": 0.9319148936170213, "train_speed(iter/s)": 1.45672 }, { "epoch": 3.747268754552076, "grad_norm": 4.447630405426025, "learning_rate": 1.4705385990127446e-05, "loss": 0.11045809984207153, "memory(GiB)": 73.83, "step": 87465, "token_acc": 0.9702970297029703, "train_speed(iter/s)": 1.456726 }, { "epoch": 3.747482969881325, "grad_norm": 1.1629797220230103, "learning_rate": 1.4700619482380406e-05, "loss": 0.23867359161376953, "memory(GiB)": 73.83, "step": 87470, "token_acc": 0.9396825396825397, "train_speed(iter/s)": 1.456734 }, { "epoch": 3.7476971852105736, "grad_norm": 2.6955411434173584, "learning_rate": 1.4695853614114763e-05, "loss": 0.24375741481781005, "memory(GiB)": 73.83, "step": 87475, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.456733 }, { "epoch": 3.747911400539823, "grad_norm": 0.7290510535240173, "learning_rate": 1.4691088385416857e-05, "loss": 0.08554428219795226, "memory(GiB)": 73.83, "step": 87480, "token_acc": 0.9883268482490273, "train_speed(iter/s)": 1.456745 }, { "epoch": 3.7481256158690717, "grad_norm": 0.2929089665412903, "learning_rate": 1.4686323796373014e-05, "loss": 0.3663517475128174, "memory(GiB)": 73.83, "step": 87485, "token_acc": 0.9330543933054394, "train_speed(iter/s)": 1.45675 }, { "epoch": 3.7483398311983205, "grad_norm": 3.5109753608703613, "learning_rate": 1.4681559847069537e-05, "loss": 0.3950511932373047, "memory(GiB)": 73.83, "step": 87490, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.456761 }, { "epoch": 3.7485540465275697, "grad_norm": 3.5530755519866943, "learning_rate": 1.467679653759274e-05, "loss": 0.3169077157974243, "memory(GiB)": 73.83, "step": 87495, "token_acc": 0.9429824561403509, "train_speed(iter/s)": 1.456763 }, { "epoch": 3.7487682618568186, "grad_norm": 0.23445384204387665, "learning_rate": 1.4672033868028907e-05, "loss": 0.08226295709609985, "memory(GiB)": 73.83, "step": 87500, "token_acc": 0.9815384615384616, "train_speed(iter/s)": 1.456766 }, { "epoch": 3.7487682618568186, "eval_loss": 2.355958938598633, "eval_runtime": 10.9104, "eval_samples_per_second": 9.166, "eval_steps_per_second": 9.166, "eval_token_acc": 0.468586387434555, "step": 87500 }, { "epoch": 3.7489824771860674, "grad_norm": 4.302363395690918, "learning_rate": 1.4667271838464303e-05, "loss": 0.6356454372406006, "memory(GiB)": 73.83, "step": 87505, "token_acc": 0.581042654028436, "train_speed(iter/s)": 1.45648 }, { "epoch": 3.7491966925153166, "grad_norm": 3.457827091217041, "learning_rate": 1.4662510448985234e-05, "loss": 0.5029096603393555, "memory(GiB)": 73.83, "step": 87510, "token_acc": 0.8931297709923665, "train_speed(iter/s)": 1.456487 }, { "epoch": 3.7494109078445654, "grad_norm": 8.956269264221191, "learning_rate": 1.4657749699677937e-05, "loss": 0.40855989456176756, "memory(GiB)": 73.83, "step": 87515, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.456486 }, { "epoch": 3.7496251231738142, "grad_norm": 3.3349664211273193, "learning_rate": 1.4652989590628658e-05, "loss": 0.5076042652130127, "memory(GiB)": 73.83, "step": 87520, "token_acc": 0.9016949152542373, "train_speed(iter/s)": 1.456495 }, { "epoch": 3.7498393385030635, "grad_norm": 2.267692804336548, "learning_rate": 1.4648230121923629e-05, "loss": 0.23217968940734862, "memory(GiB)": 73.83, "step": 87525, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.456501 }, { "epoch": 3.7500535538323123, "grad_norm": 0.03194371610879898, "learning_rate": 1.4643471293649059e-05, "loss": 0.15650721788406372, "memory(GiB)": 73.83, "step": 87530, "token_acc": 0.9777777777777777, "train_speed(iter/s)": 1.456503 }, { "epoch": 3.750267769161561, "grad_norm": 2.4598782062530518, "learning_rate": 1.4638713105891188e-05, "loss": 0.373477840423584, "memory(GiB)": 73.83, "step": 87535, "token_acc": 0.9075907590759076, "train_speed(iter/s)": 1.456504 }, { "epoch": 3.7504819844908104, "grad_norm": 0.7451271414756775, "learning_rate": 1.4633955558736201e-05, "loss": 0.4612100124359131, "memory(GiB)": 73.83, "step": 87540, "token_acc": 0.8872727272727273, "train_speed(iter/s)": 1.456508 }, { "epoch": 3.750696199820059, "grad_norm": 4.130038738250732, "learning_rate": 1.4629198652270288e-05, "loss": 0.4262916088104248, "memory(GiB)": 73.83, "step": 87545, "token_acc": 0.9012738853503185, "train_speed(iter/s)": 1.456508 }, { "epoch": 3.750910415149308, "grad_norm": 0.630781352519989, "learning_rate": 1.4624442386579601e-05, "loss": 0.19932146072387696, "memory(GiB)": 73.83, "step": 87550, "token_acc": 0.9538043478260869, "train_speed(iter/s)": 1.456513 }, { "epoch": 3.7511246304785573, "grad_norm": 2.108792543411255, "learning_rate": 1.4619686761750345e-05, "loss": 0.5292571544647217, "memory(GiB)": 73.83, "step": 87555, "token_acc": 0.8954703832752613, "train_speed(iter/s)": 1.456511 }, { "epoch": 3.751338845807806, "grad_norm": 1.8905482292175293, "learning_rate": 1.4614931777868634e-05, "loss": 0.46469573974609374, "memory(GiB)": 73.83, "step": 87560, "token_acc": 0.8996282527881041, "train_speed(iter/s)": 1.456523 }, { "epoch": 3.751553061137055, "grad_norm": 2.830050230026245, "learning_rate": 1.4610177435020645e-05, "loss": 0.25890612602233887, "memory(GiB)": 73.83, "step": 87565, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.456527 }, { "epoch": 3.751767276466304, "grad_norm": 4.319506645202637, "learning_rate": 1.4605423733292494e-05, "loss": 0.49800901412963866, "memory(GiB)": 73.83, "step": 87570, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.456529 }, { "epoch": 3.751981491795553, "grad_norm": 0.44231173396110535, "learning_rate": 1.460067067277029e-05, "loss": 0.5321436405181885, "memory(GiB)": 73.83, "step": 87575, "token_acc": 0.8814814814814815, "train_speed(iter/s)": 1.456532 }, { "epoch": 3.7521957071248018, "grad_norm": 2.3129098415374756, "learning_rate": 1.4595918253540147e-05, "loss": 0.3941753625869751, "memory(GiB)": 73.83, "step": 87580, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.45653 }, { "epoch": 3.752409922454051, "grad_norm": 1.0056757926940918, "learning_rate": 1.459116647568815e-05, "loss": 0.4815830230712891, "memory(GiB)": 73.83, "step": 87585, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.456532 }, { "epoch": 3.7526241377833, "grad_norm": 2.361565113067627, "learning_rate": 1.458641533930038e-05, "loss": 0.2665820598602295, "memory(GiB)": 73.83, "step": 87590, "token_acc": 0.9409448818897638, "train_speed(iter/s)": 1.45654 }, { "epoch": 3.7528383531125487, "grad_norm": 3.797078847885132, "learning_rate": 1.4581664844462929e-05, "loss": 0.299461817741394, "memory(GiB)": 73.83, "step": 87595, "token_acc": 0.9404388714733543, "train_speed(iter/s)": 1.456541 }, { "epoch": 3.753052568441798, "grad_norm": 3.0146853923797607, "learning_rate": 1.4576914991261848e-05, "loss": 0.5967714786529541, "memory(GiB)": 73.83, "step": 87600, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.456546 }, { "epoch": 3.7532667837710467, "grad_norm": 0.7296326756477356, "learning_rate": 1.4572165779783176e-05, "loss": 0.5864129543304444, "memory(GiB)": 73.83, "step": 87605, "token_acc": 0.8504983388704319, "train_speed(iter/s)": 1.456542 }, { "epoch": 3.7534809991002955, "grad_norm": 0.5881764888763428, "learning_rate": 1.456741721011296e-05, "loss": 0.27382075786590576, "memory(GiB)": 73.83, "step": 87610, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.456544 }, { "epoch": 3.753695214429545, "grad_norm": 4.688806533813477, "learning_rate": 1.4562669282337198e-05, "loss": 0.3062721252441406, "memory(GiB)": 73.83, "step": 87615, "token_acc": 0.9204545454545454, "train_speed(iter/s)": 1.456549 }, { "epoch": 3.7539094297587936, "grad_norm": 3.069723129272461, "learning_rate": 1.4557921996541946e-05, "loss": 0.29157233238220215, "memory(GiB)": 73.83, "step": 87620, "token_acc": 0.9541984732824428, "train_speed(iter/s)": 1.456552 }, { "epoch": 3.7541236450880424, "grad_norm": 2.554424285888672, "learning_rate": 1.455317535281317e-05, "loss": 0.48914251327514646, "memory(GiB)": 73.83, "step": 87625, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.456558 }, { "epoch": 3.7543378604172917, "grad_norm": 1.0003759860992432, "learning_rate": 1.4548429351236886e-05, "loss": 0.2539220333099365, "memory(GiB)": 73.83, "step": 87630, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.456573 }, { "epoch": 3.7545520757465405, "grad_norm": 2.9295177459716797, "learning_rate": 1.454368399189906e-05, "loss": 0.38617041110992434, "memory(GiB)": 73.83, "step": 87635, "token_acc": 0.889967637540453, "train_speed(iter/s)": 1.456574 }, { "epoch": 3.7547662910757893, "grad_norm": 2.82039475440979, "learning_rate": 1.4538939274885665e-05, "loss": 0.14877382516860962, "memory(GiB)": 73.83, "step": 87640, "token_acc": 0.9739130434782609, "train_speed(iter/s)": 1.456576 }, { "epoch": 3.7549805064050386, "grad_norm": 3.6667234897613525, "learning_rate": 1.4534195200282646e-05, "loss": 0.31948564052581785, "memory(GiB)": 73.83, "step": 87645, "token_acc": 0.9362549800796812, "train_speed(iter/s)": 1.45658 }, { "epoch": 3.7551947217342874, "grad_norm": 3.439580202102661, "learning_rate": 1.4529451768175933e-05, "loss": 0.4665986061096191, "memory(GiB)": 73.83, "step": 87650, "token_acc": 0.9018987341772152, "train_speed(iter/s)": 1.456584 }, { "epoch": 3.755408937063536, "grad_norm": 3.7969000339508057, "learning_rate": 1.4524708978651491e-05, "loss": 0.537476921081543, "memory(GiB)": 73.83, "step": 87655, "token_acc": 0.9039145907473309, "train_speed(iter/s)": 1.456589 }, { "epoch": 3.7556231523927854, "grad_norm": 2.5495691299438477, "learning_rate": 1.4519966831795228e-05, "loss": 0.2508310079574585, "memory(GiB)": 73.83, "step": 87660, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.456599 }, { "epoch": 3.7558373677220342, "grad_norm": 4.260451793670654, "learning_rate": 1.4515225327693049e-05, "loss": 0.18839993476867675, "memory(GiB)": 73.83, "step": 87665, "token_acc": 0.9461279461279462, "train_speed(iter/s)": 1.456605 }, { "epoch": 3.756051583051283, "grad_norm": 2.704655647277832, "learning_rate": 1.4510484466430846e-05, "loss": 0.3708745002746582, "memory(GiB)": 73.83, "step": 87670, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.4566 }, { "epoch": 3.7562657983805323, "grad_norm": 4.131824970245361, "learning_rate": 1.45057442480945e-05, "loss": 0.21156842708587648, "memory(GiB)": 73.83, "step": 87675, "token_acc": 0.956953642384106, "train_speed(iter/s)": 1.456601 }, { "epoch": 3.756480013709781, "grad_norm": 2.923067331314087, "learning_rate": 1.4501004672769903e-05, "loss": 0.3456393241882324, "memory(GiB)": 73.83, "step": 87680, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.456604 }, { "epoch": 3.75669422903903, "grad_norm": 4.505685329437256, "learning_rate": 1.4496265740542908e-05, "loss": 0.6423911094665528, "memory(GiB)": 73.83, "step": 87685, "token_acc": 0.864516129032258, "train_speed(iter/s)": 1.456608 }, { "epoch": 3.756908444368279, "grad_norm": 3.1176929473876953, "learning_rate": 1.4491527451499365e-05, "loss": 0.535536813735962, "memory(GiB)": 73.83, "step": 87690, "token_acc": 0.8877551020408163, "train_speed(iter/s)": 1.45661 }, { "epoch": 3.757122659697528, "grad_norm": 2.1930131912231445, "learning_rate": 1.448678980572511e-05, "loss": 0.28474903106689453, "memory(GiB)": 73.83, "step": 87695, "token_acc": 0.9345794392523364, "train_speed(iter/s)": 1.456613 }, { "epoch": 3.757336875026777, "grad_norm": 6.469902992248535, "learning_rate": 1.4482052803305962e-05, "loss": 0.8751742362976074, "memory(GiB)": 73.83, "step": 87700, "token_acc": 0.8428571428571429, "train_speed(iter/s)": 1.456614 }, { "epoch": 3.757551090356026, "grad_norm": 2.2639458179473877, "learning_rate": 1.4477316444327738e-05, "loss": 0.22298948764801024, "memory(GiB)": 73.83, "step": 87705, "token_acc": 0.9543973941368078, "train_speed(iter/s)": 1.456618 }, { "epoch": 3.757765305685275, "grad_norm": 5.239924430847168, "learning_rate": 1.4472580728876272e-05, "loss": 0.31403541564941406, "memory(GiB)": 73.83, "step": 87710, "token_acc": 0.9190283400809717, "train_speed(iter/s)": 1.456622 }, { "epoch": 3.7579795210145237, "grad_norm": 2.773214817047119, "learning_rate": 1.4467845657037332e-05, "loss": 0.27083351612091067, "memory(GiB)": 73.83, "step": 87715, "token_acc": 0.9298780487804879, "train_speed(iter/s)": 1.456631 }, { "epoch": 3.758193736343773, "grad_norm": 2.818932294845581, "learning_rate": 1.4463111228896697e-05, "loss": 0.528998327255249, "memory(GiB)": 73.83, "step": 87720, "token_acc": 0.8737201365187713, "train_speed(iter/s)": 1.456636 }, { "epoch": 3.7584079516730218, "grad_norm": 3.6785099506378174, "learning_rate": 1.4458377444540139e-05, "loss": 0.4241046905517578, "memory(GiB)": 73.83, "step": 87725, "token_acc": 0.9108527131782945, "train_speed(iter/s)": 1.456638 }, { "epoch": 3.7586221670022706, "grad_norm": 0.22854885458946228, "learning_rate": 1.4453644304053415e-05, "loss": 0.20421693325042725, "memory(GiB)": 73.83, "step": 87730, "token_acc": 0.9663299663299664, "train_speed(iter/s)": 1.45664 }, { "epoch": 3.75883638233152, "grad_norm": 3.467942237854004, "learning_rate": 1.4448911807522253e-05, "loss": 0.6610170841217041, "memory(GiB)": 73.83, "step": 87735, "token_acc": 0.8648648648648649, "train_speed(iter/s)": 1.456651 }, { "epoch": 3.7590505976607687, "grad_norm": 5.370941638946533, "learning_rate": 1.4444179955032422e-05, "loss": 0.2924372911453247, "memory(GiB)": 73.83, "step": 87740, "token_acc": 0.9158576051779935, "train_speed(iter/s)": 1.456659 }, { "epoch": 3.7592648129900175, "grad_norm": 2.785581350326538, "learning_rate": 1.443944874666962e-05, "loss": 0.221519136428833, "memory(GiB)": 73.83, "step": 87745, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.456672 }, { "epoch": 3.7594790283192667, "grad_norm": 2.768505096435547, "learning_rate": 1.443471818251957e-05, "loss": 0.2962604284286499, "memory(GiB)": 73.83, "step": 87750, "token_acc": 0.9304635761589404, "train_speed(iter/s)": 1.456676 }, { "epoch": 3.7596932436485155, "grad_norm": 1.0384190082550049, "learning_rate": 1.4429988262667959e-05, "loss": 0.2202453851699829, "memory(GiB)": 73.83, "step": 87755, "token_acc": 0.9515151515151515, "train_speed(iter/s)": 1.456684 }, { "epoch": 3.7599074589777643, "grad_norm": 2.450810194015503, "learning_rate": 1.4425258987200463e-05, "loss": 0.5325818061828613, "memory(GiB)": 73.83, "step": 87760, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.456685 }, { "epoch": 3.7601216743070136, "grad_norm": 8.352216720581055, "learning_rate": 1.4420530356202783e-05, "loss": 0.24083528518676758, "memory(GiB)": 73.83, "step": 87765, "token_acc": 0.9490740740740741, "train_speed(iter/s)": 1.456689 }, { "epoch": 3.7603358896362624, "grad_norm": 1.9395887851715088, "learning_rate": 1.4415802369760562e-05, "loss": 0.3411598205566406, "memory(GiB)": 73.83, "step": 87770, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.456694 }, { "epoch": 3.7605501049655112, "grad_norm": 5.777861595153809, "learning_rate": 1.4411075027959475e-05, "loss": 0.24050405025482177, "memory(GiB)": 73.83, "step": 87775, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.456701 }, { "epoch": 3.7607643202947605, "grad_norm": 5.6315741539001465, "learning_rate": 1.440634833088515e-05, "loss": 0.34117560386657714, "memory(GiB)": 73.83, "step": 87780, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.456706 }, { "epoch": 3.7609785356240093, "grad_norm": 3.592973470687866, "learning_rate": 1.4401622278623217e-05, "loss": 0.518140172958374, "memory(GiB)": 73.83, "step": 87785, "token_acc": 0.8769230769230769, "train_speed(iter/s)": 1.456712 }, { "epoch": 3.761192750953258, "grad_norm": 1.1872650384902954, "learning_rate": 1.4396896871259286e-05, "loss": 0.49545788764953613, "memory(GiB)": 73.83, "step": 87790, "token_acc": 0.9115853658536586, "train_speed(iter/s)": 1.456724 }, { "epoch": 3.7614069662825074, "grad_norm": 0.3333216607570648, "learning_rate": 1.4392172108878954e-05, "loss": 0.30185015201568605, "memory(GiB)": 73.83, "step": 87795, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.456729 }, { "epoch": 3.761621181611756, "grad_norm": 4.299720764160156, "learning_rate": 1.4387447991567838e-05, "loss": 0.16833313703536987, "memory(GiB)": 73.83, "step": 87800, "token_acc": 0.9721254355400697, "train_speed(iter/s)": 1.456731 }, { "epoch": 3.761835396941005, "grad_norm": 3.7686710357666016, "learning_rate": 1.438272451941151e-05, "loss": 0.3017200231552124, "memory(GiB)": 73.83, "step": 87805, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.456741 }, { "epoch": 3.7620496122702543, "grad_norm": 2.9464759826660156, "learning_rate": 1.4378001692495546e-05, "loss": 0.2535681247711182, "memory(GiB)": 73.83, "step": 87810, "token_acc": 0.9588014981273408, "train_speed(iter/s)": 1.45675 }, { "epoch": 3.762263827599503, "grad_norm": 2.542750835418701, "learning_rate": 1.437327951090549e-05, "loss": 0.3238539218902588, "memory(GiB)": 73.83, "step": 87815, "token_acc": 0.9384164222873901, "train_speed(iter/s)": 1.45676 }, { "epoch": 3.762478042928752, "grad_norm": 2.521824836730957, "learning_rate": 1.4368557974726882e-05, "loss": 0.3465795755386353, "memory(GiB)": 73.83, "step": 87820, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.456768 }, { "epoch": 3.762692258258001, "grad_norm": 2.2558207511901855, "learning_rate": 1.436383708404529e-05, "loss": 0.32413363456726074, "memory(GiB)": 73.83, "step": 87825, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.456778 }, { "epoch": 3.76290647358725, "grad_norm": 2.0757899284362793, "learning_rate": 1.4359116838946219e-05, "loss": 0.2908461093902588, "memory(GiB)": 73.83, "step": 87830, "token_acc": 0.92578125, "train_speed(iter/s)": 1.4568 }, { "epoch": 3.7631206889164988, "grad_norm": 1.0663639307022095, "learning_rate": 1.4354397239515177e-05, "loss": 0.12890830039978027, "memory(GiB)": 73.83, "step": 87835, "token_acc": 0.9714285714285714, "train_speed(iter/s)": 1.456803 }, { "epoch": 3.763334904245748, "grad_norm": 5.8623785972595215, "learning_rate": 1.434967828583767e-05, "loss": 0.4113141059875488, "memory(GiB)": 73.83, "step": 87840, "token_acc": 0.9157894736842105, "train_speed(iter/s)": 1.456803 }, { "epoch": 3.763549119574997, "grad_norm": 4.112277030944824, "learning_rate": 1.4344959977999162e-05, "loss": 0.3031275749206543, "memory(GiB)": 73.83, "step": 87845, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.45681 }, { "epoch": 3.7637633349042456, "grad_norm": 4.169903755187988, "learning_rate": 1.4340242316085156e-05, "loss": 0.21757643222808837, "memory(GiB)": 73.83, "step": 87850, "token_acc": 0.9632352941176471, "train_speed(iter/s)": 1.456821 }, { "epoch": 3.763977550233495, "grad_norm": 1.9107385873794556, "learning_rate": 1.433552530018113e-05, "loss": 0.4290291786193848, "memory(GiB)": 73.83, "step": 87855, "token_acc": 0.8823529411764706, "train_speed(iter/s)": 1.456835 }, { "epoch": 3.7641917655627437, "grad_norm": 3.043807029724121, "learning_rate": 1.4330808930372514e-05, "loss": 0.4385784149169922, "memory(GiB)": 73.83, "step": 87860, "token_acc": 0.9262820512820513, "train_speed(iter/s)": 1.456842 }, { "epoch": 3.7644059808919925, "grad_norm": 5.3874993324279785, "learning_rate": 1.4326093206744756e-05, "loss": 0.16175079345703125, "memory(GiB)": 73.83, "step": 87865, "token_acc": 0.961764705882353, "train_speed(iter/s)": 1.456847 }, { "epoch": 3.7646201962212418, "grad_norm": 4.5297112464904785, "learning_rate": 1.432137812938329e-05, "loss": 0.5458257675170899, "memory(GiB)": 73.83, "step": 87870, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.456861 }, { "epoch": 3.7648344115504906, "grad_norm": 2.4850258827209473, "learning_rate": 1.4316663698373522e-05, "loss": 0.43190755844116213, "memory(GiB)": 73.83, "step": 87875, "token_acc": 0.9087719298245615, "train_speed(iter/s)": 1.456864 }, { "epoch": 3.7650486268797394, "grad_norm": 3.8851559162139893, "learning_rate": 1.4311949913800848e-05, "loss": 0.3703805923461914, "memory(GiB)": 73.83, "step": 87880, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.456867 }, { "epoch": 3.7652628422089887, "grad_norm": 0.8414949178695679, "learning_rate": 1.4307236775750699e-05, "loss": 0.2154920816421509, "memory(GiB)": 73.83, "step": 87885, "token_acc": 0.9346153846153846, "train_speed(iter/s)": 1.45687 }, { "epoch": 3.7654770575382375, "grad_norm": 5.23362922668457, "learning_rate": 1.4302524284308438e-05, "loss": 0.4534431457519531, "memory(GiB)": 73.83, "step": 87890, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.456885 }, { "epoch": 3.7656912728674863, "grad_norm": 1.0402021408081055, "learning_rate": 1.4297812439559439e-05, "loss": 0.3450049161911011, "memory(GiB)": 73.83, "step": 87895, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.456897 }, { "epoch": 3.7659054881967355, "grad_norm": 1.9804376363754272, "learning_rate": 1.4293101241589051e-05, "loss": 0.42007598876953123, "memory(GiB)": 73.83, "step": 87900, "token_acc": 0.9099099099099099, "train_speed(iter/s)": 1.456898 }, { "epoch": 3.7661197035259844, "grad_norm": 3.1921463012695312, "learning_rate": 1.4288390690482622e-05, "loss": 0.23211932182312012, "memory(GiB)": 73.83, "step": 87905, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.456897 }, { "epoch": 3.766333918855233, "grad_norm": 2.653989315032959, "learning_rate": 1.428368078632551e-05, "loss": 0.34328346252441405, "memory(GiB)": 73.83, "step": 87910, "token_acc": 0.9338842975206612, "train_speed(iter/s)": 1.456898 }, { "epoch": 3.7665481341844824, "grad_norm": 2.986644744873047, "learning_rate": 1.4278971529203023e-05, "loss": 0.23361246585845946, "memory(GiB)": 73.83, "step": 87915, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.456898 }, { "epoch": 3.7667623495137312, "grad_norm": 0.08987720310688019, "learning_rate": 1.427426291920046e-05, "loss": 0.17053804397583008, "memory(GiB)": 73.83, "step": 87920, "token_acc": 0.956140350877193, "train_speed(iter/s)": 1.456898 }, { "epoch": 3.76697656484298, "grad_norm": 5.258199214935303, "learning_rate": 1.4269554956403153e-05, "loss": 0.36530652046203616, "memory(GiB)": 73.83, "step": 87925, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.4569 }, { "epoch": 3.7671907801722293, "grad_norm": 1.8349214792251587, "learning_rate": 1.4264847640896378e-05, "loss": 0.33083009719848633, "memory(GiB)": 73.83, "step": 87930, "token_acc": 0.9313725490196079, "train_speed(iter/s)": 1.456915 }, { "epoch": 3.767404995501478, "grad_norm": 3.3721964359283447, "learning_rate": 1.4260140972765407e-05, "loss": 0.29479274749755857, "memory(GiB)": 73.83, "step": 87935, "token_acc": 0.9343629343629344, "train_speed(iter/s)": 1.456925 }, { "epoch": 3.767619210830727, "grad_norm": 4.605485439300537, "learning_rate": 1.4255434952095498e-05, "loss": 0.5179016590118408, "memory(GiB)": 73.83, "step": 87940, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.456937 }, { "epoch": 3.767833426159976, "grad_norm": 4.55964469909668, "learning_rate": 1.4250729578971927e-05, "loss": 0.5088517189025878, "memory(GiB)": 73.83, "step": 87945, "token_acc": 0.8905109489051095, "train_speed(iter/s)": 1.456958 }, { "epoch": 3.768047641489225, "grad_norm": 2.2303807735443115, "learning_rate": 1.4246024853479928e-05, "loss": 0.4008045196533203, "memory(GiB)": 73.83, "step": 87950, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.456965 }, { "epoch": 3.768261856818474, "grad_norm": 2.5739431381225586, "learning_rate": 1.4241320775704725e-05, "loss": 0.14822659492492676, "memory(GiB)": 73.83, "step": 87955, "token_acc": 0.9696969696969697, "train_speed(iter/s)": 1.456973 }, { "epoch": 3.768476072147723, "grad_norm": 0.7277286052703857, "learning_rate": 1.4236617345731546e-05, "loss": 0.2368946075439453, "memory(GiB)": 73.83, "step": 87960, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.456975 }, { "epoch": 3.768690287476972, "grad_norm": 0.488859623670578, "learning_rate": 1.4231914563645576e-05, "loss": 0.3030246734619141, "memory(GiB)": 73.83, "step": 87965, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.456976 }, { "epoch": 3.7689045028062207, "grad_norm": 2.1749751567840576, "learning_rate": 1.4227212429532038e-05, "loss": 0.20795366764068604, "memory(GiB)": 73.83, "step": 87970, "token_acc": 0.952054794520548, "train_speed(iter/s)": 1.456976 }, { "epoch": 3.76911871813547, "grad_norm": 3.9091906547546387, "learning_rate": 1.4222510943476109e-05, "loss": 0.31670193672180175, "memory(GiB)": 73.83, "step": 87975, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.456979 }, { "epoch": 3.7693329334647188, "grad_norm": 1.4561642408370972, "learning_rate": 1.4217810105562957e-05, "loss": 0.16806299686431886, "memory(GiB)": 73.83, "step": 87980, "token_acc": 0.9720496894409938, "train_speed(iter/s)": 1.456981 }, { "epoch": 3.7695471487939676, "grad_norm": 2.6203086376190186, "learning_rate": 1.4213109915877736e-05, "loss": 0.36385014057159426, "memory(GiB)": 73.83, "step": 87985, "token_acc": 0.9205776173285198, "train_speed(iter/s)": 1.456982 }, { "epoch": 3.769761364123217, "grad_norm": 1.917945384979248, "learning_rate": 1.4208410374505587e-05, "loss": 0.28604276180267335, "memory(GiB)": 73.83, "step": 87990, "token_acc": 0.9331306990881459, "train_speed(iter/s)": 1.456981 }, { "epoch": 3.7699755794524656, "grad_norm": 2.253056526184082, "learning_rate": 1.4203711481531662e-05, "loss": 0.4649319648742676, "memory(GiB)": 73.83, "step": 87995, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.456988 }, { "epoch": 3.7701897947817145, "grad_norm": 1.3679131269454956, "learning_rate": 1.4199013237041093e-05, "loss": 0.3008065938949585, "memory(GiB)": 73.83, "step": 88000, "token_acc": 0.9326599326599326, "train_speed(iter/s)": 1.456987 }, { "epoch": 3.7701897947817145, "eval_loss": 2.4800615310668945, "eval_runtime": 11.5404, "eval_samples_per_second": 8.665, "eval_steps_per_second": 8.665, "eval_token_acc": 0.4675324675324675, "step": 88000 }, { "epoch": 3.7704040101109637, "grad_norm": 2.7231647968292236, "learning_rate": 1.4194315641118989e-05, "loss": 0.21390280723571778, "memory(GiB)": 73.83, "step": 88005, "token_acc": 0.6057513914656771, "train_speed(iter/s)": 1.456684 }, { "epoch": 3.7706182254402125, "grad_norm": 3.206044912338257, "learning_rate": 1.4189618693850444e-05, "loss": 0.5781827449798584, "memory(GiB)": 73.83, "step": 88010, "token_acc": 0.899390243902439, "train_speed(iter/s)": 1.456707 }, { "epoch": 3.7708324407694613, "grad_norm": 2.323753595352173, "learning_rate": 1.4184922395320543e-05, "loss": 0.32648696899414065, "memory(GiB)": 73.83, "step": 88015, "token_acc": 0.922360248447205, "train_speed(iter/s)": 1.456713 }, { "epoch": 3.7710466560987106, "grad_norm": 0.33933207392692566, "learning_rate": 1.4180226745614379e-05, "loss": 0.16078503131866456, "memory(GiB)": 73.83, "step": 88020, "token_acc": 0.9715302491103203, "train_speed(iter/s)": 1.456707 }, { "epoch": 3.7712608714279594, "grad_norm": 3.771108627319336, "learning_rate": 1.4175531744816989e-05, "loss": 0.4059595584869385, "memory(GiB)": 73.83, "step": 88025, "token_acc": 0.9208860759493671, "train_speed(iter/s)": 1.456711 }, { "epoch": 3.771475086757208, "grad_norm": 4.587892055511475, "learning_rate": 1.4170837393013465e-05, "loss": 0.37561323642730715, "memory(GiB)": 73.83, "step": 88030, "token_acc": 0.9216300940438872, "train_speed(iter/s)": 1.456714 }, { "epoch": 3.7716893020864575, "grad_norm": 1.8240737915039062, "learning_rate": 1.416614369028883e-05, "loss": 0.14492737054824828, "memory(GiB)": 73.83, "step": 88035, "token_acc": 0.967741935483871, "train_speed(iter/s)": 1.456718 }, { "epoch": 3.7719035174157063, "grad_norm": 2.6911439895629883, "learning_rate": 1.4161450636728119e-05, "loss": 0.18010106086730956, "memory(GiB)": 73.83, "step": 88040, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.456719 }, { "epoch": 3.772117732744955, "grad_norm": 0.2951025068759918, "learning_rate": 1.4156758232416345e-05, "loss": 0.5817102432250977, "memory(GiB)": 73.83, "step": 88045, "token_acc": 0.878419452887538, "train_speed(iter/s)": 1.456721 }, { "epoch": 3.7723319480742044, "grad_norm": 3.4875078201293945, "learning_rate": 1.4152066477438508e-05, "loss": 0.23476829528808593, "memory(GiB)": 73.83, "step": 88050, "token_acc": 0.9586466165413534, "train_speed(iter/s)": 1.456724 }, { "epoch": 3.772546163403453, "grad_norm": 3.5090222358703613, "learning_rate": 1.4147375371879628e-05, "loss": 0.3248572826385498, "memory(GiB)": 73.83, "step": 88055, "token_acc": 0.924187725631769, "train_speed(iter/s)": 1.456732 }, { "epoch": 3.772760378732702, "grad_norm": 6.331042766571045, "learning_rate": 1.4142684915824678e-05, "loss": 0.24439048767089844, "memory(GiB)": 73.83, "step": 88060, "token_acc": 0.9343065693430657, "train_speed(iter/s)": 1.456748 }, { "epoch": 3.7729745940619512, "grad_norm": 4.909460067749023, "learning_rate": 1.413799510935861e-05, "loss": 0.6453730583190918, "memory(GiB)": 73.83, "step": 88065, "token_acc": 0.8613138686131386, "train_speed(iter/s)": 1.456752 }, { "epoch": 3.7731888093912, "grad_norm": 5.145352840423584, "learning_rate": 1.4133305952566416e-05, "loss": 0.31262590885162356, "memory(GiB)": 73.83, "step": 88070, "token_acc": 0.9360902255639098, "train_speed(iter/s)": 1.456758 }, { "epoch": 3.773403024720449, "grad_norm": 2.757838726043701, "learning_rate": 1.4128617445533037e-05, "loss": 0.21394648551940917, "memory(GiB)": 73.83, "step": 88075, "token_acc": 0.933649289099526, "train_speed(iter/s)": 1.456757 }, { "epoch": 3.773617240049698, "grad_norm": 0.11779265105724335, "learning_rate": 1.4123929588343398e-05, "loss": 0.27238690853118896, "memory(GiB)": 73.83, "step": 88080, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.456759 }, { "epoch": 3.773831455378947, "grad_norm": 0.06996823847293854, "learning_rate": 1.4119242381082415e-05, "loss": 0.36380393505096437, "memory(GiB)": 73.83, "step": 88085, "token_acc": 0.9358490566037736, "train_speed(iter/s)": 1.456761 }, { "epoch": 3.7740456707081957, "grad_norm": 8.139211654663086, "learning_rate": 1.411455582383503e-05, "loss": 0.8194849967956543, "memory(GiB)": 73.83, "step": 88090, "token_acc": 0.832, "train_speed(iter/s)": 1.456772 }, { "epoch": 3.774259886037445, "grad_norm": 2.216822624206543, "learning_rate": 1.4109869916686125e-05, "loss": 0.536731481552124, "memory(GiB)": 73.83, "step": 88095, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.456774 }, { "epoch": 3.774474101366694, "grad_norm": 0.2714904844760895, "learning_rate": 1.41051846597206e-05, "loss": 0.3082357168197632, "memory(GiB)": 73.83, "step": 88100, "token_acc": 0.9453924914675768, "train_speed(iter/s)": 1.45678 }, { "epoch": 3.7746883166959426, "grad_norm": 0.4630705714225769, "learning_rate": 1.4100500053023324e-05, "loss": 0.37295663356781006, "memory(GiB)": 73.83, "step": 88105, "token_acc": 0.896797153024911, "train_speed(iter/s)": 1.456793 }, { "epoch": 3.774902532025192, "grad_norm": 2.247687578201294, "learning_rate": 1.4095816096679155e-05, "loss": 0.28163738250732423, "memory(GiB)": 73.83, "step": 88110, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.456793 }, { "epoch": 3.7751167473544407, "grad_norm": 3.450664520263672, "learning_rate": 1.409113279077297e-05, "loss": 0.6055199146270752, "memory(GiB)": 73.83, "step": 88115, "token_acc": 0.8773841961852861, "train_speed(iter/s)": 1.456792 }, { "epoch": 3.7753309626836895, "grad_norm": 3.0984714031219482, "learning_rate": 1.40864501353896e-05, "loss": 0.3809164047241211, "memory(GiB)": 73.83, "step": 88120, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.456792 }, { "epoch": 3.7755451780129388, "grad_norm": 3.3192341327667236, "learning_rate": 1.4081768130613877e-05, "loss": 0.3897262096405029, "memory(GiB)": 73.83, "step": 88125, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.456801 }, { "epoch": 3.7757593933421876, "grad_norm": 5.356209754943848, "learning_rate": 1.4077086776530618e-05, "loss": 0.5977663516998291, "memory(GiB)": 73.83, "step": 88130, "token_acc": 0.8613861386138614, "train_speed(iter/s)": 1.456801 }, { "epoch": 3.7759736086714364, "grad_norm": 2.903724431991577, "learning_rate": 1.4072406073224608e-05, "loss": 0.45613579750061034, "memory(GiB)": 73.83, "step": 88135, "token_acc": 0.896875, "train_speed(iter/s)": 1.456804 }, { "epoch": 3.7761878240006856, "grad_norm": 4.213232040405273, "learning_rate": 1.4067726020780675e-05, "loss": 0.37814922332763673, "memory(GiB)": 73.83, "step": 88140, "token_acc": 0.9358490566037736, "train_speed(iter/s)": 1.456815 }, { "epoch": 3.7764020393299345, "grad_norm": 2.580904006958008, "learning_rate": 1.4063046619283604e-05, "loss": 0.35132005214691164, "memory(GiB)": 73.83, "step": 88145, "token_acc": 0.9292307692307692, "train_speed(iter/s)": 1.456814 }, { "epoch": 3.7766162546591833, "grad_norm": 3.8372132778167725, "learning_rate": 1.4058367868818156e-05, "loss": 0.38877644538879397, "memory(GiB)": 73.83, "step": 88150, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.456815 }, { "epoch": 3.7768304699884325, "grad_norm": 1.4773447513580322, "learning_rate": 1.4053689769469085e-05, "loss": 0.4701677322387695, "memory(GiB)": 73.83, "step": 88155, "token_acc": 0.8870967741935484, "train_speed(iter/s)": 1.456818 }, { "epoch": 3.7770446853176813, "grad_norm": 3.652554512023926, "learning_rate": 1.4049012321321147e-05, "loss": 0.29754798412322997, "memory(GiB)": 73.83, "step": 88160, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.456821 }, { "epoch": 3.77725890064693, "grad_norm": 1.2534102201461792, "learning_rate": 1.4044335524459078e-05, "loss": 0.19804657697677613, "memory(GiB)": 73.83, "step": 88165, "token_acc": 0.953125, "train_speed(iter/s)": 1.456836 }, { "epoch": 3.7774731159761794, "grad_norm": 3.5923354625701904, "learning_rate": 1.4039659378967584e-05, "loss": 0.46805105209350584, "memory(GiB)": 73.83, "step": 88170, "token_acc": 0.8900343642611683, "train_speed(iter/s)": 1.456859 }, { "epoch": 3.777687331305428, "grad_norm": 2.3633527755737305, "learning_rate": 1.4034983884931407e-05, "loss": 0.5085732460021972, "memory(GiB)": 73.83, "step": 88175, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.45687 }, { "epoch": 3.777901546634677, "grad_norm": 4.123544216156006, "learning_rate": 1.4030309042435236e-05, "loss": 0.2081777572631836, "memory(GiB)": 73.83, "step": 88180, "token_acc": 0.9470404984423676, "train_speed(iter/s)": 1.456873 }, { "epoch": 3.7781157619639263, "grad_norm": 4.17136812210083, "learning_rate": 1.4025634851563762e-05, "loss": 0.3354024410247803, "memory(GiB)": 73.83, "step": 88185, "token_acc": 0.9186746987951807, "train_speed(iter/s)": 1.456874 }, { "epoch": 3.778329977293175, "grad_norm": 1.6605355739593506, "learning_rate": 1.4020961312401653e-05, "loss": 0.2461474895477295, "memory(GiB)": 73.83, "step": 88190, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.456879 }, { "epoch": 3.778544192622424, "grad_norm": 4.635175704956055, "learning_rate": 1.4016288425033574e-05, "loss": 0.4268338680267334, "memory(GiB)": 73.83, "step": 88195, "token_acc": 0.92, "train_speed(iter/s)": 1.456881 }, { "epoch": 3.778758407951673, "grad_norm": 4.6166090965271, "learning_rate": 1.4011616189544197e-05, "loss": 0.6041233539581299, "memory(GiB)": 73.83, "step": 88200, "token_acc": 0.8782894736842105, "train_speed(iter/s)": 1.456882 }, { "epoch": 3.778972623280922, "grad_norm": 2.845432996749878, "learning_rate": 1.4006944606018147e-05, "loss": 0.2894142627716064, "memory(GiB)": 73.83, "step": 88205, "token_acc": 0.9407665505226481, "train_speed(iter/s)": 1.456886 }, { "epoch": 3.779186838610171, "grad_norm": 2.505445957183838, "learning_rate": 1.4002273674540067e-05, "loss": 0.3616512298583984, "memory(GiB)": 73.83, "step": 88210, "token_acc": 0.9233038348082596, "train_speed(iter/s)": 1.456889 }, { "epoch": 3.77940105393942, "grad_norm": 3.2208778858184814, "learning_rate": 1.3997603395194548e-05, "loss": 0.1678795576095581, "memory(GiB)": 73.83, "step": 88215, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.456897 }, { "epoch": 3.779615269268669, "grad_norm": 0.7334513664245605, "learning_rate": 1.3992933768066235e-05, "loss": 0.1099295973777771, "memory(GiB)": 73.83, "step": 88220, "token_acc": 0.9706959706959707, "train_speed(iter/s)": 1.456901 }, { "epoch": 3.7798294845979177, "grad_norm": 2.411384105682373, "learning_rate": 1.3988264793239702e-05, "loss": 0.24616823196411133, "memory(GiB)": 73.83, "step": 88225, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.456906 }, { "epoch": 3.780043699927167, "grad_norm": 5.995630741119385, "learning_rate": 1.398359647079952e-05, "loss": 0.3973590850830078, "memory(GiB)": 73.83, "step": 88230, "token_acc": 0.9312977099236641, "train_speed(iter/s)": 1.456923 }, { "epoch": 3.7802579152564157, "grad_norm": 4.394443035125732, "learning_rate": 1.3978928800830288e-05, "loss": 0.38198940753936766, "memory(GiB)": 73.83, "step": 88235, "token_acc": 0.920265780730897, "train_speed(iter/s)": 1.456933 }, { "epoch": 3.7804721305856646, "grad_norm": 3.2235360145568848, "learning_rate": 1.3974261783416554e-05, "loss": 0.4461069583892822, "memory(GiB)": 73.83, "step": 88240, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.456948 }, { "epoch": 3.780686345914914, "grad_norm": 0.6115776300430298, "learning_rate": 1.3969595418642862e-05, "loss": 0.2951629877090454, "memory(GiB)": 73.83, "step": 88245, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.456949 }, { "epoch": 3.7809005612441626, "grad_norm": 4.836719512939453, "learning_rate": 1.396492970659375e-05, "loss": 0.39173736572265627, "memory(GiB)": 73.83, "step": 88250, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.456948 }, { "epoch": 3.7811147765734114, "grad_norm": 1.4876501560211182, "learning_rate": 1.3960264647353721e-05, "loss": 0.07967473864555359, "memory(GiB)": 73.83, "step": 88255, "token_acc": 0.9851301115241635, "train_speed(iter/s)": 1.456945 }, { "epoch": 3.7813289919026607, "grad_norm": 2.6014058589935303, "learning_rate": 1.3955600241007322e-05, "loss": 0.29705469608306884, "memory(GiB)": 73.83, "step": 88260, "token_acc": 0.95, "train_speed(iter/s)": 1.456953 }, { "epoch": 3.7815432072319095, "grad_norm": 7.084413051605225, "learning_rate": 1.3950936487639039e-05, "loss": 0.3924772500991821, "memory(GiB)": 73.83, "step": 88265, "token_acc": 0.92, "train_speed(iter/s)": 1.456966 }, { "epoch": 3.7817574225611583, "grad_norm": 4.621966361999512, "learning_rate": 1.3946273387333352e-05, "loss": 0.2480093002319336, "memory(GiB)": 73.83, "step": 88270, "token_acc": 0.9591194968553459, "train_speed(iter/s)": 1.456973 }, { "epoch": 3.7819716378904076, "grad_norm": 4.75291109085083, "learning_rate": 1.3941610940174748e-05, "loss": 0.48856382369995116, "memory(GiB)": 73.83, "step": 88275, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.456975 }, { "epoch": 3.7821858532196564, "grad_norm": 2.7756266593933105, "learning_rate": 1.3936949146247675e-05, "loss": 0.43760342597961427, "memory(GiB)": 73.83, "step": 88280, "token_acc": 0.90625, "train_speed(iter/s)": 1.456988 }, { "epoch": 3.782400068548905, "grad_norm": 0.9339724183082581, "learning_rate": 1.3932288005636607e-05, "loss": 0.22516329288482667, "memory(GiB)": 73.83, "step": 88285, "token_acc": 0.946843853820598, "train_speed(iter/s)": 1.456997 }, { "epoch": 3.7826142838781545, "grad_norm": 0.45739319920539856, "learning_rate": 1.3927627518425967e-05, "loss": 0.11948771476745605, "memory(GiB)": 73.83, "step": 88290, "token_acc": 0.9676258992805755, "train_speed(iter/s)": 1.457005 }, { "epoch": 3.7828284992074033, "grad_norm": 2.1895751953125, "learning_rate": 1.3922967684700206e-05, "loss": 0.21545469760894775, "memory(GiB)": 73.83, "step": 88295, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.457016 }, { "epoch": 3.783042714536652, "grad_norm": 2.644301176071167, "learning_rate": 1.3918308504543725e-05, "loss": 0.25476765632629395, "memory(GiB)": 73.83, "step": 88300, "token_acc": 0.9371069182389937, "train_speed(iter/s)": 1.457022 }, { "epoch": 3.7832569298659013, "grad_norm": 4.195491790771484, "learning_rate": 1.3913649978040939e-05, "loss": 0.416339635848999, "memory(GiB)": 73.83, "step": 88305, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.457022 }, { "epoch": 3.78347114519515, "grad_norm": 4.5467376708984375, "learning_rate": 1.390899210527623e-05, "loss": 0.5892419815063477, "memory(GiB)": 73.83, "step": 88310, "token_acc": 0.882943143812709, "train_speed(iter/s)": 1.457027 }, { "epoch": 3.783685360524399, "grad_norm": 3.3766067028045654, "learning_rate": 1.3904334886333975e-05, "loss": 0.33979482650756837, "memory(GiB)": 73.83, "step": 88315, "token_acc": 0.9263565891472868, "train_speed(iter/s)": 1.457031 }, { "epoch": 3.7838995758536482, "grad_norm": 2.1209523677825928, "learning_rate": 1.3899678321298565e-05, "loss": 0.21984195709228516, "memory(GiB)": 73.83, "step": 88320, "token_acc": 0.9477911646586346, "train_speed(iter/s)": 1.457034 }, { "epoch": 3.784113791182897, "grad_norm": 2.34822940826416, "learning_rate": 1.3895022410254348e-05, "loss": 0.16399235725402833, "memory(GiB)": 73.83, "step": 88325, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.457035 }, { "epoch": 3.784328006512146, "grad_norm": 3.1278631687164307, "learning_rate": 1.3890367153285672e-05, "loss": 0.4156914234161377, "memory(GiB)": 73.83, "step": 88330, "token_acc": 0.923728813559322, "train_speed(iter/s)": 1.457049 }, { "epoch": 3.784542221841395, "grad_norm": 5.097813129425049, "learning_rate": 1.3885712550476864e-05, "loss": 0.3119208812713623, "memory(GiB)": 73.83, "step": 88335, "token_acc": 0.9304635761589404, "train_speed(iter/s)": 1.457052 }, { "epoch": 3.784756437170644, "grad_norm": 3.241579294204712, "learning_rate": 1.388105860191224e-05, "loss": 0.538060998916626, "memory(GiB)": 73.83, "step": 88340, "token_acc": 0.8949044585987261, "train_speed(iter/s)": 1.457052 }, { "epoch": 3.7849706524998927, "grad_norm": 2.766648292541504, "learning_rate": 1.3876405307676133e-05, "loss": 0.3017817735671997, "memory(GiB)": 73.83, "step": 88345, "token_acc": 0.9272151898734177, "train_speed(iter/s)": 1.457052 }, { "epoch": 3.785184867829142, "grad_norm": 1.519771933555603, "learning_rate": 1.3871752667852833e-05, "loss": 0.12503559589385987, "memory(GiB)": 73.83, "step": 88350, "token_acc": 0.9648562300319489, "train_speed(iter/s)": 1.457057 }, { "epoch": 3.785399083158391, "grad_norm": 8.596531867980957, "learning_rate": 1.3867100682526624e-05, "loss": 0.6158252716064453, "memory(GiB)": 73.83, "step": 88355, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.457057 }, { "epoch": 3.7856132984876396, "grad_norm": 3.0548582077026367, "learning_rate": 1.386244935178178e-05, "loss": 0.4094383716583252, "memory(GiB)": 73.83, "step": 88360, "token_acc": 0.8942598187311178, "train_speed(iter/s)": 1.457054 }, { "epoch": 3.785827513816889, "grad_norm": 3.7040438652038574, "learning_rate": 1.3857798675702555e-05, "loss": 0.503433084487915, "memory(GiB)": 73.83, "step": 88365, "token_acc": 0.9127906976744186, "train_speed(iter/s)": 1.457053 }, { "epoch": 3.7860417291461377, "grad_norm": 4.602436065673828, "learning_rate": 1.385314865437322e-05, "loss": 0.43120441436767576, "memory(GiB)": 73.83, "step": 88370, "token_acc": 0.900398406374502, "train_speed(iter/s)": 1.457063 }, { "epoch": 3.7862559444753865, "grad_norm": 2.964221477508545, "learning_rate": 1.3848499287877998e-05, "loss": 0.4771317481994629, "memory(GiB)": 73.83, "step": 88375, "token_acc": 0.898360655737705, "train_speed(iter/s)": 1.457064 }, { "epoch": 3.7864701598046357, "grad_norm": 6.530442714691162, "learning_rate": 1.3843850576301137e-05, "loss": 0.25171163082122805, "memory(GiB)": 73.83, "step": 88380, "token_acc": 0.93, "train_speed(iter/s)": 1.457065 }, { "epoch": 3.7866843751338846, "grad_norm": 4.519328594207764, "learning_rate": 1.3839202519726835e-05, "loss": 0.21986637115478516, "memory(GiB)": 73.83, "step": 88385, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.457077 }, { "epoch": 3.7868985904631334, "grad_norm": 3.0779690742492676, "learning_rate": 1.3834555118239306e-05, "loss": 0.2490718126296997, "memory(GiB)": 73.83, "step": 88390, "token_acc": 0.9494584837545126, "train_speed(iter/s)": 1.457082 }, { "epoch": 3.7871128057923826, "grad_norm": 3.1946208477020264, "learning_rate": 1.3829908371922734e-05, "loss": 0.32241086959838866, "memory(GiB)": 73.83, "step": 88395, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.457088 }, { "epoch": 3.7873270211216314, "grad_norm": 1.5969773530960083, "learning_rate": 1.382526228086129e-05, "loss": 0.6178904056549073, "memory(GiB)": 73.83, "step": 88400, "token_acc": 0.8771929824561403, "train_speed(iter/s)": 1.457088 }, { "epoch": 3.7875412364508803, "grad_norm": 1.9816948175430298, "learning_rate": 1.3820616845139168e-05, "loss": 0.5006093978881836, "memory(GiB)": 73.83, "step": 88405, "token_acc": 0.9271255060728745, "train_speed(iter/s)": 1.457094 }, { "epoch": 3.7877554517801295, "grad_norm": 0.09972275048494339, "learning_rate": 1.381597206484051e-05, "loss": 0.2125955581665039, "memory(GiB)": 73.83, "step": 88410, "token_acc": 0.9588014981273408, "train_speed(iter/s)": 1.457095 }, { "epoch": 3.7879696671093783, "grad_norm": 3.3093931674957275, "learning_rate": 1.3811327940049462e-05, "loss": 0.2723142147064209, "memory(GiB)": 73.83, "step": 88415, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.457098 }, { "epoch": 3.788183882438627, "grad_norm": 3.6358253955841064, "learning_rate": 1.3806684470850156e-05, "loss": 0.4129199504852295, "memory(GiB)": 73.83, "step": 88420, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.457098 }, { "epoch": 3.7883980977678764, "grad_norm": 4.4815354347229, "learning_rate": 1.3802041657326698e-05, "loss": 0.5469314575195312, "memory(GiB)": 73.83, "step": 88425, "token_acc": 0.8838709677419355, "train_speed(iter/s)": 1.457102 }, { "epoch": 3.788612313097125, "grad_norm": 3.8970444202423096, "learning_rate": 1.3797399499563229e-05, "loss": 0.37824859619140627, "memory(GiB)": 73.83, "step": 88430, "token_acc": 0.8966942148760331, "train_speed(iter/s)": 1.457103 }, { "epoch": 3.788826528426374, "grad_norm": 4.211906909942627, "learning_rate": 1.3792757997643806e-05, "loss": 0.45983190536499025, "memory(GiB)": 73.83, "step": 88435, "token_acc": 0.907051282051282, "train_speed(iter/s)": 1.457115 }, { "epoch": 3.7890407437556233, "grad_norm": 0.9726155996322632, "learning_rate": 1.3788117151652563e-05, "loss": 0.3672039985656738, "memory(GiB)": 73.83, "step": 88440, "token_acc": 0.92, "train_speed(iter/s)": 1.457122 }, { "epoch": 3.789254959084872, "grad_norm": 1.9194949865341187, "learning_rate": 1.3783476961673541e-05, "loss": 0.2607538938522339, "memory(GiB)": 73.83, "step": 88445, "token_acc": 0.9765395894428153, "train_speed(iter/s)": 1.457133 }, { "epoch": 3.789469174414121, "grad_norm": 3.9736742973327637, "learning_rate": 1.3778837427790809e-05, "loss": 0.4226045608520508, "memory(GiB)": 73.83, "step": 88450, "token_acc": 0.9173553719008265, "train_speed(iter/s)": 1.457133 }, { "epoch": 3.78968338974337, "grad_norm": 1.555812120437622, "learning_rate": 1.3774198550088413e-05, "loss": 0.2770125150680542, "memory(GiB)": 73.83, "step": 88455, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.457134 }, { "epoch": 3.789897605072619, "grad_norm": 1.008170247077942, "learning_rate": 1.3769560328650383e-05, "loss": 0.12698500156402587, "memory(GiB)": 73.83, "step": 88460, "token_acc": 0.9724137931034482, "train_speed(iter/s)": 1.457131 }, { "epoch": 3.790111820401868, "grad_norm": 1.4525030851364136, "learning_rate": 1.3764922763560767e-05, "loss": 0.15052647590637208, "memory(GiB)": 73.83, "step": 88465, "token_acc": 0.9537953795379538, "train_speed(iter/s)": 1.457136 }, { "epoch": 3.790326035731117, "grad_norm": 8.258538246154785, "learning_rate": 1.3760285854903566e-05, "loss": 0.38093137741088867, "memory(GiB)": 73.83, "step": 88470, "token_acc": 0.9094202898550725, "train_speed(iter/s)": 1.457146 }, { "epoch": 3.790540251060366, "grad_norm": 2.407921314239502, "learning_rate": 1.3755649602762777e-05, "loss": 0.24075047969818114, "memory(GiB)": 73.83, "step": 88475, "token_acc": 0.946843853820598, "train_speed(iter/s)": 1.457149 }, { "epoch": 3.7907544663896147, "grad_norm": 1.6778743267059326, "learning_rate": 1.3751014007222402e-05, "loss": 0.3443167686462402, "memory(GiB)": 73.83, "step": 88480, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.457168 }, { "epoch": 3.790968681718864, "grad_norm": 4.132898807525635, "learning_rate": 1.3746379068366394e-05, "loss": 0.39044556617736814, "memory(GiB)": 73.83, "step": 88485, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.457171 }, { "epoch": 3.7911828970481127, "grad_norm": 3.4288530349731445, "learning_rate": 1.374174478627876e-05, "loss": 0.39546856880187986, "memory(GiB)": 73.83, "step": 88490, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.457174 }, { "epoch": 3.7913971123773615, "grad_norm": 4.62682580947876, "learning_rate": 1.3737111161043426e-05, "loss": 0.4137721061706543, "memory(GiB)": 73.83, "step": 88495, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.45718 }, { "epoch": 3.791611327706611, "grad_norm": 3.7608323097229004, "learning_rate": 1.3732478192744342e-05, "loss": 0.23770203590393066, "memory(GiB)": 73.83, "step": 88500, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.457188 }, { "epoch": 3.791611327706611, "eval_loss": 2.520193099975586, "eval_runtime": 11.4377, "eval_samples_per_second": 8.743, "eval_steps_per_second": 8.743, "eval_token_acc": 0.46473029045643155, "step": 88500 }, { "epoch": 3.7918255430358596, "grad_norm": 2.9907424449920654, "learning_rate": 1.3727845881465434e-05, "loss": 0.16600488424301146, "memory(GiB)": 73.83, "step": 88505, "token_acc": 0.604, "train_speed(iter/s)": 1.456891 }, { "epoch": 3.7920397583651084, "grad_norm": 7.0753326416015625, "learning_rate": 1.3723214227290615e-05, "loss": 0.4355946063995361, "memory(GiB)": 73.83, "step": 88510, "token_acc": 0.9159663865546218, "train_speed(iter/s)": 1.456897 }, { "epoch": 3.7922539736943577, "grad_norm": 5.118392467498779, "learning_rate": 1.3718583230303794e-05, "loss": 0.3744483947753906, "memory(GiB)": 73.83, "step": 88515, "token_acc": 0.9186746987951807, "train_speed(iter/s)": 1.456894 }, { "epoch": 3.7924681890236065, "grad_norm": 1.5211447477340698, "learning_rate": 1.3713952890588894e-05, "loss": 0.29867141246795653, "memory(GiB)": 73.83, "step": 88520, "token_acc": 0.9376947040498442, "train_speed(iter/s)": 1.456897 }, { "epoch": 3.7926824043528553, "grad_norm": 6.081250190734863, "learning_rate": 1.370932320822977e-05, "loss": 0.5975732803344727, "memory(GiB)": 73.83, "step": 88525, "token_acc": 0.9029850746268657, "train_speed(iter/s)": 1.456905 }, { "epoch": 3.7928966196821046, "grad_norm": 4.0585174560546875, "learning_rate": 1.3704694183310301e-05, "loss": 0.2027266502380371, "memory(GiB)": 73.83, "step": 88530, "token_acc": 0.953307392996109, "train_speed(iter/s)": 1.456911 }, { "epoch": 3.7931108350113534, "grad_norm": 3.2841854095458984, "learning_rate": 1.3700065815914348e-05, "loss": 0.47589359283447263, "memory(GiB)": 73.83, "step": 88535, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.456927 }, { "epoch": 3.793325050340602, "grad_norm": 4.710651874542236, "learning_rate": 1.3695438106125757e-05, "loss": 0.7293900966644287, "memory(GiB)": 73.83, "step": 88540, "token_acc": 0.8549382716049383, "train_speed(iter/s)": 1.456932 }, { "epoch": 3.7935392656698514, "grad_norm": 4.143630504608154, "learning_rate": 1.369081105402834e-05, "loss": 0.2016227960586548, "memory(GiB)": 73.83, "step": 88545, "token_acc": 0.9698996655518395, "train_speed(iter/s)": 1.456935 }, { "epoch": 3.7937534809991003, "grad_norm": 2.046172857284546, "learning_rate": 1.3686184659705959e-05, "loss": 0.1062273383140564, "memory(GiB)": 73.83, "step": 88550, "token_acc": 0.9685314685314685, "train_speed(iter/s)": 1.456941 }, { "epoch": 3.793967696328349, "grad_norm": 4.243168354034424, "learning_rate": 1.368155892324241e-05, "loss": 0.4026055335998535, "memory(GiB)": 73.83, "step": 88555, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.45694 }, { "epoch": 3.7941819116575983, "grad_norm": 2.436600923538208, "learning_rate": 1.3676933844721484e-05, "loss": 0.2866849899291992, "memory(GiB)": 73.83, "step": 88560, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.45694 }, { "epoch": 3.794396126986847, "grad_norm": 3.3253512382507324, "learning_rate": 1.3672309424226981e-05, "loss": 0.42428159713745117, "memory(GiB)": 73.83, "step": 88565, "token_acc": 0.8969072164948454, "train_speed(iter/s)": 1.456943 }, { "epoch": 3.794610342316096, "grad_norm": 2.539125442504883, "learning_rate": 1.3667685661842655e-05, "loss": 0.15588984489440919, "memory(GiB)": 73.83, "step": 88570, "token_acc": 0.9550561797752809, "train_speed(iter/s)": 1.456943 }, { "epoch": 3.794824557645345, "grad_norm": 4.428744316101074, "learning_rate": 1.36630625576523e-05, "loss": 0.4510500431060791, "memory(GiB)": 73.83, "step": 88575, "token_acc": 0.9044117647058824, "train_speed(iter/s)": 1.456944 }, { "epoch": 3.795038772974594, "grad_norm": 2.417034387588501, "learning_rate": 1.3658440111739657e-05, "loss": 0.31004602909088136, "memory(GiB)": 73.83, "step": 88580, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.456947 }, { "epoch": 3.795252988303843, "grad_norm": 3.53369140625, "learning_rate": 1.3653818324188444e-05, "loss": 0.27495059967041013, "memory(GiB)": 73.83, "step": 88585, "token_acc": 0.9409937888198758, "train_speed(iter/s)": 1.456947 }, { "epoch": 3.795467203633092, "grad_norm": 2.9974000453948975, "learning_rate": 1.364919719508242e-05, "loss": 0.6425052642822265, "memory(GiB)": 73.83, "step": 88590, "token_acc": 0.8866396761133604, "train_speed(iter/s)": 1.456949 }, { "epoch": 3.795681418962341, "grad_norm": 1.710260272026062, "learning_rate": 1.3644576724505292e-05, "loss": 0.19031543731689454, "memory(GiB)": 73.83, "step": 88595, "token_acc": 0.9638989169675091, "train_speed(iter/s)": 1.456954 }, { "epoch": 3.7958956342915897, "grad_norm": 2.6909546852111816, "learning_rate": 1.3639956912540764e-05, "loss": 0.29423458576202394, "memory(GiB)": 73.83, "step": 88600, "token_acc": 0.9431818181818182, "train_speed(iter/s)": 1.456956 }, { "epoch": 3.796109849620839, "grad_norm": 2.1830363273620605, "learning_rate": 1.3635337759272503e-05, "loss": 0.14689942598342895, "memory(GiB)": 73.83, "step": 88605, "token_acc": 0.9644268774703557, "train_speed(iter/s)": 1.456959 }, { "epoch": 3.796324064950088, "grad_norm": 11.354378700256348, "learning_rate": 1.363071926478423e-05, "loss": 0.5218536376953125, "memory(GiB)": 73.83, "step": 88610, "token_acc": 0.9205020920502092, "train_speed(iter/s)": 1.456957 }, { "epoch": 3.7965382802793366, "grad_norm": 3.13045597076416, "learning_rate": 1.3626101429159593e-05, "loss": 0.33273146152496336, "memory(GiB)": 73.83, "step": 88615, "token_acc": 0.9151785714285714, "train_speed(iter/s)": 1.456959 }, { "epoch": 3.796752495608586, "grad_norm": 2.1521685123443604, "learning_rate": 1.3621484252482252e-05, "loss": 0.29184784889221194, "memory(GiB)": 73.83, "step": 88620, "token_acc": 0.9367588932806324, "train_speed(iter/s)": 1.456969 }, { "epoch": 3.7969667109378347, "grad_norm": 1.8804973363876343, "learning_rate": 1.3616867734835854e-05, "loss": 0.41937432289123533, "memory(GiB)": 73.83, "step": 88625, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.456974 }, { "epoch": 3.7971809262670835, "grad_norm": 1.758719563484192, "learning_rate": 1.3612251876304005e-05, "loss": 0.2676429510116577, "memory(GiB)": 73.83, "step": 88630, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.456974 }, { "epoch": 3.7973951415963327, "grad_norm": 1.9822144508361816, "learning_rate": 1.3607636676970365e-05, "loss": 0.30718460083007815, "memory(GiB)": 73.83, "step": 88635, "token_acc": 0.9426751592356688, "train_speed(iter/s)": 1.456977 }, { "epoch": 3.7976093569255815, "grad_norm": 0.9961543679237366, "learning_rate": 1.3603022136918526e-05, "loss": 0.21139798164367676, "memory(GiB)": 73.83, "step": 88640, "token_acc": 0.94, "train_speed(iter/s)": 1.456979 }, { "epoch": 3.7978235722548304, "grad_norm": 4.012458801269531, "learning_rate": 1.3598408256232082e-05, "loss": 0.3961158275604248, "memory(GiB)": 73.83, "step": 88645, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.45698 }, { "epoch": 3.7980377875840796, "grad_norm": 2.2158424854278564, "learning_rate": 1.3593795034994622e-05, "loss": 0.34885568618774415, "memory(GiB)": 73.83, "step": 88650, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.45698 }, { "epoch": 3.7982520029133284, "grad_norm": 1.9314872026443481, "learning_rate": 1.3589182473289703e-05, "loss": 0.4627446174621582, "memory(GiB)": 73.83, "step": 88655, "token_acc": 0.892128279883382, "train_speed(iter/s)": 1.45698 }, { "epoch": 3.7984662182425772, "grad_norm": 3.0533666610717773, "learning_rate": 1.3584570571200895e-05, "loss": 0.3048647165298462, "memory(GiB)": 73.83, "step": 88660, "token_acc": 0.9190283400809717, "train_speed(iter/s)": 1.456983 }, { "epoch": 3.7986804335718265, "grad_norm": 3.3210456371307373, "learning_rate": 1.357995932881177e-05, "loss": 0.34237587451934814, "memory(GiB)": 73.83, "step": 88665, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.456993 }, { "epoch": 3.7988946489010753, "grad_norm": 5.287013530731201, "learning_rate": 1.3575348746205847e-05, "loss": 0.44275712966918945, "memory(GiB)": 73.83, "step": 88670, "token_acc": 0.9029126213592233, "train_speed(iter/s)": 1.456989 }, { "epoch": 3.799108864230324, "grad_norm": 6.218090057373047, "learning_rate": 1.3570738823466645e-05, "loss": 0.4933112144470215, "memory(GiB)": 73.83, "step": 88675, "token_acc": 0.9143730886850153, "train_speed(iter/s)": 1.456985 }, { "epoch": 3.7993230795595734, "grad_norm": 2.9756417274475098, "learning_rate": 1.3566129560677687e-05, "loss": 0.2312488079071045, "memory(GiB)": 73.83, "step": 88680, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.456998 }, { "epoch": 3.799537294888822, "grad_norm": 3.540557861328125, "learning_rate": 1.3561520957922469e-05, "loss": 0.4166999340057373, "memory(GiB)": 73.83, "step": 88685, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.456998 }, { "epoch": 3.799751510218071, "grad_norm": 1.8116025924682617, "learning_rate": 1.3556913015284467e-05, "loss": 0.28646020889282225, "memory(GiB)": 73.83, "step": 88690, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.457001 }, { "epoch": 3.7999657255473203, "grad_norm": 2.4937336444854736, "learning_rate": 1.3552305732847186e-05, "loss": 0.3876781940460205, "memory(GiB)": 73.83, "step": 88695, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.457004 }, { "epoch": 3.800179940876569, "grad_norm": 3.5831198692321777, "learning_rate": 1.3547699110694079e-05, "loss": 0.34786272048950195, "memory(GiB)": 73.83, "step": 88700, "token_acc": 0.9051987767584098, "train_speed(iter/s)": 1.457024 }, { "epoch": 3.800394156205818, "grad_norm": 3.5419509410858154, "learning_rate": 1.3543093148908597e-05, "loss": 0.19025396108627318, "memory(GiB)": 73.83, "step": 88705, "token_acc": 0.946360153256705, "train_speed(iter/s)": 1.457021 }, { "epoch": 3.800608371535067, "grad_norm": 3.1547646522521973, "learning_rate": 1.3538487847574183e-05, "loss": 0.17297773361206054, "memory(GiB)": 73.83, "step": 88710, "token_acc": 0.95, "train_speed(iter/s)": 1.457024 }, { "epoch": 3.800822586864316, "grad_norm": 2.9235212802886963, "learning_rate": 1.353388320677425e-05, "loss": 0.4811316967010498, "memory(GiB)": 73.83, "step": 88715, "token_acc": 0.9040590405904059, "train_speed(iter/s)": 1.457029 }, { "epoch": 3.8010368021935648, "grad_norm": 1.8567113876342773, "learning_rate": 1.3529279226592245e-05, "loss": 0.26682870388031005, "memory(GiB)": 73.83, "step": 88720, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.457034 }, { "epoch": 3.801251017522814, "grad_norm": 0.38733482360839844, "learning_rate": 1.3524675907111555e-05, "loss": 0.18547893762588502, "memory(GiB)": 73.83, "step": 88725, "token_acc": 0.9494584837545126, "train_speed(iter/s)": 1.457045 }, { "epoch": 3.801465232852063, "grad_norm": 4.087436676025391, "learning_rate": 1.3520073248415565e-05, "loss": 0.16862367391586303, "memory(GiB)": 73.83, "step": 88730, "token_acc": 0.9506172839506173, "train_speed(iter/s)": 1.457047 }, { "epoch": 3.8016794481813116, "grad_norm": 2.153421401977539, "learning_rate": 1.3515471250587686e-05, "loss": 0.19620712995529174, "memory(GiB)": 73.83, "step": 88735, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.457047 }, { "epoch": 3.801893663510561, "grad_norm": 6.074466228485107, "learning_rate": 1.3510869913711272e-05, "loss": 0.1866615891456604, "memory(GiB)": 73.83, "step": 88740, "token_acc": 0.9649122807017544, "train_speed(iter/s)": 1.457045 }, { "epoch": 3.8021078788398097, "grad_norm": 6.06599235534668, "learning_rate": 1.3506269237869679e-05, "loss": 0.35361537933349607, "memory(GiB)": 73.83, "step": 88745, "token_acc": 0.9367588932806324, "train_speed(iter/s)": 1.457045 }, { "epoch": 3.8023220941690585, "grad_norm": 1.9426521062850952, "learning_rate": 1.3501669223146236e-05, "loss": 0.3799251079559326, "memory(GiB)": 73.83, "step": 88750, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.457052 }, { "epoch": 3.802536309498308, "grad_norm": 2.195702075958252, "learning_rate": 1.3497069869624307e-05, "loss": 0.47948784828186036, "memory(GiB)": 73.83, "step": 88755, "token_acc": 0.8942857142857142, "train_speed(iter/s)": 1.457054 }, { "epoch": 3.8027505248275566, "grad_norm": 1.1096007823944092, "learning_rate": 1.34924711773872e-05, "loss": 0.08100058436393738, "memory(GiB)": 73.83, "step": 88760, "token_acc": 0.972972972972973, "train_speed(iter/s)": 1.457058 }, { "epoch": 3.8029647401568054, "grad_norm": 4.006720542907715, "learning_rate": 1.348787314651822e-05, "loss": 0.5187687397003173, "memory(GiB)": 73.83, "step": 88765, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.457057 }, { "epoch": 3.8031789554860547, "grad_norm": 5.188313961029053, "learning_rate": 1.3483275777100679e-05, "loss": 0.36913561820983887, "memory(GiB)": 73.83, "step": 88770, "token_acc": 0.9235294117647059, "train_speed(iter/s)": 1.457059 }, { "epoch": 3.8033931708153035, "grad_norm": 4.449752330780029, "learning_rate": 1.3478679069217831e-05, "loss": 0.2544647216796875, "memory(GiB)": 73.83, "step": 88775, "token_acc": 0.9282700421940928, "train_speed(iter/s)": 1.457061 }, { "epoch": 3.8036073861445523, "grad_norm": 1.4977693557739258, "learning_rate": 1.3474083022952987e-05, "loss": 0.3329331398010254, "memory(GiB)": 73.83, "step": 88780, "token_acc": 0.9260355029585798, "train_speed(iter/s)": 1.457072 }, { "epoch": 3.8038216014738016, "grad_norm": 4.614362716674805, "learning_rate": 1.3469487638389394e-05, "loss": 0.3831953525543213, "memory(GiB)": 73.83, "step": 88785, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.457076 }, { "epoch": 3.8040358168030504, "grad_norm": 2.6943013668060303, "learning_rate": 1.3464892915610305e-05, "loss": 0.38463678359985354, "memory(GiB)": 73.83, "step": 88790, "token_acc": 0.909967845659164, "train_speed(iter/s)": 1.457079 }, { "epoch": 3.804250032132299, "grad_norm": 2.5188777446746826, "learning_rate": 1.346029885469895e-05, "loss": 0.27452127933502196, "memory(GiB)": 73.83, "step": 88795, "token_acc": 0.9345238095238095, "train_speed(iter/s)": 1.457082 }, { "epoch": 3.8044642474615484, "grad_norm": 3.9475252628326416, "learning_rate": 1.3455705455738543e-05, "loss": 0.4682624816894531, "memory(GiB)": 73.83, "step": 88800, "token_acc": 0.9035714285714286, "train_speed(iter/s)": 1.457086 }, { "epoch": 3.8046784627907972, "grad_norm": 1.7576967477798462, "learning_rate": 1.3451112718812308e-05, "loss": 0.3423571825027466, "memory(GiB)": 73.83, "step": 88805, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.457087 }, { "epoch": 3.804892678120046, "grad_norm": 2.1742186546325684, "learning_rate": 1.3446520644003474e-05, "loss": 0.408040714263916, "memory(GiB)": 73.83, "step": 88810, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.457087 }, { "epoch": 3.8051068934492953, "grad_norm": 3.5067648887634277, "learning_rate": 1.3441929231395207e-05, "loss": 0.3419168949127197, "memory(GiB)": 73.83, "step": 88815, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.45709 }, { "epoch": 3.805321108778544, "grad_norm": 2.324113130569458, "learning_rate": 1.3437338481070689e-05, "loss": 0.3187922477722168, "memory(GiB)": 73.83, "step": 88820, "token_acc": 0.9329268292682927, "train_speed(iter/s)": 1.457097 }, { "epoch": 3.805535324107793, "grad_norm": 1.1809388399124146, "learning_rate": 1.3432748393113075e-05, "loss": 0.48137450218200684, "memory(GiB)": 73.83, "step": 88825, "token_acc": 0.9033232628398792, "train_speed(iter/s)": 1.457113 }, { "epoch": 3.805749539437042, "grad_norm": 2.7040963172912598, "learning_rate": 1.342815896760553e-05, "loss": 0.17008495330810547, "memory(GiB)": 73.83, "step": 88830, "token_acc": 0.9771241830065359, "train_speed(iter/s)": 1.45712 }, { "epoch": 3.805963754766291, "grad_norm": 16.200450897216797, "learning_rate": 1.342357020463118e-05, "loss": 0.3219730854034424, "memory(GiB)": 73.83, "step": 88835, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.457134 }, { "epoch": 3.80617797009554, "grad_norm": 2.23276948928833, "learning_rate": 1.3418982104273182e-05, "loss": 0.35723395347595216, "memory(GiB)": 73.83, "step": 88840, "token_acc": 0.9025270758122743, "train_speed(iter/s)": 1.45713 }, { "epoch": 3.806392185424789, "grad_norm": 2.711451292037964, "learning_rate": 1.3414394666614638e-05, "loss": 0.3587924957275391, "memory(GiB)": 73.83, "step": 88845, "token_acc": 0.928, "train_speed(iter/s)": 1.45713 }, { "epoch": 3.806606400754038, "grad_norm": 1.5188441276550293, "learning_rate": 1.3409807891738652e-05, "loss": 0.3756037712097168, "memory(GiB)": 73.83, "step": 88850, "token_acc": 0.9034267912772586, "train_speed(iter/s)": 1.45714 }, { "epoch": 3.8068206160832867, "grad_norm": 2.898495674133301, "learning_rate": 1.3405221779728327e-05, "loss": 0.1174772024154663, "memory(GiB)": 73.83, "step": 88855, "token_acc": 0.9779874213836478, "train_speed(iter/s)": 1.457144 }, { "epoch": 3.807034831412536, "grad_norm": 1.0761914253234863, "learning_rate": 1.3400636330666716e-05, "loss": 0.2598790884017944, "memory(GiB)": 73.83, "step": 88860, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.457146 }, { "epoch": 3.8072490467417848, "grad_norm": 2.0361828804016113, "learning_rate": 1.3396051544636928e-05, "loss": 0.3408914089202881, "memory(GiB)": 73.83, "step": 88865, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.45715 }, { "epoch": 3.8074632620710336, "grad_norm": 1.6532117128372192, "learning_rate": 1.3391467421722004e-05, "loss": 0.3617713212966919, "memory(GiB)": 73.83, "step": 88870, "token_acc": 0.9085173501577287, "train_speed(iter/s)": 1.457163 }, { "epoch": 3.807677477400283, "grad_norm": 4.851116180419922, "learning_rate": 1.3386883962004992e-05, "loss": 0.4590874195098877, "memory(GiB)": 73.83, "step": 88875, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.457164 }, { "epoch": 3.8078916927295317, "grad_norm": 5.64520788192749, "learning_rate": 1.3382301165568905e-05, "loss": 0.3979153871536255, "memory(GiB)": 73.83, "step": 88880, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.457174 }, { "epoch": 3.8081059080587805, "grad_norm": 2.3750674724578857, "learning_rate": 1.33777190324968e-05, "loss": 0.2935717344284058, "memory(GiB)": 73.83, "step": 88885, "token_acc": 0.941358024691358, "train_speed(iter/s)": 1.457174 }, { "epoch": 3.8083201233880297, "grad_norm": 5.567019462585449, "learning_rate": 1.3373137562871668e-05, "loss": 0.4402009010314941, "memory(GiB)": 73.83, "step": 88890, "token_acc": 0.8900343642611683, "train_speed(iter/s)": 1.457179 }, { "epoch": 3.8085343387172785, "grad_norm": 1.6836751699447632, "learning_rate": 1.336855675677649e-05, "loss": 0.2916707038879395, "memory(GiB)": 73.83, "step": 88895, "token_acc": 0.928125, "train_speed(iter/s)": 1.45718 }, { "epoch": 3.8087485540465273, "grad_norm": 2.5001449584960938, "learning_rate": 1.3363976614294288e-05, "loss": 0.1948784112930298, "memory(GiB)": 73.83, "step": 88900, "token_acc": 0.9551020408163265, "train_speed(iter/s)": 1.457181 }, { "epoch": 3.8089627693757766, "grad_norm": 2.848118305206299, "learning_rate": 1.3359397135508017e-05, "loss": 0.20049328804016114, "memory(GiB)": 73.83, "step": 88905, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.457178 }, { "epoch": 3.8091769847050254, "grad_norm": 4.302458763122559, "learning_rate": 1.3354818320500634e-05, "loss": 0.30415565967559816, "memory(GiB)": 73.83, "step": 88910, "token_acc": 0.9274193548387096, "train_speed(iter/s)": 1.45718 }, { "epoch": 3.8093912000342742, "grad_norm": 2.962663412094116, "learning_rate": 1.3350240169355099e-05, "loss": 0.3467595100402832, "memory(GiB)": 73.83, "step": 88915, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.45718 }, { "epoch": 3.8096054153635235, "grad_norm": 3.9264228343963623, "learning_rate": 1.3345662682154326e-05, "loss": 0.7652600288391114, "memory(GiB)": 73.83, "step": 88920, "token_acc": 0.8765822784810127, "train_speed(iter/s)": 1.457185 }, { "epoch": 3.8098196306927723, "grad_norm": 2.215611457824707, "learning_rate": 1.3341085858981273e-05, "loss": 0.25749294757843016, "memory(GiB)": 73.83, "step": 88925, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.45719 }, { "epoch": 3.810033846022021, "grad_norm": 1.9741681814193726, "learning_rate": 1.3336509699918837e-05, "loss": 0.4165189743041992, "memory(GiB)": 73.83, "step": 88930, "token_acc": 0.9126506024096386, "train_speed(iter/s)": 1.457189 }, { "epoch": 3.8102480613512704, "grad_norm": 1.1905468702316284, "learning_rate": 1.3331934205049924e-05, "loss": 0.19900121688842773, "memory(GiB)": 73.83, "step": 88935, "token_acc": 0.9596774193548387, "train_speed(iter/s)": 1.45719 }, { "epoch": 3.810462276680519, "grad_norm": 3.0563340187072754, "learning_rate": 1.3327359374457415e-05, "loss": 0.37336030006408694, "memory(GiB)": 73.83, "step": 88940, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.457194 }, { "epoch": 3.810676492009768, "grad_norm": 1.0985363721847534, "learning_rate": 1.3322785208224181e-05, "loss": 0.25724425315856936, "memory(GiB)": 73.83, "step": 88945, "token_acc": 0.9520295202952029, "train_speed(iter/s)": 1.457194 }, { "epoch": 3.8108907073390172, "grad_norm": 3.9021739959716797, "learning_rate": 1.3318211706433114e-05, "loss": 0.1916749358177185, "memory(GiB)": 73.83, "step": 88950, "token_acc": 0.9483870967741935, "train_speed(iter/s)": 1.457206 }, { "epoch": 3.811104922668266, "grad_norm": 5.484462261199951, "learning_rate": 1.331363886916704e-05, "loss": 0.3795084238052368, "memory(GiB)": 73.83, "step": 88955, "token_acc": 0.9273927392739274, "train_speed(iter/s)": 1.457205 }, { "epoch": 3.811319137997515, "grad_norm": 5.836116790771484, "learning_rate": 1.3309066696508826e-05, "loss": 0.6640644073486328, "memory(GiB)": 73.83, "step": 88960, "token_acc": 0.8628048780487805, "train_speed(iter/s)": 1.457233 }, { "epoch": 3.811533353326764, "grad_norm": 4.165044784545898, "learning_rate": 1.3304495188541283e-05, "loss": 0.3203861713409424, "memory(GiB)": 73.83, "step": 88965, "token_acc": 0.9363636363636364, "train_speed(iter/s)": 1.457228 }, { "epoch": 3.811747568656013, "grad_norm": 5.877482891082764, "learning_rate": 1.3299924345347237e-05, "loss": 0.4524073600769043, "memory(GiB)": 73.83, "step": 88970, "token_acc": 0.918918918918919, "train_speed(iter/s)": 1.457233 }, { "epoch": 3.8119617839852618, "grad_norm": 6.697021484375, "learning_rate": 1.3295354167009489e-05, "loss": 0.5628479957580567, "memory(GiB)": 73.83, "step": 88975, "token_acc": 0.8909090909090909, "train_speed(iter/s)": 1.457243 }, { "epoch": 3.812175999314511, "grad_norm": 0.12722107768058777, "learning_rate": 1.3290784653610817e-05, "loss": 0.5682216167449952, "memory(GiB)": 73.83, "step": 88980, "token_acc": 0.8885714285714286, "train_speed(iter/s)": 1.457238 }, { "epoch": 3.81239021464376, "grad_norm": 2.2524654865264893, "learning_rate": 1.328621580523403e-05, "loss": 0.25486633777618406, "memory(GiB)": 73.83, "step": 88985, "token_acc": 0.9315068493150684, "train_speed(iter/s)": 1.457247 }, { "epoch": 3.8126044299730086, "grad_norm": 2.8049750328063965, "learning_rate": 1.3281647621961885e-05, "loss": 0.26191470623016355, "memory(GiB)": 73.83, "step": 88990, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.457252 }, { "epoch": 3.812818645302258, "grad_norm": 2.0797922611236572, "learning_rate": 1.3277080103877138e-05, "loss": 0.3286327123641968, "memory(GiB)": 73.83, "step": 88995, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.457266 }, { "epoch": 3.8130328606315067, "grad_norm": 0.7407163381576538, "learning_rate": 1.3272513251062535e-05, "loss": 0.19406720399856567, "memory(GiB)": 73.83, "step": 89000, "token_acc": 0.9586466165413534, "train_speed(iter/s)": 1.45727 }, { "epoch": 3.8130328606315067, "eval_loss": 2.250037431716919, "eval_runtime": 11.7355, "eval_samples_per_second": 8.521, "eval_steps_per_second": 8.521, "eval_token_acc": 0.4807692307692308, "step": 89000 }, { "epoch": 3.8132470759607555, "grad_norm": 3.7816162109375, "learning_rate": 1.3267947063600794e-05, "loss": 0.38748555183410643, "memory(GiB)": 73.83, "step": 89005, "token_acc": 0.6071428571428571, "train_speed(iter/s)": 1.456978 }, { "epoch": 3.8134612912900048, "grad_norm": 2.1009175777435303, "learning_rate": 1.3263381541574659e-05, "loss": 0.27103486061096194, "memory(GiB)": 73.83, "step": 89010, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.456985 }, { "epoch": 3.8136755066192536, "grad_norm": 1.385169506072998, "learning_rate": 1.3258816685066832e-05, "loss": 0.669259786605835, "memory(GiB)": 73.83, "step": 89015, "token_acc": 0.8910256410256411, "train_speed(iter/s)": 1.457004 }, { "epoch": 3.8138897219485024, "grad_norm": 1.638204574584961, "learning_rate": 1.3254252494160007e-05, "loss": 0.21292293071746826, "memory(GiB)": 73.83, "step": 89020, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.457006 }, { "epoch": 3.8141039372777517, "grad_norm": 4.090872287750244, "learning_rate": 1.3249688968936869e-05, "loss": 0.38696064949035647, "memory(GiB)": 73.83, "step": 89025, "token_acc": 0.9026845637583892, "train_speed(iter/s)": 1.457008 }, { "epoch": 3.8143181526070005, "grad_norm": 4.3688201904296875, "learning_rate": 1.3245126109480076e-05, "loss": 0.4872419834136963, "memory(GiB)": 73.83, "step": 89030, "token_acc": 0.9029411764705882, "train_speed(iter/s)": 1.457009 }, { "epoch": 3.8145323679362493, "grad_norm": 1.3836475610733032, "learning_rate": 1.3240563915872316e-05, "loss": 0.28295533657073973, "memory(GiB)": 73.83, "step": 89035, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.457013 }, { "epoch": 3.8147465832654985, "grad_norm": 3.51057767868042, "learning_rate": 1.3236002388196206e-05, "loss": 0.24318327903747558, "memory(GiB)": 73.83, "step": 89040, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.457012 }, { "epoch": 3.8149607985947473, "grad_norm": 2.2884654998779297, "learning_rate": 1.3231441526534422e-05, "loss": 0.4687757968902588, "memory(GiB)": 73.83, "step": 89045, "token_acc": 0.9094650205761317, "train_speed(iter/s)": 1.457014 }, { "epoch": 3.815175013923996, "grad_norm": 3.6652345657348633, "learning_rate": 1.3226881330969559e-05, "loss": 0.442633056640625, "memory(GiB)": 73.83, "step": 89050, "token_acc": 0.9003831417624522, "train_speed(iter/s)": 1.457027 }, { "epoch": 3.8153892292532454, "grad_norm": 4.668616771697998, "learning_rate": 1.3222321801584242e-05, "loss": 0.32612946033477785, "memory(GiB)": 73.83, "step": 89055, "token_acc": 0.9057377049180327, "train_speed(iter/s)": 1.457038 }, { "epoch": 3.8156034445824942, "grad_norm": 2.317929983139038, "learning_rate": 1.3217762938461065e-05, "loss": 0.47237472534179686, "memory(GiB)": 73.83, "step": 89060, "token_acc": 0.8827586206896552, "train_speed(iter/s)": 1.45705 }, { "epoch": 3.815817659911743, "grad_norm": 2.8983778953552246, "learning_rate": 1.32132047416826e-05, "loss": 0.31597537994384767, "memory(GiB)": 73.83, "step": 89065, "token_acc": 0.952, "train_speed(iter/s)": 1.45705 }, { "epoch": 3.8160318752409923, "grad_norm": 0.9461668133735657, "learning_rate": 1.3208647211331454e-05, "loss": 0.5149258136749267, "memory(GiB)": 73.83, "step": 89070, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.457054 }, { "epoch": 3.816246090570241, "grad_norm": 3.0785584449768066, "learning_rate": 1.3204090347490178e-05, "loss": 0.3748481273651123, "memory(GiB)": 73.83, "step": 89075, "token_acc": 0.9245901639344263, "train_speed(iter/s)": 1.457053 }, { "epoch": 3.81646030589949, "grad_norm": 0.31157186627388, "learning_rate": 1.3199534150241321e-05, "loss": 0.3090924501419067, "memory(GiB)": 73.83, "step": 89080, "token_acc": 0.9494584837545126, "train_speed(iter/s)": 1.457058 }, { "epoch": 3.816674521228739, "grad_norm": 0.7575863599777222, "learning_rate": 1.319497861966743e-05, "loss": 0.3888145685195923, "memory(GiB)": 73.83, "step": 89085, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.457074 }, { "epoch": 3.816888736557988, "grad_norm": 3.702726364135742, "learning_rate": 1.3190423755851005e-05, "loss": 0.40952210426330565, "memory(GiB)": 73.83, "step": 89090, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.457075 }, { "epoch": 3.817102951887237, "grad_norm": 1.4493169784545898, "learning_rate": 1.3185869558874602e-05, "loss": 0.3389871597290039, "memory(GiB)": 73.83, "step": 89095, "token_acc": 0.9422492401215805, "train_speed(iter/s)": 1.457074 }, { "epoch": 3.817317167216486, "grad_norm": 8.771236419677734, "learning_rate": 1.3181316028820695e-05, "loss": 0.6651852607727051, "memory(GiB)": 73.83, "step": 89100, "token_acc": 0.8666666666666667, "train_speed(iter/s)": 1.457074 }, { "epoch": 3.817531382545735, "grad_norm": 3.491645574569702, "learning_rate": 1.31767631657718e-05, "loss": 0.5450372695922852, "memory(GiB)": 73.83, "step": 89105, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.457096 }, { "epoch": 3.8177455978749837, "grad_norm": 2.1068716049194336, "learning_rate": 1.3172210969810378e-05, "loss": 0.4559269905090332, "memory(GiB)": 73.83, "step": 89110, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.457101 }, { "epoch": 3.817959813204233, "grad_norm": 3.072139263153076, "learning_rate": 1.3167659441018909e-05, "loss": 0.41988444328308105, "memory(GiB)": 73.83, "step": 89115, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.457107 }, { "epoch": 3.8181740285334818, "grad_norm": 2.357112169265747, "learning_rate": 1.3163108579479838e-05, "loss": 0.2728205919265747, "memory(GiB)": 73.83, "step": 89120, "token_acc": 0.9315068493150684, "train_speed(iter/s)": 1.457108 }, { "epoch": 3.8183882438627306, "grad_norm": 3.6715071201324463, "learning_rate": 1.3158558385275594e-05, "loss": 0.37722508907318114, "memory(GiB)": 73.83, "step": 89125, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.457115 }, { "epoch": 3.81860245919198, "grad_norm": 3.2456910610198975, "learning_rate": 1.3154008858488643e-05, "loss": 0.0854350209236145, "memory(GiB)": 73.83, "step": 89130, "token_acc": 0.9779179810725552, "train_speed(iter/s)": 1.457121 }, { "epoch": 3.8188166745212286, "grad_norm": 6.080156326293945, "learning_rate": 1.314945999920138e-05, "loss": 0.4016472339630127, "memory(GiB)": 73.83, "step": 89135, "token_acc": 0.9096774193548387, "train_speed(iter/s)": 1.457123 }, { "epoch": 3.8190308898504775, "grad_norm": 1.558573603630066, "learning_rate": 1.3144911807496224e-05, "loss": 0.18009812831878663, "memory(GiB)": 73.83, "step": 89140, "token_acc": 0.9556451612903226, "train_speed(iter/s)": 1.457136 }, { "epoch": 3.8192451051797267, "grad_norm": 4.053117752075195, "learning_rate": 1.3140364283455558e-05, "loss": 0.26087982654571534, "memory(GiB)": 73.83, "step": 89145, "token_acc": 0.948948948948949, "train_speed(iter/s)": 1.457136 }, { "epoch": 3.8194593205089755, "grad_norm": 3.0979626178741455, "learning_rate": 1.3135817427161762e-05, "loss": 0.29939894676208495, "memory(GiB)": 73.83, "step": 89150, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.457139 }, { "epoch": 3.8196735358382243, "grad_norm": 2.464858293533325, "learning_rate": 1.3131271238697223e-05, "loss": 0.38623623847961425, "memory(GiB)": 73.83, "step": 89155, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.457142 }, { "epoch": 3.8198877511674736, "grad_norm": 1.6810930967330933, "learning_rate": 1.3126725718144295e-05, "loss": 0.22565348148345948, "memory(GiB)": 73.83, "step": 89160, "token_acc": 0.9373040752351097, "train_speed(iter/s)": 1.457147 }, { "epoch": 3.8201019664967224, "grad_norm": 4.196789264678955, "learning_rate": 1.312218086558532e-05, "loss": 0.27126944065093994, "memory(GiB)": 73.83, "step": 89165, "token_acc": 0.9222614840989399, "train_speed(iter/s)": 1.457151 }, { "epoch": 3.820316181825971, "grad_norm": 5.363186836242676, "learning_rate": 1.3117636681102635e-05, "loss": 0.3128215789794922, "memory(GiB)": 73.83, "step": 89170, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.45716 }, { "epoch": 3.8205303971552205, "grad_norm": 8.287461280822754, "learning_rate": 1.311309316477854e-05, "loss": 0.3651684284210205, "memory(GiB)": 73.83, "step": 89175, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.457163 }, { "epoch": 3.8207446124844693, "grad_norm": 1.8943554162979126, "learning_rate": 1.3108550316695383e-05, "loss": 0.3448534727096558, "memory(GiB)": 73.83, "step": 89180, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.457166 }, { "epoch": 3.820958827813718, "grad_norm": 2.913262128829956, "learning_rate": 1.3104008136935431e-05, "loss": 0.2641591548919678, "memory(GiB)": 73.83, "step": 89185, "token_acc": 0.928, "train_speed(iter/s)": 1.457166 }, { "epoch": 3.8211730431429674, "grad_norm": 2.6010143756866455, "learning_rate": 1.3099466625580992e-05, "loss": 0.26800026893615725, "memory(GiB)": 73.83, "step": 89190, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.457169 }, { "epoch": 3.821387258472216, "grad_norm": 0.1260673701763153, "learning_rate": 1.3094925782714335e-05, "loss": 0.2829886913299561, "memory(GiB)": 73.83, "step": 89195, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.457173 }, { "epoch": 3.821601473801465, "grad_norm": 2.617450475692749, "learning_rate": 1.3090385608417716e-05, "loss": 0.40515689849853515, "memory(GiB)": 73.83, "step": 89200, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.457177 }, { "epoch": 3.8218156891307142, "grad_norm": 2.763673782348633, "learning_rate": 1.3085846102773386e-05, "loss": 0.38201351165771485, "memory(GiB)": 73.83, "step": 89205, "token_acc": 0.9153094462540716, "train_speed(iter/s)": 1.457179 }, { "epoch": 3.822029904459963, "grad_norm": 0.609691858291626, "learning_rate": 1.3081307265863568e-05, "loss": 0.2393643856048584, "memory(GiB)": 73.83, "step": 89210, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.457179 }, { "epoch": 3.822244119789212, "grad_norm": 2.9034485816955566, "learning_rate": 1.3076769097770514e-05, "loss": 0.518011999130249, "memory(GiB)": 73.83, "step": 89215, "token_acc": 0.9029411764705882, "train_speed(iter/s)": 1.457186 }, { "epoch": 3.822458335118461, "grad_norm": 2.9681813716888428, "learning_rate": 1.3072231598576423e-05, "loss": 0.5492692947387695, "memory(GiB)": 73.83, "step": 89220, "token_acc": 0.9013605442176871, "train_speed(iter/s)": 1.457185 }, { "epoch": 3.82267255044771, "grad_norm": 1.4662150144577026, "learning_rate": 1.3067694768363502e-05, "loss": 0.28319127559661866, "memory(GiB)": 73.83, "step": 89225, "token_acc": 0.9477124183006536, "train_speed(iter/s)": 1.457187 }, { "epoch": 3.8228867657769587, "grad_norm": 3.4911723136901855, "learning_rate": 1.3063158607213938e-05, "loss": 0.33071844577789306, "memory(GiB)": 73.83, "step": 89230, "token_acc": 0.9264214046822743, "train_speed(iter/s)": 1.457186 }, { "epoch": 3.823100981106208, "grad_norm": 5.073183059692383, "learning_rate": 1.3058623115209884e-05, "loss": 0.6934484481811524, "memory(GiB)": 73.83, "step": 89235, "token_acc": 0.8653198653198653, "train_speed(iter/s)": 1.457195 }, { "epoch": 3.823315196435457, "grad_norm": 3.4178853034973145, "learning_rate": 1.3054088292433542e-05, "loss": 0.4502855777740479, "memory(GiB)": 73.83, "step": 89240, "token_acc": 0.8839285714285714, "train_speed(iter/s)": 1.457197 }, { "epoch": 3.8235294117647056, "grad_norm": 4.59379768371582, "learning_rate": 1.3049554138967051e-05, "loss": 0.6210983276367188, "memory(GiB)": 73.83, "step": 89245, "token_acc": 0.8709677419354839, "train_speed(iter/s)": 1.4572 }, { "epoch": 3.823743627093955, "grad_norm": 4.176398277282715, "learning_rate": 1.3045020654892537e-05, "loss": 0.32607581615448, "memory(GiB)": 73.83, "step": 89250, "token_acc": 0.9013157894736842, "train_speed(iter/s)": 1.457213 }, { "epoch": 3.8239578424232037, "grad_norm": 0.5092312097549438, "learning_rate": 1.3040487840292148e-05, "loss": 0.4953601360321045, "memory(GiB)": 73.83, "step": 89255, "token_acc": 0.917981072555205, "train_speed(iter/s)": 1.457216 }, { "epoch": 3.8241720577524525, "grad_norm": 0.2534432113170624, "learning_rate": 1.3035955695247998e-05, "loss": 0.2859842777252197, "memory(GiB)": 73.83, "step": 89260, "token_acc": 0.9331103678929766, "train_speed(iter/s)": 1.457217 }, { "epoch": 3.8243862730817018, "grad_norm": 1.7772616147994995, "learning_rate": 1.3031424219842186e-05, "loss": 0.30834777355194093, "memory(GiB)": 73.83, "step": 89265, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.457219 }, { "epoch": 3.8246004884109506, "grad_norm": 2.5155391693115234, "learning_rate": 1.3026893414156783e-05, "loss": 0.32396886348724363, "memory(GiB)": 73.83, "step": 89270, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.45722 }, { "epoch": 3.8248147037401994, "grad_norm": 3.467865228652954, "learning_rate": 1.3022363278273908e-05, "loss": 0.31321940422058103, "memory(GiB)": 73.83, "step": 89275, "token_acc": 0.9246987951807228, "train_speed(iter/s)": 1.457236 }, { "epoch": 3.8250289190694486, "grad_norm": 6.733109951019287, "learning_rate": 1.3017833812275609e-05, "loss": 0.5125834465026855, "memory(GiB)": 73.83, "step": 89280, "token_acc": 0.8782287822878229, "train_speed(iter/s)": 1.457242 }, { "epoch": 3.8252431343986975, "grad_norm": 1.9977303743362427, "learning_rate": 1.3013305016243949e-05, "loss": 0.5859471797943115, "memory(GiB)": 73.83, "step": 89285, "token_acc": 0.9158576051779935, "train_speed(iter/s)": 1.457252 }, { "epoch": 3.8254573497279463, "grad_norm": 3.511303186416626, "learning_rate": 1.300877689026096e-05, "loss": 0.175292706489563, "memory(GiB)": 73.83, "step": 89290, "token_acc": 0.952755905511811, "train_speed(iter/s)": 1.45726 }, { "epoch": 3.8256715650571955, "grad_norm": 4.128093719482422, "learning_rate": 1.300424943440866e-05, "loss": 0.2367017984390259, "memory(GiB)": 73.83, "step": 89295, "token_acc": 0.9367088607594937, "train_speed(iter/s)": 1.457262 }, { "epoch": 3.8258857803864443, "grad_norm": 3.275277853012085, "learning_rate": 1.2999722648769103e-05, "loss": 0.4583712577819824, "memory(GiB)": 73.83, "step": 89300, "token_acc": 0.9208333333333333, "train_speed(iter/s)": 1.457268 }, { "epoch": 3.826099995715693, "grad_norm": 2.284900188446045, "learning_rate": 1.299519653342428e-05, "loss": 0.1951335310935974, "memory(GiB)": 73.83, "step": 89305, "token_acc": 0.9584775086505191, "train_speed(iter/s)": 1.457268 }, { "epoch": 3.8263142110449424, "grad_norm": 6.079990386962891, "learning_rate": 1.2990671088456186e-05, "loss": 0.32642784118652346, "memory(GiB)": 73.83, "step": 89310, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.45727 }, { "epoch": 3.826528426374191, "grad_norm": 1.8517518043518066, "learning_rate": 1.2986146313946807e-05, "loss": 0.4126152038574219, "memory(GiB)": 73.83, "step": 89315, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.457279 }, { "epoch": 3.82674264170344, "grad_norm": 3.0715341567993164, "learning_rate": 1.298162220997809e-05, "loss": 0.1662973642349243, "memory(GiB)": 73.83, "step": 89320, "token_acc": 0.9536423841059603, "train_speed(iter/s)": 1.457278 }, { "epoch": 3.8269568570326893, "grad_norm": 1.126139521598816, "learning_rate": 1.2977098776632023e-05, "loss": 0.25049331188201907, "memory(GiB)": 73.83, "step": 89325, "token_acc": 0.9630681818181818, "train_speed(iter/s)": 1.457279 }, { "epoch": 3.827171072361938, "grad_norm": 2.728590726852417, "learning_rate": 1.2972576013990534e-05, "loss": 0.23315703868865967, "memory(GiB)": 73.83, "step": 89330, "token_acc": 0.9515570934256056, "train_speed(iter/s)": 1.457279 }, { "epoch": 3.827385287691187, "grad_norm": 2.963805675506592, "learning_rate": 1.2968053922135576e-05, "loss": 0.4190267562866211, "memory(GiB)": 73.83, "step": 89335, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.457275 }, { "epoch": 3.827599503020436, "grad_norm": 1.7101678848266602, "learning_rate": 1.2963532501149062e-05, "loss": 0.36665635108947753, "memory(GiB)": 73.83, "step": 89340, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.457281 }, { "epoch": 3.827813718349685, "grad_norm": 0.3867921233177185, "learning_rate": 1.2959011751112898e-05, "loss": 0.10416609048843384, "memory(GiB)": 73.83, "step": 89345, "token_acc": 0.9711538461538461, "train_speed(iter/s)": 1.457297 }, { "epoch": 3.828027933678934, "grad_norm": 1.4983128309249878, "learning_rate": 1.2954491672108987e-05, "loss": 0.19866631031036378, "memory(GiB)": 73.83, "step": 89350, "token_acc": 0.9635761589403974, "train_speed(iter/s)": 1.457301 }, { "epoch": 3.828242149008183, "grad_norm": 4.405649185180664, "learning_rate": 1.294997226421919e-05, "loss": 0.7467461585998535, "memory(GiB)": 73.83, "step": 89355, "token_acc": 0.827922077922078, "train_speed(iter/s)": 1.457301 }, { "epoch": 3.828456364337432, "grad_norm": 4.933469772338867, "learning_rate": 1.294545352752542e-05, "loss": 0.3518479347229004, "memory(GiB)": 73.83, "step": 89360, "token_acc": 0.9139072847682119, "train_speed(iter/s)": 1.457302 }, { "epoch": 3.8286705796666807, "grad_norm": 4.621524810791016, "learning_rate": 1.2940935462109522e-05, "loss": 0.32778158187866213, "memory(GiB)": 73.83, "step": 89365, "token_acc": 0.9352941176470588, "train_speed(iter/s)": 1.457306 }, { "epoch": 3.82888479499593, "grad_norm": 3.836827278137207, "learning_rate": 1.2936418068053335e-05, "loss": 0.6267382621765136, "memory(GiB)": 73.83, "step": 89370, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.457309 }, { "epoch": 3.8290990103251787, "grad_norm": 2.2992801666259766, "learning_rate": 1.2931901345438708e-05, "loss": 0.4928610801696777, "memory(GiB)": 73.83, "step": 89375, "token_acc": 0.8984615384615384, "train_speed(iter/s)": 1.457323 }, { "epoch": 3.8293132256544276, "grad_norm": 4.265336990356445, "learning_rate": 1.2927385294347444e-05, "loss": 0.4345077991485596, "memory(GiB)": 73.83, "step": 89380, "token_acc": 0.9129129129129129, "train_speed(iter/s)": 1.457327 }, { "epoch": 3.829527440983677, "grad_norm": 3.3082540035247803, "learning_rate": 1.292286991486139e-05, "loss": 0.33842039108276367, "memory(GiB)": 73.83, "step": 89385, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.45733 }, { "epoch": 3.8297416563129256, "grad_norm": 3.8641653060913086, "learning_rate": 1.2918355207062327e-05, "loss": 0.5227192401885986, "memory(GiB)": 73.83, "step": 89390, "token_acc": 0.9097222222222222, "train_speed(iter/s)": 1.457328 }, { "epoch": 3.829955871642175, "grad_norm": 2.1945009231567383, "learning_rate": 1.2913841171032033e-05, "loss": 0.34328086376190187, "memory(GiB)": 73.83, "step": 89395, "token_acc": 0.9273356401384083, "train_speed(iter/s)": 1.457331 }, { "epoch": 3.8301700869714237, "grad_norm": 2.546400547027588, "learning_rate": 1.2909327806852305e-05, "loss": 0.8315006256103515, "memory(GiB)": 73.83, "step": 89400, "token_acc": 0.8269230769230769, "train_speed(iter/s)": 1.457339 }, { "epoch": 3.8303843023006725, "grad_norm": 5.322629451751709, "learning_rate": 1.2904815114604901e-05, "loss": 0.46747546195983886, "memory(GiB)": 73.83, "step": 89405, "token_acc": 0.908284023668639, "train_speed(iter/s)": 1.457342 }, { "epoch": 3.8305985176299218, "grad_norm": 3.5516486167907715, "learning_rate": 1.2900303094371568e-05, "loss": 0.8484651565551757, "memory(GiB)": 73.83, "step": 89410, "token_acc": 0.8047945205479452, "train_speed(iter/s)": 1.457341 }, { "epoch": 3.8308127329591706, "grad_norm": 4.358623504638672, "learning_rate": 1.2895791746234032e-05, "loss": 0.3979602098464966, "memory(GiB)": 73.83, "step": 89415, "token_acc": 0.919732441471572, "train_speed(iter/s)": 1.457343 }, { "epoch": 3.8310269482884194, "grad_norm": 3.762190818786621, "learning_rate": 1.2891281070274052e-05, "loss": 0.2910939693450928, "memory(GiB)": 73.83, "step": 89420, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.45734 }, { "epoch": 3.8312411636176686, "grad_norm": 0.9614093899726868, "learning_rate": 1.2886771066573323e-05, "loss": 0.19378029108047484, "memory(GiB)": 73.83, "step": 89425, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.457346 }, { "epoch": 3.8314553789469175, "grad_norm": 2.6606173515319824, "learning_rate": 1.2882261735213552e-05, "loss": 0.25302674770355227, "memory(GiB)": 73.83, "step": 89430, "token_acc": 0.954248366013072, "train_speed(iter/s)": 1.457348 }, { "epoch": 3.8316695942761663, "grad_norm": 0.15792740881443024, "learning_rate": 1.2877753076276434e-05, "loss": 0.44028172492980955, "memory(GiB)": 73.83, "step": 89435, "token_acc": 0.9068825910931174, "train_speed(iter/s)": 1.457353 }, { "epoch": 3.8318838096054155, "grad_norm": 2.6175031661987305, "learning_rate": 1.2873245089843621e-05, "loss": 0.19942970275878907, "memory(GiB)": 73.83, "step": 89440, "token_acc": 0.9671532846715328, "train_speed(iter/s)": 1.457349 }, { "epoch": 3.8320980249346643, "grad_norm": 4.621310710906982, "learning_rate": 1.286873777599682e-05, "loss": 0.4914806842803955, "memory(GiB)": 73.83, "step": 89445, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.45735 }, { "epoch": 3.832312240263913, "grad_norm": 3.3213818073272705, "learning_rate": 1.2864231134817661e-05, "loss": 0.4458664894104004, "memory(GiB)": 73.83, "step": 89450, "token_acc": 0.9087837837837838, "train_speed(iter/s)": 1.457353 }, { "epoch": 3.8325264555931624, "grad_norm": 6.759099006652832, "learning_rate": 1.2859725166387798e-05, "loss": 0.35695490837097166, "memory(GiB)": 73.83, "step": 89455, "token_acc": 0.9212121212121213, "train_speed(iter/s)": 1.457356 }, { "epoch": 3.832740670922411, "grad_norm": 0.2380647212266922, "learning_rate": 1.2855219870788849e-05, "loss": 0.16485223770141602, "memory(GiB)": 73.83, "step": 89460, "token_acc": 0.9682539682539683, "train_speed(iter/s)": 1.457363 }, { "epoch": 3.83295488625166, "grad_norm": 2.0765609741210938, "learning_rate": 1.2850715248102425e-05, "loss": 0.27645819187164306, "memory(GiB)": 73.83, "step": 89465, "token_acc": 0.9516616314199395, "train_speed(iter/s)": 1.457369 }, { "epoch": 3.8331691015809093, "grad_norm": 3.9927217960357666, "learning_rate": 1.2846211298410155e-05, "loss": 0.4372418403625488, "memory(GiB)": 73.83, "step": 89470, "token_acc": 0.9205776173285198, "train_speed(iter/s)": 1.457374 }, { "epoch": 3.833383316910158, "grad_norm": 2.753727912902832, "learning_rate": 1.2841708021793608e-05, "loss": 0.3481900691986084, "memory(GiB)": 73.83, "step": 89475, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.457381 }, { "epoch": 3.833597532239407, "grad_norm": 2.9406144618988037, "learning_rate": 1.2837205418334392e-05, "loss": 0.24243764877319335, "memory(GiB)": 73.83, "step": 89480, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.457397 }, { "epoch": 3.833811747568656, "grad_norm": 3.204026222229004, "learning_rate": 1.2832703488114055e-05, "loss": 0.269146466255188, "memory(GiB)": 73.83, "step": 89485, "token_acc": 0.9461279461279462, "train_speed(iter/s)": 1.457398 }, { "epoch": 3.834025962897905, "grad_norm": 3.1283466815948486, "learning_rate": 1.2828202231214171e-05, "loss": 0.251559853553772, "memory(GiB)": 73.83, "step": 89490, "token_acc": 0.943217665615142, "train_speed(iter/s)": 1.457398 }, { "epoch": 3.834240178227154, "grad_norm": 4.645019054412842, "learning_rate": 1.2823701647716269e-05, "loss": 0.2473212480545044, "memory(GiB)": 73.83, "step": 89495, "token_acc": 0.966996699669967, "train_speed(iter/s)": 1.4574 }, { "epoch": 3.834454393556403, "grad_norm": 0.5934627056121826, "learning_rate": 1.281920173770187e-05, "loss": 0.5117452144622803, "memory(GiB)": 73.83, "step": 89500, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.457403 }, { "epoch": 3.834454393556403, "eval_loss": 2.394327163696289, "eval_runtime": 11.1893, "eval_samples_per_second": 8.937, "eval_steps_per_second": 8.937, "eval_token_acc": 0.4522875816993464, "step": 89500 }, { "epoch": 3.834668608885652, "grad_norm": 1.673627257347107, "learning_rate": 1.2814702501252524e-05, "loss": 0.26978416442871095, "memory(GiB)": 73.83, "step": 89505, "token_acc": 0.5807984790874525, "train_speed(iter/s)": 1.457118 }, { "epoch": 3.8348828242149007, "grad_norm": 6.506279945373535, "learning_rate": 1.2810203938449721e-05, "loss": 0.4459816932678223, "memory(GiB)": 73.83, "step": 89510, "token_acc": 0.884375, "train_speed(iter/s)": 1.457121 }, { "epoch": 3.83509703954415, "grad_norm": 3.3007731437683105, "learning_rate": 1.2805706049374967e-05, "loss": 0.5556180000305175, "memory(GiB)": 73.83, "step": 89515, "token_acc": 0.8885793871866295, "train_speed(iter/s)": 1.457127 }, { "epoch": 3.8353112548733987, "grad_norm": 0.39006999135017395, "learning_rate": 1.280120883410974e-05, "loss": 0.27123658657073973, "memory(GiB)": 73.83, "step": 89520, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.457135 }, { "epoch": 3.8355254702026476, "grad_norm": 0.416657030582428, "learning_rate": 1.2796712292735491e-05, "loss": 0.23994085788726807, "memory(GiB)": 73.83, "step": 89525, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.457142 }, { "epoch": 3.835739685531897, "grad_norm": 1.410352110862732, "learning_rate": 1.2792216425333715e-05, "loss": 0.2811130523681641, "memory(GiB)": 73.83, "step": 89530, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.457149 }, { "epoch": 3.8359539008611456, "grad_norm": 2.6053686141967773, "learning_rate": 1.2787721231985845e-05, "loss": 0.2443610191345215, "memory(GiB)": 73.83, "step": 89535, "token_acc": 0.9363057324840764, "train_speed(iter/s)": 1.457152 }, { "epoch": 3.8361681161903944, "grad_norm": 3.5066258907318115, "learning_rate": 1.2783226712773305e-05, "loss": 0.2823799133300781, "memory(GiB)": 73.83, "step": 89540, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.457161 }, { "epoch": 3.8363823315196437, "grad_norm": 0.8300315737724304, "learning_rate": 1.2778732867777515e-05, "loss": 0.2661010980606079, "memory(GiB)": 73.83, "step": 89545, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.457168 }, { "epoch": 3.8365965468488925, "grad_norm": 5.818176746368408, "learning_rate": 1.2774239697079915e-05, "loss": 0.6362083435058594, "memory(GiB)": 73.83, "step": 89550, "token_acc": 0.823170731707317, "train_speed(iter/s)": 1.45717 }, { "epoch": 3.8368107621781413, "grad_norm": 0.7993121147155762, "learning_rate": 1.2769747200761878e-05, "loss": 0.21700546741485596, "memory(GiB)": 73.83, "step": 89555, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.457176 }, { "epoch": 3.8370249775073906, "grad_norm": 4.048269748687744, "learning_rate": 1.2765255378904778e-05, "loss": 0.2583456993103027, "memory(GiB)": 73.83, "step": 89560, "token_acc": 0.9415204678362573, "train_speed(iter/s)": 1.457178 }, { "epoch": 3.8372391928366394, "grad_norm": 2.287994623184204, "learning_rate": 1.2760764231590027e-05, "loss": 0.15938153266906738, "memory(GiB)": 73.83, "step": 89565, "token_acc": 0.9735973597359736, "train_speed(iter/s)": 1.45718 }, { "epoch": 3.837453408165888, "grad_norm": 3.525359630584717, "learning_rate": 1.2756273758898962e-05, "loss": 0.5944365501403809, "memory(GiB)": 73.83, "step": 89570, "token_acc": 0.8725868725868726, "train_speed(iter/s)": 1.457188 }, { "epoch": 3.8376676234951375, "grad_norm": 2.4666686058044434, "learning_rate": 1.2751783960912933e-05, "loss": 0.31646368503570554, "memory(GiB)": 73.83, "step": 89575, "token_acc": 0.9227941176470589, "train_speed(iter/s)": 1.457187 }, { "epoch": 3.8378818388243863, "grad_norm": 1.5351396799087524, "learning_rate": 1.2747294837713285e-05, "loss": 0.6580258369445801, "memory(GiB)": 73.83, "step": 89580, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.457197 }, { "epoch": 3.838096054153635, "grad_norm": 3.2422242164611816, "learning_rate": 1.2742806389381318e-05, "loss": 0.37071616649627687, "memory(GiB)": 73.83, "step": 89585, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.457201 }, { "epoch": 3.8383102694828843, "grad_norm": 2.4957330226898193, "learning_rate": 1.2738318615998379e-05, "loss": 0.5837108135223389, "memory(GiB)": 73.83, "step": 89590, "token_acc": 0.8880208333333334, "train_speed(iter/s)": 1.45722 }, { "epoch": 3.838524484812133, "grad_norm": 2.7270233631134033, "learning_rate": 1.273383151764575e-05, "loss": 0.44471397399902346, "memory(GiB)": 73.83, "step": 89595, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.457222 }, { "epoch": 3.838738700141382, "grad_norm": 4.087234020233154, "learning_rate": 1.2729345094404727e-05, "loss": 0.4069051265716553, "memory(GiB)": 73.83, "step": 89600, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.457219 }, { "epoch": 3.838952915470631, "grad_norm": 0.26143571734428406, "learning_rate": 1.2724859346356577e-05, "loss": 0.2635300636291504, "memory(GiB)": 73.83, "step": 89605, "token_acc": 0.9448529411764706, "train_speed(iter/s)": 1.457219 }, { "epoch": 3.83916713079988, "grad_norm": 2.8486554622650146, "learning_rate": 1.2720374273582553e-05, "loss": 0.32081735134124756, "memory(GiB)": 73.83, "step": 89610, "token_acc": 0.9539473684210527, "train_speed(iter/s)": 1.457224 }, { "epoch": 3.839381346129129, "grad_norm": 0.05879000946879387, "learning_rate": 1.2715889876163933e-05, "loss": 0.30605862140655515, "memory(GiB)": 73.83, "step": 89615, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.457242 }, { "epoch": 3.839595561458378, "grad_norm": 2.4468789100646973, "learning_rate": 1.2711406154181933e-05, "loss": 0.3797661304473877, "memory(GiB)": 73.83, "step": 89620, "token_acc": 0.9153846153846154, "train_speed(iter/s)": 1.457251 }, { "epoch": 3.839809776787627, "grad_norm": 1.7187321186065674, "learning_rate": 1.27069231077178e-05, "loss": 0.375709080696106, "memory(GiB)": 73.83, "step": 89625, "token_acc": 0.9440993788819876, "train_speed(iter/s)": 1.457251 }, { "epoch": 3.8400239921168757, "grad_norm": 4.413370609283447, "learning_rate": 1.270244073685274e-05, "loss": 0.34961373805999757, "memory(GiB)": 73.83, "step": 89630, "token_acc": 0.9147727272727273, "train_speed(iter/s)": 1.457254 }, { "epoch": 3.840238207446125, "grad_norm": 0.5263944268226624, "learning_rate": 1.269795904166795e-05, "loss": 0.2742105722427368, "memory(GiB)": 73.83, "step": 89635, "token_acc": 0.9380530973451328, "train_speed(iter/s)": 1.45726 }, { "epoch": 3.840452422775374, "grad_norm": 0.6510595083236694, "learning_rate": 1.2693478022244631e-05, "loss": 0.3288670778274536, "memory(GiB)": 73.83, "step": 89640, "token_acc": 0.9205776173285198, "train_speed(iter/s)": 1.457267 }, { "epoch": 3.8406666381046226, "grad_norm": 4.708770275115967, "learning_rate": 1.268899767866394e-05, "loss": 0.5544778347015381, "memory(GiB)": 73.83, "step": 89645, "token_acc": 0.8653846153846154, "train_speed(iter/s)": 1.457259 }, { "epoch": 3.840880853433872, "grad_norm": 5.914555549621582, "learning_rate": 1.2684518011007062e-05, "loss": 0.6864283084869385, "memory(GiB)": 73.83, "step": 89650, "token_acc": 0.863768115942029, "train_speed(iter/s)": 1.457278 }, { "epoch": 3.8410950687631207, "grad_norm": 4.489391326904297, "learning_rate": 1.2680039019355156e-05, "loss": 0.2920753717422485, "memory(GiB)": 73.83, "step": 89655, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.457294 }, { "epoch": 3.8413092840923695, "grad_norm": 1.480999231338501, "learning_rate": 1.2675560703789347e-05, "loss": 0.3149597406387329, "memory(GiB)": 73.83, "step": 89660, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.457303 }, { "epoch": 3.8415234994216187, "grad_norm": 4.419018745422363, "learning_rate": 1.2671083064390765e-05, "loss": 0.4292929649353027, "memory(GiB)": 73.83, "step": 89665, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.457305 }, { "epoch": 3.8417377147508676, "grad_norm": 1.6658402681350708, "learning_rate": 1.2666606101240525e-05, "loss": 0.38507676124572754, "memory(GiB)": 73.83, "step": 89670, "token_acc": 0.8988439306358381, "train_speed(iter/s)": 1.457309 }, { "epoch": 3.8419519300801164, "grad_norm": 1.477668046951294, "learning_rate": 1.2662129814419742e-05, "loss": 0.34076247215270994, "memory(GiB)": 73.83, "step": 89675, "token_acc": 0.9304635761589404, "train_speed(iter/s)": 1.457312 }, { "epoch": 3.8421661454093656, "grad_norm": 3.00964093208313, "learning_rate": 1.2657654204009512e-05, "loss": 0.22746012210845948, "memory(GiB)": 73.83, "step": 89680, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.457315 }, { "epoch": 3.8423803607386144, "grad_norm": 4.306890964508057, "learning_rate": 1.2653179270090897e-05, "loss": 0.45505437850952146, "memory(GiB)": 73.83, "step": 89685, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.457329 }, { "epoch": 3.8425945760678633, "grad_norm": 2.3975305557250977, "learning_rate": 1.2648705012744983e-05, "loss": 0.4228024482727051, "memory(GiB)": 73.83, "step": 89690, "token_acc": 0.907185628742515, "train_speed(iter/s)": 1.457326 }, { "epoch": 3.8428087913971125, "grad_norm": 2.2259974479675293, "learning_rate": 1.2644231432052794e-05, "loss": 0.3930291414260864, "memory(GiB)": 73.83, "step": 89695, "token_acc": 0.9152542372881356, "train_speed(iter/s)": 1.457336 }, { "epoch": 3.8430230067263613, "grad_norm": 2.675023078918457, "learning_rate": 1.2639758528095414e-05, "loss": 0.5865394115447998, "memory(GiB)": 73.83, "step": 89700, "token_acc": 0.8821548821548821, "train_speed(iter/s)": 1.457345 }, { "epoch": 3.84323722205561, "grad_norm": 3.54189133644104, "learning_rate": 1.2635286300953841e-05, "loss": 0.47478322982788085, "memory(GiB)": 73.83, "step": 89705, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.457349 }, { "epoch": 3.8434514373848594, "grad_norm": 2.6250834465026855, "learning_rate": 1.2630814750709119e-05, "loss": 0.2073514223098755, "memory(GiB)": 73.83, "step": 89710, "token_acc": 0.9572953736654805, "train_speed(iter/s)": 1.457357 }, { "epoch": 3.843665652714108, "grad_norm": 2.004713773727417, "learning_rate": 1.2626343877442242e-05, "loss": 0.20246787071228028, "memory(GiB)": 73.83, "step": 89715, "token_acc": 0.9447852760736196, "train_speed(iter/s)": 1.457363 }, { "epoch": 3.843879868043357, "grad_norm": 3.4132866859436035, "learning_rate": 1.2621873681234208e-05, "loss": 0.33318026065826417, "memory(GiB)": 73.83, "step": 89720, "token_acc": 0.9282868525896414, "train_speed(iter/s)": 1.457363 }, { "epoch": 3.8440940833726063, "grad_norm": 14.460667610168457, "learning_rate": 1.2617404162165997e-05, "loss": 0.4473127365112305, "memory(GiB)": 73.83, "step": 89725, "token_acc": 0.8785942492012779, "train_speed(iter/s)": 1.457377 }, { "epoch": 3.844308298701855, "grad_norm": 2.0186030864715576, "learning_rate": 1.2612935320318558e-05, "loss": 0.40214881896972654, "memory(GiB)": 73.83, "step": 89730, "token_acc": 0.9108527131782945, "train_speed(iter/s)": 1.457389 }, { "epoch": 3.8445225140311043, "grad_norm": 3.3075509071350098, "learning_rate": 1.2608467155772885e-05, "loss": 0.43241395950317385, "memory(GiB)": 73.83, "step": 89735, "token_acc": 0.8803680981595092, "train_speed(iter/s)": 1.457394 }, { "epoch": 3.844736729360353, "grad_norm": 3.651207447052002, "learning_rate": 1.2603999668609906e-05, "loss": 0.5370546340942383, "memory(GiB)": 73.83, "step": 89740, "token_acc": 0.8872727272727273, "train_speed(iter/s)": 1.457395 }, { "epoch": 3.844950944689602, "grad_norm": 2.5115163326263428, "learning_rate": 1.2599532858910556e-05, "loss": 0.26034505367279054, "memory(GiB)": 73.83, "step": 89745, "token_acc": 0.9452554744525548, "train_speed(iter/s)": 1.457398 }, { "epoch": 3.8451651600188512, "grad_norm": 2.875990629196167, "learning_rate": 1.259506672675575e-05, "loss": 0.27001171112060546, "memory(GiB)": 73.83, "step": 89750, "token_acc": 0.9346938775510204, "train_speed(iter/s)": 1.457403 }, { "epoch": 3.8453793753481, "grad_norm": 5.02173376083374, "learning_rate": 1.2590601272226383e-05, "loss": 0.31949167251586913, "memory(GiB)": 73.83, "step": 89755, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.457406 }, { "epoch": 3.845593590677349, "grad_norm": 0.8828774094581604, "learning_rate": 1.2586136495403384e-05, "loss": 0.31624276638031007, "memory(GiB)": 73.83, "step": 89760, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.457408 }, { "epoch": 3.845807806006598, "grad_norm": 3.089514970779419, "learning_rate": 1.25816723963676e-05, "loss": 0.15284936428070067, "memory(GiB)": 73.83, "step": 89765, "token_acc": 0.9609375, "train_speed(iter/s)": 1.45742 }, { "epoch": 3.846022021335847, "grad_norm": 4.875441074371338, "learning_rate": 1.257720897519994e-05, "loss": 0.1267408847808838, "memory(GiB)": 73.83, "step": 89770, "token_acc": 0.9726962457337884, "train_speed(iter/s)": 1.457418 }, { "epoch": 3.8462362366650957, "grad_norm": 3.9166054725646973, "learning_rate": 1.2572746231981247e-05, "loss": 0.26367266178131105, "memory(GiB)": 73.83, "step": 89775, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.457418 }, { "epoch": 3.846450451994345, "grad_norm": 1.8622535467147827, "learning_rate": 1.256828416679236e-05, "loss": 0.20538997650146484, "memory(GiB)": 73.83, "step": 89780, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.457423 }, { "epoch": 3.846664667323594, "grad_norm": 2.2805583477020264, "learning_rate": 1.2563822779714125e-05, "loss": 0.4130858421325684, "memory(GiB)": 73.83, "step": 89785, "token_acc": 0.9141104294478528, "train_speed(iter/s)": 1.457426 }, { "epoch": 3.8468788826528426, "grad_norm": 3.2482235431671143, "learning_rate": 1.2559362070827341e-05, "loss": 0.1562129497528076, "memory(GiB)": 73.83, "step": 89790, "token_acc": 0.959375, "train_speed(iter/s)": 1.457431 }, { "epoch": 3.847093097982092, "grad_norm": 1.6329224109649658, "learning_rate": 1.2554902040212846e-05, "loss": 0.32967779636383054, "memory(GiB)": 73.83, "step": 89795, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.457434 }, { "epoch": 3.8473073133113407, "grad_norm": 2.1721041202545166, "learning_rate": 1.2550442687951431e-05, "loss": 0.22789597511291504, "memory(GiB)": 73.83, "step": 89800, "token_acc": 0.9346938775510204, "train_speed(iter/s)": 1.457437 }, { "epoch": 3.8475215286405895, "grad_norm": 3.152730941772461, "learning_rate": 1.2545984014123879e-05, "loss": 0.42853622436523436, "memory(GiB)": 73.83, "step": 89805, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.457436 }, { "epoch": 3.8477357439698388, "grad_norm": 2.1413075923919678, "learning_rate": 1.2541526018810956e-05, "loss": 0.19325697422027588, "memory(GiB)": 73.83, "step": 89810, "token_acc": 0.9571428571428572, "train_speed(iter/s)": 1.457442 }, { "epoch": 3.8479499592990876, "grad_norm": 2.946376085281372, "learning_rate": 1.253706870209342e-05, "loss": 0.37539119720458985, "memory(GiB)": 73.83, "step": 89815, "token_acc": 0.9149560117302052, "train_speed(iter/s)": 1.457444 }, { "epoch": 3.8481641746283364, "grad_norm": 1.604448676109314, "learning_rate": 1.2532612064052036e-05, "loss": 0.23733949661254883, "memory(GiB)": 73.83, "step": 89820, "token_acc": 0.94, "train_speed(iter/s)": 1.45745 }, { "epoch": 3.8483783899575856, "grad_norm": 4.976438522338867, "learning_rate": 1.2528156104767536e-05, "loss": 0.4919173240661621, "memory(GiB)": 73.83, "step": 89825, "token_acc": 0.8820058997050148, "train_speed(iter/s)": 1.457452 }, { "epoch": 3.8485926052868344, "grad_norm": 2.5945935249328613, "learning_rate": 1.2523700824320638e-05, "loss": 0.2056877851486206, "memory(GiB)": 73.83, "step": 89830, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.457452 }, { "epoch": 3.8488068206160833, "grad_norm": 4.267788410186768, "learning_rate": 1.2519246222792053e-05, "loss": 0.38594608306884765, "memory(GiB)": 73.83, "step": 89835, "token_acc": 0.9161290322580645, "train_speed(iter/s)": 1.457455 }, { "epoch": 3.8490210359453325, "grad_norm": 3.218634843826294, "learning_rate": 1.2514792300262474e-05, "loss": 0.31998701095581056, "memory(GiB)": 73.83, "step": 89840, "token_acc": 0.9323843416370107, "train_speed(iter/s)": 1.457465 }, { "epoch": 3.8492352512745813, "grad_norm": 4.664677619934082, "learning_rate": 1.2510339056812603e-05, "loss": 0.24486105442047118, "memory(GiB)": 73.83, "step": 89845, "token_acc": 0.9457627118644067, "train_speed(iter/s)": 1.457469 }, { "epoch": 3.84944946660383, "grad_norm": 4.773383617401123, "learning_rate": 1.25058864925231e-05, "loss": 0.6946078300476074, "memory(GiB)": 73.83, "step": 89850, "token_acc": 0.863481228668942, "train_speed(iter/s)": 1.457475 }, { "epoch": 3.8496636819330794, "grad_norm": 5.824710845947266, "learning_rate": 1.2501434607474644e-05, "loss": 0.5318926811218262, "memory(GiB)": 73.83, "step": 89855, "token_acc": 0.8955823293172691, "train_speed(iter/s)": 1.457476 }, { "epoch": 3.849877897262328, "grad_norm": 4.021597862243652, "learning_rate": 1.249698340174788e-05, "loss": 0.5093765735626221, "memory(GiB)": 73.83, "step": 89860, "token_acc": 0.8879310344827587, "train_speed(iter/s)": 1.457476 }, { "epoch": 3.850092112591577, "grad_norm": 4.027347087860107, "learning_rate": 1.2492532875423445e-05, "loss": 0.349655818939209, "memory(GiB)": 73.83, "step": 89865, "token_acc": 0.8939929328621908, "train_speed(iter/s)": 1.457478 }, { "epoch": 3.8503063279208263, "grad_norm": 4.20241117477417, "learning_rate": 1.248808302858196e-05, "loss": 0.31057071685791016, "memory(GiB)": 73.83, "step": 89870, "token_acc": 0.9372822299651568, "train_speed(iter/s)": 1.457489 }, { "epoch": 3.850520543250075, "grad_norm": 2.721275568008423, "learning_rate": 1.2483633861304023e-05, "loss": 0.3758589744567871, "memory(GiB)": 73.83, "step": 89875, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.457489 }, { "epoch": 3.850734758579324, "grad_norm": 8.012887954711914, "learning_rate": 1.2479185373670272e-05, "loss": 0.2426370143890381, "memory(GiB)": 73.83, "step": 89880, "token_acc": 0.946843853820598, "train_speed(iter/s)": 1.457507 }, { "epoch": 3.850948973908573, "grad_norm": 5.142782211303711, "learning_rate": 1.2474737565761268e-05, "loss": 0.44075846672058105, "memory(GiB)": 73.83, "step": 89885, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.457518 }, { "epoch": 3.851163189237822, "grad_norm": 2.96393084526062, "learning_rate": 1.24702904376576e-05, "loss": 0.20552680492401124, "memory(GiB)": 73.83, "step": 89890, "token_acc": 0.9621993127147767, "train_speed(iter/s)": 1.457521 }, { "epoch": 3.851377404567071, "grad_norm": 2.5408027172088623, "learning_rate": 1.2465843989439824e-05, "loss": 0.4066408634185791, "memory(GiB)": 73.83, "step": 89895, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.457521 }, { "epoch": 3.85159161989632, "grad_norm": 2.881105899810791, "learning_rate": 1.246139822118848e-05, "loss": 0.26322226524353026, "memory(GiB)": 73.83, "step": 89900, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.457528 }, { "epoch": 3.851805835225569, "grad_norm": 4.234750270843506, "learning_rate": 1.2456953132984133e-05, "loss": 0.28910777568817136, "memory(GiB)": 73.83, "step": 89905, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.457527 }, { "epoch": 3.8520200505548177, "grad_norm": 5.329709529876709, "learning_rate": 1.2452508724907303e-05, "loss": 0.20626609325408934, "memory(GiB)": 73.83, "step": 89910, "token_acc": 0.9631901840490797, "train_speed(iter/s)": 1.45753 }, { "epoch": 3.852234265884067, "grad_norm": 3.0921530723571777, "learning_rate": 1.2448064997038483e-05, "loss": 0.2691296100616455, "memory(GiB)": 73.83, "step": 89915, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.457533 }, { "epoch": 3.8524484812133157, "grad_norm": 4.996209621429443, "learning_rate": 1.2443621949458206e-05, "loss": 0.4237973213195801, "memory(GiB)": 73.83, "step": 89920, "token_acc": 0.9227642276422764, "train_speed(iter/s)": 1.457536 }, { "epoch": 3.8526626965425645, "grad_norm": 4.621468544006348, "learning_rate": 1.2439179582246952e-05, "loss": 0.3503897666931152, "memory(GiB)": 73.83, "step": 89925, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.457539 }, { "epoch": 3.852876911871814, "grad_norm": 2.7529425621032715, "learning_rate": 1.2434737895485188e-05, "loss": 0.39475882053375244, "memory(GiB)": 73.83, "step": 89930, "token_acc": 0.9100719424460432, "train_speed(iter/s)": 1.457545 }, { "epoch": 3.8530911272010626, "grad_norm": 6.842846393585205, "learning_rate": 1.2430296889253378e-05, "loss": 0.3444344997406006, "memory(GiB)": 73.83, "step": 89935, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.457545 }, { "epoch": 3.8533053425303114, "grad_norm": 0.7725549340248108, "learning_rate": 1.2425856563631993e-05, "loss": 0.1819888949394226, "memory(GiB)": 73.83, "step": 89940, "token_acc": 0.9651898734177216, "train_speed(iter/s)": 1.457549 }, { "epoch": 3.8535195578595607, "grad_norm": 6.798842430114746, "learning_rate": 1.2421416918701467e-05, "loss": 0.45990500450134275, "memory(GiB)": 73.83, "step": 89945, "token_acc": 0.8996138996138996, "train_speed(iter/s)": 1.45756 }, { "epoch": 3.8537337731888095, "grad_norm": 4.396656036376953, "learning_rate": 1.2416977954542225e-05, "loss": 0.42119760513305665, "memory(GiB)": 73.83, "step": 89950, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.457561 }, { "epoch": 3.8539479885180583, "grad_norm": 4.873146057128906, "learning_rate": 1.241253967123468e-05, "loss": 0.358493709564209, "memory(GiB)": 73.83, "step": 89955, "token_acc": 0.91701244813278, "train_speed(iter/s)": 1.457572 }, { "epoch": 3.8541622038473076, "grad_norm": 1.1649448871612549, "learning_rate": 1.2408102068859228e-05, "loss": 0.26464688777923584, "memory(GiB)": 73.83, "step": 89960, "token_acc": 0.9467680608365019, "train_speed(iter/s)": 1.457577 }, { "epoch": 3.8543764191765564, "grad_norm": 4.948370456695557, "learning_rate": 1.2403665147496285e-05, "loss": 0.36743528842926027, "memory(GiB)": 73.83, "step": 89965, "token_acc": 0.9267399267399268, "train_speed(iter/s)": 1.457584 }, { "epoch": 3.854590634505805, "grad_norm": 2.861097812652588, "learning_rate": 1.2399228907226212e-05, "loss": 0.37211833000183103, "memory(GiB)": 73.83, "step": 89970, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.457589 }, { "epoch": 3.8548048498350544, "grad_norm": 1.4576095342636108, "learning_rate": 1.2394793348129385e-05, "loss": 0.5923669815063477, "memory(GiB)": 73.83, "step": 89975, "token_acc": 0.8828337874659401, "train_speed(iter/s)": 1.457602 }, { "epoch": 3.8550190651643033, "grad_norm": 3.8015475273132324, "learning_rate": 1.2390358470286151e-05, "loss": 0.2943552494049072, "memory(GiB)": 73.83, "step": 89980, "token_acc": 0.95578231292517, "train_speed(iter/s)": 1.457602 }, { "epoch": 3.855233280493552, "grad_norm": 3.0733988285064697, "learning_rate": 1.2385924273776839e-05, "loss": 0.1820844292640686, "memory(GiB)": 73.83, "step": 89985, "token_acc": 0.95703125, "train_speed(iter/s)": 1.457605 }, { "epoch": 3.8554474958228013, "grad_norm": 4.699697494506836, "learning_rate": 1.2381490758681808e-05, "loss": 0.45168018341064453, "memory(GiB)": 73.83, "step": 89990, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.457608 }, { "epoch": 3.85566171115205, "grad_norm": 4.739465713500977, "learning_rate": 1.2377057925081342e-05, "loss": 0.28493614196777345, "memory(GiB)": 73.83, "step": 89995, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.457615 }, { "epoch": 3.855875926481299, "grad_norm": 0.1979711949825287, "learning_rate": 1.2372625773055784e-05, "loss": 0.4288204193115234, "memory(GiB)": 73.83, "step": 90000, "token_acc": 0.8940809968847352, "train_speed(iter/s)": 1.457621 }, { "epoch": 3.855875926481299, "eval_loss": 2.4813294410705566, "eval_runtime": 11.1955, "eval_samples_per_second": 8.932, "eval_steps_per_second": 8.932, "eval_token_acc": 0.461133069828722, "step": 90000 }, { "epoch": 3.856090141810548, "grad_norm": 4.806739807128906, "learning_rate": 1.236819430268541e-05, "loss": 0.40329957008361816, "memory(GiB)": 73.83, "step": 90005, "token_acc": 0.5827751196172248, "train_speed(iter/s)": 1.457338 }, { "epoch": 3.856304357139797, "grad_norm": 5.686683654785156, "learning_rate": 1.2363763514050496e-05, "loss": 0.500183391571045, "memory(GiB)": 73.83, "step": 90010, "token_acc": 0.9099378881987578, "train_speed(iter/s)": 1.457348 }, { "epoch": 3.856518572469046, "grad_norm": 1.7681432962417603, "learning_rate": 1.235933340723131e-05, "loss": 0.15034539699554444, "memory(GiB)": 73.83, "step": 90015, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.457346 }, { "epoch": 3.856732787798295, "grad_norm": 2.9349136352539062, "learning_rate": 1.2354903982308098e-05, "loss": 0.21499948501586913, "memory(GiB)": 73.83, "step": 90020, "token_acc": 0.9536423841059603, "train_speed(iter/s)": 1.457347 }, { "epoch": 3.856947003127544, "grad_norm": 2.645108461380005, "learning_rate": 1.2350475239361126e-05, "loss": 0.23597075939178466, "memory(GiB)": 73.83, "step": 90025, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.457355 }, { "epoch": 3.8571612184567927, "grad_norm": 0.41614821553230286, "learning_rate": 1.234604717847061e-05, "loss": 0.18529168367385865, "memory(GiB)": 73.83, "step": 90030, "token_acc": 0.9446808510638298, "train_speed(iter/s)": 1.457359 }, { "epoch": 3.857375433786042, "grad_norm": 3.862795114517212, "learning_rate": 1.2341619799716775e-05, "loss": 0.2588310718536377, "memory(GiB)": 73.83, "step": 90035, "token_acc": 0.9525691699604744, "train_speed(iter/s)": 1.457363 }, { "epoch": 3.857589649115291, "grad_norm": 8.077220916748047, "learning_rate": 1.2337193103179823e-05, "loss": 0.36286890506744385, "memory(GiB)": 73.83, "step": 90040, "token_acc": 0.9288888888888889, "train_speed(iter/s)": 1.457361 }, { "epoch": 3.8578038644445396, "grad_norm": 2.1984524726867676, "learning_rate": 1.2332767088939935e-05, "loss": 0.5010680198669434, "memory(GiB)": 73.83, "step": 90045, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.457364 }, { "epoch": 3.858018079773789, "grad_norm": 2.3095309734344482, "learning_rate": 1.2328341757077322e-05, "loss": 0.1698877215385437, "memory(GiB)": 73.83, "step": 90050, "token_acc": 0.9380804953560371, "train_speed(iter/s)": 1.457361 }, { "epoch": 3.8582322951030377, "grad_norm": 1.9761264324188232, "learning_rate": 1.2323917107672134e-05, "loss": 0.25360538959503176, "memory(GiB)": 73.83, "step": 90055, "token_acc": 0.946875, "train_speed(iter/s)": 1.457358 }, { "epoch": 3.8584465104322865, "grad_norm": 1.0605487823486328, "learning_rate": 1.2319493140804534e-05, "loss": 0.37806949615478513, "memory(GiB)": 73.83, "step": 90060, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.45736 }, { "epoch": 3.8586607257615357, "grad_norm": 7.771230220794678, "learning_rate": 1.2315069856554645e-05, "loss": 0.7430219650268555, "memory(GiB)": 73.83, "step": 90065, "token_acc": 0.8517110266159695, "train_speed(iter/s)": 1.457359 }, { "epoch": 3.8588749410907845, "grad_norm": 1.2878715991973877, "learning_rate": 1.2310647255002634e-05, "loss": 0.28627448081970214, "memory(GiB)": 73.83, "step": 90070, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.457361 }, { "epoch": 3.8590891564200334, "grad_norm": 2.7434139251708984, "learning_rate": 1.23062253362286e-05, "loss": 0.5407601356506347, "memory(GiB)": 73.83, "step": 90075, "token_acc": 0.8858024691358025, "train_speed(iter/s)": 1.457367 }, { "epoch": 3.8593033717492826, "grad_norm": 3.3385863304138184, "learning_rate": 1.2301804100312642e-05, "loss": 0.2633014678955078, "memory(GiB)": 73.83, "step": 90080, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.457368 }, { "epoch": 3.8595175870785314, "grad_norm": 3.1813693046569824, "learning_rate": 1.229738354733488e-05, "loss": 0.2799059391021729, "memory(GiB)": 73.83, "step": 90085, "token_acc": 0.9361022364217252, "train_speed(iter/s)": 1.457372 }, { "epoch": 3.8597318024077802, "grad_norm": 1.7382640838623047, "learning_rate": 1.229296367737538e-05, "loss": 0.26642348766326907, "memory(GiB)": 73.83, "step": 90090, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.457386 }, { "epoch": 3.8599460177370295, "grad_norm": 3.106786012649536, "learning_rate": 1.228854449051422e-05, "loss": 0.2535837411880493, "memory(GiB)": 73.83, "step": 90095, "token_acc": 0.939297124600639, "train_speed(iter/s)": 1.457383 }, { "epoch": 3.8601602330662783, "grad_norm": 1.9984363317489624, "learning_rate": 1.2284125986831452e-05, "loss": 0.3563959836959839, "memory(GiB)": 73.83, "step": 90100, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.457383 }, { "epoch": 3.860374448395527, "grad_norm": 0.08810495585203171, "learning_rate": 1.2279708166407106e-05, "loss": 0.07364892363548278, "memory(GiB)": 73.83, "step": 90105, "token_acc": 0.9862542955326461, "train_speed(iter/s)": 1.457386 }, { "epoch": 3.8605886637247764, "grad_norm": 1.7941850423812866, "learning_rate": 1.2275291029321246e-05, "loss": 0.24037108421325684, "memory(GiB)": 73.83, "step": 90110, "token_acc": 0.9470198675496688, "train_speed(iter/s)": 1.457381 }, { "epoch": 3.860802879054025, "grad_norm": 2.1794633865356445, "learning_rate": 1.227087457565388e-05, "loss": 0.27299036979675295, "memory(GiB)": 73.83, "step": 90115, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.457386 }, { "epoch": 3.861017094383274, "grad_norm": 3.2210497856140137, "learning_rate": 1.2266458805485009e-05, "loss": 0.2833712100982666, "memory(GiB)": 73.83, "step": 90120, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.457387 }, { "epoch": 3.8612313097125233, "grad_norm": 2.356083869934082, "learning_rate": 1.2262043718894633e-05, "loss": 0.3288550853729248, "memory(GiB)": 73.83, "step": 90125, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.457393 }, { "epoch": 3.861445525041772, "grad_norm": 1.2101541757583618, "learning_rate": 1.2257629315962726e-05, "loss": 0.4526925563812256, "memory(GiB)": 73.83, "step": 90130, "token_acc": 0.9194630872483222, "train_speed(iter/s)": 1.457395 }, { "epoch": 3.861659740371021, "grad_norm": 2.5838747024536133, "learning_rate": 1.2253215596769285e-05, "loss": 0.3231498718261719, "memory(GiB)": 73.83, "step": 90135, "token_acc": 0.9191616766467066, "train_speed(iter/s)": 1.457399 }, { "epoch": 3.86187395570027, "grad_norm": 2.44923996925354, "learning_rate": 1.2248802561394234e-05, "loss": 0.3713390350341797, "memory(GiB)": 73.83, "step": 90140, "token_acc": 0.9198717948717948, "train_speed(iter/s)": 1.457403 }, { "epoch": 3.862088171029519, "grad_norm": 1.7816448211669922, "learning_rate": 1.2244390209917555e-05, "loss": 0.281220531463623, "memory(GiB)": 73.83, "step": 90145, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.45741 }, { "epoch": 3.8623023863587678, "grad_norm": 1.0256550312042236, "learning_rate": 1.2239978542419162e-05, "loss": 0.17075731754302978, "memory(GiB)": 73.83, "step": 90150, "token_acc": 0.9509803921568627, "train_speed(iter/s)": 1.457412 }, { "epoch": 3.862516601688017, "grad_norm": 3.5504112243652344, "learning_rate": 1.2235567558978983e-05, "loss": 0.19468886852264405, "memory(GiB)": 73.83, "step": 90155, "token_acc": 0.9623287671232876, "train_speed(iter/s)": 1.457416 }, { "epoch": 3.862730817017266, "grad_norm": 3.670356512069702, "learning_rate": 1.2231157259676923e-05, "loss": 0.38615190982818604, "memory(GiB)": 73.83, "step": 90160, "token_acc": 0.9193083573487032, "train_speed(iter/s)": 1.457418 }, { "epoch": 3.8629450323465147, "grad_norm": 3.1300787925720215, "learning_rate": 1.2226747644592862e-05, "loss": 0.5497417449951172, "memory(GiB)": 73.83, "step": 90165, "token_acc": 0.8717948717948718, "train_speed(iter/s)": 1.45742 }, { "epoch": 3.863159247675764, "grad_norm": 4.117883205413818, "learning_rate": 1.2222338713806718e-05, "loss": 0.13011384010314941, "memory(GiB)": 73.83, "step": 90170, "token_acc": 0.9666666666666667, "train_speed(iter/s)": 1.457425 }, { "epoch": 3.8633734630050127, "grad_norm": 5.063903331756592, "learning_rate": 1.2217930467398348e-05, "loss": 0.37989728450775145, "memory(GiB)": 73.83, "step": 90175, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.457435 }, { "epoch": 3.8635876783342615, "grad_norm": 4.614716529846191, "learning_rate": 1.2213522905447612e-05, "loss": 0.4760242462158203, "memory(GiB)": 73.83, "step": 90180, "token_acc": 0.8644578313253012, "train_speed(iter/s)": 1.457436 }, { "epoch": 3.863801893663511, "grad_norm": 6.472898483276367, "learning_rate": 1.2209116028034351e-05, "loss": 0.6448237419128418, "memory(GiB)": 73.83, "step": 90185, "token_acc": 0.8916083916083916, "train_speed(iter/s)": 1.457438 }, { "epoch": 3.8640161089927596, "grad_norm": 4.828573226928711, "learning_rate": 1.2204709835238388e-05, "loss": 0.43872604370117185, "memory(GiB)": 73.83, "step": 90190, "token_acc": 0.9161676646706587, "train_speed(iter/s)": 1.457447 }, { "epoch": 3.8642303243220084, "grad_norm": 2.182910680770874, "learning_rate": 1.2200304327139578e-05, "loss": 0.2390150785446167, "memory(GiB)": 73.83, "step": 90195, "token_acc": 0.937984496124031, "train_speed(iter/s)": 1.457444 }, { "epoch": 3.8644445396512577, "grad_norm": 3.638308525085449, "learning_rate": 1.2195899503817715e-05, "loss": 0.4701512813568115, "memory(GiB)": 73.83, "step": 90200, "token_acc": 0.9012875536480687, "train_speed(iter/s)": 1.457448 }, { "epoch": 3.8646587549805065, "grad_norm": 3.2881875038146973, "learning_rate": 1.2191495365352596e-05, "loss": 0.754229211807251, "memory(GiB)": 73.83, "step": 90205, "token_acc": 0.8783783783783784, "train_speed(iter/s)": 1.457443 }, { "epoch": 3.8648729703097553, "grad_norm": 2.447166919708252, "learning_rate": 1.2187091911823989e-05, "loss": 0.4928779125213623, "memory(GiB)": 73.83, "step": 90210, "token_acc": 0.9102564102564102, "train_speed(iter/s)": 1.457438 }, { "epoch": 3.8650871856390046, "grad_norm": 3.4038572311401367, "learning_rate": 1.21826891433117e-05, "loss": 0.6780456066131592, "memory(GiB)": 73.83, "step": 90215, "token_acc": 0.8673139158576052, "train_speed(iter/s)": 1.457437 }, { "epoch": 3.8653014009682534, "grad_norm": 3.956204891204834, "learning_rate": 1.2178287059895471e-05, "loss": 0.3937422752380371, "memory(GiB)": 73.83, "step": 90220, "token_acc": 0.9087301587301587, "train_speed(iter/s)": 1.457438 }, { "epoch": 3.865515616297502, "grad_norm": 4.645510196685791, "learning_rate": 1.2173885661655038e-05, "loss": 0.5751981258392334, "memory(GiB)": 73.83, "step": 90225, "token_acc": 0.8930817610062893, "train_speed(iter/s)": 1.457441 }, { "epoch": 3.8657298316267514, "grad_norm": 1.853915810585022, "learning_rate": 1.2169484948670162e-05, "loss": 0.4810508728027344, "memory(GiB)": 73.83, "step": 90230, "token_acc": 0.9014084507042254, "train_speed(iter/s)": 1.457441 }, { "epoch": 3.8659440469560002, "grad_norm": 3.9416821002960205, "learning_rate": 1.2165084921020558e-05, "loss": 0.4966928482055664, "memory(GiB)": 73.83, "step": 90235, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.457453 }, { "epoch": 3.866158262285249, "grad_norm": 2.8704171180725098, "learning_rate": 1.216068557878593e-05, "loss": 0.2680001974105835, "memory(GiB)": 73.83, "step": 90240, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.457456 }, { "epoch": 3.8663724776144983, "grad_norm": 2.7927212715148926, "learning_rate": 1.215628692204598e-05, "loss": 0.37238397598266604, "memory(GiB)": 73.83, "step": 90245, "token_acc": 0.9233128834355828, "train_speed(iter/s)": 1.45747 }, { "epoch": 3.866586692943747, "grad_norm": 4.554950714111328, "learning_rate": 1.2151888950880374e-05, "loss": 0.14976338148117066, "memory(GiB)": 73.83, "step": 90250, "token_acc": 0.9746835443037974, "train_speed(iter/s)": 1.457473 }, { "epoch": 3.866800908272996, "grad_norm": 2.50860333442688, "learning_rate": 1.2147491665368826e-05, "loss": 0.35147318840026853, "memory(GiB)": 73.83, "step": 90255, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.457474 }, { "epoch": 3.867015123602245, "grad_norm": 3.0630247592926025, "learning_rate": 1.214309506559097e-05, "loss": 0.4210465431213379, "memory(GiB)": 73.83, "step": 90260, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.457477 }, { "epoch": 3.867229338931494, "grad_norm": 2.8269124031066895, "learning_rate": 1.213869915162646e-05, "loss": 0.18160852193832397, "memory(GiB)": 73.83, "step": 90265, "token_acc": 0.9631147540983607, "train_speed(iter/s)": 1.457476 }, { "epoch": 3.867443554260743, "grad_norm": 5.111919403076172, "learning_rate": 1.2134303923554935e-05, "loss": 0.4171594619750977, "memory(GiB)": 73.83, "step": 90270, "token_acc": 0.9145907473309609, "train_speed(iter/s)": 1.457489 }, { "epoch": 3.867657769589992, "grad_norm": 3.9220895767211914, "learning_rate": 1.2129909381455996e-05, "loss": 0.2932011127471924, "memory(GiB)": 73.83, "step": 90275, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.457508 }, { "epoch": 3.867871984919241, "grad_norm": 2.5833663940429688, "learning_rate": 1.2125515525409293e-05, "loss": 0.1876129150390625, "memory(GiB)": 73.83, "step": 90280, "token_acc": 0.9716981132075472, "train_speed(iter/s)": 1.457514 }, { "epoch": 3.8680862002484897, "grad_norm": 4.021392822265625, "learning_rate": 1.2121122355494385e-05, "loss": 0.32950944900512696, "memory(GiB)": 73.83, "step": 90285, "token_acc": 0.9341692789968652, "train_speed(iter/s)": 1.457514 }, { "epoch": 3.868300415577739, "grad_norm": 2.347498655319214, "learning_rate": 1.21167298717909e-05, "loss": 0.18596528768539428, "memory(GiB)": 73.83, "step": 90290, "token_acc": 0.9570552147239264, "train_speed(iter/s)": 1.457523 }, { "epoch": 3.8685146309069878, "grad_norm": 0.16016380488872528, "learning_rate": 1.2112338074378381e-05, "loss": 0.47811245918273926, "memory(GiB)": 73.83, "step": 90295, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.457528 }, { "epoch": 3.8687288462362366, "grad_norm": 1.2047635316848755, "learning_rate": 1.210794696333641e-05, "loss": 0.09452207684516907, "memory(GiB)": 73.83, "step": 90300, "token_acc": 0.9653846153846154, "train_speed(iter/s)": 1.457533 }, { "epoch": 3.868943061565486, "grad_norm": 3.549401044845581, "learning_rate": 1.210355653874452e-05, "loss": 0.4050013542175293, "memory(GiB)": 73.83, "step": 90305, "token_acc": 0.9136904761904762, "train_speed(iter/s)": 1.457534 }, { "epoch": 3.8691572768947347, "grad_norm": 1.0841261148452759, "learning_rate": 1.2099166800682238e-05, "loss": 0.3076289653778076, "memory(GiB)": 73.83, "step": 90310, "token_acc": 0.9246031746031746, "train_speed(iter/s)": 1.457537 }, { "epoch": 3.8693714922239835, "grad_norm": 3.77402663230896, "learning_rate": 1.209477774922912e-05, "loss": 0.5155835628509522, "memory(GiB)": 73.83, "step": 90315, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.457539 }, { "epoch": 3.8695857075532327, "grad_norm": 2.495979070663452, "learning_rate": 1.2090389384464662e-05, "loss": 0.34235801696777346, "memory(GiB)": 73.83, "step": 90320, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.457539 }, { "epoch": 3.8697999228824815, "grad_norm": 10.177538871765137, "learning_rate": 1.2086001706468358e-05, "loss": 0.39511561393737793, "memory(GiB)": 73.83, "step": 90325, "token_acc": 0.9239543726235742, "train_speed(iter/s)": 1.457554 }, { "epoch": 3.8700141382117303, "grad_norm": 0.10641231387853622, "learning_rate": 1.2081614715319706e-05, "loss": 0.36195385456085205, "memory(GiB)": 73.83, "step": 90330, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.457559 }, { "epoch": 3.8702283535409796, "grad_norm": 3.1472694873809814, "learning_rate": 1.207722841109815e-05, "loss": 0.23217742443084716, "memory(GiB)": 73.83, "step": 90335, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.457562 }, { "epoch": 3.8704425688702284, "grad_norm": 3.0688858032226562, "learning_rate": 1.2072842793883199e-05, "loss": 0.4226864814758301, "memory(GiB)": 73.83, "step": 90340, "token_acc": 0.8971061093247589, "train_speed(iter/s)": 1.457566 }, { "epoch": 3.8706567841994772, "grad_norm": 4.951725482940674, "learning_rate": 1.2068457863754273e-05, "loss": 0.3048563480377197, "memory(GiB)": 73.83, "step": 90345, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.457568 }, { "epoch": 3.8708709995287265, "grad_norm": 2.439950704574585, "learning_rate": 1.2064073620790823e-05, "loss": 0.3128660678863525, "memory(GiB)": 73.83, "step": 90350, "token_acc": 0.9349593495934959, "train_speed(iter/s)": 1.45757 }, { "epoch": 3.8710852148579753, "grad_norm": 3.8248281478881836, "learning_rate": 1.205969006507226e-05, "loss": 0.13602793216705322, "memory(GiB)": 73.83, "step": 90355, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.457581 }, { "epoch": 3.871299430187224, "grad_norm": 1.6617017984390259, "learning_rate": 1.205530719667799e-05, "loss": 0.15389457941055298, "memory(GiB)": 73.83, "step": 90360, "token_acc": 0.9755351681957186, "train_speed(iter/s)": 1.457585 }, { "epoch": 3.8715136455164734, "grad_norm": 2.4739532470703125, "learning_rate": 1.2050925015687437e-05, "loss": 0.4062896728515625, "memory(GiB)": 73.83, "step": 90365, "token_acc": 0.9056603773584906, "train_speed(iter/s)": 1.457598 }, { "epoch": 3.871727860845722, "grad_norm": 0.7302627563476562, "learning_rate": 1.2046543522179966e-05, "loss": 0.17991070747375487, "memory(GiB)": 73.83, "step": 90370, "token_acc": 0.9585987261146497, "train_speed(iter/s)": 1.457602 }, { "epoch": 3.871942076174971, "grad_norm": 5.571044445037842, "learning_rate": 1.2042162716234972e-05, "loss": 0.3249815464019775, "memory(GiB)": 73.83, "step": 90375, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.457607 }, { "epoch": 3.8721562915042202, "grad_norm": 1.0400327444076538, "learning_rate": 1.2037782597931812e-05, "loss": 0.1887362241744995, "memory(GiB)": 73.83, "step": 90380, "token_acc": 0.9601328903654485, "train_speed(iter/s)": 1.457608 }, { "epoch": 3.872370506833469, "grad_norm": 4.439963340759277, "learning_rate": 1.2033403167349833e-05, "loss": 0.24457411766052245, "memory(GiB)": 73.83, "step": 90385, "token_acc": 0.942652329749104, "train_speed(iter/s)": 1.45761 }, { "epoch": 3.872584722162718, "grad_norm": 1.2967875003814697, "learning_rate": 1.2029024424568363e-05, "loss": 0.28574085235595703, "memory(GiB)": 73.83, "step": 90390, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.457613 }, { "epoch": 3.872798937491967, "grad_norm": 3.647479295730591, "learning_rate": 1.2024646369666731e-05, "loss": 0.33872523307800295, "memory(GiB)": 73.83, "step": 90395, "token_acc": 0.9212598425196851, "train_speed(iter/s)": 1.457627 }, { "epoch": 3.873013152821216, "grad_norm": 5.6863837242126465, "learning_rate": 1.202026900272426e-05, "loss": 0.5259632110595703, "memory(GiB)": 73.83, "step": 90400, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.457628 }, { "epoch": 3.8732273681504648, "grad_norm": 2.368912935256958, "learning_rate": 1.201589232382025e-05, "loss": 0.2677757263183594, "memory(GiB)": 73.83, "step": 90405, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.457627 }, { "epoch": 3.873441583479714, "grad_norm": 5.095505714416504, "learning_rate": 1.2011516333033979e-05, "loss": 0.4998190879821777, "memory(GiB)": 73.83, "step": 90410, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.457627 }, { "epoch": 3.873655798808963, "grad_norm": 4.009832382202148, "learning_rate": 1.2007141030444723e-05, "loss": 0.45952515602111815, "memory(GiB)": 73.83, "step": 90415, "token_acc": 0.8922558922558923, "train_speed(iter/s)": 1.457628 }, { "epoch": 3.8738700141382116, "grad_norm": 1.43552565574646, "learning_rate": 1.2002766416131739e-05, "loss": 0.21014256477355958, "memory(GiB)": 73.83, "step": 90420, "token_acc": 0.964, "train_speed(iter/s)": 1.457637 }, { "epoch": 3.874084229467461, "grad_norm": 2.1989970207214355, "learning_rate": 1.1998392490174299e-05, "loss": 0.45302639007568357, "memory(GiB)": 73.83, "step": 90425, "token_acc": 0.899641577060932, "train_speed(iter/s)": 1.457648 }, { "epoch": 3.8742984447967097, "grad_norm": 0.3741726577281952, "learning_rate": 1.1994019252651611e-05, "loss": 0.15927313566207885, "memory(GiB)": 73.83, "step": 90430, "token_acc": 0.9695817490494296, "train_speed(iter/s)": 1.457658 }, { "epoch": 3.8745126601259585, "grad_norm": 6.14724063873291, "learning_rate": 1.1989646703642931e-05, "loss": 0.40214738845825193, "memory(GiB)": 73.83, "step": 90435, "token_acc": 0.9195046439628483, "train_speed(iter/s)": 1.457669 }, { "epoch": 3.8747268754552078, "grad_norm": 2.9086146354675293, "learning_rate": 1.1985274843227456e-05, "loss": 0.29502975940704346, "memory(GiB)": 73.83, "step": 90440, "token_acc": 0.9476923076923077, "train_speed(iter/s)": 1.457674 }, { "epoch": 3.8749410907844566, "grad_norm": 1.8223053216934204, "learning_rate": 1.1980903671484389e-05, "loss": 0.5855454444885254, "memory(GiB)": 73.83, "step": 90445, "token_acc": 0.897887323943662, "train_speed(iter/s)": 1.457677 }, { "epoch": 3.8751553061137054, "grad_norm": 2.3409841060638428, "learning_rate": 1.1976533188492922e-05, "loss": 0.4984877586364746, "memory(GiB)": 73.83, "step": 90450, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.457687 }, { "epoch": 3.8753695214429547, "grad_norm": 4.361268520355225, "learning_rate": 1.1972163394332203e-05, "loss": 0.40771145820617677, "memory(GiB)": 73.83, "step": 90455, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.45769 }, { "epoch": 3.8755837367722035, "grad_norm": 1.1491026878356934, "learning_rate": 1.1967794289081436e-05, "loss": 0.3543394088745117, "memory(GiB)": 73.83, "step": 90460, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.457695 }, { "epoch": 3.8757979521014523, "grad_norm": 1.9203346967697144, "learning_rate": 1.1963425872819755e-05, "loss": 0.3337294578552246, "memory(GiB)": 73.83, "step": 90465, "token_acc": 0.9329073482428115, "train_speed(iter/s)": 1.457696 }, { "epoch": 3.8760121674307015, "grad_norm": 4.244836807250977, "learning_rate": 1.1959058145626289e-05, "loss": 0.18789503574371338, "memory(GiB)": 73.83, "step": 90470, "token_acc": 0.9675324675324676, "train_speed(iter/s)": 1.457701 }, { "epoch": 3.8762263827599504, "grad_norm": 5.271153926849365, "learning_rate": 1.1954691107580174e-05, "loss": 0.4623249053955078, "memory(GiB)": 73.83, "step": 90475, "token_acc": 0.8961538461538462, "train_speed(iter/s)": 1.457696 }, { "epoch": 3.876440598089199, "grad_norm": 4.489410877227783, "learning_rate": 1.1950324758760507e-05, "loss": 0.43621373176574707, "memory(GiB)": 73.83, "step": 90480, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.457696 }, { "epoch": 3.8766548134184484, "grad_norm": 2.939131021499634, "learning_rate": 1.1945959099246407e-05, "loss": 0.38207640647888186, "memory(GiB)": 73.83, "step": 90485, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.457713 }, { "epoch": 3.8768690287476972, "grad_norm": 4.192682266235352, "learning_rate": 1.1941594129116962e-05, "loss": 0.27038209438323973, "memory(GiB)": 73.83, "step": 90490, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.457724 }, { "epoch": 3.877083244076946, "grad_norm": 2.1259169578552246, "learning_rate": 1.1937229848451237e-05, "loss": 0.39382214546203614, "memory(GiB)": 73.83, "step": 90495, "token_acc": 0.9263456090651558, "train_speed(iter/s)": 1.457729 }, { "epoch": 3.8772974594061953, "grad_norm": 6.3020758628845215, "learning_rate": 1.1932866257328302e-05, "loss": 0.4188670635223389, "memory(GiB)": 73.83, "step": 90500, "token_acc": 0.9221183800623053, "train_speed(iter/s)": 1.457739 }, { "epoch": 3.8772974594061953, "eval_loss": 2.4842469692230225, "eval_runtime": 11.1068, "eval_samples_per_second": 9.003, "eval_steps_per_second": 9.003, "eval_token_acc": 0.45174825174825173, "step": 90500 }, { "epoch": 3.877511674735444, "grad_norm": 3.378347396850586, "learning_rate": 1.1928503355827192e-05, "loss": 0.26643321514129636, "memory(GiB)": 73.83, "step": 90505, "token_acc": 0.5895372233400402, "train_speed(iter/s)": 1.457458 }, { "epoch": 3.877725890064693, "grad_norm": 4.694216251373291, "learning_rate": 1.1924141144026969e-05, "loss": 0.3961156368255615, "memory(GiB)": 73.83, "step": 90510, "token_acc": 0.9126506024096386, "train_speed(iter/s)": 1.457457 }, { "epoch": 3.877940105393942, "grad_norm": 3.4190866947174072, "learning_rate": 1.1919779622006632e-05, "loss": 0.38622446060180665, "memory(GiB)": 73.83, "step": 90515, "token_acc": 0.9142857142857143, "train_speed(iter/s)": 1.457456 }, { "epoch": 3.878154320723191, "grad_norm": 3.268723487854004, "learning_rate": 1.1915418789845229e-05, "loss": 0.5263411521911621, "memory(GiB)": 73.83, "step": 90520, "token_acc": 0.9041533546325878, "train_speed(iter/s)": 1.457465 }, { "epoch": 3.87836853605244, "grad_norm": 5.56354284286499, "learning_rate": 1.1911058647621737e-05, "loss": 0.7750755310058594, "memory(GiB)": 73.83, "step": 90525, "token_acc": 0.8424657534246576, "train_speed(iter/s)": 1.457467 }, { "epoch": 3.878582751381689, "grad_norm": 3.199101686477661, "learning_rate": 1.1906699195415144e-05, "loss": 0.31588382720947267, "memory(GiB)": 73.83, "step": 90530, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.457465 }, { "epoch": 3.878796966710938, "grad_norm": 4.295632362365723, "learning_rate": 1.1902340433304431e-05, "loss": 0.3535521745681763, "memory(GiB)": 73.83, "step": 90535, "token_acc": 0.9176029962546817, "train_speed(iter/s)": 1.457472 }, { "epoch": 3.8790111820401867, "grad_norm": 3.7717227935791016, "learning_rate": 1.1897982361368548e-05, "loss": 0.3499160289764404, "memory(GiB)": 73.83, "step": 90540, "token_acc": 0.9240506329113924, "train_speed(iter/s)": 1.457484 }, { "epoch": 3.879225397369436, "grad_norm": 3.9248483180999756, "learning_rate": 1.1893624979686474e-05, "loss": 0.5079047679901123, "memory(GiB)": 73.83, "step": 90545, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.457497 }, { "epoch": 3.8794396126986848, "grad_norm": 4.80728006362915, "learning_rate": 1.1889268288337124e-05, "loss": 0.1805684447288513, "memory(GiB)": 73.83, "step": 90550, "token_acc": 0.9532710280373832, "train_speed(iter/s)": 1.457498 }, { "epoch": 3.8796538280279336, "grad_norm": 1.228669285774231, "learning_rate": 1.1884912287399436e-05, "loss": 0.40908002853393555, "memory(GiB)": 73.83, "step": 90555, "token_acc": 0.9116719242902208, "train_speed(iter/s)": 1.457498 }, { "epoch": 3.879868043357183, "grad_norm": 2.6514434814453125, "learning_rate": 1.1880556976952312e-05, "loss": 0.33432927131652834, "memory(GiB)": 73.83, "step": 90560, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.457504 }, { "epoch": 3.8800822586864316, "grad_norm": 3.4750449657440186, "learning_rate": 1.1876202357074645e-05, "loss": 0.2889508008956909, "memory(GiB)": 73.83, "step": 90565, "token_acc": 0.9457364341085271, "train_speed(iter/s)": 1.457504 }, { "epoch": 3.8802964740156805, "grad_norm": 2.7290427684783936, "learning_rate": 1.187184842784535e-05, "loss": 0.39651989936828613, "memory(GiB)": 73.83, "step": 90570, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.457507 }, { "epoch": 3.8805106893449297, "grad_norm": 2.2924935817718506, "learning_rate": 1.1867495189343286e-05, "loss": 0.2579569101333618, "memory(GiB)": 73.83, "step": 90575, "token_acc": 0.9465875370919882, "train_speed(iter/s)": 1.45751 }, { "epoch": 3.8807249046741785, "grad_norm": 1.5062179565429688, "learning_rate": 1.1863142641647307e-05, "loss": 0.16184935569763184, "memory(GiB)": 73.83, "step": 90580, "token_acc": 0.9645390070921985, "train_speed(iter/s)": 1.45751 }, { "epoch": 3.8809391200034273, "grad_norm": 3.763275623321533, "learning_rate": 1.1858790784836282e-05, "loss": 0.32626590728759763, "memory(GiB)": 73.83, "step": 90585, "token_acc": 0.9308176100628931, "train_speed(iter/s)": 1.45751 }, { "epoch": 3.8811533353326766, "grad_norm": 1.9954270124435425, "learning_rate": 1.185443961898905e-05, "loss": 0.11104137897491455, "memory(GiB)": 73.83, "step": 90590, "token_acc": 0.9713261648745519, "train_speed(iter/s)": 1.457513 }, { "epoch": 3.8813675506619254, "grad_norm": 3.4293577671051025, "learning_rate": 1.1850089144184423e-05, "loss": 0.5828761577606201, "memory(GiB)": 73.83, "step": 90595, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.45752 }, { "epoch": 3.881581765991174, "grad_norm": 1.4560145139694214, "learning_rate": 1.18457393605012e-05, "loss": 0.21939258575439452, "memory(GiB)": 73.83, "step": 90600, "token_acc": 0.940959409594096, "train_speed(iter/s)": 1.457525 }, { "epoch": 3.8817959813204235, "grad_norm": 1.29267418384552, "learning_rate": 1.184139026801822e-05, "loss": 0.3058335542678833, "memory(GiB)": 73.83, "step": 90605, "token_acc": 0.9381107491856677, "train_speed(iter/s)": 1.457532 }, { "epoch": 3.8820101966496723, "grad_norm": 3.2843403816223145, "learning_rate": 1.1837041866814252e-05, "loss": 0.5483071804046631, "memory(GiB)": 73.83, "step": 90610, "token_acc": 0.8814102564102564, "train_speed(iter/s)": 1.457525 }, { "epoch": 3.882224411978921, "grad_norm": 0.7982764840126038, "learning_rate": 1.1832694156968066e-05, "loss": 0.4335508346557617, "memory(GiB)": 73.83, "step": 90615, "token_acc": 0.90625, "train_speed(iter/s)": 1.457528 }, { "epoch": 3.8824386273081704, "grad_norm": 4.200196266174316, "learning_rate": 1.1828347138558432e-05, "loss": 0.5126582145690918, "memory(GiB)": 73.83, "step": 90620, "token_acc": 0.8740458015267175, "train_speed(iter/s)": 1.457529 }, { "epoch": 3.882652842637419, "grad_norm": 0.384503573179245, "learning_rate": 1.1824000811664083e-05, "loss": 0.4376936435699463, "memory(GiB)": 73.83, "step": 90625, "token_acc": 0.8850574712643678, "train_speed(iter/s)": 1.457528 }, { "epoch": 3.882867057966668, "grad_norm": 2.879732608795166, "learning_rate": 1.1819655176363786e-05, "loss": 0.30875391960144044, "memory(GiB)": 73.83, "step": 90630, "token_acc": 0.9228486646884273, "train_speed(iter/s)": 1.457528 }, { "epoch": 3.8830812732959172, "grad_norm": 0.9128987789154053, "learning_rate": 1.1815310232736249e-05, "loss": 0.24763970375061034, "memory(GiB)": 73.83, "step": 90635, "token_acc": 0.9511278195488722, "train_speed(iter/s)": 1.457536 }, { "epoch": 3.883295488625166, "grad_norm": 6.16156005859375, "learning_rate": 1.1810965980860189e-05, "loss": 0.5324638366699219, "memory(GiB)": 73.83, "step": 90640, "token_acc": 0.889967637540453, "train_speed(iter/s)": 1.457546 }, { "epoch": 3.883509703954415, "grad_norm": 3.5398502349853516, "learning_rate": 1.1806622420814306e-05, "loss": 0.34369065761566164, "memory(GiB)": 73.83, "step": 90645, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.457559 }, { "epoch": 3.883723919283664, "grad_norm": 3.94158935546875, "learning_rate": 1.1802279552677282e-05, "loss": 0.1725207209587097, "memory(GiB)": 73.83, "step": 90650, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.457561 }, { "epoch": 3.883938134612913, "grad_norm": 2.3705971240997314, "learning_rate": 1.1797937376527784e-05, "loss": 0.4201171875, "memory(GiB)": 73.83, "step": 90655, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.457569 }, { "epoch": 3.8841523499421617, "grad_norm": 4.576180458068848, "learning_rate": 1.1793595892444492e-05, "loss": 0.15951237678527833, "memory(GiB)": 73.83, "step": 90660, "token_acc": 0.9651162790697675, "train_speed(iter/s)": 1.457571 }, { "epoch": 3.884366565271411, "grad_norm": 4.026871681213379, "learning_rate": 1.1789255100506057e-05, "loss": 0.29544754028320314, "memory(GiB)": 73.83, "step": 90665, "token_acc": 0.9462809917355371, "train_speed(iter/s)": 1.457574 }, { "epoch": 3.88458078060066, "grad_norm": 5.779103755950928, "learning_rate": 1.1784915000791114e-05, "loss": 0.5278478145599366, "memory(GiB)": 73.83, "step": 90670, "token_acc": 0.8850574712643678, "train_speed(iter/s)": 1.457591 }, { "epoch": 3.8847949959299086, "grad_norm": 5.528655052185059, "learning_rate": 1.1780575593378284e-05, "loss": 0.4569265842437744, "memory(GiB)": 73.83, "step": 90675, "token_acc": 0.9044368600682594, "train_speed(iter/s)": 1.457592 }, { "epoch": 3.885009211259158, "grad_norm": 4.000006675720215, "learning_rate": 1.177623687834618e-05, "loss": 0.28462882041931153, "memory(GiB)": 73.83, "step": 90680, "token_acc": 0.938953488372093, "train_speed(iter/s)": 1.457591 }, { "epoch": 3.8852234265884067, "grad_norm": 2.9860682487487793, "learning_rate": 1.1771898855773388e-05, "loss": 0.3386664867401123, "memory(GiB)": 73.83, "step": 90685, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.457592 }, { "epoch": 3.8854376419176555, "grad_norm": 1.9474713802337646, "learning_rate": 1.1767561525738525e-05, "loss": 0.22866222858428956, "memory(GiB)": 73.83, "step": 90690, "token_acc": 0.948339483394834, "train_speed(iter/s)": 1.457601 }, { "epoch": 3.8856518572469048, "grad_norm": 3.789435625076294, "learning_rate": 1.1763224888320145e-05, "loss": 0.2423105239868164, "memory(GiB)": 73.83, "step": 90695, "token_acc": 0.9541984732824428, "train_speed(iter/s)": 1.457614 }, { "epoch": 3.8858660725761536, "grad_norm": 0.9913832545280457, "learning_rate": 1.1758888943596818e-05, "loss": 0.3358945846557617, "memory(GiB)": 73.83, "step": 90700, "token_acc": 0.9465648854961832, "train_speed(iter/s)": 1.457614 }, { "epoch": 3.8860802879054024, "grad_norm": 0.5320296287536621, "learning_rate": 1.1754553691647092e-05, "loss": 0.27064101696014403, "memory(GiB)": 73.83, "step": 90705, "token_acc": 0.9455128205128205, "train_speed(iter/s)": 1.457626 }, { "epoch": 3.8862945032346516, "grad_norm": 1.6522294282913208, "learning_rate": 1.1750219132549489e-05, "loss": 0.2916949510574341, "memory(GiB)": 73.83, "step": 90710, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.457629 }, { "epoch": 3.8865087185639005, "grad_norm": 10.367413520812988, "learning_rate": 1.1745885266382561e-05, "loss": 0.31367380619049073, "memory(GiB)": 73.83, "step": 90715, "token_acc": 0.9339080459770115, "train_speed(iter/s)": 1.457632 }, { "epoch": 3.8867229338931493, "grad_norm": 3.011582374572754, "learning_rate": 1.1741552093224805e-05, "loss": 0.35669121742248533, "memory(GiB)": 73.83, "step": 90720, "token_acc": 0.926984126984127, "train_speed(iter/s)": 1.457638 }, { "epoch": 3.8869371492223985, "grad_norm": 3.762279987335205, "learning_rate": 1.1737219613154727e-05, "loss": 0.5395762920379639, "memory(GiB)": 73.83, "step": 90725, "token_acc": 0.8782051282051282, "train_speed(iter/s)": 1.457645 }, { "epoch": 3.8871513645516473, "grad_norm": 3.7383968830108643, "learning_rate": 1.1732887826250787e-05, "loss": 0.3326548099517822, "memory(GiB)": 73.83, "step": 90730, "token_acc": 0.9383116883116883, "train_speed(iter/s)": 1.457661 }, { "epoch": 3.887365579880896, "grad_norm": 0.2810438871383667, "learning_rate": 1.1728556732591501e-05, "loss": 0.34749057292938235, "memory(GiB)": 73.83, "step": 90735, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.457665 }, { "epoch": 3.8875797952101454, "grad_norm": 2.4828131198883057, "learning_rate": 1.172422633225531e-05, "loss": 0.25553069114685056, "memory(GiB)": 73.83, "step": 90740, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.457671 }, { "epoch": 3.887794010539394, "grad_norm": 6.071589946746826, "learning_rate": 1.1719896625320654e-05, "loss": 0.21765007972717285, "memory(GiB)": 73.83, "step": 90745, "token_acc": 0.956386292834891, "train_speed(iter/s)": 1.457684 }, { "epoch": 3.888008225868643, "grad_norm": 1.4205396175384521, "learning_rate": 1.1715567611865991e-05, "loss": 0.38486812114715574, "memory(GiB)": 73.83, "step": 90750, "token_acc": 0.9172661870503597, "train_speed(iter/s)": 1.457689 }, { "epoch": 3.8882224411978923, "grad_norm": 3.4763662815093994, "learning_rate": 1.1711239291969733e-05, "loss": 0.2986193418502808, "memory(GiB)": 73.83, "step": 90755, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.457698 }, { "epoch": 3.888436656527141, "grad_norm": 3.5906896591186523, "learning_rate": 1.1706911665710296e-05, "loss": 0.24403109550476074, "memory(GiB)": 73.83, "step": 90760, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.457694 }, { "epoch": 3.88865087185639, "grad_norm": 1.6510474681854248, "learning_rate": 1.1702584733166073e-05, "loss": 0.2805623531341553, "memory(GiB)": 73.83, "step": 90765, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.457697 }, { "epoch": 3.888865087185639, "grad_norm": 3.885110378265381, "learning_rate": 1.1698258494415443e-05, "loss": 0.357115650177002, "memory(GiB)": 73.83, "step": 90770, "token_acc": 0.9215017064846417, "train_speed(iter/s)": 1.457699 }, { "epoch": 3.889079302514888, "grad_norm": 3.3046727180480957, "learning_rate": 1.1693932949536801e-05, "loss": 0.2732023954391479, "memory(GiB)": 73.83, "step": 90775, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.457705 }, { "epoch": 3.889293517844137, "grad_norm": 3.9724385738372803, "learning_rate": 1.1689608098608495e-05, "loss": 0.26155757904052734, "memory(GiB)": 73.83, "step": 90780, "token_acc": 0.9346938775510204, "train_speed(iter/s)": 1.457707 }, { "epoch": 3.889507733173386, "grad_norm": 3.007251501083374, "learning_rate": 1.168528394170888e-05, "loss": 0.4491099834442139, "memory(GiB)": 73.83, "step": 90785, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.457714 }, { "epoch": 3.889721948502635, "grad_norm": 5.536174297332764, "learning_rate": 1.1680960478916292e-05, "loss": 0.37458722591400145, "memory(GiB)": 73.83, "step": 90790, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.45772 }, { "epoch": 3.8899361638318837, "grad_norm": 3.3828043937683105, "learning_rate": 1.1676637710309047e-05, "loss": 0.4107633113861084, "memory(GiB)": 73.83, "step": 90795, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.45772 }, { "epoch": 3.890150379161133, "grad_norm": 4.32712984085083, "learning_rate": 1.1672315635965447e-05, "loss": 0.6891708850860596, "memory(GiB)": 73.83, "step": 90800, "token_acc": 0.84375, "train_speed(iter/s)": 1.457712 }, { "epoch": 3.8903645944903817, "grad_norm": 4.63887357711792, "learning_rate": 1.1667994255963805e-05, "loss": 0.3176675796508789, "memory(GiB)": 73.83, "step": 90805, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.457713 }, { "epoch": 3.8905788098196306, "grad_norm": 0.8413729071617126, "learning_rate": 1.1663673570382416e-05, "loss": 0.31972532272338866, "memory(GiB)": 73.83, "step": 90810, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.457715 }, { "epoch": 3.89079302514888, "grad_norm": 4.970249176025391, "learning_rate": 1.1659353579299543e-05, "loss": 0.3181858777999878, "memory(GiB)": 73.83, "step": 90815, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.457712 }, { "epoch": 3.8910072404781286, "grad_norm": 3.5169949531555176, "learning_rate": 1.1655034282793448e-05, "loss": 0.20919620990753174, "memory(GiB)": 73.83, "step": 90820, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.457722 }, { "epoch": 3.8912214558073774, "grad_norm": 4.84558629989624, "learning_rate": 1.1650715680942381e-05, "loss": 0.5051758766174317, "memory(GiB)": 73.83, "step": 90825, "token_acc": 0.8932038834951457, "train_speed(iter/s)": 1.45772 }, { "epoch": 3.8914356711366267, "grad_norm": 1.319471001625061, "learning_rate": 1.1646397773824553e-05, "loss": 0.2046119451522827, "memory(GiB)": 73.83, "step": 90830, "token_acc": 0.9580838323353293, "train_speed(iter/s)": 1.457725 }, { "epoch": 3.8916498864658755, "grad_norm": 2.8459391593933105, "learning_rate": 1.1642080561518226e-05, "loss": 0.26075472831726076, "memory(GiB)": 73.83, "step": 90835, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.457729 }, { "epoch": 3.8918641017951243, "grad_norm": 2.512054681777954, "learning_rate": 1.163776404410159e-05, "loss": 0.24284090995788574, "memory(GiB)": 73.83, "step": 90840, "token_acc": 0.9590163934426229, "train_speed(iter/s)": 1.457732 }, { "epoch": 3.8920783171243736, "grad_norm": 0.5419874787330627, "learning_rate": 1.1633448221652848e-05, "loss": 0.39919726848602294, "memory(GiB)": 73.83, "step": 90845, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.457734 }, { "epoch": 3.8922925324536224, "grad_norm": 0.7300463318824768, "learning_rate": 1.1629133094250183e-05, "loss": 0.2945871353149414, "memory(GiB)": 73.83, "step": 90850, "token_acc": 0.9363636363636364, "train_speed(iter/s)": 1.457739 }, { "epoch": 3.892506747782871, "grad_norm": 0.1734209954738617, "learning_rate": 1.1624818661971747e-05, "loss": 0.26758565902709963, "memory(GiB)": 73.83, "step": 90855, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.457739 }, { "epoch": 3.8927209631121205, "grad_norm": 3.174577474594116, "learning_rate": 1.1620504924895737e-05, "loss": 0.26487016677856445, "memory(GiB)": 73.83, "step": 90860, "token_acc": 0.9327485380116959, "train_speed(iter/s)": 1.457752 }, { "epoch": 3.8929351784413693, "grad_norm": 2.960054874420166, "learning_rate": 1.1616191883100285e-05, "loss": 0.5078456401824951, "memory(GiB)": 73.83, "step": 90865, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.45776 }, { "epoch": 3.893149393770618, "grad_norm": 5.287675857543945, "learning_rate": 1.1611879536663523e-05, "loss": 0.20877516269683838, "memory(GiB)": 73.83, "step": 90870, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.457775 }, { "epoch": 3.8933636090998673, "grad_norm": 0.6956615447998047, "learning_rate": 1.1607567885663555e-05, "loss": 0.21232635974884034, "memory(GiB)": 73.83, "step": 90875, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.45778 }, { "epoch": 3.893577824429116, "grad_norm": 7.241737365722656, "learning_rate": 1.160325693017853e-05, "loss": 0.3612092971801758, "memory(GiB)": 73.83, "step": 90880, "token_acc": 0.9196141479099679, "train_speed(iter/s)": 1.457783 }, { "epoch": 3.893792039758365, "grad_norm": 2.8794517517089844, "learning_rate": 1.1598946670286525e-05, "loss": 0.26961510181427, "memory(GiB)": 73.83, "step": 90885, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.45779 }, { "epoch": 3.894006255087614, "grad_norm": 0.19230271875858307, "learning_rate": 1.1594637106065608e-05, "loss": 0.41100587844848635, "memory(GiB)": 73.83, "step": 90890, "token_acc": 0.9110320284697508, "train_speed(iter/s)": 1.457794 }, { "epoch": 3.894220470416863, "grad_norm": 0.7278115153312683, "learning_rate": 1.1590328237593878e-05, "loss": 0.2054546594619751, "memory(GiB)": 73.83, "step": 90895, "token_acc": 0.959409594095941, "train_speed(iter/s)": 1.457795 }, { "epoch": 3.894434685746112, "grad_norm": 2.9802396297454834, "learning_rate": 1.1586020064949387e-05, "loss": 0.19142534732818603, "memory(GiB)": 73.83, "step": 90900, "token_acc": 0.9501661129568106, "train_speed(iter/s)": 1.457807 }, { "epoch": 3.894648901075361, "grad_norm": 3.0349340438842773, "learning_rate": 1.1581712588210181e-05, "loss": 0.26067638397216797, "memory(GiB)": 73.83, "step": 90905, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.457818 }, { "epoch": 3.89486311640461, "grad_norm": 3.469395399093628, "learning_rate": 1.1577405807454283e-05, "loss": 0.2861616134643555, "memory(GiB)": 73.83, "step": 90910, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.457825 }, { "epoch": 3.8950773317338587, "grad_norm": 1.405258059501648, "learning_rate": 1.1573099722759712e-05, "loss": 0.2463066577911377, "memory(GiB)": 73.83, "step": 90915, "token_acc": 0.9391891891891891, "train_speed(iter/s)": 1.457839 }, { "epoch": 3.895291547063108, "grad_norm": 0.6816351413726807, "learning_rate": 1.1568794334204502e-05, "loss": 0.218516206741333, "memory(GiB)": 73.83, "step": 90920, "token_acc": 0.9547038327526133, "train_speed(iter/s)": 1.457843 }, { "epoch": 3.895505762392357, "grad_norm": 2.554910659790039, "learning_rate": 1.1564489641866633e-05, "loss": 0.4748044967651367, "memory(GiB)": 73.83, "step": 90925, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.45785 }, { "epoch": 3.8957199777216056, "grad_norm": 4.0920729637146, "learning_rate": 1.156018564582409e-05, "loss": 0.5031590461730957, "memory(GiB)": 73.83, "step": 90930, "token_acc": 0.8859060402684564, "train_speed(iter/s)": 1.457864 }, { "epoch": 3.895934193050855, "grad_norm": 2.680917978286743, "learning_rate": 1.1555882346154845e-05, "loss": 0.3709068775177002, "memory(GiB)": 73.83, "step": 90935, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.457871 }, { "epoch": 3.8961484083801037, "grad_norm": 2.5939857959747314, "learning_rate": 1.1551579742936852e-05, "loss": 0.38845648765563967, "memory(GiB)": 73.83, "step": 90940, "token_acc": 0.9126984126984127, "train_speed(iter/s)": 1.457873 }, { "epoch": 3.8963626237093525, "grad_norm": 3.087005615234375, "learning_rate": 1.1547277836248043e-05, "loss": 0.479155445098877, "memory(GiB)": 73.83, "step": 90945, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.457885 }, { "epoch": 3.8965768390386017, "grad_norm": 3.8949804306030273, "learning_rate": 1.1542976626166374e-05, "loss": 0.3188114881515503, "memory(GiB)": 73.83, "step": 90950, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.457891 }, { "epoch": 3.8967910543678506, "grad_norm": 6.156152725219727, "learning_rate": 1.153867611276977e-05, "loss": 0.4296678066253662, "memory(GiB)": 73.83, "step": 90955, "token_acc": 0.8993055555555556, "train_speed(iter/s)": 1.457889 }, { "epoch": 3.8970052696970994, "grad_norm": 5.166787624359131, "learning_rate": 1.1534376296136124e-05, "loss": 0.4322836399078369, "memory(GiB)": 73.83, "step": 90960, "token_acc": 0.9119718309859155, "train_speed(iter/s)": 1.45789 }, { "epoch": 3.8972194850263486, "grad_norm": 2.869427442550659, "learning_rate": 1.153007717634334e-05, "loss": 0.4284999847412109, "memory(GiB)": 73.83, "step": 90965, "token_acc": 0.8847352024922118, "train_speed(iter/s)": 1.457888 }, { "epoch": 3.8974337003555974, "grad_norm": 0.6782225370407104, "learning_rate": 1.1525778753469297e-05, "loss": 0.1420462965965271, "memory(GiB)": 73.83, "step": 90970, "token_acc": 0.9780701754385965, "train_speed(iter/s)": 1.45789 }, { "epoch": 3.8976479156848463, "grad_norm": 2.9080607891082764, "learning_rate": 1.1521481027591847e-05, "loss": 0.5227378368377685, "memory(GiB)": 73.83, "step": 90975, "token_acc": 0.8837920489296636, "train_speed(iter/s)": 1.457909 }, { "epoch": 3.8978621310140955, "grad_norm": 1.8099223375320435, "learning_rate": 1.1517183998788877e-05, "loss": 0.5329712390899658, "memory(GiB)": 73.83, "step": 90980, "token_acc": 0.8695652173913043, "train_speed(iter/s)": 1.457915 }, { "epoch": 3.8980763463433443, "grad_norm": 4.930373668670654, "learning_rate": 1.1512887667138217e-05, "loss": 0.4073484420776367, "memory(GiB)": 73.83, "step": 90985, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.457915 }, { "epoch": 3.898290561672593, "grad_norm": 4.958105564117432, "learning_rate": 1.1508592032717701e-05, "loss": 0.5263996124267578, "memory(GiB)": 73.83, "step": 90990, "token_acc": 0.905511811023622, "train_speed(iter/s)": 1.457928 }, { "epoch": 3.8985047770018424, "grad_norm": 3.6781303882598877, "learning_rate": 1.1504297095605154e-05, "loss": 0.2759371280670166, "memory(GiB)": 73.83, "step": 90995, "token_acc": 0.9438202247191011, "train_speed(iter/s)": 1.45794 }, { "epoch": 3.898718992331091, "grad_norm": 3.056676149368286, "learning_rate": 1.1500002855878362e-05, "loss": 0.3003842353820801, "memory(GiB)": 73.83, "step": 91000, "token_acc": 0.9439655172413793, "train_speed(iter/s)": 1.457941 }, { "epoch": 3.898718992331091, "eval_loss": 2.283008098602295, "eval_runtime": 11.2978, "eval_samples_per_second": 8.851, "eval_steps_per_second": 8.851, "eval_token_acc": 0.4743935309973046, "step": 91000 }, { "epoch": 3.89893320766034, "grad_norm": 1.0669194459915161, "learning_rate": 1.1495709313615143e-05, "loss": 0.3634622812271118, "memory(GiB)": 73.83, "step": 91005, "token_acc": 0.5842583249243188, "train_speed(iter/s)": 1.457656 }, { "epoch": 3.8991474229895893, "grad_norm": 4.1106696128845215, "learning_rate": 1.1491416468893274e-05, "loss": 0.5322070121765137, "memory(GiB)": 73.83, "step": 91010, "token_acc": 0.8732394366197183, "train_speed(iter/s)": 1.457669 }, { "epoch": 3.899361638318838, "grad_norm": 0.5300968885421753, "learning_rate": 1.1487124321790515e-05, "loss": 0.426801061630249, "memory(GiB)": 73.83, "step": 91015, "token_acc": 0.8982456140350877, "train_speed(iter/s)": 1.457679 }, { "epoch": 3.899575853648087, "grad_norm": 3.7978427410125732, "learning_rate": 1.148283287238463e-05, "loss": 0.41280522346496584, "memory(GiB)": 73.83, "step": 91020, "token_acc": 0.9108910891089109, "train_speed(iter/s)": 1.457677 }, { "epoch": 3.899790068977336, "grad_norm": 5.917712211608887, "learning_rate": 1.1478542120753344e-05, "loss": 0.4478609561920166, "memory(GiB)": 73.83, "step": 91025, "token_acc": 0.9330708661417323, "train_speed(iter/s)": 1.457686 }, { "epoch": 3.900004284306585, "grad_norm": 0.37611451745033264, "learning_rate": 1.1474252066974422e-05, "loss": 0.13512693643569945, "memory(GiB)": 73.83, "step": 91030, "token_acc": 0.967391304347826, "train_speed(iter/s)": 1.457689 }, { "epoch": 3.900218499635834, "grad_norm": 5.373564720153809, "learning_rate": 1.1469962711125548e-05, "loss": 0.3215423822402954, "memory(GiB)": 73.83, "step": 91035, "token_acc": 0.941747572815534, "train_speed(iter/s)": 1.457694 }, { "epoch": 3.900432714965083, "grad_norm": 2.056446075439453, "learning_rate": 1.1465674053284452e-05, "loss": 0.19418163299560548, "memory(GiB)": 73.83, "step": 91040, "token_acc": 0.9536423841059603, "train_speed(iter/s)": 1.457702 }, { "epoch": 3.900646930294332, "grad_norm": 3.2290022373199463, "learning_rate": 1.1461386093528826e-05, "loss": 0.43697481155395507, "memory(GiB)": 73.83, "step": 91045, "token_acc": 0.91875, "train_speed(iter/s)": 1.457705 }, { "epoch": 3.9008611456235807, "grad_norm": 1.9609496593475342, "learning_rate": 1.1457098831936342e-05, "loss": 0.20117173194885254, "memory(GiB)": 73.83, "step": 91050, "token_acc": 0.9614147909967846, "train_speed(iter/s)": 1.457717 }, { "epoch": 3.90107536095283, "grad_norm": 2.2066526412963867, "learning_rate": 1.1452812268584667e-05, "loss": 0.23923470973968505, "memory(GiB)": 73.83, "step": 91055, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.457721 }, { "epoch": 3.9012895762820787, "grad_norm": 4.608396053314209, "learning_rate": 1.1448526403551441e-05, "loss": 0.5464173316955566, "memory(GiB)": 73.83, "step": 91060, "token_acc": 0.8853868194842407, "train_speed(iter/s)": 1.457718 }, { "epoch": 3.9015037916113275, "grad_norm": 2.861891746520996, "learning_rate": 1.1444241236914343e-05, "loss": 0.22468876838684082, "memory(GiB)": 73.83, "step": 91065, "token_acc": 0.9429657794676806, "train_speed(iter/s)": 1.45772 }, { "epoch": 3.901718006940577, "grad_norm": 4.112911224365234, "learning_rate": 1.143995676875098e-05, "loss": 0.36719536781311035, "memory(GiB)": 73.83, "step": 91070, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.457726 }, { "epoch": 3.9019322222698256, "grad_norm": 3.509002923965454, "learning_rate": 1.1435672999138975e-05, "loss": 0.5159715175628662, "memory(GiB)": 73.83, "step": 91075, "token_acc": 0.8932384341637011, "train_speed(iter/s)": 1.457742 }, { "epoch": 3.9021464375990744, "grad_norm": 0.48365306854248047, "learning_rate": 1.1431389928155922e-05, "loss": 0.3299422025680542, "memory(GiB)": 73.83, "step": 91080, "token_acc": 0.904320987654321, "train_speed(iter/s)": 1.457748 }, { "epoch": 3.9023606529283237, "grad_norm": 3.2347378730773926, "learning_rate": 1.1427107555879412e-05, "loss": 0.3947983026504517, "memory(GiB)": 73.83, "step": 91085, "token_acc": 0.934984520123839, "train_speed(iter/s)": 1.457752 }, { "epoch": 3.9025748682575725, "grad_norm": 2.268125057220459, "learning_rate": 1.142282588238705e-05, "loss": 0.40955724716186526, "memory(GiB)": 73.83, "step": 91090, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.457752 }, { "epoch": 3.9027890835868213, "grad_norm": 3.7068190574645996, "learning_rate": 1.1418544907756368e-05, "loss": 0.483968448638916, "memory(GiB)": 73.83, "step": 91095, "token_acc": 0.8871473354231975, "train_speed(iter/s)": 1.457762 }, { "epoch": 3.9030032989160706, "grad_norm": 4.165232181549072, "learning_rate": 1.1414264632064952e-05, "loss": 0.1699204683303833, "memory(GiB)": 73.83, "step": 91100, "token_acc": 0.9690402476780186, "train_speed(iter/s)": 1.457768 }, { "epoch": 3.9032175142453194, "grad_norm": 0.6733167171478271, "learning_rate": 1.1409985055390332e-05, "loss": 0.197745418548584, "memory(GiB)": 73.83, "step": 91105, "token_acc": 0.9437229437229437, "train_speed(iter/s)": 1.457775 }, { "epoch": 3.903431729574568, "grad_norm": 6.096571445465088, "learning_rate": 1.1405706177810027e-05, "loss": 0.5168612003326416, "memory(GiB)": 73.83, "step": 91110, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.457786 }, { "epoch": 3.9036459449038174, "grad_norm": 2.3120319843292236, "learning_rate": 1.1401427999401565e-05, "loss": 0.38667471408843995, "memory(GiB)": 73.83, "step": 91115, "token_acc": 0.8976897689768977, "train_speed(iter/s)": 1.4578 }, { "epoch": 3.9038601602330663, "grad_norm": 3.131805181503296, "learning_rate": 1.1397150520242422e-05, "loss": 0.5390504837036133, "memory(GiB)": 73.83, "step": 91120, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.457807 }, { "epoch": 3.904074375562315, "grad_norm": 4.741940021514893, "learning_rate": 1.1392873740410132e-05, "loss": 0.420134162902832, "memory(GiB)": 73.83, "step": 91125, "token_acc": 0.8996282527881041, "train_speed(iter/s)": 1.457812 }, { "epoch": 3.9042885908915643, "grad_norm": 7.278970241546631, "learning_rate": 1.1388597659982148e-05, "loss": 0.5865985870361328, "memory(GiB)": 73.83, "step": 91130, "token_acc": 0.9006410256410257, "train_speed(iter/s)": 1.457824 }, { "epoch": 3.904502806220813, "grad_norm": 4.308842182159424, "learning_rate": 1.138432227903593e-05, "loss": 0.25224077701568604, "memory(GiB)": 73.83, "step": 91135, "token_acc": 0.927007299270073, "train_speed(iter/s)": 1.457824 }, { "epoch": 3.904717021550062, "grad_norm": 3.3313984870910645, "learning_rate": 1.1380047597648946e-05, "loss": 0.46082444190979005, "memory(GiB)": 73.83, "step": 91140, "token_acc": 0.9007633587786259, "train_speed(iter/s)": 1.457836 }, { "epoch": 3.904931236879311, "grad_norm": 2.733034610748291, "learning_rate": 1.1375773615898617e-05, "loss": 0.39096105098724365, "memory(GiB)": 73.83, "step": 91145, "token_acc": 0.925531914893617, "train_speed(iter/s)": 1.457847 }, { "epoch": 3.90514545220856, "grad_norm": 2.086001396179199, "learning_rate": 1.137150033386239e-05, "loss": 0.2681437492370605, "memory(GiB)": 73.83, "step": 91150, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.457861 }, { "epoch": 3.905359667537809, "grad_norm": 3.09391188621521, "learning_rate": 1.1367227751617671e-05, "loss": 0.23886497020721437, "memory(GiB)": 73.83, "step": 91155, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.457863 }, { "epoch": 3.905573882867058, "grad_norm": 3.0024495124816895, "learning_rate": 1.1362955869241865e-05, "loss": 0.5457503318786621, "memory(GiB)": 73.83, "step": 91160, "token_acc": 0.8818897637795275, "train_speed(iter/s)": 1.457873 }, { "epoch": 3.905788098196307, "grad_norm": 4.002335071563721, "learning_rate": 1.1358684686812354e-05, "loss": 0.38543190956115725, "memory(GiB)": 73.83, "step": 91165, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.457876 }, { "epoch": 3.9060023135255557, "grad_norm": 0.83725905418396, "learning_rate": 1.1354414204406505e-05, "loss": 0.26732423305511477, "memory(GiB)": 73.83, "step": 91170, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.457881 }, { "epoch": 3.906216528854805, "grad_norm": 3.2213079929351807, "learning_rate": 1.135014442210171e-05, "loss": 0.2430692195892334, "memory(GiB)": 73.83, "step": 91175, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.457886 }, { "epoch": 3.906430744184054, "grad_norm": 4.467784404754639, "learning_rate": 1.1345875339975292e-05, "loss": 0.25074281692504885, "memory(GiB)": 73.83, "step": 91180, "token_acc": 0.9427710843373494, "train_speed(iter/s)": 1.457886 }, { "epoch": 3.9066449595133026, "grad_norm": 1.904608130455017, "learning_rate": 1.1341606958104616e-05, "loss": 0.25010297298431394, "memory(GiB)": 73.83, "step": 91185, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.457883 }, { "epoch": 3.906859174842552, "grad_norm": 2.6763086318969727, "learning_rate": 1.1337339276566993e-05, "loss": 0.4181967735290527, "memory(GiB)": 73.83, "step": 91190, "token_acc": 0.8987730061349694, "train_speed(iter/s)": 1.457902 }, { "epoch": 3.9070733901718007, "grad_norm": 5.5242919921875, "learning_rate": 1.1333072295439739e-05, "loss": 0.9844871520996094, "memory(GiB)": 73.83, "step": 91195, "token_acc": 0.8202247191011236, "train_speed(iter/s)": 1.457902 }, { "epoch": 3.9072876055010495, "grad_norm": 4.098672389984131, "learning_rate": 1.1328806014800158e-05, "loss": 0.1638599991798401, "memory(GiB)": 73.83, "step": 91200, "token_acc": 0.9682539682539683, "train_speed(iter/s)": 1.457911 }, { "epoch": 3.9075018208302987, "grad_norm": 0.7275869846343994, "learning_rate": 1.132454043472551e-05, "loss": 0.30180404186248777, "memory(GiB)": 73.83, "step": 91205, "token_acc": 0.9028776978417267, "train_speed(iter/s)": 1.45791 }, { "epoch": 3.9077160361595475, "grad_norm": 4.1869401931762695, "learning_rate": 1.1320275555293113e-05, "loss": 0.2046203851699829, "memory(GiB)": 73.83, "step": 91210, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.457919 }, { "epoch": 3.9079302514887964, "grad_norm": 3.054114818572998, "learning_rate": 1.1316011376580204e-05, "loss": 0.3405191421508789, "memory(GiB)": 73.83, "step": 91215, "token_acc": 0.9377777777777778, "train_speed(iter/s)": 1.457915 }, { "epoch": 3.9081444668180456, "grad_norm": 2.747586727142334, "learning_rate": 1.1311747898664038e-05, "loss": 0.4966853141784668, "memory(GiB)": 73.83, "step": 91220, "token_acc": 0.8694267515923567, "train_speed(iter/s)": 1.457919 }, { "epoch": 3.9083586821472944, "grad_norm": 0.6367446780204773, "learning_rate": 1.1307485121621858e-05, "loss": 0.18903902769088746, "memory(GiB)": 73.83, "step": 91225, "token_acc": 0.9513888888888888, "train_speed(iter/s)": 1.457926 }, { "epoch": 3.9085728974765432, "grad_norm": 3.996934652328491, "learning_rate": 1.1303223045530859e-05, "loss": 0.4915400505065918, "memory(GiB)": 73.83, "step": 91230, "token_acc": 0.9067055393586005, "train_speed(iter/s)": 1.457936 }, { "epoch": 3.9087871128057925, "grad_norm": 0.12196636945009232, "learning_rate": 1.1298961670468294e-05, "loss": 0.1647907018661499, "memory(GiB)": 73.83, "step": 91235, "token_acc": 0.9702602230483272, "train_speed(iter/s)": 1.457932 }, { "epoch": 3.9090013281350413, "grad_norm": 4.840312957763672, "learning_rate": 1.1294700996511342e-05, "loss": 0.26951000690460203, "memory(GiB)": 73.83, "step": 91240, "token_acc": 0.9528985507246377, "train_speed(iter/s)": 1.457937 }, { "epoch": 3.90921554346429, "grad_norm": 2.312631845474243, "learning_rate": 1.1290441023737175e-05, "loss": 0.27956862449645997, "memory(GiB)": 73.83, "step": 91245, "token_acc": 0.9294117647058824, "train_speed(iter/s)": 1.457942 }, { "epoch": 3.9094297587935394, "grad_norm": 3.925098180770874, "learning_rate": 1.1286181752222996e-05, "loss": 0.7830999374389649, "memory(GiB)": 73.83, "step": 91250, "token_acc": 0.8539325842696629, "train_speed(iter/s)": 1.457941 }, { "epoch": 3.909643974122788, "grad_norm": 4.322506427764893, "learning_rate": 1.128192318204595e-05, "loss": 0.557285213470459, "memory(GiB)": 73.83, "step": 91255, "token_acc": 0.8562300319488818, "train_speed(iter/s)": 1.457941 }, { "epoch": 3.909858189452037, "grad_norm": 7.242366313934326, "learning_rate": 1.1277665313283187e-05, "loss": 0.6659724235534668, "memory(GiB)": 73.83, "step": 91260, "token_acc": 0.8618181818181818, "train_speed(iter/s)": 1.457948 }, { "epoch": 3.9100724047812863, "grad_norm": 4.049862861633301, "learning_rate": 1.1273408146011827e-05, "loss": 0.39203662872314454, "memory(GiB)": 73.83, "step": 91265, "token_acc": 0.9060402684563759, "train_speed(iter/s)": 1.45795 }, { "epoch": 3.910286620110535, "grad_norm": 0.2829808294773102, "learning_rate": 1.1269151680309021e-05, "loss": 0.35770719051361083, "memory(GiB)": 73.83, "step": 91270, "token_acc": 0.9192307692307692, "train_speed(iter/s)": 1.457962 }, { "epoch": 3.910500835439784, "grad_norm": 2.342261552810669, "learning_rate": 1.126489591625186e-05, "loss": 0.4231730937957764, "memory(GiB)": 73.83, "step": 91275, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.457963 }, { "epoch": 3.910715050769033, "grad_norm": 12.165470123291016, "learning_rate": 1.1260640853917453e-05, "loss": 0.71180100440979, "memory(GiB)": 73.83, "step": 91280, "token_acc": 0.8645833333333334, "train_speed(iter/s)": 1.457977 }, { "epoch": 3.910929266098282, "grad_norm": 6.365111351013184, "learning_rate": 1.1256386493382876e-05, "loss": 0.5071557998657227, "memory(GiB)": 73.83, "step": 91285, "token_acc": 0.8931297709923665, "train_speed(iter/s)": 1.457986 }, { "epoch": 3.9111434814275308, "grad_norm": 4.296713829040527, "learning_rate": 1.1252132834725187e-05, "loss": 0.4073168754577637, "memory(GiB)": 73.83, "step": 91290, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.457998 }, { "epoch": 3.91135769675678, "grad_norm": 4.150421142578125, "learning_rate": 1.1247879878021472e-05, "loss": 0.5954475402832031, "memory(GiB)": 73.83, "step": 91295, "token_acc": 0.8469387755102041, "train_speed(iter/s)": 1.457998 }, { "epoch": 3.911571912086029, "grad_norm": 1.7464042901992798, "learning_rate": 1.1243627623348769e-05, "loss": 0.376607608795166, "memory(GiB)": 73.83, "step": 91300, "token_acc": 0.9183006535947712, "train_speed(iter/s)": 1.458004 }, { "epoch": 3.9117861274152776, "grad_norm": 3.4221487045288086, "learning_rate": 1.1239376070784108e-05, "loss": 0.335605525970459, "memory(GiB)": 73.83, "step": 91305, "token_acc": 0.942652329749104, "train_speed(iter/s)": 1.458002 }, { "epoch": 3.912000342744527, "grad_norm": 2.3293142318725586, "learning_rate": 1.1235125220404507e-05, "loss": 0.2641822338104248, "memory(GiB)": 73.83, "step": 91310, "token_acc": 0.9435483870967742, "train_speed(iter/s)": 1.458004 }, { "epoch": 3.9122145580737757, "grad_norm": 1.999759554862976, "learning_rate": 1.1230875072286979e-05, "loss": 0.15882608890533448, "memory(GiB)": 73.83, "step": 91315, "token_acc": 0.9529914529914529, "train_speed(iter/s)": 1.458004 }, { "epoch": 3.9124287734030245, "grad_norm": 0.7291763424873352, "learning_rate": 1.1226625626508502e-05, "loss": 0.27457003593444823, "memory(GiB)": 73.83, "step": 91320, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.458014 }, { "epoch": 3.912642988732274, "grad_norm": 3.723565101623535, "learning_rate": 1.1222376883146079e-05, "loss": 0.49408931732177735, "memory(GiB)": 73.83, "step": 91325, "token_acc": 0.919732441471572, "train_speed(iter/s)": 1.458019 }, { "epoch": 3.9128572040615226, "grad_norm": 3.158301830291748, "learning_rate": 1.1218128842276688e-05, "loss": 0.6796288013458252, "memory(GiB)": 73.83, "step": 91330, "token_acc": 0.8534798534798534, "train_speed(iter/s)": 1.458022 }, { "epoch": 3.9130714193907714, "grad_norm": 4.605672836303711, "learning_rate": 1.121388150397727e-05, "loss": 0.22416868209838867, "memory(GiB)": 73.83, "step": 91335, "token_acc": 0.9497907949790795, "train_speed(iter/s)": 1.458024 }, { "epoch": 3.9132856347200207, "grad_norm": 3.3996944427490234, "learning_rate": 1.120963486832477e-05, "loss": 0.35319702625274657, "memory(GiB)": 73.83, "step": 91340, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.458044 }, { "epoch": 3.9134998500492695, "grad_norm": 4.632585048675537, "learning_rate": 1.1205388935396127e-05, "loss": 0.3825770139694214, "memory(GiB)": 73.83, "step": 91345, "token_acc": 0.9191176470588235, "train_speed(iter/s)": 1.458041 }, { "epoch": 3.9137140653785183, "grad_norm": 2.985145092010498, "learning_rate": 1.120114370526824e-05, "loss": 0.23938283920288086, "memory(GiB)": 73.83, "step": 91350, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.458039 }, { "epoch": 3.9139282807077675, "grad_norm": 4.880465507507324, "learning_rate": 1.119689917801805e-05, "loss": 0.47259230613708497, "memory(GiB)": 73.83, "step": 91355, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.458039 }, { "epoch": 3.9141424960370164, "grad_norm": 3.552344799041748, "learning_rate": 1.1192655353722425e-05, "loss": 0.4804570198059082, "memory(GiB)": 73.83, "step": 91360, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.458051 }, { "epoch": 3.914356711366265, "grad_norm": 0.8001114130020142, "learning_rate": 1.1188412232458256e-05, "loss": 0.2270660877227783, "memory(GiB)": 73.83, "step": 91365, "token_acc": 0.9446254071661238, "train_speed(iter/s)": 1.458058 }, { "epoch": 3.9145709266955144, "grad_norm": 2.203639268875122, "learning_rate": 1.1184169814302409e-05, "loss": 0.30090980529785155, "memory(GiB)": 73.83, "step": 91370, "token_acc": 0.9300911854103343, "train_speed(iter/s)": 1.458065 }, { "epoch": 3.9147851420247632, "grad_norm": 2.647027015686035, "learning_rate": 1.117992809933172e-05, "loss": 0.13991138935089112, "memory(GiB)": 73.83, "step": 91375, "token_acc": 0.9638989169675091, "train_speed(iter/s)": 1.458068 }, { "epoch": 3.914999357354012, "grad_norm": 2.902872323989868, "learning_rate": 1.1175687087623066e-05, "loss": 0.4930086135864258, "memory(GiB)": 73.83, "step": 91380, "token_acc": 0.8869257950530035, "train_speed(iter/s)": 1.458078 }, { "epoch": 3.9152135726832613, "grad_norm": 0.16884928941726685, "learning_rate": 1.1171446779253258e-05, "loss": 0.32490570545196534, "memory(GiB)": 73.83, "step": 91385, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.458082 }, { "epoch": 3.91542778801251, "grad_norm": 2.0970873832702637, "learning_rate": 1.116720717429912e-05, "loss": 0.3508862257003784, "memory(GiB)": 73.83, "step": 91390, "token_acc": 0.9291044776119403, "train_speed(iter/s)": 1.458087 }, { "epoch": 3.915642003341759, "grad_norm": 5.189492225646973, "learning_rate": 1.1162968272837438e-05, "loss": 0.3059358596801758, "memory(GiB)": 73.83, "step": 91395, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.458085 }, { "epoch": 3.915856218671008, "grad_norm": 0.8638033270835876, "learning_rate": 1.1158730074945029e-05, "loss": 0.3315077781677246, "memory(GiB)": 73.83, "step": 91400, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.458084 }, { "epoch": 3.916070434000257, "grad_norm": 2.037128210067749, "learning_rate": 1.1154492580698661e-05, "loss": 0.2710346460342407, "memory(GiB)": 73.83, "step": 91405, "token_acc": 0.9548387096774194, "train_speed(iter/s)": 1.458097 }, { "epoch": 3.916284649329506, "grad_norm": 1.7496336698532104, "learning_rate": 1.1150255790175086e-05, "loss": 0.2638402462005615, "memory(GiB)": 73.83, "step": 91410, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.458099 }, { "epoch": 3.916498864658755, "grad_norm": 2.890815019607544, "learning_rate": 1.1146019703451083e-05, "loss": 0.2876424312591553, "memory(GiB)": 73.83, "step": 91415, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.458107 }, { "epoch": 3.916713079988004, "grad_norm": 3.205042600631714, "learning_rate": 1.1141784320603382e-05, "loss": 0.5849930286407471, "memory(GiB)": 73.83, "step": 91420, "token_acc": 0.8687943262411347, "train_speed(iter/s)": 1.458109 }, { "epoch": 3.9169272953172527, "grad_norm": 0.6587882041931152, "learning_rate": 1.1137549641708717e-05, "loss": 0.3379584789276123, "memory(GiB)": 73.83, "step": 91425, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.458107 }, { "epoch": 3.917141510646502, "grad_norm": 2.67872953414917, "learning_rate": 1.1133315666843791e-05, "loss": 0.1787131905555725, "memory(GiB)": 73.83, "step": 91430, "token_acc": 0.958041958041958, "train_speed(iter/s)": 1.458111 }, { "epoch": 3.9173557259757508, "grad_norm": 3.1058096885681152, "learning_rate": 1.1129082396085294e-05, "loss": 0.2664060592651367, "memory(GiB)": 73.83, "step": 91435, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.458129 }, { "epoch": 3.9175699413049996, "grad_norm": 2.396226644515991, "learning_rate": 1.1124849829509953e-05, "loss": 0.22731118202209472, "memory(GiB)": 73.83, "step": 91440, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.458132 }, { "epoch": 3.917784156634249, "grad_norm": 3.872570037841797, "learning_rate": 1.1120617967194425e-05, "loss": 0.1873868227005005, "memory(GiB)": 73.83, "step": 91445, "token_acc": 0.9695652173913043, "train_speed(iter/s)": 1.458149 }, { "epoch": 3.9179983719634977, "grad_norm": 5.92330265045166, "learning_rate": 1.1116386809215368e-05, "loss": 0.6381815433502197, "memory(GiB)": 73.83, "step": 91450, "token_acc": 0.8701298701298701, "train_speed(iter/s)": 1.458159 }, { "epoch": 3.9182125872927465, "grad_norm": 1.3099873065948486, "learning_rate": 1.1112156355649444e-05, "loss": 0.3329931735992432, "memory(GiB)": 73.83, "step": 91455, "token_acc": 0.9156976744186046, "train_speed(iter/s)": 1.458158 }, { "epoch": 3.9184268026219957, "grad_norm": 2.1500282287597656, "learning_rate": 1.1107926606573287e-05, "loss": 0.37689805030822754, "memory(GiB)": 73.83, "step": 91460, "token_acc": 0.9115646258503401, "train_speed(iter/s)": 1.458164 }, { "epoch": 3.9186410179512445, "grad_norm": 3.1329219341278076, "learning_rate": 1.1103697562063513e-05, "loss": 0.271016263961792, "memory(GiB)": 73.83, "step": 91465, "token_acc": 0.9385665529010239, "train_speed(iter/s)": 1.458166 }, { "epoch": 3.9188552332804933, "grad_norm": 0.15283501148223877, "learning_rate": 1.109946922219674e-05, "loss": 0.15472275018692017, "memory(GiB)": 73.83, "step": 91470, "token_acc": 0.9646643109540636, "train_speed(iter/s)": 1.458175 }, { "epoch": 3.9190694486097426, "grad_norm": 1.3507678508758545, "learning_rate": 1.1095241587049593e-05, "loss": 0.3030203342437744, "memory(GiB)": 73.83, "step": 91475, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.458186 }, { "epoch": 3.9192836639389914, "grad_norm": 6.726649284362793, "learning_rate": 1.1091014656698634e-05, "loss": 0.27059574127197267, "memory(GiB)": 73.83, "step": 91480, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.458187 }, { "epoch": 3.9194978792682402, "grad_norm": 2.6539368629455566, "learning_rate": 1.1086788431220446e-05, "loss": 0.19522749185562133, "memory(GiB)": 73.83, "step": 91485, "token_acc": 0.9563758389261745, "train_speed(iter/s)": 1.458199 }, { "epoch": 3.9197120945974895, "grad_norm": 0.40296903252601624, "learning_rate": 1.108256291069159e-05, "loss": 0.2910623550415039, "memory(GiB)": 73.83, "step": 91490, "token_acc": 0.9251101321585903, "train_speed(iter/s)": 1.458204 }, { "epoch": 3.9199263099267383, "grad_norm": 2.706145763397217, "learning_rate": 1.1078338095188601e-05, "loss": 0.5221336364746094, "memory(GiB)": 73.83, "step": 91495, "token_acc": 0.8884892086330936, "train_speed(iter/s)": 1.458213 }, { "epoch": 3.920140525255987, "grad_norm": 4.028011322021484, "learning_rate": 1.1074113984788043e-05, "loss": 0.308853006362915, "memory(GiB)": 73.83, "step": 91500, "token_acc": 0.9437751004016064, "train_speed(iter/s)": 1.458215 }, { "epoch": 3.920140525255987, "eval_loss": 2.5758440494537354, "eval_runtime": 11.4481, "eval_samples_per_second": 8.735, "eval_steps_per_second": 8.735, "eval_token_acc": 0.4457070707070707, "step": 91500 }, { "epoch": 3.9203547405852364, "grad_norm": 3.9008448123931885, "learning_rate": 1.1069890579566427e-05, "loss": 0.3023218631744385, "memory(GiB)": 73.83, "step": 91505, "token_acc": 0.5840071877807727, "train_speed(iter/s)": 1.457918 }, { "epoch": 3.920568955914485, "grad_norm": 1.292372465133667, "learning_rate": 1.106566787960026e-05, "loss": 0.2808968544006348, "memory(GiB)": 73.83, "step": 91510, "token_acc": 0.9596774193548387, "train_speed(iter/s)": 1.457918 }, { "epoch": 3.920783171243734, "grad_norm": 2.2550652027130127, "learning_rate": 1.1061445884966042e-05, "loss": 0.12767796516418456, "memory(GiB)": 73.83, "step": 91515, "token_acc": 0.967741935483871, "train_speed(iter/s)": 1.45792 }, { "epoch": 3.9209973865729832, "grad_norm": 2.222951889038086, "learning_rate": 1.1057224595740246e-05, "loss": 0.39107584953308105, "memory(GiB)": 73.83, "step": 91520, "token_acc": 0.9201277955271565, "train_speed(iter/s)": 1.457921 }, { "epoch": 3.921211601902232, "grad_norm": 4.765552520751953, "learning_rate": 1.1053004011999374e-05, "loss": 0.45908393859863283, "memory(GiB)": 73.83, "step": 91525, "token_acc": 0.900355871886121, "train_speed(iter/s)": 1.457927 }, { "epoch": 3.921425817231481, "grad_norm": 3.7131199836730957, "learning_rate": 1.1048784133819867e-05, "loss": 0.22529094219207763, "memory(GiB)": 73.83, "step": 91530, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.457929 }, { "epoch": 3.92164003256073, "grad_norm": 3.2461938858032227, "learning_rate": 1.1044564961278175e-05, "loss": 0.473720121383667, "memory(GiB)": 73.83, "step": 91535, "token_acc": 0.898989898989899, "train_speed(iter/s)": 1.457937 }, { "epoch": 3.921854247889979, "grad_norm": 2.257159948348999, "learning_rate": 1.104034649445072e-05, "loss": 0.35536043643951415, "memory(GiB)": 73.83, "step": 91540, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.457937 }, { "epoch": 3.9220684632192278, "grad_norm": 4.146340847015381, "learning_rate": 1.1036128733413948e-05, "loss": 0.20813589096069335, "memory(GiB)": 73.83, "step": 91545, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.45794 }, { "epoch": 3.922282678548477, "grad_norm": 3.267580509185791, "learning_rate": 1.1031911678244255e-05, "loss": 0.40640764236450194, "memory(GiB)": 73.83, "step": 91550, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.457941 }, { "epoch": 3.922496893877726, "grad_norm": 0.32521113753318787, "learning_rate": 1.1027695329018023e-05, "loss": 0.08435572385787964, "memory(GiB)": 73.83, "step": 91555, "token_acc": 0.9651162790697675, "train_speed(iter/s)": 1.457946 }, { "epoch": 3.9227111092069746, "grad_norm": 0.7093426585197449, "learning_rate": 1.102347968581166e-05, "loss": 0.15180329084396363, "memory(GiB)": 73.83, "step": 91560, "token_acc": 0.9541666666666667, "train_speed(iter/s)": 1.457951 }, { "epoch": 3.922925324536224, "grad_norm": 4.173666954040527, "learning_rate": 1.101926474870153e-05, "loss": 0.41053028106689454, "memory(GiB)": 73.83, "step": 91565, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.457961 }, { "epoch": 3.9231395398654727, "grad_norm": 2.570300817489624, "learning_rate": 1.1015050517763987e-05, "loss": 0.22688181400299073, "memory(GiB)": 73.83, "step": 91570, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.457969 }, { "epoch": 3.9233537551947215, "grad_norm": 1.6287587881088257, "learning_rate": 1.101083699307537e-05, "loss": 0.2198075532913208, "memory(GiB)": 73.83, "step": 91575, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.457966 }, { "epoch": 3.9235679705239708, "grad_norm": 4.619167327880859, "learning_rate": 1.1006624174711999e-05, "loss": 0.52133207321167, "memory(GiB)": 73.83, "step": 91580, "token_acc": 0.887240356083086, "train_speed(iter/s)": 1.457975 }, { "epoch": 3.9237821858532196, "grad_norm": 6.403381824493408, "learning_rate": 1.1002412062750233e-05, "loss": 0.35586249828338623, "memory(GiB)": 73.83, "step": 91585, "token_acc": 0.9196787148594378, "train_speed(iter/s)": 1.457987 }, { "epoch": 3.9239964011824684, "grad_norm": 0.6540128588676453, "learning_rate": 1.099820065726635e-05, "loss": 0.13758621215820313, "memory(GiB)": 73.83, "step": 91590, "token_acc": 0.9692307692307692, "train_speed(iter/s)": 1.457991 }, { "epoch": 3.9242106165117177, "grad_norm": 5.357930660247803, "learning_rate": 1.0993989958336648e-05, "loss": 0.5887109279632569, "memory(GiB)": 73.83, "step": 91595, "token_acc": 0.8944281524926686, "train_speed(iter/s)": 1.45799 }, { "epoch": 3.9244248318409665, "grad_norm": 1.540175437927246, "learning_rate": 1.098977996603741e-05, "loss": 0.3847651958465576, "memory(GiB)": 73.83, "step": 91600, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.457991 }, { "epoch": 3.9246390471702153, "grad_norm": 1.8642610311508179, "learning_rate": 1.09855706804449e-05, "loss": 0.40784592628479005, "memory(GiB)": 73.83, "step": 91605, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.457999 }, { "epoch": 3.9248532624994645, "grad_norm": 1.7369176149368286, "learning_rate": 1.0981362101635367e-05, "loss": 0.23652191162109376, "memory(GiB)": 73.83, "step": 91610, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.458001 }, { "epoch": 3.9250674778287133, "grad_norm": 3.2242937088012695, "learning_rate": 1.0977154229685055e-05, "loss": 0.45053863525390625, "memory(GiB)": 73.83, "step": 91615, "token_acc": 0.8983050847457628, "train_speed(iter/s)": 1.458006 }, { "epoch": 3.925281693157962, "grad_norm": 1.3247929811477661, "learning_rate": 1.097294706467022e-05, "loss": 0.42120065689086916, "memory(GiB)": 73.83, "step": 91620, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.458007 }, { "epoch": 3.9254959084872114, "grad_norm": 2.686821699142456, "learning_rate": 1.096874060666706e-05, "loss": 0.3599431276321411, "memory(GiB)": 73.83, "step": 91625, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.458016 }, { "epoch": 3.9257101238164602, "grad_norm": 0.33125001192092896, "learning_rate": 1.0964534855751774e-05, "loss": 0.3414464950561523, "memory(GiB)": 73.83, "step": 91630, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.458023 }, { "epoch": 3.925924339145709, "grad_norm": 1.9407265186309814, "learning_rate": 1.0960329812000557e-05, "loss": 0.4116572380065918, "memory(GiB)": 73.83, "step": 91635, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.45803 }, { "epoch": 3.9261385544749583, "grad_norm": 2.994199514389038, "learning_rate": 1.0956125475489575e-05, "loss": 0.2259655475616455, "memory(GiB)": 73.83, "step": 91640, "token_acc": 0.9404388714733543, "train_speed(iter/s)": 1.458031 }, { "epoch": 3.926352769804207, "grad_norm": 6.596823692321777, "learning_rate": 1.095192184629502e-05, "loss": 0.7681463241577149, "memory(GiB)": 73.83, "step": 91645, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.458029 }, { "epoch": 3.926566985133456, "grad_norm": 4.970283031463623, "learning_rate": 1.0947718924493034e-05, "loss": 0.5388926029205322, "memory(GiB)": 73.83, "step": 91650, "token_acc": 0.875, "train_speed(iter/s)": 1.458041 }, { "epoch": 3.926781200462705, "grad_norm": 4.2132134437561035, "learning_rate": 1.0943516710159751e-05, "loss": 0.4341888427734375, "memory(GiB)": 73.83, "step": 91655, "token_acc": 0.9174917491749175, "train_speed(iter/s)": 1.458041 }, { "epoch": 3.926995415791954, "grad_norm": 4.49338960647583, "learning_rate": 1.0939315203371303e-05, "loss": 0.31499905586242677, "memory(GiB)": 73.83, "step": 91660, "token_acc": 0.9329608938547486, "train_speed(iter/s)": 1.458042 }, { "epoch": 3.927209631121203, "grad_norm": 2.347440719604492, "learning_rate": 1.0935114404203783e-05, "loss": 0.34077303409576415, "memory(GiB)": 73.83, "step": 91665, "token_acc": 0.9076433121019108, "train_speed(iter/s)": 1.45804 }, { "epoch": 3.927423846450452, "grad_norm": 2.778613805770874, "learning_rate": 1.0930914312733332e-05, "loss": 0.5651789665222168, "memory(GiB)": 73.83, "step": 91670, "token_acc": 0.9, "train_speed(iter/s)": 1.458046 }, { "epoch": 3.927638061779701, "grad_norm": 3.5154995918273926, "learning_rate": 1.0926714929036014e-05, "loss": 0.17024749517440796, "memory(GiB)": 73.83, "step": 91675, "token_acc": 0.9695817490494296, "train_speed(iter/s)": 1.45805 }, { "epoch": 3.9278522771089497, "grad_norm": 2.5934159755706787, "learning_rate": 1.0922516253187909e-05, "loss": 0.5703732967376709, "memory(GiB)": 73.83, "step": 91680, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.458061 }, { "epoch": 3.928066492438199, "grad_norm": 3.65692400932312, "learning_rate": 1.0918318285265078e-05, "loss": 0.30271828174591064, "memory(GiB)": 73.83, "step": 91685, "token_acc": 0.9375, "train_speed(iter/s)": 1.458065 }, { "epoch": 3.9282807077674478, "grad_norm": 3.431220293045044, "learning_rate": 1.091412102534356e-05, "loss": 0.39194815158843993, "memory(GiB)": 73.83, "step": 91690, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.458067 }, { "epoch": 3.9284949230966966, "grad_norm": 1.8125990629196167, "learning_rate": 1.0909924473499423e-05, "loss": 0.28984191417694094, "memory(GiB)": 73.83, "step": 91695, "token_acc": 0.9402390438247012, "train_speed(iter/s)": 1.458073 }, { "epoch": 3.928709138425946, "grad_norm": 3.257910966873169, "learning_rate": 1.0905728629808654e-05, "loss": 0.2887981176376343, "memory(GiB)": 73.83, "step": 91700, "token_acc": 0.9476923076923077, "train_speed(iter/s)": 1.458074 }, { "epoch": 3.9289233537551946, "grad_norm": 4.957387924194336, "learning_rate": 1.0901533494347304e-05, "loss": 0.586701774597168, "memory(GiB)": 73.83, "step": 91705, "token_acc": 0.8836565096952909, "train_speed(iter/s)": 1.458078 }, { "epoch": 3.9291375690844434, "grad_norm": 3.5388522148132324, "learning_rate": 1.0897339067191347e-05, "loss": 0.8287006378173828, "memory(GiB)": 73.83, "step": 91710, "token_acc": 0.8350515463917526, "train_speed(iter/s)": 1.45808 }, { "epoch": 3.9293517844136927, "grad_norm": 8.285316467285156, "learning_rate": 1.0893145348416778e-05, "loss": 0.5036337852478028, "memory(GiB)": 73.83, "step": 91715, "token_acc": 0.8767772511848341, "train_speed(iter/s)": 1.458088 }, { "epoch": 3.9295659997429415, "grad_norm": 2.8332388401031494, "learning_rate": 1.0888952338099561e-05, "loss": 0.5213450908660888, "memory(GiB)": 73.83, "step": 91720, "token_acc": 0.90234375, "train_speed(iter/s)": 1.458108 }, { "epoch": 3.9297802150721903, "grad_norm": 0.0724354162812233, "learning_rate": 1.088476003631565e-05, "loss": 0.33218839168548586, "memory(GiB)": 73.83, "step": 91725, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.458121 }, { "epoch": 3.9299944304014396, "grad_norm": 4.580667018890381, "learning_rate": 1.088056844314102e-05, "loss": 0.5888550281524658, "memory(GiB)": 73.83, "step": 91730, "token_acc": 0.90625, "train_speed(iter/s)": 1.458119 }, { "epoch": 3.9302086457306884, "grad_norm": 2.321489095687866, "learning_rate": 1.087637755865159e-05, "loss": 0.4488957405090332, "memory(GiB)": 73.83, "step": 91735, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.458122 }, { "epoch": 3.930422861059937, "grad_norm": 1.7353183031082153, "learning_rate": 1.0872187382923277e-05, "loss": 0.37646892070770266, "memory(GiB)": 73.83, "step": 91740, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.458122 }, { "epoch": 3.9306370763891865, "grad_norm": 0.49422168731689453, "learning_rate": 1.0867997916031997e-05, "loss": 0.2183070421218872, "memory(GiB)": 73.83, "step": 91745, "token_acc": 0.9672727272727273, "train_speed(iter/s)": 1.458128 }, { "epoch": 3.9308512917184353, "grad_norm": 7.483141899108887, "learning_rate": 1.0863809158053645e-05, "loss": 0.5710744857788086, "memory(GiB)": 73.83, "step": 91750, "token_acc": 0.8728323699421965, "train_speed(iter/s)": 1.458139 }, { "epoch": 3.931065507047684, "grad_norm": 5.349090099334717, "learning_rate": 1.0859621109064089e-05, "loss": 0.34262771606445314, "memory(GiB)": 73.83, "step": 91755, "token_acc": 0.9176470588235294, "train_speed(iter/s)": 1.458152 }, { "epoch": 3.9312797223769333, "grad_norm": 1.865323543548584, "learning_rate": 1.0855433769139223e-05, "loss": 0.33566510677337646, "memory(GiB)": 73.83, "step": 91760, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.458153 }, { "epoch": 3.931493937706182, "grad_norm": 2.492424249649048, "learning_rate": 1.0851247138354886e-05, "loss": 0.18887959718704223, "memory(GiB)": 73.83, "step": 91765, "token_acc": 0.9638989169675091, "train_speed(iter/s)": 1.458157 }, { "epoch": 3.931708153035431, "grad_norm": 5.626111030578613, "learning_rate": 1.084706121678694e-05, "loss": 0.47156682014465334, "memory(GiB)": 73.83, "step": 91770, "token_acc": 0.8984615384615384, "train_speed(iter/s)": 1.45816 }, { "epoch": 3.9319223683646802, "grad_norm": 0.5531324148178101, "learning_rate": 1.0842876004511215e-05, "loss": 0.6185770511627198, "memory(GiB)": 73.83, "step": 91775, "token_acc": 0.8929889298892989, "train_speed(iter/s)": 1.458167 }, { "epoch": 3.932136583693929, "grad_norm": 4.476040840148926, "learning_rate": 1.083869150160352e-05, "loss": 0.37564663887023925, "memory(GiB)": 73.83, "step": 91780, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.458164 }, { "epoch": 3.932350799023178, "grad_norm": 3.644111156463623, "learning_rate": 1.0834507708139651e-05, "loss": 0.4385236740112305, "memory(GiB)": 73.83, "step": 91785, "token_acc": 0.8879668049792531, "train_speed(iter/s)": 1.458173 }, { "epoch": 3.932565014352427, "grad_norm": 3.798424005508423, "learning_rate": 1.083032462419543e-05, "loss": 0.38988988399505614, "memory(GiB)": 73.83, "step": 91790, "token_acc": 0.925, "train_speed(iter/s)": 1.458176 }, { "epoch": 3.932779229681676, "grad_norm": 1.6732984781265259, "learning_rate": 1.0826142249846621e-05, "loss": 0.21159737110137938, "memory(GiB)": 73.83, "step": 91795, "token_acc": 0.9463087248322147, "train_speed(iter/s)": 1.45818 }, { "epoch": 3.9329934450109247, "grad_norm": 2.150635242462158, "learning_rate": 1.0821960585168994e-05, "loss": 0.1822056531906128, "memory(GiB)": 73.83, "step": 91800, "token_acc": 0.9559322033898305, "train_speed(iter/s)": 1.45818 }, { "epoch": 3.933207660340174, "grad_norm": 1.3695580959320068, "learning_rate": 1.0817779630238301e-05, "loss": 0.22789301872253417, "memory(GiB)": 73.83, "step": 91805, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.458181 }, { "epoch": 3.933421875669423, "grad_norm": 2.2031073570251465, "learning_rate": 1.0813599385130274e-05, "loss": 0.2973299264907837, "memory(GiB)": 73.83, "step": 91810, "token_acc": 0.9307958477508651, "train_speed(iter/s)": 1.458181 }, { "epoch": 3.9336360909986716, "grad_norm": 4.603542804718018, "learning_rate": 1.0809419849920671e-05, "loss": 0.35037319660186766, "memory(GiB)": 73.83, "step": 91815, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.45819 }, { "epoch": 3.933850306327921, "grad_norm": 2.092822551727295, "learning_rate": 1.0805241024685186e-05, "loss": 0.33410236835479734, "memory(GiB)": 73.83, "step": 91820, "token_acc": 0.924187725631769, "train_speed(iter/s)": 1.458187 }, { "epoch": 3.9340645216571697, "grad_norm": 6.465167045593262, "learning_rate": 1.0801062909499526e-05, "loss": 0.5786396026611328, "memory(GiB)": 73.83, "step": 91825, "token_acc": 0.8996138996138996, "train_speed(iter/s)": 1.458181 }, { "epoch": 3.9342787369864185, "grad_norm": 3.4842307567596436, "learning_rate": 1.0796885504439391e-05, "loss": 0.42449150085449217, "memory(GiB)": 73.83, "step": 91830, "token_acc": 0.9169675090252708, "train_speed(iter/s)": 1.458191 }, { "epoch": 3.9344929523156678, "grad_norm": 2.196397066116333, "learning_rate": 1.0792708809580431e-05, "loss": 0.36484899520874026, "memory(GiB)": 73.83, "step": 91835, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.458196 }, { "epoch": 3.9347071676449166, "grad_norm": 4.147243976593018, "learning_rate": 1.0788532824998343e-05, "loss": 0.5388880729675293, "memory(GiB)": 73.83, "step": 91840, "token_acc": 0.8803680981595092, "train_speed(iter/s)": 1.458195 }, { "epoch": 3.9349213829741654, "grad_norm": 2.141339063644409, "learning_rate": 1.0784357550768753e-05, "loss": 0.4251612663269043, "memory(GiB)": 73.83, "step": 91845, "token_acc": 0.8970099667774086, "train_speed(iter/s)": 1.458197 }, { "epoch": 3.9351355983034146, "grad_norm": 5.231916427612305, "learning_rate": 1.078018298696733e-05, "loss": 0.4741776466369629, "memory(GiB)": 73.83, "step": 91850, "token_acc": 0.8865248226950354, "train_speed(iter/s)": 1.458209 }, { "epoch": 3.9353498136326635, "grad_norm": 2.660727024078369, "learning_rate": 1.0776009133669684e-05, "loss": 0.39154829978942873, "memory(GiB)": 73.83, "step": 91855, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.458209 }, { "epoch": 3.9355640289619123, "grad_norm": 3.7976598739624023, "learning_rate": 1.0771835990951424e-05, "loss": 0.4729097843170166, "memory(GiB)": 73.83, "step": 91860, "token_acc": 0.9049180327868852, "train_speed(iter/s)": 1.458209 }, { "epoch": 3.9357782442911615, "grad_norm": 7.704154014587402, "learning_rate": 1.0767663558888159e-05, "loss": 0.5629953384399414, "memory(GiB)": 73.83, "step": 91865, "token_acc": 0.8949579831932774, "train_speed(iter/s)": 1.45822 }, { "epoch": 3.9359924596204103, "grad_norm": 1.7600975036621094, "learning_rate": 1.076349183755545e-05, "loss": 0.11985470056533813, "memory(GiB)": 73.83, "step": 91870, "token_acc": 0.9772727272727273, "train_speed(iter/s)": 1.458226 }, { "epoch": 3.936206674949659, "grad_norm": 4.641913890838623, "learning_rate": 1.0759320827028913e-05, "loss": 0.22133755683898926, "memory(GiB)": 73.83, "step": 91875, "token_acc": 0.9471544715447154, "train_speed(iter/s)": 1.458229 }, { "epoch": 3.9364208902789084, "grad_norm": 4.311718940734863, "learning_rate": 1.0755150527384089e-05, "loss": 0.14660476446151732, "memory(GiB)": 73.83, "step": 91880, "token_acc": 0.9559322033898305, "train_speed(iter/s)": 1.458232 }, { "epoch": 3.936635105608157, "grad_norm": 3.0836613178253174, "learning_rate": 1.0750980938696525e-05, "loss": 0.1871357798576355, "memory(GiB)": 73.83, "step": 91885, "token_acc": 0.9493670886075949, "train_speed(iter/s)": 1.458229 }, { "epoch": 3.936849320937406, "grad_norm": 4.354429244995117, "learning_rate": 1.0746812061041761e-05, "loss": 0.3549583911895752, "memory(GiB)": 73.83, "step": 91890, "token_acc": 0.911864406779661, "train_speed(iter/s)": 1.45823 }, { "epoch": 3.9370635362666553, "grad_norm": 4.656513214111328, "learning_rate": 1.0742643894495319e-05, "loss": 0.35665230751037597, "memory(GiB)": 73.83, "step": 91895, "token_acc": 0.9078498293515358, "train_speed(iter/s)": 1.458239 }, { "epoch": 3.937277751595904, "grad_norm": 1.1920605897903442, "learning_rate": 1.07384764391327e-05, "loss": 0.2665479898452759, "memory(GiB)": 73.83, "step": 91900, "token_acc": 0.9527272727272728, "train_speed(iter/s)": 1.458244 }, { "epoch": 3.937491966925153, "grad_norm": 2.7819643020629883, "learning_rate": 1.0734309695029422e-05, "loss": 0.31835570335388186, "memory(GiB)": 73.83, "step": 91905, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.458251 }, { "epoch": 3.937706182254402, "grad_norm": 5.406821250915527, "learning_rate": 1.0730143662260938e-05, "loss": 0.5297123908996582, "memory(GiB)": 73.83, "step": 91910, "token_acc": 0.8838709677419355, "train_speed(iter/s)": 1.458252 }, { "epoch": 3.937920397583651, "grad_norm": 2.6077048778533936, "learning_rate": 1.0725978340902758e-05, "loss": 0.39276459217071535, "memory(GiB)": 73.83, "step": 91915, "token_acc": 0.9409722222222222, "train_speed(iter/s)": 1.458253 }, { "epoch": 3.9381346129129, "grad_norm": 3.0141048431396484, "learning_rate": 1.0721813731030328e-05, "loss": 0.4537950992584229, "memory(GiB)": 73.83, "step": 91920, "token_acc": 0.89375, "train_speed(iter/s)": 1.458257 }, { "epoch": 3.938348828242149, "grad_norm": 3.9463205337524414, "learning_rate": 1.071764983271908e-05, "loss": 0.4366273880004883, "memory(GiB)": 73.83, "step": 91925, "token_acc": 0.8918918918918919, "train_speed(iter/s)": 1.458258 }, { "epoch": 3.938563043571398, "grad_norm": 4.291007041931152, "learning_rate": 1.0713486646044446e-05, "loss": 0.25856590270996094, "memory(GiB)": 73.83, "step": 91930, "token_acc": 0.9495268138801262, "train_speed(iter/s)": 1.45826 }, { "epoch": 3.9387772589006467, "grad_norm": 7.359781742095947, "learning_rate": 1.0709324171081863e-05, "loss": 0.4291681289672852, "memory(GiB)": 73.83, "step": 91935, "token_acc": 0.9256965944272446, "train_speed(iter/s)": 1.458265 }, { "epoch": 3.938991474229896, "grad_norm": 3.5364532470703125, "learning_rate": 1.0705162407906739e-05, "loss": 0.5646511077880859, "memory(GiB)": 73.83, "step": 91940, "token_acc": 0.8773584905660378, "train_speed(iter/s)": 1.458273 }, { "epoch": 3.9392056895591447, "grad_norm": 3.812408447265625, "learning_rate": 1.0701001356594453e-05, "loss": 0.3190268039703369, "memory(GiB)": 73.83, "step": 91945, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.458272 }, { "epoch": 3.9394199048883936, "grad_norm": 4.0262603759765625, "learning_rate": 1.0696841017220394e-05, "loss": 0.577215051651001, "memory(GiB)": 73.83, "step": 91950, "token_acc": 0.9038461538461539, "train_speed(iter/s)": 1.458277 }, { "epoch": 3.939634120217643, "grad_norm": 4.692511558532715, "learning_rate": 1.0692681389859916e-05, "loss": 0.21852402687072753, "memory(GiB)": 73.83, "step": 91955, "token_acc": 0.9632352941176471, "train_speed(iter/s)": 1.458278 }, { "epoch": 3.9398483355468916, "grad_norm": 2.1052279472351074, "learning_rate": 1.0688522474588403e-05, "loss": 0.17277097702026367, "memory(GiB)": 73.83, "step": 91960, "token_acc": 0.961038961038961, "train_speed(iter/s)": 1.458284 }, { "epoch": 3.9400625508761404, "grad_norm": 2.2497546672821045, "learning_rate": 1.0684364271481184e-05, "loss": 0.3116642713546753, "memory(GiB)": 73.83, "step": 91965, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.458287 }, { "epoch": 3.9402767662053897, "grad_norm": 2.3484268188476562, "learning_rate": 1.0680206780613583e-05, "loss": 0.2698577880859375, "memory(GiB)": 73.83, "step": 91970, "token_acc": 0.9335260115606936, "train_speed(iter/s)": 1.458287 }, { "epoch": 3.9404909815346385, "grad_norm": 5.977365493774414, "learning_rate": 1.0676050002060922e-05, "loss": 0.5378568172454834, "memory(GiB)": 73.83, "step": 91975, "token_acc": 0.8758169934640523, "train_speed(iter/s)": 1.458286 }, { "epoch": 3.9407051968638873, "grad_norm": 4.692793846130371, "learning_rate": 1.0671893935898509e-05, "loss": 0.3439932823181152, "memory(GiB)": 73.83, "step": 91980, "token_acc": 0.9205607476635514, "train_speed(iter/s)": 1.45829 }, { "epoch": 3.9409194121931366, "grad_norm": 3.3791568279266357, "learning_rate": 1.0667738582201608e-05, "loss": 0.5396737098693848, "memory(GiB)": 73.83, "step": 91985, "token_acc": 0.8742331288343558, "train_speed(iter/s)": 1.458291 }, { "epoch": 3.9411336275223854, "grad_norm": 1.580499529838562, "learning_rate": 1.0663583941045525e-05, "loss": 0.3632943153381348, "memory(GiB)": 73.83, "step": 91990, "token_acc": 0.9125475285171103, "train_speed(iter/s)": 1.458294 }, { "epoch": 3.941347842851634, "grad_norm": 1.3945199251174927, "learning_rate": 1.0659430012505534e-05, "loss": 0.22328295707702636, "memory(GiB)": 73.83, "step": 91995, "token_acc": 0.9572192513368984, "train_speed(iter/s)": 1.458294 }, { "epoch": 3.9415620581808835, "grad_norm": 2.8940165042877197, "learning_rate": 1.0655276796656871e-05, "loss": 0.3357869625091553, "memory(GiB)": 73.83, "step": 92000, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.458295 }, { "epoch": 3.9415620581808835, "eval_loss": 2.466050148010254, "eval_runtime": 11.2912, "eval_samples_per_second": 8.856, "eval_steps_per_second": 8.856, "eval_token_acc": 0.44906166219839144, "step": 92000 }, { "epoch": 3.9417762735101323, "grad_norm": 0.307390958070755, "learning_rate": 1.0651124293574777e-05, "loss": 0.29802689552307127, "memory(GiB)": 73.83, "step": 92005, "token_acc": 0.5903041825095057, "train_speed(iter/s)": 1.458015 }, { "epoch": 3.941990488839381, "grad_norm": 5.098630905151367, "learning_rate": 1.0646972503334474e-05, "loss": 0.3144557476043701, "memory(GiB)": 73.83, "step": 92010, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.458014 }, { "epoch": 3.9422047041686303, "grad_norm": 4.086143493652344, "learning_rate": 1.0642821426011174e-05, "loss": 0.4205376625061035, "memory(GiB)": 73.83, "step": 92015, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.458014 }, { "epoch": 3.942418919497879, "grad_norm": 8.585102081298828, "learning_rate": 1.0638671061680095e-05, "loss": 0.2811316728591919, "memory(GiB)": 73.83, "step": 92020, "token_acc": 0.9395161290322581, "train_speed(iter/s)": 1.458019 }, { "epoch": 3.942633134827128, "grad_norm": 2.991922378540039, "learning_rate": 1.063452141041641e-05, "loss": 0.12433822154998779, "memory(GiB)": 73.83, "step": 92025, "token_acc": 0.9665551839464883, "train_speed(iter/s)": 1.45802 }, { "epoch": 3.942847350156377, "grad_norm": 2.5897274017333984, "learning_rate": 1.0630372472295303e-05, "loss": 0.27375576496124265, "memory(GiB)": 73.83, "step": 92030, "token_acc": 0.9489795918367347, "train_speed(iter/s)": 1.458029 }, { "epoch": 3.943061565485626, "grad_norm": 0.11095355451107025, "learning_rate": 1.0626224247391924e-05, "loss": 0.3405829668045044, "memory(GiB)": 73.83, "step": 92035, "token_acc": 0.9108108108108108, "train_speed(iter/s)": 1.458028 }, { "epoch": 3.943275780814875, "grad_norm": 1.7361444234848022, "learning_rate": 1.0622076735781433e-05, "loss": 0.3717683792114258, "memory(GiB)": 73.83, "step": 92040, "token_acc": 0.9148148148148149, "train_speed(iter/s)": 1.458029 }, { "epoch": 3.943489996144124, "grad_norm": 0.25377124547958374, "learning_rate": 1.0617929937538951e-05, "loss": 0.2572933673858643, "memory(GiB)": 73.83, "step": 92045, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.458027 }, { "epoch": 3.943704211473373, "grad_norm": 0.14430515468120575, "learning_rate": 1.0613783852739617e-05, "loss": 0.34719860553741455, "memory(GiB)": 73.83, "step": 92050, "token_acc": 0.9442622950819672, "train_speed(iter/s)": 1.458036 }, { "epoch": 3.9439184268026217, "grad_norm": 2.989487886428833, "learning_rate": 1.0609638481458545e-05, "loss": 0.30168392658233645, "memory(GiB)": 73.83, "step": 92055, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.45804 }, { "epoch": 3.944132642131871, "grad_norm": 5.5336384773254395, "learning_rate": 1.0605493823770801e-05, "loss": 0.43814849853515625, "memory(GiB)": 73.83, "step": 92060, "token_acc": 0.9158878504672897, "train_speed(iter/s)": 1.458047 }, { "epoch": 3.94434685746112, "grad_norm": 1.4845309257507324, "learning_rate": 1.060134987975151e-05, "loss": 0.223112154006958, "memory(GiB)": 73.83, "step": 92065, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.458051 }, { "epoch": 3.9445610727903686, "grad_norm": 6.505062103271484, "learning_rate": 1.0597206649475721e-05, "loss": 0.44588298797607423, "memory(GiB)": 73.83, "step": 92070, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.458051 }, { "epoch": 3.944775288119618, "grad_norm": 4.1430559158325195, "learning_rate": 1.0593064133018488e-05, "loss": 0.2278979778289795, "memory(GiB)": 73.83, "step": 92075, "token_acc": 0.9572953736654805, "train_speed(iter/s)": 1.458051 }, { "epoch": 3.9449895034488667, "grad_norm": 1.8072991371154785, "learning_rate": 1.0588922330454876e-05, "loss": 0.4149333000183105, "memory(GiB)": 73.83, "step": 92080, "token_acc": 0.9196428571428571, "train_speed(iter/s)": 1.458054 }, { "epoch": 3.9452037187781155, "grad_norm": 1.2612733840942383, "learning_rate": 1.0584781241859903e-05, "loss": 0.1928694248199463, "memory(GiB)": 73.83, "step": 92085, "token_acc": 0.9694656488549618, "train_speed(iter/s)": 1.458055 }, { "epoch": 3.9454179341073647, "grad_norm": 2.9599416255950928, "learning_rate": 1.05806408673086e-05, "loss": 0.34561135768890383, "memory(GiB)": 73.83, "step": 92090, "token_acc": 0.9246031746031746, "train_speed(iter/s)": 1.45806 }, { "epoch": 3.9456321494366136, "grad_norm": 2.3018686771392822, "learning_rate": 1.057650120687596e-05, "loss": 0.374979567527771, "memory(GiB)": 73.83, "step": 92095, "token_acc": 0.9258064516129032, "train_speed(iter/s)": 1.458061 }, { "epoch": 3.9458463647658624, "grad_norm": 3.81473708152771, "learning_rate": 1.057236226063697e-05, "loss": 0.22179090976715088, "memory(GiB)": 73.83, "step": 92100, "token_acc": 0.9432835820895522, "train_speed(iter/s)": 1.458083 }, { "epoch": 3.9460605800951116, "grad_norm": 2.850142240524292, "learning_rate": 1.056822402866664e-05, "loss": 0.34001023769378663, "memory(GiB)": 73.83, "step": 92105, "token_acc": 0.9171974522292994, "train_speed(iter/s)": 1.458091 }, { "epoch": 3.9462747954243604, "grad_norm": 5.859430313110352, "learning_rate": 1.0564086511039917e-05, "loss": 0.5409473419189453, "memory(GiB)": 73.83, "step": 92110, "token_acc": 0.887240356083086, "train_speed(iter/s)": 1.458103 }, { "epoch": 3.9464890107536097, "grad_norm": 2.4363861083984375, "learning_rate": 1.0559949707831762e-05, "loss": 0.4343570709228516, "memory(GiB)": 73.83, "step": 92115, "token_acc": 0.9206798866855525, "train_speed(iter/s)": 1.458107 }, { "epoch": 3.9467032260828585, "grad_norm": 4.017660140991211, "learning_rate": 1.0555813619117122e-05, "loss": 0.30574536323547363, "memory(GiB)": 73.83, "step": 92120, "token_acc": 0.9402390438247012, "train_speed(iter/s)": 1.458118 }, { "epoch": 3.9469174414121073, "grad_norm": 3.7990074157714844, "learning_rate": 1.0551678244970914e-05, "loss": 0.32439794540405276, "memory(GiB)": 73.83, "step": 92125, "token_acc": 0.9307692307692308, "train_speed(iter/s)": 1.458124 }, { "epoch": 3.9471316567413566, "grad_norm": 3.5529696941375732, "learning_rate": 1.0547543585468044e-05, "loss": 0.41310648918151854, "memory(GiB)": 73.83, "step": 92130, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.458136 }, { "epoch": 3.9473458720706054, "grad_norm": 2.724929094314575, "learning_rate": 1.0543409640683439e-05, "loss": 0.17263193130493165, "memory(GiB)": 73.83, "step": 92135, "token_acc": 0.948, "train_speed(iter/s)": 1.458149 }, { "epoch": 3.947560087399854, "grad_norm": 3.05900502204895, "learning_rate": 1.053927641069199e-05, "loss": 0.36132285594940183, "memory(GiB)": 73.83, "step": 92140, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.458158 }, { "epoch": 3.9477743027291035, "grad_norm": 1.6091015338897705, "learning_rate": 1.0535143895568566e-05, "loss": 0.2578601360321045, "memory(GiB)": 73.83, "step": 92145, "token_acc": 0.9530386740331491, "train_speed(iter/s)": 1.458159 }, { "epoch": 3.9479885180583523, "grad_norm": 2.392016649246216, "learning_rate": 1.0531012095388032e-05, "loss": 0.34826323986053465, "memory(GiB)": 73.83, "step": 92150, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.458162 }, { "epoch": 3.948202733387601, "grad_norm": 2.8935139179229736, "learning_rate": 1.0526881010225242e-05, "loss": 0.1045226812362671, "memory(GiB)": 73.83, "step": 92155, "token_acc": 0.9783393501805054, "train_speed(iter/s)": 1.458165 }, { "epoch": 3.9484169487168503, "grad_norm": 2.8860890865325928, "learning_rate": 1.0522750640155016e-05, "loss": 0.2406538486480713, "memory(GiB)": 73.83, "step": 92160, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.458166 }, { "epoch": 3.948631164046099, "grad_norm": 6.720494270324707, "learning_rate": 1.0518620985252208e-05, "loss": 0.4827274322509766, "memory(GiB)": 73.83, "step": 92165, "token_acc": 0.9263565891472868, "train_speed(iter/s)": 1.458162 }, { "epoch": 3.948845379375348, "grad_norm": 0.24428491294384003, "learning_rate": 1.0514492045591617e-05, "loss": 0.1384440302848816, "memory(GiB)": 73.83, "step": 92170, "token_acc": 0.9692832764505119, "train_speed(iter/s)": 1.458162 }, { "epoch": 3.949059594704597, "grad_norm": 3.1829946041107178, "learning_rate": 1.0510363821248042e-05, "loss": 0.24751641750335693, "memory(GiB)": 73.83, "step": 92175, "token_acc": 0.951310861423221, "train_speed(iter/s)": 1.458162 }, { "epoch": 3.949273810033846, "grad_norm": 1.6949812173843384, "learning_rate": 1.050623631229627e-05, "loss": 0.3307502031326294, "memory(GiB)": 73.83, "step": 92180, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.458172 }, { "epoch": 3.949488025363095, "grad_norm": 6.001053810119629, "learning_rate": 1.0502109518811065e-05, "loss": 0.45359277725219727, "memory(GiB)": 73.83, "step": 92185, "token_acc": 0.9158249158249159, "train_speed(iter/s)": 1.458175 }, { "epoch": 3.949702240692344, "grad_norm": 3.8597371578216553, "learning_rate": 1.049798344086721e-05, "loss": 0.344010066986084, "memory(GiB)": 73.83, "step": 92190, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.458174 }, { "epoch": 3.949916456021593, "grad_norm": 0.40956366062164307, "learning_rate": 1.0493858078539442e-05, "loss": 0.27058389186859133, "memory(GiB)": 73.83, "step": 92195, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.458181 }, { "epoch": 3.9501306713508417, "grad_norm": 5.368690490722656, "learning_rate": 1.0489733431902494e-05, "loss": 0.3126669406890869, "memory(GiB)": 73.83, "step": 92200, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.458181 }, { "epoch": 3.950344886680091, "grad_norm": 2.5240182876586914, "learning_rate": 1.0485609501031074e-05, "loss": 0.30764193534851075, "memory(GiB)": 73.83, "step": 92205, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.458191 }, { "epoch": 3.95055910200934, "grad_norm": 5.936232089996338, "learning_rate": 1.048148628599992e-05, "loss": 0.542138147354126, "memory(GiB)": 73.83, "step": 92210, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.458191 }, { "epoch": 3.9507733173385886, "grad_norm": 2.831481695175171, "learning_rate": 1.0477363786883709e-05, "loss": 0.4616283416748047, "memory(GiB)": 73.83, "step": 92215, "token_acc": 0.9035714285714286, "train_speed(iter/s)": 1.4582 }, { "epoch": 3.950987532667838, "grad_norm": 5.454102993011475, "learning_rate": 1.0473242003757123e-05, "loss": 0.36570186614990235, "memory(GiB)": 73.83, "step": 92220, "token_acc": 0.925531914893617, "train_speed(iter/s)": 1.458201 }, { "epoch": 3.9512017479970867, "grad_norm": 4.384995937347412, "learning_rate": 1.0469120936694843e-05, "loss": 0.3359992027282715, "memory(GiB)": 73.83, "step": 92225, "token_acc": 0.9138576779026217, "train_speed(iter/s)": 1.458211 }, { "epoch": 3.9514159633263355, "grad_norm": 3.4364101886749268, "learning_rate": 1.0465000585771522e-05, "loss": 0.3610379219055176, "memory(GiB)": 73.83, "step": 92230, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.458212 }, { "epoch": 3.9516301786555847, "grad_norm": 0.7378700375556946, "learning_rate": 1.0460880951061807e-05, "loss": 0.15783169269561767, "memory(GiB)": 73.83, "step": 92235, "token_acc": 0.9544072948328267, "train_speed(iter/s)": 1.458217 }, { "epoch": 3.9518443939848336, "grad_norm": 6.650496006011963, "learning_rate": 1.0456762032640322e-05, "loss": 0.3515087842941284, "memory(GiB)": 73.83, "step": 92240, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.458219 }, { "epoch": 3.9520586093140824, "grad_norm": 4.153378009796143, "learning_rate": 1.0452643830581672e-05, "loss": 0.3619898557662964, "memory(GiB)": 73.83, "step": 92245, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.458233 }, { "epoch": 3.9522728246433316, "grad_norm": 0.2780580520629883, "learning_rate": 1.0448526344960491e-05, "loss": 0.604233980178833, "memory(GiB)": 73.83, "step": 92250, "token_acc": 0.8972332015810277, "train_speed(iter/s)": 1.458237 }, { "epoch": 3.9524870399725804, "grad_norm": 2.925341844558716, "learning_rate": 1.0444409575851355e-05, "loss": 0.4277377128601074, "memory(GiB)": 73.83, "step": 92255, "token_acc": 0.8907849829351536, "train_speed(iter/s)": 1.458241 }, { "epoch": 3.9527012553018293, "grad_norm": 1.5810261964797974, "learning_rate": 1.0440293523328847e-05, "loss": 0.33086304664611815, "memory(GiB)": 73.83, "step": 92260, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.458246 }, { "epoch": 3.9529154706310785, "grad_norm": 6.718324661254883, "learning_rate": 1.0436178187467538e-05, "loss": 0.7319259643554688, "memory(GiB)": 73.83, "step": 92265, "token_acc": 0.8507462686567164, "train_speed(iter/s)": 1.458254 }, { "epoch": 3.9531296859603273, "grad_norm": 9.120609283447266, "learning_rate": 1.043206356834197e-05, "loss": 0.5497824668884277, "memory(GiB)": 73.83, "step": 92270, "token_acc": 0.9105058365758755, "train_speed(iter/s)": 1.458259 }, { "epoch": 3.953343901289576, "grad_norm": 0.6339030265808105, "learning_rate": 1.0427949666026677e-05, "loss": 0.5640312194824219, "memory(GiB)": 73.83, "step": 92275, "token_acc": 0.8901960784313725, "train_speed(iter/s)": 1.45827 }, { "epoch": 3.9535581166188254, "grad_norm": 3.5111119747161865, "learning_rate": 1.0423836480596195e-05, "loss": 0.5866907596588135, "memory(GiB)": 73.83, "step": 92280, "token_acc": 0.8774928774928775, "train_speed(iter/s)": 1.458271 }, { "epoch": 3.953772331948074, "grad_norm": 3.15960693359375, "learning_rate": 1.0419724012125053e-05, "loss": 0.650423812866211, "memory(GiB)": 73.83, "step": 92285, "token_acc": 0.8736842105263158, "train_speed(iter/s)": 1.458272 }, { "epoch": 3.953986547277323, "grad_norm": 2.4097793102264404, "learning_rate": 1.041561226068774e-05, "loss": 0.45790901184082033, "memory(GiB)": 73.83, "step": 92290, "token_acc": 0.903954802259887, "train_speed(iter/s)": 1.458276 }, { "epoch": 3.9542007626065723, "grad_norm": 4.113375663757324, "learning_rate": 1.041150122635875e-05, "loss": 0.33031044006347654, "memory(GiB)": 73.83, "step": 92295, "token_acc": 0.9145569620253164, "train_speed(iter/s)": 1.458281 }, { "epoch": 3.954414977935821, "grad_norm": 3.629880905151367, "learning_rate": 1.0407390909212544e-05, "loss": 0.48749375343322754, "memory(GiB)": 73.83, "step": 92300, "token_acc": 0.9050847457627119, "train_speed(iter/s)": 1.458284 }, { "epoch": 3.95462919326507, "grad_norm": 7.1313910484313965, "learning_rate": 1.0403281309323587e-05, "loss": 0.7189833641052246, "memory(GiB)": 73.83, "step": 92305, "token_acc": 0.7965517241379311, "train_speed(iter/s)": 1.45827 }, { "epoch": 3.954843408594319, "grad_norm": 2.8810789585113525, "learning_rate": 1.0399172426766341e-05, "loss": 0.2228269100189209, "memory(GiB)": 73.83, "step": 92310, "token_acc": 0.9441340782122905, "train_speed(iter/s)": 1.458274 }, { "epoch": 3.955057623923568, "grad_norm": 10.447203636169434, "learning_rate": 1.039506426161524e-05, "loss": 0.3526034116744995, "memory(GiB)": 73.83, "step": 92315, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.458273 }, { "epoch": 3.955271839252817, "grad_norm": 1.7845966815948486, "learning_rate": 1.0390956813944697e-05, "loss": 0.18762779235839844, "memory(GiB)": 73.83, "step": 92320, "token_acc": 0.9575971731448764, "train_speed(iter/s)": 1.458274 }, { "epoch": 3.955486054582066, "grad_norm": 4.2458271980285645, "learning_rate": 1.038685008382913e-05, "loss": 0.3321269750595093, "memory(GiB)": 73.83, "step": 92325, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.458286 }, { "epoch": 3.955700269911315, "grad_norm": 3.0590806007385254, "learning_rate": 1.038274407134292e-05, "loss": 0.2685011625289917, "memory(GiB)": 73.83, "step": 92330, "token_acc": 0.934375, "train_speed(iter/s)": 1.458286 }, { "epoch": 3.9559144852405637, "grad_norm": 1.9036005735397339, "learning_rate": 1.0378638776560483e-05, "loss": 0.4175253391265869, "memory(GiB)": 73.83, "step": 92335, "token_acc": 0.9051094890510949, "train_speed(iter/s)": 1.458288 }, { "epoch": 3.956128700569813, "grad_norm": 0.16929462552070618, "learning_rate": 1.0374534199556168e-05, "loss": 0.25741019248962405, "memory(GiB)": 73.83, "step": 92340, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.458293 }, { "epoch": 3.9563429158990617, "grad_norm": 1.0377558469772339, "learning_rate": 1.0370430340404336e-05, "loss": 0.2893967628479004, "memory(GiB)": 73.83, "step": 92345, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.458297 }, { "epoch": 3.9565571312283105, "grad_norm": 4.4159770011901855, "learning_rate": 1.0366327199179338e-05, "loss": 0.3498523712158203, "memory(GiB)": 73.83, "step": 92350, "token_acc": 0.9372549019607843, "train_speed(iter/s)": 1.458301 }, { "epoch": 3.95677134655756, "grad_norm": 1.5684224367141724, "learning_rate": 1.0362224775955486e-05, "loss": 0.25644793510437014, "memory(GiB)": 73.83, "step": 92355, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.458312 }, { "epoch": 3.9569855618868086, "grad_norm": 1.570952296257019, "learning_rate": 1.035812307080713e-05, "loss": 0.2928192138671875, "memory(GiB)": 73.83, "step": 92360, "token_acc": 0.9329608938547486, "train_speed(iter/s)": 1.458327 }, { "epoch": 3.9571997772160574, "grad_norm": 4.223345756530762, "learning_rate": 1.0354022083808551e-05, "loss": 0.4730949401855469, "memory(GiB)": 73.83, "step": 92365, "token_acc": 0.8900709219858156, "train_speed(iter/s)": 1.45833 }, { "epoch": 3.9574139925453067, "grad_norm": 5.907159805297852, "learning_rate": 1.0349921815034063e-05, "loss": 0.663557767868042, "memory(GiB)": 73.83, "step": 92370, "token_acc": 0.8656716417910447, "train_speed(iter/s)": 1.45833 }, { "epoch": 3.9576282078745555, "grad_norm": 2.916459083557129, "learning_rate": 1.0345822264557936e-05, "loss": 0.2770711898803711, "memory(GiB)": 73.83, "step": 92375, "token_acc": 0.9398496240601504, "train_speed(iter/s)": 1.458337 }, { "epoch": 3.9578424232038043, "grad_norm": 4.338794708251953, "learning_rate": 1.0341723432454437e-05, "loss": 0.8423773765563964, "memory(GiB)": 73.83, "step": 92380, "token_acc": 0.8118466898954704, "train_speed(iter/s)": 1.458339 }, { "epoch": 3.9580566385330536, "grad_norm": 5.849497318267822, "learning_rate": 1.0337625318797822e-05, "loss": 0.3981311559677124, "memory(GiB)": 73.83, "step": 92385, "token_acc": 0.9074733096085409, "train_speed(iter/s)": 1.458351 }, { "epoch": 3.9582708538623024, "grad_norm": 3.1614551544189453, "learning_rate": 1.0333527923662318e-05, "loss": 0.37824230194091796, "memory(GiB)": 73.83, "step": 92390, "token_acc": 0.9198717948717948, "train_speed(iter/s)": 1.458354 }, { "epoch": 3.958485069191551, "grad_norm": 3.61651349067688, "learning_rate": 1.0329431247122179e-05, "loss": 0.35536966323852537, "memory(GiB)": 73.83, "step": 92395, "token_acc": 0.927536231884058, "train_speed(iter/s)": 1.45836 }, { "epoch": 3.9586992845208004, "grad_norm": 1.8378973007202148, "learning_rate": 1.0325335289251602e-05, "loss": 0.4828056335449219, "memory(GiB)": 73.83, "step": 92400, "token_acc": 0.8815331010452961, "train_speed(iter/s)": 1.458364 }, { "epoch": 3.9589134998500493, "grad_norm": 3.6257519721984863, "learning_rate": 1.0321240050124797e-05, "loss": 0.3946316957473755, "memory(GiB)": 73.83, "step": 92405, "token_acc": 0.905829596412556, "train_speed(iter/s)": 1.458379 }, { "epoch": 3.959127715179298, "grad_norm": 4.813309669494629, "learning_rate": 1.0317145529815952e-05, "loss": 0.2569340705871582, "memory(GiB)": 73.83, "step": 92410, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.458381 }, { "epoch": 3.9593419305085473, "grad_norm": 2.7887043952941895, "learning_rate": 1.0313051728399243e-05, "loss": 0.18359026908874512, "memory(GiB)": 73.83, "step": 92415, "token_acc": 0.9695945945945946, "train_speed(iter/s)": 1.458389 }, { "epoch": 3.959556145837796, "grad_norm": 2.6403355598449707, "learning_rate": 1.0308958645948814e-05, "loss": 0.21345221996307373, "memory(GiB)": 73.83, "step": 92420, "token_acc": 0.9400921658986175, "train_speed(iter/s)": 1.458404 }, { "epoch": 3.959770361167045, "grad_norm": 5.237696170806885, "learning_rate": 1.030486628253885e-05, "loss": 0.5520557403564453, "memory(GiB)": 73.83, "step": 92425, "token_acc": 0.89171974522293, "train_speed(iter/s)": 1.458404 }, { "epoch": 3.959984576496294, "grad_norm": 3.1873974800109863, "learning_rate": 1.0300774638243454e-05, "loss": 0.40987114906311034, "memory(GiB)": 73.83, "step": 92430, "token_acc": 0.906832298136646, "train_speed(iter/s)": 1.458406 }, { "epoch": 3.960198791825543, "grad_norm": 2.1203501224517822, "learning_rate": 1.0296683713136779e-05, "loss": 0.2407849073410034, "memory(GiB)": 73.83, "step": 92435, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.458416 }, { "epoch": 3.960413007154792, "grad_norm": 2.935182571411133, "learning_rate": 1.0292593507292924e-05, "loss": 0.4136331558227539, "memory(GiB)": 73.83, "step": 92440, "token_acc": 0.9077490774907749, "train_speed(iter/s)": 1.458427 }, { "epoch": 3.960627222484041, "grad_norm": 4.088423728942871, "learning_rate": 1.0288504020785983e-05, "loss": 0.3175637245178223, "memory(GiB)": 73.83, "step": 92445, "token_acc": 0.9261538461538461, "train_speed(iter/s)": 1.458428 }, { "epoch": 3.96084143781329, "grad_norm": 2.262399911880493, "learning_rate": 1.0284415253690032e-05, "loss": 0.48650393486022947, "memory(GiB)": 73.83, "step": 92450, "token_acc": 0.9, "train_speed(iter/s)": 1.458425 }, { "epoch": 3.961055653142539, "grad_norm": 1.811157464981079, "learning_rate": 1.0280327206079172e-05, "loss": 0.3032386064529419, "memory(GiB)": 73.83, "step": 92455, "token_acc": 0.9380530973451328, "train_speed(iter/s)": 1.458428 }, { "epoch": 3.961269868471788, "grad_norm": 0.47404035925865173, "learning_rate": 1.0276239878027443e-05, "loss": 0.19175275564193725, "memory(GiB)": 73.83, "step": 92460, "token_acc": 0.95578231292517, "train_speed(iter/s)": 1.45844 }, { "epoch": 3.961484083801037, "grad_norm": 7.246684551239014, "learning_rate": 1.0272153269608892e-05, "loss": 0.6605988502502441, "memory(GiB)": 73.83, "step": 92465, "token_acc": 0.8601823708206687, "train_speed(iter/s)": 1.45846 }, { "epoch": 3.961698299130286, "grad_norm": 1.4843446016311646, "learning_rate": 1.0268067380897556e-05, "loss": 0.4642601490020752, "memory(GiB)": 73.83, "step": 92470, "token_acc": 0.9078212290502793, "train_speed(iter/s)": 1.458455 }, { "epoch": 3.961912514459535, "grad_norm": 1.6381276845932007, "learning_rate": 1.0263982211967433e-05, "loss": 0.2150951623916626, "memory(GiB)": 73.83, "step": 92475, "token_acc": 0.9498432601880877, "train_speed(iter/s)": 1.458456 }, { "epoch": 3.9621267297887837, "grad_norm": 2.506582021713257, "learning_rate": 1.0259897762892562e-05, "loss": 0.3798380374908447, "memory(GiB)": 73.83, "step": 92480, "token_acc": 0.9358490566037736, "train_speed(iter/s)": 1.458464 }, { "epoch": 3.962340945118033, "grad_norm": 2.2125155925750732, "learning_rate": 1.0255814033746925e-05, "loss": 0.25460176467895507, "memory(GiB)": 73.83, "step": 92485, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.45847 }, { "epoch": 3.9625551604472817, "grad_norm": 2.380335569381714, "learning_rate": 1.0251731024604495e-05, "loss": 0.3954993009567261, "memory(GiB)": 73.83, "step": 92490, "token_acc": 0.9184397163120568, "train_speed(iter/s)": 1.458469 }, { "epoch": 3.9627693757765305, "grad_norm": 0.6796833276748657, "learning_rate": 1.0247648735539245e-05, "loss": 0.21683621406555176, "memory(GiB)": 73.83, "step": 92495, "token_acc": 0.9563758389261745, "train_speed(iter/s)": 1.458484 }, { "epoch": 3.96298359110578, "grad_norm": 3.916163206100464, "learning_rate": 1.0243567166625117e-05, "loss": 0.24865176677703857, "memory(GiB)": 73.83, "step": 92500, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.458495 }, { "epoch": 3.96298359110578, "eval_loss": 2.3006350994110107, "eval_runtime": 11.4307, "eval_samples_per_second": 8.748, "eval_steps_per_second": 8.748, "eval_token_acc": 0.4518950437317784, "step": 92500 }, { "epoch": 3.9631978064350286, "grad_norm": 3.208667755126953, "learning_rate": 1.0239486317936081e-05, "loss": 0.5928923130035401, "memory(GiB)": 73.83, "step": 92505, "token_acc": 0.6035120147874307, "train_speed(iter/s)": 1.458221 }, { "epoch": 3.9634120217642774, "grad_norm": 5.235823631286621, "learning_rate": 1.0235406189546031e-05, "loss": 0.3054505348205566, "memory(GiB)": 73.83, "step": 92510, "token_acc": 0.9267399267399268, "train_speed(iter/s)": 1.458225 }, { "epoch": 3.9636262370935267, "grad_norm": 3.298004150390625, "learning_rate": 1.0231326781528916e-05, "loss": 0.10134743452072144, "memory(GiB)": 73.83, "step": 92515, "token_acc": 0.9698996655518395, "train_speed(iter/s)": 1.458228 }, { "epoch": 3.9638404524227755, "grad_norm": 5.271418571472168, "learning_rate": 1.022724809395862e-05, "loss": 0.4250916004180908, "memory(GiB)": 73.83, "step": 92520, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.458233 }, { "epoch": 3.9640546677520243, "grad_norm": 0.19133611023426056, "learning_rate": 1.0223170126909038e-05, "loss": 0.11166917085647583, "memory(GiB)": 73.83, "step": 92525, "token_acc": 0.9733333333333334, "train_speed(iter/s)": 1.458235 }, { "epoch": 3.9642688830812736, "grad_norm": 0.601818323135376, "learning_rate": 1.021909288045404e-05, "loss": 0.28193514347076415, "memory(GiB)": 73.83, "step": 92530, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.458243 }, { "epoch": 3.9644830984105224, "grad_norm": 5.3390793800354, "learning_rate": 1.0215016354667477e-05, "loss": 0.44277005195617675, "memory(GiB)": 73.83, "step": 92535, "token_acc": 0.8963210702341137, "train_speed(iter/s)": 1.458251 }, { "epoch": 3.964697313739771, "grad_norm": 2.647008180618286, "learning_rate": 1.0210940549623227e-05, "loss": 0.18514593839645385, "memory(GiB)": 73.83, "step": 92540, "token_acc": 0.9615384615384616, "train_speed(iter/s)": 1.458249 }, { "epoch": 3.9649115290690204, "grad_norm": 2.945587396621704, "learning_rate": 1.020686546539512e-05, "loss": 0.25972685813903806, "memory(GiB)": 73.83, "step": 92545, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.458253 }, { "epoch": 3.9651257443982693, "grad_norm": 1.7511917352676392, "learning_rate": 1.0202791102056969e-05, "loss": 0.15670106410980225, "memory(GiB)": 73.83, "step": 92550, "token_acc": 0.9679715302491103, "train_speed(iter/s)": 1.458257 }, { "epoch": 3.965339959727518, "grad_norm": 1.1686538457870483, "learning_rate": 1.0198717459682594e-05, "loss": 0.17904587984085082, "memory(GiB)": 73.83, "step": 92555, "token_acc": 0.9568965517241379, "train_speed(iter/s)": 1.458264 }, { "epoch": 3.9655541750567673, "grad_norm": 1.8520727157592773, "learning_rate": 1.0194644538345787e-05, "loss": 0.1343826174736023, "memory(GiB)": 73.83, "step": 92560, "token_acc": 0.9635627530364372, "train_speed(iter/s)": 1.458272 }, { "epoch": 3.965768390386016, "grad_norm": 0.33481818437576294, "learning_rate": 1.0190572338120318e-05, "loss": 0.4152646064758301, "memory(GiB)": 73.83, "step": 92565, "token_acc": 0.9149659863945578, "train_speed(iter/s)": 1.458276 }, { "epoch": 3.965982605715265, "grad_norm": 3.1921305656433105, "learning_rate": 1.0186500859079995e-05, "loss": 0.384043025970459, "memory(GiB)": 73.83, "step": 92570, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.45828 }, { "epoch": 3.966196821044514, "grad_norm": 3.5475127696990967, "learning_rate": 1.018243010129854e-05, "loss": 0.35916838645935056, "memory(GiB)": 73.83, "step": 92575, "token_acc": 0.9297520661157025, "train_speed(iter/s)": 1.458281 }, { "epoch": 3.966411036373763, "grad_norm": 2.098214864730835, "learning_rate": 1.017836006484973e-05, "loss": 0.39766950607299806, "memory(GiB)": 73.83, "step": 92580, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.458281 }, { "epoch": 3.966625251703012, "grad_norm": 4.819207191467285, "learning_rate": 1.0174290749807281e-05, "loss": 0.3205896854400635, "memory(GiB)": 73.83, "step": 92585, "token_acc": 0.9105691056910569, "train_speed(iter/s)": 1.458281 }, { "epoch": 3.966839467032261, "grad_norm": 0.5328640341758728, "learning_rate": 1.0170222156244914e-05, "loss": 0.1851873517036438, "memory(GiB)": 73.83, "step": 92590, "token_acc": 0.9520958083832335, "train_speed(iter/s)": 1.458286 }, { "epoch": 3.96705368236151, "grad_norm": 4.5671234130859375, "learning_rate": 1.0166154284236324e-05, "loss": 0.40915713310241697, "memory(GiB)": 73.83, "step": 92595, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.458292 }, { "epoch": 3.9672678976907587, "grad_norm": 2.4836487770080566, "learning_rate": 1.016208713385523e-05, "loss": 0.39462316036224365, "memory(GiB)": 73.83, "step": 92600, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.458302 }, { "epoch": 3.967482113020008, "grad_norm": 1.6868109703063965, "learning_rate": 1.0158020705175298e-05, "loss": 0.34497761726379395, "memory(GiB)": 73.83, "step": 92605, "token_acc": 0.9135802469135802, "train_speed(iter/s)": 1.458314 }, { "epoch": 3.967696328349257, "grad_norm": 1.3532942533493042, "learning_rate": 1.01539549982702e-05, "loss": 0.3855916976928711, "memory(GiB)": 73.83, "step": 92610, "token_acc": 0.9127272727272727, "train_speed(iter/s)": 1.458332 }, { "epoch": 3.9679105436785056, "grad_norm": 0.6402592658996582, "learning_rate": 1.014989001321358e-05, "loss": 0.1783931612968445, "memory(GiB)": 73.83, "step": 92615, "token_acc": 0.9560439560439561, "train_speed(iter/s)": 1.458332 }, { "epoch": 3.968124759007755, "grad_norm": 3.7956337928771973, "learning_rate": 1.0145825750079075e-05, "loss": 0.6137558937072753, "memory(GiB)": 73.83, "step": 92620, "token_acc": 0.8487972508591065, "train_speed(iter/s)": 1.458336 }, { "epoch": 3.9683389743370037, "grad_norm": 3.8072266578674316, "learning_rate": 1.0141762208940337e-05, "loss": 0.42281455993652345, "memory(GiB)": 73.83, "step": 92625, "token_acc": 0.9118942731277533, "train_speed(iter/s)": 1.458342 }, { "epoch": 3.9685531896662525, "grad_norm": 2.0093886852264404, "learning_rate": 1.0137699389870963e-05, "loss": 0.30936102867126464, "memory(GiB)": 73.83, "step": 92630, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.458346 }, { "epoch": 3.9687674049955017, "grad_norm": 1.1863210201263428, "learning_rate": 1.013363729294456e-05, "loss": 0.257472562789917, "memory(GiB)": 73.83, "step": 92635, "token_acc": 0.9477124183006536, "train_speed(iter/s)": 1.458351 }, { "epoch": 3.9689816203247505, "grad_norm": 0.5442156195640564, "learning_rate": 1.0129575918234713e-05, "loss": 0.4649616241455078, "memory(GiB)": 73.83, "step": 92640, "token_acc": 0.9111969111969112, "train_speed(iter/s)": 1.458351 }, { "epoch": 3.9691958356539994, "grad_norm": 2.5520901679992676, "learning_rate": 1.0125515265815005e-05, "loss": 0.36203484535217284, "memory(GiB)": 73.83, "step": 92645, "token_acc": 0.9250814332247557, "train_speed(iter/s)": 1.458355 }, { "epoch": 3.9694100509832486, "grad_norm": 3.8289952278137207, "learning_rate": 1.0121455335758978e-05, "loss": 0.5139838218688965, "memory(GiB)": 73.83, "step": 92650, "token_acc": 0.8821548821548821, "train_speed(iter/s)": 1.45836 }, { "epoch": 3.9696242663124974, "grad_norm": 2.704388380050659, "learning_rate": 1.0117396128140194e-05, "loss": 0.3351579189300537, "memory(GiB)": 73.83, "step": 92655, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.458365 }, { "epoch": 3.9698384816417462, "grad_norm": 0.364583820104599, "learning_rate": 1.0113337643032206e-05, "loss": 0.20801684856414795, "memory(GiB)": 73.83, "step": 92660, "token_acc": 0.9576271186440678, "train_speed(iter/s)": 1.458369 }, { "epoch": 3.9700526969709955, "grad_norm": 4.954829216003418, "learning_rate": 1.0109279880508527e-05, "loss": 0.3350972175598145, "memory(GiB)": 73.83, "step": 92665, "token_acc": 0.9366197183098591, "train_speed(iter/s)": 1.45838 }, { "epoch": 3.9702669123002443, "grad_norm": 1.769304633140564, "learning_rate": 1.010522284064266e-05, "loss": 0.3373068571090698, "memory(GiB)": 73.83, "step": 92670, "token_acc": 0.9131832797427653, "train_speed(iter/s)": 1.458385 }, { "epoch": 3.970481127629493, "grad_norm": 1.8145276308059692, "learning_rate": 1.0101166523508104e-05, "loss": 0.22074730396270753, "memory(GiB)": 73.83, "step": 92675, "token_acc": 0.955719557195572, "train_speed(iter/s)": 1.458386 }, { "epoch": 3.9706953429587424, "grad_norm": 1.923877239227295, "learning_rate": 1.0097110929178332e-05, "loss": 0.35237860679626465, "memory(GiB)": 73.83, "step": 92680, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.458395 }, { "epoch": 3.970909558287991, "grad_norm": 3.722052574157715, "learning_rate": 1.009305605772684e-05, "loss": 0.34524056911468504, "memory(GiB)": 73.83, "step": 92685, "token_acc": 0.9144736842105263, "train_speed(iter/s)": 1.458398 }, { "epoch": 3.97112377361724, "grad_norm": 5.170253276824951, "learning_rate": 1.008900190922708e-05, "loss": 0.520012092590332, "memory(GiB)": 73.83, "step": 92690, "token_acc": 0.8977635782747604, "train_speed(iter/s)": 1.458398 }, { "epoch": 3.9713379889464893, "grad_norm": 3.830428123474121, "learning_rate": 1.0084948483752482e-05, "loss": 0.29186277389526366, "memory(GiB)": 73.83, "step": 92695, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.458402 }, { "epoch": 3.971552204275738, "grad_norm": 0.20663076639175415, "learning_rate": 1.0080895781376493e-05, "loss": 0.3154005765914917, "memory(GiB)": 73.83, "step": 92700, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.458408 }, { "epoch": 3.971766419604987, "grad_norm": 2.6747145652770996, "learning_rate": 1.007684380217252e-05, "loss": 0.13770754337310792, "memory(GiB)": 73.83, "step": 92705, "token_acc": 0.9591078066914498, "train_speed(iter/s)": 1.458412 }, { "epoch": 3.971980634934236, "grad_norm": 2.1820173263549805, "learning_rate": 1.007279254621396e-05, "loss": 0.41775174140930177, "memory(GiB)": 73.83, "step": 92710, "token_acc": 0.9018987341772152, "train_speed(iter/s)": 1.458413 }, { "epoch": 3.972194850263485, "grad_norm": 6.143526077270508, "learning_rate": 1.0068742013574229e-05, "loss": 0.5782939910888671, "memory(GiB)": 73.83, "step": 92715, "token_acc": 0.8582677165354331, "train_speed(iter/s)": 1.458423 }, { "epoch": 3.9724090655927338, "grad_norm": 2.8790132999420166, "learning_rate": 1.0064692204326699e-05, "loss": 0.25708780288696287, "memory(GiB)": 73.83, "step": 92720, "token_acc": 0.9425287356321839, "train_speed(iter/s)": 1.458424 }, { "epoch": 3.972623280921983, "grad_norm": 6.0841288566589355, "learning_rate": 1.006064311854471e-05, "loss": 0.4435241222381592, "memory(GiB)": 73.83, "step": 92725, "token_acc": 0.9169960474308301, "train_speed(iter/s)": 1.45843 }, { "epoch": 3.972837496251232, "grad_norm": 4.923411846160889, "learning_rate": 1.0056594756301658e-05, "loss": 0.35997135639190675, "memory(GiB)": 73.83, "step": 92730, "token_acc": 0.904320987654321, "train_speed(iter/s)": 1.458439 }, { "epoch": 3.9730517115804806, "grad_norm": 1.5304800271987915, "learning_rate": 1.0052547117670863e-05, "loss": 0.2614882946014404, "memory(GiB)": 73.83, "step": 92735, "token_acc": 0.9302949061662198, "train_speed(iter/s)": 1.458439 }, { "epoch": 3.97326592690973, "grad_norm": 1.6848472356796265, "learning_rate": 1.004850020272563e-05, "loss": 0.2304358959197998, "memory(GiB)": 73.83, "step": 92740, "token_acc": 0.9552238805970149, "train_speed(iter/s)": 1.45844 }, { "epoch": 3.9734801422389787, "grad_norm": 2.339149236679077, "learning_rate": 1.0044454011539317e-05, "loss": 0.281096887588501, "memory(GiB)": 73.83, "step": 92745, "token_acc": 0.9352226720647774, "train_speed(iter/s)": 1.458441 }, { "epoch": 3.9736943575682275, "grad_norm": 4.135396957397461, "learning_rate": 1.0040408544185192e-05, "loss": 0.42684712409973147, "memory(GiB)": 73.83, "step": 92750, "token_acc": 0.8962962962962963, "train_speed(iter/s)": 1.458439 }, { "epoch": 3.973908572897477, "grad_norm": 2.7473461627960205, "learning_rate": 1.0036363800736554e-05, "loss": 0.29967057704925537, "memory(GiB)": 73.83, "step": 92755, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.458449 }, { "epoch": 3.9741227882267256, "grad_norm": 2.381883144378662, "learning_rate": 1.0032319781266674e-05, "loss": 0.21881027221679689, "memory(GiB)": 73.83, "step": 92760, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.458453 }, { "epoch": 3.9743370035559744, "grad_norm": 3.0337517261505127, "learning_rate": 1.0028276485848803e-05, "loss": 0.3863114595413208, "memory(GiB)": 73.83, "step": 92765, "token_acc": 0.9390681003584229, "train_speed(iter/s)": 1.458461 }, { "epoch": 3.9745512188852237, "grad_norm": 4.188570499420166, "learning_rate": 1.0024233914556213e-05, "loss": 0.35970778465270997, "memory(GiB)": 73.83, "step": 92770, "token_acc": 0.9186046511627907, "train_speed(iter/s)": 1.458455 }, { "epoch": 3.9747654342144725, "grad_norm": 1.997742772102356, "learning_rate": 1.0020192067462125e-05, "loss": 0.44922075271606443, "memory(GiB)": 73.83, "step": 92775, "token_acc": 0.9065155807365439, "train_speed(iter/s)": 1.458456 }, { "epoch": 3.9749796495437213, "grad_norm": 3.2650489807128906, "learning_rate": 1.0016150944639763e-05, "loss": 0.5257848262786865, "memory(GiB)": 73.83, "step": 92780, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.458461 }, { "epoch": 3.9751938648729706, "grad_norm": 1.6735663414001465, "learning_rate": 1.0012110546162334e-05, "loss": 0.18579860925674438, "memory(GiB)": 73.83, "step": 92785, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.458463 }, { "epoch": 3.9754080802022194, "grad_norm": 1.7263730764389038, "learning_rate": 1.0008070872103032e-05, "loss": 0.3944389820098877, "memory(GiB)": 73.83, "step": 92790, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.458469 }, { "epoch": 3.975622295531468, "grad_norm": 4.040563583374023, "learning_rate": 1.0004031922535029e-05, "loss": 0.2931070327758789, "memory(GiB)": 73.83, "step": 92795, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.45847 }, { "epoch": 3.9758365108607174, "grad_norm": 1.5992316007614136, "learning_rate": 9.999993697531512e-06, "loss": 0.22988884449005126, "memory(GiB)": 73.83, "step": 92800, "token_acc": 0.950530035335689, "train_speed(iter/s)": 1.458468 }, { "epoch": 3.9760507261899662, "grad_norm": 1.985144019126892, "learning_rate": 9.99595619716564e-06, "loss": 0.43945822715759275, "memory(GiB)": 73.83, "step": 92805, "token_acc": 0.9028571428571428, "train_speed(iter/s)": 1.458467 }, { "epoch": 3.976264941519215, "grad_norm": 4.836609840393066, "learning_rate": 9.991919421510547e-06, "loss": 0.31491737365722655, "memory(GiB)": 73.83, "step": 92810, "token_acc": 0.9233576642335767, "train_speed(iter/s)": 1.458467 }, { "epoch": 3.9764791568484643, "grad_norm": 4.353375434875488, "learning_rate": 9.987883370639368e-06, "loss": 0.23467001914978028, "memory(GiB)": 73.83, "step": 92815, "token_acc": 0.954225352112676, "train_speed(iter/s)": 1.458478 }, { "epoch": 3.976693372177713, "grad_norm": 1.3454228639602661, "learning_rate": 9.983848044625216e-06, "loss": 0.23791189193725587, "memory(GiB)": 73.83, "step": 92820, "token_acc": 0.952112676056338, "train_speed(iter/s)": 1.458493 }, { "epoch": 3.976907587506962, "grad_norm": 3.425621271133423, "learning_rate": 9.97981344354118e-06, "loss": 0.3122744083404541, "memory(GiB)": 73.83, "step": 92825, "token_acc": 0.9144736842105263, "train_speed(iter/s)": 1.458503 }, { "epoch": 3.977121802836211, "grad_norm": 2.0792055130004883, "learning_rate": 9.97577956746038e-06, "loss": 0.3410488605499268, "memory(GiB)": 73.83, "step": 92830, "token_acc": 0.9271523178807947, "train_speed(iter/s)": 1.458501 }, { "epoch": 3.97733601816546, "grad_norm": 1.9698941707611084, "learning_rate": 9.971746416455879e-06, "loss": 0.09859130382537842, "memory(GiB)": 73.83, "step": 92835, "token_acc": 0.9753521126760564, "train_speed(iter/s)": 1.458512 }, { "epoch": 3.977550233494709, "grad_norm": 3.278618812561035, "learning_rate": 9.96771399060074e-06, "loss": 0.2561699390411377, "memory(GiB)": 73.83, "step": 92840, "token_acc": 0.932806324110672, "train_speed(iter/s)": 1.458525 }, { "epoch": 3.977764448823958, "grad_norm": 1.0933654308319092, "learning_rate": 9.963682289968018e-06, "loss": 0.2734501838684082, "memory(GiB)": 73.83, "step": 92845, "token_acc": 0.9419354838709677, "train_speed(iter/s)": 1.458529 }, { "epoch": 3.977978664153207, "grad_norm": 4.557270526885986, "learning_rate": 9.959651314630747e-06, "loss": 0.2414107322692871, "memory(GiB)": 73.83, "step": 92850, "token_acc": 0.9404255319148936, "train_speed(iter/s)": 1.458541 }, { "epoch": 3.9781928794824557, "grad_norm": 3.144547700881958, "learning_rate": 9.955621064661935e-06, "loss": 0.24893105030059814, "memory(GiB)": 73.83, "step": 92855, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.458545 }, { "epoch": 3.978407094811705, "grad_norm": 2.1108238697052, "learning_rate": 9.95159154013463e-06, "loss": 0.28245632648468016, "memory(GiB)": 73.83, "step": 92860, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.458542 }, { "epoch": 3.9786213101409538, "grad_norm": 2.910305976867676, "learning_rate": 9.947562741121813e-06, "loss": 0.3240018367767334, "memory(GiB)": 73.83, "step": 92865, "token_acc": 0.9087301587301587, "train_speed(iter/s)": 1.45855 }, { "epoch": 3.9788355254702026, "grad_norm": 4.537869453430176, "learning_rate": 9.943534667696459e-06, "loss": 0.47360405921936033, "memory(GiB)": 73.83, "step": 92870, "token_acc": 0.8918032786885246, "train_speed(iter/s)": 1.458552 }, { "epoch": 3.979049740799452, "grad_norm": 1.594789743423462, "learning_rate": 9.939507319931557e-06, "loss": 0.21985912322998047, "memory(GiB)": 73.83, "step": 92875, "token_acc": 0.9543859649122807, "train_speed(iter/s)": 1.458554 }, { "epoch": 3.9792639561287007, "grad_norm": 4.012551307678223, "learning_rate": 9.935480697900063e-06, "loss": 0.3999828100204468, "memory(GiB)": 73.83, "step": 92880, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.458559 }, { "epoch": 3.9794781714579495, "grad_norm": 4.502296447753906, "learning_rate": 9.931454801674906e-06, "loss": 0.336596155166626, "memory(GiB)": 73.83, "step": 92885, "token_acc": 0.9288135593220339, "train_speed(iter/s)": 1.45856 }, { "epoch": 3.9796923867871987, "grad_norm": 7.0745110511779785, "learning_rate": 9.92742963132905e-06, "loss": 0.4553853511810303, "memory(GiB)": 73.83, "step": 92890, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.458569 }, { "epoch": 3.9799066021164475, "grad_norm": 6.2436394691467285, "learning_rate": 9.923405186935392e-06, "loss": 0.41385812759399415, "memory(GiB)": 73.83, "step": 92895, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.458583 }, { "epoch": 3.9801208174456963, "grad_norm": 2.010312795639038, "learning_rate": 9.919381468566846e-06, "loss": 0.47264814376831055, "memory(GiB)": 73.83, "step": 92900, "token_acc": 0.9161490683229814, "train_speed(iter/s)": 1.458586 }, { "epoch": 3.9803350327749456, "grad_norm": 1.3403801918029785, "learning_rate": 9.9153584762963e-06, "loss": 0.20572166442871093, "memory(GiB)": 73.83, "step": 92905, "token_acc": 0.9603960396039604, "train_speed(iter/s)": 1.458587 }, { "epoch": 3.9805492481041944, "grad_norm": 2.7936453819274902, "learning_rate": 9.911336210196626e-06, "loss": 0.1591116189956665, "memory(GiB)": 73.83, "step": 92910, "token_acc": 0.9585062240663901, "train_speed(iter/s)": 1.458594 }, { "epoch": 3.9807634634334432, "grad_norm": 2.1348323822021484, "learning_rate": 9.907314670340717e-06, "loss": 0.33284406661987304, "memory(GiB)": 73.83, "step": 92915, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.458594 }, { "epoch": 3.9809776787626925, "grad_norm": 2.6088593006134033, "learning_rate": 9.903293856801415e-06, "loss": 0.4111870288848877, "memory(GiB)": 73.83, "step": 92920, "token_acc": 0.8914473684210527, "train_speed(iter/s)": 1.458597 }, { "epoch": 3.9811918940919413, "grad_norm": 6.075390338897705, "learning_rate": 9.899273769651557e-06, "loss": 0.3000878572463989, "memory(GiB)": 73.83, "step": 92925, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.4586 }, { "epoch": 3.98140610942119, "grad_norm": 6.6653900146484375, "learning_rate": 9.895254408963972e-06, "loss": 0.3254518508911133, "memory(GiB)": 73.83, "step": 92930, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.458602 }, { "epoch": 3.9816203247504394, "grad_norm": 3.2101171016693115, "learning_rate": 9.891235774811474e-06, "loss": 0.22819674015045166, "memory(GiB)": 73.83, "step": 92935, "token_acc": 0.9439252336448598, "train_speed(iter/s)": 1.458604 }, { "epoch": 3.981834540079688, "grad_norm": 1.3484232425689697, "learning_rate": 9.88721786726685e-06, "loss": 0.20009124279022217, "memory(GiB)": 73.83, "step": 92940, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.458612 }, { "epoch": 3.982048755408937, "grad_norm": 1.7389353513717651, "learning_rate": 9.883200686402905e-06, "loss": 0.11466193199157715, "memory(GiB)": 73.83, "step": 92945, "token_acc": 0.9686274509803922, "train_speed(iter/s)": 1.458616 }, { "epoch": 3.9822629707381862, "grad_norm": 4.322993755340576, "learning_rate": 9.879184232292426e-06, "loss": 0.3901206016540527, "memory(GiB)": 73.83, "step": 92950, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.458622 }, { "epoch": 3.982477186067435, "grad_norm": 4.862278461456299, "learning_rate": 9.875168505008159e-06, "loss": 0.34976911544799805, "memory(GiB)": 73.83, "step": 92955, "token_acc": 0.9141104294478528, "train_speed(iter/s)": 1.458626 }, { "epoch": 3.982691401396684, "grad_norm": 3.5191123485565186, "learning_rate": 9.87115350462286e-06, "loss": 0.21760244369506837, "memory(GiB)": 73.83, "step": 92960, "token_acc": 0.9540229885057471, "train_speed(iter/s)": 1.45863 }, { "epoch": 3.982905616725933, "grad_norm": 3.7881901264190674, "learning_rate": 9.867139231209255e-06, "loss": 0.4231268405914307, "memory(GiB)": 73.83, "step": 92965, "token_acc": 0.9240121580547113, "train_speed(iter/s)": 1.458632 }, { "epoch": 3.983119832055182, "grad_norm": 2.282968044281006, "learning_rate": 9.863125684840059e-06, "loss": 0.43421497344970705, "memory(GiB)": 73.83, "step": 92970, "token_acc": 0.895910780669145, "train_speed(iter/s)": 1.458633 }, { "epoch": 3.9833340473844308, "grad_norm": 3.6180107593536377, "learning_rate": 9.859112865588e-06, "loss": 0.3724878549575806, "memory(GiB)": 73.83, "step": 92975, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.458637 }, { "epoch": 3.98354826271368, "grad_norm": 4.068251609802246, "learning_rate": 9.855100773525771e-06, "loss": 0.30672035217285154, "memory(GiB)": 73.83, "step": 92980, "token_acc": 0.9361022364217252, "train_speed(iter/s)": 1.458643 }, { "epoch": 3.983762478042929, "grad_norm": 5.796994209289551, "learning_rate": 9.851089408726055e-06, "loss": 0.32082836627960204, "memory(GiB)": 73.83, "step": 92985, "token_acc": 0.9274193548387096, "train_speed(iter/s)": 1.458645 }, { "epoch": 3.9839766933721776, "grad_norm": 0.5429520606994629, "learning_rate": 9.847078771261509e-06, "loss": 0.19980701208114623, "memory(GiB)": 73.83, "step": 92990, "token_acc": 0.9571428571428572, "train_speed(iter/s)": 1.458649 }, { "epoch": 3.984190908701427, "grad_norm": 3.3081729412078857, "learning_rate": 9.843068861204801e-06, "loss": 0.1755264639854431, "memory(GiB)": 73.83, "step": 92995, "token_acc": 0.9724409448818898, "train_speed(iter/s)": 1.458656 }, { "epoch": 3.9844051240306757, "grad_norm": 1.6102442741394043, "learning_rate": 9.839059678628559e-06, "loss": 0.28764240741729735, "memory(GiB)": 73.83, "step": 93000, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.458673 }, { "epoch": 3.9844051240306757, "eval_loss": 2.442375421524048, "eval_runtime": 11.2114, "eval_samples_per_second": 8.92, "eval_steps_per_second": 8.92, "eval_token_acc": 0.44, "step": 93000 }, { "epoch": 3.9846193393599245, "grad_norm": 5.601236820220947, "learning_rate": 9.83505122360544e-06, "loss": 0.37116830348968505, "memory(GiB)": 73.83, "step": 93005, "token_acc": 0.5642633228840125, "train_speed(iter/s)": 1.458396 }, { "epoch": 3.9848335546891738, "grad_norm": 2.198046922683716, "learning_rate": 9.831043496208042e-06, "loss": 0.25783958435058596, "memory(GiB)": 73.83, "step": 93010, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.4584 }, { "epoch": 3.9850477700184226, "grad_norm": 3.2466633319854736, "learning_rate": 9.827036496508973e-06, "loss": 0.6105743408203125, "memory(GiB)": 73.83, "step": 93015, "token_acc": 0.875, "train_speed(iter/s)": 1.458411 }, { "epoch": 3.9852619853476714, "grad_norm": 2.6678099632263184, "learning_rate": 9.823030224580809e-06, "loss": 0.2881613254547119, "memory(GiB)": 73.83, "step": 93020, "token_acc": 0.9348534201954397, "train_speed(iter/s)": 1.458412 }, { "epoch": 3.9854762006769207, "grad_norm": 2.995086193084717, "learning_rate": 9.819024680496159e-06, "loss": 0.33956863880157473, "memory(GiB)": 73.83, "step": 93025, "token_acc": 0.9317507418397626, "train_speed(iter/s)": 1.458415 }, { "epoch": 3.9856904160061695, "grad_norm": 3.1987063884735107, "learning_rate": 9.815019864327551e-06, "loss": 0.2315971612930298, "memory(GiB)": 73.83, "step": 93030, "token_acc": 0.9367469879518072, "train_speed(iter/s)": 1.458418 }, { "epoch": 3.9859046313354183, "grad_norm": 3.8487956523895264, "learning_rate": 9.81101577614757e-06, "loss": 0.2592129945755005, "memory(GiB)": 73.83, "step": 93035, "token_acc": 0.932258064516129, "train_speed(iter/s)": 1.458416 }, { "epoch": 3.9861188466646675, "grad_norm": 6.435823440551758, "learning_rate": 9.807012416028739e-06, "loss": 0.4902402400970459, "memory(GiB)": 73.83, "step": 93040, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.458421 }, { "epoch": 3.9863330619939163, "grad_norm": 1.7811022996902466, "learning_rate": 9.803009784043581e-06, "loss": 0.4464315414428711, "memory(GiB)": 73.83, "step": 93045, "token_acc": 0.9225806451612903, "train_speed(iter/s)": 1.458426 }, { "epoch": 3.986547277323165, "grad_norm": 2.809879779815674, "learning_rate": 9.799007880264611e-06, "loss": 0.40581607818603516, "memory(GiB)": 73.83, "step": 93050, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.45843 }, { "epoch": 3.9867614926524144, "grad_norm": 4.030426979064941, "learning_rate": 9.79500670476431e-06, "loss": 0.3927032709121704, "memory(GiB)": 73.83, "step": 93055, "token_acc": 0.9163763066202091, "train_speed(iter/s)": 1.458437 }, { "epoch": 3.9869757079816632, "grad_norm": 4.6223673820495605, "learning_rate": 9.791006257615192e-06, "loss": 0.6183012008666993, "memory(GiB)": 73.83, "step": 93060, "token_acc": 0.8688046647230321, "train_speed(iter/s)": 1.45844 }, { "epoch": 3.987189923310912, "grad_norm": 5.612214088439941, "learning_rate": 9.787006538889714e-06, "loss": 0.4575809478759766, "memory(GiB)": 73.83, "step": 93065, "token_acc": 0.875968992248062, "train_speed(iter/s)": 1.458439 }, { "epoch": 3.9874041386401613, "grad_norm": 4.7367353439331055, "learning_rate": 9.783007548660338e-06, "loss": 0.24752445220947267, "memory(GiB)": 73.83, "step": 93070, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.458444 }, { "epoch": 3.98761835396941, "grad_norm": 3.0838499069213867, "learning_rate": 9.779009286999501e-06, "loss": 0.20063726902008056, "memory(GiB)": 73.83, "step": 93075, "token_acc": 0.9519650655021834, "train_speed(iter/s)": 1.458445 }, { "epoch": 3.987832569298659, "grad_norm": 3.9276552200317383, "learning_rate": 9.775011753979646e-06, "loss": 0.530812931060791, "memory(GiB)": 73.83, "step": 93080, "token_acc": 0.8812949640287769, "train_speed(iter/s)": 1.458442 }, { "epoch": 3.988046784627908, "grad_norm": 2.0181422233581543, "learning_rate": 9.771014949673174e-06, "loss": 0.3383587121963501, "memory(GiB)": 73.83, "step": 93085, "token_acc": 0.8956228956228957, "train_speed(iter/s)": 1.458451 }, { "epoch": 3.988260999957157, "grad_norm": 0.18907420337200165, "learning_rate": 9.767018874152517e-06, "loss": 0.2865665197372437, "memory(GiB)": 73.83, "step": 93090, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.458456 }, { "epoch": 3.988475215286406, "grad_norm": 1.0803462266921997, "learning_rate": 9.76302352749004e-06, "loss": 0.49110636711120603, "memory(GiB)": 73.83, "step": 93095, "token_acc": 0.9046052631578947, "train_speed(iter/s)": 1.458468 }, { "epoch": 3.988689430615655, "grad_norm": 5.432915687561035, "learning_rate": 9.75902890975815e-06, "loss": 0.5720970153808593, "memory(GiB)": 73.83, "step": 93100, "token_acc": 0.8683274021352313, "train_speed(iter/s)": 1.458469 }, { "epoch": 3.988903645944904, "grad_norm": 7.08799409866333, "learning_rate": 9.755035021029202e-06, "loss": 0.5607736110687256, "memory(GiB)": 73.83, "step": 93105, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.458469 }, { "epoch": 3.9891178612741527, "grad_norm": 3.442924737930298, "learning_rate": 9.751041861375549e-06, "loss": 0.3985152244567871, "memory(GiB)": 73.83, "step": 93110, "token_acc": 0.9, "train_speed(iter/s)": 1.458467 }, { "epoch": 3.989332076603402, "grad_norm": 2.1890339851379395, "learning_rate": 9.74704943086951e-06, "loss": 0.23418068885803223, "memory(GiB)": 73.83, "step": 93115, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.458467 }, { "epoch": 3.9895462919326508, "grad_norm": 4.335849285125732, "learning_rate": 9.74305772958345e-06, "loss": 0.49205813407897947, "memory(GiB)": 73.83, "step": 93120, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.458479 }, { "epoch": 3.9897605072618996, "grad_norm": 1.310845971107483, "learning_rate": 9.73906675758966e-06, "loss": 0.29150710105895994, "memory(GiB)": 73.83, "step": 93125, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.458479 }, { "epoch": 3.989974722591149, "grad_norm": 0.40125882625579834, "learning_rate": 9.735076514960445e-06, "loss": 0.39795398712158203, "memory(GiB)": 73.83, "step": 93130, "token_acc": 0.9170305676855895, "train_speed(iter/s)": 1.4585 }, { "epoch": 3.9901889379203976, "grad_norm": 1.5966051816940308, "learning_rate": 9.731087001768085e-06, "loss": 0.3828551769256592, "memory(GiB)": 73.83, "step": 93135, "token_acc": 0.9118942731277533, "train_speed(iter/s)": 1.458504 }, { "epoch": 3.9904031532496465, "grad_norm": 1.3962452411651611, "learning_rate": 9.727098218084868e-06, "loss": 0.32921576499938965, "memory(GiB)": 73.83, "step": 93140, "token_acc": 0.9390243902439024, "train_speed(iter/s)": 1.458505 }, { "epoch": 3.9906173685788957, "grad_norm": 6.005207538604736, "learning_rate": 9.72311016398303e-06, "loss": 0.9285135269165039, "memory(GiB)": 73.83, "step": 93145, "token_acc": 0.8057324840764332, "train_speed(iter/s)": 1.458533 }, { "epoch": 3.9908315839081445, "grad_norm": 0.45455634593963623, "learning_rate": 9.719122839534844e-06, "loss": 0.21018919944763184, "memory(GiB)": 73.83, "step": 93150, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.458537 }, { "epoch": 3.9910457992373933, "grad_norm": 2.8301820755004883, "learning_rate": 9.715136244812533e-06, "loss": 0.39438316822052, "memory(GiB)": 73.83, "step": 93155, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.458542 }, { "epoch": 3.9912600145666426, "grad_norm": 3.4190773963928223, "learning_rate": 9.711150379888328e-06, "loss": 0.24300537109375, "memory(GiB)": 73.83, "step": 93160, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.458544 }, { "epoch": 3.9914742298958914, "grad_norm": 5.705650329589844, "learning_rate": 9.70716524483441e-06, "loss": 0.2915026187896729, "memory(GiB)": 73.83, "step": 93165, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.458542 }, { "epoch": 3.99168844522514, "grad_norm": 4.797740459442139, "learning_rate": 9.703180839723003e-06, "loss": 0.3527646541595459, "memory(GiB)": 73.83, "step": 93170, "token_acc": 0.909375, "train_speed(iter/s)": 1.458548 }, { "epoch": 3.9919026605543895, "grad_norm": 4.1988677978515625, "learning_rate": 9.699197164626261e-06, "loss": 0.4048208236694336, "memory(GiB)": 73.83, "step": 93175, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.458542 }, { "epoch": 3.9921168758836383, "grad_norm": 4.587080478668213, "learning_rate": 9.695214219616383e-06, "loss": 0.33542752265930176, "memory(GiB)": 73.83, "step": 93180, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.458543 }, { "epoch": 3.992331091212887, "grad_norm": 2.6883511543273926, "learning_rate": 9.69123200476551e-06, "loss": 0.24370856285095216, "memory(GiB)": 73.83, "step": 93185, "token_acc": 0.9528619528619529, "train_speed(iter/s)": 1.458542 }, { "epoch": 3.9925453065421364, "grad_norm": 2.5199596881866455, "learning_rate": 9.687250520145785e-06, "loss": 0.3558960914611816, "memory(GiB)": 73.83, "step": 93190, "token_acc": 0.924901185770751, "train_speed(iter/s)": 1.458543 }, { "epoch": 3.992759521871385, "grad_norm": 0.8902509808540344, "learning_rate": 9.683269765829328e-06, "loss": 0.34066627025604246, "memory(GiB)": 73.83, "step": 93195, "token_acc": 0.9243243243243243, "train_speed(iter/s)": 1.458545 }, { "epoch": 3.992973737200634, "grad_norm": 2.4178473949432373, "learning_rate": 9.679289741888248e-06, "loss": 0.366081690788269, "memory(GiB)": 73.83, "step": 93200, "token_acc": 0.9274924471299094, "train_speed(iter/s)": 1.458545 }, { "epoch": 3.9931879525298832, "grad_norm": 0.9989567399024963, "learning_rate": 9.67531044839467e-06, "loss": 0.26615917682647705, "memory(GiB)": 73.83, "step": 93205, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.458548 }, { "epoch": 3.993402167859132, "grad_norm": 3.0208961963653564, "learning_rate": 9.67133188542067e-06, "loss": 0.4207324981689453, "memory(GiB)": 73.83, "step": 93210, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.458556 }, { "epoch": 3.993616383188381, "grad_norm": 4.284745693206787, "learning_rate": 9.667354053038329e-06, "loss": 0.5243640899658203, "memory(GiB)": 73.83, "step": 93215, "token_acc": 0.889589905362776, "train_speed(iter/s)": 1.458568 }, { "epoch": 3.99383059851763, "grad_norm": 2.510524272918701, "learning_rate": 9.663376951319702e-06, "loss": 0.42661342620849607, "memory(GiB)": 73.83, "step": 93220, "token_acc": 0.900398406374502, "train_speed(iter/s)": 1.458572 }, { "epoch": 3.994044813846879, "grad_norm": 2.117854356765747, "learning_rate": 9.659400580336836e-06, "loss": 0.4046923160552979, "memory(GiB)": 73.83, "step": 93225, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.45857 }, { "epoch": 3.9942590291761277, "grad_norm": 5.335406303405762, "learning_rate": 9.655424940161761e-06, "loss": 0.5262259006500244, "memory(GiB)": 73.83, "step": 93230, "token_acc": 0.8763250883392226, "train_speed(iter/s)": 1.458569 }, { "epoch": 3.994473244505377, "grad_norm": 1.563000202178955, "learning_rate": 9.651450030866527e-06, "loss": 0.24569201469421387, "memory(GiB)": 73.83, "step": 93235, "token_acc": 0.9400544959128065, "train_speed(iter/s)": 1.458573 }, { "epoch": 3.994687459834626, "grad_norm": 5.185432434082031, "learning_rate": 9.647475852523109e-06, "loss": 0.615963363647461, "memory(GiB)": 73.83, "step": 93240, "token_acc": 0.8854166666666666, "train_speed(iter/s)": 1.45857 }, { "epoch": 3.9949016751638746, "grad_norm": 2.4824483394622803, "learning_rate": 9.64350240520353e-06, "loss": 0.41025490760803224, "memory(GiB)": 73.83, "step": 93245, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.458574 }, { "epoch": 3.995115890493124, "grad_norm": 4.036127090454102, "learning_rate": 9.639529688979765e-06, "loss": 0.21744813919067382, "memory(GiB)": 73.83, "step": 93250, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.458587 }, { "epoch": 3.9953301058223727, "grad_norm": 0.21154797077178955, "learning_rate": 9.63555770392378e-06, "loss": 0.6604944229125976, "memory(GiB)": 73.83, "step": 93255, "token_acc": 0.8401486988847584, "train_speed(iter/s)": 1.458592 }, { "epoch": 3.9955443211516215, "grad_norm": 5.095943450927734, "learning_rate": 9.63158645010752e-06, "loss": 0.45545291900634766, "memory(GiB)": 73.83, "step": 93260, "token_acc": 0.8846153846153846, "train_speed(iter/s)": 1.458587 }, { "epoch": 3.9957585364808708, "grad_norm": 2.6238598823547363, "learning_rate": 9.627615927602957e-06, "loss": 0.16402019262313844, "memory(GiB)": 73.83, "step": 93265, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.458589 }, { "epoch": 3.9959727518101196, "grad_norm": 1.1541974544525146, "learning_rate": 9.623646136482e-06, "loss": 0.2592662811279297, "memory(GiB)": 73.83, "step": 93270, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.458608 }, { "epoch": 3.9961869671393684, "grad_norm": 2.788583993911743, "learning_rate": 9.619677076816568e-06, "loss": 0.35111203193664553, "memory(GiB)": 73.83, "step": 93275, "token_acc": 0.9096774193548387, "train_speed(iter/s)": 1.45861 }, { "epoch": 3.9964011824686176, "grad_norm": 4.898138046264648, "learning_rate": 9.615708748678565e-06, "loss": 0.2878578186035156, "memory(GiB)": 73.83, "step": 93280, "token_acc": 0.9454022988505747, "train_speed(iter/s)": 1.458613 }, { "epoch": 3.9966153977978665, "grad_norm": 0.6068550944328308, "learning_rate": 9.611741152139886e-06, "loss": 0.3334269762039185, "memory(GiB)": 73.83, "step": 93285, "token_acc": 0.9255663430420712, "train_speed(iter/s)": 1.458615 }, { "epoch": 3.9968296131271153, "grad_norm": 0.10267291218042374, "learning_rate": 9.607774287272386e-06, "loss": 0.3733291387557983, "memory(GiB)": 73.83, "step": 93290, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.45862 }, { "epoch": 3.9970438284563645, "grad_norm": 2.9866318702697754, "learning_rate": 9.603808154147958e-06, "loss": 0.37077620029449465, "memory(GiB)": 73.83, "step": 93295, "token_acc": 0.9023569023569024, "train_speed(iter/s)": 1.458626 }, { "epoch": 3.9972580437856133, "grad_norm": 3.380307674407959, "learning_rate": 9.599842752838444e-06, "loss": 0.5926393032073974, "memory(GiB)": 73.83, "step": 93300, "token_acc": 0.8898550724637682, "train_speed(iter/s)": 1.458629 }, { "epoch": 3.997472259114862, "grad_norm": 4.445250511169434, "learning_rate": 9.595878083415676e-06, "loss": 0.42943425178527833, "memory(GiB)": 73.83, "step": 93305, "token_acc": 0.8945454545454545, "train_speed(iter/s)": 1.458636 }, { "epoch": 3.9976864744441114, "grad_norm": 1.9646614789962769, "learning_rate": 9.59191414595148e-06, "loss": 0.28893356323242186, "memory(GiB)": 73.83, "step": 93310, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.458642 }, { "epoch": 3.99790068977336, "grad_norm": 1.1868969202041626, "learning_rate": 9.587950940517648e-06, "loss": 0.1778350591659546, "memory(GiB)": 73.83, "step": 93315, "token_acc": 0.9656357388316151, "train_speed(iter/s)": 1.458645 }, { "epoch": 3.998114905102609, "grad_norm": 4.2819905281066895, "learning_rate": 9.583988467185995e-06, "loss": 0.31158583164215087, "memory(GiB)": 73.83, "step": 93320, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.458652 }, { "epoch": 3.9983291204318583, "grad_norm": 0.6234617829322815, "learning_rate": 9.580026726028318e-06, "loss": 0.3350935459136963, "memory(GiB)": 73.83, "step": 93325, "token_acc": 0.9303030303030303, "train_speed(iter/s)": 1.458652 }, { "epoch": 3.998543335761107, "grad_norm": 2.7107503414154053, "learning_rate": 9.576065717116378e-06, "loss": 0.28310861587524416, "memory(GiB)": 73.83, "step": 93330, "token_acc": 0.9353846153846154, "train_speed(iter/s)": 1.458653 }, { "epoch": 3.998757551090356, "grad_norm": 3.0142312049865723, "learning_rate": 9.572105440521922e-06, "loss": 0.7036429882049561, "memory(GiB)": 73.83, "step": 93335, "token_acc": 0.8692579505300353, "train_speed(iter/s)": 1.458668 }, { "epoch": 3.998971766419605, "grad_norm": 2.1198885440826416, "learning_rate": 9.568145896316704e-06, "loss": 0.2729969024658203, "memory(GiB)": 73.83, "step": 93340, "token_acc": 0.9367816091954023, "train_speed(iter/s)": 1.45867 }, { "epoch": 3.999185981748854, "grad_norm": 3.959475040435791, "learning_rate": 9.56418708457244e-06, "loss": 0.2647205352783203, "memory(GiB)": 73.83, "step": 93345, "token_acc": 0.9409282700421941, "train_speed(iter/s)": 1.458678 }, { "epoch": 3.999400197078103, "grad_norm": 3.5694737434387207, "learning_rate": 9.560229005360866e-06, "loss": 0.20955891609191896, "memory(GiB)": 73.83, "step": 93350, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.458698 }, { "epoch": 3.999614412407352, "grad_norm": 3.8273561000823975, "learning_rate": 9.556271658753685e-06, "loss": 0.24987831115722656, "memory(GiB)": 73.83, "step": 93355, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.458696 }, { "epoch": 3.999828627736601, "grad_norm": 0.1563795655965805, "learning_rate": 9.552315044822579e-06, "loss": 0.38870813846588137, "memory(GiB)": 73.83, "step": 93360, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.458696 }, { "epoch": 4.00004284306585, "grad_norm": 3.916023015975952, "learning_rate": 9.548359163639226e-06, "loss": 0.29577367305755614, "memory(GiB)": 73.83, "step": 93365, "token_acc": 0.9392712550607287, "train_speed(iter/s)": 1.4587 }, { "epoch": 4.000257058395099, "grad_norm": 0.4506331980228424, "learning_rate": 9.5444040152753e-06, "loss": 0.3477515459060669, "memory(GiB)": 73.83, "step": 93370, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.458713 }, { "epoch": 4.000471273724347, "grad_norm": 4.63006591796875, "learning_rate": 9.540449599802426e-06, "loss": 0.728610897064209, "memory(GiB)": 73.83, "step": 93375, "token_acc": 0.85, "train_speed(iter/s)": 1.458728 }, { "epoch": 4.0006854890535966, "grad_norm": 4.47650146484375, "learning_rate": 9.536495917292283e-06, "loss": 0.434998083114624, "memory(GiB)": 73.83, "step": 93380, "token_acc": 0.9280575539568345, "train_speed(iter/s)": 1.45874 }, { "epoch": 4.000899704382846, "grad_norm": 3.7281570434570312, "learning_rate": 9.532542967816466e-06, "loss": 0.32059602737426757, "memory(GiB)": 73.83, "step": 93385, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.45874 }, { "epoch": 4.001113919712094, "grad_norm": 0.5371716618537903, "learning_rate": 9.528590751446587e-06, "loss": 0.23748533725738524, "memory(GiB)": 73.83, "step": 93390, "token_acc": 0.9346938775510204, "train_speed(iter/s)": 1.458741 }, { "epoch": 4.001328135041343, "grad_norm": 6.396268844604492, "learning_rate": 9.524639268254259e-06, "loss": 0.548110055923462, "memory(GiB)": 73.83, "step": 93395, "token_acc": 0.9035714285714286, "train_speed(iter/s)": 1.458753 }, { "epoch": 4.001542350370593, "grad_norm": 5.089548110961914, "learning_rate": 9.520688518311065e-06, "loss": 0.3861522674560547, "memory(GiB)": 73.83, "step": 93400, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.458763 }, { "epoch": 4.001756565699841, "grad_norm": 4.393570423126221, "learning_rate": 9.51673850168856e-06, "loss": 0.23543233871459962, "memory(GiB)": 73.83, "step": 93405, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.458775 }, { "epoch": 4.00197078102909, "grad_norm": 1.887742042541504, "learning_rate": 9.512789218458323e-06, "loss": 0.3425728797912598, "memory(GiB)": 73.83, "step": 93410, "token_acc": 0.9202127659574468, "train_speed(iter/s)": 1.458776 }, { "epoch": 4.00218499635834, "grad_norm": 2.7362582683563232, "learning_rate": 9.508840668691887e-06, "loss": 0.2525345325469971, "memory(GiB)": 73.83, "step": 93415, "token_acc": 0.9213483146067416, "train_speed(iter/s)": 1.458793 }, { "epoch": 4.002399211687588, "grad_norm": 4.050759792327881, "learning_rate": 9.50489285246079e-06, "loss": 0.42478628158569337, "memory(GiB)": 73.83, "step": 93420, "token_acc": 0.9135446685878963, "train_speed(iter/s)": 1.458798 }, { "epoch": 4.002613427016837, "grad_norm": 0.381220281124115, "learning_rate": 9.500945769836545e-06, "loss": 0.21056513786315917, "memory(GiB)": 73.83, "step": 93425, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.458808 }, { "epoch": 4.0028276423460865, "grad_norm": 2.8244640827178955, "learning_rate": 9.496999420890646e-06, "loss": 0.18346045017242432, "memory(GiB)": 73.83, "step": 93430, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.458808 }, { "epoch": 4.003041857675335, "grad_norm": 5.253096580505371, "learning_rate": 9.493053805694608e-06, "loss": 0.17649508714675904, "memory(GiB)": 73.83, "step": 93435, "token_acc": 0.9768339768339769, "train_speed(iter/s)": 1.458817 }, { "epoch": 4.003256073004584, "grad_norm": 0.8488630056381226, "learning_rate": 9.4891089243199e-06, "loss": 0.34626216888427735, "memory(GiB)": 73.83, "step": 93440, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.458815 }, { "epoch": 4.003470288333833, "grad_norm": 5.417370796203613, "learning_rate": 9.48516477683799e-06, "loss": 0.499052095413208, "memory(GiB)": 73.83, "step": 93445, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.458816 }, { "epoch": 4.003684503663082, "grad_norm": 2.142089366912842, "learning_rate": 9.48122136332032e-06, "loss": 0.2743516445159912, "memory(GiB)": 73.83, "step": 93450, "token_acc": 0.9228295819935691, "train_speed(iter/s)": 1.458819 }, { "epoch": 4.003898718992331, "grad_norm": 5.5771918296813965, "learning_rate": 9.477278683838341e-06, "loss": 0.3277240514755249, "memory(GiB)": 73.83, "step": 93455, "token_acc": 0.9572953736654805, "train_speed(iter/s)": 1.458827 }, { "epoch": 4.00411293432158, "grad_norm": 3.3601415157318115, "learning_rate": 9.473336738463457e-06, "loss": 0.2802829027175903, "memory(GiB)": 73.83, "step": 93460, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.458835 }, { "epoch": 4.004327149650829, "grad_norm": 6.004004955291748, "learning_rate": 9.469395527267089e-06, "loss": 0.2626879453659058, "memory(GiB)": 73.83, "step": 93465, "token_acc": 0.9419354838709677, "train_speed(iter/s)": 1.458838 }, { "epoch": 4.004541364980078, "grad_norm": 4.2246880531311035, "learning_rate": 9.46545505032066e-06, "loss": 0.2990840196609497, "memory(GiB)": 73.83, "step": 93470, "token_acc": 0.92578125, "train_speed(iter/s)": 1.458838 }, { "epoch": 4.004755580309327, "grad_norm": 10.853462219238281, "learning_rate": 9.46151530769554e-06, "loss": 0.49813504219055177, "memory(GiB)": 73.83, "step": 93475, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.45884 }, { "epoch": 4.0049697956385755, "grad_norm": 2.1875975131988525, "learning_rate": 9.457576299463094e-06, "loss": 0.3097047805786133, "memory(GiB)": 73.83, "step": 93480, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.458839 }, { "epoch": 4.005184010967825, "grad_norm": 3.589834690093994, "learning_rate": 9.45363802569469e-06, "loss": 0.2982046604156494, "memory(GiB)": 73.83, "step": 93485, "token_acc": 0.9425675675675675, "train_speed(iter/s)": 1.45884 }, { "epoch": 4.005398226297074, "grad_norm": 3.389155864715576, "learning_rate": 9.449700486461649e-06, "loss": 0.41328582763671873, "memory(GiB)": 73.83, "step": 93490, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.45884 }, { "epoch": 4.005612441626322, "grad_norm": 0.4820370078086853, "learning_rate": 9.44576368183534e-06, "loss": 0.612665319442749, "memory(GiB)": 73.83, "step": 93495, "token_acc": 0.8683385579937304, "train_speed(iter/s)": 1.45885 }, { "epoch": 4.005826656955572, "grad_norm": 4.048168659210205, "learning_rate": 9.44182761188706e-06, "loss": 0.3150007247924805, "memory(GiB)": 73.83, "step": 93500, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.458847 }, { "epoch": 4.005826656955572, "eval_loss": 2.2783761024475098, "eval_runtime": 12.0493, "eval_samples_per_second": 8.299, "eval_steps_per_second": 8.299, "eval_token_acc": 0.4879032258064516, "step": 93500 }, { "epoch": 4.006040872284821, "grad_norm": 3.5521578788757324, "learning_rate": 9.437892276688126e-06, "loss": 0.376718544960022, "memory(GiB)": 73.83, "step": 93505, "token_acc": 0.6218721037998146, "train_speed(iter/s)": 1.458555 }, { "epoch": 4.006255087614069, "grad_norm": 5.210806369781494, "learning_rate": 9.433957676309823e-06, "loss": 0.4067685604095459, "memory(GiB)": 73.83, "step": 93510, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.458566 }, { "epoch": 4.0064693029433185, "grad_norm": 3.425699234008789, "learning_rate": 9.43002381082343e-06, "loss": 0.35170233249664307, "memory(GiB)": 73.83, "step": 93515, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.458573 }, { "epoch": 4.006683518272568, "grad_norm": 4.446146488189697, "learning_rate": 9.426090680300197e-06, "loss": 0.5149882316589356, "memory(GiB)": 73.83, "step": 93520, "token_acc": 0.912751677852349, "train_speed(iter/s)": 1.458574 }, { "epoch": 4.006897733601816, "grad_norm": 2.739867687225342, "learning_rate": 9.422158284811406e-06, "loss": 0.33726844787597654, "memory(GiB)": 73.83, "step": 93525, "token_acc": 0.9368131868131868, "train_speed(iter/s)": 1.458573 }, { "epoch": 4.007111948931065, "grad_norm": 0.4762301743030548, "learning_rate": 9.41822662442829e-06, "loss": 0.20419113636016845, "memory(GiB)": 73.83, "step": 93530, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.458586 }, { "epoch": 4.007326164260315, "grad_norm": 3.2680375576019287, "learning_rate": 9.414295699222048e-06, "loss": 0.3540801525115967, "memory(GiB)": 73.83, "step": 93535, "token_acc": 0.9052287581699346, "train_speed(iter/s)": 1.458586 }, { "epoch": 4.007540379589563, "grad_norm": 4.533607482910156, "learning_rate": 9.410365509263924e-06, "loss": 0.1103330373764038, "memory(GiB)": 73.83, "step": 93540, "token_acc": 0.9854014598540146, "train_speed(iter/s)": 1.458587 }, { "epoch": 4.007754594918812, "grad_norm": 3.8703627586364746, "learning_rate": 9.406436054625106e-06, "loss": 0.4382792949676514, "memory(GiB)": 73.83, "step": 93545, "token_acc": 0.9296875, "train_speed(iter/s)": 1.458598 }, { "epoch": 4.0079688102480615, "grad_norm": 2.679769515991211, "learning_rate": 9.40250733537677e-06, "loss": 0.4223890781402588, "memory(GiB)": 73.83, "step": 93550, "token_acc": 0.9172932330827067, "train_speed(iter/s)": 1.458613 }, { "epoch": 4.00818302557731, "grad_norm": 1.4475817680358887, "learning_rate": 9.398579351590103e-06, "loss": 0.3539857864379883, "memory(GiB)": 73.83, "step": 93555, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.458615 }, { "epoch": 4.008397240906559, "grad_norm": 4.246394157409668, "learning_rate": 9.394652103336265e-06, "loss": 0.5706164360046386, "memory(GiB)": 73.83, "step": 93560, "token_acc": 0.8910891089108911, "train_speed(iter/s)": 1.458627 }, { "epoch": 4.008611456235808, "grad_norm": 3.9084489345550537, "learning_rate": 9.39072559068639e-06, "loss": 0.3348174810409546, "memory(GiB)": 73.83, "step": 93565, "token_acc": 0.9426229508196722, "train_speed(iter/s)": 1.458635 }, { "epoch": 4.008825671565057, "grad_norm": 4.397092819213867, "learning_rate": 9.386799813711617e-06, "loss": 0.541400146484375, "memory(GiB)": 73.83, "step": 93570, "token_acc": 0.9219330855018587, "train_speed(iter/s)": 1.458644 }, { "epoch": 4.009039886894306, "grad_norm": 2.6127676963806152, "learning_rate": 9.382874772483052e-06, "loss": 0.3688701391220093, "memory(GiB)": 73.83, "step": 93575, "token_acc": 0.910958904109589, "train_speed(iter/s)": 1.458649 }, { "epoch": 4.009254102223555, "grad_norm": 2.3174703121185303, "learning_rate": 9.378950467071829e-06, "loss": 0.2734518051147461, "memory(GiB)": 73.83, "step": 93580, "token_acc": 0.9337539432176656, "train_speed(iter/s)": 1.45865 }, { "epoch": 4.009468317552804, "grad_norm": 1.418569803237915, "learning_rate": 9.375026897549017e-06, "loss": 0.34462239742279055, "memory(GiB)": 73.83, "step": 93585, "token_acc": 0.9294478527607362, "train_speed(iter/s)": 1.458651 }, { "epoch": 4.009682532882053, "grad_norm": 3.0033833980560303, "learning_rate": 9.371104063985702e-06, "loss": 0.30479295253753663, "memory(GiB)": 73.83, "step": 93590, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.458657 }, { "epoch": 4.009896748211302, "grad_norm": 3.3660402297973633, "learning_rate": 9.367181966452953e-06, "loss": 0.34887094497680665, "memory(GiB)": 73.83, "step": 93595, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.458652 }, { "epoch": 4.0101109635405505, "grad_norm": 0.40253347158432007, "learning_rate": 9.363260605021824e-06, "loss": 0.26095645427703856, "memory(GiB)": 73.83, "step": 93600, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.458656 }, { "epoch": 4.0103251788698, "grad_norm": 1.8917664289474487, "learning_rate": 9.359339979763326e-06, "loss": 0.4101395606994629, "memory(GiB)": 73.83, "step": 93605, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.458663 }, { "epoch": 4.010539394199049, "grad_norm": 1.7363628149032593, "learning_rate": 9.355420090748518e-06, "loss": 0.1633205771446228, "memory(GiB)": 73.83, "step": 93610, "token_acc": 0.966542750929368, "train_speed(iter/s)": 1.458664 }, { "epoch": 4.010753609528297, "grad_norm": 2.600647449493408, "learning_rate": 9.351500938048408e-06, "loss": 0.4201503753662109, "memory(GiB)": 73.83, "step": 93615, "token_acc": 0.9082278481012658, "train_speed(iter/s)": 1.458674 }, { "epoch": 4.010967824857547, "grad_norm": 2.679264545440674, "learning_rate": 9.347582521733994e-06, "loss": 0.23682219982147218, "memory(GiB)": 73.83, "step": 93620, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.458675 }, { "epoch": 4.011182040186796, "grad_norm": 2.947183132171631, "learning_rate": 9.34366484187626e-06, "loss": 0.3066393852233887, "memory(GiB)": 73.83, "step": 93625, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.458681 }, { "epoch": 4.011396255516045, "grad_norm": 3.586968183517456, "learning_rate": 9.339747898546169e-06, "loss": 0.2517491340637207, "memory(GiB)": 73.83, "step": 93630, "token_acc": 0.9517684887459807, "train_speed(iter/s)": 1.458676 }, { "epoch": 4.0116104708452935, "grad_norm": 1.5459405183792114, "learning_rate": 9.335831691814684e-06, "loss": 0.23121790885925292, "memory(GiB)": 73.83, "step": 93635, "token_acc": 0.9550898203592815, "train_speed(iter/s)": 1.458681 }, { "epoch": 4.011824686174543, "grad_norm": 2.1160244941711426, "learning_rate": 9.331916221752756e-06, "loss": 0.21334149837493896, "memory(GiB)": 73.83, "step": 93640, "token_acc": 0.9438596491228071, "train_speed(iter/s)": 1.458683 }, { "epoch": 4.012038901503792, "grad_norm": 3.4494283199310303, "learning_rate": 9.328001488431326e-06, "loss": 0.36775186061859133, "memory(GiB)": 73.83, "step": 93645, "token_acc": 0.9291338582677166, "train_speed(iter/s)": 1.458681 }, { "epoch": 4.01225311683304, "grad_norm": 4.273740768432617, "learning_rate": 9.324087491921296e-06, "loss": 0.47138171195983886, "memory(GiB)": 73.83, "step": 93650, "token_acc": 0.9070422535211268, "train_speed(iter/s)": 1.458687 }, { "epoch": 4.01246733216229, "grad_norm": 2.144360065460205, "learning_rate": 9.320174232293576e-06, "loss": 0.2099216938018799, "memory(GiB)": 73.83, "step": 93655, "token_acc": 0.9504132231404959, "train_speed(iter/s)": 1.458692 }, { "epoch": 4.012681547491539, "grad_norm": 2.0558924674987793, "learning_rate": 9.316261709619068e-06, "loss": 0.3360844612121582, "memory(GiB)": 73.83, "step": 93660, "token_acc": 0.9227799227799228, "train_speed(iter/s)": 1.458703 }, { "epoch": 4.012895762820787, "grad_norm": 4.390480995178223, "learning_rate": 9.312349923968628e-06, "loss": 0.34025580883026124, "memory(GiB)": 73.83, "step": 93665, "token_acc": 0.9267515923566879, "train_speed(iter/s)": 1.458709 }, { "epoch": 4.013109978150037, "grad_norm": 0.14915616810321808, "learning_rate": 9.308438875413145e-06, "loss": 0.2920227527618408, "memory(GiB)": 73.83, "step": 93670, "token_acc": 0.934640522875817, "train_speed(iter/s)": 1.45871 }, { "epoch": 4.013324193479286, "grad_norm": 2.4761803150177, "learning_rate": 9.304528564023468e-06, "loss": 0.17565653324127198, "memory(GiB)": 73.83, "step": 93675, "token_acc": 0.9630996309963099, "train_speed(iter/s)": 1.458708 }, { "epoch": 4.013538408808534, "grad_norm": 1.5071589946746826, "learning_rate": 9.30061898987043e-06, "loss": 0.40970163345336913, "memory(GiB)": 73.83, "step": 93680, "token_acc": 0.9138461538461539, "train_speed(iter/s)": 1.458721 }, { "epoch": 4.013752624137783, "grad_norm": 2.7098588943481445, "learning_rate": 9.296710153024846e-06, "loss": 0.3602594375610352, "memory(GiB)": 73.83, "step": 93685, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.458732 }, { "epoch": 4.013966839467033, "grad_norm": 2.4198267459869385, "learning_rate": 9.292802053557553e-06, "loss": 0.3577301502227783, "memory(GiB)": 73.83, "step": 93690, "token_acc": 0.9173228346456693, "train_speed(iter/s)": 1.458733 }, { "epoch": 4.014181054796281, "grad_norm": 5.253167629241943, "learning_rate": 9.288894691539323e-06, "loss": 0.47867870330810547, "memory(GiB)": 73.83, "step": 93695, "token_acc": 0.9012345679012346, "train_speed(iter/s)": 1.45874 }, { "epoch": 4.01439527012553, "grad_norm": 2.291123628616333, "learning_rate": 9.284988067040973e-06, "loss": 0.2149869203567505, "memory(GiB)": 73.83, "step": 93700, "token_acc": 0.9462025316455697, "train_speed(iter/s)": 1.458744 }, { "epoch": 4.01460948545478, "grad_norm": 1.982826828956604, "learning_rate": 9.281082180133254e-06, "loss": 0.17784380912780762, "memory(GiB)": 73.83, "step": 93705, "token_acc": 0.9634146341463414, "train_speed(iter/s)": 1.458743 }, { "epoch": 4.014823700784028, "grad_norm": 1.4393064975738525, "learning_rate": 9.27717703088693e-06, "loss": 0.4407931327819824, "memory(GiB)": 73.83, "step": 93710, "token_acc": 0.9125, "train_speed(iter/s)": 1.458745 }, { "epoch": 4.015037916113277, "grad_norm": 5.081395626068115, "learning_rate": 9.273272619372748e-06, "loss": 0.5993137359619141, "memory(GiB)": 73.83, "step": 93715, "token_acc": 0.8456591639871383, "train_speed(iter/s)": 1.458753 }, { "epoch": 4.0152521314425265, "grad_norm": 1.9459296464920044, "learning_rate": 9.269368945661422e-06, "loss": 0.34437873363494875, "memory(GiB)": 73.83, "step": 93720, "token_acc": 0.9124579124579124, "train_speed(iter/s)": 1.458768 }, { "epoch": 4.015466346771775, "grad_norm": 5.398425102233887, "learning_rate": 9.265466009823698e-06, "loss": 0.647445821762085, "memory(GiB)": 73.83, "step": 93725, "token_acc": 0.8713235294117647, "train_speed(iter/s)": 1.458779 }, { "epoch": 4.015680562101024, "grad_norm": 2.0753884315490723, "learning_rate": 9.261563811930269e-06, "loss": 0.1722138524055481, "memory(GiB)": 73.83, "step": 93730, "token_acc": 0.9641693811074918, "train_speed(iter/s)": 1.458781 }, { "epoch": 4.015894777430273, "grad_norm": 10.17336654663086, "learning_rate": 9.257662352051827e-06, "loss": 0.6999724864959717, "memory(GiB)": 73.83, "step": 93735, "token_acc": 0.8659420289855072, "train_speed(iter/s)": 1.458787 }, { "epoch": 4.016108992759522, "grad_norm": 1.6832313537597656, "learning_rate": 9.25376163025905e-06, "loss": 0.2635433912277222, "memory(GiB)": 73.83, "step": 93740, "token_acc": 0.9398280802292264, "train_speed(iter/s)": 1.458792 }, { "epoch": 4.016323208088771, "grad_norm": 2.9630467891693115, "learning_rate": 9.24986164662261e-06, "loss": 0.18351426124572753, "memory(GiB)": 73.83, "step": 93745, "token_acc": 0.9568965517241379, "train_speed(iter/s)": 1.458794 }, { "epoch": 4.01653742341802, "grad_norm": 1.9704985618591309, "learning_rate": 9.245962401213132e-06, "loss": 0.18205126523971557, "memory(GiB)": 73.83, "step": 93750, "token_acc": 0.9652509652509652, "train_speed(iter/s)": 1.458795 }, { "epoch": 4.016751638747269, "grad_norm": 4.193221092224121, "learning_rate": 9.242063894101294e-06, "loss": 0.5133508682250977, "memory(GiB)": 73.83, "step": 93755, "token_acc": 0.9137380191693291, "train_speed(iter/s)": 1.458797 }, { "epoch": 4.016965854076518, "grad_norm": 1.171485424041748, "learning_rate": 9.238166125357689e-06, "loss": 0.20653038024902343, "memory(GiB)": 73.83, "step": 93760, "token_acc": 0.9579579579579579, "train_speed(iter/s)": 1.458799 }, { "epoch": 4.017180069405767, "grad_norm": 4.719894886016846, "learning_rate": 9.234269095052956e-06, "loss": 0.7073173522949219, "memory(GiB)": 73.83, "step": 93765, "token_acc": 0.8503401360544217, "train_speed(iter/s)": 1.458799 }, { "epoch": 4.0173942847350155, "grad_norm": 0.11948703974485397, "learning_rate": 9.230372803257675e-06, "loss": 0.18896018266677855, "memory(GiB)": 73.83, "step": 93770, "token_acc": 0.9572953736654805, "train_speed(iter/s)": 1.458804 }, { "epoch": 4.017608500064265, "grad_norm": 5.759969234466553, "learning_rate": 9.226477250042443e-06, "loss": 0.5369710922241211, "memory(GiB)": 73.83, "step": 93775, "token_acc": 0.9063545150501672, "train_speed(iter/s)": 1.458809 }, { "epoch": 4.017822715393514, "grad_norm": 1.9720990657806396, "learning_rate": 9.222582435477805e-06, "loss": 0.3201334476470947, "memory(GiB)": 73.83, "step": 93780, "token_acc": 0.9226190476190477, "train_speed(iter/s)": 1.458814 }, { "epoch": 4.018036930722762, "grad_norm": 4.847124099731445, "learning_rate": 9.218688359634352e-06, "loss": 0.6200182914733887, "memory(GiB)": 73.83, "step": 93785, "token_acc": 0.8587786259541985, "train_speed(iter/s)": 1.458817 }, { "epoch": 4.018251146052012, "grad_norm": 1.554137945175171, "learning_rate": 9.214795022582617e-06, "loss": 0.1670989513397217, "memory(GiB)": 73.83, "step": 93790, "token_acc": 0.946843853820598, "train_speed(iter/s)": 1.45882 }, { "epoch": 4.018465361381261, "grad_norm": 10.671019554138184, "learning_rate": 9.210902424393131e-06, "loss": 0.55215425491333, "memory(GiB)": 73.83, "step": 93795, "token_acc": 0.8676470588235294, "train_speed(iter/s)": 1.458815 }, { "epoch": 4.018679576710509, "grad_norm": 5.8683390617370605, "learning_rate": 9.207010565136409e-06, "loss": 0.3129847049713135, "memory(GiB)": 73.83, "step": 93800, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.458824 }, { "epoch": 4.0188937920397585, "grad_norm": 2.287123441696167, "learning_rate": 9.203119444882958e-06, "loss": 0.3820496559143066, "memory(GiB)": 73.83, "step": 93805, "token_acc": 0.9221311475409836, "train_speed(iter/s)": 1.458829 }, { "epoch": 4.019108007369008, "grad_norm": 3.470235586166382, "learning_rate": 9.199229063703257e-06, "loss": 0.32242627143859864, "memory(GiB)": 73.83, "step": 93810, "token_acc": 0.9300699300699301, "train_speed(iter/s)": 1.45883 }, { "epoch": 4.019322222698256, "grad_norm": 2.97017240524292, "learning_rate": 9.195339421667804e-06, "loss": 0.3659482479095459, "memory(GiB)": 73.83, "step": 93815, "token_acc": 0.9337349397590361, "train_speed(iter/s)": 1.458833 }, { "epoch": 4.019536438027505, "grad_norm": 1.530998945236206, "learning_rate": 9.191450518847062e-06, "loss": 0.2826169490814209, "memory(GiB)": 73.83, "step": 93820, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.458833 }, { "epoch": 4.019750653356755, "grad_norm": 3.6302616596221924, "learning_rate": 9.187562355311469e-06, "loss": 0.3090542793273926, "memory(GiB)": 73.83, "step": 93825, "token_acc": 0.9075342465753424, "train_speed(iter/s)": 1.458835 }, { "epoch": 4.019964868686003, "grad_norm": 3.2926125526428223, "learning_rate": 9.183674931131464e-06, "loss": 0.42302913665771485, "memory(GiB)": 73.83, "step": 93830, "token_acc": 0.9246031746031746, "train_speed(iter/s)": 1.458843 }, { "epoch": 4.020179084015252, "grad_norm": 2.2265803813934326, "learning_rate": 9.17978824637748e-06, "loss": 0.39337456226348877, "memory(GiB)": 73.83, "step": 93835, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.458849 }, { "epoch": 4.0203932993445015, "grad_norm": 3.2284281253814697, "learning_rate": 9.175902301119922e-06, "loss": 0.43983283042907717, "memory(GiB)": 73.83, "step": 93840, "token_acc": 0.9012345679012346, "train_speed(iter/s)": 1.45886 }, { "epoch": 4.02060751467375, "grad_norm": 0.2257789820432663, "learning_rate": 9.172017095429191e-06, "loss": 0.2482088565826416, "memory(GiB)": 73.83, "step": 93845, "token_acc": 0.9510204081632653, "train_speed(iter/s)": 1.458865 }, { "epoch": 4.020821730002999, "grad_norm": 1.510593295097351, "learning_rate": 9.16813262937568e-06, "loss": 0.4217970371246338, "memory(GiB)": 73.83, "step": 93850, "token_acc": 0.9186746987951807, "train_speed(iter/s)": 1.458869 }, { "epoch": 4.021035945332248, "grad_norm": 0.49009743332862854, "learning_rate": 9.164248903029742e-06, "loss": 0.38755571842193604, "memory(GiB)": 73.83, "step": 93855, "token_acc": 0.9085365853658537, "train_speed(iter/s)": 1.458881 }, { "epoch": 4.021250160661497, "grad_norm": 3.8158884048461914, "learning_rate": 9.16036591646175e-06, "loss": 0.3686375617980957, "memory(GiB)": 73.83, "step": 93860, "token_acc": 0.9149797570850202, "train_speed(iter/s)": 1.458901 }, { "epoch": 4.021464375990746, "grad_norm": 0.7451112270355225, "learning_rate": 9.156483669742022e-06, "loss": 0.30981709957122805, "memory(GiB)": 73.83, "step": 93865, "token_acc": 0.9391891891891891, "train_speed(iter/s)": 1.4589 }, { "epoch": 4.021678591319995, "grad_norm": 2.7416231632232666, "learning_rate": 9.152602162940921e-06, "loss": 0.21076323986053466, "memory(GiB)": 73.83, "step": 93870, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.458903 }, { "epoch": 4.021892806649244, "grad_norm": 3.619824171066284, "learning_rate": 9.148721396128751e-06, "loss": 0.5106104850769043, "memory(GiB)": 73.83, "step": 93875, "token_acc": 0.8897338403041825, "train_speed(iter/s)": 1.458909 }, { "epoch": 4.022107021978493, "grad_norm": 1.1775803565979004, "learning_rate": 9.144841369375811e-06, "loss": 0.12093185186386109, "memory(GiB)": 73.83, "step": 93880, "token_acc": 0.9619883040935673, "train_speed(iter/s)": 1.458908 }, { "epoch": 4.022321237307742, "grad_norm": 2.2953622341156006, "learning_rate": 9.140962082752397e-06, "loss": 0.14148982763290405, "memory(GiB)": 73.83, "step": 93885, "token_acc": 0.969811320754717, "train_speed(iter/s)": 1.45891 }, { "epoch": 4.0225354526369905, "grad_norm": 3.8488662242889404, "learning_rate": 9.137083536328783e-06, "loss": 0.3508529901504517, "memory(GiB)": 73.83, "step": 93890, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.45891 }, { "epoch": 4.02274966796624, "grad_norm": 2.8734140396118164, "learning_rate": 9.133205730175221e-06, "loss": 0.18443963527679444, "memory(GiB)": 73.83, "step": 93895, "token_acc": 0.9522058823529411, "train_speed(iter/s)": 1.458913 }, { "epoch": 4.022963883295489, "grad_norm": 2.8390345573425293, "learning_rate": 9.129328664361985e-06, "loss": 0.43355398178100585, "memory(GiB)": 73.83, "step": 93900, "token_acc": 0.916083916083916, "train_speed(iter/s)": 1.458915 }, { "epoch": 4.023178098624737, "grad_norm": 3.719041109085083, "learning_rate": 9.12545233895929e-06, "loss": 0.19454102516174315, "memory(GiB)": 73.83, "step": 93905, "token_acc": 0.9492753623188406, "train_speed(iter/s)": 1.458926 }, { "epoch": 4.023392313953987, "grad_norm": 6.7378692626953125, "learning_rate": 9.121576754037381e-06, "loss": 0.39220528602600097, "memory(GiB)": 73.83, "step": 93910, "token_acc": 0.9372384937238494, "train_speed(iter/s)": 1.45893 }, { "epoch": 4.023606529283236, "grad_norm": 5.406418800354004, "learning_rate": 9.117701909666453e-06, "loss": 0.7567895412445068, "memory(GiB)": 73.83, "step": 93915, "token_acc": 0.8589341692789969, "train_speed(iter/s)": 1.458932 }, { "epoch": 4.023820744612484, "grad_norm": 4.908551216125488, "learning_rate": 9.113827805916714e-06, "loss": 0.3342757225036621, "memory(GiB)": 73.83, "step": 93920, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.458934 }, { "epoch": 4.0240349599417335, "grad_norm": 2.09194016456604, "learning_rate": 9.109954442858315e-06, "loss": 0.22366511821746826, "memory(GiB)": 73.83, "step": 93925, "token_acc": 0.952054794520548, "train_speed(iter/s)": 1.458937 }, { "epoch": 4.024249175270983, "grad_norm": 0.8858455419540405, "learning_rate": 9.106081820561469e-06, "loss": 0.23601288795471193, "memory(GiB)": 73.83, "step": 93930, "token_acc": 0.9483870967741935, "train_speed(iter/s)": 1.458938 }, { "epoch": 4.024463390600231, "grad_norm": 2.1370344161987305, "learning_rate": 9.102209939096311e-06, "loss": 0.3294321060180664, "memory(GiB)": 73.83, "step": 93935, "token_acc": 0.913312693498452, "train_speed(iter/s)": 1.458942 }, { "epoch": 4.02467760592948, "grad_norm": 4.513490676879883, "learning_rate": 9.098338798532985e-06, "loss": 0.6548812389373779, "memory(GiB)": 73.83, "step": 93940, "token_acc": 0.857707509881423, "train_speed(iter/s)": 1.458956 }, { "epoch": 4.02489182125873, "grad_norm": 2.5564026832580566, "learning_rate": 9.094468398941618e-06, "loss": 0.27661316394805907, "memory(GiB)": 73.83, "step": 93945, "token_acc": 0.9319526627218935, "train_speed(iter/s)": 1.45896 }, { "epoch": 4.025106036587978, "grad_norm": 6.314724445343018, "learning_rate": 9.090598740392331e-06, "loss": 0.3194943189620972, "memory(GiB)": 73.83, "step": 93950, "token_acc": 0.9250814332247557, "train_speed(iter/s)": 1.458963 }, { "epoch": 4.025320251917227, "grad_norm": 1.9995371103286743, "learning_rate": 9.086729822955214e-06, "loss": 0.3777758598327637, "memory(GiB)": 73.83, "step": 93955, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.458978 }, { "epoch": 4.025534467246477, "grad_norm": 3.9305455684661865, "learning_rate": 9.082861646700374e-06, "loss": 0.3221175193786621, "memory(GiB)": 73.83, "step": 93960, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.458975 }, { "epoch": 4.025748682575725, "grad_norm": 0.9986673593521118, "learning_rate": 9.078994211697883e-06, "loss": 0.2506579399108887, "memory(GiB)": 73.83, "step": 93965, "token_acc": 0.9480122324159022, "train_speed(iter/s)": 1.458978 }, { "epoch": 4.025962897904974, "grad_norm": 3.3245816230773926, "learning_rate": 9.075127518017796e-06, "loss": 0.30237200260162356, "memory(GiB)": 73.83, "step": 93970, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.45898 }, { "epoch": 4.0261771132342234, "grad_norm": 2.2480571269989014, "learning_rate": 9.07126156573016e-06, "loss": 0.225325345993042, "memory(GiB)": 73.83, "step": 93975, "token_acc": 0.9633333333333334, "train_speed(iter/s)": 1.458994 }, { "epoch": 4.026391328563472, "grad_norm": 3.4584572315216064, "learning_rate": 9.06739635490501e-06, "loss": 0.2691656589508057, "memory(GiB)": 73.83, "step": 93980, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.458998 }, { "epoch": 4.026605543892721, "grad_norm": 2.620944023132324, "learning_rate": 9.06353188561237e-06, "loss": 0.3555187702178955, "memory(GiB)": 73.83, "step": 93985, "token_acc": 0.9136212624584718, "train_speed(iter/s)": 1.459017 }, { "epoch": 4.02681975922197, "grad_norm": 4.272947788238525, "learning_rate": 9.059668157922262e-06, "loss": 0.5021571636199951, "memory(GiB)": 73.83, "step": 93990, "token_acc": 0.8883928571428571, "train_speed(iter/s)": 1.459019 }, { "epoch": 4.027033974551219, "grad_norm": 2.8513028621673584, "learning_rate": 9.055805171904674e-06, "loss": 0.17015581130981444, "memory(GiB)": 73.83, "step": 93995, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.459033 }, { "epoch": 4.027248189880468, "grad_norm": 1.3602616786956787, "learning_rate": 9.051942927629586e-06, "loss": 0.4189635753631592, "memory(GiB)": 73.83, "step": 94000, "token_acc": 0.9102990033222591, "train_speed(iter/s)": 1.459043 }, { "epoch": 4.027248189880468, "eval_loss": 2.4830918312072754, "eval_runtime": 11.9482, "eval_samples_per_second": 8.369, "eval_steps_per_second": 8.369, "eval_token_acc": 0.45535714285714285, "step": 94000 }, { "epoch": 4.027462405209717, "grad_norm": 2.3647851943969727, "learning_rate": 9.04808142516696e-06, "loss": 0.2789682149887085, "memory(GiB)": 73.83, "step": 94005, "token_acc": 0.5843429636533085, "train_speed(iter/s)": 1.458753 }, { "epoch": 4.027676620538966, "grad_norm": 1.6201688051223755, "learning_rate": 9.044220664586745e-06, "loss": 0.17028721570968627, "memory(GiB)": 73.83, "step": 94010, "token_acc": 0.9625850340136054, "train_speed(iter/s)": 1.458761 }, { "epoch": 4.027890835868215, "grad_norm": 2.2632064819335938, "learning_rate": 9.0403606459589e-06, "loss": 0.3291167259216309, "memory(GiB)": 73.83, "step": 94015, "token_acc": 0.9399293286219081, "train_speed(iter/s)": 1.458764 }, { "epoch": 4.028105051197464, "grad_norm": 5.059566974639893, "learning_rate": 9.03650136935335e-06, "loss": 0.45277252197265627, "memory(GiB)": 73.83, "step": 94020, "token_acc": 0.9035369774919614, "train_speed(iter/s)": 1.458766 }, { "epoch": 4.0283192665267125, "grad_norm": 2.1205737590789795, "learning_rate": 9.03264283484e-06, "loss": 0.25932950973510743, "memory(GiB)": 73.83, "step": 94025, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.458767 }, { "epoch": 4.028533481855962, "grad_norm": 3.0614559650421143, "learning_rate": 9.028785042488758e-06, "loss": 0.3686488628387451, "memory(GiB)": 73.83, "step": 94030, "token_acc": 0.91701244813278, "train_speed(iter/s)": 1.458766 }, { "epoch": 4.028747697185211, "grad_norm": 3.9212300777435303, "learning_rate": 9.024927992369514e-06, "loss": 0.2776022911071777, "memory(GiB)": 73.83, "step": 94035, "token_acc": 0.9330708661417323, "train_speed(iter/s)": 1.458776 }, { "epoch": 4.028961912514459, "grad_norm": 6.155084133148193, "learning_rate": 9.021071684552117e-06, "loss": 0.45780210494995116, "memory(GiB)": 73.83, "step": 94040, "token_acc": 0.9071428571428571, "train_speed(iter/s)": 1.458778 }, { "epoch": 4.029176127843709, "grad_norm": 3.8857874870300293, "learning_rate": 9.017216119106464e-06, "loss": 0.3071354866027832, "memory(GiB)": 73.83, "step": 94045, "token_acc": 0.9299363057324841, "train_speed(iter/s)": 1.458777 }, { "epoch": 4.029390343172958, "grad_norm": 4.754788398742676, "learning_rate": 9.013361296102385e-06, "loss": 0.6797051429748535, "memory(GiB)": 73.83, "step": 94050, "token_acc": 0.870253164556962, "train_speed(iter/s)": 1.458789 }, { "epoch": 4.029604558502206, "grad_norm": 2.345161199569702, "learning_rate": 9.009507215609702e-06, "loss": 0.28306915760040285, "memory(GiB)": 73.83, "step": 94055, "token_acc": 0.9356913183279743, "train_speed(iter/s)": 1.458789 }, { "epoch": 4.0298187738314555, "grad_norm": 0.2944100797176361, "learning_rate": 9.005653877698261e-06, "loss": 0.4675439357757568, "memory(GiB)": 73.83, "step": 94060, "token_acc": 0.8989547038327527, "train_speed(iter/s)": 1.458796 }, { "epoch": 4.030032989160705, "grad_norm": 1.9697754383087158, "learning_rate": 9.001801282437854e-06, "loss": 0.328992223739624, "memory(GiB)": 73.83, "step": 94065, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.45881 }, { "epoch": 4.030247204489953, "grad_norm": 2.6892964839935303, "learning_rate": 8.997949429898267e-06, "loss": 0.2343555212020874, "memory(GiB)": 73.83, "step": 94070, "token_acc": 0.9330985915492958, "train_speed(iter/s)": 1.458817 }, { "epoch": 4.030461419819202, "grad_norm": 2.7413313388824463, "learning_rate": 8.994098320149303e-06, "loss": 0.2260861873626709, "memory(GiB)": 73.83, "step": 94075, "token_acc": 0.9488054607508533, "train_speed(iter/s)": 1.45882 }, { "epoch": 4.030675635148452, "grad_norm": 1.7590839862823486, "learning_rate": 8.99024795326071e-06, "loss": 0.16048816442489625, "memory(GiB)": 73.83, "step": 94080, "token_acc": 0.952054794520548, "train_speed(iter/s)": 1.458818 }, { "epoch": 4.0308898504777, "grad_norm": 6.902778148651123, "learning_rate": 8.986398329302248e-06, "loss": 0.28037495613098146, "memory(GiB)": 73.83, "step": 94085, "token_acc": 0.9420849420849421, "train_speed(iter/s)": 1.458818 }, { "epoch": 4.031104065806949, "grad_norm": 1.5175105333328247, "learning_rate": 8.982549448343653e-06, "loss": 0.1729580283164978, "memory(GiB)": 73.83, "step": 94090, "token_acc": 0.9568627450980393, "train_speed(iter/s)": 1.458821 }, { "epoch": 4.0313182811361985, "grad_norm": 5.042522430419922, "learning_rate": 8.978701310454652e-06, "loss": 0.6236751556396485, "memory(GiB)": 73.83, "step": 94095, "token_acc": 0.88268156424581, "train_speed(iter/s)": 1.458832 }, { "epoch": 4.031532496465447, "grad_norm": 1.0283516645431519, "learning_rate": 8.974853915704945e-06, "loss": 0.4614863395690918, "memory(GiB)": 73.83, "step": 94100, "token_acc": 0.8861111111111111, "train_speed(iter/s)": 1.458839 }, { "epoch": 4.031746711794696, "grad_norm": 1.1713731288909912, "learning_rate": 8.97100726416426e-06, "loss": 0.3100429058074951, "memory(GiB)": 73.83, "step": 94105, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.458835 }, { "epoch": 4.031960927123945, "grad_norm": 0.3815225660800934, "learning_rate": 8.967161355902265e-06, "loss": 0.6341402530670166, "memory(GiB)": 73.83, "step": 94110, "token_acc": 0.8640483383685801, "train_speed(iter/s)": 1.458844 }, { "epoch": 4.032175142453194, "grad_norm": 1.5039565563201904, "learning_rate": 8.96331619098863e-06, "loss": 0.35896193981170654, "memory(GiB)": 73.83, "step": 94115, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.458847 }, { "epoch": 4.032389357782443, "grad_norm": 4.125894069671631, "learning_rate": 8.959471769493022e-06, "loss": 0.17508904933929442, "memory(GiB)": 73.83, "step": 94120, "token_acc": 0.9612903225806452, "train_speed(iter/s)": 1.458852 }, { "epoch": 4.032603573111692, "grad_norm": 1.9931126832962036, "learning_rate": 8.955628091485068e-06, "loss": 0.0871682345867157, "memory(GiB)": 73.83, "step": 94125, "token_acc": 0.9863760217983651, "train_speed(iter/s)": 1.458851 }, { "epoch": 4.032817788440941, "grad_norm": 4.187533855438232, "learning_rate": 8.951785157034415e-06, "loss": 0.2639153480529785, "memory(GiB)": 73.83, "step": 94130, "token_acc": 0.9460431654676259, "train_speed(iter/s)": 1.45885 }, { "epoch": 4.03303200377019, "grad_norm": 2.202911138534546, "learning_rate": 8.947942966210692e-06, "loss": 0.3052999496459961, "memory(GiB)": 73.83, "step": 94135, "token_acc": 0.924901185770751, "train_speed(iter/s)": 1.458851 }, { "epoch": 4.033246219099439, "grad_norm": 1.3217809200286865, "learning_rate": 8.944101519083492e-06, "loss": 0.2902100086212158, "memory(GiB)": 73.83, "step": 94140, "token_acc": 0.9471947194719472, "train_speed(iter/s)": 1.458852 }, { "epoch": 4.0334604344286875, "grad_norm": 3.5471267700195312, "learning_rate": 8.940260815722407e-06, "loss": 0.26657638549804685, "memory(GiB)": 73.83, "step": 94145, "token_acc": 0.9671052631578947, "train_speed(iter/s)": 1.458867 }, { "epoch": 4.033674649757937, "grad_norm": 0.22471259534358978, "learning_rate": 8.936420856197015e-06, "loss": 0.33357398509979247, "memory(GiB)": 73.83, "step": 94150, "token_acc": 0.9315068493150684, "train_speed(iter/s)": 1.458867 }, { "epoch": 4.033888865087186, "grad_norm": 2.0673882961273193, "learning_rate": 8.932581640576865e-06, "loss": 0.2528877258300781, "memory(GiB)": 73.83, "step": 94155, "token_acc": 0.963855421686747, "train_speed(iter/s)": 1.458868 }, { "epoch": 4.034103080416434, "grad_norm": 2.839963912963867, "learning_rate": 8.928743168931535e-06, "loss": 0.5337448120117188, "memory(GiB)": 73.83, "step": 94160, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.458874 }, { "epoch": 4.034317295745684, "grad_norm": 3.514407157897949, "learning_rate": 8.924905441330556e-06, "loss": 0.41772637367248533, "memory(GiB)": 73.83, "step": 94165, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.458877 }, { "epoch": 4.034531511074933, "grad_norm": 3.2629518508911133, "learning_rate": 8.921068457843445e-06, "loss": 0.473248291015625, "memory(GiB)": 73.83, "step": 94170, "token_acc": 0.9003021148036254, "train_speed(iter/s)": 1.458889 }, { "epoch": 4.034745726404181, "grad_norm": 0.1452031433582306, "learning_rate": 8.917232218539712e-06, "loss": 0.39846134185791016, "memory(GiB)": 73.83, "step": 94175, "token_acc": 0.8988095238095238, "train_speed(iter/s)": 1.458895 }, { "epoch": 4.0349599417334305, "grad_norm": 4.241239070892334, "learning_rate": 8.913396723488854e-06, "loss": 0.306516170501709, "memory(GiB)": 73.83, "step": 94180, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.4589 }, { "epoch": 4.03517415706268, "grad_norm": 1.9860142469406128, "learning_rate": 8.909561972760343e-06, "loss": 0.2389678955078125, "memory(GiB)": 73.83, "step": 94185, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.458907 }, { "epoch": 4.035388372391928, "grad_norm": 3.1230130195617676, "learning_rate": 8.905727966423682e-06, "loss": 0.28873209953308104, "memory(GiB)": 73.83, "step": 94190, "token_acc": 0.9637462235649547, "train_speed(iter/s)": 1.458908 }, { "epoch": 4.035602587721177, "grad_norm": 4.9669413566589355, "learning_rate": 8.901894704548308e-06, "loss": 0.47850475311279295, "memory(GiB)": 73.83, "step": 94195, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.458911 }, { "epoch": 4.035816803050427, "grad_norm": 1.2116732597351074, "learning_rate": 8.89806218720365e-06, "loss": 0.18047311305999755, "memory(GiB)": 73.83, "step": 94200, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.458911 }, { "epoch": 4.036031018379675, "grad_norm": 2.2920126914978027, "learning_rate": 8.89423041445916e-06, "loss": 0.27266905307769773, "memory(GiB)": 73.83, "step": 94205, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.458913 }, { "epoch": 4.036245233708924, "grad_norm": 5.357195854187012, "learning_rate": 8.890399386384251e-06, "loss": 0.4772057056427002, "memory(GiB)": 73.83, "step": 94210, "token_acc": 0.8839590443686007, "train_speed(iter/s)": 1.458922 }, { "epoch": 4.0364594490381736, "grad_norm": 0.8708167672157288, "learning_rate": 8.88656910304831e-06, "loss": 0.084649658203125, "memory(GiB)": 73.83, "step": 94215, "token_acc": 0.9701492537313433, "train_speed(iter/s)": 1.458927 }, { "epoch": 4.036673664367422, "grad_norm": 2.6980652809143066, "learning_rate": 8.882739564520743e-06, "loss": 0.4395749568939209, "memory(GiB)": 73.83, "step": 94220, "token_acc": 0.8838951310861424, "train_speed(iter/s)": 1.458928 }, { "epoch": 4.036887879696671, "grad_norm": 4.425510883331299, "learning_rate": 8.878910770870924e-06, "loss": 0.4454970836639404, "memory(GiB)": 73.83, "step": 94225, "token_acc": 0.9161490683229814, "train_speed(iter/s)": 1.458929 }, { "epoch": 4.03710209502592, "grad_norm": 0.2307976335287094, "learning_rate": 8.875082722168215e-06, "loss": 0.6056137084960938, "memory(GiB)": 73.83, "step": 94230, "token_acc": 0.9053497942386831, "train_speed(iter/s)": 1.458942 }, { "epoch": 4.037316310355169, "grad_norm": 5.958781719207764, "learning_rate": 8.871255418481955e-06, "loss": 0.3458531379699707, "memory(GiB)": 73.83, "step": 94235, "token_acc": 0.9246861924686193, "train_speed(iter/s)": 1.458949 }, { "epoch": 4.037530525684418, "grad_norm": 3.086613178253174, "learning_rate": 8.867428859881487e-06, "loss": 0.4370893478393555, "memory(GiB)": 73.83, "step": 94240, "token_acc": 0.9079754601226994, "train_speed(iter/s)": 1.458962 }, { "epoch": 4.037744741013667, "grad_norm": 3.3685038089752197, "learning_rate": 8.863603046436119e-06, "loss": 0.21783242225646973, "memory(GiB)": 73.83, "step": 94245, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.458969 }, { "epoch": 4.037958956342916, "grad_norm": 4.480193614959717, "learning_rate": 8.859777978215184e-06, "loss": 0.6241944789886474, "memory(GiB)": 73.83, "step": 94250, "token_acc": 0.8725490196078431, "train_speed(iter/s)": 1.458975 }, { "epoch": 4.038173171672165, "grad_norm": 1.9771709442138672, "learning_rate": 8.855953655287962e-06, "loss": 0.16295881271362306, "memory(GiB)": 73.83, "step": 94255, "token_acc": 0.9573770491803278, "train_speed(iter/s)": 1.458985 }, { "epoch": 4.038387387001414, "grad_norm": 1.4734861850738525, "learning_rate": 8.852130077723736e-06, "loss": 0.22051804065704345, "memory(GiB)": 73.83, "step": 94260, "token_acc": 0.9482071713147411, "train_speed(iter/s)": 1.458986 }, { "epoch": 4.038601602330663, "grad_norm": 1.6960914134979248, "learning_rate": 8.848307245591774e-06, "loss": 0.3043177604675293, "memory(GiB)": 73.83, "step": 94265, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.458991 }, { "epoch": 4.038815817659912, "grad_norm": 1.756989598274231, "learning_rate": 8.844485158961318e-06, "loss": 0.2188943147659302, "memory(GiB)": 73.83, "step": 94270, "token_acc": 0.9488636363636364, "train_speed(iter/s)": 1.458993 }, { "epoch": 4.039030032989161, "grad_norm": 3.5582668781280518, "learning_rate": 8.840663817901618e-06, "loss": 0.7024284362792969, "memory(GiB)": 73.83, "step": 94275, "token_acc": 0.8681672025723473, "train_speed(iter/s)": 1.458998 }, { "epoch": 4.0392442483184094, "grad_norm": 2.153148889541626, "learning_rate": 8.836843222481917e-06, "loss": 0.17574696540832518, "memory(GiB)": 73.83, "step": 94280, "token_acc": 0.9537366548042705, "train_speed(iter/s)": 1.459004 }, { "epoch": 4.039458463647659, "grad_norm": 7.539645671844482, "learning_rate": 8.833023372771415e-06, "loss": 0.2797266960144043, "memory(GiB)": 73.83, "step": 94285, "token_acc": 0.9397590361445783, "train_speed(iter/s)": 1.459007 }, { "epoch": 4.039672678976908, "grad_norm": 5.173285961151123, "learning_rate": 8.829204268839314e-06, "loss": 0.438076114654541, "memory(GiB)": 73.83, "step": 94290, "token_acc": 0.8988326848249028, "train_speed(iter/s)": 1.459022 }, { "epoch": 4.039886894306156, "grad_norm": 4.110470771789551, "learning_rate": 8.825385910754796e-06, "loss": 0.41046485900878904, "memory(GiB)": 73.83, "step": 94295, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.459026 }, { "epoch": 4.040101109635406, "grad_norm": 1.5213004350662231, "learning_rate": 8.82156829858703e-06, "loss": 0.3968977451324463, "memory(GiB)": 73.83, "step": 94300, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.45904 }, { "epoch": 4.040315324964655, "grad_norm": 0.5391095876693726, "learning_rate": 8.817751432405186e-06, "loss": 0.20167410373687744, "memory(GiB)": 73.83, "step": 94305, "token_acc": 0.9482071713147411, "train_speed(iter/s)": 1.459051 }, { "epoch": 4.040529540293903, "grad_norm": 4.131378650665283, "learning_rate": 8.813935312278414e-06, "loss": 0.3813158512115479, "memory(GiB)": 73.83, "step": 94310, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.459063 }, { "epoch": 4.0407437556231525, "grad_norm": 4.240681171417236, "learning_rate": 8.81011993827584e-06, "loss": 0.6454259872436523, "memory(GiB)": 73.83, "step": 94315, "token_acc": 0.8536585365853658, "train_speed(iter/s)": 1.459083 }, { "epoch": 4.040957970952402, "grad_norm": 4.089972019195557, "learning_rate": 8.806305310466579e-06, "loss": 0.2707887411117554, "memory(GiB)": 73.83, "step": 94320, "token_acc": 0.9381107491856677, "train_speed(iter/s)": 1.459086 }, { "epoch": 4.04117218628165, "grad_norm": 2.7952792644500732, "learning_rate": 8.80249142891974e-06, "loss": 0.3036825656890869, "memory(GiB)": 73.83, "step": 94325, "token_acc": 0.9452554744525548, "train_speed(iter/s)": 1.459089 }, { "epoch": 4.041386401610899, "grad_norm": 3.0905323028564453, "learning_rate": 8.798678293704405e-06, "loss": 0.5307093620300293, "memory(GiB)": 73.83, "step": 94330, "token_acc": 0.8819188191881919, "train_speed(iter/s)": 1.459094 }, { "epoch": 4.041600616940149, "grad_norm": 4.471896171569824, "learning_rate": 8.794865904889677e-06, "loss": 0.31331419944763184, "memory(GiB)": 73.83, "step": 94335, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.459099 }, { "epoch": 4.041814832269397, "grad_norm": 2.2757441997528076, "learning_rate": 8.791054262544601e-06, "loss": 0.16694031953811644, "memory(GiB)": 73.83, "step": 94340, "token_acc": 0.9621212121212122, "train_speed(iter/s)": 1.459111 }, { "epoch": 4.042029047598646, "grad_norm": 2.7549235820770264, "learning_rate": 8.78724336673824e-06, "loss": 0.36051373481750487, "memory(GiB)": 73.83, "step": 94345, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.459113 }, { "epoch": 4.0422432629278955, "grad_norm": 3.9048426151275635, "learning_rate": 8.783433217539611e-06, "loss": 0.2359778881072998, "memory(GiB)": 73.83, "step": 94350, "token_acc": 0.9634551495016611, "train_speed(iter/s)": 1.459111 }, { "epoch": 4.042457478257144, "grad_norm": 4.395294189453125, "learning_rate": 8.779623815017763e-06, "loss": 0.5536608219146728, "memory(GiB)": 73.83, "step": 94355, "token_acc": 0.8993288590604027, "train_speed(iter/s)": 1.459113 }, { "epoch": 4.042671693586393, "grad_norm": 6.062656879425049, "learning_rate": 8.77581515924169e-06, "loss": 0.39990160465240476, "memory(GiB)": 73.83, "step": 94360, "token_acc": 0.9208860759493671, "train_speed(iter/s)": 1.459121 }, { "epoch": 4.042885908915642, "grad_norm": 3.271389961242676, "learning_rate": 8.772007250280407e-06, "loss": 0.45854840278625486, "memory(GiB)": 73.83, "step": 94365, "token_acc": 0.9208211143695014, "train_speed(iter/s)": 1.459126 }, { "epoch": 4.043100124244891, "grad_norm": 9.682509422302246, "learning_rate": 8.768200088202888e-06, "loss": 0.47858619689941406, "memory(GiB)": 73.83, "step": 94370, "token_acc": 0.9058823529411765, "train_speed(iter/s)": 1.459127 }, { "epoch": 4.04331433957414, "grad_norm": 2.5874593257904053, "learning_rate": 8.764393673078102e-06, "loss": 0.3736170768737793, "memory(GiB)": 73.83, "step": 94375, "token_acc": 0.8986301369863013, "train_speed(iter/s)": 1.459131 }, { "epoch": 4.043528554903389, "grad_norm": 2.0265324115753174, "learning_rate": 8.760588004975007e-06, "loss": 0.12209019660949708, "memory(GiB)": 73.83, "step": 94380, "token_acc": 0.9626168224299065, "train_speed(iter/s)": 1.459136 }, { "epoch": 4.043742770232638, "grad_norm": 1.5778141021728516, "learning_rate": 8.756783083962545e-06, "loss": 0.22197554111480713, "memory(GiB)": 73.83, "step": 94385, "token_acc": 0.9551282051282052, "train_speed(iter/s)": 1.459138 }, { "epoch": 4.043956985561887, "grad_norm": 1.9997766017913818, "learning_rate": 8.752978910109638e-06, "loss": 0.49269757270812986, "memory(GiB)": 73.83, "step": 94390, "token_acc": 0.9144981412639405, "train_speed(iter/s)": 1.459139 }, { "epoch": 4.044171200891136, "grad_norm": 1.6213899850845337, "learning_rate": 8.749175483485223e-06, "loss": 0.2754401683807373, "memory(GiB)": 73.83, "step": 94395, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.45914 }, { "epoch": 4.0443854162203845, "grad_norm": 3.9057021141052246, "learning_rate": 8.745372804158187e-06, "loss": 0.1659135937690735, "memory(GiB)": 73.83, "step": 94400, "token_acc": 0.9455782312925171, "train_speed(iter/s)": 1.459141 }, { "epoch": 4.044599631549634, "grad_norm": 3.654238224029541, "learning_rate": 8.741570872197424e-06, "loss": 0.3710469245910645, "memory(GiB)": 73.83, "step": 94405, "token_acc": 0.9012345679012346, "train_speed(iter/s)": 1.45915 }, { "epoch": 4.044813846878883, "grad_norm": 6.465725898742676, "learning_rate": 8.73776968767181e-06, "loss": 0.3207651376724243, "memory(GiB)": 73.83, "step": 94410, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.459156 }, { "epoch": 4.045028062208131, "grad_norm": 3.5060555934906006, "learning_rate": 8.733969250650187e-06, "loss": 0.36670513153076173, "memory(GiB)": 73.83, "step": 94415, "token_acc": 0.9258160237388724, "train_speed(iter/s)": 1.459168 }, { "epoch": 4.045242277537381, "grad_norm": 0.978593647480011, "learning_rate": 8.73016956120144e-06, "loss": 0.3646113395690918, "memory(GiB)": 73.83, "step": 94420, "token_acc": 0.9339622641509434, "train_speed(iter/s)": 1.459169 }, { "epoch": 4.04545649286663, "grad_norm": 1.669148564338684, "learning_rate": 8.72637061939437e-06, "loss": 0.2729646682739258, "memory(GiB)": 73.83, "step": 94425, "token_acc": 0.9329268292682927, "train_speed(iter/s)": 1.459167 }, { "epoch": 4.045670708195878, "grad_norm": 1.9471533298492432, "learning_rate": 8.722572425297832e-06, "loss": 0.2523669719696045, "memory(GiB)": 73.83, "step": 94430, "token_acc": 0.938337801608579, "train_speed(iter/s)": 1.459167 }, { "epoch": 4.0458849235251275, "grad_norm": 1.7915067672729492, "learning_rate": 8.71877497898061e-06, "loss": 0.2749507188796997, "memory(GiB)": 73.83, "step": 94435, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.45917 }, { "epoch": 4.046099138854377, "grad_norm": 2.835419178009033, "learning_rate": 8.714978280511505e-06, "loss": 0.3349249601364136, "memory(GiB)": 73.83, "step": 94440, "token_acc": 0.933933933933934, "train_speed(iter/s)": 1.459173 }, { "epoch": 4.046313354183625, "grad_norm": 5.324923515319824, "learning_rate": 8.711182329959284e-06, "loss": 0.40597825050354003, "memory(GiB)": 73.83, "step": 94445, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.459175 }, { "epoch": 4.046527569512874, "grad_norm": 5.521371841430664, "learning_rate": 8.70738712739274e-06, "loss": 0.36487417221069335, "memory(GiB)": 73.83, "step": 94450, "token_acc": 0.9204892966360856, "train_speed(iter/s)": 1.459175 }, { "epoch": 4.046741784842124, "grad_norm": 4.534594535827637, "learning_rate": 8.703592672880617e-06, "loss": 0.19505653381347657, "memory(GiB)": 73.83, "step": 94455, "token_acc": 0.9601226993865031, "train_speed(iter/s)": 1.459176 }, { "epoch": 4.046956000171372, "grad_norm": 3.100179672241211, "learning_rate": 8.699798966491651e-06, "loss": 0.30712075233459474, "memory(GiB)": 73.83, "step": 94460, "token_acc": 0.9294871794871795, "train_speed(iter/s)": 1.459192 }, { "epoch": 4.047170215500621, "grad_norm": 3.9290273189544678, "learning_rate": 8.696006008294566e-06, "loss": 0.5102787494659424, "memory(GiB)": 73.83, "step": 94465, "token_acc": 0.9014925373134328, "train_speed(iter/s)": 1.459189 }, { "epoch": 4.0473844308298705, "grad_norm": 5.531606197357178, "learning_rate": 8.692213798358084e-06, "loss": 0.31549155712127686, "memory(GiB)": 73.83, "step": 94470, "token_acc": 0.9252336448598131, "train_speed(iter/s)": 1.459193 }, { "epoch": 4.047598646159119, "grad_norm": 1.9264291524887085, "learning_rate": 8.688422336750879e-06, "loss": 0.3105078935623169, "memory(GiB)": 73.83, "step": 94475, "token_acc": 0.9291784702549575, "train_speed(iter/s)": 1.459199 }, { "epoch": 4.047812861488368, "grad_norm": 2.9648609161376953, "learning_rate": 8.684631623541672e-06, "loss": 0.42391128540039064, "memory(GiB)": 73.83, "step": 94480, "token_acc": 0.9096989966555183, "train_speed(iter/s)": 1.459204 }, { "epoch": 4.048027076817617, "grad_norm": 4.217563629150391, "learning_rate": 8.680841658799127e-06, "loss": 0.2492506504058838, "memory(GiB)": 73.83, "step": 94485, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.459209 }, { "epoch": 4.048241292146866, "grad_norm": 0.8074445724487305, "learning_rate": 8.677052442591888e-06, "loss": 0.1806087851524353, "memory(GiB)": 73.83, "step": 94490, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.459213 }, { "epoch": 4.048455507476115, "grad_norm": 3.6499288082122803, "learning_rate": 8.6732639749886e-06, "loss": 0.5545611381530762, "memory(GiB)": 73.83, "step": 94495, "token_acc": 0.8821548821548821, "train_speed(iter/s)": 1.45921 }, { "epoch": 4.048669722805364, "grad_norm": 0.6969799399375916, "learning_rate": 8.669476256057913e-06, "loss": 0.2069333791732788, "memory(GiB)": 73.83, "step": 94500, "token_acc": 0.9406528189910979, "train_speed(iter/s)": 1.459217 }, { "epoch": 4.048669722805364, "eval_loss": 2.4598844051361084, "eval_runtime": 12.0551, "eval_samples_per_second": 8.295, "eval_steps_per_second": 8.295, "eval_token_acc": 0.45113788487282463, "step": 94500 }, { "epoch": 4.048883938134613, "grad_norm": 3.1718027591705322, "learning_rate": 8.665689285868416e-06, "loss": 0.4152851581573486, "memory(GiB)": 73.83, "step": 94505, "token_acc": 0.5727002967359051, "train_speed(iter/s)": 1.458931 }, { "epoch": 4.049098153463862, "grad_norm": 9.879657745361328, "learning_rate": 8.661903064488753e-06, "loss": 0.5091351509094239, "memory(GiB)": 73.83, "step": 94510, "token_acc": 0.8622754491017964, "train_speed(iter/s)": 1.458945 }, { "epoch": 4.049312368793111, "grad_norm": 11.86551284790039, "learning_rate": 8.65811759198749e-06, "loss": 0.342805814743042, "memory(GiB)": 73.83, "step": 94515, "token_acc": 0.9136690647482014, "train_speed(iter/s)": 1.458958 }, { "epoch": 4.0495265841223596, "grad_norm": 4.481446266174316, "learning_rate": 8.654332868433213e-06, "loss": 0.3493818998336792, "memory(GiB)": 73.83, "step": 94520, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.458967 }, { "epoch": 4.049740799451609, "grad_norm": 3.2994091510772705, "learning_rate": 8.650548893894484e-06, "loss": 0.21556975841522216, "memory(GiB)": 73.83, "step": 94525, "token_acc": 0.9628252788104089, "train_speed(iter/s)": 1.458965 }, { "epoch": 4.049955014780858, "grad_norm": 1.1359230279922485, "learning_rate": 8.646765668439844e-06, "loss": 0.4142917633056641, "memory(GiB)": 73.83, "step": 94530, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.458966 }, { "epoch": 4.050169230110106, "grad_norm": 2.8050472736358643, "learning_rate": 8.642983192137827e-06, "loss": 0.3015427589416504, "memory(GiB)": 73.83, "step": 94535, "token_acc": 0.9290123456790124, "train_speed(iter/s)": 1.458971 }, { "epoch": 4.050383445439356, "grad_norm": 2.2959353923797607, "learning_rate": 8.639201465056973e-06, "loss": 0.17113955020904542, "memory(GiB)": 73.83, "step": 94540, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.458972 }, { "epoch": 4.050597660768605, "grad_norm": 0.6320818662643433, "learning_rate": 8.635420487265788e-06, "loss": 0.40349268913269043, "memory(GiB)": 73.83, "step": 94545, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.458975 }, { "epoch": 4.050811876097853, "grad_norm": 4.604569911956787, "learning_rate": 8.631640258832769e-06, "loss": 0.41286630630493165, "memory(GiB)": 73.83, "step": 94550, "token_acc": 0.9116719242902208, "train_speed(iter/s)": 1.45898 }, { "epoch": 4.051026091427103, "grad_norm": 1.4797866344451904, "learning_rate": 8.627860779826385e-06, "loss": 0.10122368335723878, "memory(GiB)": 73.83, "step": 94555, "token_acc": 0.9798657718120806, "train_speed(iter/s)": 1.458986 }, { "epoch": 4.051240306756352, "grad_norm": 1.6808749437332153, "learning_rate": 8.624082050315107e-06, "loss": 0.2323539972305298, "memory(GiB)": 73.83, "step": 94560, "token_acc": 0.9602649006622517, "train_speed(iter/s)": 1.458991 }, { "epoch": 4.0514545220856, "grad_norm": 2.8526813983917236, "learning_rate": 8.620304070367401e-06, "loss": 0.3330716609954834, "memory(GiB)": 73.83, "step": 94565, "token_acc": 0.9389830508474576, "train_speed(iter/s)": 1.458992 }, { "epoch": 4.0516687374148495, "grad_norm": 4.674051284790039, "learning_rate": 8.616526840051697e-06, "loss": 0.29731833934783936, "memory(GiB)": 73.83, "step": 94570, "token_acc": 0.9494584837545126, "train_speed(iter/s)": 1.458999 }, { "epoch": 4.051882952744099, "grad_norm": 7.85341739654541, "learning_rate": 8.612750359436444e-06, "loss": 0.3932304859161377, "memory(GiB)": 73.83, "step": 94575, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.459002 }, { "epoch": 4.052097168073347, "grad_norm": 4.492949962615967, "learning_rate": 8.608974628590044e-06, "loss": 0.27326431274414065, "memory(GiB)": 73.83, "step": 94580, "token_acc": 0.9446640316205533, "train_speed(iter/s)": 1.459009 }, { "epoch": 4.052311383402596, "grad_norm": 2.541766881942749, "learning_rate": 8.605199647580892e-06, "loss": 0.17637259960174562, "memory(GiB)": 73.83, "step": 94585, "token_acc": 0.9611650485436893, "train_speed(iter/s)": 1.459015 }, { "epoch": 4.052525598731846, "grad_norm": 0.647005021572113, "learning_rate": 8.601425416477366e-06, "loss": 0.3748570680618286, "memory(GiB)": 73.83, "step": 94590, "token_acc": 0.9226361031518625, "train_speed(iter/s)": 1.459024 }, { "epoch": 4.052739814061094, "grad_norm": 4.002673149108887, "learning_rate": 8.597651935347872e-06, "loss": 0.2957625865936279, "memory(GiB)": 73.83, "step": 94595, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.459035 }, { "epoch": 4.052954029390343, "grad_norm": 2.156463146209717, "learning_rate": 8.593879204260747e-06, "loss": 0.23447415828704835, "memory(GiB)": 73.83, "step": 94600, "token_acc": 0.9579579579579579, "train_speed(iter/s)": 1.459033 }, { "epoch": 4.0531682447195925, "grad_norm": 0.8658608794212341, "learning_rate": 8.590107223284344e-06, "loss": 0.35891697406768797, "memory(GiB)": 73.83, "step": 94605, "token_acc": 0.903010033444816, "train_speed(iter/s)": 1.459021 }, { "epoch": 4.053382460048841, "grad_norm": 0.5919279456138611, "learning_rate": 8.586335992486994e-06, "loss": 0.3789587259292603, "memory(GiB)": 73.83, "step": 94610, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.459023 }, { "epoch": 4.05359667537809, "grad_norm": 0.3200875520706177, "learning_rate": 8.582565511937019e-06, "loss": 0.3572386741638184, "memory(GiB)": 73.83, "step": 94615, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.459037 }, { "epoch": 4.053810890707339, "grad_norm": 2.280395269393921, "learning_rate": 8.578795781702703e-06, "loss": 0.2881666660308838, "memory(GiB)": 73.83, "step": 94620, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.459039 }, { "epoch": 4.054025106036588, "grad_norm": 1.9400264024734497, "learning_rate": 8.575026801852375e-06, "loss": 0.39533085823059083, "memory(GiB)": 73.83, "step": 94625, "token_acc": 0.9183098591549296, "train_speed(iter/s)": 1.45904 }, { "epoch": 4.054239321365837, "grad_norm": 2.067600727081299, "learning_rate": 8.571258572454299e-06, "loss": 0.44566993713378905, "memory(GiB)": 73.83, "step": 94630, "token_acc": 0.9076305220883534, "train_speed(iter/s)": 1.459062 }, { "epoch": 4.054453536695086, "grad_norm": 2.4379496574401855, "learning_rate": 8.567491093576735e-06, "loss": 0.3115605592727661, "memory(GiB)": 73.83, "step": 94635, "token_acc": 0.9190283400809717, "train_speed(iter/s)": 1.459063 }, { "epoch": 4.054667752024335, "grad_norm": 3.550441265106201, "learning_rate": 8.563724365287934e-06, "loss": 0.2508277893066406, "memory(GiB)": 73.83, "step": 94640, "token_acc": 0.9453781512605042, "train_speed(iter/s)": 1.459063 }, { "epoch": 4.054881967353584, "grad_norm": 3.1987504959106445, "learning_rate": 8.559958387656125e-06, "loss": 0.4144002914428711, "memory(GiB)": 73.83, "step": 94645, "token_acc": 0.9011976047904192, "train_speed(iter/s)": 1.459067 }, { "epoch": 4.055096182682833, "grad_norm": 2.1679561138153076, "learning_rate": 8.556193160749549e-06, "loss": 0.18547964096069336, "memory(GiB)": 73.83, "step": 94650, "token_acc": 0.9548611111111112, "train_speed(iter/s)": 1.459071 }, { "epoch": 4.0553103980120815, "grad_norm": 2.020491600036621, "learning_rate": 8.552428684636416e-06, "loss": 0.21617679595947265, "memory(GiB)": 73.83, "step": 94655, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.459074 }, { "epoch": 4.055524613341331, "grad_norm": 3.6884982585906982, "learning_rate": 8.548664959384922e-06, "loss": 0.3074381351470947, "memory(GiB)": 73.83, "step": 94660, "token_acc": 0.9574468085106383, "train_speed(iter/s)": 1.459075 }, { "epoch": 4.05573882867058, "grad_norm": 2.6279430389404297, "learning_rate": 8.544901985063247e-06, "loss": 0.3600595951080322, "memory(GiB)": 73.83, "step": 94665, "token_acc": 0.9235668789808917, "train_speed(iter/s)": 1.45908 }, { "epoch": 4.055953043999828, "grad_norm": 4.559435844421387, "learning_rate": 8.541139761739559e-06, "loss": 0.3931111812591553, "memory(GiB)": 73.83, "step": 94670, "token_acc": 0.9158878504672897, "train_speed(iter/s)": 1.459091 }, { "epoch": 4.056167259329078, "grad_norm": 7.208067417144775, "learning_rate": 8.537378289482006e-06, "loss": 0.31506805419921874, "memory(GiB)": 73.83, "step": 94675, "token_acc": 0.9267399267399268, "train_speed(iter/s)": 1.459096 }, { "epoch": 4.056381474658327, "grad_norm": 2.3270018100738525, "learning_rate": 8.533617568358749e-06, "loss": 0.14976625442504882, "memory(GiB)": 73.83, "step": 94680, "token_acc": 0.9591078066914498, "train_speed(iter/s)": 1.459098 }, { "epoch": 4.056595689987575, "grad_norm": 2.189542055130005, "learning_rate": 8.529857598437912e-06, "loss": 0.3140183210372925, "memory(GiB)": 73.83, "step": 94685, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.459099 }, { "epoch": 4.0568099053168245, "grad_norm": 0.6800786256790161, "learning_rate": 8.526098379787606e-06, "loss": 0.34580144882202146, "memory(GiB)": 73.83, "step": 94690, "token_acc": 0.9483695652173914, "train_speed(iter/s)": 1.459105 }, { "epoch": 4.057024120646074, "grad_norm": 2.3894965648651123, "learning_rate": 8.522339912475936e-06, "loss": 0.2695406436920166, "memory(GiB)": 73.83, "step": 94695, "token_acc": 0.9440559440559441, "train_speed(iter/s)": 1.459105 }, { "epoch": 4.057238335975322, "grad_norm": 4.342015266418457, "learning_rate": 8.51858219657098e-06, "loss": 0.400211763381958, "memory(GiB)": 73.83, "step": 94700, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.45911 }, { "epoch": 4.057452551304571, "grad_norm": 4.001551628112793, "learning_rate": 8.514825232140816e-06, "loss": 0.24038348197937012, "memory(GiB)": 73.83, "step": 94705, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.459114 }, { "epoch": 4.057666766633821, "grad_norm": 4.127545356750488, "learning_rate": 8.511069019253515e-06, "loss": 0.32306675910949706, "memory(GiB)": 73.83, "step": 94710, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.459118 }, { "epoch": 4.057880981963069, "grad_norm": 2.9358808994293213, "learning_rate": 8.507313557977121e-06, "loss": 0.41014962196350097, "memory(GiB)": 73.83, "step": 94715, "token_acc": 0.9006410256410257, "train_speed(iter/s)": 1.459117 }, { "epoch": 4.058095197292318, "grad_norm": 2.040619134902954, "learning_rate": 8.503558848379656e-06, "loss": 0.13885436058044434, "memory(GiB)": 73.83, "step": 94720, "token_acc": 0.9594594594594594, "train_speed(iter/s)": 1.459116 }, { "epoch": 4.0583094126215675, "grad_norm": 5.856706619262695, "learning_rate": 8.499804890529157e-06, "loss": 0.5395654201507568, "memory(GiB)": 73.83, "step": 94725, "token_acc": 0.8934707903780069, "train_speed(iter/s)": 1.45912 }, { "epoch": 4.058523627950816, "grad_norm": 1.9229322671890259, "learning_rate": 8.496051684493627e-06, "loss": 0.41640310287475585, "memory(GiB)": 73.83, "step": 94730, "token_acc": 0.9058441558441559, "train_speed(iter/s)": 1.459117 }, { "epoch": 4.058737843280065, "grad_norm": 3.882939577102661, "learning_rate": 8.492299230341039e-06, "loss": 0.1514428973197937, "memory(GiB)": 73.83, "step": 94735, "token_acc": 0.9693593314763231, "train_speed(iter/s)": 1.459118 }, { "epoch": 4.058952058609314, "grad_norm": 2.7537968158721924, "learning_rate": 8.488547528139402e-06, "loss": 0.45378456115722654, "memory(GiB)": 73.83, "step": 94740, "token_acc": 0.9096573208722741, "train_speed(iter/s)": 1.459123 }, { "epoch": 4.059166273938563, "grad_norm": 2.9966139793395996, "learning_rate": 8.484796577956666e-06, "loss": 0.25401978492736815, "memory(GiB)": 73.83, "step": 94745, "token_acc": 0.9386973180076629, "train_speed(iter/s)": 1.459126 }, { "epoch": 4.059380489267812, "grad_norm": 1.66119384765625, "learning_rate": 8.481046379860785e-06, "loss": 0.2513139247894287, "memory(GiB)": 73.83, "step": 94750, "token_acc": 0.9463722397476341, "train_speed(iter/s)": 1.459131 }, { "epoch": 4.059594704597061, "grad_norm": 1.2487210035324097, "learning_rate": 8.477296933919693e-06, "loss": 0.37492148876190184, "memory(GiB)": 73.83, "step": 94755, "token_acc": 0.9051383399209486, "train_speed(iter/s)": 1.459148 }, { "epoch": 4.05980891992631, "grad_norm": 2.165255308151245, "learning_rate": 8.473548240201323e-06, "loss": 0.28378450870513916, "memory(GiB)": 73.83, "step": 94760, "token_acc": 0.9386973180076629, "train_speed(iter/s)": 1.459154 }, { "epoch": 4.060023135255559, "grad_norm": 5.584085941314697, "learning_rate": 8.469800298773567e-06, "loss": 0.41950459480285646, "memory(GiB)": 73.83, "step": 94765, "token_acc": 0.909433962264151, "train_speed(iter/s)": 1.459156 }, { "epoch": 4.060237350584808, "grad_norm": 2.986248016357422, "learning_rate": 8.466053109704353e-06, "loss": 0.40255074501037597, "memory(GiB)": 73.83, "step": 94770, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.459159 }, { "epoch": 4.0604515659140565, "grad_norm": 3.4316534996032715, "learning_rate": 8.462306673061548e-06, "loss": 0.701026725769043, "memory(GiB)": 73.83, "step": 94775, "token_acc": 0.840782122905028, "train_speed(iter/s)": 1.459156 }, { "epoch": 4.060665781243306, "grad_norm": 3.6478524208068848, "learning_rate": 8.458560988913022e-06, "loss": 0.13635393381118774, "memory(GiB)": 73.83, "step": 94780, "token_acc": 0.9547169811320755, "train_speed(iter/s)": 1.459161 }, { "epoch": 4.060879996572555, "grad_norm": 3.400446653366089, "learning_rate": 8.45481605732663e-06, "loss": 0.47162456512451173, "memory(GiB)": 73.83, "step": 94785, "token_acc": 0.9, "train_speed(iter/s)": 1.45917 }, { "epoch": 4.061094211901803, "grad_norm": 2.4036788940429688, "learning_rate": 8.451071878370209e-06, "loss": 0.305405592918396, "memory(GiB)": 73.83, "step": 94790, "token_acc": 0.9294478527607362, "train_speed(iter/s)": 1.459173 }, { "epoch": 4.061308427231053, "grad_norm": 3.6939849853515625, "learning_rate": 8.447328452111597e-06, "loss": 0.4589071273803711, "memory(GiB)": 73.83, "step": 94795, "token_acc": 0.8943894389438944, "train_speed(iter/s)": 1.45917 }, { "epoch": 4.061522642560302, "grad_norm": 4.448829650878906, "learning_rate": 8.443585778618618e-06, "loss": 0.5423956394195557, "memory(GiB)": 73.83, "step": 94800, "token_acc": 0.8685897435897436, "train_speed(iter/s)": 1.459171 }, { "epoch": 4.06173685788955, "grad_norm": 3.5462143421173096, "learning_rate": 8.439843857959074e-06, "loss": 0.29996528625488283, "memory(GiB)": 73.83, "step": 94805, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.459169 }, { "epoch": 4.0619510732188, "grad_norm": 3.9275360107421875, "learning_rate": 8.436102690200737e-06, "loss": 0.5358775138854981, "memory(GiB)": 73.83, "step": 94810, "token_acc": 0.8852941176470588, "train_speed(iter/s)": 1.459174 }, { "epoch": 4.062165288548049, "grad_norm": 2.4594273567199707, "learning_rate": 8.432362275411392e-06, "loss": 0.1826917052268982, "memory(GiB)": 73.83, "step": 94815, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.459177 }, { "epoch": 4.062379503877297, "grad_norm": 4.148278713226318, "learning_rate": 8.428622613658788e-06, "loss": 0.4634083271026611, "memory(GiB)": 73.83, "step": 94820, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.459184 }, { "epoch": 4.062593719206546, "grad_norm": 2.448709487915039, "learning_rate": 8.424883705010689e-06, "loss": 0.3579734802246094, "memory(GiB)": 73.83, "step": 94825, "token_acc": 0.9131736526946108, "train_speed(iter/s)": 1.459194 }, { "epoch": 4.062807934535796, "grad_norm": 2.4377553462982178, "learning_rate": 8.421145549534826e-06, "loss": 0.169279944896698, "memory(GiB)": 73.83, "step": 94830, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.459199 }, { "epoch": 4.063022149865044, "grad_norm": 2.1212308406829834, "learning_rate": 8.417408147298916e-06, "loss": 0.3209673881530762, "memory(GiB)": 73.83, "step": 94835, "token_acc": 0.9294478527607362, "train_speed(iter/s)": 1.459202 }, { "epoch": 4.063236365194293, "grad_norm": 1.0430527925491333, "learning_rate": 8.413671498370667e-06, "loss": 0.32699849605560305, "memory(GiB)": 73.83, "step": 94840, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.459201 }, { "epoch": 4.063450580523543, "grad_norm": 3.1724209785461426, "learning_rate": 8.409935602817764e-06, "loss": 0.3317121028900146, "memory(GiB)": 73.83, "step": 94845, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.459214 }, { "epoch": 4.063664795852791, "grad_norm": 3.392547369003296, "learning_rate": 8.406200460707885e-06, "loss": 0.33837237358093264, "memory(GiB)": 73.83, "step": 94850, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.459219 }, { "epoch": 4.06387901118204, "grad_norm": 5.128413200378418, "learning_rate": 8.402466072108716e-06, "loss": 0.48804798126220705, "memory(GiB)": 73.83, "step": 94855, "token_acc": 0.8746438746438746, "train_speed(iter/s)": 1.459219 }, { "epoch": 4.0640932265112895, "grad_norm": 4.09347677230835, "learning_rate": 8.398732437087892e-06, "loss": 0.5727322101593018, "memory(GiB)": 73.83, "step": 94860, "token_acc": 0.8942307692307693, "train_speed(iter/s)": 1.459224 }, { "epoch": 4.064307441840538, "grad_norm": 6.22737979888916, "learning_rate": 8.394999555713045e-06, "loss": 0.6408977508544922, "memory(GiB)": 73.83, "step": 94865, "token_acc": 0.844311377245509, "train_speed(iter/s)": 1.459245 }, { "epoch": 4.064521657169787, "grad_norm": 0.5791308879852295, "learning_rate": 8.391267428051819e-06, "loss": 0.20346412658691407, "memory(GiB)": 73.83, "step": 94870, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.459249 }, { "epoch": 4.064735872499036, "grad_norm": 2.399118423461914, "learning_rate": 8.387536054171818e-06, "loss": 0.5012077331542969, "memory(GiB)": 73.83, "step": 94875, "token_acc": 0.9108910891089109, "train_speed(iter/s)": 1.459261 }, { "epoch": 4.064950087828285, "grad_norm": 0.32009878754615784, "learning_rate": 8.383805434140623e-06, "loss": 0.40251898765563965, "memory(GiB)": 73.83, "step": 94880, "token_acc": 0.9125874125874126, "train_speed(iter/s)": 1.459271 }, { "epoch": 4.065164303157534, "grad_norm": 2.175851583480835, "learning_rate": 8.380075568025847e-06, "loss": 0.42680039405822756, "memory(GiB)": 73.83, "step": 94885, "token_acc": 0.9009009009009009, "train_speed(iter/s)": 1.459277 }, { "epoch": 4.065378518486783, "grad_norm": 1.743965744972229, "learning_rate": 8.376346455895046e-06, "loss": 0.4447620391845703, "memory(GiB)": 73.83, "step": 94890, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.459285 }, { "epoch": 4.065592733816032, "grad_norm": 1.2708451747894287, "learning_rate": 8.37261809781577e-06, "loss": 0.3749372482299805, "memory(GiB)": 73.83, "step": 94895, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.459288 }, { "epoch": 4.065806949145281, "grad_norm": 2.038888931274414, "learning_rate": 8.368890493855568e-06, "loss": 0.6000860691070556, "memory(GiB)": 73.83, "step": 94900, "token_acc": 0.8580246913580247, "train_speed(iter/s)": 1.459285 }, { "epoch": 4.06602116447453, "grad_norm": 2.374433755874634, "learning_rate": 8.365163644081969e-06, "loss": 0.235890793800354, "memory(GiB)": 73.83, "step": 94905, "token_acc": 0.9427480916030534, "train_speed(iter/s)": 1.459284 }, { "epoch": 4.0662353798037785, "grad_norm": 3.337151527404785, "learning_rate": 8.361437548562478e-06, "loss": 0.5997297763824463, "memory(GiB)": 73.83, "step": 94910, "token_acc": 0.8596491228070176, "train_speed(iter/s)": 1.459294 }, { "epoch": 4.066449595133028, "grad_norm": 4.57769250869751, "learning_rate": 8.357712207364615e-06, "loss": 0.21681790351867675, "memory(GiB)": 73.83, "step": 94915, "token_acc": 0.9501661129568106, "train_speed(iter/s)": 1.459298 }, { "epoch": 4.066663810462277, "grad_norm": 3.5142886638641357, "learning_rate": 8.353987620555859e-06, "loss": 0.28319242000579836, "memory(GiB)": 73.83, "step": 94920, "token_acc": 0.941747572815534, "train_speed(iter/s)": 1.4593 }, { "epoch": 4.066878025791525, "grad_norm": 4.440555572509766, "learning_rate": 8.350263788203682e-06, "loss": 0.3210457801818848, "memory(GiB)": 73.83, "step": 94925, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.459293 }, { "epoch": 4.067092241120775, "grad_norm": 1.9844880104064941, "learning_rate": 8.34654071037555e-06, "loss": 0.18674273490905763, "memory(GiB)": 73.83, "step": 94930, "token_acc": 0.9575289575289575, "train_speed(iter/s)": 1.459298 }, { "epoch": 4.067306456450024, "grad_norm": 1.5410308837890625, "learning_rate": 8.342818387138895e-06, "loss": 0.20923206806182862, "memory(GiB)": 73.83, "step": 94935, "token_acc": 0.9484848484848485, "train_speed(iter/s)": 1.459296 }, { "epoch": 4.067520671779272, "grad_norm": 1.9704434871673584, "learning_rate": 8.339096818561165e-06, "loss": 0.14085712432861328, "memory(GiB)": 73.83, "step": 94940, "token_acc": 0.9638157894736842, "train_speed(iter/s)": 1.459296 }, { "epoch": 4.0677348871085215, "grad_norm": 6.038177967071533, "learning_rate": 8.335376004709784e-06, "loss": 0.6667991161346436, "memory(GiB)": 73.83, "step": 94945, "token_acc": 0.8655737704918033, "train_speed(iter/s)": 1.459294 }, { "epoch": 4.067949102437771, "grad_norm": 4.514578819274902, "learning_rate": 8.331655945652155e-06, "loss": 0.3563373565673828, "memory(GiB)": 73.83, "step": 94950, "token_acc": 0.9291553133514986, "train_speed(iter/s)": 1.459293 }, { "epoch": 4.068163317767019, "grad_norm": 1.893851637840271, "learning_rate": 8.327936641455663e-06, "loss": 0.18136779069900513, "memory(GiB)": 73.83, "step": 94955, "token_acc": 0.9665551839464883, "train_speed(iter/s)": 1.459304 }, { "epoch": 4.068377533096268, "grad_norm": 3.708503246307373, "learning_rate": 8.324218092187696e-06, "loss": 0.3468088865280151, "memory(GiB)": 73.83, "step": 94960, "token_acc": 0.9104477611940298, "train_speed(iter/s)": 1.459303 }, { "epoch": 4.068591748425518, "grad_norm": 2.8483102321624756, "learning_rate": 8.320500297915595e-06, "loss": 0.4846210479736328, "memory(GiB)": 73.83, "step": 94965, "token_acc": 0.8829787234042553, "train_speed(iter/s)": 1.459304 }, { "epoch": 4.068805963754766, "grad_norm": 2.1402676105499268, "learning_rate": 8.316783258706745e-06, "loss": 0.3061125040054321, "memory(GiB)": 73.83, "step": 94970, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.45931 }, { "epoch": 4.069020179084015, "grad_norm": 1.9672914743423462, "learning_rate": 8.31306697462847e-06, "loss": 0.3152065992355347, "memory(GiB)": 73.83, "step": 94975, "token_acc": 0.9173789173789174, "train_speed(iter/s)": 1.459314 }, { "epoch": 4.0692343944132645, "grad_norm": 4.772182464599609, "learning_rate": 8.309351445748086e-06, "loss": 0.42961735725402833, "memory(GiB)": 73.83, "step": 94980, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.459315 }, { "epoch": 4.069448609742513, "grad_norm": 4.24347448348999, "learning_rate": 8.305636672132916e-06, "loss": 0.3930068492889404, "memory(GiB)": 73.83, "step": 94985, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.459336 }, { "epoch": 4.069662825071762, "grad_norm": 4.3791279792785645, "learning_rate": 8.301922653850247e-06, "loss": 0.13179302215576172, "memory(GiB)": 73.83, "step": 94990, "token_acc": 0.9728682170542635, "train_speed(iter/s)": 1.459337 }, { "epoch": 4.069877040401011, "grad_norm": 4.188531875610352, "learning_rate": 8.298209390967355e-06, "loss": 0.5280294418334961, "memory(GiB)": 73.83, "step": 94995, "token_acc": 0.8933717579250721, "train_speed(iter/s)": 1.459358 }, { "epoch": 4.07009125573026, "grad_norm": 3.364837646484375, "learning_rate": 8.294496883551528e-06, "loss": 0.31493828296661375, "memory(GiB)": 73.83, "step": 95000, "token_acc": 0.9397163120567376, "train_speed(iter/s)": 1.459357 }, { "epoch": 4.07009125573026, "eval_loss": 2.5441412925720215, "eval_runtime": 11.4634, "eval_samples_per_second": 8.723, "eval_steps_per_second": 8.723, "eval_token_acc": 0.45545796737766625, "step": 95000 }, { "epoch": 4.070305471059509, "grad_norm": 1.3328547477722168, "learning_rate": 8.290785131670014e-06, "loss": 0.32445602416992186, "memory(GiB)": 73.83, "step": 95005, "token_acc": 0.5992939099735216, "train_speed(iter/s)": 1.459082 }, { "epoch": 4.070519686388758, "grad_norm": 3.5792291164398193, "learning_rate": 8.287074135390061e-06, "loss": 0.32961244583129884, "memory(GiB)": 73.83, "step": 95010, "token_acc": 0.9192546583850931, "train_speed(iter/s)": 1.45908 }, { "epoch": 4.070733901718007, "grad_norm": 3.5275256633758545, "learning_rate": 8.283363894778868e-06, "loss": 0.39865407943725584, "memory(GiB)": 73.83, "step": 95015, "token_acc": 0.9388489208633094, "train_speed(iter/s)": 1.45908 }, { "epoch": 4.070948117047256, "grad_norm": 3.5896222591400146, "learning_rate": 8.279654409903687e-06, "loss": 0.5170844554901123, "memory(GiB)": 73.83, "step": 95020, "token_acc": 0.8909090909090909, "train_speed(iter/s)": 1.459089 }, { "epoch": 4.071162332376505, "grad_norm": 5.376219749450684, "learning_rate": 8.275945680831693e-06, "loss": 0.4055911064147949, "memory(GiB)": 73.83, "step": 95025, "token_acc": 0.9113475177304965, "train_speed(iter/s)": 1.459091 }, { "epoch": 4.0713765477057535, "grad_norm": 0.9616581797599792, "learning_rate": 8.272237707630092e-06, "loss": 0.20093767642974852, "memory(GiB)": 73.83, "step": 95030, "token_acc": 0.957983193277311, "train_speed(iter/s)": 1.459095 }, { "epoch": 4.071590763035003, "grad_norm": 5.09306001663208, "learning_rate": 8.26853049036605e-06, "loss": 0.3072676181793213, "memory(GiB)": 73.83, "step": 95035, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.459099 }, { "epoch": 4.071804978364252, "grad_norm": 5.325514316558838, "learning_rate": 8.264824029106727e-06, "loss": 0.509259033203125, "memory(GiB)": 73.83, "step": 95040, "token_acc": 0.8932584269662921, "train_speed(iter/s)": 1.459104 }, { "epoch": 4.0720191936935, "grad_norm": 0.48458221554756165, "learning_rate": 8.261118323919265e-06, "loss": 0.3933782339096069, "memory(GiB)": 73.83, "step": 95045, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.459106 }, { "epoch": 4.07223340902275, "grad_norm": 1.9528136253356934, "learning_rate": 8.257413374870804e-06, "loss": 0.2033074140548706, "memory(GiB)": 73.83, "step": 95050, "token_acc": 0.9594594594594594, "train_speed(iter/s)": 1.459111 }, { "epoch": 4.072447624351999, "grad_norm": 3.917776584625244, "learning_rate": 8.25370918202844e-06, "loss": 0.25169711112976073, "memory(GiB)": 73.83, "step": 95055, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.459114 }, { "epoch": 4.072661839681247, "grad_norm": 2.4241111278533936, "learning_rate": 8.25000574545931e-06, "loss": 0.25664567947387695, "memory(GiB)": 73.83, "step": 95060, "token_acc": 0.9566787003610109, "train_speed(iter/s)": 1.459114 }, { "epoch": 4.0728760550104965, "grad_norm": 0.07838138937950134, "learning_rate": 8.246303065230488e-06, "loss": 0.14446728229522704, "memory(GiB)": 73.83, "step": 95065, "token_acc": 0.9609120521172638, "train_speed(iter/s)": 1.459114 }, { "epoch": 4.073090270339746, "grad_norm": 4.6520562171936035, "learning_rate": 8.242601141409056e-06, "loss": 0.4601278781890869, "memory(GiB)": 73.83, "step": 95070, "token_acc": 0.9015384615384615, "train_speed(iter/s)": 1.45912 }, { "epoch": 4.073304485668994, "grad_norm": 1.564586877822876, "learning_rate": 8.238899974062076e-06, "loss": 0.30851476192474364, "memory(GiB)": 73.83, "step": 95075, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.459122 }, { "epoch": 4.073518700998243, "grad_norm": 4.889163494110107, "learning_rate": 8.235199563256585e-06, "loss": 0.6681359291076661, "memory(GiB)": 73.83, "step": 95080, "token_acc": 0.8715596330275229, "train_speed(iter/s)": 1.459122 }, { "epoch": 4.073732916327493, "grad_norm": 2.519007921218872, "learning_rate": 8.231499909059642e-06, "loss": 0.26997199058532717, "memory(GiB)": 73.83, "step": 95085, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.459127 }, { "epoch": 4.073947131656741, "grad_norm": 0.774601936340332, "learning_rate": 8.227801011538256e-06, "loss": 0.522744083404541, "memory(GiB)": 74.99, "step": 95090, "token_acc": 0.8810975609756098, "train_speed(iter/s)": 1.459122 }, { "epoch": 4.07416134698599, "grad_norm": 3.640941858291626, "learning_rate": 8.224102870759448e-06, "loss": 0.32269361019134524, "memory(GiB)": 74.99, "step": 95095, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.459122 }, { "epoch": 4.07437556231524, "grad_norm": 3.1225202083587646, "learning_rate": 8.2204054867902e-06, "loss": 0.1981595754623413, "memory(GiB)": 74.99, "step": 95100, "token_acc": 0.9528619528619529, "train_speed(iter/s)": 1.459124 }, { "epoch": 4.074589777644488, "grad_norm": 2.0717341899871826, "learning_rate": 8.216708859697508e-06, "loss": 0.11667983531951905, "memory(GiB)": 74.99, "step": 95105, "token_acc": 0.9644268774703557, "train_speed(iter/s)": 1.459125 }, { "epoch": 4.074803992973737, "grad_norm": 2.615107297897339, "learning_rate": 8.213012989548313e-06, "loss": 0.4383289337158203, "memory(GiB)": 74.99, "step": 95110, "token_acc": 0.9153846153846154, "train_speed(iter/s)": 1.45913 }, { "epoch": 4.0750182083029864, "grad_norm": 6.02342414855957, "learning_rate": 8.209317876409601e-06, "loss": 0.32882449626922605, "memory(GiB)": 74.99, "step": 95115, "token_acc": 0.914179104477612, "train_speed(iter/s)": 1.459134 }, { "epoch": 4.075232423632235, "grad_norm": 3.533629894256592, "learning_rate": 8.205623520348299e-06, "loss": 0.42243137359619143, "memory(GiB)": 74.99, "step": 95120, "token_acc": 0.9140625, "train_speed(iter/s)": 1.459137 }, { "epoch": 4.075446638961484, "grad_norm": 2.982111692428589, "learning_rate": 8.201929921431334e-06, "loss": 0.3868952989578247, "memory(GiB)": 74.99, "step": 95125, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.459139 }, { "epoch": 4.075660854290733, "grad_norm": 3.8286008834838867, "learning_rate": 8.198237079725613e-06, "loss": 0.5449553489685058, "memory(GiB)": 74.99, "step": 95130, "token_acc": 0.8960244648318043, "train_speed(iter/s)": 1.459149 }, { "epoch": 4.075875069619982, "grad_norm": 0.7302953600883484, "learning_rate": 8.194544995298048e-06, "loss": 0.21487174034118653, "memory(GiB)": 74.99, "step": 95135, "token_acc": 0.9556451612903226, "train_speed(iter/s)": 1.459159 }, { "epoch": 4.076089284949231, "grad_norm": 2.6025354862213135, "learning_rate": 8.190853668215498e-06, "loss": 0.30245113372802734, "memory(GiB)": 74.99, "step": 95140, "token_acc": 0.9287925696594427, "train_speed(iter/s)": 1.459163 }, { "epoch": 4.07630350027848, "grad_norm": 1.5727744102478027, "learning_rate": 8.18716309854487e-06, "loss": 0.35186138153076174, "memory(GiB)": 74.99, "step": 95145, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.459176 }, { "epoch": 4.076517715607729, "grad_norm": 4.26010799407959, "learning_rate": 8.183473286353005e-06, "loss": 0.11042428016662598, "memory(GiB)": 74.99, "step": 95150, "token_acc": 0.9708029197080292, "train_speed(iter/s)": 1.459181 }, { "epoch": 4.076731930936978, "grad_norm": 0.20303519070148468, "learning_rate": 8.179784231706749e-06, "loss": 0.4326455593109131, "memory(GiB)": 74.99, "step": 95155, "token_acc": 0.9006622516556292, "train_speed(iter/s)": 1.459184 }, { "epoch": 4.076946146266227, "grad_norm": 1.5999103784561157, "learning_rate": 8.176095934672933e-06, "loss": 0.47518601417541506, "memory(GiB)": 74.99, "step": 95160, "token_acc": 0.9204545454545454, "train_speed(iter/s)": 1.459202 }, { "epoch": 4.0771603615954755, "grad_norm": 3.448047637939453, "learning_rate": 8.172408395318359e-06, "loss": 0.214288330078125, "memory(GiB)": 74.99, "step": 95165, "token_acc": 0.9475806451612904, "train_speed(iter/s)": 1.459204 }, { "epoch": 4.077374576924725, "grad_norm": 3.5919671058654785, "learning_rate": 8.168721613709846e-06, "loss": 0.20907297134399414, "memory(GiB)": 74.99, "step": 95170, "token_acc": 0.9507042253521126, "train_speed(iter/s)": 1.45921 }, { "epoch": 4.077588792253974, "grad_norm": 5.9466681480407715, "learning_rate": 8.165035589914193e-06, "loss": 0.36966121196746826, "memory(GiB)": 74.99, "step": 95175, "token_acc": 0.9202453987730062, "train_speed(iter/s)": 1.459215 }, { "epoch": 4.077803007583222, "grad_norm": 0.22602276504039764, "learning_rate": 8.161350323998169e-06, "loss": 0.15633761882781982, "memory(GiB)": 74.99, "step": 95180, "token_acc": 0.9621848739495799, "train_speed(iter/s)": 1.459215 }, { "epoch": 4.078017222912472, "grad_norm": 2.5482382774353027, "learning_rate": 8.157665816028525e-06, "loss": 0.1974120020866394, "memory(GiB)": 74.99, "step": 95185, "token_acc": 0.9617021276595744, "train_speed(iter/s)": 1.459219 }, { "epoch": 4.078231438241721, "grad_norm": 4.609391689300537, "learning_rate": 8.153982066072018e-06, "loss": 0.38431720733642577, "memory(GiB)": 74.99, "step": 95190, "token_acc": 0.9078498293515358, "train_speed(iter/s)": 1.459219 }, { "epoch": 4.078445653570969, "grad_norm": 4.873157024383545, "learning_rate": 8.150299074195383e-06, "loss": 0.29031686782836913, "memory(GiB)": 74.99, "step": 95195, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.459221 }, { "epoch": 4.0786598689002185, "grad_norm": 5.394603252410889, "learning_rate": 8.146616840465326e-06, "loss": 0.3296787738800049, "memory(GiB)": 74.99, "step": 95200, "token_acc": 0.9221311475409836, "train_speed(iter/s)": 1.459227 }, { "epoch": 4.078874084229468, "grad_norm": 2.6882221698760986, "learning_rate": 8.142935364948579e-06, "loss": 0.37382824420928956, "memory(GiB)": 74.99, "step": 95205, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.459225 }, { "epoch": 4.079088299558716, "grad_norm": 2.917966365814209, "learning_rate": 8.13925464771182e-06, "loss": 0.2521966934204102, "memory(GiB)": 74.99, "step": 95210, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.459226 }, { "epoch": 4.079302514887965, "grad_norm": 0.8527376055717468, "learning_rate": 8.135574688821734e-06, "loss": 0.2530005693435669, "memory(GiB)": 74.99, "step": 95215, "token_acc": 0.9464285714285714, "train_speed(iter/s)": 1.459232 }, { "epoch": 4.079516730217215, "grad_norm": 3.2810325622558594, "learning_rate": 8.131895488344981e-06, "loss": 0.39339699745178225, "memory(GiB)": 74.99, "step": 95220, "token_acc": 0.9236641221374046, "train_speed(iter/s)": 1.459245 }, { "epoch": 4.079730945546463, "grad_norm": 2.70963191986084, "learning_rate": 8.128217046348207e-06, "loss": 0.370645546913147, "memory(GiB)": 74.99, "step": 95225, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.459244 }, { "epoch": 4.079945160875712, "grad_norm": 3.019094467163086, "learning_rate": 8.124539362898076e-06, "loss": 0.2622556686401367, "memory(GiB)": 74.99, "step": 95230, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.459242 }, { "epoch": 4.0801593762049615, "grad_norm": 2.9281294345855713, "learning_rate": 8.120862438061183e-06, "loss": 0.21417136192321778, "memory(GiB)": 74.99, "step": 95235, "token_acc": 0.9531772575250836, "train_speed(iter/s)": 1.459245 }, { "epoch": 4.08037359153421, "grad_norm": 2.418473720550537, "learning_rate": 8.117186271904158e-06, "loss": 0.25544157028198244, "memory(GiB)": 74.99, "step": 95240, "token_acc": 0.9492537313432836, "train_speed(iter/s)": 1.459249 }, { "epoch": 4.080587806863459, "grad_norm": 0.7573021054267883, "learning_rate": 8.113510864493602e-06, "loss": 0.2054819345474243, "memory(GiB)": 74.99, "step": 95245, "token_acc": 0.9628647214854111, "train_speed(iter/s)": 1.459253 }, { "epoch": 4.080802022192708, "grad_norm": 5.688795566558838, "learning_rate": 8.109836215896083e-06, "loss": 0.2360250473022461, "memory(GiB)": 74.99, "step": 95250, "token_acc": 0.95, "train_speed(iter/s)": 1.459266 }, { "epoch": 4.081016237521957, "grad_norm": 4.495298385620117, "learning_rate": 8.106162326178168e-06, "loss": 0.31391894817352295, "memory(GiB)": 74.99, "step": 95255, "token_acc": 0.948, "train_speed(iter/s)": 1.459269 }, { "epoch": 4.081230452851206, "grad_norm": 0.11899235099554062, "learning_rate": 8.102489195406432e-06, "loss": 0.3744157552719116, "memory(GiB)": 74.99, "step": 95260, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.459266 }, { "epoch": 4.081444668180455, "grad_norm": 2.6136069297790527, "learning_rate": 8.09881682364741e-06, "loss": 0.38498196601867674, "memory(GiB)": 74.99, "step": 95265, "token_acc": 0.8933823529411765, "train_speed(iter/s)": 1.459262 }, { "epoch": 4.081658883509704, "grad_norm": 0.789708137512207, "learning_rate": 8.095145210967624e-06, "loss": 0.23293497562408447, "memory(GiB)": 74.99, "step": 95270, "token_acc": 0.9450171821305842, "train_speed(iter/s)": 1.459267 }, { "epoch": 4.081873098838953, "grad_norm": 3.0887527465820312, "learning_rate": 8.091474357433598e-06, "loss": 0.4267579555511475, "memory(GiB)": 74.99, "step": 95275, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.459274 }, { "epoch": 4.082087314168202, "grad_norm": 2.465855836868286, "learning_rate": 8.087804263111825e-06, "loss": 0.5578618049621582, "memory(GiB)": 74.99, "step": 95280, "token_acc": 0.8885448916408669, "train_speed(iter/s)": 1.459286 }, { "epoch": 4.0823015294974505, "grad_norm": 3.8573174476623535, "learning_rate": 8.084134928068782e-06, "loss": 0.2759145021438599, "memory(GiB)": 74.99, "step": 95285, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.459286 }, { "epoch": 4.0825157448267, "grad_norm": 2.8330047130584717, "learning_rate": 8.080466352370968e-06, "loss": 0.26651673316955565, "memory(GiB)": 74.99, "step": 95290, "token_acc": 0.9539748953974896, "train_speed(iter/s)": 1.459286 }, { "epoch": 4.082729960155949, "grad_norm": 3.3047332763671875, "learning_rate": 8.076798536084828e-06, "loss": 0.4084175109863281, "memory(GiB)": 74.99, "step": 95295, "token_acc": 0.904320987654321, "train_speed(iter/s)": 1.459282 }, { "epoch": 4.082944175485197, "grad_norm": 5.1594719886779785, "learning_rate": 8.073131479276807e-06, "loss": 0.26407885551452637, "memory(GiB)": 74.99, "step": 95300, "token_acc": 0.939622641509434, "train_speed(iter/s)": 1.459285 }, { "epoch": 4.083158390814447, "grad_norm": 6.488198757171631, "learning_rate": 8.069465182013347e-06, "loss": 0.33022513389587405, "memory(GiB)": 74.99, "step": 95305, "token_acc": 0.9342105263157895, "train_speed(iter/s)": 1.459283 }, { "epoch": 4.083372606143696, "grad_norm": 5.509545803070068, "learning_rate": 8.065799644360844e-06, "loss": 0.49832968711853026, "memory(GiB)": 74.99, "step": 95310, "token_acc": 0.8810408921933085, "train_speed(iter/s)": 1.459282 }, { "epoch": 4.083586821472944, "grad_norm": 5.084054470062256, "learning_rate": 8.062134866385717e-06, "loss": 0.40526657104492186, "memory(GiB)": 74.99, "step": 95315, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.459288 }, { "epoch": 4.0838010368021935, "grad_norm": 3.7691683769226074, "learning_rate": 8.05847084815437e-06, "loss": 0.5015369415283203, "memory(GiB)": 74.99, "step": 95320, "token_acc": 0.8817891373801917, "train_speed(iter/s)": 1.459291 }, { "epoch": 4.084015252131443, "grad_norm": 3.890848159790039, "learning_rate": 8.054807589733166e-06, "loss": 0.4461824417114258, "memory(GiB)": 74.99, "step": 95325, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.459289 }, { "epoch": 4.084229467460691, "grad_norm": 3.3179831504821777, "learning_rate": 8.051145091188472e-06, "loss": 0.29629228115081785, "memory(GiB)": 74.99, "step": 95330, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.459287 }, { "epoch": 4.08444368278994, "grad_norm": 4.372819900512695, "learning_rate": 8.047483352586638e-06, "loss": 0.7560223579406739, "memory(GiB)": 74.99, "step": 95335, "token_acc": 0.8571428571428571, "train_speed(iter/s)": 1.459296 }, { "epoch": 4.08465789811919, "grad_norm": 1.5899078845977783, "learning_rate": 8.043822373993997e-06, "loss": 0.3388272047042847, "memory(GiB)": 74.99, "step": 95340, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.4593 }, { "epoch": 4.084872113448438, "grad_norm": 6.556164264678955, "learning_rate": 8.04016215547686e-06, "loss": 0.7282149314880371, "memory(GiB)": 74.99, "step": 95345, "token_acc": 0.8681672025723473, "train_speed(iter/s)": 1.459311 }, { "epoch": 4.085086328777687, "grad_norm": 0.15498441457748413, "learning_rate": 8.036502697101555e-06, "loss": 0.12994301319122314, "memory(GiB)": 74.99, "step": 95350, "token_acc": 0.9771863117870723, "train_speed(iter/s)": 1.459318 }, { "epoch": 4.0853005441069365, "grad_norm": 2.9848859310150146, "learning_rate": 8.032843998934369e-06, "loss": 0.2078075647354126, "memory(GiB)": 74.99, "step": 95355, "token_acc": 0.9375, "train_speed(iter/s)": 1.459321 }, { "epoch": 4.085514759436185, "grad_norm": 3.9357402324676514, "learning_rate": 8.029186061041588e-06, "loss": 0.6222160339355469, "memory(GiB)": 74.99, "step": 95360, "token_acc": 0.8627450980392157, "train_speed(iter/s)": 1.459317 }, { "epoch": 4.085728974765434, "grad_norm": 4.568761348724365, "learning_rate": 8.025528883489468e-06, "loss": 0.29935636520385744, "memory(GiB)": 74.99, "step": 95365, "token_acc": 0.9472049689440993, "train_speed(iter/s)": 1.459314 }, { "epoch": 4.085943190094683, "grad_norm": 4.837918758392334, "learning_rate": 8.02187246634426e-06, "loss": 0.3748573064804077, "memory(GiB)": 74.99, "step": 95370, "token_acc": 0.9190140845070423, "train_speed(iter/s)": 1.459322 }, { "epoch": 4.086157405423932, "grad_norm": 3.2594029903411865, "learning_rate": 8.018216809672219e-06, "loss": 0.32382545471191404, "memory(GiB)": 74.99, "step": 95375, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.459324 }, { "epoch": 4.086371620753181, "grad_norm": 3.675894021987915, "learning_rate": 8.014561913539565e-06, "loss": 0.5687693119049072, "memory(GiB)": 74.99, "step": 95380, "token_acc": 0.8875, "train_speed(iter/s)": 1.459344 }, { "epoch": 4.08658583608243, "grad_norm": 4.664277076721191, "learning_rate": 8.010907778012494e-06, "loss": 0.7576786041259765, "memory(GiB)": 74.99, "step": 95385, "token_acc": 0.8445945945945946, "train_speed(iter/s)": 1.459364 }, { "epoch": 4.086800051411679, "grad_norm": 3.180696964263916, "learning_rate": 8.007254403157233e-06, "loss": 0.17918635606765748, "memory(GiB)": 74.99, "step": 95390, "token_acc": 0.9659863945578231, "train_speed(iter/s)": 1.459366 }, { "epoch": 4.087014266740928, "grad_norm": 3.245243787765503, "learning_rate": 8.003601789039944e-06, "loss": 0.5129621982574463, "memory(GiB)": 74.99, "step": 95395, "token_acc": 0.881578947368421, "train_speed(iter/s)": 1.459372 }, { "epoch": 4.087228482070177, "grad_norm": 2.939243793487549, "learning_rate": 7.999949935726797e-06, "loss": 0.27016420364379884, "memory(GiB)": 74.99, "step": 95400, "token_acc": 0.9535603715170279, "train_speed(iter/s)": 1.459376 }, { "epoch": 4.087442697399426, "grad_norm": 3.804514169692993, "learning_rate": 7.996298843283967e-06, "loss": 0.23581264019012452, "memory(GiB)": 74.99, "step": 95405, "token_acc": 0.9401408450704225, "train_speed(iter/s)": 1.459379 }, { "epoch": 4.087656912728675, "grad_norm": 2.266146659851074, "learning_rate": 7.99264851177759e-06, "loss": 0.30835297107696535, "memory(GiB)": 74.99, "step": 95410, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.45939 }, { "epoch": 4.087871128057924, "grad_norm": 3.565189838409424, "learning_rate": 7.988998941273784e-06, "loss": 0.3021169900894165, "memory(GiB)": 74.99, "step": 95415, "token_acc": 0.9486404833836858, "train_speed(iter/s)": 1.459391 }, { "epoch": 4.088085343387172, "grad_norm": 3.6662607192993164, "learning_rate": 7.985350131838676e-06, "loss": 0.3782604694366455, "memory(GiB)": 74.99, "step": 95420, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.459399 }, { "epoch": 4.088299558716422, "grad_norm": 4.41990852355957, "learning_rate": 7.981702083538368e-06, "loss": 0.350165867805481, "memory(GiB)": 74.99, "step": 95425, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.459404 }, { "epoch": 4.088513774045671, "grad_norm": 5.50639533996582, "learning_rate": 7.978054796438922e-06, "loss": 0.5301513195037841, "memory(GiB)": 74.99, "step": 95430, "token_acc": 0.9035714285714286, "train_speed(iter/s)": 1.459421 }, { "epoch": 4.088727989374919, "grad_norm": 2.1291472911834717, "learning_rate": 7.974408270606448e-06, "loss": 0.27247982025146483, "memory(GiB)": 74.99, "step": 95435, "token_acc": 0.9472049689440993, "train_speed(iter/s)": 1.459429 }, { "epoch": 4.088942204704169, "grad_norm": 4.263675689697266, "learning_rate": 7.970762506106993e-06, "loss": 0.15077046155929566, "memory(GiB)": 74.99, "step": 95440, "token_acc": 0.967032967032967, "train_speed(iter/s)": 1.459435 }, { "epoch": 4.089156420033418, "grad_norm": 4.6705193519592285, "learning_rate": 7.967117503006604e-06, "loss": 0.4063271999359131, "memory(GiB)": 74.99, "step": 95445, "token_acc": 0.9057377049180327, "train_speed(iter/s)": 1.459449 }, { "epoch": 4.089370635362666, "grad_norm": 0.10464057326316833, "learning_rate": 7.963473261371307e-06, "loss": 0.2506770133972168, "memory(GiB)": 74.99, "step": 95450, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.459449 }, { "epoch": 4.0895848506919155, "grad_norm": 2.941500186920166, "learning_rate": 7.959829781267114e-06, "loss": 0.40933547019958494, "memory(GiB)": 74.99, "step": 95455, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.459451 }, { "epoch": 4.089799066021165, "grad_norm": 3.1093223094940186, "learning_rate": 7.956187062760045e-06, "loss": 0.39646234512329104, "memory(GiB)": 74.99, "step": 95460, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.459469 }, { "epoch": 4.090013281350413, "grad_norm": 2.7446253299713135, "learning_rate": 7.952545105916098e-06, "loss": 0.23188281059265137, "memory(GiB)": 74.99, "step": 95465, "token_acc": 0.9536423841059603, "train_speed(iter/s)": 1.459473 }, { "epoch": 4.090227496679662, "grad_norm": 1.9752618074417114, "learning_rate": 7.948903910801236e-06, "loss": 0.29972195625305176, "memory(GiB)": 74.99, "step": 95470, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.459482 }, { "epoch": 4.090441712008912, "grad_norm": 1.7174022197723389, "learning_rate": 7.945263477481434e-06, "loss": 0.3273813247680664, "memory(GiB)": 74.99, "step": 95475, "token_acc": 0.9331103678929766, "train_speed(iter/s)": 1.459486 }, { "epoch": 4.09065592733816, "grad_norm": 6.629561901092529, "learning_rate": 7.94162380602263e-06, "loss": 0.37445650100708006, "memory(GiB)": 74.99, "step": 95480, "token_acc": 0.9067164179104478, "train_speed(iter/s)": 1.459489 }, { "epoch": 4.090870142667409, "grad_norm": 2.453061819076538, "learning_rate": 7.937984896490763e-06, "loss": 0.3029945850372314, "memory(GiB)": 74.99, "step": 95485, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.459491 }, { "epoch": 4.0910843579966585, "grad_norm": 2.7395241260528564, "learning_rate": 7.934346748951748e-06, "loss": 0.3437433004379272, "memory(GiB)": 74.99, "step": 95490, "token_acc": 0.9343065693430657, "train_speed(iter/s)": 1.459493 }, { "epoch": 4.091298573325907, "grad_norm": 2.10855770111084, "learning_rate": 7.930709363471517e-06, "loss": 0.2916827440261841, "memory(GiB)": 74.99, "step": 95495, "token_acc": 0.9578947368421052, "train_speed(iter/s)": 1.459498 }, { "epoch": 4.091512788655156, "grad_norm": 1.8800917863845825, "learning_rate": 7.927072740115943e-06, "loss": 0.2850929021835327, "memory(GiB)": 77.59, "step": 95500, "token_acc": 0.9240121580547113, "train_speed(iter/s)": 1.459499 }, { "epoch": 4.091512788655156, "eval_loss": 2.489314079284668, "eval_runtime": 11.3116, "eval_samples_per_second": 8.84, "eval_steps_per_second": 8.84, "eval_token_acc": 0.4596541786743516, "step": 95500 }, { "epoch": 4.091727003984405, "grad_norm": 4.127394676208496, "learning_rate": 7.923436878950919e-06, "loss": 0.46116008758544924, "memory(GiB)": 77.59, "step": 95505, "token_acc": 0.5911379657603223, "train_speed(iter/s)": 1.45923 }, { "epoch": 4.091941219313654, "grad_norm": 4.929025650024414, "learning_rate": 7.919801780042307e-06, "loss": 0.44260377883911134, "memory(GiB)": 77.59, "step": 95510, "token_acc": 0.9134078212290503, "train_speed(iter/s)": 1.459231 }, { "epoch": 4.092155434642903, "grad_norm": 3.907477617263794, "learning_rate": 7.916167443455946e-06, "loss": 0.13401589393615723, "memory(GiB)": 77.59, "step": 95515, "token_acc": 0.9580838323353293, "train_speed(iter/s)": 1.459229 }, { "epoch": 4.092369649972152, "grad_norm": 2.9354958534240723, "learning_rate": 7.912533869257704e-06, "loss": 0.3555779457092285, "memory(GiB)": 77.59, "step": 95520, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.459231 }, { "epoch": 4.092583865301401, "grad_norm": 3.2529678344726562, "learning_rate": 7.90890105751339e-06, "loss": 0.3384925365447998, "memory(GiB)": 77.59, "step": 95525, "token_acc": 0.9331103678929766, "train_speed(iter/s)": 1.45923 }, { "epoch": 4.09279808063065, "grad_norm": 3.580012798309326, "learning_rate": 7.905269008288807e-06, "loss": 0.2582880735397339, "memory(GiB)": 77.59, "step": 95530, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.459236 }, { "epoch": 4.093012295959899, "grad_norm": 3.4759857654571533, "learning_rate": 7.901637721649774e-06, "loss": 0.507399320602417, "memory(GiB)": 77.59, "step": 95535, "token_acc": 0.8902077151335311, "train_speed(iter/s)": 1.45925 }, { "epoch": 4.0932265112891475, "grad_norm": 1.3344734907150269, "learning_rate": 7.898007197662066e-06, "loss": 0.2737112522125244, "memory(GiB)": 77.59, "step": 95540, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.459249 }, { "epoch": 4.093440726618397, "grad_norm": 3.3454933166503906, "learning_rate": 7.894377436391443e-06, "loss": 0.20959157943725587, "memory(GiB)": 77.59, "step": 95545, "token_acc": 0.9525423728813559, "train_speed(iter/s)": 1.459252 }, { "epoch": 4.093654941947646, "grad_norm": 2.4536585807800293, "learning_rate": 7.890748437903677e-06, "loss": 0.3971314668655396, "memory(GiB)": 77.59, "step": 95550, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.459268 }, { "epoch": 4.093869157276894, "grad_norm": 2.1151907444000244, "learning_rate": 7.887120202264514e-06, "loss": 0.13536022901535033, "memory(GiB)": 77.59, "step": 95555, "token_acc": 0.9633699633699634, "train_speed(iter/s)": 1.459268 }, { "epoch": 4.094083372606144, "grad_norm": 7.5615386962890625, "learning_rate": 7.883492729539665e-06, "loss": 0.7297562599182129, "memory(GiB)": 77.59, "step": 95560, "token_acc": 0.8440860215053764, "train_speed(iter/s)": 1.45927 }, { "epoch": 4.094297587935393, "grad_norm": 4.030489921569824, "learning_rate": 7.879866019794858e-06, "loss": 0.5652758121490479, "memory(GiB)": 77.59, "step": 95565, "token_acc": 0.8882175226586103, "train_speed(iter/s)": 1.459275 }, { "epoch": 4.094511803264641, "grad_norm": 5.2988691329956055, "learning_rate": 7.876240073095793e-06, "loss": 0.46048769950866697, "memory(GiB)": 77.59, "step": 95570, "token_acc": 0.89501312335958, "train_speed(iter/s)": 1.45928 }, { "epoch": 4.0947260185938905, "grad_norm": 1.74123215675354, "learning_rate": 7.872614889508134e-06, "loss": 0.2931062698364258, "memory(GiB)": 77.59, "step": 95575, "token_acc": 0.9518900343642611, "train_speed(iter/s)": 1.459279 }, { "epoch": 4.09494023392314, "grad_norm": 3.3610455989837646, "learning_rate": 7.868990469097593e-06, "loss": 0.17586956024169922, "memory(GiB)": 77.59, "step": 95580, "token_acc": 0.9702602230483272, "train_speed(iter/s)": 1.45928 }, { "epoch": 4.095154449252388, "grad_norm": 2.5299994945526123, "learning_rate": 7.86536681192981e-06, "loss": 0.16356064081192018, "memory(GiB)": 77.59, "step": 95585, "token_acc": 0.9664179104477612, "train_speed(iter/s)": 1.459284 }, { "epoch": 4.095368664581637, "grad_norm": 8.426236152648926, "learning_rate": 7.861743918070435e-06, "loss": 0.28490843772888186, "memory(GiB)": 77.59, "step": 95590, "token_acc": 0.9382716049382716, "train_speed(iter/s)": 1.459286 }, { "epoch": 4.095582879910887, "grad_norm": 4.005906105041504, "learning_rate": 7.858121787585093e-06, "loss": 0.28820004463195803, "memory(GiB)": 77.59, "step": 95595, "token_acc": 0.9391891891891891, "train_speed(iter/s)": 1.459303 }, { "epoch": 4.095797095240135, "grad_norm": 2.947199821472168, "learning_rate": 7.854500420539401e-06, "loss": 0.39296650886535645, "memory(GiB)": 77.59, "step": 95600, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.459308 }, { "epoch": 4.096011310569384, "grad_norm": 5.1750078201293945, "learning_rate": 7.850879816998962e-06, "loss": 0.48421173095703124, "memory(GiB)": 77.59, "step": 95605, "token_acc": 0.8880597014925373, "train_speed(iter/s)": 1.459314 }, { "epoch": 4.0962255258986335, "grad_norm": 3.616422414779663, "learning_rate": 7.847259977029392e-06, "loss": 0.39400854110717776, "memory(GiB)": 77.59, "step": 95610, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.459308 }, { "epoch": 4.096439741227882, "grad_norm": 5.302975177764893, "learning_rate": 7.843640900696247e-06, "loss": 0.6641236782073975, "memory(GiB)": 77.59, "step": 95615, "token_acc": 0.8677966101694915, "train_speed(iter/s)": 1.459315 }, { "epoch": 4.096653956557131, "grad_norm": 3.728583812713623, "learning_rate": 7.840022588065098e-06, "loss": 0.33673644065856934, "memory(GiB)": 77.59, "step": 95620, "token_acc": 0.9108527131782945, "train_speed(iter/s)": 1.459318 }, { "epoch": 4.09686817188638, "grad_norm": 1.5449786186218262, "learning_rate": 7.836405039201483e-06, "loss": 0.6151815891265869, "memory(GiB)": 77.59, "step": 95625, "token_acc": 0.904, "train_speed(iter/s)": 1.45932 }, { "epoch": 4.097082387215629, "grad_norm": 3.3594040870666504, "learning_rate": 7.832788254170948e-06, "loss": 0.2632344961166382, "memory(GiB)": 77.59, "step": 95630, "token_acc": 0.9548192771084337, "train_speed(iter/s)": 1.459324 }, { "epoch": 4.097296602544878, "grad_norm": 2.2127857208251953, "learning_rate": 7.829172233038996e-06, "loss": 0.145699143409729, "memory(GiB)": 77.59, "step": 95635, "token_acc": 0.9660493827160493, "train_speed(iter/s)": 1.459328 }, { "epoch": 4.097510817874127, "grad_norm": 4.861756324768066, "learning_rate": 7.825556975871156e-06, "loss": 0.22035348415374756, "memory(GiB)": 77.59, "step": 95640, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.45933 }, { "epoch": 4.097725033203376, "grad_norm": 2.414721727371216, "learning_rate": 7.821942482732918e-06, "loss": 0.33896560668945314, "memory(GiB)": 77.59, "step": 95645, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.459333 }, { "epoch": 4.097939248532625, "grad_norm": 1.8543485403060913, "learning_rate": 7.818328753689763e-06, "loss": 0.2844693183898926, "memory(GiB)": 77.59, "step": 95650, "token_acc": 0.9370629370629371, "train_speed(iter/s)": 1.459332 }, { "epoch": 4.098153463861874, "grad_norm": 2.3303892612457275, "learning_rate": 7.814715788807148e-06, "loss": 0.34804956912994384, "memory(GiB)": 77.59, "step": 95655, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.459333 }, { "epoch": 4.0983676791911225, "grad_norm": 1.2789169549942017, "learning_rate": 7.811103588150514e-06, "loss": 0.3702098846435547, "memory(GiB)": 77.59, "step": 95660, "token_acc": 0.9070631970260223, "train_speed(iter/s)": 1.459333 }, { "epoch": 4.098581894520372, "grad_norm": 5.300588130950928, "learning_rate": 7.807492151785334e-06, "loss": 0.39889798164367674, "memory(GiB)": 77.59, "step": 95665, "token_acc": 0.9201277955271565, "train_speed(iter/s)": 1.459334 }, { "epoch": 4.098796109849621, "grad_norm": 3.5880093574523926, "learning_rate": 7.803881479777008e-06, "loss": 0.310950493812561, "memory(GiB)": 77.59, "step": 95670, "token_acc": 0.9207920792079208, "train_speed(iter/s)": 1.459336 }, { "epoch": 4.099010325178869, "grad_norm": 3.2640459537506104, "learning_rate": 7.800271572190954e-06, "loss": 0.4254180908203125, "memory(GiB)": 77.59, "step": 95675, "token_acc": 0.9078014184397163, "train_speed(iter/s)": 1.459339 }, { "epoch": 4.099224540508119, "grad_norm": 3.5911436080932617, "learning_rate": 7.796662429092555e-06, "loss": 0.30486040115356444, "memory(GiB)": 77.59, "step": 95680, "token_acc": 0.926984126984127, "train_speed(iter/s)": 1.459346 }, { "epoch": 4.099438755837368, "grad_norm": 4.467776298522949, "learning_rate": 7.793054050547215e-06, "loss": 0.2671544551849365, "memory(GiB)": 77.59, "step": 95685, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.459357 }, { "epoch": 4.099652971166616, "grad_norm": 3.164794683456421, "learning_rate": 7.78944643662029e-06, "loss": 0.14559892416000367, "memory(GiB)": 77.59, "step": 95690, "token_acc": 0.9716312056737588, "train_speed(iter/s)": 1.459365 }, { "epoch": 4.099867186495866, "grad_norm": 6.384521961212158, "learning_rate": 7.785839587377141e-06, "loss": 0.5343860149383545, "memory(GiB)": 77.59, "step": 95695, "token_acc": 0.8904899135446686, "train_speed(iter/s)": 1.459368 }, { "epoch": 4.100081401825115, "grad_norm": 1.9732855558395386, "learning_rate": 7.782233502883119e-06, "loss": 0.39347655773162843, "memory(GiB)": 77.59, "step": 95700, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.459382 }, { "epoch": 4.100295617154363, "grad_norm": 4.600942134857178, "learning_rate": 7.778628183203535e-06, "loss": 0.49395103454589845, "memory(GiB)": 77.59, "step": 95705, "token_acc": 0.8961038961038961, "train_speed(iter/s)": 1.459386 }, { "epoch": 4.1005098324836124, "grad_norm": 0.9322373867034912, "learning_rate": 7.775023628403705e-06, "loss": 0.09937276244163513, "memory(GiB)": 77.59, "step": 95710, "token_acc": 0.9776119402985075, "train_speed(iter/s)": 1.459385 }, { "epoch": 4.100724047812862, "grad_norm": 2.3317487239837646, "learning_rate": 7.771419838548938e-06, "loss": 0.3388197422027588, "memory(GiB)": 77.59, "step": 95715, "token_acc": 0.9156626506024096, "train_speed(iter/s)": 1.459397 }, { "epoch": 4.10093826314211, "grad_norm": 2.767971992492676, "learning_rate": 7.767816813704499e-06, "loss": 0.40311470031738283, "memory(GiB)": 77.59, "step": 95720, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.459405 }, { "epoch": 4.101152478471359, "grad_norm": 3.533876419067383, "learning_rate": 7.76421455393569e-06, "loss": 0.2966663122177124, "memory(GiB)": 77.59, "step": 95725, "token_acc": 0.9491525423728814, "train_speed(iter/s)": 1.459409 }, { "epoch": 4.101366693800609, "grad_norm": 0.337054580450058, "learning_rate": 7.760613059307748e-06, "loss": 0.2164156198501587, "memory(GiB)": 77.59, "step": 95730, "token_acc": 0.9410029498525073, "train_speed(iter/s)": 1.459412 }, { "epoch": 4.101580909129857, "grad_norm": 2.6222338676452637, "learning_rate": 7.757012329885933e-06, "loss": 0.2603895902633667, "memory(GiB)": 77.59, "step": 95735, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.459433 }, { "epoch": 4.101795124459106, "grad_norm": 2.985623598098755, "learning_rate": 7.753412365735463e-06, "loss": 0.3509040117263794, "memory(GiB)": 77.59, "step": 95740, "token_acc": 0.9223880597014925, "train_speed(iter/s)": 1.459438 }, { "epoch": 4.1020093397883555, "grad_norm": 3.3432202339172363, "learning_rate": 7.749813166921543e-06, "loss": 0.24336590766906738, "memory(GiB)": 77.59, "step": 95745, "token_acc": 0.9348659003831418, "train_speed(iter/s)": 1.459443 }, { "epoch": 4.102223555117604, "grad_norm": 3.8133928775787354, "learning_rate": 7.746214733509411e-06, "loss": 0.5139685153961182, "memory(GiB)": 77.59, "step": 95750, "token_acc": 0.9060402684563759, "train_speed(iter/s)": 1.459441 }, { "epoch": 4.102437770446853, "grad_norm": 1.3730350732803345, "learning_rate": 7.742617065564217e-06, "loss": 0.2520288944244385, "memory(GiB)": 77.59, "step": 95755, "token_acc": 0.946843853820598, "train_speed(iter/s)": 1.459445 }, { "epoch": 4.102651985776102, "grad_norm": 4.466381072998047, "learning_rate": 7.739020163151173e-06, "loss": 0.3802527904510498, "memory(GiB)": 77.59, "step": 95760, "token_acc": 0.9053030303030303, "train_speed(iter/s)": 1.459445 }, { "epoch": 4.102866201105351, "grad_norm": 2.1476211547851562, "learning_rate": 7.73542402633542e-06, "loss": 0.5196426391601563, "memory(GiB)": 77.59, "step": 95765, "token_acc": 0.904, "train_speed(iter/s)": 1.459448 }, { "epoch": 4.1030804164346, "grad_norm": 5.6233296394348145, "learning_rate": 7.731828655182105e-06, "loss": 0.12532123327255248, "memory(GiB)": 77.59, "step": 95770, "token_acc": 0.9701492537313433, "train_speed(iter/s)": 1.459453 }, { "epoch": 4.103294631763849, "grad_norm": 8.761807441711426, "learning_rate": 7.728234049756372e-06, "loss": 0.31251742839813235, "memory(GiB)": 77.59, "step": 95775, "token_acc": 0.9375, "train_speed(iter/s)": 1.459457 }, { "epoch": 4.103508847093098, "grad_norm": 0.41424494981765747, "learning_rate": 7.724640210123312e-06, "loss": 0.5262028217315674, "memory(GiB)": 77.59, "step": 95780, "token_acc": 0.8759398496240601, "train_speed(iter/s)": 1.459466 }, { "epoch": 4.103723062422347, "grad_norm": 2.9872801303863525, "learning_rate": 7.721047136348076e-06, "loss": 0.6202744007110595, "memory(GiB)": 77.59, "step": 95785, "token_acc": 0.8741007194244604, "train_speed(iter/s)": 1.459479 }, { "epoch": 4.103937277751596, "grad_norm": 4.00942850112915, "learning_rate": 7.717454828495724e-06, "loss": 0.5279306888580322, "memory(GiB)": 77.59, "step": 95790, "token_acc": 0.8828828828828829, "train_speed(iter/s)": 1.459481 }, { "epoch": 4.1041514930808445, "grad_norm": 2.673140525817871, "learning_rate": 7.71386328663134e-06, "loss": 0.4416008472442627, "memory(GiB)": 77.59, "step": 95795, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.459495 }, { "epoch": 4.104365708410094, "grad_norm": 3.1606342792510986, "learning_rate": 7.710272510819993e-06, "loss": 0.3506616592407227, "memory(GiB)": 77.59, "step": 95800, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.459496 }, { "epoch": 4.104579923739343, "grad_norm": 2.7584280967712402, "learning_rate": 7.706682501126722e-06, "loss": 0.3518896341323853, "memory(GiB)": 77.59, "step": 95805, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.459509 }, { "epoch": 4.104794139068591, "grad_norm": 5.9847636222839355, "learning_rate": 7.703093257616579e-06, "loss": 0.35430316925048827, "memory(GiB)": 77.59, "step": 95810, "token_acc": 0.9153094462540716, "train_speed(iter/s)": 1.459512 }, { "epoch": 4.105008354397841, "grad_norm": 3.0027198791503906, "learning_rate": 7.699504780354582e-06, "loss": 0.4687209129333496, "memory(GiB)": 77.59, "step": 95815, "token_acc": 0.9253246753246753, "train_speed(iter/s)": 1.459526 }, { "epoch": 4.10522256972709, "grad_norm": 2.109652042388916, "learning_rate": 7.69591706940574e-06, "loss": 0.38176865577697755, "memory(GiB)": 77.59, "step": 95820, "token_acc": 0.9363295880149812, "train_speed(iter/s)": 1.459538 }, { "epoch": 4.105436785056338, "grad_norm": 1.309617519378662, "learning_rate": 7.69233012483504e-06, "loss": 0.2930804967880249, "memory(GiB)": 77.59, "step": 95825, "token_acc": 0.9107142857142857, "train_speed(iter/s)": 1.459541 }, { "epoch": 4.1056510003855875, "grad_norm": 6.602456569671631, "learning_rate": 7.688743946707455e-06, "loss": 0.34668750762939454, "memory(GiB)": 77.59, "step": 95830, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.459542 }, { "epoch": 4.105865215714837, "grad_norm": 0.563772439956665, "learning_rate": 7.685158535087966e-06, "loss": 0.3276612997055054, "memory(GiB)": 77.59, "step": 95835, "token_acc": 0.9466192170818505, "train_speed(iter/s)": 1.459547 }, { "epoch": 4.106079431044085, "grad_norm": 5.045650959014893, "learning_rate": 7.681573890041539e-06, "loss": 0.49578680992126467, "memory(GiB)": 77.59, "step": 95840, "token_acc": 0.9, "train_speed(iter/s)": 1.459546 }, { "epoch": 4.106293646373334, "grad_norm": 2.157841205596924, "learning_rate": 7.677990011633096e-06, "loss": 0.19752026796340943, "memory(GiB)": 77.59, "step": 95845, "token_acc": 0.9554794520547946, "train_speed(iter/s)": 1.459547 }, { "epoch": 4.106507861702584, "grad_norm": 3.79158878326416, "learning_rate": 7.674406899927567e-06, "loss": 0.5302028179168701, "memory(GiB)": 77.59, "step": 95850, "token_acc": 0.8838028169014085, "train_speed(iter/s)": 1.459552 }, { "epoch": 4.106722077031832, "grad_norm": 5.438638687133789, "learning_rate": 7.670824554989858e-06, "loss": 0.35875544548034666, "memory(GiB)": 77.59, "step": 95855, "token_acc": 0.9217081850533808, "train_speed(iter/s)": 1.459565 }, { "epoch": 4.106936292361081, "grad_norm": 3.6664726734161377, "learning_rate": 7.667242976884875e-06, "loss": 0.3466258764266968, "memory(GiB)": 77.59, "step": 95860, "token_acc": 0.9272151898734177, "train_speed(iter/s)": 1.459568 }, { "epoch": 4.1071505076903305, "grad_norm": 0.3406655788421631, "learning_rate": 7.663662165677481e-06, "loss": 0.08716933727264405, "memory(GiB)": 77.59, "step": 95865, "token_acc": 0.9651567944250871, "train_speed(iter/s)": 1.45958 }, { "epoch": 4.107364723019579, "grad_norm": 1.6851065158843994, "learning_rate": 7.660082121432577e-06, "loss": 0.163592529296875, "memory(GiB)": 77.59, "step": 95870, "token_acc": 0.97265625, "train_speed(iter/s)": 1.459582 }, { "epoch": 4.107578938348828, "grad_norm": 2.255568504333496, "learning_rate": 7.656502844214996e-06, "loss": 0.2603933572769165, "memory(GiB)": 77.59, "step": 95875, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.459595 }, { "epoch": 4.107793153678077, "grad_norm": 5.404271602630615, "learning_rate": 7.652924334089594e-06, "loss": 0.28052358627319335, "memory(GiB)": 77.59, "step": 95880, "token_acc": 0.9372822299651568, "train_speed(iter/s)": 1.459605 }, { "epoch": 4.108007369007326, "grad_norm": 2.949974775314331, "learning_rate": 7.649346591121193e-06, "loss": 0.4396796703338623, "memory(GiB)": 77.59, "step": 95885, "token_acc": 0.9051094890510949, "train_speed(iter/s)": 1.459608 }, { "epoch": 4.108221584336575, "grad_norm": 1.76039457321167, "learning_rate": 7.645769615374592e-06, "loss": 0.28672792911529543, "memory(GiB)": 77.59, "step": 95890, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.45961 }, { "epoch": 4.108435799665824, "grad_norm": 4.2074103355407715, "learning_rate": 7.642193406914621e-06, "loss": 0.22196571826934813, "memory(GiB)": 77.59, "step": 95895, "token_acc": 0.953307392996109, "train_speed(iter/s)": 1.459611 }, { "epoch": 4.108650014995073, "grad_norm": 1.6357988119125366, "learning_rate": 7.638617965806038e-06, "loss": 0.22944726943969726, "memory(GiB)": 77.59, "step": 95900, "token_acc": 0.9522058823529411, "train_speed(iter/s)": 1.459615 }, { "epoch": 4.108864230324322, "grad_norm": 5.14609956741333, "learning_rate": 7.635043292113637e-06, "loss": 0.4139768123626709, "memory(GiB)": 77.59, "step": 95905, "token_acc": 0.9120879120879121, "train_speed(iter/s)": 1.45962 }, { "epoch": 4.109078445653571, "grad_norm": 2.485102415084839, "learning_rate": 7.631469385902173e-06, "loss": 0.42506103515625, "memory(GiB)": 77.59, "step": 95910, "token_acc": 0.9143730886850153, "train_speed(iter/s)": 1.459625 }, { "epoch": 4.1092926609828195, "grad_norm": 2.463040828704834, "learning_rate": 7.627896247236382e-06, "loss": 0.0961795449256897, "memory(GiB)": 77.59, "step": 95915, "token_acc": 0.9829059829059829, "train_speed(iter/s)": 1.459626 }, { "epoch": 4.109506876312069, "grad_norm": 3.8547871112823486, "learning_rate": 7.624323876180989e-06, "loss": 0.22413954734802247, "memory(GiB)": 77.59, "step": 95920, "token_acc": 0.9450867052023122, "train_speed(iter/s)": 1.459631 }, { "epoch": 4.109721091641318, "grad_norm": 3.2810988426208496, "learning_rate": 7.6207522728007354e-06, "loss": 0.359229588508606, "memory(GiB)": 77.59, "step": 95925, "token_acc": 0.8963210702341137, "train_speed(iter/s)": 1.459631 }, { "epoch": 4.109935306970566, "grad_norm": 3.3233697414398193, "learning_rate": 7.6171814371603035e-06, "loss": 0.3049289703369141, "memory(GiB)": 77.59, "step": 95930, "token_acc": 0.9182389937106918, "train_speed(iter/s)": 1.459633 }, { "epoch": 4.110149522299816, "grad_norm": 6.171472072601318, "learning_rate": 7.613611369324392e-06, "loss": 0.5367675304412842, "memory(GiB)": 77.59, "step": 95935, "token_acc": 0.876, "train_speed(iter/s)": 1.459638 }, { "epoch": 4.110363737629065, "grad_norm": 4.257186412811279, "learning_rate": 7.61004206935767e-06, "loss": 0.4161534786224365, "memory(GiB)": 77.59, "step": 95940, "token_acc": 0.9073482428115016, "train_speed(iter/s)": 1.45964 }, { "epoch": 4.110577952958313, "grad_norm": 2.749199151992798, "learning_rate": 7.606473537324804e-06, "loss": 0.2886219024658203, "memory(GiB)": 77.59, "step": 95945, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.459645 }, { "epoch": 4.1107921682875626, "grad_norm": 5.870785713195801, "learning_rate": 7.602905773290425e-06, "loss": 0.31442010402679443, "memory(GiB)": 77.59, "step": 95950, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.459644 }, { "epoch": 4.111006383616812, "grad_norm": 3.1145882606506348, "learning_rate": 7.599338777319192e-06, "loss": 0.28844165802001953, "memory(GiB)": 77.59, "step": 95955, "token_acc": 0.9304635761589404, "train_speed(iter/s)": 1.459649 }, { "epoch": 4.11122059894606, "grad_norm": 1.9343202114105225, "learning_rate": 7.59577254947571e-06, "loss": 0.19529086351394653, "memory(GiB)": 77.59, "step": 95960, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.459648 }, { "epoch": 4.111434814275309, "grad_norm": 1.8779592514038086, "learning_rate": 7.592207089824588e-06, "loss": 0.521448802947998, "memory(GiB)": 77.59, "step": 95965, "token_acc": 0.8901515151515151, "train_speed(iter/s)": 1.459656 }, { "epoch": 4.111649029604559, "grad_norm": 3.779987335205078, "learning_rate": 7.5886423984304125e-06, "loss": 0.4492375373840332, "memory(GiB)": 77.59, "step": 95970, "token_acc": 0.9132947976878613, "train_speed(iter/s)": 1.459664 }, { "epoch": 4.111863244933807, "grad_norm": 3.402118682861328, "learning_rate": 7.58507847535776e-06, "loss": 0.3751258850097656, "memory(GiB)": 77.59, "step": 95975, "token_acc": 0.9347181008902077, "train_speed(iter/s)": 1.45967 }, { "epoch": 4.112077460263056, "grad_norm": 2.6913416385650635, "learning_rate": 7.5815153206711955e-06, "loss": 0.33190891742706297, "memory(GiB)": 77.59, "step": 95980, "token_acc": 0.9179331306990881, "train_speed(iter/s)": 1.45967 }, { "epoch": 4.112291675592306, "grad_norm": 6.933797359466553, "learning_rate": 7.577952934435284e-06, "loss": 0.3557994842529297, "memory(GiB)": 77.59, "step": 95985, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.459675 }, { "epoch": 4.112505890921554, "grad_norm": 3.8386645317077637, "learning_rate": 7.574391316714552e-06, "loss": 0.27078871726989745, "memory(GiB)": 77.59, "step": 95990, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.459681 }, { "epoch": 4.112720106250803, "grad_norm": 1.6881234645843506, "learning_rate": 7.57083046757352e-06, "loss": 0.3227804660797119, "memory(GiB)": 77.59, "step": 95995, "token_acc": 0.9462025316455697, "train_speed(iter/s)": 1.459682 }, { "epoch": 4.1129343215800525, "grad_norm": 0.6807956099510193, "learning_rate": 7.567270387076692e-06, "loss": 0.28503572940826416, "memory(GiB)": 77.59, "step": 96000, "token_acc": 0.9518072289156626, "train_speed(iter/s)": 1.459692 }, { "epoch": 4.1129343215800525, "eval_loss": 2.2138919830322266, "eval_runtime": 11.6636, "eval_samples_per_second": 8.574, "eval_steps_per_second": 8.574, "eval_token_acc": 0.4821648216482165, "step": 96000 }, { "epoch": 4.113148536909301, "grad_norm": 2.0897743701934814, "learning_rate": 7.563711075288571e-06, "loss": 0.21464641094207765, "memory(GiB)": 77.59, "step": 96005, "token_acc": 0.6128456735057984, "train_speed(iter/s)": 1.459414 }, { "epoch": 4.11336275223855, "grad_norm": 1.4704391956329346, "learning_rate": 7.560152532273618e-06, "loss": 0.397432279586792, "memory(GiB)": 77.59, "step": 96010, "token_acc": 0.9012345679012346, "train_speed(iter/s)": 1.459428 }, { "epoch": 4.113576967567799, "grad_norm": 4.191709518432617, "learning_rate": 7.556594758096325e-06, "loss": 0.7234666824340821, "memory(GiB)": 77.59, "step": 96015, "token_acc": 0.8424908424908425, "train_speed(iter/s)": 1.459437 }, { "epoch": 4.113791182897048, "grad_norm": 7.047314167022705, "learning_rate": 7.553037752821135e-06, "loss": 0.5141180992126465, "memory(GiB)": 77.59, "step": 96020, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.459441 }, { "epoch": 4.114005398226297, "grad_norm": 2.298607587814331, "learning_rate": 7.549481516512485e-06, "loss": 0.22899222373962402, "memory(GiB)": 77.59, "step": 96025, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.459456 }, { "epoch": 4.114219613555546, "grad_norm": 0.33425456285476685, "learning_rate": 7.5459260492347975e-06, "loss": 0.24473872184753417, "memory(GiB)": 77.59, "step": 96030, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.45946 }, { "epoch": 4.114433828884795, "grad_norm": 3.45466685295105, "learning_rate": 7.5423713510524745e-06, "loss": 0.5416419029235839, "memory(GiB)": 77.59, "step": 96035, "token_acc": 0.8853503184713376, "train_speed(iter/s)": 1.45946 }, { "epoch": 4.114648044214044, "grad_norm": 2.943100690841675, "learning_rate": 7.538817422029931e-06, "loss": 0.3913369655609131, "memory(GiB)": 77.59, "step": 96040, "token_acc": 0.903114186851211, "train_speed(iter/s)": 1.459462 }, { "epoch": 4.114862259543293, "grad_norm": 3.685288667678833, "learning_rate": 7.535264262231545e-06, "loss": 0.37957148551940917, "memory(GiB)": 77.59, "step": 96045, "token_acc": 0.8981818181818182, "train_speed(iter/s)": 1.459474 }, { "epoch": 4.1150764748725415, "grad_norm": 2.9879441261291504, "learning_rate": 7.531711871721669e-06, "loss": 0.26218347549438475, "memory(GiB)": 77.59, "step": 96050, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.459474 }, { "epoch": 4.115290690201791, "grad_norm": 4.533926963806152, "learning_rate": 7.528160250564681e-06, "loss": 0.18371137380599975, "memory(GiB)": 77.59, "step": 96055, "token_acc": 0.9490445859872612, "train_speed(iter/s)": 1.459474 }, { "epoch": 4.11550490553104, "grad_norm": 4.094937324523926, "learning_rate": 7.524609398824916e-06, "loss": 0.5571147918701171, "memory(GiB)": 77.59, "step": 96060, "token_acc": 0.8945686900958466, "train_speed(iter/s)": 1.459483 }, { "epoch": 4.115719120860288, "grad_norm": 2.1566760540008545, "learning_rate": 7.521059316566681e-06, "loss": 0.15958573818206787, "memory(GiB)": 77.59, "step": 96065, "token_acc": 0.9581881533101045, "train_speed(iter/s)": 1.45948 }, { "epoch": 4.115933336189538, "grad_norm": 6.740076541900635, "learning_rate": 7.517510003854322e-06, "loss": 0.5067630767822265, "memory(GiB)": 77.59, "step": 96070, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.459493 }, { "epoch": 4.116147551518787, "grad_norm": 1.824054479598999, "learning_rate": 7.513961460752117e-06, "loss": 0.24074060916900636, "memory(GiB)": 77.59, "step": 96075, "token_acc": 0.9288537549407114, "train_speed(iter/s)": 1.459496 }, { "epoch": 4.116361766848035, "grad_norm": 4.2435383796691895, "learning_rate": 7.510413687324358e-06, "loss": 0.36558306217193604, "memory(GiB)": 77.59, "step": 96080, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.459493 }, { "epoch": 4.1165759821772845, "grad_norm": 4.906698703765869, "learning_rate": 7.506866683635311e-06, "loss": 0.4813060760498047, "memory(GiB)": 77.59, "step": 96085, "token_acc": 0.8664259927797834, "train_speed(iter/s)": 1.459493 }, { "epoch": 4.116790197506534, "grad_norm": 2.718520164489746, "learning_rate": 7.503320449749235e-06, "loss": 0.294072699546814, "memory(GiB)": 77.59, "step": 96090, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.459497 }, { "epoch": 4.117004412835782, "grad_norm": 4.372956275939941, "learning_rate": 7.499774985730363e-06, "loss": 0.29252893924713136, "memory(GiB)": 77.59, "step": 96095, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.4595 }, { "epoch": 4.117218628165031, "grad_norm": 2.889756679534912, "learning_rate": 7.496230291642947e-06, "loss": 0.5981541156768799, "memory(GiB)": 77.59, "step": 96100, "token_acc": 0.8776758409785933, "train_speed(iter/s)": 1.459498 }, { "epoch": 4.117432843494281, "grad_norm": 2.3575849533081055, "learning_rate": 7.492686367551194e-06, "loss": 0.2508898973464966, "memory(GiB)": 77.59, "step": 96105, "token_acc": 0.9455128205128205, "train_speed(iter/s)": 1.459504 }, { "epoch": 4.117647058823529, "grad_norm": 0.2702142596244812, "learning_rate": 7.489143213519301e-06, "loss": 0.40923585891723635, "memory(GiB)": 77.59, "step": 96110, "token_acc": 0.9263565891472868, "train_speed(iter/s)": 1.459508 }, { "epoch": 4.117861274152778, "grad_norm": 1.8716295957565308, "learning_rate": 7.485600829611456e-06, "loss": 0.36980471611022947, "memory(GiB)": 77.59, "step": 96115, "token_acc": 0.9134948096885813, "train_speed(iter/s)": 1.459511 }, { "epoch": 4.1180754894820275, "grad_norm": 2.869042158126831, "learning_rate": 7.482059215891823e-06, "loss": 0.3460922002792358, "memory(GiB)": 77.59, "step": 96120, "token_acc": 0.9288389513108615, "train_speed(iter/s)": 1.459513 }, { "epoch": 4.118289704811276, "grad_norm": 0.6275534629821777, "learning_rate": 7.478518372424576e-06, "loss": 0.3245592832565308, "memory(GiB)": 77.59, "step": 96125, "token_acc": 0.917910447761194, "train_speed(iter/s)": 1.459514 }, { "epoch": 4.118503920140525, "grad_norm": 2.1310527324676514, "learning_rate": 7.474978299273861e-06, "loss": 0.35756916999816896, "memory(GiB)": 77.59, "step": 96130, "token_acc": 0.9015384615384615, "train_speed(iter/s)": 1.459521 }, { "epoch": 4.118718135469774, "grad_norm": 2.219191789627075, "learning_rate": 7.47143899650381e-06, "loss": 0.25843513011932373, "memory(GiB)": 77.59, "step": 96135, "token_acc": 0.9513888888888888, "train_speed(iter/s)": 1.459519 }, { "epoch": 4.118932350799023, "grad_norm": 2.3265321254730225, "learning_rate": 7.467900464178534e-06, "loss": 0.32662386894226075, "memory(GiB)": 77.59, "step": 96140, "token_acc": 0.937984496124031, "train_speed(iter/s)": 1.459519 }, { "epoch": 4.119146566128272, "grad_norm": 3.47356915473938, "learning_rate": 7.4643627023621435e-06, "loss": 0.2193305730819702, "memory(GiB)": 77.59, "step": 96145, "token_acc": 0.9448818897637795, "train_speed(iter/s)": 1.45952 }, { "epoch": 4.119360781457521, "grad_norm": 2.042346954345703, "learning_rate": 7.460825711118724e-06, "loss": 0.2316370964050293, "memory(GiB)": 77.59, "step": 96150, "token_acc": 0.9371069182389937, "train_speed(iter/s)": 1.459519 }, { "epoch": 4.11957499678677, "grad_norm": 3.816188097000122, "learning_rate": 7.457289490512337e-06, "loss": 0.46909432411193847, "memory(GiB)": 77.59, "step": 96155, "token_acc": 0.9178470254957507, "train_speed(iter/s)": 1.459518 }, { "epoch": 4.119789212116019, "grad_norm": 3.507511854171753, "learning_rate": 7.453754040607075e-06, "loss": 0.41955065727233887, "memory(GiB)": 77.59, "step": 96160, "token_acc": 0.9196787148594378, "train_speed(iter/s)": 1.459519 }, { "epoch": 4.120003427445268, "grad_norm": 5.561809062957764, "learning_rate": 7.450219361466965e-06, "loss": 0.3066245555877686, "memory(GiB)": 77.59, "step": 96165, "token_acc": 0.9281045751633987, "train_speed(iter/s)": 1.459523 }, { "epoch": 4.1202176427745165, "grad_norm": 1.873422384262085, "learning_rate": 7.44668545315605e-06, "loss": 0.3027423143386841, "memory(GiB)": 77.59, "step": 96170, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.459524 }, { "epoch": 4.120431858103766, "grad_norm": 2.903149127960205, "learning_rate": 7.443152315738344e-06, "loss": 0.5142850875854492, "memory(GiB)": 77.59, "step": 96175, "token_acc": 0.9075342465753424, "train_speed(iter/s)": 1.459525 }, { "epoch": 4.120646073433015, "grad_norm": 2.202873706817627, "learning_rate": 7.439619949277848e-06, "loss": 0.21478683948516847, "memory(GiB)": 77.59, "step": 96180, "token_acc": 0.9554655870445344, "train_speed(iter/s)": 1.459527 }, { "epoch": 4.120860288762263, "grad_norm": 3.0754528045654297, "learning_rate": 7.436088353838566e-06, "loss": 0.5497416973114013, "memory(GiB)": 77.59, "step": 96185, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.459527 }, { "epoch": 4.121074504091513, "grad_norm": 3.742068290710449, "learning_rate": 7.432557529484479e-06, "loss": 0.4993183135986328, "memory(GiB)": 77.59, "step": 96190, "token_acc": 0.9058171745152355, "train_speed(iter/s)": 1.459536 }, { "epoch": 4.121288719420762, "grad_norm": 3.714625597000122, "learning_rate": 7.4290274762795265e-06, "loss": 0.31265738010406496, "memory(GiB)": 77.59, "step": 96195, "token_acc": 0.9049586776859504, "train_speed(iter/s)": 1.459538 }, { "epoch": 4.12150293475001, "grad_norm": 0.719196081161499, "learning_rate": 7.425498194287689e-06, "loss": 0.4114699363708496, "memory(GiB)": 77.59, "step": 96200, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.459561 }, { "epoch": 4.1217171500792595, "grad_norm": 5.391012191772461, "learning_rate": 7.421969683572894e-06, "loss": 0.3543818235397339, "memory(GiB)": 77.59, "step": 96205, "token_acc": 0.912, "train_speed(iter/s)": 1.459565 }, { "epoch": 4.121931365408509, "grad_norm": 2.8996658325195312, "learning_rate": 7.418441944199045e-06, "loss": 0.33328382968902587, "memory(GiB)": 77.59, "step": 96210, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.459568 }, { "epoch": 4.122145580737757, "grad_norm": 5.442002296447754, "learning_rate": 7.414914976230075e-06, "loss": 0.29532060623168943, "memory(GiB)": 77.59, "step": 96215, "token_acc": 0.9367088607594937, "train_speed(iter/s)": 1.459574 }, { "epoch": 4.122359796067006, "grad_norm": 2.009814500808716, "learning_rate": 7.411388779729872e-06, "loss": 0.4960793018341064, "memory(GiB)": 77.59, "step": 96220, "token_acc": 0.8828337874659401, "train_speed(iter/s)": 1.459573 }, { "epoch": 4.122574011396256, "grad_norm": 5.829534530639648, "learning_rate": 7.4078633547623065e-06, "loss": 0.4248184680938721, "memory(GiB)": 77.59, "step": 96225, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.459583 }, { "epoch": 4.122788226725504, "grad_norm": 4.031980037689209, "learning_rate": 7.404338701391256e-06, "loss": 0.21199193000793456, "memory(GiB)": 77.59, "step": 96230, "token_acc": 0.9553264604810997, "train_speed(iter/s)": 1.459591 }, { "epoch": 4.123002442054753, "grad_norm": 2.35026478767395, "learning_rate": 7.400814819680568e-06, "loss": 0.14841647148132325, "memory(GiB)": 77.59, "step": 96235, "token_acc": 0.9514925373134329, "train_speed(iter/s)": 1.459603 }, { "epoch": 4.123216657384003, "grad_norm": 2.0219197273254395, "learning_rate": 7.397291709694065e-06, "loss": 0.38639862537384034, "memory(GiB)": 77.59, "step": 96240, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.459609 }, { "epoch": 4.123430872713251, "grad_norm": 4.215966701507568, "learning_rate": 7.393769371495602e-06, "loss": 0.5459198951721191, "memory(GiB)": 77.59, "step": 96245, "token_acc": 0.8885245901639345, "train_speed(iter/s)": 1.459608 }, { "epoch": 4.1236450880425, "grad_norm": 3.4303529262542725, "learning_rate": 7.390247805148976e-06, "loss": 0.4477085113525391, "memory(GiB)": 77.59, "step": 96250, "token_acc": 0.9084249084249084, "train_speed(iter/s)": 1.459616 }, { "epoch": 4.123859303371749, "grad_norm": 5.288018226623535, "learning_rate": 7.386727010717981e-06, "loss": 0.37794744968414307, "memory(GiB)": 77.59, "step": 96255, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.459632 }, { "epoch": 4.124073518700998, "grad_norm": 3.763664722442627, "learning_rate": 7.3832069882664e-06, "loss": 0.32861392498016356, "memory(GiB)": 77.59, "step": 96260, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.459636 }, { "epoch": 4.124287734030247, "grad_norm": 2.0185210704803467, "learning_rate": 7.379687737857987e-06, "loss": 0.29864494800567626, "memory(GiB)": 77.59, "step": 96265, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.45965 }, { "epoch": 4.124501949359496, "grad_norm": 2.273698568344116, "learning_rate": 7.37616925955652e-06, "loss": 0.4743161201477051, "memory(GiB)": 77.59, "step": 96270, "token_acc": 0.8958990536277602, "train_speed(iter/s)": 1.459649 }, { "epoch": 4.124716164688745, "grad_norm": 1.4936145544052124, "learning_rate": 7.372651553425735e-06, "loss": 0.3514869689941406, "memory(GiB)": 77.59, "step": 96275, "token_acc": 0.9339622641509434, "train_speed(iter/s)": 1.459658 }, { "epoch": 4.124930380017994, "grad_norm": 4.2255859375, "learning_rate": 7.369134619529361e-06, "loss": 0.3003955841064453, "memory(GiB)": 77.59, "step": 96280, "token_acc": 0.9257950530035336, "train_speed(iter/s)": 1.459669 }, { "epoch": 4.125144595347243, "grad_norm": 1.693981409072876, "learning_rate": 7.365618457931101e-06, "loss": 0.2673905372619629, "memory(GiB)": 77.59, "step": 96285, "token_acc": 0.935672514619883, "train_speed(iter/s)": 1.459668 }, { "epoch": 4.125358810676492, "grad_norm": 2.308337926864624, "learning_rate": 7.362103068694653e-06, "loss": 0.26192440986633303, "memory(GiB)": 77.59, "step": 96290, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.459667 }, { "epoch": 4.125573026005741, "grad_norm": 5.600395679473877, "learning_rate": 7.358588451883714e-06, "loss": 0.48440232276916506, "memory(GiB)": 77.59, "step": 96295, "token_acc": 0.875, "train_speed(iter/s)": 1.459666 }, { "epoch": 4.12578724133499, "grad_norm": 2.1284070014953613, "learning_rate": 7.355074607561929e-06, "loss": 0.36704487800598146, "memory(GiB)": 77.59, "step": 96300, "token_acc": 0.9346590909090909, "train_speed(iter/s)": 1.45967 }, { "epoch": 4.126001456664239, "grad_norm": 2.362717390060425, "learning_rate": 7.351561535792984e-06, "loss": 0.12691934108734132, "memory(GiB)": 77.59, "step": 96305, "token_acc": 0.9635036496350365, "train_speed(iter/s)": 1.459674 }, { "epoch": 4.126215671993488, "grad_norm": 0.23203186690807343, "learning_rate": 7.348049236640509e-06, "loss": 0.20752642154693604, "memory(GiB)": 77.59, "step": 96310, "token_acc": 0.9567901234567902, "train_speed(iter/s)": 1.459685 }, { "epoch": 4.126429887322737, "grad_norm": 3.606544256210327, "learning_rate": 7.344537710168136e-06, "loss": 0.3387579441070557, "memory(GiB)": 77.59, "step": 96315, "token_acc": 0.94921875, "train_speed(iter/s)": 1.459695 }, { "epoch": 4.126644102651985, "grad_norm": 3.076935291290283, "learning_rate": 7.3410269564394725e-06, "loss": 0.43465166091918944, "memory(GiB)": 77.59, "step": 96320, "token_acc": 0.9093959731543624, "train_speed(iter/s)": 1.459697 }, { "epoch": 4.126858317981235, "grad_norm": 3.831350803375244, "learning_rate": 7.337516975518116e-06, "loss": 0.32822585105895996, "memory(GiB)": 77.59, "step": 96325, "token_acc": 0.92, "train_speed(iter/s)": 1.459711 }, { "epoch": 4.127072533310484, "grad_norm": 2.396674871444702, "learning_rate": 7.334007767467666e-06, "loss": 0.511937141418457, "memory(GiB)": 77.59, "step": 96330, "token_acc": 0.881578947368421, "train_speed(iter/s)": 1.459715 }, { "epoch": 4.127286748639733, "grad_norm": 2.3970015048980713, "learning_rate": 7.330499332351692e-06, "loss": 0.49962940216064455, "memory(GiB)": 77.59, "step": 96335, "token_acc": 0.889967637540453, "train_speed(iter/s)": 1.459726 }, { "epoch": 4.1275009639689815, "grad_norm": 4.5045623779296875, "learning_rate": 7.326991670233751e-06, "loss": 0.6093682765960693, "memory(GiB)": 77.59, "step": 96340, "token_acc": 0.8775510204081632, "train_speed(iter/s)": 1.459731 }, { "epoch": 4.127715179298231, "grad_norm": 5.565774917602539, "learning_rate": 7.3234847811773755e-06, "loss": 0.30778956413269043, "memory(GiB)": 77.59, "step": 96345, "token_acc": 0.9535714285714286, "train_speed(iter/s)": 1.459742 }, { "epoch": 4.127929394627479, "grad_norm": 3.1203606128692627, "learning_rate": 7.319978665246113e-06, "loss": 0.37822566032409666, "memory(GiB)": 77.59, "step": 96350, "token_acc": 0.9051094890510949, "train_speed(iter/s)": 1.459745 }, { "epoch": 4.128143609956728, "grad_norm": 0.1746746152639389, "learning_rate": 7.31647332250347e-06, "loss": 0.16248605251312256, "memory(GiB)": 77.59, "step": 96355, "token_acc": 0.968503937007874, "train_speed(iter/s)": 1.459749 }, { "epoch": 4.128357825285978, "grad_norm": 3.348982572555542, "learning_rate": 7.312968753012961e-06, "loss": 0.2970393180847168, "memory(GiB)": 77.59, "step": 96360, "token_acc": 0.9387096774193548, "train_speed(iter/s)": 1.459754 }, { "epoch": 4.128572040615227, "grad_norm": 3.4793410301208496, "learning_rate": 7.309464956838063e-06, "loss": 0.25054900646209716, "memory(GiB)": 77.59, "step": 96365, "token_acc": 0.9316546762589928, "train_speed(iter/s)": 1.459753 }, { "epoch": 4.128786255944475, "grad_norm": 7.026834964752197, "learning_rate": 7.30596193404226e-06, "loss": 0.5968217849731445, "memory(GiB)": 77.59, "step": 96370, "token_acc": 0.879245283018868, "train_speed(iter/s)": 1.459761 }, { "epoch": 4.1290004712737245, "grad_norm": 3.632438898086548, "learning_rate": 7.302459684689006e-06, "loss": 0.3491785526275635, "memory(GiB)": 77.59, "step": 96375, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.459769 }, { "epoch": 4.129214686602973, "grad_norm": 6.652271270751953, "learning_rate": 7.298958208841744e-06, "loss": 0.46164937019348146, "memory(GiB)": 77.59, "step": 96380, "token_acc": 0.9086021505376344, "train_speed(iter/s)": 1.45978 }, { "epoch": 4.129428901932222, "grad_norm": 1.8816431760787964, "learning_rate": 7.295457506563902e-06, "loss": 0.4554474353790283, "memory(GiB)": 77.59, "step": 96385, "token_acc": 0.8941605839416058, "train_speed(iter/s)": 1.459793 }, { "epoch": 4.129643117261471, "grad_norm": 2.940006971359253, "learning_rate": 7.291957577918923e-06, "loss": 0.4032088279724121, "memory(GiB)": 77.59, "step": 96390, "token_acc": 0.9329268292682927, "train_speed(iter/s)": 1.459798 }, { "epoch": 4.129857332590721, "grad_norm": 2.275182008743286, "learning_rate": 7.288458422970191e-06, "loss": 0.2792295694351196, "memory(GiB)": 77.59, "step": 96395, "token_acc": 0.9260700389105059, "train_speed(iter/s)": 1.459809 }, { "epoch": 4.130071547919969, "grad_norm": 3.639805316925049, "learning_rate": 7.284960041781102e-06, "loss": 0.4269711971282959, "memory(GiB)": 77.59, "step": 96400, "token_acc": 0.9072164948453608, "train_speed(iter/s)": 1.459813 }, { "epoch": 4.130285763249218, "grad_norm": 1.9992176294326782, "learning_rate": 7.281462434415032e-06, "loss": 0.25500895977020266, "memory(GiB)": 77.59, "step": 96405, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.459817 }, { "epoch": 4.130499978578467, "grad_norm": 0.45860573649406433, "learning_rate": 7.277965600935333e-06, "loss": 0.5269485950469971, "memory(GiB)": 77.59, "step": 96410, "token_acc": 0.8726114649681529, "train_speed(iter/s)": 1.459825 }, { "epoch": 4.130714193907716, "grad_norm": 4.093488693237305, "learning_rate": 7.274469541405376e-06, "loss": 0.20254085063934327, "memory(GiB)": 77.59, "step": 96415, "token_acc": 0.9519650655021834, "train_speed(iter/s)": 1.459829 }, { "epoch": 4.130928409236965, "grad_norm": 4.170108318328857, "learning_rate": 7.270974255888469e-06, "loss": 0.21731324195861818, "memory(GiB)": 77.59, "step": 96420, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.459833 }, { "epoch": 4.131142624566214, "grad_norm": 1.9929355382919312, "learning_rate": 7.267479744447958e-06, "loss": 0.30321598052978516, "memory(GiB)": 77.59, "step": 96425, "token_acc": 0.9554655870445344, "train_speed(iter/s)": 1.459845 }, { "epoch": 4.131356839895463, "grad_norm": 0.773847758769989, "learning_rate": 7.263986007147139e-06, "loss": 0.2157459020614624, "memory(GiB)": 77.59, "step": 96430, "token_acc": 0.9368029739776952, "train_speed(iter/s)": 1.459847 }, { "epoch": 4.131571055224712, "grad_norm": 3.9934890270233154, "learning_rate": 7.260493044049299e-06, "loss": 0.6771902084350586, "memory(GiB)": 77.59, "step": 96435, "token_acc": 0.8706293706293706, "train_speed(iter/s)": 1.459845 }, { "epoch": 4.13178527055396, "grad_norm": 3.710667848587036, "learning_rate": 7.257000855217727e-06, "loss": 0.3264575958251953, "memory(GiB)": 77.59, "step": 96440, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.459845 }, { "epoch": 4.13199948588321, "grad_norm": 2.8145952224731445, "learning_rate": 7.253509440715667e-06, "loss": 0.3412631511688232, "memory(GiB)": 77.59, "step": 96445, "token_acc": 0.89568345323741, "train_speed(iter/s)": 1.459843 }, { "epoch": 4.132213701212459, "grad_norm": 2.4131031036376953, "learning_rate": 7.25001880060639e-06, "loss": 0.26494498252868653, "memory(GiB)": 77.59, "step": 96450, "token_acc": 0.9472049689440993, "train_speed(iter/s)": 1.459846 }, { "epoch": 4.132427916541708, "grad_norm": 0.2843936085700989, "learning_rate": 7.246528934953128e-06, "loss": 0.23514223098754883, "memory(GiB)": 77.59, "step": 96455, "token_acc": 0.9504950495049505, "train_speed(iter/s)": 1.459851 }, { "epoch": 4.1326421318709565, "grad_norm": 1.965881109237671, "learning_rate": 7.243039843819105e-06, "loss": 0.4938494682312012, "memory(GiB)": 77.59, "step": 96460, "token_acc": 0.8939393939393939, "train_speed(iter/s)": 1.459845 }, { "epoch": 4.132856347200206, "grad_norm": 7.20706844329834, "learning_rate": 7.2395515272675186e-06, "loss": 0.14912973642349242, "memory(GiB)": 77.59, "step": 96465, "token_acc": 0.9595141700404858, "train_speed(iter/s)": 1.459851 }, { "epoch": 4.133070562529454, "grad_norm": 3.2789433002471924, "learning_rate": 7.236063985361563e-06, "loss": 0.5888936519622803, "memory(GiB)": 77.59, "step": 96470, "token_acc": 0.8698224852071006, "train_speed(iter/s)": 1.459857 }, { "epoch": 4.133284777858703, "grad_norm": 2.896785020828247, "learning_rate": 7.232577218164427e-06, "loss": 0.3819873809814453, "memory(GiB)": 77.59, "step": 96475, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.459859 }, { "epoch": 4.133498993187953, "grad_norm": 2.7690935134887695, "learning_rate": 7.229091225739282e-06, "loss": 0.30672905445098875, "memory(GiB)": 77.59, "step": 96480, "token_acc": 0.9158576051779935, "train_speed(iter/s)": 1.459864 }, { "epoch": 4.133713208517202, "grad_norm": 7.1715803146362305, "learning_rate": 7.225606008149266e-06, "loss": 0.43659701347351076, "memory(GiB)": 77.59, "step": 96485, "token_acc": 0.918429003021148, "train_speed(iter/s)": 1.459867 }, { "epoch": 4.13392742384645, "grad_norm": 4.819521903991699, "learning_rate": 7.222121565457529e-06, "loss": 0.583427619934082, "memory(GiB)": 77.59, "step": 96490, "token_acc": 0.8700361010830325, "train_speed(iter/s)": 1.459879 }, { "epoch": 4.1341416391756995, "grad_norm": 3.294630765914917, "learning_rate": 7.2186378977271705e-06, "loss": 0.19144665002822875, "memory(GiB)": 77.59, "step": 96495, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.459882 }, { "epoch": 4.134355854504949, "grad_norm": 3.091024875640869, "learning_rate": 7.215155005021323e-06, "loss": 0.14338622093200684, "memory(GiB)": 77.59, "step": 96500, "token_acc": 0.9747292418772563, "train_speed(iter/s)": 1.459883 }, { "epoch": 4.134355854504949, "eval_loss": 2.6473097801208496, "eval_runtime": 11.9987, "eval_samples_per_second": 8.334, "eval_steps_per_second": 8.334, "eval_token_acc": 0.4336283185840708, "step": 96500 }, { "epoch": 4.134570069834197, "grad_norm": 6.101513862609863, "learning_rate": 7.211672887403087e-06, "loss": 0.4457258224487305, "memory(GiB)": 77.59, "step": 96505, "token_acc": 0.5542056074766355, "train_speed(iter/s)": 1.459601 }, { "epoch": 4.134784285163446, "grad_norm": 4.609130382537842, "learning_rate": 7.208191544935538e-06, "loss": 0.2711066246032715, "memory(GiB)": 77.59, "step": 96510, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.459599 }, { "epoch": 4.134998500492696, "grad_norm": 0.7249208688735962, "learning_rate": 7.204710977681739e-06, "loss": 0.31815998554229735, "memory(GiB)": 77.59, "step": 96515, "token_acc": 0.9373134328358209, "train_speed(iter/s)": 1.459596 }, { "epoch": 4.135212715821944, "grad_norm": 6.076536178588867, "learning_rate": 7.201231185704749e-06, "loss": 0.4149313449859619, "memory(GiB)": 77.59, "step": 96520, "token_acc": 0.8943396226415095, "train_speed(iter/s)": 1.459605 }, { "epoch": 4.135426931151193, "grad_norm": 3.8825459480285645, "learning_rate": 7.197752169067601e-06, "loss": 0.3322422504425049, "memory(GiB)": 77.59, "step": 96525, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.459609 }, { "epoch": 4.135641146480443, "grad_norm": 1.02797532081604, "learning_rate": 7.194273927833312e-06, "loss": 0.5541154861450195, "memory(GiB)": 77.59, "step": 96530, "token_acc": 0.8447368421052631, "train_speed(iter/s)": 1.459619 }, { "epoch": 4.135855361809691, "grad_norm": 3.902383327484131, "learning_rate": 7.190796462064919e-06, "loss": 0.4552985668182373, "memory(GiB)": 77.59, "step": 96535, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.459621 }, { "epoch": 4.13606957713894, "grad_norm": 3.3150691986083984, "learning_rate": 7.187319771825407e-06, "loss": 0.3798069953918457, "memory(GiB)": 77.59, "step": 96540, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.459623 }, { "epoch": 4.1362837924681894, "grad_norm": 2.2497286796569824, "learning_rate": 7.183843857177757e-06, "loss": 0.44881601333618165, "memory(GiB)": 77.59, "step": 96545, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.459627 }, { "epoch": 4.136498007797438, "grad_norm": 2.3926868438720703, "learning_rate": 7.180368718184943e-06, "loss": 0.27299702167510986, "memory(GiB)": 77.59, "step": 96550, "token_acc": 0.9451476793248945, "train_speed(iter/s)": 1.459633 }, { "epoch": 4.136712223126687, "grad_norm": 2.3682851791381836, "learning_rate": 7.176894354909908e-06, "loss": 0.24278151988983154, "memory(GiB)": 77.59, "step": 96555, "token_acc": 0.968944099378882, "train_speed(iter/s)": 1.459635 }, { "epoch": 4.136926438455936, "grad_norm": 1.1285955905914307, "learning_rate": 7.173420767415611e-06, "loss": 0.2919813394546509, "memory(GiB)": 77.59, "step": 96560, "token_acc": 0.939297124600639, "train_speed(iter/s)": 1.459632 }, { "epoch": 4.137140653785185, "grad_norm": 3.3118269443511963, "learning_rate": 7.169947955764961e-06, "loss": 0.2774558782577515, "memory(GiB)": 77.59, "step": 96565, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.459636 }, { "epoch": 4.137354869114434, "grad_norm": 6.967087268829346, "learning_rate": 7.166475920020888e-06, "loss": 0.5034801483154296, "memory(GiB)": 77.59, "step": 96570, "token_acc": 0.9061488673139159, "train_speed(iter/s)": 1.459637 }, { "epoch": 4.137569084443683, "grad_norm": 5.60983943939209, "learning_rate": 7.163004660246292e-06, "loss": 0.3364851951599121, "memory(GiB)": 77.59, "step": 96575, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.459643 }, { "epoch": 4.137783299772932, "grad_norm": 4.234689712524414, "learning_rate": 7.159534176504046e-06, "loss": 0.2353134870529175, "memory(GiB)": 77.59, "step": 96580, "token_acc": 0.9407407407407408, "train_speed(iter/s)": 1.459647 }, { "epoch": 4.137997515102181, "grad_norm": 3.690058946609497, "learning_rate": 7.1560644688570324e-06, "loss": 0.39130964279174807, "memory(GiB)": 77.59, "step": 96585, "token_acc": 0.9088050314465409, "train_speed(iter/s)": 1.45965 }, { "epoch": 4.13821173043143, "grad_norm": 5.2561140060424805, "learning_rate": 7.152595537368084e-06, "loss": 0.4312914848327637, "memory(GiB)": 77.59, "step": 96590, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.45966 }, { "epoch": 4.1384259457606785, "grad_norm": 1.8836129903793335, "learning_rate": 7.149127382100074e-06, "loss": 0.20585041046142577, "memory(GiB)": 77.59, "step": 96595, "token_acc": 0.959349593495935, "train_speed(iter/s)": 1.459658 }, { "epoch": 4.138640161089928, "grad_norm": 4.1508684158325195, "learning_rate": 7.145660003115822e-06, "loss": 0.34194629192352294, "memory(GiB)": 77.59, "step": 96600, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.459667 }, { "epoch": 4.138854376419177, "grad_norm": 5.91136360168457, "learning_rate": 7.142193400478136e-06, "loss": 0.5142143249511719, "memory(GiB)": 77.59, "step": 96605, "token_acc": 0.8993055555555556, "train_speed(iter/s)": 1.459678 }, { "epoch": 4.139068591748425, "grad_norm": 3.523458242416382, "learning_rate": 7.138727574249821e-06, "loss": 0.2664497852325439, "memory(GiB)": 77.59, "step": 96610, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.459678 }, { "epoch": 4.139282807077675, "grad_norm": 4.4443864822387695, "learning_rate": 7.135262524493652e-06, "loss": 0.3432452201843262, "memory(GiB)": 77.59, "step": 96615, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.459678 }, { "epoch": 4.139497022406924, "grad_norm": 3.9003894329071045, "learning_rate": 7.131798251272426e-06, "loss": 0.40209450721740725, "memory(GiB)": 77.59, "step": 96620, "token_acc": 0.9212328767123288, "train_speed(iter/s)": 1.45968 }, { "epoch": 4.139711237736172, "grad_norm": 0.5714368224143982, "learning_rate": 7.128334754648891e-06, "loss": 0.1781236171722412, "memory(GiB)": 77.59, "step": 96625, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.459681 }, { "epoch": 4.1399254530654215, "grad_norm": 2.257314443588257, "learning_rate": 7.124872034685781e-06, "loss": 0.14120662212371826, "memory(GiB)": 77.59, "step": 96630, "token_acc": 0.9781931464174455, "train_speed(iter/s)": 1.459682 }, { "epoch": 4.140139668394671, "grad_norm": 0.12901563942432404, "learning_rate": 7.1214100914458404e-06, "loss": 0.19160878658294678, "memory(GiB)": 77.59, "step": 96635, "token_acc": 0.9425675675675675, "train_speed(iter/s)": 1.459689 }, { "epoch": 4.140353883723919, "grad_norm": 0.7760873436927795, "learning_rate": 7.117948924991769e-06, "loss": 0.3810176134109497, "memory(GiB)": 77.59, "step": 96640, "token_acc": 0.9300411522633745, "train_speed(iter/s)": 1.459694 }, { "epoch": 4.140568099053168, "grad_norm": 2.3177602291107178, "learning_rate": 7.1144885353862714e-06, "loss": 0.3932011365890503, "memory(GiB)": 77.59, "step": 96645, "token_acc": 0.9020771513353115, "train_speed(iter/s)": 1.459697 }, { "epoch": 4.140782314382418, "grad_norm": 2.0314249992370605, "learning_rate": 7.111028922692065e-06, "loss": 0.45679507255554197, "memory(GiB)": 77.59, "step": 96650, "token_acc": 0.903125, "train_speed(iter/s)": 1.4597 }, { "epoch": 4.140996529711666, "grad_norm": 3.9489657878875732, "learning_rate": 7.107570086971793e-06, "loss": 0.3982600212097168, "memory(GiB)": 77.59, "step": 96655, "token_acc": 0.9204545454545454, "train_speed(iter/s)": 1.459706 }, { "epoch": 4.141210745040915, "grad_norm": 4.7429704666137695, "learning_rate": 7.104112028288135e-06, "loss": 0.3518668174743652, "memory(GiB)": 77.59, "step": 96660, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.459716 }, { "epoch": 4.1414249603701645, "grad_norm": 2.896178722381592, "learning_rate": 7.100654746703722e-06, "loss": 0.26163079738616946, "memory(GiB)": 77.59, "step": 96665, "token_acc": 0.9508670520231214, "train_speed(iter/s)": 1.459724 }, { "epoch": 4.141639175699413, "grad_norm": 6.041184425354004, "learning_rate": 7.097198242281189e-06, "loss": 0.41567130088806153, "memory(GiB)": 77.59, "step": 96670, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.459726 }, { "epoch": 4.141853391028662, "grad_norm": 5.084014892578125, "learning_rate": 7.093742515083146e-06, "loss": 0.33705227375030516, "memory(GiB)": 77.59, "step": 96675, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.459728 }, { "epoch": 4.142067606357911, "grad_norm": 4.293116092681885, "learning_rate": 7.0902875651722215e-06, "loss": 0.38639767169952394, "memory(GiB)": 77.59, "step": 96680, "token_acc": 0.9248554913294798, "train_speed(iter/s)": 1.45973 }, { "epoch": 4.14228182168716, "grad_norm": 0.5389736294746399, "learning_rate": 7.0868333926109865e-06, "loss": 0.23350563049316406, "memory(GiB)": 77.59, "step": 96685, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.459739 }, { "epoch": 4.142496037016409, "grad_norm": 5.201595783233643, "learning_rate": 7.08337999746202e-06, "loss": 0.2991873025894165, "memory(GiB)": 77.59, "step": 96690, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.459746 }, { "epoch": 4.142710252345658, "grad_norm": 4.483455181121826, "learning_rate": 7.079927379787887e-06, "loss": 0.3675997734069824, "memory(GiB)": 77.59, "step": 96695, "token_acc": 0.9046153846153846, "train_speed(iter/s)": 1.459744 }, { "epoch": 4.142924467674907, "grad_norm": 2.738240957260132, "learning_rate": 7.076475539651117e-06, "loss": 0.45646028518676757, "memory(GiB)": 77.59, "step": 96700, "token_acc": 0.8908450704225352, "train_speed(iter/s)": 1.459747 }, { "epoch": 4.143138683004156, "grad_norm": 2.458845615386963, "learning_rate": 7.073024477114276e-06, "loss": 0.29805445671081543, "memory(GiB)": 77.59, "step": 96705, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.459754 }, { "epoch": 4.143352898333405, "grad_norm": 4.766048431396484, "learning_rate": 7.069574192239858e-06, "loss": 0.5774873733520508, "memory(GiB)": 77.59, "step": 96710, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.459759 }, { "epoch": 4.1435671136626535, "grad_norm": 0.34822842478752136, "learning_rate": 7.06612468509037e-06, "loss": 0.2793873310089111, "memory(GiB)": 77.59, "step": 96715, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.459759 }, { "epoch": 4.143781328991903, "grad_norm": 4.96731424331665, "learning_rate": 7.062675955728315e-06, "loss": 0.39994828701019286, "memory(GiB)": 77.59, "step": 96720, "token_acc": 0.9161290322580645, "train_speed(iter/s)": 1.459762 }, { "epoch": 4.143995544321152, "grad_norm": 0.6317316889762878, "learning_rate": 7.0592280042161675e-06, "loss": 0.12954236268997193, "memory(GiB)": 77.59, "step": 96725, "token_acc": 0.9602888086642599, "train_speed(iter/s)": 1.459767 }, { "epoch": 4.1442097596504, "grad_norm": 2.4795026779174805, "learning_rate": 7.055780830616382e-06, "loss": 0.32052125930786135, "memory(GiB)": 77.59, "step": 96730, "token_acc": 0.9251700680272109, "train_speed(iter/s)": 1.459777 }, { "epoch": 4.14442397497965, "grad_norm": 3.7027323246002197, "learning_rate": 7.052334434991403e-06, "loss": 0.2221242904663086, "memory(GiB)": 77.59, "step": 96735, "token_acc": 0.9501661129568106, "train_speed(iter/s)": 1.45978 }, { "epoch": 4.144638190308899, "grad_norm": 3.2260425090789795, "learning_rate": 7.048888817403687e-06, "loss": 0.3074974060058594, "memory(GiB)": 77.59, "step": 96740, "token_acc": 0.9381107491856677, "train_speed(iter/s)": 1.459783 }, { "epoch": 4.144852405638147, "grad_norm": 3.834015130996704, "learning_rate": 7.045443977915639e-06, "loss": 0.2174443006515503, "memory(GiB)": 77.59, "step": 96745, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.459782 }, { "epoch": 4.1450666209673965, "grad_norm": 5.710634708404541, "learning_rate": 7.04199991658967e-06, "loss": 0.22425053119659424, "memory(GiB)": 77.59, "step": 96750, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.459797 }, { "epoch": 4.145280836296646, "grad_norm": 0.4395253658294678, "learning_rate": 7.038556633488169e-06, "loss": 0.20714905261993408, "memory(GiB)": 77.59, "step": 96755, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.459799 }, { "epoch": 4.145495051625894, "grad_norm": 4.711230278015137, "learning_rate": 7.035114128673503e-06, "loss": 0.23834648132324218, "memory(GiB)": 77.59, "step": 96760, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.459804 }, { "epoch": 4.145709266955143, "grad_norm": 2.8886475563049316, "learning_rate": 7.031672402208061e-06, "loss": 0.44417858123779297, "memory(GiB)": 77.59, "step": 96765, "token_acc": 0.9115646258503401, "train_speed(iter/s)": 1.459805 }, { "epoch": 4.145923482284393, "grad_norm": 2.3028340339660645, "learning_rate": 7.028231454154183e-06, "loss": 0.15273754596710204, "memory(GiB)": 77.59, "step": 96770, "token_acc": 0.9590443686006825, "train_speed(iter/s)": 1.459813 }, { "epoch": 4.146137697613641, "grad_norm": 5.159353733062744, "learning_rate": 7.0247912845742005e-06, "loss": 0.38460679054260255, "memory(GiB)": 77.59, "step": 96775, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.459818 }, { "epoch": 4.14635191294289, "grad_norm": 3.101559638977051, "learning_rate": 7.021351893530437e-06, "loss": 0.38289604187011717, "memory(GiB)": 77.59, "step": 96780, "token_acc": 0.9076433121019108, "train_speed(iter/s)": 1.459819 }, { "epoch": 4.1465661282721396, "grad_norm": 0.5486238598823547, "learning_rate": 7.0179132810851924e-06, "loss": 0.2079148769378662, "memory(GiB)": 77.59, "step": 96785, "token_acc": 0.956989247311828, "train_speed(iter/s)": 1.459827 }, { "epoch": 4.146780343601388, "grad_norm": 6.8495378494262695, "learning_rate": 7.014475447300767e-06, "loss": 0.6590514183044434, "memory(GiB)": 77.59, "step": 96790, "token_acc": 0.8686567164179104, "train_speed(iter/s)": 1.459829 }, { "epoch": 4.146994558930637, "grad_norm": 2.6532301902770996, "learning_rate": 7.011038392239455e-06, "loss": 0.6696853637695312, "memory(GiB)": 77.59, "step": 96795, "token_acc": 0.8067885117493473, "train_speed(iter/s)": 1.459831 }, { "epoch": 4.147208774259886, "grad_norm": 0.16093991696834564, "learning_rate": 7.007602115963513e-06, "loss": 0.4989445686340332, "memory(GiB)": 77.59, "step": 96800, "token_acc": 0.9221311475409836, "train_speed(iter/s)": 1.459837 }, { "epoch": 4.147422989589135, "grad_norm": 0.2759771943092346, "learning_rate": 7.004166618535185e-06, "loss": 0.33355154991149905, "memory(GiB)": 77.59, "step": 96805, "token_acc": 0.9246861924686193, "train_speed(iter/s)": 1.45985 }, { "epoch": 4.147637204918384, "grad_norm": 2.901329278945923, "learning_rate": 7.00073190001671e-06, "loss": 0.5570498943328858, "memory(GiB)": 77.59, "step": 96810, "token_acc": 0.8869257950530035, "train_speed(iter/s)": 1.459857 }, { "epoch": 4.147851420247633, "grad_norm": 2.408752918243408, "learning_rate": 6.997297960470317e-06, "loss": 0.31950814723968507, "memory(GiB)": 77.59, "step": 96815, "token_acc": 0.952, "train_speed(iter/s)": 1.459862 }, { "epoch": 4.148065635576882, "grad_norm": 3.4434633255004883, "learning_rate": 6.9938647999582e-06, "loss": 0.42647337913513184, "memory(GiB)": 77.59, "step": 96820, "token_acc": 0.8847583643122676, "train_speed(iter/s)": 1.459864 }, { "epoch": 4.148279850906131, "grad_norm": 5.046436309814453, "learning_rate": 6.990432418542575e-06, "loss": 0.42771759033203127, "memory(GiB)": 77.59, "step": 96825, "token_acc": 0.9239543726235742, "train_speed(iter/s)": 1.459863 }, { "epoch": 4.14849406623538, "grad_norm": 2.300387382507324, "learning_rate": 6.987000816285611e-06, "loss": 0.46449804306030273, "memory(GiB)": 77.59, "step": 96830, "token_acc": 0.909433962264151, "train_speed(iter/s)": 1.459866 }, { "epoch": 4.148708281564629, "grad_norm": 5.9207587242126465, "learning_rate": 6.983569993249478e-06, "loss": 0.3196870803833008, "memory(GiB)": 77.59, "step": 96835, "token_acc": 0.9171597633136095, "train_speed(iter/s)": 1.45988 }, { "epoch": 4.148922496893878, "grad_norm": 2.7450249195098877, "learning_rate": 6.9801399494963285e-06, "loss": 0.5243780612945557, "memory(GiB)": 77.59, "step": 96840, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.459884 }, { "epoch": 4.149136712223127, "grad_norm": 5.710129737854004, "learning_rate": 6.976710685088289e-06, "loss": 0.356002140045166, "memory(GiB)": 77.59, "step": 96845, "token_acc": 0.9262295081967213, "train_speed(iter/s)": 1.459889 }, { "epoch": 4.1493509275523754, "grad_norm": 1.8480194807052612, "learning_rate": 6.973282200087506e-06, "loss": 0.23385119438171387, "memory(GiB)": 77.59, "step": 96850, "token_acc": 0.9575971731448764, "train_speed(iter/s)": 1.459892 }, { "epoch": 4.149565142881625, "grad_norm": 2.774425506591797, "learning_rate": 6.969854494556077e-06, "loss": 0.21279301643371581, "memory(GiB)": 77.59, "step": 96855, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.459899 }, { "epoch": 4.149779358210874, "grad_norm": 3.7982091903686523, "learning_rate": 6.966427568556089e-06, "loss": 0.471677303314209, "memory(GiB)": 77.59, "step": 96860, "token_acc": 0.9067796610169492, "train_speed(iter/s)": 1.45991 }, { "epoch": 4.149993573540122, "grad_norm": 3.3276379108428955, "learning_rate": 6.963001422149646e-06, "loss": 0.434310245513916, "memory(GiB)": 77.59, "step": 96865, "token_acc": 0.8957654723127035, "train_speed(iter/s)": 1.459911 }, { "epoch": 4.150207788869372, "grad_norm": 4.3723039627075195, "learning_rate": 6.959576055398798e-06, "loss": 0.4975308895111084, "memory(GiB)": 77.59, "step": 96870, "token_acc": 0.8915254237288136, "train_speed(iter/s)": 1.459912 }, { "epoch": 4.150422004198621, "grad_norm": 2.4318695068359375, "learning_rate": 6.956151468365613e-06, "loss": 0.3813388109207153, "memory(GiB)": 77.59, "step": 96875, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.459917 }, { "epoch": 4.150636219527869, "grad_norm": 0.33844709396362305, "learning_rate": 6.952727661112107e-06, "loss": 0.12493792772293091, "memory(GiB)": 77.59, "step": 96880, "token_acc": 0.9704918032786886, "train_speed(iter/s)": 1.459919 }, { "epoch": 4.1508504348571185, "grad_norm": 3.564300775527954, "learning_rate": 6.949304633700332e-06, "loss": 0.2563016891479492, "memory(GiB)": 77.59, "step": 96885, "token_acc": 0.946969696969697, "train_speed(iter/s)": 1.459918 }, { "epoch": 4.151064650186368, "grad_norm": 1.5722054243087769, "learning_rate": 6.945882386192293e-06, "loss": 0.29379293918609617, "memory(GiB)": 77.59, "step": 96890, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.459921 }, { "epoch": 4.151278865515616, "grad_norm": 6.043455600738525, "learning_rate": 6.942460918649979e-06, "loss": 0.2375568151473999, "memory(GiB)": 77.59, "step": 96895, "token_acc": 0.9634146341463414, "train_speed(iter/s)": 1.459926 }, { "epoch": 4.151493080844865, "grad_norm": 1.963138461112976, "learning_rate": 6.939040231135374e-06, "loss": 0.3089661836624146, "memory(GiB)": 77.59, "step": 96900, "token_acc": 0.9186440677966101, "train_speed(iter/s)": 1.459921 }, { "epoch": 4.151707296174115, "grad_norm": 2.659320116043091, "learning_rate": 6.935620323710446e-06, "loss": 0.32090167999267577, "memory(GiB)": 77.59, "step": 96905, "token_acc": 0.9033457249070632, "train_speed(iter/s)": 1.459927 }, { "epoch": 4.151921511503363, "grad_norm": 4.566887855529785, "learning_rate": 6.9322011964371605e-06, "loss": 0.42245917320251464, "memory(GiB)": 77.59, "step": 96910, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.459937 }, { "epoch": 4.152135726832612, "grad_norm": 5.504729270935059, "learning_rate": 6.928782849377447e-06, "loss": 0.3895397186279297, "memory(GiB)": 77.59, "step": 96915, "token_acc": 0.9148936170212766, "train_speed(iter/s)": 1.459939 }, { "epoch": 4.1523499421618615, "grad_norm": 5.5536088943481445, "learning_rate": 6.925365282593244e-06, "loss": 0.3940650463104248, "memory(GiB)": 77.59, "step": 96920, "token_acc": 0.8921568627450981, "train_speed(iter/s)": 1.459938 }, { "epoch": 4.15256415749111, "grad_norm": 2.017331123352051, "learning_rate": 6.921948496146452e-06, "loss": 0.46739020347595217, "memory(GiB)": 77.59, "step": 96925, "token_acc": 0.9240924092409241, "train_speed(iter/s)": 1.459948 }, { "epoch": 4.152778372820359, "grad_norm": 1.9591145515441895, "learning_rate": 6.918532490098961e-06, "loss": 0.2857907772064209, "memory(GiB)": 77.59, "step": 96930, "token_acc": 0.9522388059701492, "train_speed(iter/s)": 1.459953 }, { "epoch": 4.152992588149608, "grad_norm": 4.652282238006592, "learning_rate": 6.915117264512672e-06, "loss": 0.5403074264526367, "memory(GiB)": 77.59, "step": 96935, "token_acc": 0.9249146757679181, "train_speed(iter/s)": 1.459957 }, { "epoch": 4.153206803478857, "grad_norm": 4.020087718963623, "learning_rate": 6.911702819449456e-06, "loss": 0.42928352355957033, "memory(GiB)": 77.59, "step": 96940, "token_acc": 0.889795918367347, "train_speed(iter/s)": 1.459962 }, { "epoch": 4.153421018808106, "grad_norm": 1.2521945238113403, "learning_rate": 6.9082891549711705e-06, "loss": 0.3644322156906128, "memory(GiB)": 77.59, "step": 96945, "token_acc": 0.9233576642335767, "train_speed(iter/s)": 1.45997 }, { "epoch": 4.153635234137355, "grad_norm": 2.030139446258545, "learning_rate": 6.904876271139643e-06, "loss": 0.2359405755996704, "memory(GiB)": 77.59, "step": 96950, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.459972 }, { "epoch": 4.153849449466604, "grad_norm": 4.133825778961182, "learning_rate": 6.901464168016713e-06, "loss": 0.39000983238220216, "memory(GiB)": 77.59, "step": 96955, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.459975 }, { "epoch": 4.154063664795853, "grad_norm": 1.35000741481781, "learning_rate": 6.898052845664188e-06, "loss": 0.09985465407371522, "memory(GiB)": 77.59, "step": 96960, "token_acc": 0.9887640449438202, "train_speed(iter/s)": 1.45998 }, { "epoch": 4.154277880125102, "grad_norm": 6.073630332946777, "learning_rate": 6.894642304143856e-06, "loss": 0.35747315883636477, "memory(GiB)": 77.59, "step": 96965, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.459981 }, { "epoch": 4.1544920954543505, "grad_norm": 2.3014447689056396, "learning_rate": 6.891232543517529e-06, "loss": 0.3382347822189331, "memory(GiB)": 77.59, "step": 96970, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.459993 }, { "epoch": 4.1547063107836, "grad_norm": 3.5237269401550293, "learning_rate": 6.887823563846962e-06, "loss": 0.5407455921173095, "memory(GiB)": 77.59, "step": 96975, "token_acc": 0.8746438746438746, "train_speed(iter/s)": 1.46 }, { "epoch": 4.154920526112849, "grad_norm": 2.9422240257263184, "learning_rate": 6.884415365193913e-06, "loss": 0.6022521018981933, "memory(GiB)": 77.59, "step": 96980, "token_acc": 0.870253164556962, "train_speed(iter/s)": 1.46001 }, { "epoch": 4.155134741442097, "grad_norm": 0.9564924836158752, "learning_rate": 6.881007947620127e-06, "loss": 0.1840253710746765, "memory(GiB)": 77.59, "step": 96985, "token_acc": 0.9590643274853801, "train_speed(iter/s)": 1.460011 }, { "epoch": 4.155348956771347, "grad_norm": 2.748431921005249, "learning_rate": 6.877601311187321e-06, "loss": 0.36679089069366455, "memory(GiB)": 77.59, "step": 96990, "token_acc": 0.9265734265734266, "train_speed(iter/s)": 1.460009 }, { "epoch": 4.155563172100596, "grad_norm": 5.371917247772217, "learning_rate": 6.874195455957227e-06, "loss": 0.37884974479675293, "memory(GiB)": 77.59, "step": 96995, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.460007 }, { "epoch": 4.155777387429844, "grad_norm": 3.076990842819214, "learning_rate": 6.870790381991538e-06, "loss": 0.41310672760009765, "memory(GiB)": 77.59, "step": 97000, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.460011 }, { "epoch": 4.155777387429844, "eval_loss": 2.3518643379211426, "eval_runtime": 11.6586, "eval_samples_per_second": 8.577, "eval_steps_per_second": 8.577, "eval_token_acc": 0.46142433234421365, "step": 97000 }, { "epoch": 4.1559916027590935, "grad_norm": 4.257608890533447, "learning_rate": 6.86738608935194e-06, "loss": 0.7286584854125977, "memory(GiB)": 77.59, "step": 97005, "token_acc": 0.5823170731707317, "train_speed(iter/s)": 1.459729 }, { "epoch": 4.156205818088343, "grad_norm": 1.7465745210647583, "learning_rate": 6.863982578100098e-06, "loss": 0.2922464370727539, "memory(GiB)": 77.59, "step": 97010, "token_acc": 0.9262820512820513, "train_speed(iter/s)": 1.459742 }, { "epoch": 4.156420033417591, "grad_norm": 5.572153568267822, "learning_rate": 6.860579848297683e-06, "loss": 0.3873037576675415, "memory(GiB)": 77.59, "step": 97015, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.459743 }, { "epoch": 4.15663424874684, "grad_norm": 5.500073432922363, "learning_rate": 6.857177900006317e-06, "loss": 0.3708317518234253, "memory(GiB)": 77.59, "step": 97020, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.459757 }, { "epoch": 4.15684846407609, "grad_norm": 3.523866653442383, "learning_rate": 6.853776733287665e-06, "loss": 0.4259768486022949, "memory(GiB)": 77.59, "step": 97025, "token_acc": 0.888135593220339, "train_speed(iter/s)": 1.459756 }, { "epoch": 4.157062679405338, "grad_norm": 3.1055803298950195, "learning_rate": 6.850376348203313e-06, "loss": 0.35391721725463865, "memory(GiB)": 77.59, "step": 97030, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.459776 }, { "epoch": 4.157276894734587, "grad_norm": 4.450345516204834, "learning_rate": 6.8469767448148755e-06, "loss": 0.4507895946502686, "memory(GiB)": 77.59, "step": 97035, "token_acc": 0.8842975206611571, "train_speed(iter/s)": 1.459779 }, { "epoch": 4.1574911100638365, "grad_norm": 4.5472636222839355, "learning_rate": 6.843577923183936e-06, "loss": 0.3659677505493164, "memory(GiB)": 77.59, "step": 97040, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.459781 }, { "epoch": 4.157705325393085, "grad_norm": 2.155871629714966, "learning_rate": 6.840179883372066e-06, "loss": 0.37338135242462156, "memory(GiB)": 77.59, "step": 97045, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.459781 }, { "epoch": 4.157919540722334, "grad_norm": 2.628673553466797, "learning_rate": 6.836782625440813e-06, "loss": 0.14011192321777344, "memory(GiB)": 77.59, "step": 97050, "token_acc": 0.9626556016597511, "train_speed(iter/s)": 1.459794 }, { "epoch": 4.158133756051583, "grad_norm": 5.427844524383545, "learning_rate": 6.833386149451748e-06, "loss": 0.20210881233215333, "memory(GiB)": 77.59, "step": 97055, "token_acc": 0.9516129032258065, "train_speed(iter/s)": 1.459798 }, { "epoch": 4.158347971380832, "grad_norm": 4.248945236206055, "learning_rate": 6.8299904554663785e-06, "loss": 0.549284839630127, "memory(GiB)": 77.59, "step": 97060, "token_acc": 0.8791208791208791, "train_speed(iter/s)": 1.459804 }, { "epoch": 4.158562186710081, "grad_norm": 3.9288039207458496, "learning_rate": 6.8265955435462325e-06, "loss": 0.627753734588623, "memory(GiB)": 77.59, "step": 97065, "token_acc": 0.8434163701067615, "train_speed(iter/s)": 1.459812 }, { "epoch": 4.15877640203933, "grad_norm": 0.10822742432355881, "learning_rate": 6.823201413752811e-06, "loss": 0.3769437551498413, "memory(GiB)": 77.59, "step": 97070, "token_acc": 0.9117647058823529, "train_speed(iter/s)": 1.459824 }, { "epoch": 4.158990617368579, "grad_norm": 0.055102039128541946, "learning_rate": 6.819808066147587e-06, "loss": 0.2964228868484497, "memory(GiB)": 77.59, "step": 97075, "token_acc": 0.9248366013071896, "train_speed(iter/s)": 1.459824 }, { "epoch": 4.159204832697828, "grad_norm": 3.5827651023864746, "learning_rate": 6.816415500792056e-06, "loss": 0.3246347665786743, "memory(GiB)": 77.59, "step": 97080, "token_acc": 0.9205298013245033, "train_speed(iter/s)": 1.459826 }, { "epoch": 4.159419048027077, "grad_norm": 2.370065927505493, "learning_rate": 6.813023717747652e-06, "loss": 0.3765261173248291, "memory(GiB)": 77.59, "step": 97085, "token_acc": 0.9228070175438596, "train_speed(iter/s)": 1.459829 }, { "epoch": 4.1596332633563255, "grad_norm": 1.3338936567306519, "learning_rate": 6.8096327170758535e-06, "loss": 0.15258713960647582, "memory(GiB)": 77.59, "step": 97090, "token_acc": 0.965034965034965, "train_speed(iter/s)": 1.45983 }, { "epoch": 4.159847478685575, "grad_norm": 5.640203952789307, "learning_rate": 6.806242498838072e-06, "loss": 0.9925989151000977, "memory(GiB)": 77.59, "step": 97095, "token_acc": 0.8327526132404182, "train_speed(iter/s)": 1.459841 }, { "epoch": 4.160061694014824, "grad_norm": 6.236618995666504, "learning_rate": 6.802853063095727e-06, "loss": 0.251239013671875, "memory(GiB)": 77.59, "step": 97100, "token_acc": 0.9375, "train_speed(iter/s)": 1.459846 }, { "epoch": 4.160275909344072, "grad_norm": 1.651166319847107, "learning_rate": 6.799464409910222e-06, "loss": 0.25877859592437746, "memory(GiB)": 77.59, "step": 97105, "token_acc": 0.9421221864951769, "train_speed(iter/s)": 1.459851 }, { "epoch": 4.160490124673322, "grad_norm": 1.9501855373382568, "learning_rate": 6.796076539342933e-06, "loss": 0.41582489013671875, "memory(GiB)": 77.59, "step": 97110, "token_acc": 0.9080882352941176, "train_speed(iter/s)": 1.459869 }, { "epoch": 4.160704340002571, "grad_norm": 0.7531179189682007, "learning_rate": 6.792689451455253e-06, "loss": 0.17758466005325318, "memory(GiB)": 77.59, "step": 97115, "token_acc": 0.9531772575250836, "train_speed(iter/s)": 1.45987 }, { "epoch": 4.160918555331819, "grad_norm": 2.9715750217437744, "learning_rate": 6.78930314630854e-06, "loss": 0.4001638412475586, "memory(GiB)": 77.59, "step": 97120, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.459875 }, { "epoch": 4.161132770661069, "grad_norm": 4.357032299041748, "learning_rate": 6.785917623964133e-06, "loss": 0.24576287269592284, "memory(GiB)": 77.59, "step": 97125, "token_acc": 0.9375, "train_speed(iter/s)": 1.459878 }, { "epoch": 4.161346985990318, "grad_norm": 2.3069190979003906, "learning_rate": 6.782532884483367e-06, "loss": 0.7355796337127686, "memory(GiB)": 77.59, "step": 97130, "token_acc": 0.839344262295082, "train_speed(iter/s)": 1.459888 }, { "epoch": 4.161561201319566, "grad_norm": 1.8695300817489624, "learning_rate": 6.779148927927548e-06, "loss": 0.45167860984802244, "memory(GiB)": 77.59, "step": 97135, "token_acc": 0.899641577060932, "train_speed(iter/s)": 1.459894 }, { "epoch": 4.1617754166488155, "grad_norm": 3.4552624225616455, "learning_rate": 6.775765754358004e-06, "loss": 0.3216372489929199, "memory(GiB)": 77.59, "step": 97140, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.459897 }, { "epoch": 4.161989631978065, "grad_norm": 3.9364166259765625, "learning_rate": 6.77238336383601e-06, "loss": 0.3852393627166748, "memory(GiB)": 77.59, "step": 97145, "token_acc": 0.9137931034482759, "train_speed(iter/s)": 1.459908 }, { "epoch": 4.162203847307313, "grad_norm": 2.1573445796966553, "learning_rate": 6.769001756422838e-06, "loss": 0.34531173706054685, "memory(GiB)": 77.59, "step": 97150, "token_acc": 0.9432835820895522, "train_speed(iter/s)": 1.459911 }, { "epoch": 4.162418062636562, "grad_norm": 2.638077735900879, "learning_rate": 6.765620932179756e-06, "loss": 0.18468594551086426, "memory(GiB)": 77.59, "step": 97155, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.459913 }, { "epoch": 4.162632277965812, "grad_norm": 5.189519882202148, "learning_rate": 6.7622408911679976e-06, "loss": 0.16034319400787353, "memory(GiB)": 77.59, "step": 97160, "token_acc": 0.962457337883959, "train_speed(iter/s)": 1.459912 }, { "epoch": 4.16284649329506, "grad_norm": 4.947395324707031, "learning_rate": 6.7588616334488045e-06, "loss": 0.23345434665679932, "memory(GiB)": 77.59, "step": 97165, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.459919 }, { "epoch": 4.163060708624309, "grad_norm": 2.0649266242980957, "learning_rate": 6.7554831590834086e-06, "loss": 0.33362009525299074, "memory(GiB)": 77.59, "step": 97170, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.459922 }, { "epoch": 4.1632749239535585, "grad_norm": 3.7910022735595703, "learning_rate": 6.752105468133002e-06, "loss": 0.349067234992981, "memory(GiB)": 77.59, "step": 97175, "token_acc": 0.9223300970873787, "train_speed(iter/s)": 1.459925 }, { "epoch": 4.163489139282807, "grad_norm": 2.9535024166107178, "learning_rate": 6.748728560658774e-06, "loss": 0.29617819786071775, "memory(GiB)": 77.59, "step": 97180, "token_acc": 0.9292604501607717, "train_speed(iter/s)": 1.45993 }, { "epoch": 4.163703354612056, "grad_norm": 0.11881634593009949, "learning_rate": 6.745352436721902e-06, "loss": 0.17481690645217896, "memory(GiB)": 77.59, "step": 97185, "token_acc": 0.9616519174041298, "train_speed(iter/s)": 1.459936 }, { "epoch": 4.163917569941305, "grad_norm": 3.278655767440796, "learning_rate": 6.741977096383545e-06, "loss": 0.6864329814910889, "memory(GiB)": 77.59, "step": 97190, "token_acc": 0.8357348703170029, "train_speed(iter/s)": 1.459948 }, { "epoch": 4.164131785270554, "grad_norm": 1.47878897190094, "learning_rate": 6.738602539704842e-06, "loss": 0.3028780221939087, "memory(GiB)": 77.59, "step": 97195, "token_acc": 0.9244604316546763, "train_speed(iter/s)": 1.459951 }, { "epoch": 4.164346000599803, "grad_norm": 0.19924937188625336, "learning_rate": 6.735228766746948e-06, "loss": 0.2019261598587036, "memory(GiB)": 77.59, "step": 97200, "token_acc": 0.9578947368421052, "train_speed(iter/s)": 1.459962 }, { "epoch": 4.164560215929052, "grad_norm": 2.9159016609191895, "learning_rate": 6.731855777570972e-06, "loss": 0.2678964614868164, "memory(GiB)": 77.59, "step": 97205, "token_acc": 0.9372693726937269, "train_speed(iter/s)": 1.459966 }, { "epoch": 4.164774431258301, "grad_norm": 1.161870002746582, "learning_rate": 6.728483572238015e-06, "loss": 0.3604951620101929, "memory(GiB)": 77.59, "step": 97210, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.45997 }, { "epoch": 4.16498864658755, "grad_norm": 5.406956672668457, "learning_rate": 6.725112150809171e-06, "loss": 0.6941930294036865, "memory(GiB)": 77.59, "step": 97215, "token_acc": 0.8666666666666667, "train_speed(iter/s)": 1.459971 }, { "epoch": 4.165202861916799, "grad_norm": 3.223749876022339, "learning_rate": 6.721741513345503e-06, "loss": 0.47463397979736327, "memory(GiB)": 77.59, "step": 97220, "token_acc": 0.8940397350993378, "train_speed(iter/s)": 1.459975 }, { "epoch": 4.1654170772460475, "grad_norm": 3.3509268760681152, "learning_rate": 6.718371659908101e-06, "loss": 0.39056308269500734, "memory(GiB)": 77.59, "step": 97225, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.459975 }, { "epoch": 4.165631292575297, "grad_norm": 2.321324348449707, "learning_rate": 6.715002590557984e-06, "loss": 0.3346219539642334, "memory(GiB)": 77.59, "step": 97230, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.459977 }, { "epoch": 4.165845507904546, "grad_norm": 6.853032112121582, "learning_rate": 6.711634305356207e-06, "loss": 0.3427662134170532, "memory(GiB)": 77.59, "step": 97235, "token_acc": 0.9203187250996016, "train_speed(iter/s)": 1.459972 }, { "epoch": 4.166059723233794, "grad_norm": 5.876855373382568, "learning_rate": 6.708266804363789e-06, "loss": 0.5529674053192138, "memory(GiB)": 77.59, "step": 97240, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.459989 }, { "epoch": 4.166273938563044, "grad_norm": 7.173551559448242, "learning_rate": 6.704900087641725e-06, "loss": 0.44930109977722166, "memory(GiB)": 77.59, "step": 97245, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.45999 }, { "epoch": 4.166488153892293, "grad_norm": 4.944094181060791, "learning_rate": 6.70153415525101e-06, "loss": 0.4515075206756592, "memory(GiB)": 77.59, "step": 97250, "token_acc": 0.9148264984227129, "train_speed(iter/s)": 1.45999 }, { "epoch": 4.166702369221541, "grad_norm": 5.0121588706970215, "learning_rate": 6.6981690072526085e-06, "loss": 0.3951236724853516, "memory(GiB)": 77.59, "step": 97255, "token_acc": 0.9132231404958677, "train_speed(iter/s)": 1.459992 }, { "epoch": 4.1669165845507905, "grad_norm": 1.8242756128311157, "learning_rate": 6.694804643707509e-06, "loss": 0.25155665874481203, "memory(GiB)": 77.59, "step": 97260, "token_acc": 0.943217665615142, "train_speed(iter/s)": 1.460001 }, { "epoch": 4.16713079988004, "grad_norm": 0.5136867761611938, "learning_rate": 6.691441064676651e-06, "loss": 0.2824327707290649, "memory(GiB)": 77.59, "step": 97265, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.460002 }, { "epoch": 4.167345015209288, "grad_norm": 2.1567745208740234, "learning_rate": 6.68807827022096e-06, "loss": 0.2568269014358521, "memory(GiB)": 77.59, "step": 97270, "token_acc": 0.943089430894309, "train_speed(iter/s)": 1.460006 }, { "epoch": 4.167559230538537, "grad_norm": 6.3242082595825195, "learning_rate": 6.684716260401358e-06, "loss": 0.2938757658004761, "memory(GiB)": 77.59, "step": 97275, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.46001 }, { "epoch": 4.167773445867787, "grad_norm": 1.593278408050537, "learning_rate": 6.681355035278747e-06, "loss": 0.11464307308197022, "memory(GiB)": 77.59, "step": 97280, "token_acc": 0.9685314685314685, "train_speed(iter/s)": 1.460009 }, { "epoch": 4.167987661197035, "grad_norm": 3.4629294872283936, "learning_rate": 6.677994594914039e-06, "loss": 0.4117640495300293, "memory(GiB)": 77.59, "step": 97285, "token_acc": 0.9253731343283582, "train_speed(iter/s)": 1.460014 }, { "epoch": 4.168201876526284, "grad_norm": 3.178605556488037, "learning_rate": 6.674634939368096e-06, "loss": 0.34734961986541746, "memory(GiB)": 77.59, "step": 97290, "token_acc": 0.9087301587301587, "train_speed(iter/s)": 1.460033 }, { "epoch": 4.1684160918555335, "grad_norm": 1.6664519309997559, "learning_rate": 6.671276068701782e-06, "loss": 0.3989849328994751, "memory(GiB)": 77.59, "step": 97295, "token_acc": 0.9141914191419142, "train_speed(iter/s)": 1.460043 }, { "epoch": 4.168630307184782, "grad_norm": 4.829763412475586, "learning_rate": 6.667917982975947e-06, "loss": 0.5140761375427246, "memory(GiB)": 77.59, "step": 97300, "token_acc": 0.8950819672131147, "train_speed(iter/s)": 1.460054 }, { "epoch": 4.168844522514031, "grad_norm": 0.06836356222629547, "learning_rate": 6.66456068225142e-06, "loss": 0.2605807542800903, "memory(GiB)": 77.59, "step": 97305, "token_acc": 0.9372937293729373, "train_speed(iter/s)": 1.460058 }, { "epoch": 4.16905873784328, "grad_norm": 3.500898838043213, "learning_rate": 6.661204166589025e-06, "loss": 0.28860950469970703, "memory(GiB)": 77.59, "step": 97310, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.460068 }, { "epoch": 4.169272953172529, "grad_norm": 2.185178756713867, "learning_rate": 6.657848436049585e-06, "loss": 0.36973569393157957, "memory(GiB)": 77.59, "step": 97315, "token_acc": 0.9137254901960784, "train_speed(iter/s)": 1.460071 }, { "epoch": 4.169487168501778, "grad_norm": 3.1548943519592285, "learning_rate": 6.654493490693875e-06, "loss": 0.330818772315979, "memory(GiB)": 77.59, "step": 97320, "token_acc": 0.9375, "train_speed(iter/s)": 1.460075 }, { "epoch": 4.169701383831027, "grad_norm": 3.9836368560791016, "learning_rate": 6.651139330582679e-06, "loss": 0.4275322437286377, "memory(GiB)": 77.59, "step": 97325, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.460074 }, { "epoch": 4.169915599160276, "grad_norm": 0.9981362223625183, "learning_rate": 6.647785955776759e-06, "loss": 0.36172165870666506, "memory(GiB)": 77.59, "step": 97330, "token_acc": 0.9227941176470589, "train_speed(iter/s)": 1.460083 }, { "epoch": 4.170129814489525, "grad_norm": 1.3591649532318115, "learning_rate": 6.644433366336861e-06, "loss": 0.5058945178985595, "memory(GiB)": 77.59, "step": 97335, "token_acc": 0.8771929824561403, "train_speed(iter/s)": 1.460096 }, { "epoch": 4.170344029818774, "grad_norm": 5.092177867889404, "learning_rate": 6.6410815623237145e-06, "loss": 0.3518717050552368, "memory(GiB)": 77.59, "step": 97340, "token_acc": 0.9233449477351916, "train_speed(iter/s)": 1.460095 }, { "epoch": 4.1705582451480225, "grad_norm": 2.606449604034424, "learning_rate": 6.6377305437980596e-06, "loss": 0.22081334590911866, "memory(GiB)": 77.59, "step": 97345, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.460099 }, { "epoch": 4.170772460477272, "grad_norm": 11.376921653747559, "learning_rate": 6.634380310820598e-06, "loss": 0.44518508911132815, "memory(GiB)": 77.59, "step": 97350, "token_acc": 0.8881789137380192, "train_speed(iter/s)": 1.460117 }, { "epoch": 4.170986675806521, "grad_norm": 6.1446733474731445, "learning_rate": 6.6310308634520144e-06, "loss": 0.4732032775878906, "memory(GiB)": 77.59, "step": 97355, "token_acc": 0.9080882352941176, "train_speed(iter/s)": 1.460119 }, { "epoch": 4.171200891135769, "grad_norm": 2.9128036499023438, "learning_rate": 6.627682201752988e-06, "loss": 0.21711349487304688, "memory(GiB)": 77.59, "step": 97360, "token_acc": 0.95, "train_speed(iter/s)": 1.460131 }, { "epoch": 4.171415106465019, "grad_norm": 4.217676639556885, "learning_rate": 6.6243343257841765e-06, "loss": 0.30523488521575926, "memory(GiB)": 77.59, "step": 97365, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.46013 }, { "epoch": 4.171629321794268, "grad_norm": 1.6031062602996826, "learning_rate": 6.620987235606246e-06, "loss": 0.27183189392089846, "memory(GiB)": 77.59, "step": 97370, "token_acc": 0.9416342412451362, "train_speed(iter/s)": 1.460132 }, { "epoch": 4.171843537123516, "grad_norm": 3.2560617923736572, "learning_rate": 6.617640931279828e-06, "loss": 0.3387740612030029, "memory(GiB)": 77.59, "step": 97375, "token_acc": 0.9181818181818182, "train_speed(iter/s)": 1.460135 }, { "epoch": 4.1720577524527656, "grad_norm": 1.8174837827682495, "learning_rate": 6.614295412865523e-06, "loss": 0.25264854431152345, "memory(GiB)": 77.59, "step": 97380, "token_acc": 0.9409937888198758, "train_speed(iter/s)": 1.460137 }, { "epoch": 4.172271967782015, "grad_norm": 4.771573066711426, "learning_rate": 6.610950680423972e-06, "loss": 0.5029104709625244, "memory(GiB)": 77.59, "step": 97385, "token_acc": 0.9009009009009009, "train_speed(iter/s)": 1.460154 }, { "epoch": 4.172486183111263, "grad_norm": 4.443476676940918, "learning_rate": 6.607606734015753e-06, "loss": 0.252191948890686, "memory(GiB)": 77.59, "step": 97390, "token_acc": 0.953125, "train_speed(iter/s)": 1.460155 }, { "epoch": 4.172700398440512, "grad_norm": 3.202730894088745, "learning_rate": 6.604263573701441e-06, "loss": 0.4086909770965576, "memory(GiB)": 77.59, "step": 97395, "token_acc": 0.8952380952380953, "train_speed(iter/s)": 1.460172 }, { "epoch": 4.172914613769762, "grad_norm": 3.5525829792022705, "learning_rate": 6.600921199541593e-06, "loss": 0.36516823768615725, "memory(GiB)": 77.59, "step": 97400, "token_acc": 0.9171779141104295, "train_speed(iter/s)": 1.460188 }, { "epoch": 4.17312882909901, "grad_norm": 0.09657716006040573, "learning_rate": 6.597579611596782e-06, "loss": 0.21336562633514405, "memory(GiB)": 77.59, "step": 97405, "token_acc": 0.9625, "train_speed(iter/s)": 1.460192 }, { "epoch": 4.173343044428259, "grad_norm": 1.358163833618164, "learning_rate": 6.594238809927527e-06, "loss": 0.3435099124908447, "memory(GiB)": 77.59, "step": 97410, "token_acc": 0.9305993690851735, "train_speed(iter/s)": 1.460193 }, { "epoch": 4.173557259757509, "grad_norm": 0.5975470542907715, "learning_rate": 6.590898794594358e-06, "loss": 0.21276743412017823, "memory(GiB)": 77.59, "step": 97415, "token_acc": 0.9583333333333334, "train_speed(iter/s)": 1.460195 }, { "epoch": 4.173771475086757, "grad_norm": 2.3451809883117676, "learning_rate": 6.587559565657775e-06, "loss": 0.36565585136413575, "memory(GiB)": 77.59, "step": 97420, "token_acc": 0.9080459770114943, "train_speed(iter/s)": 1.460207 }, { "epoch": 4.173985690416006, "grad_norm": 0.16196109354496002, "learning_rate": 6.584221123178263e-06, "loss": 0.17001349925994874, "memory(GiB)": 77.59, "step": 97425, "token_acc": 0.9535603715170279, "train_speed(iter/s)": 1.460208 }, { "epoch": 4.1741999057452555, "grad_norm": 3.2360713481903076, "learning_rate": 6.580883467216326e-06, "loss": 0.35758304595947266, "memory(GiB)": 77.59, "step": 97430, "token_acc": 0.920863309352518, "train_speed(iter/s)": 1.460209 }, { "epoch": 4.174414121074504, "grad_norm": 5.221468925476074, "learning_rate": 6.577546597832413e-06, "loss": 0.5322786331176758, "memory(GiB)": 77.59, "step": 97435, "token_acc": 0.8719512195121951, "train_speed(iter/s)": 1.460204 }, { "epoch": 4.174628336403753, "grad_norm": 1.437172532081604, "learning_rate": 6.574210515086982e-06, "loss": 0.16237348318099976, "memory(GiB)": 77.59, "step": 97440, "token_acc": 0.9669669669669669, "train_speed(iter/s)": 1.460205 }, { "epoch": 4.174842551733002, "grad_norm": 4.122325897216797, "learning_rate": 6.570875219040457e-06, "loss": 0.5404956817626954, "memory(GiB)": 77.59, "step": 97445, "token_acc": 0.8949044585987261, "train_speed(iter/s)": 1.460211 }, { "epoch": 4.175056767062251, "grad_norm": 1.7078391313552856, "learning_rate": 6.567540709753262e-06, "loss": 0.283498477935791, "memory(GiB)": 77.59, "step": 97450, "token_acc": 0.9226190476190477, "train_speed(iter/s)": 1.460205 }, { "epoch": 4.1752709823915, "grad_norm": 0.9430018067359924, "learning_rate": 6.564206987285809e-06, "loss": 0.2563011884689331, "memory(GiB)": 77.59, "step": 97455, "token_acc": 0.9473684210526315, "train_speed(iter/s)": 1.460215 }, { "epoch": 4.175485197720749, "grad_norm": 4.2265801429748535, "learning_rate": 6.560874051698502e-06, "loss": 0.5684188842773438, "memory(GiB)": 77.59, "step": 97460, "token_acc": 0.884272997032641, "train_speed(iter/s)": 1.460225 }, { "epoch": 4.175699413049998, "grad_norm": 2.1178979873657227, "learning_rate": 6.557541903051712e-06, "loss": 0.4700815200805664, "memory(GiB)": 77.59, "step": 97465, "token_acc": 0.8856088560885609, "train_speed(iter/s)": 1.460226 }, { "epoch": 4.175913628379247, "grad_norm": 0.6396270394325256, "learning_rate": 6.554210541405803e-06, "loss": 0.3201985597610474, "memory(GiB)": 77.59, "step": 97470, "token_acc": 0.9215686274509803, "train_speed(iter/s)": 1.460228 }, { "epoch": 4.176127843708496, "grad_norm": 3.5686988830566406, "learning_rate": 6.550879966821122e-06, "loss": 0.5309565544128418, "memory(GiB)": 77.59, "step": 97475, "token_acc": 0.9063545150501672, "train_speed(iter/s)": 1.46023 }, { "epoch": 4.1763420590377445, "grad_norm": 2.7273800373077393, "learning_rate": 6.547550179358014e-06, "loss": 0.3641794681549072, "memory(GiB)": 77.59, "step": 97480, "token_acc": 0.8997289972899729, "train_speed(iter/s)": 1.460238 }, { "epoch": 4.176556274366994, "grad_norm": 1.6447190046310425, "learning_rate": 6.544221179076782e-06, "loss": 0.3765552282333374, "memory(GiB)": 77.59, "step": 97485, "token_acc": 0.9088235294117647, "train_speed(iter/s)": 1.460242 }, { "epoch": 4.176770489696243, "grad_norm": 3.699047088623047, "learning_rate": 6.5408929660377595e-06, "loss": 0.2707526206970215, "memory(GiB)": 77.59, "step": 97490, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.460251 }, { "epoch": 4.176984705025491, "grad_norm": 2.7553257942199707, "learning_rate": 6.537565540301227e-06, "loss": 0.3330109119415283, "memory(GiB)": 77.59, "step": 97495, "token_acc": 0.9398496240601504, "train_speed(iter/s)": 1.460256 }, { "epoch": 4.177198920354741, "grad_norm": 3.6314079761505127, "learning_rate": 6.534238901927469e-06, "loss": 0.5373851299285889, "memory(GiB)": 77.59, "step": 97500, "token_acc": 0.8908045977011494, "train_speed(iter/s)": 1.460258 }, { "epoch": 4.177198920354741, "eval_loss": 2.583786964416504, "eval_runtime": 11.2419, "eval_samples_per_second": 8.895, "eval_steps_per_second": 8.895, "eval_token_acc": 0.4436416184971098, "step": 97500 }, { "epoch": 4.17741313568399, "grad_norm": 4.204164981842041, "learning_rate": 6.530913050976744e-06, "loss": 0.658365535736084, "memory(GiB)": 77.59, "step": 97505, "token_acc": 0.5731462925851704, "train_speed(iter/s)": 1.46 }, { "epoch": 4.177627351013238, "grad_norm": 4.643906593322754, "learning_rate": 6.527587987509299e-06, "loss": 0.45300636291503904, "memory(GiB)": 77.59, "step": 97510, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.46 }, { "epoch": 4.1778415663424875, "grad_norm": 2.5250635147094727, "learning_rate": 6.5242637115853876e-06, "loss": 0.18197418451309205, "memory(GiB)": 77.59, "step": 97515, "token_acc": 0.9629629629629629, "train_speed(iter/s)": 1.460003 }, { "epoch": 4.178055781671737, "grad_norm": 2.913274049758911, "learning_rate": 6.520940223265226e-06, "loss": 0.28318376541137696, "memory(GiB)": 77.59, "step": 97520, "token_acc": 0.9425287356321839, "train_speed(iter/s)": 1.460002 }, { "epoch": 4.178269997000985, "grad_norm": 2.3135437965393066, "learning_rate": 6.517617522609015e-06, "loss": 0.47270889282226564, "memory(GiB)": 77.59, "step": 97525, "token_acc": 0.9, "train_speed(iter/s)": 1.460003 }, { "epoch": 4.178484212330234, "grad_norm": 5.325996398925781, "learning_rate": 6.51429560967694e-06, "loss": 0.47706069946289065, "memory(GiB)": 77.59, "step": 97530, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.460008 }, { "epoch": 4.178698427659484, "grad_norm": 7.4018425941467285, "learning_rate": 6.510974484529209e-06, "loss": 0.32614195346832275, "memory(GiB)": 77.59, "step": 97535, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.460013 }, { "epoch": 4.178912642988732, "grad_norm": 4.2571187019348145, "learning_rate": 6.5076541472259666e-06, "loss": 0.3398735046386719, "memory(GiB)": 77.59, "step": 97540, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.460024 }, { "epoch": 4.179126858317981, "grad_norm": 1.1020060777664185, "learning_rate": 6.50433459782736e-06, "loss": 0.5331015586853027, "memory(GiB)": 77.59, "step": 97545, "token_acc": 0.8940397350993378, "train_speed(iter/s)": 1.460029 }, { "epoch": 4.1793410736472305, "grad_norm": 0.1424068659543991, "learning_rate": 6.501015836393543e-06, "loss": 0.3665480136871338, "memory(GiB)": 77.59, "step": 97550, "token_acc": 0.9067164179104478, "train_speed(iter/s)": 1.460031 }, { "epoch": 4.179555288976479, "grad_norm": 5.178808212280273, "learning_rate": 6.497697862984631e-06, "loss": 0.41159806251525877, "memory(GiB)": 77.59, "step": 97555, "token_acc": 0.9004329004329005, "train_speed(iter/s)": 1.460038 }, { "epoch": 4.179769504305728, "grad_norm": 2.1409592628479004, "learning_rate": 6.494380677660733e-06, "loss": 0.40760278701782227, "memory(GiB)": 77.59, "step": 97560, "token_acc": 0.9050632911392406, "train_speed(iter/s)": 1.460048 }, { "epoch": 4.179983719634977, "grad_norm": 3.835376501083374, "learning_rate": 6.491064280481934e-06, "loss": 0.4240875720977783, "memory(GiB)": 77.59, "step": 97565, "token_acc": 0.908284023668639, "train_speed(iter/s)": 1.460046 }, { "epoch": 4.180197934964226, "grad_norm": 2.033799648284912, "learning_rate": 6.4877486715083144e-06, "loss": 0.12782635688781738, "memory(GiB)": 77.59, "step": 97570, "token_acc": 0.967741935483871, "train_speed(iter/s)": 1.460045 }, { "epoch": 4.180412150293475, "grad_norm": 2.1786141395568848, "learning_rate": 6.484433850799959e-06, "loss": 0.3486099481582642, "memory(GiB)": 77.59, "step": 97575, "token_acc": 0.9081272084805654, "train_speed(iter/s)": 1.460053 }, { "epoch": 4.180626365622724, "grad_norm": 3.089102029800415, "learning_rate": 6.481119818416898e-06, "loss": 0.08281692266464233, "memory(GiB)": 77.59, "step": 97580, "token_acc": 0.9795918367346939, "train_speed(iter/s)": 1.46006 }, { "epoch": 4.180840580951973, "grad_norm": 5.708587646484375, "learning_rate": 6.477806574419182e-06, "loss": 0.48087358474731445, "memory(GiB)": 77.59, "step": 97585, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.460063 }, { "epoch": 4.181054796281222, "grad_norm": 2.8505542278289795, "learning_rate": 6.474494118866825e-06, "loss": 0.2611204147338867, "memory(GiB)": 77.59, "step": 97590, "token_acc": 0.9426229508196722, "train_speed(iter/s)": 1.460058 }, { "epoch": 4.181269011610471, "grad_norm": 0.487552672624588, "learning_rate": 6.471182451819824e-06, "loss": 0.11936368942260742, "memory(GiB)": 77.59, "step": 97595, "token_acc": 0.9618320610687023, "train_speed(iter/s)": 1.460061 }, { "epoch": 4.1814832269397195, "grad_norm": 3.7532525062561035, "learning_rate": 6.467871573338186e-06, "loss": 0.1650167226791382, "memory(GiB)": 77.59, "step": 97600, "token_acc": 0.953405017921147, "train_speed(iter/s)": 1.460064 }, { "epoch": 4.181697442268969, "grad_norm": 3.358043670654297, "learning_rate": 6.464561483481907e-06, "loss": 0.5403341293334961, "memory(GiB)": 77.59, "step": 97605, "token_acc": 0.8910081743869209, "train_speed(iter/s)": 1.460078 }, { "epoch": 4.181911657598218, "grad_norm": 3.1714885234832764, "learning_rate": 6.461252182310929e-06, "loss": 0.3714768886566162, "memory(GiB)": 77.59, "step": 97610, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.460083 }, { "epoch": 4.182125872927466, "grad_norm": 2.693624258041382, "learning_rate": 6.457943669885219e-06, "loss": 0.4526193141937256, "memory(GiB)": 77.59, "step": 97615, "token_acc": 0.8832116788321168, "train_speed(iter/s)": 1.460082 }, { "epoch": 4.182340088256716, "grad_norm": 4.4694671630859375, "learning_rate": 6.4546359462646975e-06, "loss": 0.34573955535888673, "memory(GiB)": 77.59, "step": 97620, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.460088 }, { "epoch": 4.182554303585965, "grad_norm": 3.577923536300659, "learning_rate": 6.4513290115093e-06, "loss": 0.2870840311050415, "memory(GiB)": 77.59, "step": 97625, "token_acc": 0.9251968503937008, "train_speed(iter/s)": 1.46009 }, { "epoch": 4.182768518915213, "grad_norm": 8.053380012512207, "learning_rate": 6.448022865678916e-06, "loss": 0.37318460941314696, "memory(GiB)": 77.59, "step": 97630, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.460098 }, { "epoch": 4.1829827342444625, "grad_norm": 4.7150373458862305, "learning_rate": 6.444717508833464e-06, "loss": 0.4212969779968262, "memory(GiB)": 77.59, "step": 97635, "token_acc": 0.9007633587786259, "train_speed(iter/s)": 1.460096 }, { "epoch": 4.183196949573712, "grad_norm": 4.007823467254639, "learning_rate": 6.441412941032809e-06, "loss": 0.4210054874420166, "memory(GiB)": 77.59, "step": 97640, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.460096 }, { "epoch": 4.18341116490296, "grad_norm": 4.1203107833862305, "learning_rate": 6.4381091623368254e-06, "loss": 0.3103918552398682, "memory(GiB)": 77.59, "step": 97645, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.460098 }, { "epoch": 4.183625380232209, "grad_norm": 1.2829504013061523, "learning_rate": 6.434806172805358e-06, "loss": 0.2604515075683594, "memory(GiB)": 77.59, "step": 97650, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.460102 }, { "epoch": 4.183839595561459, "grad_norm": 3.282407522201538, "learning_rate": 6.431503972498232e-06, "loss": 0.4477139949798584, "memory(GiB)": 77.59, "step": 97655, "token_acc": 0.8996539792387543, "train_speed(iter/s)": 1.460115 }, { "epoch": 4.184053810890707, "grad_norm": 4.048049449920654, "learning_rate": 6.428202561475288e-06, "loss": 0.5166503429412842, "memory(GiB)": 77.59, "step": 97660, "token_acc": 0.9212598425196851, "train_speed(iter/s)": 1.46012 }, { "epoch": 4.184268026219956, "grad_norm": 4.927420139312744, "learning_rate": 6.424901939796335e-06, "loss": 0.42885580062866213, "memory(GiB)": 77.59, "step": 97665, "token_acc": 0.9005847953216374, "train_speed(iter/s)": 1.460122 }, { "epoch": 4.184482241549206, "grad_norm": 0.4785492718219757, "learning_rate": 6.421602107521157e-06, "loss": 0.23147976398468018, "memory(GiB)": 77.59, "step": 97670, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.460128 }, { "epoch": 4.184696456878454, "grad_norm": 2.466228723526001, "learning_rate": 6.41830306470953e-06, "loss": 0.171213698387146, "memory(GiB)": 77.59, "step": 97675, "token_acc": 0.950530035335689, "train_speed(iter/s)": 1.460146 }, { "epoch": 4.184910672207703, "grad_norm": 2.548906087875366, "learning_rate": 6.415004811421232e-06, "loss": 0.19051543474197388, "memory(GiB)": 77.59, "step": 97680, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.46015 }, { "epoch": 4.185124887536952, "grad_norm": 3.1325454711914062, "learning_rate": 6.411707347716012e-06, "loss": 0.22157719135284423, "memory(GiB)": 77.59, "step": 97685, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.460154 }, { "epoch": 4.185339102866201, "grad_norm": 4.170884609222412, "learning_rate": 6.408410673653586e-06, "loss": 0.4933638572692871, "memory(GiB)": 77.59, "step": 97690, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.46015 }, { "epoch": 4.18555331819545, "grad_norm": 3.838067054748535, "learning_rate": 6.405114789293709e-06, "loss": 0.37092061042785646, "memory(GiB)": 77.59, "step": 97695, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.460153 }, { "epoch": 4.185767533524699, "grad_norm": 1.9812541007995605, "learning_rate": 6.40181969469607e-06, "loss": 0.3119025707244873, "memory(GiB)": 77.59, "step": 97700, "token_acc": 0.9225589225589226, "train_speed(iter/s)": 1.460152 }, { "epoch": 4.185981748853948, "grad_norm": 4.311748504638672, "learning_rate": 6.398525389920368e-06, "loss": 0.2817831516265869, "memory(GiB)": 77.59, "step": 97705, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.460161 }, { "epoch": 4.186195964183197, "grad_norm": 4.39789342880249, "learning_rate": 6.395231875026275e-06, "loss": 0.362914514541626, "memory(GiB)": 77.59, "step": 97710, "token_acc": 0.9288135593220339, "train_speed(iter/s)": 1.460168 }, { "epoch": 4.186410179512446, "grad_norm": 5.5988569259643555, "learning_rate": 6.391939150073456e-06, "loss": 0.220673131942749, "memory(GiB)": 77.59, "step": 97715, "token_acc": 0.9495798319327731, "train_speed(iter/s)": 1.460165 }, { "epoch": 4.186624394841695, "grad_norm": 6.242227554321289, "learning_rate": 6.388647215121579e-06, "loss": 0.2356261968612671, "memory(GiB)": 77.59, "step": 97720, "token_acc": 0.9421487603305785, "train_speed(iter/s)": 1.460166 }, { "epoch": 4.186838610170944, "grad_norm": 2.904934883117676, "learning_rate": 6.385356070230264e-06, "loss": 0.21695225238800048, "memory(GiB)": 77.59, "step": 97725, "token_acc": 0.9442379182156134, "train_speed(iter/s)": 1.460166 }, { "epoch": 4.187052825500193, "grad_norm": 3.7167327404022217, "learning_rate": 6.38206571545914e-06, "loss": 0.3375232696533203, "memory(GiB)": 77.59, "step": 97730, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.460173 }, { "epoch": 4.1872670408294415, "grad_norm": 1.3004860877990723, "learning_rate": 6.37877615086781e-06, "loss": 0.09643903374671936, "memory(GiB)": 77.59, "step": 97735, "token_acc": 0.9698996655518395, "train_speed(iter/s)": 1.460175 }, { "epoch": 4.187481256158691, "grad_norm": 5.951595306396484, "learning_rate": 6.375487376515859e-06, "loss": 0.36009559631347654, "memory(GiB)": 77.59, "step": 97740, "token_acc": 0.933852140077821, "train_speed(iter/s)": 1.460179 }, { "epoch": 4.18769547148794, "grad_norm": 1.9411615133285522, "learning_rate": 6.372199392462891e-06, "loss": 0.37661037445068357, "memory(GiB)": 77.59, "step": 97745, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.460187 }, { "epoch": 4.187909686817188, "grad_norm": 2.522258758544922, "learning_rate": 6.368912198768445e-06, "loss": 0.19776941537857057, "memory(GiB)": 77.59, "step": 97750, "token_acc": 0.9440993788819876, "train_speed(iter/s)": 1.46019 }, { "epoch": 4.188123902146438, "grad_norm": 2.1099977493286133, "learning_rate": 6.365625795492092e-06, "loss": 0.48793964385986327, "memory(GiB)": 77.59, "step": 97755, "token_acc": 0.8844444444444445, "train_speed(iter/s)": 1.460203 }, { "epoch": 4.188338117475687, "grad_norm": 3.2931838035583496, "learning_rate": 6.362340182693366e-06, "loss": 0.5163224220275879, "memory(GiB)": 77.59, "step": 97760, "token_acc": 0.8919753086419753, "train_speed(iter/s)": 1.460204 }, { "epoch": 4.188552332804935, "grad_norm": 2.1474239826202393, "learning_rate": 6.3590553604317756e-06, "loss": 0.3873679876327515, "memory(GiB)": 77.59, "step": 97765, "token_acc": 0.9131832797427653, "train_speed(iter/s)": 1.460202 }, { "epoch": 4.1887665481341845, "grad_norm": 5.39483642578125, "learning_rate": 6.3557713287668405e-06, "loss": 0.3716289520263672, "memory(GiB)": 77.59, "step": 97770, "token_acc": 0.9305555555555556, "train_speed(iter/s)": 1.460207 }, { "epoch": 4.188980763463434, "grad_norm": 2.084623336791992, "learning_rate": 6.3524880877580405e-06, "loss": 0.23407597541809083, "memory(GiB)": 77.59, "step": 97775, "token_acc": 0.9620689655172414, "train_speed(iter/s)": 1.460207 }, { "epoch": 4.189194978792682, "grad_norm": 0.9902254939079285, "learning_rate": 6.349205637464872e-06, "loss": 0.24070391654968262, "memory(GiB)": 77.59, "step": 97780, "token_acc": 0.9307958477508651, "train_speed(iter/s)": 1.460214 }, { "epoch": 4.189409194121931, "grad_norm": 3.182434320449829, "learning_rate": 6.345923977946789e-06, "loss": 0.26826000213623047, "memory(GiB)": 77.59, "step": 97785, "token_acc": 0.9529411764705882, "train_speed(iter/s)": 1.460214 }, { "epoch": 4.189623409451181, "grad_norm": 3.3424391746520996, "learning_rate": 6.3426431092632465e-06, "loss": 0.23312735557556152, "memory(GiB)": 77.59, "step": 97790, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.460213 }, { "epoch": 4.189837624780429, "grad_norm": 3.9559199810028076, "learning_rate": 6.339363031473677e-06, "loss": 0.1419013261795044, "memory(GiB)": 77.59, "step": 97795, "token_acc": 0.967391304347826, "train_speed(iter/s)": 1.46021 }, { "epoch": 4.190051840109678, "grad_norm": 3.2174971103668213, "learning_rate": 6.3360837446374945e-06, "loss": 0.35105156898498535, "memory(GiB)": 77.59, "step": 97800, "token_acc": 0.9312977099236641, "train_speed(iter/s)": 1.460211 }, { "epoch": 4.1902660554389275, "grad_norm": 4.2937092781066895, "learning_rate": 6.3328052488141225e-06, "loss": 0.6402896881103516, "memory(GiB)": 77.59, "step": 97805, "token_acc": 0.8796296296296297, "train_speed(iter/s)": 1.46022 }, { "epoch": 4.190480270768176, "grad_norm": 4.231352806091309, "learning_rate": 6.329527544062952e-06, "loss": 0.6518398284912109, "memory(GiB)": 77.59, "step": 97810, "token_acc": 0.8704318936877077, "train_speed(iter/s)": 1.460228 }, { "epoch": 4.190694486097425, "grad_norm": 2.599597930908203, "learning_rate": 6.326250630443348e-06, "loss": 0.3621212005615234, "memory(GiB)": 77.59, "step": 97815, "token_acc": 0.9242424242424242, "train_speed(iter/s)": 1.46023 }, { "epoch": 4.190908701426674, "grad_norm": 3.691052198410034, "learning_rate": 6.32297450801469e-06, "loss": 0.40732507705688475, "memory(GiB)": 77.59, "step": 97820, "token_acc": 0.9032258064516129, "train_speed(iter/s)": 1.460234 }, { "epoch": 4.191122916755923, "grad_norm": 3.036773920059204, "learning_rate": 6.319699176836308e-06, "loss": 0.18678412437438965, "memory(GiB)": 77.59, "step": 97825, "token_acc": 0.9531772575250836, "train_speed(iter/s)": 1.460247 }, { "epoch": 4.191337132085172, "grad_norm": 0.6137109398841858, "learning_rate": 6.316424636967561e-06, "loss": 0.23461430072784423, "memory(GiB)": 77.59, "step": 97830, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.460247 }, { "epoch": 4.191551347414421, "grad_norm": 2.381547212600708, "learning_rate": 6.313150888467751e-06, "loss": 0.2145379066467285, "memory(GiB)": 77.59, "step": 97835, "token_acc": 0.9502369668246445, "train_speed(iter/s)": 1.460247 }, { "epoch": 4.19176556274367, "grad_norm": 3.448763132095337, "learning_rate": 6.309877931396202e-06, "loss": 0.2681333065032959, "memory(GiB)": 77.59, "step": 97840, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.460242 }, { "epoch": 4.191979778072919, "grad_norm": 2.8452088832855225, "learning_rate": 6.306605765812202e-06, "loss": 0.33008403778076173, "memory(GiB)": 77.59, "step": 97845, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.460245 }, { "epoch": 4.192193993402168, "grad_norm": 3.194904088973999, "learning_rate": 6.3033343917750235e-06, "loss": 0.2719533443450928, "memory(GiB)": 77.59, "step": 97850, "token_acc": 0.935251798561151, "train_speed(iter/s)": 1.460248 }, { "epoch": 4.1924082087314165, "grad_norm": 3.2260468006134033, "learning_rate": 6.300063809343936e-06, "loss": 0.3483938455581665, "memory(GiB)": 77.59, "step": 97855, "token_acc": 0.9461538461538461, "train_speed(iter/s)": 1.460245 }, { "epoch": 4.192622424060666, "grad_norm": 2.8232271671295166, "learning_rate": 6.2967940185781785e-06, "loss": 0.3098812818527222, "memory(GiB)": 77.59, "step": 97860, "token_acc": 0.9225352112676056, "train_speed(iter/s)": 1.460259 }, { "epoch": 4.192836639389915, "grad_norm": 6.607365131378174, "learning_rate": 6.293525019537e-06, "loss": 0.4239799976348877, "memory(GiB)": 77.59, "step": 97865, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.460259 }, { "epoch": 4.193050854719163, "grad_norm": 1.1906187534332275, "learning_rate": 6.290256812279616e-06, "loss": 0.4301448822021484, "memory(GiB)": 77.59, "step": 97870, "token_acc": 0.9003690036900369, "train_speed(iter/s)": 1.460262 }, { "epoch": 4.193265070048413, "grad_norm": 2.8714754581451416, "learning_rate": 6.286989396865234e-06, "loss": 0.472912073135376, "memory(GiB)": 77.59, "step": 97875, "token_acc": 0.8791946308724832, "train_speed(iter/s)": 1.460272 }, { "epoch": 4.193479285377662, "grad_norm": 1.2754883766174316, "learning_rate": 6.283722773353046e-06, "loss": 0.10838866233825684, "memory(GiB)": 77.59, "step": 97880, "token_acc": 0.9668874172185431, "train_speed(iter/s)": 1.460277 }, { "epoch": 4.19369350070691, "grad_norm": 5.246139049530029, "learning_rate": 6.280456941802215e-06, "loss": 0.5106232166290283, "memory(GiB)": 77.59, "step": 97885, "token_acc": 0.9171428571428571, "train_speed(iter/s)": 1.460289 }, { "epoch": 4.1939077160361595, "grad_norm": 4.70322322845459, "learning_rate": 6.277191902271934e-06, "loss": 0.40793418884277344, "memory(GiB)": 77.59, "step": 97890, "token_acc": 0.922509225092251, "train_speed(iter/s)": 1.460295 }, { "epoch": 4.194121931365409, "grad_norm": 3.8737030029296875, "learning_rate": 6.273927654821321e-06, "loss": 0.21413168907165528, "memory(GiB)": 77.59, "step": 97895, "token_acc": 0.9497907949790795, "train_speed(iter/s)": 1.460304 }, { "epoch": 4.194336146694657, "grad_norm": 4.02509880065918, "learning_rate": 6.2706641995095405e-06, "loss": 0.1903510570526123, "memory(GiB)": 77.59, "step": 97900, "token_acc": 0.9612068965517241, "train_speed(iter/s)": 1.460311 }, { "epoch": 4.194550362023906, "grad_norm": 4.010931968688965, "learning_rate": 6.267401536395701e-06, "loss": 0.3762238025665283, "memory(GiB)": 77.59, "step": 97905, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.460314 }, { "epoch": 4.194764577353156, "grad_norm": 4.0380377769470215, "learning_rate": 6.2641396655389095e-06, "loss": 0.3393506288528442, "memory(GiB)": 77.59, "step": 97910, "token_acc": 0.9235127478753541, "train_speed(iter/s)": 1.460324 }, { "epoch": 4.194978792682404, "grad_norm": 2.6324524879455566, "learning_rate": 6.2608785869982475e-06, "loss": 0.3885503768920898, "memory(GiB)": 77.59, "step": 97915, "token_acc": 0.9003831417624522, "train_speed(iter/s)": 1.460328 }, { "epoch": 4.195193008011653, "grad_norm": 0.46595272421836853, "learning_rate": 6.257618300832796e-06, "loss": 0.09880998134613037, "memory(GiB)": 77.59, "step": 97920, "token_acc": 0.9691780821917808, "train_speed(iter/s)": 1.460332 }, { "epoch": 4.1954072233409025, "grad_norm": 2.5683486461639404, "learning_rate": 6.254358807101635e-06, "loss": 0.15531188249588013, "memory(GiB)": 77.59, "step": 97925, "token_acc": 0.9704797047970479, "train_speed(iter/s)": 1.460333 }, { "epoch": 4.195621438670151, "grad_norm": 2.6194188594818115, "learning_rate": 6.251100105863794e-06, "loss": 0.3720966100692749, "memory(GiB)": 77.59, "step": 97930, "token_acc": 0.9154518950437318, "train_speed(iter/s)": 1.460336 }, { "epoch": 4.1958356539994, "grad_norm": 3.7578253746032715, "learning_rate": 6.247842197178317e-06, "loss": 0.2867238759994507, "memory(GiB)": 77.59, "step": 97935, "token_acc": 0.9283276450511946, "train_speed(iter/s)": 1.460336 }, { "epoch": 4.196049869328649, "grad_norm": 3.7614517211914062, "learning_rate": 6.2445850811042264e-06, "loss": 0.5793591976165772, "memory(GiB)": 77.59, "step": 97940, "token_acc": 0.8754716981132076, "train_speed(iter/s)": 1.460341 }, { "epoch": 4.196264084657898, "grad_norm": 2.745328903198242, "learning_rate": 6.241328757700505e-06, "loss": 0.22961721420288086, "memory(GiB)": 77.59, "step": 97945, "token_acc": 0.9451612903225807, "train_speed(iter/s)": 1.460345 }, { "epoch": 4.196478299987147, "grad_norm": 2.240875244140625, "learning_rate": 6.238073227026176e-06, "loss": 0.16062201261520387, "memory(GiB)": 77.59, "step": 97950, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.460342 }, { "epoch": 4.196692515316396, "grad_norm": 2.522005796432495, "learning_rate": 6.234818489140204e-06, "loss": 0.22144055366516113, "memory(GiB)": 77.59, "step": 97955, "token_acc": 0.947565543071161, "train_speed(iter/s)": 1.460348 }, { "epoch": 4.196906730645645, "grad_norm": 5.536969184875488, "learning_rate": 6.231564544101548e-06, "loss": 0.2158492088317871, "memory(GiB)": 77.59, "step": 97960, "token_acc": 0.954954954954955, "train_speed(iter/s)": 1.460349 }, { "epoch": 4.197120945974894, "grad_norm": 1.3729010820388794, "learning_rate": 6.22831139196916e-06, "loss": 0.3408372163772583, "memory(GiB)": 77.59, "step": 97965, "token_acc": 0.9196428571428571, "train_speed(iter/s)": 1.460353 }, { "epoch": 4.197335161304143, "grad_norm": 3.5917844772338867, "learning_rate": 6.225059032801961e-06, "loss": 0.2202216386795044, "memory(GiB)": 77.59, "step": 97970, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.460357 }, { "epoch": 4.197549376633392, "grad_norm": 5.986090660095215, "learning_rate": 6.221807466658891e-06, "loss": 0.37030792236328125, "memory(GiB)": 77.59, "step": 97975, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.460366 }, { "epoch": 4.197763591962641, "grad_norm": 3.23225736618042, "learning_rate": 6.2185566935988375e-06, "loss": 0.6040616035461426, "memory(GiB)": 77.59, "step": 97980, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.460379 }, { "epoch": 4.19797780729189, "grad_norm": 1.4490723609924316, "learning_rate": 6.2153067136807076e-06, "loss": 0.28947622776031495, "memory(GiB)": 77.59, "step": 97985, "token_acc": 0.926923076923077, "train_speed(iter/s)": 1.460393 }, { "epoch": 4.198192022621138, "grad_norm": 4.648755073547363, "learning_rate": 6.212057526963372e-06, "loss": 0.338886022567749, "memory(GiB)": 77.59, "step": 97990, "token_acc": 0.9320987654320988, "train_speed(iter/s)": 1.460393 }, { "epoch": 4.198406237950388, "grad_norm": 3.765591859817505, "learning_rate": 6.2088091335056945e-06, "loss": 0.4257965087890625, "memory(GiB)": 77.59, "step": 97995, "token_acc": 0.9022346368715084, "train_speed(iter/s)": 1.460394 }, { "epoch": 4.198620453279637, "grad_norm": 3.0799202919006348, "learning_rate": 6.205561533366511e-06, "loss": 0.2652931213378906, "memory(GiB)": 77.59, "step": 98000, "token_acc": 0.9418960244648318, "train_speed(iter/s)": 1.460399 }, { "epoch": 4.198620453279637, "eval_loss": 2.3386666774749756, "eval_runtime": 11.0206, "eval_samples_per_second": 9.074, "eval_steps_per_second": 9.074, "eval_token_acc": 0.48200312989045385, "step": 98000 }, { "epoch": 4.198834668608885, "grad_norm": 1.8059052228927612, "learning_rate": 6.202314726604658e-06, "loss": 0.32872135639190675, "memory(GiB)": 77.59, "step": 98005, "token_acc": 0.6314699792960663, "train_speed(iter/s)": 1.46014 }, { "epoch": 4.199048883938135, "grad_norm": 3.9678468704223633, "learning_rate": 6.199068713278966e-06, "loss": 0.3490708112716675, "memory(GiB)": 77.59, "step": 98010, "token_acc": 0.9330855018587361, "train_speed(iter/s)": 1.460149 }, { "epoch": 4.199263099267384, "grad_norm": 2.0733749866485596, "learning_rate": 6.1958234934482384e-06, "loss": 0.3047593832015991, "memory(GiB)": 77.59, "step": 98015, "token_acc": 0.9375, "train_speed(iter/s)": 1.460147 }, { "epoch": 4.199477314596632, "grad_norm": 2.571464776992798, "learning_rate": 6.192579067171256e-06, "loss": 0.28133187294006345, "memory(GiB)": 77.59, "step": 98020, "token_acc": 0.9389067524115756, "train_speed(iter/s)": 1.460147 }, { "epoch": 4.1996915299258815, "grad_norm": 0.6197360754013062, "learning_rate": 6.189335434506799e-06, "loss": 0.3245791673660278, "memory(GiB)": 77.59, "step": 98025, "token_acc": 0.9263157894736842, "train_speed(iter/s)": 1.460149 }, { "epoch": 4.199905745255131, "grad_norm": 0.8720895051956177, "learning_rate": 6.186092595513615e-06, "loss": 0.3316923141479492, "memory(GiB)": 77.59, "step": 98030, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.460162 }, { "epoch": 4.200119960584379, "grad_norm": 2.657533884048462, "learning_rate": 6.18285055025048e-06, "loss": 0.3757708787918091, "memory(GiB)": 77.59, "step": 98035, "token_acc": 0.9205882352941176, "train_speed(iter/s)": 1.460164 }, { "epoch": 4.200334175913628, "grad_norm": 3.12629771232605, "learning_rate": 6.179609298776106e-06, "loss": 0.29407515525817873, "memory(GiB)": 77.59, "step": 98040, "token_acc": 0.9492063492063492, "train_speed(iter/s)": 1.460164 }, { "epoch": 4.200548391242878, "grad_norm": 13.645951271057129, "learning_rate": 6.176368841149211e-06, "loss": 0.41391916275024415, "memory(GiB)": 77.59, "step": 98045, "token_acc": 0.875, "train_speed(iter/s)": 1.460164 }, { "epoch": 4.200762606572126, "grad_norm": 1.9651498794555664, "learning_rate": 6.17312917742851e-06, "loss": 0.14906247854232788, "memory(GiB)": 77.59, "step": 98050, "token_acc": 0.9691119691119691, "train_speed(iter/s)": 1.460162 }, { "epoch": 4.200976821901375, "grad_norm": 2.1097352504730225, "learning_rate": 6.169890307672693e-06, "loss": 0.31005961894989015, "memory(GiB)": 77.59, "step": 98055, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.460165 }, { "epoch": 4.2011910372306245, "grad_norm": 0.13332220911979675, "learning_rate": 6.166652231940423e-06, "loss": 0.19098480939865112, "memory(GiB)": 77.59, "step": 98060, "token_acc": 0.9532374100719424, "train_speed(iter/s)": 1.460169 }, { "epoch": 4.201405252559873, "grad_norm": 0.9627434611320496, "learning_rate": 6.163414950290358e-06, "loss": 0.1452128529548645, "memory(GiB)": 77.59, "step": 98065, "token_acc": 0.9566929133858267, "train_speed(iter/s)": 1.460174 }, { "epoch": 4.201619467889122, "grad_norm": 3.24282169342041, "learning_rate": 6.1601784627811684e-06, "loss": 0.3717356204986572, "memory(GiB)": 77.59, "step": 98070, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.460174 }, { "epoch": 4.201833683218371, "grad_norm": 1.8840986490249634, "learning_rate": 6.1569427694714635e-06, "loss": 0.47037954330444337, "memory(GiB)": 77.59, "step": 98075, "token_acc": 0.9197080291970803, "train_speed(iter/s)": 1.460175 }, { "epoch": 4.20204789854762, "grad_norm": 3.0803091526031494, "learning_rate": 6.1537078704198726e-06, "loss": 0.5272412776947022, "memory(GiB)": 77.59, "step": 98080, "token_acc": 0.8984375, "train_speed(iter/s)": 1.460183 }, { "epoch": 4.202262113876869, "grad_norm": 6.262923240661621, "learning_rate": 6.150473765684994e-06, "loss": 0.2853280782699585, "memory(GiB)": 77.59, "step": 98085, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.460186 }, { "epoch": 4.202476329206118, "grad_norm": 1.5645930767059326, "learning_rate": 6.147240455325409e-06, "loss": 0.13065292835235595, "memory(GiB)": 77.59, "step": 98090, "token_acc": 0.9551724137931035, "train_speed(iter/s)": 1.460199 }, { "epoch": 4.202690544535367, "grad_norm": 5.716933250427246, "learning_rate": 6.14400793939971e-06, "loss": 0.3632649898529053, "memory(GiB)": 77.59, "step": 98095, "token_acc": 0.9275862068965517, "train_speed(iter/s)": 1.460211 }, { "epoch": 4.202904759864616, "grad_norm": 0.6317568421363831, "learning_rate": 6.140776217966443e-06, "loss": 0.03927308917045593, "memory(GiB)": 77.59, "step": 98100, "token_acc": 0.9843260188087775, "train_speed(iter/s)": 1.460215 }, { "epoch": 4.203118975193865, "grad_norm": 2.001603126525879, "learning_rate": 6.137545291084162e-06, "loss": 0.4550739288330078, "memory(GiB)": 77.59, "step": 98105, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.460225 }, { "epoch": 4.2033331905231135, "grad_norm": 4.520556449890137, "learning_rate": 6.134315158811393e-06, "loss": 0.6394781589508056, "memory(GiB)": 77.59, "step": 98110, "token_acc": 0.8769716088328076, "train_speed(iter/s)": 1.460237 }, { "epoch": 4.203547405852363, "grad_norm": 5.785590648651123, "learning_rate": 6.1310858212066445e-06, "loss": 0.4844368934631348, "memory(GiB)": 77.59, "step": 98115, "token_acc": 0.8933823529411765, "train_speed(iter/s)": 1.460246 }, { "epoch": 4.203761621181612, "grad_norm": 1.6158615350723267, "learning_rate": 6.127857278328442e-06, "loss": 0.23742752075195311, "memory(GiB)": 77.59, "step": 98120, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.460254 }, { "epoch": 4.20397583651086, "grad_norm": 2.10612154006958, "learning_rate": 6.124629530235249e-06, "loss": 0.3526984930038452, "memory(GiB)": 77.59, "step": 98125, "token_acc": 0.9106529209621993, "train_speed(iter/s)": 1.460258 }, { "epoch": 4.20419005184011, "grad_norm": 4.3854193687438965, "learning_rate": 6.121402576985558e-06, "loss": 0.40417070388793946, "memory(GiB)": 77.59, "step": 98130, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.460267 }, { "epoch": 4.204404267169359, "grad_norm": 6.884286403656006, "learning_rate": 6.118176418637822e-06, "loss": 0.559720516204834, "memory(GiB)": 77.59, "step": 98135, "token_acc": 0.8805970149253731, "train_speed(iter/s)": 1.460279 }, { "epoch": 4.204618482498607, "grad_norm": 3.6299490928649902, "learning_rate": 6.114951055250484e-06, "loss": 0.2233105182647705, "memory(GiB)": 77.59, "step": 98140, "token_acc": 0.9485530546623794, "train_speed(iter/s)": 1.460284 }, { "epoch": 4.2048326978278565, "grad_norm": 3.066171407699585, "learning_rate": 6.111726486881975e-06, "loss": 0.3955944776535034, "memory(GiB)": 77.59, "step": 98145, "token_acc": 0.9219219219219219, "train_speed(iter/s)": 1.460291 }, { "epoch": 4.205046913157106, "grad_norm": 1.6231837272644043, "learning_rate": 6.108502713590702e-06, "loss": 0.1952950119972229, "memory(GiB)": 77.59, "step": 98150, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.460294 }, { "epoch": 4.205261128486354, "grad_norm": 4.427287578582764, "learning_rate": 6.105279735435082e-06, "loss": 0.5234605312347412, "memory(GiB)": 77.59, "step": 98155, "token_acc": 0.9017543859649123, "train_speed(iter/s)": 1.460296 }, { "epoch": 4.205475343815603, "grad_norm": 4.5213847160339355, "learning_rate": 6.1020575524735005e-06, "loss": 0.18679825067520142, "memory(GiB)": 77.59, "step": 98160, "token_acc": 0.9517241379310345, "train_speed(iter/s)": 1.460297 }, { "epoch": 4.205689559144853, "grad_norm": 2.896451473236084, "learning_rate": 6.098836164764326e-06, "loss": 0.13123548030853271, "memory(GiB)": 77.59, "step": 98165, "token_acc": 0.9699248120300752, "train_speed(iter/s)": 1.460302 }, { "epoch": 4.205903774474101, "grad_norm": 3.6990931034088135, "learning_rate": 6.095615572365921e-06, "loss": 0.14367743730545043, "memory(GiB)": 77.59, "step": 98170, "token_acc": 0.9695817490494296, "train_speed(iter/s)": 1.460302 }, { "epoch": 4.20611798980335, "grad_norm": 4.157937526702881, "learning_rate": 6.092395775336607e-06, "loss": 0.23412373065948486, "memory(GiB)": 77.59, "step": 98175, "token_acc": 0.9494163424124513, "train_speed(iter/s)": 1.460309 }, { "epoch": 4.2063322051325995, "grad_norm": 4.09370231628418, "learning_rate": 6.089176773734745e-06, "loss": 0.45462779998779296, "memory(GiB)": 77.59, "step": 98180, "token_acc": 0.9018987341772152, "train_speed(iter/s)": 1.460311 }, { "epoch": 4.206546420461848, "grad_norm": 9.246460914611816, "learning_rate": 6.085958567618638e-06, "loss": 0.45557370185852053, "memory(GiB)": 77.59, "step": 98185, "token_acc": 0.9094076655052264, "train_speed(iter/s)": 1.460319 }, { "epoch": 4.206760635791097, "grad_norm": 2.724729299545288, "learning_rate": 6.08274115704659e-06, "loss": 0.430684232711792, "memory(GiB)": 77.59, "step": 98190, "token_acc": 0.9163879598662207, "train_speed(iter/s)": 1.460323 }, { "epoch": 4.206974851120346, "grad_norm": 1.4742372035980225, "learning_rate": 6.07952454207687e-06, "loss": 0.18971855640411378, "memory(GiB)": 77.59, "step": 98195, "token_acc": 0.941320293398533, "train_speed(iter/s)": 1.460331 }, { "epoch": 4.207189066449595, "grad_norm": 3.6978464126586914, "learning_rate": 6.076308722767776e-06, "loss": 0.28788158893585203, "memory(GiB)": 77.59, "step": 98200, "token_acc": 0.9403973509933775, "train_speed(iter/s)": 1.460347 }, { "epoch": 4.207403281778844, "grad_norm": 1.8474230766296387, "learning_rate": 6.073093699177551e-06, "loss": 0.542391300201416, "memory(GiB)": 77.59, "step": 98205, "token_acc": 0.8980891719745223, "train_speed(iter/s)": 1.460352 }, { "epoch": 4.207617497108093, "grad_norm": 2.9289681911468506, "learning_rate": 6.069879471364426e-06, "loss": 0.16756672859191896, "memory(GiB)": 77.59, "step": 98210, "token_acc": 0.9529411764705882, "train_speed(iter/s)": 1.46035 }, { "epoch": 4.207831712437342, "grad_norm": 3.33054518699646, "learning_rate": 6.0666660393866596e-06, "loss": 0.1819087505340576, "memory(GiB)": 77.59, "step": 98215, "token_acc": 0.962457337883959, "train_speed(iter/s)": 1.460352 }, { "epoch": 4.208045927766591, "grad_norm": 0.3311339020729065, "learning_rate": 6.063453403302449e-06, "loss": 0.2694936990737915, "memory(GiB)": 77.59, "step": 98220, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.460355 }, { "epoch": 4.20826014309584, "grad_norm": 3.6944830417633057, "learning_rate": 6.06024156317e-06, "loss": 0.5174617767333984, "memory(GiB)": 77.59, "step": 98225, "token_acc": 0.8896797153024911, "train_speed(iter/s)": 1.460359 }, { "epoch": 4.2084743584250885, "grad_norm": 4.786106109619141, "learning_rate": 6.057030519047491e-06, "loss": 0.33652405738830565, "memory(GiB)": 77.59, "step": 98230, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.460368 }, { "epoch": 4.208688573754338, "grad_norm": 2.831531524658203, "learning_rate": 6.053820270993082e-06, "loss": 0.5005819797515869, "memory(GiB)": 77.59, "step": 98235, "token_acc": 0.886986301369863, "train_speed(iter/s)": 1.460373 }, { "epoch": 4.208902789083587, "grad_norm": 1.9027533531188965, "learning_rate": 6.050610819064961e-06, "loss": 0.25626211166381835, "memory(GiB)": 77.59, "step": 98240, "token_acc": 0.9522292993630573, "train_speed(iter/s)": 1.460375 }, { "epoch": 4.209117004412835, "grad_norm": 0.6037582755088806, "learning_rate": 6.047402163321248e-06, "loss": 0.3858400583267212, "memory(GiB)": 77.59, "step": 98245, "token_acc": 0.9033333333333333, "train_speed(iter/s)": 1.460378 }, { "epoch": 4.209331219742085, "grad_norm": 6.199275970458984, "learning_rate": 6.044194303820078e-06, "loss": 0.3889246225357056, "memory(GiB)": 77.59, "step": 98250, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.460392 }, { "epoch": 4.209545435071334, "grad_norm": 2.5593135356903076, "learning_rate": 6.040987240619561e-06, "loss": 0.29590065479278566, "memory(GiB)": 77.59, "step": 98255, "token_acc": 0.9363057324840764, "train_speed(iter/s)": 1.460397 }, { "epoch": 4.209759650400582, "grad_norm": 3.1672117710113525, "learning_rate": 6.037780973777785e-06, "loss": 0.27024378776550295, "memory(GiB)": 77.59, "step": 98260, "token_acc": 0.9432835820895522, "train_speed(iter/s)": 1.460396 }, { "epoch": 4.209973865729832, "grad_norm": 2.8979005813598633, "learning_rate": 6.0345755033528514e-06, "loss": 0.24535608291625977, "memory(GiB)": 77.59, "step": 98265, "token_acc": 0.9459459459459459, "train_speed(iter/s)": 1.460393 }, { "epoch": 4.210188081059081, "grad_norm": 1.855757236480713, "learning_rate": 6.031370829402832e-06, "loss": 0.42762365341186526, "memory(GiB)": 77.59, "step": 98270, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.460391 }, { "epoch": 4.210402296388329, "grad_norm": 6.797255039215088, "learning_rate": 6.0281669519857755e-06, "loss": 0.29749908447265627, "memory(GiB)": 77.59, "step": 98275, "token_acc": 0.9309090909090909, "train_speed(iter/s)": 1.460393 }, { "epoch": 4.2106165117175784, "grad_norm": 2.5197594165802, "learning_rate": 6.024963871159722e-06, "loss": 0.3211709499359131, "memory(GiB)": 77.59, "step": 98280, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.460393 }, { "epoch": 4.210830727046828, "grad_norm": 2.6888070106506348, "learning_rate": 6.021761586982705e-06, "loss": 0.4436773777008057, "memory(GiB)": 77.59, "step": 98285, "token_acc": 0.891566265060241, "train_speed(iter/s)": 1.460397 }, { "epoch": 4.211044942376076, "grad_norm": 4.275557518005371, "learning_rate": 6.018560099512732e-06, "loss": 0.298696231842041, "memory(GiB)": 77.59, "step": 98290, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.460404 }, { "epoch": 4.211259157705325, "grad_norm": 0.7055520415306091, "learning_rate": 6.015359408807786e-06, "loss": 0.08551610708236694, "memory(GiB)": 77.59, "step": 98295, "token_acc": 0.9738805970149254, "train_speed(iter/s)": 1.460409 }, { "epoch": 4.211473373034575, "grad_norm": 5.079324245452881, "learning_rate": 6.012159514925875e-06, "loss": 0.3427137851715088, "memory(GiB)": 77.59, "step": 98300, "token_acc": 0.9239543726235742, "train_speed(iter/s)": 1.460416 }, { "epoch": 4.211687588363823, "grad_norm": 3.1973769664764404, "learning_rate": 6.008960417924964e-06, "loss": 0.3401891469955444, "memory(GiB)": 77.59, "step": 98305, "token_acc": 0.9317269076305221, "train_speed(iter/s)": 1.460415 }, { "epoch": 4.211901803693072, "grad_norm": 3.694179058074951, "learning_rate": 6.005762117862995e-06, "loss": 0.488649320602417, "memory(GiB)": 77.59, "step": 98310, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.460413 }, { "epoch": 4.2121160190223215, "grad_norm": 2.642925500869751, "learning_rate": 6.002564614797923e-06, "loss": 0.326021671295166, "memory(GiB)": 77.59, "step": 98315, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.460415 }, { "epoch": 4.21233023435157, "grad_norm": 4.401251792907715, "learning_rate": 5.999367908787651e-06, "loss": 0.49544081687927244, "memory(GiB)": 77.59, "step": 98320, "token_acc": 0.8764044943820225, "train_speed(iter/s)": 1.460427 }, { "epoch": 4.212544449680819, "grad_norm": 2.8843636512756348, "learning_rate": 5.996171999890116e-06, "loss": 0.20623476505279542, "memory(GiB)": 77.59, "step": 98325, "token_acc": 0.9481481481481482, "train_speed(iter/s)": 1.460425 }, { "epoch": 4.212758665010068, "grad_norm": 3.801729440689087, "learning_rate": 5.992976888163204e-06, "loss": 0.4091972827911377, "memory(GiB)": 77.59, "step": 98330, "token_acc": 0.8954703832752613, "train_speed(iter/s)": 1.460441 }, { "epoch": 4.212972880339317, "grad_norm": 2.0670435428619385, "learning_rate": 5.989782573664798e-06, "loss": 0.1710578203201294, "memory(GiB)": 77.59, "step": 98335, "token_acc": 0.960960960960961, "train_speed(iter/s)": 1.460437 }, { "epoch": 4.213187095668566, "grad_norm": 5.776317119598389, "learning_rate": 5.9865890564527515e-06, "loss": 0.45784711837768555, "memory(GiB)": 77.59, "step": 98340, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.460434 }, { "epoch": 4.213401310997815, "grad_norm": 6.743402004241943, "learning_rate": 5.983396336584945e-06, "loss": 0.5236430168151855, "memory(GiB)": 77.59, "step": 98345, "token_acc": 0.8870292887029289, "train_speed(iter/s)": 1.460436 }, { "epoch": 4.213615526327064, "grad_norm": 4.665584087371826, "learning_rate": 5.9802044141192056e-06, "loss": 0.4332111358642578, "memory(GiB)": 77.59, "step": 98350, "token_acc": 0.9120521172638436, "train_speed(iter/s)": 1.460435 }, { "epoch": 4.213829741656313, "grad_norm": 1.7149096727371216, "learning_rate": 5.977013289113348e-06, "loss": 0.1784236788749695, "memory(GiB)": 77.59, "step": 98355, "token_acc": 0.9571984435797666, "train_speed(iter/s)": 1.460437 }, { "epoch": 4.214043956985562, "grad_norm": 0.2040080577135086, "learning_rate": 5.973822961625203e-06, "loss": 0.565197229385376, "memory(GiB)": 77.59, "step": 98360, "token_acc": 0.9003215434083601, "train_speed(iter/s)": 1.460449 }, { "epoch": 4.2142581723148105, "grad_norm": 1.6076951026916504, "learning_rate": 5.970633431712552e-06, "loss": 0.25562376976013185, "memory(GiB)": 77.59, "step": 98365, "token_acc": 0.9323308270676691, "train_speed(iter/s)": 1.46045 }, { "epoch": 4.21447238764406, "grad_norm": 4.2371416091918945, "learning_rate": 5.967444699433183e-06, "loss": 0.5039235591888428, "memory(GiB)": 77.59, "step": 98370, "token_acc": 0.8920863309352518, "train_speed(iter/s)": 1.460459 }, { "epoch": 4.214686602973309, "grad_norm": 1.7140982151031494, "learning_rate": 5.964256764844855e-06, "loss": 0.44909348487854006, "memory(GiB)": 77.59, "step": 98375, "token_acc": 0.8814102564102564, "train_speed(iter/s)": 1.460457 }, { "epoch": 4.214900818302557, "grad_norm": 1.8835740089416504, "learning_rate": 5.961069628005317e-06, "loss": 0.14329066276550292, "memory(GiB)": 77.59, "step": 98380, "token_acc": 0.973568281938326, "train_speed(iter/s)": 1.460456 }, { "epoch": 4.215115033631807, "grad_norm": 3.9875826835632324, "learning_rate": 5.9578832889723215e-06, "loss": 0.4226044178009033, "memory(GiB)": 77.59, "step": 98385, "token_acc": 0.9204152249134948, "train_speed(iter/s)": 1.460461 }, { "epoch": 4.215329248961056, "grad_norm": 3.207207202911377, "learning_rate": 5.954697747803584e-06, "loss": 0.4239375591278076, "memory(GiB)": 77.59, "step": 98390, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.460464 }, { "epoch": 4.215543464290304, "grad_norm": 1.35182523727417, "learning_rate": 5.951513004556819e-06, "loss": 0.40343570709228516, "memory(GiB)": 77.59, "step": 98395, "token_acc": 0.9313725490196079, "train_speed(iter/s)": 1.460466 }, { "epoch": 4.2157576796195535, "grad_norm": 2.565239429473877, "learning_rate": 5.948329059289715e-06, "loss": 0.4879294395446777, "memory(GiB)": 77.59, "step": 98400, "token_acc": 0.9022801302931596, "train_speed(iter/s)": 1.460474 }, { "epoch": 4.215971894948803, "grad_norm": 2.6764397621154785, "learning_rate": 5.945145912059946e-06, "loss": 0.22307093143463136, "memory(GiB)": 77.59, "step": 98405, "token_acc": 0.9501557632398754, "train_speed(iter/s)": 1.460485 }, { "epoch": 4.216186110278051, "grad_norm": 5.858125686645508, "learning_rate": 5.94196356292519e-06, "loss": 0.35108494758605957, "memory(GiB)": 77.59, "step": 98410, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.460487 }, { "epoch": 4.2164003256073, "grad_norm": 1.6357051134109497, "learning_rate": 5.938782011943089e-06, "loss": 0.22363026142120362, "memory(GiB)": 77.59, "step": 98415, "token_acc": 0.9485294117647058, "train_speed(iter/s)": 1.46049 }, { "epoch": 4.21661454093655, "grad_norm": 6.186338901519775, "learning_rate": 5.935601259171292e-06, "loss": 0.311961030960083, "memory(GiB)": 77.59, "step": 98420, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.460493 }, { "epoch": 4.216828756265798, "grad_norm": 1.9252214431762695, "learning_rate": 5.932421304667418e-06, "loss": 0.3580860376358032, "memory(GiB)": 77.59, "step": 98425, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.460498 }, { "epoch": 4.217042971595047, "grad_norm": 0.7059862613677979, "learning_rate": 5.9292421484890674e-06, "loss": 0.4231275081634521, "memory(GiB)": 77.59, "step": 98430, "token_acc": 0.8757763975155279, "train_speed(iter/s)": 1.460494 }, { "epoch": 4.2172571869242965, "grad_norm": 4.139179706573486, "learning_rate": 5.926063790693837e-06, "loss": 0.1346617102622986, "memory(GiB)": 77.59, "step": 98435, "token_acc": 0.9759450171821306, "train_speed(iter/s)": 1.460508 }, { "epoch": 4.217471402253545, "grad_norm": 2.6003310680389404, "learning_rate": 5.922886231339297e-06, "loss": 0.19401853084564208, "memory(GiB)": 77.59, "step": 98440, "token_acc": 0.956386292834891, "train_speed(iter/s)": 1.460511 }, { "epoch": 4.217685617582794, "grad_norm": 0.36567676067352295, "learning_rate": 5.919709470483032e-06, "loss": 0.17147066593170165, "memory(GiB)": 77.59, "step": 98445, "token_acc": 0.9627118644067797, "train_speed(iter/s)": 1.460513 }, { "epoch": 4.217899832912043, "grad_norm": 4.606250286102295, "learning_rate": 5.9165335081825754e-06, "loss": 0.5336176872253418, "memory(GiB)": 77.59, "step": 98450, "token_acc": 0.8819188191881919, "train_speed(iter/s)": 1.460512 }, { "epoch": 4.218114048241292, "grad_norm": 0.5379306674003601, "learning_rate": 5.913358344495473e-06, "loss": 0.23886210918426515, "memory(GiB)": 77.59, "step": 98455, "token_acc": 0.9579579579579579, "train_speed(iter/s)": 1.460516 }, { "epoch": 4.218328263570541, "grad_norm": 5.035633087158203, "learning_rate": 5.910183979479239e-06, "loss": 0.5329433441162109, "memory(GiB)": 77.59, "step": 98460, "token_acc": 0.8706293706293706, "train_speed(iter/s)": 1.460525 }, { "epoch": 4.21854247889979, "grad_norm": 1.801478624343872, "learning_rate": 5.907010413191372e-06, "loss": 0.25093884468078614, "memory(GiB)": 77.59, "step": 98465, "token_acc": 0.9509803921568627, "train_speed(iter/s)": 1.460527 }, { "epoch": 4.218756694229039, "grad_norm": 4.97199010848999, "learning_rate": 5.903837645689381e-06, "loss": 0.4529107570648193, "memory(GiB)": 77.59, "step": 98470, "token_acc": 0.8839590443686007, "train_speed(iter/s)": 1.460527 }, { "epoch": 4.218970909558288, "grad_norm": 1.7617846727371216, "learning_rate": 5.900665677030742e-06, "loss": 0.2982935428619385, "memory(GiB)": 77.59, "step": 98475, "token_acc": 0.9315960912052117, "train_speed(iter/s)": 1.46053 }, { "epoch": 4.219185124887537, "grad_norm": 2.372936248779297, "learning_rate": 5.89749450727291e-06, "loss": 0.2840536117553711, "memory(GiB)": 77.59, "step": 98480, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.460531 }, { "epoch": 4.2193993402167855, "grad_norm": 5.008861064910889, "learning_rate": 5.894324136473334e-06, "loss": 0.336911940574646, "memory(GiB)": 77.59, "step": 98485, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.460544 }, { "epoch": 4.219613555546035, "grad_norm": 5.133538722991943, "learning_rate": 5.891154564689444e-06, "loss": 0.7812463760375976, "memory(GiB)": 77.59, "step": 98490, "token_acc": 0.8455882352941176, "train_speed(iter/s)": 1.460552 }, { "epoch": 4.219827770875284, "grad_norm": 2.4442646503448486, "learning_rate": 5.887985791978673e-06, "loss": 0.28197875022888186, "memory(GiB)": 77.59, "step": 98495, "token_acc": 0.9424920127795527, "train_speed(iter/s)": 1.460565 }, { "epoch": 4.220041986204532, "grad_norm": 6.069707870483398, "learning_rate": 5.884817818398414e-06, "loss": 0.33835902214050295, "memory(GiB)": 77.59, "step": 98500, "token_acc": 0.9356060606060606, "train_speed(iter/s)": 1.460568 }, { "epoch": 4.220041986204532, "eval_loss": 2.3698079586029053, "eval_runtime": 11.1797, "eval_samples_per_second": 8.945, "eval_steps_per_second": 8.945, "eval_token_acc": 0.4591584158415842, "step": 98500 }, { "epoch": 4.220256201533782, "grad_norm": 4.452273368835449, "learning_rate": 5.881650644006071e-06, "loss": 0.6154337406158448, "memory(GiB)": 77.59, "step": 98505, "token_acc": 0.572573463935886, "train_speed(iter/s)": 1.460302 }, { "epoch": 4.220470416863031, "grad_norm": 3.8812408447265625, "learning_rate": 5.878484268859013e-06, "loss": 0.2837276220321655, "memory(GiB)": 77.59, "step": 98510, "token_acc": 0.9330543933054394, "train_speed(iter/s)": 1.460307 }, { "epoch": 4.220684632192279, "grad_norm": 3.6574974060058594, "learning_rate": 5.875318693014603e-06, "loss": 0.26116945743560793, "memory(GiB)": 77.59, "step": 98515, "token_acc": 0.948, "train_speed(iter/s)": 1.460317 }, { "epoch": 4.2208988475215286, "grad_norm": 2.228839874267578, "learning_rate": 5.872153916530187e-06, "loss": 0.24362611770629883, "memory(GiB)": 77.59, "step": 98520, "token_acc": 0.9421221864951769, "train_speed(iter/s)": 1.46032 }, { "epoch": 4.221113062850778, "grad_norm": 1.5867958068847656, "learning_rate": 5.868989939463087e-06, "loss": 0.0874245822429657, "memory(GiB)": 77.59, "step": 98525, "token_acc": 0.982078853046595, "train_speed(iter/s)": 1.460321 }, { "epoch": 4.221327278180026, "grad_norm": 1.9845398664474487, "learning_rate": 5.865826761870646e-06, "loss": 0.3263733148574829, "memory(GiB)": 77.59, "step": 98530, "token_acc": 0.9084745762711864, "train_speed(iter/s)": 1.460326 }, { "epoch": 4.221541493509275, "grad_norm": 2.972912073135376, "learning_rate": 5.862664383810151e-06, "loss": 0.518475866317749, "memory(GiB)": 77.59, "step": 98535, "token_acc": 0.911660777385159, "train_speed(iter/s)": 1.460335 }, { "epoch": 4.221755708838525, "grad_norm": 2.6933553218841553, "learning_rate": 5.859502805338896e-06, "loss": 0.23730204105377198, "memory(GiB)": 77.59, "step": 98540, "token_acc": 0.9442231075697212, "train_speed(iter/s)": 1.460336 }, { "epoch": 4.221969924167773, "grad_norm": 5.218696594238281, "learning_rate": 5.856342026514156e-06, "loss": 0.324979567527771, "memory(GiB)": 77.59, "step": 98545, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.460339 }, { "epoch": 4.222184139497022, "grad_norm": 3.0255656242370605, "learning_rate": 5.85318204739318e-06, "loss": 0.27507891654968264, "memory(GiB)": 77.59, "step": 98550, "token_acc": 0.9375, "train_speed(iter/s)": 1.460343 }, { "epoch": 4.222398354826272, "grad_norm": 1.9940305948257446, "learning_rate": 5.850022868033234e-06, "loss": 0.24088263511657715, "memory(GiB)": 77.59, "step": 98555, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.460348 }, { "epoch": 4.22261257015552, "grad_norm": 2.7179617881774902, "learning_rate": 5.846864488491527e-06, "loss": 0.21496038436889647, "memory(GiB)": 77.59, "step": 98560, "token_acc": 0.9486404833836858, "train_speed(iter/s)": 1.460347 }, { "epoch": 4.222826785484769, "grad_norm": 1.8753952980041504, "learning_rate": 5.843706908825303e-06, "loss": 0.3633769750595093, "memory(GiB)": 77.59, "step": 98565, "token_acc": 0.9194139194139194, "train_speed(iter/s)": 1.460345 }, { "epoch": 4.2230410008140185, "grad_norm": 6.614856243133545, "learning_rate": 5.840550129091743e-06, "loss": 0.3339223861694336, "memory(GiB)": 77.59, "step": 98570, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.460346 }, { "epoch": 4.223255216143267, "grad_norm": 2.6373326778411865, "learning_rate": 5.837394149348052e-06, "loss": 0.15577929019927977, "memory(GiB)": 77.59, "step": 98575, "token_acc": 0.9632107023411371, "train_speed(iter/s)": 1.460359 }, { "epoch": 4.223469431472516, "grad_norm": 3.111429214477539, "learning_rate": 5.83423896965139e-06, "loss": 0.14420044422149658, "memory(GiB)": 77.59, "step": 98580, "token_acc": 0.9752066115702479, "train_speed(iter/s)": 1.460363 }, { "epoch": 4.223683646801765, "grad_norm": 5.654809474945068, "learning_rate": 5.831084590058905e-06, "loss": 0.31103034019470216, "memory(GiB)": 77.59, "step": 98585, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.46037 }, { "epoch": 4.223897862131014, "grad_norm": 3.144057512283325, "learning_rate": 5.827931010627774e-06, "loss": 0.5136019706726074, "memory(GiB)": 77.59, "step": 98590, "token_acc": 0.8966789667896679, "train_speed(iter/s)": 1.460376 }, { "epoch": 4.224112077460263, "grad_norm": 2.850536823272705, "learning_rate": 5.824778231415106e-06, "loss": 0.19430198669433593, "memory(GiB)": 77.59, "step": 98595, "token_acc": 0.9588014981273408, "train_speed(iter/s)": 1.460376 }, { "epoch": 4.224326292789512, "grad_norm": 5.872076034545898, "learning_rate": 5.821626252478018e-06, "loss": 0.6430590629577637, "memory(GiB)": 77.59, "step": 98600, "token_acc": 0.868, "train_speed(iter/s)": 1.460394 }, { "epoch": 4.224540508118761, "grad_norm": 6.225719928741455, "learning_rate": 5.818475073873614e-06, "loss": 0.30883588790893557, "memory(GiB)": 77.59, "step": 98605, "token_acc": 0.9269102990033222, "train_speed(iter/s)": 1.460397 }, { "epoch": 4.22475472344801, "grad_norm": 2.7378525733947754, "learning_rate": 5.815324695658975e-06, "loss": 0.28690917491912843, "memory(GiB)": 77.59, "step": 98610, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.460395 }, { "epoch": 4.224968938777259, "grad_norm": 3.567760467529297, "learning_rate": 5.812175117891183e-06, "loss": 0.31884002685546875, "memory(GiB)": 77.59, "step": 98615, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.460404 }, { "epoch": 4.2251831541065075, "grad_norm": 0.3013283908367157, "learning_rate": 5.809026340627288e-06, "loss": 0.368349289894104, "memory(GiB)": 77.59, "step": 98620, "token_acc": 0.933579335793358, "train_speed(iter/s)": 1.460406 }, { "epoch": 4.225397369435757, "grad_norm": 3.1369173526763916, "learning_rate": 5.805878363924338e-06, "loss": 0.4208204746246338, "memory(GiB)": 77.59, "step": 98625, "token_acc": 0.8934169278996865, "train_speed(iter/s)": 1.460418 }, { "epoch": 4.225611584765006, "grad_norm": 4.346953868865967, "learning_rate": 5.802731187839361e-06, "loss": 0.22518038749694824, "memory(GiB)": 77.59, "step": 98630, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.46043 }, { "epoch": 4.225825800094254, "grad_norm": 0.8391711115837097, "learning_rate": 5.799584812429354e-06, "loss": 0.43103785514831544, "memory(GiB)": 77.59, "step": 98635, "token_acc": 0.9180887372013652, "train_speed(iter/s)": 1.460427 }, { "epoch": 4.226040015423504, "grad_norm": 1.181831955909729, "learning_rate": 5.7964392377513445e-06, "loss": 0.49734997749328613, "memory(GiB)": 77.59, "step": 98640, "token_acc": 0.8884892086330936, "train_speed(iter/s)": 1.460444 }, { "epoch": 4.226254230752753, "grad_norm": 3.833901882171631, "learning_rate": 5.793294463862297e-06, "loss": 0.3075899600982666, "memory(GiB)": 77.59, "step": 98645, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.460444 }, { "epoch": 4.226468446082001, "grad_norm": 4.565054416656494, "learning_rate": 5.790150490819196e-06, "loss": 0.23805198669433594, "memory(GiB)": 77.59, "step": 98650, "token_acc": 0.9683794466403162, "train_speed(iter/s)": 1.460449 }, { "epoch": 4.2266826614112505, "grad_norm": 4.06193208694458, "learning_rate": 5.787007318678994e-06, "loss": 0.5105116367340088, "memory(GiB)": 77.59, "step": 98655, "token_acc": 0.9061371841155235, "train_speed(iter/s)": 1.460452 }, { "epoch": 4.2268968767405, "grad_norm": 2.922534942626953, "learning_rate": 5.7838649474986235e-06, "loss": 0.2080148220062256, "memory(GiB)": 77.59, "step": 98660, "token_acc": 0.9637462235649547, "train_speed(iter/s)": 1.460453 }, { "epoch": 4.227111092069748, "grad_norm": 6.410250663757324, "learning_rate": 5.7807233773350235e-06, "loss": 0.3165459394454956, "memory(GiB)": 77.59, "step": 98665, "token_acc": 0.910828025477707, "train_speed(iter/s)": 1.460452 }, { "epoch": 4.227325307398997, "grad_norm": 4.156235218048096, "learning_rate": 5.7775826082450915e-06, "loss": 0.6105090618133545, "memory(GiB)": 77.59, "step": 98670, "token_acc": 0.8632218844984803, "train_speed(iter/s)": 1.460452 }, { "epoch": 4.227539522728247, "grad_norm": 4.551898002624512, "learning_rate": 5.774442640285738e-06, "loss": 0.6724032402038574, "memory(GiB)": 77.59, "step": 98675, "token_acc": 0.8550185873605948, "train_speed(iter/s)": 1.460462 }, { "epoch": 4.227753738057495, "grad_norm": 1.329150915145874, "learning_rate": 5.77130347351385e-06, "loss": 0.25801548957824705, "memory(GiB)": 77.59, "step": 98680, "token_acc": 0.943502824858757, "train_speed(iter/s)": 1.460472 }, { "epoch": 4.227967953386744, "grad_norm": 2.826343297958374, "learning_rate": 5.768165107986284e-06, "loss": 0.43781046867370604, "memory(GiB)": 77.59, "step": 98685, "token_acc": 0.9097472924187726, "train_speed(iter/s)": 1.460478 }, { "epoch": 4.2281821687159935, "grad_norm": 3.2613465785980225, "learning_rate": 5.765027543759904e-06, "loss": 0.17825158834457397, "memory(GiB)": 77.59, "step": 98690, "token_acc": 0.9587628865979382, "train_speed(iter/s)": 1.460482 }, { "epoch": 4.228396384045242, "grad_norm": 4.264679431915283, "learning_rate": 5.7618907808915325e-06, "loss": 0.4143477439880371, "memory(GiB)": 77.59, "step": 98695, "token_acc": 0.9030470914127424, "train_speed(iter/s)": 1.460491 }, { "epoch": 4.228610599374491, "grad_norm": 0.23428717255592346, "learning_rate": 5.758754819438017e-06, "loss": 0.2311037302017212, "memory(GiB)": 77.59, "step": 98700, "token_acc": 0.9576547231270358, "train_speed(iter/s)": 1.460495 }, { "epoch": 4.22882481470374, "grad_norm": 4.7222089767456055, "learning_rate": 5.755619659456163e-06, "loss": 0.34157700538635255, "memory(GiB)": 77.59, "step": 98705, "token_acc": 0.9195804195804196, "train_speed(iter/s)": 1.460505 }, { "epoch": 4.229039030032989, "grad_norm": 4.090785503387451, "learning_rate": 5.752485301002752e-06, "loss": 0.33631985187530516, "memory(GiB)": 77.59, "step": 98710, "token_acc": 0.9270516717325228, "train_speed(iter/s)": 1.460506 }, { "epoch": 4.229253245362238, "grad_norm": 3.65490460395813, "learning_rate": 5.74935174413459e-06, "loss": 0.3639557361602783, "memory(GiB)": 77.59, "step": 98715, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.460505 }, { "epoch": 4.229467460691487, "grad_norm": 1.7750933170318604, "learning_rate": 5.746218988908431e-06, "loss": 0.2870866537094116, "memory(GiB)": 77.59, "step": 98720, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.460507 }, { "epoch": 4.229681676020736, "grad_norm": 3.298051118850708, "learning_rate": 5.743087035381028e-06, "loss": 0.348272967338562, "memory(GiB)": 77.59, "step": 98725, "token_acc": 0.9316546762589928, "train_speed(iter/s)": 1.460521 }, { "epoch": 4.229895891349985, "grad_norm": 2.44561767578125, "learning_rate": 5.739955883609105e-06, "loss": 0.38597879409790037, "memory(GiB)": 77.59, "step": 98730, "token_acc": 0.9105691056910569, "train_speed(iter/s)": 1.460529 }, { "epoch": 4.230110106679234, "grad_norm": 0.6552690863609314, "learning_rate": 5.736825533649415e-06, "loss": 0.10939161777496338, "memory(GiB)": 77.59, "step": 98735, "token_acc": 0.9797979797979798, "train_speed(iter/s)": 1.460532 }, { "epoch": 4.2303243220084825, "grad_norm": 6.23784875869751, "learning_rate": 5.733695985558651e-06, "loss": 0.48670239448547364, "memory(GiB)": 77.59, "step": 98740, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.460539 }, { "epoch": 4.230538537337732, "grad_norm": 3.5105419158935547, "learning_rate": 5.730567239393514e-06, "loss": 0.5075639247894287, "memory(GiB)": 77.59, "step": 98745, "token_acc": 0.875, "train_speed(iter/s)": 1.460542 }, { "epoch": 4.230752752666981, "grad_norm": 4.238986968994141, "learning_rate": 5.7274392952106735e-06, "loss": 0.3776716232299805, "memory(GiB)": 77.59, "step": 98750, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.460555 }, { "epoch": 4.230966967996229, "grad_norm": 2.1574151515960693, "learning_rate": 5.724312153066791e-06, "loss": 0.45610823631286623, "memory(GiB)": 77.59, "step": 98755, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.460572 }, { "epoch": 4.231181183325479, "grad_norm": 3.5169928073883057, "learning_rate": 5.721185813018543e-06, "loss": 0.42739391326904297, "memory(GiB)": 77.59, "step": 98760, "token_acc": 0.9020408163265307, "train_speed(iter/s)": 1.460573 }, { "epoch": 4.231395398654728, "grad_norm": 2.5923125743865967, "learning_rate": 5.71806027512255e-06, "loss": 0.15773770809173585, "memory(GiB)": 77.59, "step": 98765, "token_acc": 0.9653846153846154, "train_speed(iter/s)": 1.460577 }, { "epoch": 4.231609613983976, "grad_norm": 2.6478028297424316, "learning_rate": 5.714935539435429e-06, "loss": 0.16280879974365234, "memory(GiB)": 77.59, "step": 98770, "token_acc": 0.9616858237547893, "train_speed(iter/s)": 1.460582 }, { "epoch": 4.2318238293132255, "grad_norm": 3.6796998977661133, "learning_rate": 5.711811606013801e-06, "loss": 0.32240610122680663, "memory(GiB)": 77.59, "step": 98775, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.460584 }, { "epoch": 4.232038044642475, "grad_norm": 2.0940918922424316, "learning_rate": 5.708688474914237e-06, "loss": 0.38462400436401367, "memory(GiB)": 77.59, "step": 98780, "token_acc": 0.9296875, "train_speed(iter/s)": 1.46059 }, { "epoch": 4.232252259971723, "grad_norm": 4.353359699249268, "learning_rate": 5.705566146193342e-06, "loss": 0.23185746669769286, "memory(GiB)": 77.59, "step": 98785, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.46059 }, { "epoch": 4.232466475300972, "grad_norm": 0.4727589190006256, "learning_rate": 5.702444619907654e-06, "loss": 0.28174803256988523, "memory(GiB)": 77.59, "step": 98790, "token_acc": 0.9425287356321839, "train_speed(iter/s)": 1.460597 }, { "epoch": 4.232680690630222, "grad_norm": 2.2007791996002197, "learning_rate": 5.699323896113751e-06, "loss": 0.2730401039123535, "memory(GiB)": 77.59, "step": 98795, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.460602 }, { "epoch": 4.23289490595947, "grad_norm": 4.227914333343506, "learning_rate": 5.696203974868147e-06, "loss": 0.4853482723236084, "memory(GiB)": 77.59, "step": 98800, "token_acc": 0.9178082191780822, "train_speed(iter/s)": 1.460609 }, { "epoch": 4.233109121288719, "grad_norm": 2.662520408630371, "learning_rate": 5.6930848562273766e-06, "loss": 0.40438046455383303, "memory(GiB)": 77.59, "step": 98805, "token_acc": 0.9142011834319527, "train_speed(iter/s)": 1.46061 }, { "epoch": 4.233323336617969, "grad_norm": 2.2792513370513916, "learning_rate": 5.6899665402479315e-06, "loss": 0.3501986265182495, "memory(GiB)": 77.59, "step": 98810, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.460608 }, { "epoch": 4.233537551947217, "grad_norm": 0.04690990597009659, "learning_rate": 5.686849026986296e-06, "loss": 0.39158694744110106, "memory(GiB)": 77.59, "step": 98815, "token_acc": 0.9058823529411765, "train_speed(iter/s)": 1.460615 }, { "epoch": 4.233751767276466, "grad_norm": 3.6037979125976562, "learning_rate": 5.683732316498974e-06, "loss": 0.18790595531463622, "memory(GiB)": 77.59, "step": 98820, "token_acc": 0.9662162162162162, "train_speed(iter/s)": 1.460617 }, { "epoch": 4.233965982605715, "grad_norm": 5.276424407958984, "learning_rate": 5.6806164088424154e-06, "loss": 0.4726551055908203, "memory(GiB)": 77.59, "step": 98825, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.460615 }, { "epoch": 4.234180197934964, "grad_norm": 0.641029953956604, "learning_rate": 5.677501304073063e-06, "loss": 0.4694469928741455, "memory(GiB)": 77.59, "step": 98830, "token_acc": 0.8805460750853242, "train_speed(iter/s)": 1.460618 }, { "epoch": 4.234394413264213, "grad_norm": 3.8378779888153076, "learning_rate": 5.67438700224735e-06, "loss": 0.5453046321868896, "memory(GiB)": 77.59, "step": 98835, "token_acc": 0.8863636363636364, "train_speed(iter/s)": 1.460629 }, { "epoch": 4.234608628593462, "grad_norm": 4.921539783477783, "learning_rate": 5.671273503421692e-06, "loss": 0.28703956604003905, "memory(GiB)": 77.59, "step": 98840, "token_acc": 0.948905109489051, "train_speed(iter/s)": 1.460634 }, { "epoch": 4.234822843922711, "grad_norm": 0.09099418669939041, "learning_rate": 5.668160807652506e-06, "loss": 0.361665940284729, "memory(GiB)": 77.59, "step": 98845, "token_acc": 0.9093567251461988, "train_speed(iter/s)": 1.460636 }, { "epoch": 4.23503705925196, "grad_norm": 1.8570798635482788, "learning_rate": 5.6650489149961685e-06, "loss": 0.3637643575668335, "memory(GiB)": 77.59, "step": 98850, "token_acc": 0.9391025641025641, "train_speed(iter/s)": 1.460648 }, { "epoch": 4.235251274581209, "grad_norm": 3.433112144470215, "learning_rate": 5.661937825509067e-06, "loss": 0.2741391181945801, "memory(GiB)": 77.59, "step": 98855, "token_acc": 0.9128919860627178, "train_speed(iter/s)": 1.460649 }, { "epoch": 4.235465489910458, "grad_norm": 2.5715365409851074, "learning_rate": 5.658827539247541e-06, "loss": 0.2528564214706421, "memory(GiB)": 77.59, "step": 98860, "token_acc": 0.9381443298969072, "train_speed(iter/s)": 1.460654 }, { "epoch": 4.235679705239707, "grad_norm": 2.370586395263672, "learning_rate": 5.655718056267962e-06, "loss": 0.22785680294036864, "memory(GiB)": 77.59, "step": 98865, "token_acc": 0.9505703422053232, "train_speed(iter/s)": 1.460655 }, { "epoch": 4.235893920568956, "grad_norm": 2.640453815460205, "learning_rate": 5.652609376626644e-06, "loss": 0.4014315605163574, "memory(GiB)": 77.59, "step": 98870, "token_acc": 0.9338842975206612, "train_speed(iter/s)": 1.460659 }, { "epoch": 4.2361081358982045, "grad_norm": 2.551140785217285, "learning_rate": 5.6495015003799e-06, "loss": 0.3089723825454712, "memory(GiB)": 77.59, "step": 98875, "token_acc": 0.946236559139785, "train_speed(iter/s)": 1.460664 }, { "epoch": 4.236322351227454, "grad_norm": 2.0026068687438965, "learning_rate": 5.6463944275840505e-06, "loss": 0.3701270580291748, "memory(GiB)": 77.59, "step": 98880, "token_acc": 0.9362416107382551, "train_speed(iter/s)": 1.460666 }, { "epoch": 4.236536566556703, "grad_norm": 3.2718496322631836, "learning_rate": 5.643288158295373e-06, "loss": 0.3907939434051514, "memory(GiB)": 77.59, "step": 98885, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.460666 }, { "epoch": 4.236750781885951, "grad_norm": 2.3231453895568848, "learning_rate": 5.640182692570145e-06, "loss": 0.45495147705078126, "memory(GiB)": 77.59, "step": 98890, "token_acc": 0.9144736842105263, "train_speed(iter/s)": 1.460669 }, { "epoch": 4.236964997215201, "grad_norm": 1.3106743097305298, "learning_rate": 5.637078030464615e-06, "loss": 0.4961081027984619, "memory(GiB)": 77.59, "step": 98895, "token_acc": 0.8888888888888888, "train_speed(iter/s)": 1.460669 }, { "epoch": 4.23717921254445, "grad_norm": 3.912414789199829, "learning_rate": 5.633974172035023e-06, "loss": 0.381719970703125, "memory(GiB)": 77.59, "step": 98900, "token_acc": 0.8698630136986302, "train_speed(iter/s)": 1.460668 }, { "epoch": 4.237393427873698, "grad_norm": 1.9742006063461304, "learning_rate": 5.630871117337621e-06, "loss": 0.26785926818847655, "memory(GiB)": 77.59, "step": 98905, "token_acc": 0.9407008086253369, "train_speed(iter/s)": 1.460671 }, { "epoch": 4.2376076432029475, "grad_norm": 2.2781193256378174, "learning_rate": 5.627768866428606e-06, "loss": 0.2972802400588989, "memory(GiB)": 77.59, "step": 98910, "token_acc": 0.9342560553633218, "train_speed(iter/s)": 1.460674 }, { "epoch": 4.237821858532197, "grad_norm": 1.8144066333770752, "learning_rate": 5.624667419364182e-06, "loss": 0.34033238887786865, "memory(GiB)": 77.59, "step": 98915, "token_acc": 0.9190031152647975, "train_speed(iter/s)": 1.460675 }, { "epoch": 4.238036073861445, "grad_norm": 5.045583248138428, "learning_rate": 5.621566776200538e-06, "loss": 0.2978729486465454, "memory(GiB)": 77.59, "step": 98920, "token_acc": 0.9431438127090301, "train_speed(iter/s)": 1.460681 }, { "epoch": 4.238250289190694, "grad_norm": 0.5852870345115662, "learning_rate": 5.618466936993832e-06, "loss": 0.3177135467529297, "memory(GiB)": 77.59, "step": 98925, "token_acc": 0.9298780487804879, "train_speed(iter/s)": 1.460686 }, { "epoch": 4.238464504519944, "grad_norm": 3.5753440856933594, "learning_rate": 5.6153679018002394e-06, "loss": 0.2511808156967163, "memory(GiB)": 77.59, "step": 98930, "token_acc": 0.9461077844311377, "train_speed(iter/s)": 1.460687 }, { "epoch": 4.238678719849192, "grad_norm": 1.6270471811294556, "learning_rate": 5.612269670675885e-06, "loss": 0.12153782844543456, "memory(GiB)": 77.59, "step": 98935, "token_acc": 0.976271186440678, "train_speed(iter/s)": 1.46069 }, { "epoch": 4.238892935178441, "grad_norm": 4.096114635467529, "learning_rate": 5.609172243676914e-06, "loss": 0.579539442062378, "memory(GiB)": 77.59, "step": 98940, "token_acc": 0.8858024691358025, "train_speed(iter/s)": 1.460703 }, { "epoch": 4.2391071505076905, "grad_norm": 2.787522077560425, "learning_rate": 5.60607562085943e-06, "loss": 0.1848574161529541, "memory(GiB)": 77.59, "step": 98945, "token_acc": 0.9519230769230769, "train_speed(iter/s)": 1.460705 }, { "epoch": 4.23932136583694, "grad_norm": 2.466801404953003, "learning_rate": 5.602979802279534e-06, "loss": 0.25361409187316897, "memory(GiB)": 77.59, "step": 98950, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.460708 }, { "epoch": 4.239535581166188, "grad_norm": 3.2727839946746826, "learning_rate": 5.599884787993304e-06, "loss": 0.3867812156677246, "memory(GiB)": 77.59, "step": 98955, "token_acc": 0.889763779527559, "train_speed(iter/s)": 1.460711 }, { "epoch": 4.239749796495437, "grad_norm": 1.552621841430664, "learning_rate": 5.5967905780567985e-06, "loss": 0.21828792095184327, "memory(GiB)": 77.59, "step": 98960, "token_acc": 0.9486404833836858, "train_speed(iter/s)": 1.460715 }, { "epoch": 4.239964011824686, "grad_norm": 1.6285334825515747, "learning_rate": 5.593697172526097e-06, "loss": 0.10240182876586915, "memory(GiB)": 77.59, "step": 98965, "token_acc": 0.9725609756097561, "train_speed(iter/s)": 1.460721 }, { "epoch": 4.240178227153935, "grad_norm": 7.094956874847412, "learning_rate": 5.590604571457231e-06, "loss": 0.48931303024291994, "memory(GiB)": 77.59, "step": 98970, "token_acc": 0.9036544850498339, "train_speed(iter/s)": 1.460724 }, { "epoch": 4.240392442483184, "grad_norm": 3.6230616569519043, "learning_rate": 5.587512774906217e-06, "loss": 0.45249485969543457, "memory(GiB)": 77.59, "step": 98975, "token_acc": 0.9055374592833876, "train_speed(iter/s)": 1.460724 }, { "epoch": 4.2406066578124335, "grad_norm": 3.3586666584014893, "learning_rate": 5.584421782929078e-06, "loss": 0.3305421590805054, "memory(GiB)": 77.59, "step": 98980, "token_acc": 0.9295302013422819, "train_speed(iter/s)": 1.460724 }, { "epoch": 4.240820873141682, "grad_norm": 2.7416887283325195, "learning_rate": 5.58133159558179e-06, "loss": 0.34415650367736816, "memory(GiB)": 77.59, "step": 98985, "token_acc": 0.9278350515463918, "train_speed(iter/s)": 1.460733 }, { "epoch": 4.241035088470931, "grad_norm": 1.3233532905578613, "learning_rate": 5.5782422129203534e-06, "loss": 0.22794792652130128, "memory(GiB)": 77.59, "step": 98990, "token_acc": 0.9357142857142857, "train_speed(iter/s)": 1.460746 }, { "epoch": 4.2412493038001795, "grad_norm": 2.948298454284668, "learning_rate": 5.575153635000735e-06, "loss": 0.29994001388549807, "memory(GiB)": 77.59, "step": 98995, "token_acc": 0.9492753623188406, "train_speed(iter/s)": 1.46075 }, { "epoch": 4.241463519129429, "grad_norm": 0.917403519153595, "learning_rate": 5.572065861878883e-06, "loss": 0.11734390258789062, "memory(GiB)": 77.59, "step": 99000, "token_acc": 0.9571984435797666, "train_speed(iter/s)": 1.460751 }, { "epoch": 4.241463519129429, "eval_loss": 2.3904130458831787, "eval_runtime": 11.3952, "eval_samples_per_second": 8.776, "eval_steps_per_second": 8.776, "eval_token_acc": 0.4231242312423124, "step": 99000 }, { "epoch": 4.241677734458678, "grad_norm": 3.453608512878418, "learning_rate": 5.568978893610727e-06, "loss": 0.30573675632476804, "memory(GiB)": 77.59, "step": 99005, "token_acc": 0.576068376068376, "train_speed(iter/s)": 1.46049 }, { "epoch": 4.241891949787927, "grad_norm": 3.5876240730285645, "learning_rate": 5.565892730252203e-06, "loss": 0.28070242404937745, "memory(GiB)": 77.59, "step": 99010, "token_acc": 0.9288135593220339, "train_speed(iter/s)": 1.460501 }, { "epoch": 4.242106165117176, "grad_norm": 3.59676456451416, "learning_rate": 5.5628073718592184e-06, "loss": 0.4837662696838379, "memory(GiB)": 77.59, "step": 99015, "token_acc": 0.8963210702341137, "train_speed(iter/s)": 1.460503 }, { "epoch": 4.242320380446425, "grad_norm": 2.2460668087005615, "learning_rate": 5.55972281848765e-06, "loss": 0.16983033418655397, "memory(GiB)": 77.59, "step": 99020, "token_acc": 0.9644012944983819, "train_speed(iter/s)": 1.460506 }, { "epoch": 4.242534595775673, "grad_norm": 3.473839521408081, "learning_rate": 5.556639070193404e-06, "loss": 0.3442616701126099, "memory(GiB)": 77.59, "step": 99025, "token_acc": 0.936026936026936, "train_speed(iter/s)": 1.460509 }, { "epoch": 4.2427488111049225, "grad_norm": 3.477576494216919, "learning_rate": 5.553556127032333e-06, "loss": 0.3409093379974365, "memory(GiB)": 77.59, "step": 99030, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.460507 }, { "epoch": 4.242963026434172, "grad_norm": 1.6943607330322266, "learning_rate": 5.550473989060284e-06, "loss": 0.4166813850402832, "memory(GiB)": 77.59, "step": 99035, "token_acc": 0.9148264984227129, "train_speed(iter/s)": 1.460509 }, { "epoch": 4.243177241763421, "grad_norm": 4.6123175621032715, "learning_rate": 5.5473926563331e-06, "loss": 0.39700140953063967, "memory(GiB)": 77.59, "step": 99040, "token_acc": 0.9041916167664671, "train_speed(iter/s)": 1.460522 }, { "epoch": 4.243391457092669, "grad_norm": 1.4908344745635986, "learning_rate": 5.5443121289065814e-06, "loss": 0.09816011190414428, "memory(GiB)": 77.59, "step": 99045, "token_acc": 0.970873786407767, "train_speed(iter/s)": 1.460523 }, { "epoch": 4.243605672421919, "grad_norm": 3.282500982284546, "learning_rate": 5.541232406836567e-06, "loss": 0.4291842460632324, "memory(GiB)": 77.59, "step": 99050, "token_acc": 0.8877551020408163, "train_speed(iter/s)": 1.460533 }, { "epoch": 4.243819887751167, "grad_norm": 3.307806968688965, "learning_rate": 5.538153490178833e-06, "loss": 0.24405758380889891, "memory(GiB)": 77.59, "step": 99055, "token_acc": 0.9513513513513514, "train_speed(iter/s)": 1.460541 }, { "epoch": 4.244034103080416, "grad_norm": 4.021800994873047, "learning_rate": 5.5350753789891515e-06, "loss": 0.32877278327941895, "memory(GiB)": 77.59, "step": 99060, "token_acc": 0.9240506329113924, "train_speed(iter/s)": 1.460542 }, { "epoch": 4.2442483184096655, "grad_norm": 5.438872337341309, "learning_rate": 5.531998073323297e-06, "loss": 0.3604578018188477, "memory(GiB)": 77.59, "step": 99065, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.460544 }, { "epoch": 4.244462533738915, "grad_norm": 5.412907600402832, "learning_rate": 5.528921573237001e-06, "loss": 0.2004307270050049, "memory(GiB)": 77.59, "step": 99070, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.460543 }, { "epoch": 4.244676749068163, "grad_norm": 6.761722087860107, "learning_rate": 5.525845878786018e-06, "loss": 0.3526031494140625, "memory(GiB)": 77.59, "step": 99075, "token_acc": 0.9172932330827067, "train_speed(iter/s)": 1.460551 }, { "epoch": 4.244890964397412, "grad_norm": 4.720512390136719, "learning_rate": 5.522770990026044e-06, "loss": 0.362024712562561, "memory(GiB)": 77.59, "step": 99080, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.460551 }, { "epoch": 4.245105179726661, "grad_norm": 1.1930980682373047, "learning_rate": 5.519696907012811e-06, "loss": 0.3945646047592163, "memory(GiB)": 77.59, "step": 99085, "token_acc": 0.9072847682119205, "train_speed(iter/s)": 1.460553 }, { "epoch": 4.24531939505591, "grad_norm": 3.085958242416382, "learning_rate": 5.516623629801987e-06, "loss": 0.5524797439575195, "memory(GiB)": 77.59, "step": 99090, "token_acc": 0.8821656050955414, "train_speed(iter/s)": 1.460555 }, { "epoch": 4.245533610385159, "grad_norm": 4.2652177810668945, "learning_rate": 5.51355115844926e-06, "loss": 0.4594606399536133, "memory(GiB)": 77.59, "step": 99095, "token_acc": 0.9105431309904153, "train_speed(iter/s)": 1.460558 }, { "epoch": 4.245747825714409, "grad_norm": 4.223758697509766, "learning_rate": 5.510479493010285e-06, "loss": 0.4724720478057861, "memory(GiB)": 77.59, "step": 99100, "token_acc": 0.9171428571428571, "train_speed(iter/s)": 1.460554 }, { "epoch": 4.245962041043657, "grad_norm": 2.9903666973114014, "learning_rate": 5.5074086335407e-06, "loss": 0.28963356018066405, "memory(GiB)": 77.59, "step": 99105, "token_acc": 0.9320754716981132, "train_speed(iter/s)": 1.460557 }, { "epoch": 4.246176256372906, "grad_norm": 6.638394355773926, "learning_rate": 5.504338580096152e-06, "loss": 0.4154500484466553, "memory(GiB)": 77.59, "step": 99110, "token_acc": 0.914396887159533, "train_speed(iter/s)": 1.460559 }, { "epoch": 4.246390471702155, "grad_norm": 3.5337259769439697, "learning_rate": 5.501269332732256e-06, "loss": 0.2660919904708862, "memory(GiB)": 77.59, "step": 99115, "token_acc": 0.9529780564263323, "train_speed(iter/s)": 1.460563 }, { "epoch": 4.246604687031404, "grad_norm": 2.6475746631622314, "learning_rate": 5.498200891504606e-06, "loss": 0.1978609561920166, "memory(GiB)": 77.59, "step": 99120, "token_acc": 0.936026936026936, "train_speed(iter/s)": 1.460565 }, { "epoch": 4.246818902360653, "grad_norm": 2.398845911026001, "learning_rate": 5.4951332564687926e-06, "loss": 0.27705998420715333, "memory(GiB)": 77.59, "step": 99125, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.460571 }, { "epoch": 4.247033117689902, "grad_norm": 3.060739517211914, "learning_rate": 5.492066427680376e-06, "loss": 0.6145734310150146, "memory(GiB)": 77.59, "step": 99130, "token_acc": 0.8977272727272727, "train_speed(iter/s)": 1.460571 }, { "epoch": 4.247247333019151, "grad_norm": 3.657759189605713, "learning_rate": 5.489000405194944e-06, "loss": 0.3525040626525879, "memory(GiB)": 77.59, "step": 99135, "token_acc": 0.9198717948717948, "train_speed(iter/s)": 1.46057 }, { "epoch": 4.2474615483484, "grad_norm": 0.43371090292930603, "learning_rate": 5.48593518906802e-06, "loss": 0.2933589696884155, "memory(GiB)": 77.59, "step": 99140, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.460568 }, { "epoch": 4.247675763677648, "grad_norm": 3.5647120475769043, "learning_rate": 5.482870779355137e-06, "loss": 0.48013691902160643, "memory(GiB)": 77.59, "step": 99145, "token_acc": 0.8841698841698842, "train_speed(iter/s)": 1.46057 }, { "epoch": 4.247889979006898, "grad_norm": 1.285935640335083, "learning_rate": 5.479807176111812e-06, "loss": 0.5947145938873291, "memory(GiB)": 77.59, "step": 99150, "token_acc": 0.8856088560885609, "train_speed(iter/s)": 1.460579 }, { "epoch": 4.248104194336147, "grad_norm": 3.5292446613311768, "learning_rate": 5.476744379393534e-06, "loss": 0.3278569221496582, "memory(GiB)": 77.59, "step": 99155, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.46058 }, { "epoch": 4.248318409665396, "grad_norm": 3.879993200302124, "learning_rate": 5.4736823892558075e-06, "loss": 0.4710966110229492, "memory(GiB)": 77.59, "step": 99160, "token_acc": 0.8885630498533724, "train_speed(iter/s)": 1.46059 }, { "epoch": 4.2485326249946445, "grad_norm": 2.1294052600860596, "learning_rate": 5.470621205754084e-06, "loss": 0.534494686126709, "memory(GiB)": 77.59, "step": 99165, "token_acc": 0.8950819672131147, "train_speed(iter/s)": 1.460594 }, { "epoch": 4.248746840323894, "grad_norm": 0.19044864177703857, "learning_rate": 5.467560828943841e-06, "loss": 0.22045614719390869, "memory(GiB)": 77.59, "step": 99170, "token_acc": 0.9532710280373832, "train_speed(iter/s)": 1.460596 }, { "epoch": 4.248961055653142, "grad_norm": 5.63406229019165, "learning_rate": 5.464501258880505e-06, "loss": 0.4030198574066162, "memory(GiB)": 77.59, "step": 99175, "token_acc": 0.9043824701195219, "train_speed(iter/s)": 1.460593 }, { "epoch": 4.249175270982391, "grad_norm": 2.6736199855804443, "learning_rate": 5.461442495619507e-06, "loss": 0.13513541221618652, "memory(GiB)": 77.59, "step": 99180, "token_acc": 0.9660377358490566, "train_speed(iter/s)": 1.460596 }, { "epoch": 4.249389486311641, "grad_norm": 3.2597806453704834, "learning_rate": 5.45838453921626e-06, "loss": 0.17900835275650023, "memory(GiB)": 77.59, "step": 99185, "token_acc": 0.9660377358490566, "train_speed(iter/s)": 1.460599 }, { "epoch": 4.24960370164089, "grad_norm": 2.158194065093994, "learning_rate": 5.455327389726151e-06, "loss": 0.38070676326751707, "memory(GiB)": 77.59, "step": 99190, "token_acc": 0.9039039039039038, "train_speed(iter/s)": 1.460603 }, { "epoch": 4.249817916970138, "grad_norm": 4.684634208679199, "learning_rate": 5.452271047204582e-06, "loss": 0.2593016862869263, "memory(GiB)": 77.59, "step": 99195, "token_acc": 0.9341085271317829, "train_speed(iter/s)": 1.460605 }, { "epoch": 4.2500321322993875, "grad_norm": 2.2012524604797363, "learning_rate": 5.449215511706907e-06, "loss": 0.34082052707672117, "memory(GiB)": 77.59, "step": 99200, "token_acc": 0.9307958477508651, "train_speed(iter/s)": 1.460617 }, { "epoch": 4.250246347628636, "grad_norm": 3.8410797119140625, "learning_rate": 5.4461607832884895e-06, "loss": 0.2409813165664673, "memory(GiB)": 77.59, "step": 99205, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.460619 }, { "epoch": 4.250460562957885, "grad_norm": 1.6481209993362427, "learning_rate": 5.443106862004666e-06, "loss": 0.1776225447654724, "memory(GiB)": 77.59, "step": 99210, "token_acc": 0.9525547445255474, "train_speed(iter/s)": 1.460633 }, { "epoch": 4.250674778287134, "grad_norm": 0.5203295350074768, "learning_rate": 5.440053747910751e-06, "loss": 0.30366969108581543, "memory(GiB)": 77.59, "step": 99215, "token_acc": 0.9237536656891495, "train_speed(iter/s)": 1.460636 }, { "epoch": 4.250888993616384, "grad_norm": 3.0099070072174072, "learning_rate": 5.4370014410620674e-06, "loss": 0.44410881996154783, "memory(GiB)": 77.59, "step": 99220, "token_acc": 0.924187725631769, "train_speed(iter/s)": 1.460635 }, { "epoch": 4.251103208945632, "grad_norm": 0.3619317412376404, "learning_rate": 5.4339499415139105e-06, "loss": 0.41265320777893066, "memory(GiB)": 77.59, "step": 99225, "token_acc": 0.9126637554585153, "train_speed(iter/s)": 1.460643 }, { "epoch": 4.251317424274881, "grad_norm": 1.4911712408065796, "learning_rate": 5.4308992493215414e-06, "loss": 0.42545018196105955, "memory(GiB)": 77.59, "step": 99230, "token_acc": 0.9133574007220217, "train_speed(iter/s)": 1.460644 }, { "epoch": 4.25153163960413, "grad_norm": 2.730238914489746, "learning_rate": 5.427849364540255e-06, "loss": 0.28123955726623534, "memory(GiB)": 77.59, "step": 99235, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.460649 }, { "epoch": 4.251745854933379, "grad_norm": 6.953327655792236, "learning_rate": 5.4248002872252904e-06, "loss": 0.2349015474319458, "memory(GiB)": 77.59, "step": 99240, "token_acc": 0.952755905511811, "train_speed(iter/s)": 1.460649 }, { "epoch": 4.251960070262628, "grad_norm": 3.782987356185913, "learning_rate": 5.42175201743188e-06, "loss": 0.28633577823638917, "memory(GiB)": 77.59, "step": 99245, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.460649 }, { "epoch": 4.252174285591877, "grad_norm": 3.6657488346099854, "learning_rate": 5.418704555215243e-06, "loss": 0.5435770511627197, "memory(GiB)": 77.59, "step": 99250, "token_acc": 0.8710691823899371, "train_speed(iter/s)": 1.460653 }, { "epoch": 4.252388500921126, "grad_norm": 3.0204927921295166, "learning_rate": 5.415657900630605e-06, "loss": 0.1926941990852356, "memory(GiB)": 77.59, "step": 99255, "token_acc": 0.9593220338983051, "train_speed(iter/s)": 1.460656 }, { "epoch": 4.252602716250375, "grad_norm": 3.905763626098633, "learning_rate": 5.412612053733146e-06, "loss": 0.49177122116088867, "memory(GiB)": 77.59, "step": 99260, "token_acc": 0.9055944055944056, "train_speed(iter/s)": 1.460664 }, { "epoch": 4.252816931579623, "grad_norm": 1.9018237590789795, "learning_rate": 5.409567014578043e-06, "loss": 0.5189894676208496, "memory(GiB)": 77.59, "step": 99265, "token_acc": 0.9125475285171103, "train_speed(iter/s)": 1.46066 }, { "epoch": 4.253031146908873, "grad_norm": 3.225910186767578, "learning_rate": 5.406522783220463e-06, "loss": 0.2440197706222534, "memory(GiB)": 77.59, "step": 99270, "token_acc": 0.9403508771929825, "train_speed(iter/s)": 1.460671 }, { "epoch": 4.253245362238122, "grad_norm": 1.0787941217422485, "learning_rate": 5.403479359715552e-06, "loss": 0.23378307819366456, "memory(GiB)": 77.59, "step": 99275, "token_acc": 0.9380804953560371, "train_speed(iter/s)": 1.460673 }, { "epoch": 4.253459577567371, "grad_norm": 4.54672384262085, "learning_rate": 5.400436744118448e-06, "loss": 0.325193452835083, "memory(GiB)": 77.59, "step": 99280, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.460682 }, { "epoch": 4.2536737928966195, "grad_norm": 5.089395046234131, "learning_rate": 5.397394936484274e-06, "loss": 0.3960570335388184, "memory(GiB)": 77.59, "step": 99285, "token_acc": 0.9147540983606557, "train_speed(iter/s)": 1.460682 }, { "epoch": 4.253888008225869, "grad_norm": 1.3037617206573486, "learning_rate": 5.394353936868135e-06, "loss": 0.08618568181991577, "memory(GiB)": 77.59, "step": 99290, "token_acc": 0.9795221843003413, "train_speed(iter/s)": 1.460683 }, { "epoch": 4.254102223555117, "grad_norm": 4.489980697631836, "learning_rate": 5.391313745325116e-06, "loss": 0.3892933130264282, "memory(GiB)": 77.59, "step": 99295, "token_acc": 0.9327485380116959, "train_speed(iter/s)": 1.460683 }, { "epoch": 4.254316438884366, "grad_norm": 2.9777703285217285, "learning_rate": 5.3882743619102786e-06, "loss": 0.33685009479522704, "memory(GiB)": 77.59, "step": 99300, "token_acc": 0.9264705882352942, "train_speed(iter/s)": 1.460685 }, { "epoch": 4.254530654213616, "grad_norm": 2.7541799545288086, "learning_rate": 5.385235786678716e-06, "loss": 0.25134906768798826, "memory(GiB)": 77.59, "step": 99305, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.460691 }, { "epoch": 4.254744869542865, "grad_norm": 4.993149757385254, "learning_rate": 5.3821980196854475e-06, "loss": 0.3358609199523926, "memory(GiB)": 77.59, "step": 99310, "token_acc": 0.9245901639344263, "train_speed(iter/s)": 1.46069 }, { "epoch": 4.254959084872113, "grad_norm": 10.258065223693848, "learning_rate": 5.379161060985533e-06, "loss": 0.4189657211303711, "memory(GiB)": 77.59, "step": 99315, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.4607 }, { "epoch": 4.2551733002013625, "grad_norm": 1.8804173469543457, "learning_rate": 5.376124910633967e-06, "loss": 0.377756667137146, "memory(GiB)": 77.59, "step": 99320, "token_acc": 0.9262536873156342, "train_speed(iter/s)": 1.460701 }, { "epoch": 4.255387515530612, "grad_norm": 2.129577875137329, "learning_rate": 5.373089568685758e-06, "loss": 0.480744743347168, "memory(GiB)": 77.59, "step": 99325, "token_acc": 0.8978102189781022, "train_speed(iter/s)": 1.460703 }, { "epoch": 4.25560173085986, "grad_norm": 2.247936248779297, "learning_rate": 5.370055035195898e-06, "loss": 0.48265872001647947, "memory(GiB)": 77.59, "step": 99330, "token_acc": 0.898360655737705, "train_speed(iter/s)": 1.460714 }, { "epoch": 4.255815946189109, "grad_norm": 4.618083477020264, "learning_rate": 5.367021310219344e-06, "loss": 0.4677571296691895, "memory(GiB)": 77.59, "step": 99335, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.460718 }, { "epoch": 4.256030161518359, "grad_norm": 5.072110652923584, "learning_rate": 5.363988393811081e-06, "loss": 0.18441224098205566, "memory(GiB)": 77.59, "step": 99340, "token_acc": 0.957983193277311, "train_speed(iter/s)": 1.460715 }, { "epoch": 4.256244376847607, "grad_norm": 5.641822814941406, "learning_rate": 5.360956286026042e-06, "loss": 0.3643972396850586, "memory(GiB)": 77.59, "step": 99345, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.460721 }, { "epoch": 4.256458592176856, "grad_norm": 2.575453519821167, "learning_rate": 5.357924986919149e-06, "loss": 0.482479190826416, "memory(GiB)": 77.59, "step": 99350, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.460719 }, { "epoch": 4.2566728075061055, "grad_norm": 2.8800721168518066, "learning_rate": 5.354894496545326e-06, "loss": 0.2679301738739014, "memory(GiB)": 77.59, "step": 99355, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.460719 }, { "epoch": 4.256887022835354, "grad_norm": 3.5387628078460693, "learning_rate": 5.351864814959462e-06, "loss": 0.5908097267150879, "memory(GiB)": 77.59, "step": 99360, "token_acc": 0.8619631901840491, "train_speed(iter/s)": 1.460718 }, { "epoch": 4.257101238164603, "grad_norm": 1.376343011856079, "learning_rate": 5.348835942216457e-06, "loss": 0.4091130256652832, "memory(GiB)": 77.59, "step": 99365, "token_acc": 0.8972332015810277, "train_speed(iter/s)": 1.460721 }, { "epoch": 4.257315453493852, "grad_norm": 1.300049901008606, "learning_rate": 5.345807878371173e-06, "loss": 0.22927002906799315, "memory(GiB)": 77.59, "step": 99370, "token_acc": 0.959349593495935, "train_speed(iter/s)": 1.460722 }, { "epoch": 4.257529668823101, "grad_norm": 3.1282405853271484, "learning_rate": 5.3427806234784606e-06, "loss": 0.4178495407104492, "memory(GiB)": 77.59, "step": 99375, "token_acc": 0.9131736526946108, "train_speed(iter/s)": 1.460732 }, { "epoch": 4.25774388415235, "grad_norm": 3.9382073879241943, "learning_rate": 5.339754177593182e-06, "loss": 0.4154224872589111, "memory(GiB)": 77.59, "step": 99380, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.460752 }, { "epoch": 4.257958099481599, "grad_norm": 4.387333393096924, "learning_rate": 5.336728540770148e-06, "loss": 0.4085714340209961, "memory(GiB)": 77.59, "step": 99385, "token_acc": 0.8985507246376812, "train_speed(iter/s)": 1.460749 }, { "epoch": 4.258172314810848, "grad_norm": 4.6281890869140625, "learning_rate": 5.333703713064175e-06, "loss": 0.645814323425293, "memory(GiB)": 77.59, "step": 99390, "token_acc": 0.8456973293768546, "train_speed(iter/s)": 1.46075 }, { "epoch": 4.258386530140097, "grad_norm": 2.8303420543670654, "learning_rate": 5.330679694530049e-06, "loss": 0.22087137699127196, "memory(GiB)": 77.59, "step": 99395, "token_acc": 0.9492753623188406, "train_speed(iter/s)": 1.460757 }, { "epoch": 4.258600745469346, "grad_norm": 3.8760879039764404, "learning_rate": 5.327656485222576e-06, "loss": 0.3366761922836304, "memory(GiB)": 77.59, "step": 99400, "token_acc": 0.9278996865203761, "train_speed(iter/s)": 1.46076 }, { "epoch": 4.258814960798595, "grad_norm": 1.2860451936721802, "learning_rate": 5.324634085196506e-06, "loss": 0.18870819807052613, "memory(GiB)": 77.59, "step": 99405, "token_acc": 0.9522184300341296, "train_speed(iter/s)": 1.460764 }, { "epoch": 4.259029176127844, "grad_norm": 2.7602431774139404, "learning_rate": 5.321612494506606e-06, "loss": 0.37745866775512693, "memory(GiB)": 77.59, "step": 99410, "token_acc": 0.903114186851211, "train_speed(iter/s)": 1.460772 }, { "epoch": 4.259243391457093, "grad_norm": 0.7228614091873169, "learning_rate": 5.318591713207599e-06, "loss": 0.26611392498016356, "memory(GiB)": 77.59, "step": 99415, "token_acc": 0.9442379182156134, "train_speed(iter/s)": 1.460774 }, { "epoch": 4.259457606786341, "grad_norm": 5.340729713439941, "learning_rate": 5.315571741354214e-06, "loss": 0.4543898582458496, "memory(GiB)": 77.59, "step": 99420, "token_acc": 0.8852459016393442, "train_speed(iter/s)": 1.460782 }, { "epoch": 4.259671822115591, "grad_norm": 2.186314105987549, "learning_rate": 5.312552579001173e-06, "loss": 0.29874367713928224, "memory(GiB)": 77.59, "step": 99425, "token_acc": 0.932258064516129, "train_speed(iter/s)": 1.460801 }, { "epoch": 4.25988603744484, "grad_norm": 3.7764477729797363, "learning_rate": 5.309534226203161e-06, "loss": 0.4322027683258057, "memory(GiB)": 77.59, "step": 99430, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.460799 }, { "epoch": 4.260100252774088, "grad_norm": 0.5493272542953491, "learning_rate": 5.306516683014862e-06, "loss": 0.3048254013061523, "memory(GiB)": 77.59, "step": 99435, "token_acc": 0.9269005847953217, "train_speed(iter/s)": 1.460799 }, { "epoch": 4.260314468103338, "grad_norm": 4.7451910972595215, "learning_rate": 5.303499949490937e-06, "loss": 0.39146223068237307, "memory(GiB)": 77.59, "step": 99440, "token_acc": 0.9079365079365079, "train_speed(iter/s)": 1.460799 }, { "epoch": 4.260528683432587, "grad_norm": 5.15762996673584, "learning_rate": 5.300484025686037e-06, "loss": 0.4404001235961914, "memory(GiB)": 77.59, "step": 99445, "token_acc": 0.9174311926605505, "train_speed(iter/s)": 1.460797 }, { "epoch": 4.260742898761835, "grad_norm": 5.699605464935303, "learning_rate": 5.297468911654796e-06, "loss": 0.26614739894866946, "memory(GiB)": 77.59, "step": 99450, "token_acc": 0.9351145038167938, "train_speed(iter/s)": 1.460798 }, { "epoch": 4.2609571140910845, "grad_norm": 2.0094449520111084, "learning_rate": 5.294454607451838e-06, "loss": 0.3527825117111206, "memory(GiB)": 77.59, "step": 99455, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.460796 }, { "epoch": 4.261171329420334, "grad_norm": 1.756239891052246, "learning_rate": 5.291441113131779e-06, "loss": 0.12031493186950684, "memory(GiB)": 77.59, "step": 99460, "token_acc": 0.9631336405529954, "train_speed(iter/s)": 1.460808 }, { "epoch": 4.261385544749582, "grad_norm": 2.8136937618255615, "learning_rate": 5.2884284287492034e-06, "loss": 0.30382671356201174, "memory(GiB)": 77.59, "step": 99465, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.460819 }, { "epoch": 4.261599760078831, "grad_norm": 3.2136754989624023, "learning_rate": 5.285416554358696e-06, "loss": 0.717038631439209, "memory(GiB)": 77.59, "step": 99470, "token_acc": 0.8589211618257261, "train_speed(iter/s)": 1.460819 }, { "epoch": 4.261813975408081, "grad_norm": 2.6056532859802246, "learning_rate": 5.282405490014808e-06, "loss": 0.3850548267364502, "memory(GiB)": 77.59, "step": 99475, "token_acc": 0.9015151515151515, "train_speed(iter/s)": 1.460818 }, { "epoch": 4.262028190737329, "grad_norm": 2.5138978958129883, "learning_rate": 5.279395235772084e-06, "loss": 0.4926425933837891, "memory(GiB)": 77.59, "step": 99480, "token_acc": 0.8980263157894737, "train_speed(iter/s)": 1.460818 }, { "epoch": 4.262242406066578, "grad_norm": 2.4019885063171387, "learning_rate": 5.276385791685079e-06, "loss": 0.3422654151916504, "memory(GiB)": 77.59, "step": 99485, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.460819 }, { "epoch": 4.2624566213958275, "grad_norm": 2.5374789237976074, "learning_rate": 5.273377157808296e-06, "loss": 0.11084836721420288, "memory(GiB)": 77.59, "step": 99490, "token_acc": 0.9806201550387597, "train_speed(iter/s)": 1.460821 }, { "epoch": 4.262670836725076, "grad_norm": 2.123809337615967, "learning_rate": 5.270369334196246e-06, "loss": 0.09977803230285645, "memory(GiB)": 77.59, "step": 99495, "token_acc": 0.9772727272727273, "train_speed(iter/s)": 1.460833 }, { "epoch": 4.262885052054325, "grad_norm": 1.5619951486587524, "learning_rate": 5.267362320903413e-06, "loss": 0.21138787269592285, "memory(GiB)": 77.59, "step": 99500, "token_acc": 0.9396825396825397, "train_speed(iter/s)": 1.460844 }, { "epoch": 4.262885052054325, "eval_loss": 2.3955862522125244, "eval_runtime": 11.7655, "eval_samples_per_second": 8.499, "eval_steps_per_second": 8.499, "eval_token_acc": 0.46153846153846156, "step": 99500 }, { "epoch": 4.263099267383574, "grad_norm": 2.5412425994873047, "learning_rate": 5.264356117984265e-06, "loss": 0.5106884002685547, "memory(GiB)": 77.59, "step": 99505, "token_acc": 0.5801749271137027, "train_speed(iter/s)": 1.460577 }, { "epoch": 4.263313482712823, "grad_norm": 3.7940099239349365, "learning_rate": 5.261350725493286e-06, "loss": 0.4201078414916992, "memory(GiB)": 77.59, "step": 99510, "token_acc": 0.935933147632312, "train_speed(iter/s)": 1.460578 }, { "epoch": 4.263527698042072, "grad_norm": 2.2930619716644287, "learning_rate": 5.258346143484899e-06, "loss": 0.22705252170562745, "memory(GiB)": 77.59, "step": 99515, "token_acc": 0.9490196078431372, "train_speed(iter/s)": 1.460587 }, { "epoch": 4.263741913371321, "grad_norm": 4.403921127319336, "learning_rate": 5.2553423720135494e-06, "loss": 0.3951128959655762, "memory(GiB)": 77.59, "step": 99520, "token_acc": 0.9172932330827067, "train_speed(iter/s)": 1.460588 }, { "epoch": 4.26395612870057, "grad_norm": 4.792290210723877, "learning_rate": 5.2523394111336334e-06, "loss": 0.2982816696166992, "memory(GiB)": 77.59, "step": 99525, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.460591 }, { "epoch": 4.264170344029819, "grad_norm": 2.436452627182007, "learning_rate": 5.249337260899573e-06, "loss": 0.284543514251709, "memory(GiB)": 77.59, "step": 99530, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.4606 }, { "epoch": 4.264384559359068, "grad_norm": 1.5684884786605835, "learning_rate": 5.2463359213657525e-06, "loss": 0.38846089839935305, "memory(GiB)": 77.59, "step": 99535, "token_acc": 0.912621359223301, "train_speed(iter/s)": 1.460603 }, { "epoch": 4.2645987746883165, "grad_norm": 2.21769380569458, "learning_rate": 5.243335392586524e-06, "loss": 0.3381308078765869, "memory(GiB)": 77.59, "step": 99540, "token_acc": 0.9323308270676691, "train_speed(iter/s)": 1.46061 }, { "epoch": 4.264812990017566, "grad_norm": 3.364208936691284, "learning_rate": 5.240335674616276e-06, "loss": 0.32372078895568845, "memory(GiB)": 77.59, "step": 99545, "token_acc": 0.9367588932806324, "train_speed(iter/s)": 1.460625 }, { "epoch": 4.265027205346815, "grad_norm": 1.1386833190917969, "learning_rate": 5.237336767509332e-06, "loss": 0.3740227460861206, "memory(GiB)": 77.59, "step": 99550, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.460623 }, { "epoch": 4.265241420676063, "grad_norm": 2.715829372406006, "learning_rate": 5.234338671320021e-06, "loss": 0.20838632583618164, "memory(GiB)": 77.59, "step": 99555, "token_acc": 0.9592476489028213, "train_speed(iter/s)": 1.46063 }, { "epoch": 4.265455636005313, "grad_norm": 1.433476448059082, "learning_rate": 5.2313413861026614e-06, "loss": 0.33864898681640626, "memory(GiB)": 77.59, "step": 99560, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.460641 }, { "epoch": 4.265669851334562, "grad_norm": 0.5896339416503906, "learning_rate": 5.2283449119115366e-06, "loss": 0.36991283893585203, "memory(GiB)": 77.59, "step": 99565, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.460645 }, { "epoch": 4.26588406666381, "grad_norm": 3.51943039894104, "learning_rate": 5.225349248800954e-06, "loss": 0.1935302734375, "memory(GiB)": 77.59, "step": 99570, "token_acc": 0.9411764705882353, "train_speed(iter/s)": 1.460648 }, { "epoch": 4.2660982819930595, "grad_norm": 2.8167197704315186, "learning_rate": 5.222354396825174e-06, "loss": 0.4228076934814453, "memory(GiB)": 77.59, "step": 99575, "token_acc": 0.9122257053291536, "train_speed(iter/s)": 1.46065 }, { "epoch": 4.266312497322309, "grad_norm": 2.4165494441986084, "learning_rate": 5.2193603560384495e-06, "loss": 0.33481554985046386, "memory(GiB)": 77.59, "step": 99580, "token_acc": 0.9221311475409836, "train_speed(iter/s)": 1.460659 }, { "epoch": 4.266526712651557, "grad_norm": 1.3183972835540771, "learning_rate": 5.216367126495015e-06, "loss": 0.37667431831359866, "memory(GiB)": 77.59, "step": 99585, "token_acc": 0.951048951048951, "train_speed(iter/s)": 1.460663 }, { "epoch": 4.266740927980806, "grad_norm": 4.2846360206604, "learning_rate": 5.213374708249097e-06, "loss": 0.3232332944869995, "memory(GiB)": 77.59, "step": 99590, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.460673 }, { "epoch": 4.266955143310056, "grad_norm": 5.92938232421875, "learning_rate": 5.210383101354915e-06, "loss": 0.5132254123687744, "memory(GiB)": 77.59, "step": 99595, "token_acc": 0.8855218855218855, "train_speed(iter/s)": 1.460666 }, { "epoch": 4.267169358639304, "grad_norm": 3.3376073837280273, "learning_rate": 5.207392305866648e-06, "loss": 0.3741640567779541, "memory(GiB)": 77.59, "step": 99600, "token_acc": 0.9132075471698113, "train_speed(iter/s)": 1.460673 }, { "epoch": 4.267383573968553, "grad_norm": 3.525714874267578, "learning_rate": 5.204402321838503e-06, "loss": 0.3421555757522583, "memory(GiB)": 77.59, "step": 99605, "token_acc": 0.9305019305019305, "train_speed(iter/s)": 1.460678 }, { "epoch": 4.2675977892978025, "grad_norm": 3.1425952911376953, "learning_rate": 5.2014131493246246e-06, "loss": 0.19460070133209229, "memory(GiB)": 77.59, "step": 99610, "token_acc": 0.963076923076923, "train_speed(iter/s)": 1.460677 }, { "epoch": 4.267812004627051, "grad_norm": 4.047091484069824, "learning_rate": 5.198424788379175e-06, "loss": 0.28466804027557374, "memory(GiB)": 77.59, "step": 99615, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.460681 }, { "epoch": 4.2680262199563, "grad_norm": 1.5819629430770874, "learning_rate": 5.195437239056289e-06, "loss": 0.41487464904785154, "memory(GiB)": 77.59, "step": 99620, "token_acc": 0.921832884097035, "train_speed(iter/s)": 1.460685 }, { "epoch": 4.268240435285549, "grad_norm": 1.3933672904968262, "learning_rate": 5.192450501410079e-06, "loss": 0.16637845039367677, "memory(GiB)": 77.59, "step": 99625, "token_acc": 0.9624060150375939, "train_speed(iter/s)": 1.460688 }, { "epoch": 4.268454650614798, "grad_norm": 3.696650743484497, "learning_rate": 5.189464575494668e-06, "loss": 0.34290771484375, "memory(GiB)": 77.59, "step": 99630, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.460692 }, { "epoch": 4.268668865944047, "grad_norm": 3.2812552452087402, "learning_rate": 5.186479461364141e-06, "loss": 0.2942279100418091, "memory(GiB)": 77.59, "step": 99635, "token_acc": 0.9282700421940928, "train_speed(iter/s)": 1.460699 }, { "epoch": 4.268883081273296, "grad_norm": 1.077406883239746, "learning_rate": 5.183495159072583e-06, "loss": 0.29581973552703855, "memory(GiB)": 77.59, "step": 99640, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.460699 }, { "epoch": 4.269097296602545, "grad_norm": 6.157959461212158, "learning_rate": 5.180511668674043e-06, "loss": 0.3744260311126709, "memory(GiB)": 77.59, "step": 99645, "token_acc": 0.916058394160584, "train_speed(iter/s)": 1.460699 }, { "epoch": 4.269311511931794, "grad_norm": 7.448582649230957, "learning_rate": 5.177528990222574e-06, "loss": 0.5244349479675293, "memory(GiB)": 77.59, "step": 99650, "token_acc": 0.8916967509025271, "train_speed(iter/s)": 1.460719 }, { "epoch": 4.269525727261043, "grad_norm": 0.21380171179771423, "learning_rate": 5.174547123772222e-06, "loss": 0.27196803092956545, "memory(GiB)": 77.59, "step": 99655, "token_acc": 0.945054945054945, "train_speed(iter/s)": 1.460723 }, { "epoch": 4.2697399425902915, "grad_norm": 2.301433801651001, "learning_rate": 5.1715660693769985e-06, "loss": 0.31064932346343993, "memory(GiB)": 77.59, "step": 99660, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.460732 }, { "epoch": 4.269954157919541, "grad_norm": 0.6087860465049744, "learning_rate": 5.168585827090911e-06, "loss": 0.2782894134521484, "memory(GiB)": 77.59, "step": 99665, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.460733 }, { "epoch": 4.27016837324879, "grad_norm": 4.691458225250244, "learning_rate": 5.165606396967932e-06, "loss": 0.42331418991088865, "memory(GiB)": 77.59, "step": 99670, "token_acc": 0.89272030651341, "train_speed(iter/s)": 1.460745 }, { "epoch": 4.270382588578038, "grad_norm": 0.46740925312042236, "learning_rate": 5.1626277790620625e-06, "loss": 0.2097219705581665, "memory(GiB)": 77.59, "step": 99675, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.46075 }, { "epoch": 4.270596803907288, "grad_norm": 13.193229675292969, "learning_rate": 5.159649973427255e-06, "loss": 0.3154323577880859, "memory(GiB)": 77.59, "step": 99680, "token_acc": 0.9051724137931034, "train_speed(iter/s)": 1.460751 }, { "epoch": 4.270811019236537, "grad_norm": 2.4334824085235596, "learning_rate": 5.1566729801174385e-06, "loss": 0.3867526054382324, "memory(GiB)": 77.59, "step": 99685, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.46075 }, { "epoch": 4.271025234565785, "grad_norm": 1.381535291671753, "learning_rate": 5.153696799186569e-06, "loss": 0.3242532253265381, "memory(GiB)": 77.59, "step": 99690, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.46075 }, { "epoch": 4.271239449895035, "grad_norm": 0.815574586391449, "learning_rate": 5.1507214306885485e-06, "loss": 0.1908065438270569, "memory(GiB)": 77.59, "step": 99695, "token_acc": 0.9563492063492064, "train_speed(iter/s)": 1.460752 }, { "epoch": 4.271453665224284, "grad_norm": 6.343724727630615, "learning_rate": 5.147746874677284e-06, "loss": 0.392113471031189, "memory(GiB)": 77.59, "step": 99700, "token_acc": 0.9, "train_speed(iter/s)": 1.460753 }, { "epoch": 4.271667880553532, "grad_norm": 5.2909369468688965, "learning_rate": 5.144773131206659e-06, "loss": 0.6642618656158448, "memory(GiB)": 77.59, "step": 99705, "token_acc": 0.860248447204969, "train_speed(iter/s)": 1.460766 }, { "epoch": 4.2718820958827814, "grad_norm": 2.2698426246643066, "learning_rate": 5.141800200330538e-06, "loss": 0.4885410308837891, "memory(GiB)": 77.59, "step": 99710, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.460767 }, { "epoch": 4.272096311212031, "grad_norm": 2.14892578125, "learning_rate": 5.138828082102792e-06, "loss": 0.3900023937225342, "memory(GiB)": 77.59, "step": 99715, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.460772 }, { "epoch": 4.272310526541279, "grad_norm": 4.045999526977539, "learning_rate": 5.135856776577263e-06, "loss": 0.23598723411560057, "memory(GiB)": 77.59, "step": 99720, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.460778 }, { "epoch": 4.272524741870528, "grad_norm": 2.873863935470581, "learning_rate": 5.1328862838077754e-06, "loss": 0.32664592266082765, "memory(GiB)": 77.59, "step": 99725, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.460785 }, { "epoch": 4.272738957199778, "grad_norm": 3.139317512512207, "learning_rate": 5.129916603848139e-06, "loss": 0.22108983993530273, "memory(GiB)": 77.59, "step": 99730, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.460785 }, { "epoch": 4.272953172529026, "grad_norm": 3.6568915843963623, "learning_rate": 5.126947736752141e-06, "loss": 0.6049819469451905, "memory(GiB)": 77.59, "step": 99735, "token_acc": 0.868421052631579, "train_speed(iter/s)": 1.460797 }, { "epoch": 4.273167387858275, "grad_norm": 1.946705937385559, "learning_rate": 5.123979682573599e-06, "loss": 0.28012733459472655, "memory(GiB)": 77.59, "step": 99740, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.46081 }, { "epoch": 4.2733816031875245, "grad_norm": 2.9501142501831055, "learning_rate": 5.12101244136624e-06, "loss": 0.505632734298706, "memory(GiB)": 77.59, "step": 99745, "token_acc": 0.9151515151515152, "train_speed(iter/s)": 1.460814 }, { "epoch": 4.273595818516773, "grad_norm": 6.32075834274292, "learning_rate": 5.118046013183858e-06, "loss": 0.5727143287658691, "memory(GiB)": 77.59, "step": 99750, "token_acc": 0.9049079754601227, "train_speed(iter/s)": 1.460817 }, { "epoch": 4.273810033846022, "grad_norm": 6.38724422454834, "learning_rate": 5.115080398080174e-06, "loss": 0.6648005962371826, "memory(GiB)": 77.59, "step": 99755, "token_acc": 0.8811475409836066, "train_speed(iter/s)": 1.460817 }, { "epoch": 4.274024249175271, "grad_norm": 3.7132508754730225, "learning_rate": 5.112115596108918e-06, "loss": 0.3091521501541138, "memory(GiB)": 77.59, "step": 99760, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.460828 }, { "epoch": 4.27423846450452, "grad_norm": 3.154407024383545, "learning_rate": 5.109151607323792e-06, "loss": 0.1858285427093506, "memory(GiB)": 77.59, "step": 99765, "token_acc": 0.9605734767025089, "train_speed(iter/s)": 1.460832 }, { "epoch": 4.274452679833769, "grad_norm": 3.545576572418213, "learning_rate": 5.1061884317784855e-06, "loss": 0.2557882785797119, "memory(GiB)": 77.59, "step": 99770, "token_acc": 0.9337748344370861, "train_speed(iter/s)": 1.460835 }, { "epoch": 4.274666895163018, "grad_norm": 1.8169572353363037, "learning_rate": 5.103226069526701e-06, "loss": 0.18036774396896363, "memory(GiB)": 77.59, "step": 99775, "token_acc": 0.9608540925266904, "train_speed(iter/s)": 1.460833 }, { "epoch": 4.274881110492267, "grad_norm": 1.6837166547775269, "learning_rate": 5.100264520622089e-06, "loss": 0.24274101257324218, "memory(GiB)": 77.59, "step": 99780, "token_acc": 0.954225352112676, "train_speed(iter/s)": 1.460836 }, { "epoch": 4.275095325821516, "grad_norm": 4.330089569091797, "learning_rate": 5.097303785118307e-06, "loss": 0.31524410247802737, "memory(GiB)": 77.59, "step": 99785, "token_acc": 0.9482758620689655, "train_speed(iter/s)": 1.460836 }, { "epoch": 4.275309541150765, "grad_norm": 2.307032346725464, "learning_rate": 5.09434386306899e-06, "loss": 0.29250144958496094, "memory(GiB)": 77.59, "step": 99790, "token_acc": 0.9433962264150944, "train_speed(iter/s)": 1.460847 }, { "epoch": 4.2755237564800135, "grad_norm": 2.594367504119873, "learning_rate": 5.09138475452775e-06, "loss": 0.45240058898925783, "memory(GiB)": 77.59, "step": 99795, "token_acc": 0.8962025316455696, "train_speed(iter/s)": 1.460846 }, { "epoch": 4.275737971809263, "grad_norm": 5.303891181945801, "learning_rate": 5.088426459548207e-06, "loss": 0.46891107559204104, "memory(GiB)": 77.59, "step": 99800, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.460848 }, { "epoch": 4.275952187138512, "grad_norm": 3.1789190769195557, "learning_rate": 5.085468978183955e-06, "loss": 0.274338436126709, "memory(GiB)": 77.59, "step": 99805, "token_acc": 0.9377049180327869, "train_speed(iter/s)": 1.460851 }, { "epoch": 4.27616640246776, "grad_norm": 3.1071839332580566, "learning_rate": 5.082512310488563e-06, "loss": 0.40357627868652346, "memory(GiB)": 77.59, "step": 99810, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.460858 }, { "epoch": 4.27638061779701, "grad_norm": 3.7382380962371826, "learning_rate": 5.079556456515599e-06, "loss": 0.36572871208190916, "memory(GiB)": 77.59, "step": 99815, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.460863 }, { "epoch": 4.276594833126259, "grad_norm": 3.470856189727783, "learning_rate": 5.0766014163185935e-06, "loss": 0.17744266986846924, "memory(GiB)": 77.59, "step": 99820, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.460864 }, { "epoch": 4.276809048455507, "grad_norm": 2.1829745769500732, "learning_rate": 5.073647189951109e-06, "loss": 0.32458736896514895, "memory(GiB)": 77.59, "step": 99825, "token_acc": 0.9066147859922179, "train_speed(iter/s)": 1.460861 }, { "epoch": 4.2770232637847565, "grad_norm": 5.481524467468262, "learning_rate": 5.07069377746664e-06, "loss": 0.46438307762145997, "memory(GiB)": 77.59, "step": 99830, "token_acc": 0.9236947791164659, "train_speed(iter/s)": 1.460876 }, { "epoch": 4.277237479114006, "grad_norm": 4.677042007446289, "learning_rate": 5.067741178918711e-06, "loss": 0.4265859603881836, "memory(GiB)": 77.59, "step": 99835, "token_acc": 0.9018181818181819, "train_speed(iter/s)": 1.460894 }, { "epoch": 4.277451694443254, "grad_norm": 2.6622257232666016, "learning_rate": 5.0647893943607975e-06, "loss": 0.20108513832092284, "memory(GiB)": 77.59, "step": 99840, "token_acc": 0.9584569732937686, "train_speed(iter/s)": 1.460898 }, { "epoch": 4.277665909772503, "grad_norm": 4.458497047424316, "learning_rate": 5.061838423846377e-06, "loss": 0.5939162731170654, "memory(GiB)": 77.59, "step": 99845, "token_acc": 0.87, "train_speed(iter/s)": 1.460902 }, { "epoch": 4.277880125101753, "grad_norm": 2.879066228866577, "learning_rate": 5.058888267428913e-06, "loss": 0.31253871917724607, "memory(GiB)": 77.59, "step": 99850, "token_acc": 0.9297658862876255, "train_speed(iter/s)": 1.460914 }, { "epoch": 4.278094340431001, "grad_norm": 3.488481044769287, "learning_rate": 5.055938925161835e-06, "loss": 0.212174391746521, "memory(GiB)": 77.59, "step": 99855, "token_acc": 0.9548611111111112, "train_speed(iter/s)": 1.460917 }, { "epoch": 4.27830855576025, "grad_norm": 2.671111583709717, "learning_rate": 5.052990397098589e-06, "loss": 0.45183229446411133, "memory(GiB)": 77.59, "step": 99860, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.460918 }, { "epoch": 4.2785227710894995, "grad_norm": 1.7578084468841553, "learning_rate": 5.050042683292589e-06, "loss": 0.3612847805023193, "memory(GiB)": 77.59, "step": 99865, "token_acc": 0.9178571428571428, "train_speed(iter/s)": 1.46093 }, { "epoch": 4.278736986418748, "grad_norm": 2.57316517829895, "learning_rate": 5.047095783797234e-06, "loss": 0.2856208562850952, "memory(GiB)": 77.59, "step": 99870, "token_acc": 0.9555555555555556, "train_speed(iter/s)": 1.460937 }, { "epoch": 4.278951201747997, "grad_norm": 3.7962288856506348, "learning_rate": 5.044149698665906e-06, "loss": 0.2394437313079834, "memory(GiB)": 77.59, "step": 99875, "token_acc": 0.9347826086956522, "train_speed(iter/s)": 1.460949 }, { "epoch": 4.279165417077246, "grad_norm": 6.048018932342529, "learning_rate": 5.041204427951968e-06, "loss": 0.33783626556396484, "memory(GiB)": 77.59, "step": 99880, "token_acc": 0.9256198347107438, "train_speed(iter/s)": 1.460949 }, { "epoch": 4.279379632406495, "grad_norm": 3.038200616836548, "learning_rate": 5.0382599717087974e-06, "loss": 0.27928388118743896, "memory(GiB)": 77.59, "step": 99885, "token_acc": 0.9219219219219219, "train_speed(iter/s)": 1.460951 }, { "epoch": 4.279593847735744, "grad_norm": 5.3399200439453125, "learning_rate": 5.035316329989725e-06, "loss": 0.5327032566070556, "memory(GiB)": 77.59, "step": 99890, "token_acc": 0.8909657320872274, "train_speed(iter/s)": 1.460955 }, { "epoch": 4.279808063064993, "grad_norm": 9.551346778869629, "learning_rate": 5.03237350284807e-06, "loss": 0.49667882919311523, "memory(GiB)": 77.59, "step": 99895, "token_acc": 0.9221789883268483, "train_speed(iter/s)": 1.460958 }, { "epoch": 4.280022278394242, "grad_norm": 3.8982772827148438, "learning_rate": 5.029431490337156e-06, "loss": 0.43938241004943845, "memory(GiB)": 77.59, "step": 99900, "token_acc": 0.8993506493506493, "train_speed(iter/s)": 1.460958 }, { "epoch": 4.280236493723491, "grad_norm": 3.4469947814941406, "learning_rate": 5.026490292510283e-06, "loss": 0.42351016998291013, "memory(GiB)": 77.59, "step": 99905, "token_acc": 0.9169329073482428, "train_speed(iter/s)": 1.460957 }, { "epoch": 4.28045070905274, "grad_norm": 2.415527582168579, "learning_rate": 5.023549909420722e-06, "loss": 0.6654161930084228, "memory(GiB)": 77.59, "step": 99910, "token_acc": 0.8938547486033519, "train_speed(iter/s)": 1.460961 }, { "epoch": 4.2806649243819885, "grad_norm": 3.2908596992492676, "learning_rate": 5.02061034112174e-06, "loss": 0.24340598583221434, "memory(GiB)": 77.59, "step": 99915, "token_acc": 0.934375, "train_speed(iter/s)": 1.460965 }, { "epoch": 4.280879139711238, "grad_norm": 3.1084794998168945, "learning_rate": 5.0176715876666e-06, "loss": 0.4259368896484375, "memory(GiB)": 77.59, "step": 99920, "token_acc": 0.9133333333333333, "train_speed(iter/s)": 1.460979 }, { "epoch": 4.281093355040487, "grad_norm": 1.4243357181549072, "learning_rate": 5.014733649108538e-06, "loss": 0.23667747974395753, "memory(GiB)": 77.59, "step": 99925, "token_acc": 0.9298892988929889, "train_speed(iter/s)": 1.460978 }, { "epoch": 4.281307570369735, "grad_norm": 3.3944180011749268, "learning_rate": 5.011796525500778e-06, "loss": 0.27852582931518555, "memory(GiB)": 77.59, "step": 99930, "token_acc": 0.9407894736842105, "train_speed(iter/s)": 1.460982 }, { "epoch": 4.281521785698985, "grad_norm": 2.1252758502960205, "learning_rate": 5.008860216896527e-06, "loss": 0.15380915403366088, "memory(GiB)": 77.59, "step": 99935, "token_acc": 0.9477124183006536, "train_speed(iter/s)": 1.460988 }, { "epoch": 4.281736001028234, "grad_norm": 5.243740558624268, "learning_rate": 5.005924723348965e-06, "loss": 0.5058942317962647, "memory(GiB)": 77.59, "step": 99940, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.460989 }, { "epoch": 4.281950216357482, "grad_norm": 0.6776344776153564, "learning_rate": 5.002990044911299e-06, "loss": 0.20588600635528564, "memory(GiB)": 77.59, "step": 99945, "token_acc": 0.9466192170818505, "train_speed(iter/s)": 1.46099 }, { "epoch": 4.2821644316867316, "grad_norm": 2.7738289833068848, "learning_rate": 5.000056181636676e-06, "loss": 0.39241561889648435, "memory(GiB)": 77.59, "step": 99950, "token_acc": 0.9198473282442748, "train_speed(iter/s)": 1.460993 }, { "epoch": 4.282378647015981, "grad_norm": 4.51870584487915, "learning_rate": 4.9971231335782465e-06, "loss": 0.3619981050491333, "memory(GiB)": 77.59, "step": 99955, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.460999 }, { "epoch": 4.282592862345229, "grad_norm": 4.907376766204834, "learning_rate": 4.9941909007891476e-06, "loss": 0.4899251461029053, "memory(GiB)": 77.59, "step": 99960, "token_acc": 0.9, "train_speed(iter/s)": 1.46101 }, { "epoch": 4.282807077674478, "grad_norm": 4.62591028213501, "learning_rate": 4.991259483322491e-06, "loss": 0.27319319248199464, "memory(GiB)": 77.59, "step": 99965, "token_acc": 0.9486301369863014, "train_speed(iter/s)": 1.461014 }, { "epoch": 4.283021293003728, "grad_norm": 2.3805274963378906, "learning_rate": 4.988328881231402e-06, "loss": 0.2790034055709839, "memory(GiB)": 77.59, "step": 99970, "token_acc": 0.9422492401215805, "train_speed(iter/s)": 1.461017 }, { "epoch": 4.283235508332976, "grad_norm": 5.296984672546387, "learning_rate": 4.985399094568949e-06, "loss": 0.2118358612060547, "memory(GiB)": 77.59, "step": 99975, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.461019 }, { "epoch": 4.283449723662225, "grad_norm": 1.5260471105575562, "learning_rate": 4.982470123388228e-06, "loss": 0.34045627117156985, "memory(GiB)": 77.59, "step": 99980, "token_acc": 0.9244186046511628, "train_speed(iter/s)": 1.461019 }, { "epoch": 4.283663938991475, "grad_norm": 4.645987033843994, "learning_rate": 4.97954196774229e-06, "loss": 0.5635487556457519, "memory(GiB)": 77.59, "step": 99985, "token_acc": 0.8833333333333333, "train_speed(iter/s)": 1.461029 }, { "epoch": 4.283878154320723, "grad_norm": 2.4386935234069824, "learning_rate": 4.976614627684184e-06, "loss": 0.31154396533966067, "memory(GiB)": 77.59, "step": 99990, "token_acc": 0.9322709163346613, "train_speed(iter/s)": 1.461035 }, { "epoch": 4.284092369649972, "grad_norm": 0.6911927461624146, "learning_rate": 4.973688103266938e-06, "loss": 0.3273287773132324, "memory(GiB)": 77.59, "step": 99995, "token_acc": 0.9358974358974359, "train_speed(iter/s)": 1.461035 }, { "epoch": 4.2843065849792215, "grad_norm": 4.5256028175354, "learning_rate": 4.97076239454356e-06, "loss": 0.25275349617004395, "memory(GiB)": 77.59, "step": 100000, "token_acc": 0.9347079037800687, "train_speed(iter/s)": 1.461036 }, { "epoch": 4.2843065849792215, "eval_loss": 2.469208002090454, "eval_runtime": 11.2184, "eval_samples_per_second": 8.914, "eval_steps_per_second": 8.914, "eval_token_acc": 0.4560723514211886, "step": 100000 }, { "epoch": 4.28452080030847, "grad_norm": 0.5246925354003906, "learning_rate": 4.967837501567069e-06, "loss": 0.4758607387542725, "memory(GiB)": 77.59, "step": 100005, "token_acc": 0.5973094170403588, "train_speed(iter/s)": 1.46079 }, { "epoch": 4.284735015637719, "grad_norm": 0.5713053941726685, "learning_rate": 4.964913424390449e-06, "loss": 0.20188345909118652, "memory(GiB)": 77.59, "step": 100010, "token_acc": 0.9498207885304659, "train_speed(iter/s)": 1.460793 }, { "epoch": 4.284949230966968, "grad_norm": 3.245299816131592, "learning_rate": 4.961990163066671e-06, "loss": 0.2901291847229004, "memory(GiB)": 77.59, "step": 100015, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.460803 }, { "epoch": 4.285163446296217, "grad_norm": 2.6304373741149902, "learning_rate": 4.959067717648685e-06, "loss": 0.0745045006275177, "memory(GiB)": 77.59, "step": 100020, "token_acc": 0.9704797047970479, "train_speed(iter/s)": 1.460802 }, { "epoch": 4.285377661625466, "grad_norm": 0.09042610973119736, "learning_rate": 4.956146088189434e-06, "loss": 0.24014768600463868, "memory(GiB)": 77.59, "step": 100025, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.460815 }, { "epoch": 4.285591876954715, "grad_norm": 3.540282726287842, "learning_rate": 4.953225274741857e-06, "loss": 0.2688897609710693, "memory(GiB)": 77.59, "step": 100030, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.460814 }, { "epoch": 4.285806092283964, "grad_norm": 1.7145427465438843, "learning_rate": 4.9503052773588635e-06, "loss": 0.3604144811630249, "memory(GiB)": 77.59, "step": 100035, "token_acc": 0.927797833935018, "train_speed(iter/s)": 1.460824 }, { "epoch": 4.286020307613213, "grad_norm": 6.202212333679199, "learning_rate": 4.947386096093337e-06, "loss": 0.5249392032623291, "memory(GiB)": 77.59, "step": 100040, "token_acc": 0.8779220779220779, "train_speed(iter/s)": 1.460829 }, { "epoch": 4.286234522942462, "grad_norm": 2.6635053157806396, "learning_rate": 4.944467730998187e-06, "loss": 0.3964036226272583, "memory(GiB)": 77.59, "step": 100045, "token_acc": 0.9280575539568345, "train_speed(iter/s)": 1.460834 }, { "epoch": 4.2864487382717105, "grad_norm": 0.22549916803836823, "learning_rate": 4.94155018212627e-06, "loss": 0.41340999603271483, "memory(GiB)": 77.59, "step": 100050, "token_acc": 0.886039886039886, "train_speed(iter/s)": 1.460838 }, { "epoch": 4.28666295360096, "grad_norm": 2.216219186782837, "learning_rate": 4.9386334495304384e-06, "loss": 0.26362712383270265, "memory(GiB)": 77.59, "step": 100055, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.46084 }, { "epoch": 4.286877168930209, "grad_norm": 1.9179097414016724, "learning_rate": 4.935717533263523e-06, "loss": 0.5461084842681885, "memory(GiB)": 77.59, "step": 100060, "token_acc": 0.893491124260355, "train_speed(iter/s)": 1.46084 }, { "epoch": 4.287091384259457, "grad_norm": 4.199880123138428, "learning_rate": 4.93280243337837e-06, "loss": 0.4137904167175293, "memory(GiB)": 77.59, "step": 100065, "token_acc": 0.8952879581151832, "train_speed(iter/s)": 1.46084 }, { "epoch": 4.287305599588707, "grad_norm": 0.14376488327980042, "learning_rate": 4.929888149927775e-06, "loss": 0.38546042442321776, "memory(GiB)": 77.59, "step": 100070, "token_acc": 0.9163987138263665, "train_speed(iter/s)": 1.460849 }, { "epoch": 4.287519814917956, "grad_norm": 4.247796535491943, "learning_rate": 4.926974682964536e-06, "loss": 0.6695891380310058, "memory(GiB)": 77.59, "step": 100075, "token_acc": 0.8742138364779874, "train_speed(iter/s)": 1.460862 }, { "epoch": 4.287734030247204, "grad_norm": 1.944679617881775, "learning_rate": 4.924062032541432e-06, "loss": 0.45705881118774416, "memory(GiB)": 77.59, "step": 100080, "token_acc": 0.909967845659164, "train_speed(iter/s)": 1.460875 }, { "epoch": 4.2879482455764535, "grad_norm": 2.034600019454956, "learning_rate": 4.921150198711217e-06, "loss": 0.41254096031188964, "memory(GiB)": 77.59, "step": 100085, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.460878 }, { "epoch": 4.288162460905703, "grad_norm": 0.2476711869239807, "learning_rate": 4.9182391815266685e-06, "loss": 0.16849662065505983, "memory(GiB)": 77.59, "step": 100090, "token_acc": 0.9529411764705882, "train_speed(iter/s)": 1.460881 }, { "epoch": 4.288376676234951, "grad_norm": 5.541838645935059, "learning_rate": 4.915328981040501e-06, "loss": 0.3789521217346191, "memory(GiB)": 77.59, "step": 100095, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.460884 }, { "epoch": 4.2885908915642, "grad_norm": 2.4886715412139893, "learning_rate": 4.912419597305446e-06, "loss": 0.2655519485473633, "memory(GiB)": 77.59, "step": 100100, "token_acc": 0.9382239382239382, "train_speed(iter/s)": 1.460887 }, { "epoch": 4.28880510689345, "grad_norm": 3.3030171394348145, "learning_rate": 4.909511030374209e-06, "loss": 0.3263232231140137, "memory(GiB)": 77.59, "step": 100105, "token_acc": 0.927710843373494, "train_speed(iter/s)": 1.460893 }, { "epoch": 4.289019322222698, "grad_norm": 2.6854236125946045, "learning_rate": 4.906603280299471e-06, "loss": 0.31942975521087646, "memory(GiB)": 77.59, "step": 100110, "token_acc": 0.9416058394160584, "train_speed(iter/s)": 1.460893 }, { "epoch": 4.289233537551947, "grad_norm": 2.6993539333343506, "learning_rate": 4.903696347133912e-06, "loss": 0.48531789779663087, "memory(GiB)": 77.59, "step": 100115, "token_acc": 0.8847583643122676, "train_speed(iter/s)": 1.460897 }, { "epoch": 4.2894477528811965, "grad_norm": 1.5316118001937866, "learning_rate": 4.900790230930191e-06, "loss": 0.3654092073440552, "memory(GiB)": 77.59, "step": 100120, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.460899 }, { "epoch": 4.289661968210445, "grad_norm": 3.4267735481262207, "learning_rate": 4.897884931740976e-06, "loss": 0.21299281120300292, "memory(GiB)": 77.59, "step": 100125, "token_acc": 0.953405017921147, "train_speed(iter/s)": 1.460911 }, { "epoch": 4.289876183539694, "grad_norm": 4.386176109313965, "learning_rate": 4.894980449618886e-06, "loss": 0.39452438354492186, "memory(GiB)": 77.59, "step": 100130, "token_acc": 0.9083333333333333, "train_speed(iter/s)": 1.46091 }, { "epoch": 4.290090398868943, "grad_norm": 3.480156421661377, "learning_rate": 4.892076784616534e-06, "loss": 0.48459672927856445, "memory(GiB)": 77.59, "step": 100135, "token_acc": 0.9023668639053254, "train_speed(iter/s)": 1.460907 }, { "epoch": 4.290304614198192, "grad_norm": 4.2883992195129395, "learning_rate": 4.889173936786523e-06, "loss": 0.6254446983337403, "memory(GiB)": 77.59, "step": 100140, "token_acc": 0.8545454545454545, "train_speed(iter/s)": 1.460904 }, { "epoch": 4.290518829527441, "grad_norm": 4.111508369445801, "learning_rate": 4.886271906181439e-06, "loss": 0.4770817756652832, "memory(GiB)": 77.59, "step": 100145, "token_acc": 0.9134615384615384, "train_speed(iter/s)": 1.460904 }, { "epoch": 4.29073304485669, "grad_norm": 2.442129373550415, "learning_rate": 4.883370692853867e-06, "loss": 0.22514491081237792, "memory(GiB)": 77.59, "step": 100150, "token_acc": 0.9530201342281879, "train_speed(iter/s)": 1.46091 }, { "epoch": 4.290947260185939, "grad_norm": 2.753054141998291, "learning_rate": 4.8804702968563596e-06, "loss": 0.40351228713989257, "memory(GiB)": 77.59, "step": 100155, "token_acc": 0.9096774193548387, "train_speed(iter/s)": 1.460913 }, { "epoch": 4.291161475515188, "grad_norm": 4.4658403396606445, "learning_rate": 4.877570718241453e-06, "loss": 0.6003690719604492, "memory(GiB)": 77.59, "step": 100160, "token_acc": 0.8709677419354839, "train_speed(iter/s)": 1.460909 }, { "epoch": 4.291375690844437, "grad_norm": 2.6378333568573, "learning_rate": 4.8746719570616885e-06, "loss": 0.45105586051940916, "memory(GiB)": 77.59, "step": 100165, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.460921 }, { "epoch": 4.2915899061736855, "grad_norm": 4.915841579437256, "learning_rate": 4.871774013369556e-06, "loss": 0.5028673648834229, "memory(GiB)": 77.59, "step": 100170, "token_acc": 0.8820960698689956, "train_speed(iter/s)": 1.460923 }, { "epoch": 4.291804121502935, "grad_norm": 2.936469793319702, "learning_rate": 4.868876887217583e-06, "loss": 0.27214741706848145, "memory(GiB)": 77.59, "step": 100175, "token_acc": 0.9314641744548287, "train_speed(iter/s)": 1.460929 }, { "epoch": 4.292018336832184, "grad_norm": 6.31168794631958, "learning_rate": 4.865980578658241e-06, "loss": 0.6998337745666504, "memory(GiB)": 77.59, "step": 100180, "token_acc": 0.84, "train_speed(iter/s)": 1.460955 }, { "epoch": 4.292232552161432, "grad_norm": 3.2774226665496826, "learning_rate": 4.8630850877440014e-06, "loss": 0.3347806930541992, "memory(GiB)": 77.59, "step": 100185, "token_acc": 0.9278688524590164, "train_speed(iter/s)": 1.460962 }, { "epoch": 4.292446767490682, "grad_norm": 3.6247949600219727, "learning_rate": 4.86019041452731e-06, "loss": 0.27015373706817625, "memory(GiB)": 77.59, "step": 100190, "token_acc": 0.9540636042402827, "train_speed(iter/s)": 1.460965 }, { "epoch": 4.292660982819931, "grad_norm": 0.5919560790061951, "learning_rate": 4.8572965590606204e-06, "loss": 0.03601826429367065, "memory(GiB)": 77.59, "step": 100195, "token_acc": 0.9960159362549801, "train_speed(iter/s)": 1.460975 }, { "epoch": 4.292875198149179, "grad_norm": 3.5056095123291016, "learning_rate": 4.8544035213963505e-06, "loss": 0.1994800329208374, "memory(GiB)": 77.59, "step": 100200, "token_acc": 0.9590443686006825, "train_speed(iter/s)": 1.460976 }, { "epoch": 4.2930894134784285, "grad_norm": 4.301004886627197, "learning_rate": 4.851511301586903e-06, "loss": 0.29860472679138184, "memory(GiB)": 77.59, "step": 100205, "token_acc": 0.9191489361702128, "train_speed(iter/s)": 1.460979 }, { "epoch": 4.293303628807678, "grad_norm": 3.1022350788116455, "learning_rate": 4.848619899684686e-06, "loss": 0.3912771224975586, "memory(GiB)": 77.59, "step": 100210, "token_acc": 0.91, "train_speed(iter/s)": 1.460983 }, { "epoch": 4.293517844136926, "grad_norm": 6.505673408508301, "learning_rate": 4.845729315742081e-06, "loss": 0.6483358383178711, "memory(GiB)": 77.59, "step": 100215, "token_acc": 0.8586206896551725, "train_speed(iter/s)": 1.460986 }, { "epoch": 4.293732059466175, "grad_norm": 0.7988331317901611, "learning_rate": 4.8428395498114455e-06, "loss": 0.2527296781539917, "memory(GiB)": 77.59, "step": 100220, "token_acc": 0.9207547169811321, "train_speed(iter/s)": 1.460986 }, { "epoch": 4.293946274795425, "grad_norm": 4.438408851623535, "learning_rate": 4.839950601945131e-06, "loss": 0.290722918510437, "memory(GiB)": 77.59, "step": 100225, "token_acc": 0.9338235294117647, "train_speed(iter/s)": 1.460986 }, { "epoch": 4.294160490124673, "grad_norm": 4.109919548034668, "learning_rate": 4.837062472195469e-06, "loss": 0.6590603351593017, "memory(GiB)": 77.59, "step": 100230, "token_acc": 0.8735294117647059, "train_speed(iter/s)": 1.460995 }, { "epoch": 4.294374705453922, "grad_norm": 0.018743207678198814, "learning_rate": 4.83417516061479e-06, "loss": 0.11493091583251953, "memory(GiB)": 77.59, "step": 100235, "token_acc": 0.9797297297297297, "train_speed(iter/s)": 1.460996 }, { "epoch": 4.294588920783172, "grad_norm": 3.5401644706726074, "learning_rate": 4.831288667255401e-06, "loss": 0.5309223175048828, "memory(GiB)": 77.59, "step": 100240, "token_acc": 0.9003831417624522, "train_speed(iter/s)": 1.461005 }, { "epoch": 4.29480313611242, "grad_norm": 1.3376479148864746, "learning_rate": 4.82840299216959e-06, "loss": 0.26873929500579835, "memory(GiB)": 77.59, "step": 100245, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.461016 }, { "epoch": 4.295017351441669, "grad_norm": 2.6709330081939697, "learning_rate": 4.8255181354096345e-06, "loss": 0.18372204303741455, "memory(GiB)": 77.59, "step": 100250, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.46102 }, { "epoch": 4.295231566770918, "grad_norm": 4.941420555114746, "learning_rate": 4.822634097027789e-06, "loss": 0.6161649703979493, "memory(GiB)": 77.59, "step": 100255, "token_acc": 0.8692579505300353, "train_speed(iter/s)": 1.461027 }, { "epoch": 4.295445782100167, "grad_norm": 1.404225468635559, "learning_rate": 4.819750877076301e-06, "loss": 0.22935953140258789, "memory(GiB)": 77.59, "step": 100260, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.461034 }, { "epoch": 4.295659997429416, "grad_norm": 2.388123035430908, "learning_rate": 4.816868475607411e-06, "loss": 0.2507441997528076, "memory(GiB)": 77.59, "step": 100265, "token_acc": 0.9506172839506173, "train_speed(iter/s)": 1.461047 }, { "epoch": 4.295874212758665, "grad_norm": 0.06864850223064423, "learning_rate": 4.813986892673339e-06, "loss": 0.4565541744232178, "memory(GiB)": 77.59, "step": 100270, "token_acc": 0.9154929577464789, "train_speed(iter/s)": 1.46105 }, { "epoch": 4.296088428087914, "grad_norm": 1.9201817512512207, "learning_rate": 4.811106128326281e-06, "loss": 0.3399663925170898, "memory(GiB)": 77.59, "step": 100275, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.461058 }, { "epoch": 4.296302643417163, "grad_norm": 2.032546043395996, "learning_rate": 4.808226182618431e-06, "loss": 0.22009963989257814, "memory(GiB)": 77.59, "step": 100280, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.46107 }, { "epoch": 4.296516858746412, "grad_norm": 0.2158326804637909, "learning_rate": 4.8053470556019555e-06, "loss": 0.2941922664642334, "memory(GiB)": 77.59, "step": 100285, "token_acc": 0.9418604651162791, "train_speed(iter/s)": 1.461069 }, { "epoch": 4.296731074075661, "grad_norm": 4.1693034172058105, "learning_rate": 4.802468747329003e-06, "loss": 0.3276723861694336, "memory(GiB)": 77.59, "step": 100290, "token_acc": 0.9206349206349206, "train_speed(iter/s)": 1.461089 }, { "epoch": 4.29694528940491, "grad_norm": 4.073807239532471, "learning_rate": 4.799591257851738e-06, "loss": 0.41713504791259765, "memory(GiB)": 77.59, "step": 100295, "token_acc": 0.9204545454545454, "train_speed(iter/s)": 1.46109 }, { "epoch": 4.297159504734159, "grad_norm": 2.8388915061950684, "learning_rate": 4.796714587222278e-06, "loss": 0.4311513900756836, "memory(GiB)": 77.59, "step": 100300, "token_acc": 0.9209726443768997, "train_speed(iter/s)": 1.461092 }, { "epoch": 4.2973737200634075, "grad_norm": 2.019962787628174, "learning_rate": 4.7938387354927396e-06, "loss": 0.33185243606567383, "memory(GiB)": 77.59, "step": 100305, "token_acc": 0.8989169675090253, "train_speed(iter/s)": 1.461102 }, { "epoch": 4.297587935392657, "grad_norm": 6.960842609405518, "learning_rate": 4.790963702715218e-06, "loss": 0.47779226303100586, "memory(GiB)": 77.59, "step": 100310, "token_acc": 0.906896551724138, "train_speed(iter/s)": 1.461104 }, { "epoch": 4.297802150721906, "grad_norm": 3.614431142807007, "learning_rate": 4.788089488941788e-06, "loss": 0.4054481029510498, "memory(GiB)": 77.59, "step": 100315, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.461115 }, { "epoch": 4.298016366051154, "grad_norm": 3.4085071086883545, "learning_rate": 4.785216094224543e-06, "loss": 0.3685122489929199, "memory(GiB)": 77.59, "step": 100320, "token_acc": 0.9238754325259516, "train_speed(iter/s)": 1.461116 }, { "epoch": 4.298230581380404, "grad_norm": 4.252023220062256, "learning_rate": 4.782343518615517e-06, "loss": 0.2873069763183594, "memory(GiB)": 77.59, "step": 100325, "token_acc": 0.943609022556391, "train_speed(iter/s)": 1.46112 }, { "epoch": 4.298444796709653, "grad_norm": 0.23780351877212524, "learning_rate": 4.779471762166759e-06, "loss": 0.31420202255249025, "memory(GiB)": 77.59, "step": 100330, "token_acc": 0.9233226837060703, "train_speed(iter/s)": 1.46112 }, { "epoch": 4.298659012038901, "grad_norm": 1.2410835027694702, "learning_rate": 4.776600824930283e-06, "loss": 0.3056577444076538, "memory(GiB)": 77.59, "step": 100335, "token_acc": 0.91, "train_speed(iter/s)": 1.461129 }, { "epoch": 4.2988732273681505, "grad_norm": 0.6102287769317627, "learning_rate": 4.773730706958113e-06, "loss": 0.2771978616714478, "memory(GiB)": 77.59, "step": 100340, "token_acc": 0.9402985074626866, "train_speed(iter/s)": 1.46113 }, { "epoch": 4.2990874426974, "grad_norm": 9.000707626342773, "learning_rate": 4.770861408302235e-06, "loss": 0.4045722007751465, "memory(GiB)": 77.59, "step": 100345, "token_acc": 0.8958333333333334, "train_speed(iter/s)": 1.46114 }, { "epoch": 4.299301658026648, "grad_norm": 2.857699155807495, "learning_rate": 4.767992929014625e-06, "loss": 0.24372999668121337, "memory(GiB)": 77.59, "step": 100350, "token_acc": 0.9415584415584416, "train_speed(iter/s)": 1.46114 }, { "epoch": 4.299515873355897, "grad_norm": 1.8933675289154053, "learning_rate": 4.765125269147264e-06, "loss": 0.39977226257324217, "memory(GiB)": 77.59, "step": 100355, "token_acc": 0.8977635782747604, "train_speed(iter/s)": 1.461151 }, { "epoch": 4.299730088685147, "grad_norm": 5.041592597961426, "learning_rate": 4.762258428752092e-06, "loss": 0.23574502468109132, "memory(GiB)": 77.59, "step": 100360, "token_acc": 0.9440298507462687, "train_speed(iter/s)": 1.461155 }, { "epoch": 4.299944304014395, "grad_norm": 0.2198748141527176, "learning_rate": 4.759392407881041e-06, "loss": 0.21856930255889892, "memory(GiB)": 77.59, "step": 100365, "token_acc": 0.9503311258278145, "train_speed(iter/s)": 1.461155 }, { "epoch": 4.300158519343644, "grad_norm": 4.356518745422363, "learning_rate": 4.756527206586037e-06, "loss": 0.2642316102981567, "memory(GiB)": 77.59, "step": 100370, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.461171 }, { "epoch": 4.3003727346728935, "grad_norm": 3.028245210647583, "learning_rate": 4.7536628249189805e-06, "loss": 0.2689563751220703, "memory(GiB)": 77.59, "step": 100375, "token_acc": 0.9442379182156134, "train_speed(iter/s)": 1.461173 }, { "epoch": 4.300586950002142, "grad_norm": 2.2201085090637207, "learning_rate": 4.750799262931771e-06, "loss": 0.09055392742156983, "memory(GiB)": 77.59, "step": 100380, "token_acc": 0.9705882352941176, "train_speed(iter/s)": 1.461175 }, { "epoch": 4.300801165331391, "grad_norm": 1.9024966955184937, "learning_rate": 4.747936520676277e-06, "loss": 0.22392992973327636, "memory(GiB)": 77.59, "step": 100385, "token_acc": 0.9363636363636364, "train_speed(iter/s)": 1.461182 }, { "epoch": 4.30101538066064, "grad_norm": 3.1653530597686768, "learning_rate": 4.745074598204369e-06, "loss": 0.4649835109710693, "memory(GiB)": 77.59, "step": 100390, "token_acc": 0.8981132075471698, "train_speed(iter/s)": 1.461184 }, { "epoch": 4.301229595989889, "grad_norm": 1.671795129776001, "learning_rate": 4.742213495567882e-06, "loss": 0.18340083360671997, "memory(GiB)": 77.59, "step": 100395, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.461187 }, { "epoch": 4.301443811319138, "grad_norm": 3.554013252258301, "learning_rate": 4.739353212818659e-06, "loss": 0.3920236587524414, "memory(GiB)": 77.59, "step": 100400, "token_acc": 0.9020979020979021, "train_speed(iter/s)": 1.46119 }, { "epoch": 4.301658026648387, "grad_norm": 4.788381576538086, "learning_rate": 4.736493750008497e-06, "loss": 0.4030032157897949, "memory(GiB)": 77.59, "step": 100405, "token_acc": 0.9116465863453815, "train_speed(iter/s)": 1.461193 }, { "epoch": 4.301872241977636, "grad_norm": 0.7231872081756592, "learning_rate": 4.7336351071892105e-06, "loss": 0.20106472969055175, "memory(GiB)": 77.59, "step": 100410, "token_acc": 0.966542750929368, "train_speed(iter/s)": 1.461205 }, { "epoch": 4.302086457306885, "grad_norm": 1.888725757598877, "learning_rate": 4.730777284412596e-06, "loss": 0.4521568775177002, "memory(GiB)": 77.59, "step": 100415, "token_acc": 0.9108635097493036, "train_speed(iter/s)": 1.461212 }, { "epoch": 4.302300672636134, "grad_norm": 2.604102611541748, "learning_rate": 4.727920281730425e-06, "loss": 0.3216772794723511, "memory(GiB)": 77.59, "step": 100420, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.461217 }, { "epoch": 4.3025148879653825, "grad_norm": 1.7964081764221191, "learning_rate": 4.7250640991944375e-06, "loss": 0.2470013380050659, "memory(GiB)": 77.59, "step": 100425, "token_acc": 0.9425675675675675, "train_speed(iter/s)": 1.461216 }, { "epoch": 4.302729103294632, "grad_norm": 5.310136318206787, "learning_rate": 4.722208736856387e-06, "loss": 0.523933219909668, "memory(GiB)": 77.59, "step": 100430, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.461218 }, { "epoch": 4.302943318623881, "grad_norm": 0.33963480591773987, "learning_rate": 4.719354194767994e-06, "loss": 0.24396934509277343, "memory(GiB)": 77.59, "step": 100435, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.461226 }, { "epoch": 4.303157533953129, "grad_norm": 3.827596426010132, "learning_rate": 4.716500472980983e-06, "loss": 0.2562158823013306, "memory(GiB)": 77.59, "step": 100440, "token_acc": 0.945054945054945, "train_speed(iter/s)": 1.461232 }, { "epoch": 4.303371749282379, "grad_norm": 5.252817630767822, "learning_rate": 4.713647571547048e-06, "loss": 0.4651162147521973, "memory(GiB)": 77.59, "step": 100445, "token_acc": 0.8903654485049833, "train_speed(iter/s)": 1.461236 }, { "epoch": 4.303585964611628, "grad_norm": 3.289031982421875, "learning_rate": 4.710795490517861e-06, "loss": 0.6308228492736816, "memory(GiB)": 77.59, "step": 100450, "token_acc": 0.8468468468468469, "train_speed(iter/s)": 1.461245 }, { "epoch": 4.303800179940876, "grad_norm": 4.205815315246582, "learning_rate": 4.707944229945105e-06, "loss": 0.4784702777862549, "memory(GiB)": 77.59, "step": 100455, "token_acc": 0.924187725631769, "train_speed(iter/s)": 1.461249 }, { "epoch": 4.3040143952701255, "grad_norm": 2.253157615661621, "learning_rate": 4.705093789880416e-06, "loss": 0.3500782012939453, "memory(GiB)": 77.59, "step": 100460, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.461251 }, { "epoch": 4.304228610599375, "grad_norm": 3.9922096729278564, "learning_rate": 4.702244170375453e-06, "loss": 0.4313839912414551, "memory(GiB)": 77.59, "step": 100465, "token_acc": 0.9155405405405406, "train_speed(iter/s)": 1.461262 }, { "epoch": 4.304442825928623, "grad_norm": 2.539498805999756, "learning_rate": 4.699395371481829e-06, "loss": 0.26502768993377684, "memory(GiB)": 77.59, "step": 100470, "token_acc": 0.9387755102040817, "train_speed(iter/s)": 1.461265 }, { "epoch": 4.304657041257872, "grad_norm": 3.60587215423584, "learning_rate": 4.696547393251155e-06, "loss": 0.33789238929748533, "memory(GiB)": 77.59, "step": 100475, "token_acc": 0.9233128834355828, "train_speed(iter/s)": 1.461268 }, { "epoch": 4.304871256587122, "grad_norm": 1.1419280767440796, "learning_rate": 4.693700235735021e-06, "loss": 0.5283544540405274, "memory(GiB)": 77.59, "step": 100480, "token_acc": 0.8817891373801917, "train_speed(iter/s)": 1.461267 }, { "epoch": 4.30508547191637, "grad_norm": 4.242347240447998, "learning_rate": 4.690853898985004e-06, "loss": 0.40081019401550294, "memory(GiB)": 77.59, "step": 100485, "token_acc": 0.916955017301038, "train_speed(iter/s)": 1.461278 }, { "epoch": 4.305299687245619, "grad_norm": 1.0340981483459473, "learning_rate": 4.688008383052672e-06, "loss": 0.14282019138336183, "memory(GiB)": 77.59, "step": 100490, "token_acc": 0.9704797047970479, "train_speed(iter/s)": 1.461279 }, { "epoch": 4.3055139025748685, "grad_norm": 2.7789015769958496, "learning_rate": 4.685163687989569e-06, "loss": 0.45093650817871095, "memory(GiB)": 77.59, "step": 100495, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.461289 }, { "epoch": 4.305728117904117, "grad_norm": 3.376027822494507, "learning_rate": 4.682319813847247e-06, "loss": 0.408339786529541, "memory(GiB)": 77.59, "step": 100500, "token_acc": 0.8996655518394648, "train_speed(iter/s)": 1.461289 }, { "epoch": 4.305728117904117, "eval_loss": 2.4880826473236084, "eval_runtime": 11.5705, "eval_samples_per_second": 8.643, "eval_steps_per_second": 8.643, "eval_token_acc": 0.45092460881934565, "step": 100500 }, { "epoch": 4.305942333233366, "grad_norm": 2.458803653717041, "learning_rate": 4.679476760677209e-06, "loss": 0.5391776084899902, "memory(GiB)": 77.59, "step": 100505, "token_acc": 0.5770392749244713, "train_speed(iter/s)": 1.461034 }, { "epoch": 4.306156548562615, "grad_norm": 3.0457122325897217, "learning_rate": 4.676634528530965e-06, "loss": 0.29138336181640623, "memory(GiB)": 77.59, "step": 100510, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.461035 }, { "epoch": 4.306370763891864, "grad_norm": 4.480109214782715, "learning_rate": 4.673793117460007e-06, "loss": 0.38513622283935545, "memory(GiB)": 77.59, "step": 100515, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.46104 }, { "epoch": 4.306584979221113, "grad_norm": 2.2003836631774902, "learning_rate": 4.670952527515793e-06, "loss": 0.14810459613800048, "memory(GiB)": 77.59, "step": 100520, "token_acc": 0.9689265536723164, "train_speed(iter/s)": 1.461045 }, { "epoch": 4.306799194550362, "grad_norm": 2.0506532192230225, "learning_rate": 4.668112758749804e-06, "loss": 0.2755594730377197, "memory(GiB)": 77.59, "step": 100525, "token_acc": 0.9562043795620438, "train_speed(iter/s)": 1.46105 }, { "epoch": 4.307013409879611, "grad_norm": 2.137821912765503, "learning_rate": 4.665273811213478e-06, "loss": 0.2759788990020752, "memory(GiB)": 77.59, "step": 100530, "token_acc": 0.945054945054945, "train_speed(iter/s)": 1.461056 }, { "epoch": 4.30722762520886, "grad_norm": 6.978790760040283, "learning_rate": 4.662435684958244e-06, "loss": 0.6287755489349365, "memory(GiB)": 77.59, "step": 100535, "token_acc": 0.8864468864468864, "train_speed(iter/s)": 1.461056 }, { "epoch": 4.307441840538109, "grad_norm": 2.7239739894866943, "learning_rate": 4.659598380035518e-06, "loss": 0.2537358522415161, "memory(GiB)": 77.59, "step": 100540, "token_acc": 0.9216417910447762, "train_speed(iter/s)": 1.461058 }, { "epoch": 4.307656055867358, "grad_norm": 6.027720928192139, "learning_rate": 4.656761896496703e-06, "loss": 0.6760228157043457, "memory(GiB)": 77.59, "step": 100545, "token_acc": 0.8645833333333334, "train_speed(iter/s)": 1.461056 }, { "epoch": 4.307870271196607, "grad_norm": 2.018143653869629, "learning_rate": 4.653926234393169e-06, "loss": 0.1510690450668335, "memory(GiB)": 77.59, "step": 100550, "token_acc": 0.9682539682539683, "train_speed(iter/s)": 1.461059 }, { "epoch": 4.308084486525856, "grad_norm": 4.950108051300049, "learning_rate": 4.651091393776308e-06, "loss": 0.3131261348724365, "memory(GiB)": 77.59, "step": 100555, "token_acc": 0.9315589353612167, "train_speed(iter/s)": 1.461065 }, { "epoch": 4.308298701855104, "grad_norm": 0.2865028977394104, "learning_rate": 4.648257374697462e-06, "loss": 0.2355654239654541, "memory(GiB)": 77.59, "step": 100560, "token_acc": 0.9366666666666666, "train_speed(iter/s)": 1.461071 }, { "epoch": 4.308512917184354, "grad_norm": 1.1695367097854614, "learning_rate": 4.645424177207985e-06, "loss": 0.34927024841308596, "memory(GiB)": 77.59, "step": 100565, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.461075 }, { "epoch": 4.308727132513603, "grad_norm": 4.2099714279174805, "learning_rate": 4.6425918013591916e-06, "loss": 0.3285334825515747, "memory(GiB)": 77.59, "step": 100570, "token_acc": 0.9395017793594306, "train_speed(iter/s)": 1.461078 }, { "epoch": 4.308941347842851, "grad_norm": 4.817563056945801, "learning_rate": 4.639760247202396e-06, "loss": 0.3963630676269531, "memory(GiB)": 77.59, "step": 100575, "token_acc": 0.8932384341637011, "train_speed(iter/s)": 1.461092 }, { "epoch": 4.309155563172101, "grad_norm": 2.385380983352661, "learning_rate": 4.636929514788891e-06, "loss": 0.3321181058883667, "memory(GiB)": 77.59, "step": 100580, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.461094 }, { "epoch": 4.30936977850135, "grad_norm": 2.03545880317688, "learning_rate": 4.6340996041699665e-06, "loss": 0.33314459323883056, "memory(GiB)": 77.59, "step": 100585, "token_acc": 0.9363057324840764, "train_speed(iter/s)": 1.461106 }, { "epoch": 4.309583993830598, "grad_norm": 4.064988136291504, "learning_rate": 4.631270515396891e-06, "loss": 0.3984776258468628, "memory(GiB)": 77.59, "step": 100590, "token_acc": 0.925, "train_speed(iter/s)": 1.461104 }, { "epoch": 4.3097982091598475, "grad_norm": 2.524524688720703, "learning_rate": 4.628442248520904e-06, "loss": 0.2597846508026123, "memory(GiB)": 77.59, "step": 100595, "token_acc": 0.9328859060402684, "train_speed(iter/s)": 1.461105 }, { "epoch": 4.310012424489097, "grad_norm": 4.012593746185303, "learning_rate": 4.625614803593248e-06, "loss": 0.45181427001953123, "memory(GiB)": 77.59, "step": 100600, "token_acc": 0.8938356164383562, "train_speed(iter/s)": 1.461105 }, { "epoch": 4.310226639818345, "grad_norm": 3.522904872894287, "learning_rate": 4.622788180665133e-06, "loss": 0.4781993865966797, "memory(GiB)": 77.59, "step": 100605, "token_acc": 0.9351535836177475, "train_speed(iter/s)": 1.461113 }, { "epoch": 4.310440855147594, "grad_norm": 5.8968377113342285, "learning_rate": 4.61996237978779e-06, "loss": 0.14647285938262938, "memory(GiB)": 77.59, "step": 100610, "token_acc": 0.9656488549618321, "train_speed(iter/s)": 1.46111 }, { "epoch": 4.310655070476844, "grad_norm": 2.569706439971924, "learning_rate": 4.6171374010123945e-06, "loss": 0.49536700248718263, "memory(GiB)": 77.59, "step": 100615, "token_acc": 0.896969696969697, "train_speed(iter/s)": 1.461113 }, { "epoch": 4.310869285806092, "grad_norm": 2.083752393722534, "learning_rate": 4.614313244390133e-06, "loss": 0.24295647144317628, "memory(GiB)": 77.59, "step": 100620, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.461117 }, { "epoch": 4.311083501135341, "grad_norm": 3.7356550693511963, "learning_rate": 4.611489909972161e-06, "loss": 0.25446391105651855, "memory(GiB)": 77.59, "step": 100625, "token_acc": 0.9446494464944649, "train_speed(iter/s)": 1.46112 }, { "epoch": 4.3112977164645905, "grad_norm": 1.9047225713729858, "learning_rate": 4.6086673978096125e-06, "loss": 0.35773801803588867, "memory(GiB)": 77.59, "step": 100630, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.461122 }, { "epoch": 4.311511931793839, "grad_norm": 1.7162189483642578, "learning_rate": 4.605845707953649e-06, "loss": 0.5305943012237548, "memory(GiB)": 77.59, "step": 100635, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.461135 }, { "epoch": 4.311726147123088, "grad_norm": 3.109921455383301, "learning_rate": 4.603024840455367e-06, "loss": 0.6071090698242188, "memory(GiB)": 77.59, "step": 100640, "token_acc": 0.8725490196078431, "train_speed(iter/s)": 1.461142 }, { "epoch": 4.311940362452337, "grad_norm": 1.615151047706604, "learning_rate": 4.600204795365881e-06, "loss": 0.3243889331817627, "memory(GiB)": 77.59, "step": 100645, "token_acc": 0.9311377245508982, "train_speed(iter/s)": 1.461146 }, { "epoch": 4.312154577781586, "grad_norm": 1.6289187669754028, "learning_rate": 4.597385572736273e-06, "loss": 0.23530447483062744, "memory(GiB)": 77.59, "step": 100650, "token_acc": 0.9321428571428572, "train_speed(iter/s)": 1.461146 }, { "epoch": 4.312368793110835, "grad_norm": 3.0427262783050537, "learning_rate": 4.59456717261762e-06, "loss": 0.2984498977661133, "memory(GiB)": 77.59, "step": 100655, "token_acc": 0.9370860927152318, "train_speed(iter/s)": 1.461152 }, { "epoch": 4.312583008440084, "grad_norm": 1.813575267791748, "learning_rate": 4.591749595060974e-06, "loss": 0.22679240703582765, "memory(GiB)": 77.59, "step": 100660, "token_acc": 0.9715189873417721, "train_speed(iter/s)": 1.461156 }, { "epoch": 4.312797223769333, "grad_norm": 2.778456926345825, "learning_rate": 4.588932840117366e-06, "loss": 0.19686053991317748, "memory(GiB)": 77.59, "step": 100665, "token_acc": 0.9451612903225807, "train_speed(iter/s)": 1.46116 }, { "epoch": 4.313011439098582, "grad_norm": 3.577562093734741, "learning_rate": 4.586116907837857e-06, "loss": 0.3107320785522461, "memory(GiB)": 77.59, "step": 100670, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.46116 }, { "epoch": 4.313225654427831, "grad_norm": 1.653132438659668, "learning_rate": 4.583301798273437e-06, "loss": 0.1895901918411255, "memory(GiB)": 77.59, "step": 100675, "token_acc": 0.9554794520547946, "train_speed(iter/s)": 1.461163 }, { "epoch": 4.3134398697570795, "grad_norm": 2.4563610553741455, "learning_rate": 4.580487511475112e-06, "loss": 0.14462391138076783, "memory(GiB)": 77.59, "step": 100680, "token_acc": 0.9570552147239264, "train_speed(iter/s)": 1.461166 }, { "epoch": 4.313654085086329, "grad_norm": 3.678021192550659, "learning_rate": 4.577674047493857e-06, "loss": 0.4390141010284424, "memory(GiB)": 77.59, "step": 100685, "token_acc": 0.9153225806451613, "train_speed(iter/s)": 1.461167 }, { "epoch": 4.313868300415578, "grad_norm": 3.23001766204834, "learning_rate": 4.574861406380654e-06, "loss": 0.35112462043762205, "memory(GiB)": 77.59, "step": 100690, "token_acc": 0.9355932203389831, "train_speed(iter/s)": 1.461172 }, { "epoch": 4.314082515744826, "grad_norm": 3.361239433288574, "learning_rate": 4.572049588186433e-06, "loss": 0.35380139350891116, "memory(GiB)": 77.59, "step": 100695, "token_acc": 0.921875, "train_speed(iter/s)": 1.461184 }, { "epoch": 4.314296731074076, "grad_norm": 3.0594863891601562, "learning_rate": 4.5692385929621604e-06, "loss": 0.28484060764312746, "memory(GiB)": 77.59, "step": 100700, "token_acc": 0.9418960244648318, "train_speed(iter/s)": 1.461189 }, { "epoch": 4.314510946403325, "grad_norm": 1.7511886358261108, "learning_rate": 4.566428420758739e-06, "loss": 0.24542388916015626, "memory(GiB)": 77.59, "step": 100705, "token_acc": 0.9442815249266863, "train_speed(iter/s)": 1.46119 }, { "epoch": 4.314725161732573, "grad_norm": 3.386218547821045, "learning_rate": 4.563619071627096e-06, "loss": 0.4286513328552246, "memory(GiB)": 77.59, "step": 100710, "token_acc": 0.9213836477987422, "train_speed(iter/s)": 1.461191 }, { "epoch": 4.3149393770618225, "grad_norm": 1.4254324436187744, "learning_rate": 4.560810545618116e-06, "loss": 0.4372988224029541, "memory(GiB)": 77.59, "step": 100715, "token_acc": 0.9368029739776952, "train_speed(iter/s)": 1.461196 }, { "epoch": 4.315153592391072, "grad_norm": 0.027389120310544968, "learning_rate": 4.558002842782677e-06, "loss": 0.24805657863616942, "memory(GiB)": 77.59, "step": 100720, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.461198 }, { "epoch": 4.31536780772032, "grad_norm": 1.221103549003601, "learning_rate": 4.5551959631716376e-06, "loss": 0.3351656436920166, "memory(GiB)": 77.59, "step": 100725, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.461203 }, { "epoch": 4.315582023049569, "grad_norm": 3.8684873580932617, "learning_rate": 4.552389906835863e-06, "loss": 0.6937335014343262, "memory(GiB)": 77.59, "step": 100730, "token_acc": 0.8657718120805369, "train_speed(iter/s)": 1.461206 }, { "epoch": 4.315796238378819, "grad_norm": 2.663374185562134, "learning_rate": 4.549584673826179e-06, "loss": 0.38849873542785646, "memory(GiB)": 77.59, "step": 100735, "token_acc": 0.9153094462540716, "train_speed(iter/s)": 1.461213 }, { "epoch": 4.316010453708067, "grad_norm": 1.5251408815383911, "learning_rate": 4.546780264193406e-06, "loss": 0.14507675170898438, "memory(GiB)": 77.59, "step": 100740, "token_acc": 0.972809667673716, "train_speed(iter/s)": 1.461212 }, { "epoch": 4.316224669037316, "grad_norm": 2.2470109462738037, "learning_rate": 4.543976677988343e-06, "loss": 0.2901093244552612, "memory(GiB)": 77.59, "step": 100745, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.461212 }, { "epoch": 4.3164388843665655, "grad_norm": 2.9538323879241943, "learning_rate": 4.541173915261776e-06, "loss": 0.4041624069213867, "memory(GiB)": 77.59, "step": 100750, "token_acc": 0.90625, "train_speed(iter/s)": 1.461215 }, { "epoch": 4.316653099695814, "grad_norm": 5.533748626708984, "learning_rate": 4.538371976064499e-06, "loss": 0.6741133689880371, "memory(GiB)": 77.59, "step": 100755, "token_acc": 0.8802816901408451, "train_speed(iter/s)": 1.461215 }, { "epoch": 4.316867315025063, "grad_norm": 3.660484552383423, "learning_rate": 4.535570860447258e-06, "loss": 0.29795782566070556, "memory(GiB)": 77.59, "step": 100760, "token_acc": 0.941747572815534, "train_speed(iter/s)": 1.46122 }, { "epoch": 4.317081530354312, "grad_norm": 5.526294708251953, "learning_rate": 4.532770568460798e-06, "loss": 0.496384859085083, "memory(GiB)": 77.59, "step": 100765, "token_acc": 0.9140893470790378, "train_speed(iter/s)": 1.461219 }, { "epoch": 4.317295745683561, "grad_norm": 2.201631784439087, "learning_rate": 4.529971100155855e-06, "loss": 0.5056455612182618, "memory(GiB)": 77.59, "step": 100770, "token_acc": 0.8878787878787879, "train_speed(iter/s)": 1.461239 }, { "epoch": 4.31750996101281, "grad_norm": 6.0772199630737305, "learning_rate": 4.52717245558314e-06, "loss": 0.3601662635803223, "memory(GiB)": 77.59, "step": 100775, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.461242 }, { "epoch": 4.317724176342059, "grad_norm": 3.213275671005249, "learning_rate": 4.524374634793338e-06, "loss": 0.23218610286712646, "memory(GiB)": 77.59, "step": 100780, "token_acc": 0.9604863221884499, "train_speed(iter/s)": 1.461246 }, { "epoch": 4.317938391671308, "grad_norm": 11.824914932250977, "learning_rate": 4.521577637837154e-06, "loss": 0.2628467559814453, "memory(GiB)": 77.59, "step": 100785, "token_acc": 0.9452054794520548, "train_speed(iter/s)": 1.461268 }, { "epoch": 4.318152607000557, "grad_norm": 3.34602689743042, "learning_rate": 4.518781464765259e-06, "loss": 0.4136794567108154, "memory(GiB)": 77.59, "step": 100790, "token_acc": 0.9242902208201893, "train_speed(iter/s)": 1.461268 }, { "epoch": 4.318366822329806, "grad_norm": 2.5231573581695557, "learning_rate": 4.515986115628306e-06, "loss": 0.1548193097114563, "memory(GiB)": 77.59, "step": 100795, "token_acc": 0.9678571428571429, "train_speed(iter/s)": 1.46127 }, { "epoch": 4.3185810376590545, "grad_norm": 2.319842576980591, "learning_rate": 4.513191590476934e-06, "loss": 0.10340893268585205, "memory(GiB)": 77.59, "step": 100800, "token_acc": 0.9795221843003413, "train_speed(iter/s)": 1.461273 }, { "epoch": 4.318795252988304, "grad_norm": 4.0338134765625, "learning_rate": 4.510397889361761e-06, "loss": 0.32441067695617676, "memory(GiB)": 77.59, "step": 100805, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.461276 }, { "epoch": 4.319009468317553, "grad_norm": 0.12153308838605881, "learning_rate": 4.507605012333394e-06, "loss": 0.18989487886428832, "memory(GiB)": 77.59, "step": 100810, "token_acc": 0.969811320754717, "train_speed(iter/s)": 1.461281 }, { "epoch": 4.319223683646801, "grad_norm": 3.710505723953247, "learning_rate": 4.504812959442451e-06, "loss": 0.3857694387435913, "memory(GiB)": 77.59, "step": 100815, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.461291 }, { "epoch": 4.319437898976051, "grad_norm": 2.560133218765259, "learning_rate": 4.502021730739497e-06, "loss": 0.2843024492263794, "memory(GiB)": 77.59, "step": 100820, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.461291 }, { "epoch": 4.3196521143053, "grad_norm": 2.2524871826171875, "learning_rate": 4.4992313262751e-06, "loss": 0.22841808795928956, "memory(GiB)": 77.59, "step": 100825, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.461295 }, { "epoch": 4.319866329634548, "grad_norm": 3.7139511108398438, "learning_rate": 4.496441746099811e-06, "loss": 0.32918384075164797, "memory(GiB)": 77.59, "step": 100830, "token_acc": 0.9243421052631579, "train_speed(iter/s)": 1.461298 }, { "epoch": 4.320080544963798, "grad_norm": 1.4885276556015015, "learning_rate": 4.493652990264152e-06, "loss": 0.31391952037811277, "memory(GiB)": 77.59, "step": 100835, "token_acc": 0.9389067524115756, "train_speed(iter/s)": 1.461312 }, { "epoch": 4.320294760293047, "grad_norm": 1.9590120315551758, "learning_rate": 4.4908650588186715e-06, "loss": 0.5382940292358398, "memory(GiB)": 77.59, "step": 100840, "token_acc": 0.8912280701754386, "train_speed(iter/s)": 1.461321 }, { "epoch": 4.320508975622295, "grad_norm": 0.03482648730278015, "learning_rate": 4.488077951813863e-06, "loss": 0.5717967033386231, "memory(GiB)": 77.59, "step": 100845, "token_acc": 0.9, "train_speed(iter/s)": 1.461325 }, { "epoch": 4.3207231909515444, "grad_norm": 0.31409573554992676, "learning_rate": 4.4852916693002124e-06, "loss": 0.4753856658935547, "memory(GiB)": 77.59, "step": 100850, "token_acc": 0.8929765886287625, "train_speed(iter/s)": 1.461324 }, { "epoch": 4.320937406280794, "grad_norm": 1.6985585689544678, "learning_rate": 4.482506211328191e-06, "loss": 0.34358768463134765, "memory(GiB)": 77.59, "step": 100855, "token_acc": 0.9308943089430894, "train_speed(iter/s)": 1.461326 }, { "epoch": 4.321151621610042, "grad_norm": 3.444448232650757, "learning_rate": 4.479721577948276e-06, "loss": 0.32488360404968264, "memory(GiB)": 77.59, "step": 100860, "token_acc": 0.9433333333333334, "train_speed(iter/s)": 1.461331 }, { "epoch": 4.321365836939291, "grad_norm": 1.892751693725586, "learning_rate": 4.476937769210909e-06, "loss": 0.14936407804489135, "memory(GiB)": 77.59, "step": 100865, "token_acc": 0.9605263157894737, "train_speed(iter/s)": 1.46135 }, { "epoch": 4.321580052268541, "grad_norm": 3.02032732963562, "learning_rate": 4.474154785166506e-06, "loss": 0.24979963302612304, "memory(GiB)": 77.59, "step": 100870, "token_acc": 0.9320987654320988, "train_speed(iter/s)": 1.461354 }, { "epoch": 4.321794267597789, "grad_norm": 0.3116592764854431, "learning_rate": 4.471372625865511e-06, "loss": 0.5007205963134765, "memory(GiB)": 77.59, "step": 100875, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.461361 }, { "epoch": 4.322008482927038, "grad_norm": 4.091840744018555, "learning_rate": 4.468591291358304e-06, "loss": 0.19928252696990967, "memory(GiB)": 77.59, "step": 100880, "token_acc": 0.9466666666666667, "train_speed(iter/s)": 1.461367 }, { "epoch": 4.3222226982562875, "grad_norm": 2.176557779312134, "learning_rate": 4.465810781695279e-06, "loss": 0.5900081157684326, "memory(GiB)": 77.59, "step": 100885, "token_acc": 0.8726114649681529, "train_speed(iter/s)": 1.461369 }, { "epoch": 4.322436913585536, "grad_norm": 1.7537301778793335, "learning_rate": 4.463031096926806e-06, "loss": 0.22035224437713624, "memory(GiB)": 77.59, "step": 100890, "token_acc": 0.948220064724919, "train_speed(iter/s)": 1.461371 }, { "epoch": 4.322651128914785, "grad_norm": 3.1231815814971924, "learning_rate": 4.460252237103235e-06, "loss": 0.22144043445587158, "memory(GiB)": 77.59, "step": 100895, "token_acc": 0.9320388349514563, "train_speed(iter/s)": 1.461372 }, { "epoch": 4.322865344244034, "grad_norm": 3.207777261734009, "learning_rate": 4.457474202274925e-06, "loss": 0.16970479488372803, "memory(GiB)": 77.59, "step": 100900, "token_acc": 0.9664179104477612, "train_speed(iter/s)": 1.461378 }, { "epoch": 4.323079559573283, "grad_norm": 3.84899640083313, "learning_rate": 4.454696992492185e-06, "loss": 0.5173405647277832, "memory(GiB)": 77.59, "step": 100905, "token_acc": 0.8754716981132076, "train_speed(iter/s)": 1.461385 }, { "epoch": 4.323293774902532, "grad_norm": 3.0358824729919434, "learning_rate": 4.451920607805343e-06, "loss": 0.19220938682556152, "memory(GiB)": 77.59, "step": 100910, "token_acc": 0.95, "train_speed(iter/s)": 1.461387 }, { "epoch": 4.323507990231781, "grad_norm": 3.6651713848114014, "learning_rate": 4.44914504826468e-06, "loss": 0.39980196952819824, "memory(GiB)": 77.59, "step": 100915, "token_acc": 0.890728476821192, "train_speed(iter/s)": 1.46139 }, { "epoch": 4.32372220556103, "grad_norm": 0.5956964492797852, "learning_rate": 4.446370313920489e-06, "loss": 0.07514835000038148, "memory(GiB)": 77.59, "step": 100920, "token_acc": 0.982078853046595, "train_speed(iter/s)": 1.461405 }, { "epoch": 4.323936420890279, "grad_norm": 0.4710063636302948, "learning_rate": 4.44359640482302e-06, "loss": 0.43165016174316406, "memory(GiB)": 77.59, "step": 100925, "token_acc": 0.91015625, "train_speed(iter/s)": 1.461409 }, { "epoch": 4.324150636219528, "grad_norm": 1.9668569564819336, "learning_rate": 4.440823321022541e-06, "loss": 0.4305275917053223, "memory(GiB)": 77.59, "step": 100930, "token_acc": 0.8962962962962963, "train_speed(iter/s)": 1.461408 }, { "epoch": 4.3243648515487765, "grad_norm": 0.2992168962955475, "learning_rate": 4.438051062569293e-06, "loss": 0.17496339082717896, "memory(GiB)": 77.59, "step": 100935, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.461412 }, { "epoch": 4.324579066878026, "grad_norm": 2.2400612831115723, "learning_rate": 4.435279629513489e-06, "loss": 0.37470521926879885, "memory(GiB)": 77.59, "step": 100940, "token_acc": 0.9256965944272446, "train_speed(iter/s)": 1.461416 }, { "epoch": 4.324793282207275, "grad_norm": 6.193637371063232, "learning_rate": 4.432509021905334e-06, "loss": 0.5905871868133545, "memory(GiB)": 77.59, "step": 100945, "token_acc": 0.86, "train_speed(iter/s)": 1.461415 }, { "epoch": 4.325007497536523, "grad_norm": 1.2662891149520874, "learning_rate": 4.4297392397950334e-06, "loss": 0.3136629819869995, "memory(GiB)": 77.59, "step": 100950, "token_acc": 0.9220338983050848, "train_speed(iter/s)": 1.461418 }, { "epoch": 4.325221712865773, "grad_norm": 4.041921138763428, "learning_rate": 4.42697028323274e-06, "loss": 0.35533857345581055, "memory(GiB)": 77.59, "step": 100955, "token_acc": 0.9370860927152318, "train_speed(iter/s)": 1.46142 }, { "epoch": 4.325435928195022, "grad_norm": 3.7638750076293945, "learning_rate": 4.424202152268642e-06, "loss": 0.5863192081451416, "memory(GiB)": 77.59, "step": 100960, "token_acc": 0.89198606271777, "train_speed(iter/s)": 1.461428 }, { "epoch": 4.32565014352427, "grad_norm": 3.248467445373535, "learning_rate": 4.421434846952871e-06, "loss": 0.27729287147521975, "memory(GiB)": 77.59, "step": 100965, "token_acc": 0.9232954545454546, "train_speed(iter/s)": 1.461435 }, { "epoch": 4.3258643588535195, "grad_norm": 2.6861860752105713, "learning_rate": 4.4186683673355714e-06, "loss": 0.31302316188812257, "memory(GiB)": 77.59, "step": 100970, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.46144 }, { "epoch": 4.326078574182769, "grad_norm": 3.7817599773406982, "learning_rate": 4.415902713466846e-06, "loss": 0.19282009601593017, "memory(GiB)": 77.59, "step": 100975, "token_acc": 0.9405594405594405, "train_speed(iter/s)": 1.46144 }, { "epoch": 4.326292789512017, "grad_norm": 3.8134286403656006, "learning_rate": 4.413137885396795e-06, "loss": 0.3722784996032715, "memory(GiB)": 77.59, "step": 100980, "token_acc": 0.9150326797385621, "train_speed(iter/s)": 1.46144 }, { "epoch": 4.326507004841266, "grad_norm": 3.002703905105591, "learning_rate": 4.4103738831755285e-06, "loss": 0.166592800617218, "memory(GiB)": 77.59, "step": 100985, "token_acc": 0.9562043795620438, "train_speed(iter/s)": 1.461442 }, { "epoch": 4.326721220170516, "grad_norm": 1.4480364322662354, "learning_rate": 4.407610706853105e-06, "loss": 0.24429764747619628, "memory(GiB)": 77.59, "step": 100990, "token_acc": 0.9487870619946092, "train_speed(iter/s)": 1.461451 }, { "epoch": 4.326935435499764, "grad_norm": 4.180449485778809, "learning_rate": 4.404848356479585e-06, "loss": 0.4284348011016846, "memory(GiB)": 77.59, "step": 100995, "token_acc": 0.9121621621621622, "train_speed(iter/s)": 1.461452 }, { "epoch": 4.327149650829013, "grad_norm": 2.9180660247802734, "learning_rate": 4.402086832104996e-06, "loss": 0.18197245597839357, "memory(GiB)": 77.59, "step": 101000, "token_acc": 0.9703389830508474, "train_speed(iter/s)": 1.461452 }, { "epoch": 4.327149650829013, "eval_loss": 2.3369650840759277, "eval_runtime": 11.5046, "eval_samples_per_second": 8.692, "eval_steps_per_second": 8.692, "eval_token_acc": 0.48021108179419525, "step": 101000 }, { "epoch": 4.3273638661582625, "grad_norm": 4.092196464538574, "learning_rate": 4.399326133779386e-06, "loss": 0.3418461322784424, "memory(GiB)": 77.59, "step": 101005, "token_acc": 0.6125703564727955, "train_speed(iter/s)": 1.461188 }, { "epoch": 4.327578081487511, "grad_norm": 3.2175323963165283, "learning_rate": 4.396566261552765e-06, "loss": 0.28709983825683594, "memory(GiB)": 77.59, "step": 101010, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.461196 }, { "epoch": 4.32779229681676, "grad_norm": 2.850590705871582, "learning_rate": 4.393807215475115e-06, "loss": 0.29312796592712403, "memory(GiB)": 77.59, "step": 101015, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.4612 }, { "epoch": 4.328006512146009, "grad_norm": 4.095639705657959, "learning_rate": 4.3910489955964365e-06, "loss": 0.31951701641082764, "memory(GiB)": 77.59, "step": 101020, "token_acc": 0.9298780487804879, "train_speed(iter/s)": 1.461219 }, { "epoch": 4.328220727475258, "grad_norm": 4.6544013023376465, "learning_rate": 4.388291601966688e-06, "loss": 0.36851327419281005, "memory(GiB)": 77.59, "step": 101025, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.461227 }, { "epoch": 4.328434942804507, "grad_norm": 4.440948486328125, "learning_rate": 4.385535034635829e-06, "loss": 0.6329166889190674, "memory(GiB)": 77.59, "step": 101030, "token_acc": 0.8690095846645367, "train_speed(iter/s)": 1.461227 }, { "epoch": 4.328649158133756, "grad_norm": 0.9040910005569458, "learning_rate": 4.382779293653788e-06, "loss": 0.26689374446868896, "memory(GiB)": 77.59, "step": 101035, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.461238 }, { "epoch": 4.328863373463005, "grad_norm": 2.4270377159118652, "learning_rate": 4.380024379070491e-06, "loss": 0.6035255432128906, "memory(GiB)": 77.59, "step": 101040, "token_acc": 0.875, "train_speed(iter/s)": 1.461248 }, { "epoch": 4.329077588792254, "grad_norm": 2.821573495864868, "learning_rate": 4.377270290935848e-06, "loss": 0.47306199073791505, "memory(GiB)": 77.59, "step": 101045, "token_acc": 0.8936170212765957, "train_speed(iter/s)": 1.461251 }, { "epoch": 4.329291804121503, "grad_norm": 2.6380982398986816, "learning_rate": 4.374517029299757e-06, "loss": 0.16778790950775146, "memory(GiB)": 77.59, "step": 101050, "token_acc": 0.9575163398692811, "train_speed(iter/s)": 1.461256 }, { "epoch": 4.3295060194507515, "grad_norm": 2.576781988143921, "learning_rate": 4.3717645942120855e-06, "loss": 0.18458378314971924, "memory(GiB)": 77.59, "step": 101055, "token_acc": 0.9543973941368078, "train_speed(iter/s)": 1.46126 }, { "epoch": 4.329720234780001, "grad_norm": 2.379176616668701, "learning_rate": 4.369012985722704e-06, "loss": 0.2534276247024536, "memory(GiB)": 77.59, "step": 101060, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.461267 }, { "epoch": 4.32993445010925, "grad_norm": 3.1395390033721924, "learning_rate": 4.366262203881461e-06, "loss": 0.3719038963317871, "memory(GiB)": 77.59, "step": 101065, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.461284 }, { "epoch": 4.330148665438498, "grad_norm": 3.9730372428894043, "learning_rate": 4.363512248738172e-06, "loss": 0.3254352807998657, "memory(GiB)": 77.59, "step": 101070, "token_acc": 0.9295302013422819, "train_speed(iter/s)": 1.461282 }, { "epoch": 4.330362880767748, "grad_norm": 2.7663583755493164, "learning_rate": 4.3607631203426704e-06, "loss": 0.47878093719482423, "memory(GiB)": 77.59, "step": 101075, "token_acc": 0.9049079754601227, "train_speed(iter/s)": 1.461288 }, { "epoch": 4.330577096096997, "grad_norm": 3.6435797214508057, "learning_rate": 4.358014818744765e-06, "loss": 0.46086812019348145, "memory(GiB)": 77.59, "step": 101080, "token_acc": 0.8954703832752613, "train_speed(iter/s)": 1.461296 }, { "epoch": 4.330791311426245, "grad_norm": 2.1908936500549316, "learning_rate": 4.355267343994241e-06, "loss": 0.14663679599761964, "memory(GiB)": 77.59, "step": 101085, "token_acc": 0.96875, "train_speed(iter/s)": 1.4613 }, { "epoch": 4.3310055267554946, "grad_norm": 5.713663578033447, "learning_rate": 4.352520696140866e-06, "loss": 0.5883623123168945, "memory(GiB)": 77.59, "step": 101090, "token_acc": 0.8714285714285714, "train_speed(iter/s)": 1.461301 }, { "epoch": 4.331219742084744, "grad_norm": 5.22633695602417, "learning_rate": 4.349774875234397e-06, "loss": 0.32868423461914065, "memory(GiB)": 77.59, "step": 101095, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.461305 }, { "epoch": 4.331433957413992, "grad_norm": 1.7268236875534058, "learning_rate": 4.3470298813245715e-06, "loss": 0.232708740234375, "memory(GiB)": 77.59, "step": 101100, "token_acc": 0.9428571428571428, "train_speed(iter/s)": 1.461305 }, { "epoch": 4.331648172743241, "grad_norm": 3.7881410121917725, "learning_rate": 4.344285714461138e-06, "loss": 0.4365672588348389, "memory(GiB)": 77.59, "step": 101105, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.461306 }, { "epoch": 4.331862388072491, "grad_norm": 1.6282343864440918, "learning_rate": 4.341542374693791e-06, "loss": 0.23604695796966552, "memory(GiB)": 77.59, "step": 101110, "token_acc": 0.948948948948949, "train_speed(iter/s)": 1.461304 }, { "epoch": 4.332076603401739, "grad_norm": 3.4517922401428223, "learning_rate": 4.3387998620722405e-06, "loss": 0.3164109945297241, "memory(GiB)": 77.59, "step": 101115, "token_acc": 0.9322709163346613, "train_speed(iter/s)": 1.461304 }, { "epoch": 4.332290818730988, "grad_norm": 6.690794944763184, "learning_rate": 4.336058176646163e-06, "loss": 0.39873790740966797, "memory(GiB)": 77.59, "step": 101120, "token_acc": 0.9133858267716536, "train_speed(iter/s)": 1.461305 }, { "epoch": 4.332505034060238, "grad_norm": 3.809532642364502, "learning_rate": 4.3333173184652185e-06, "loss": 0.6285267829895019, "memory(GiB)": 77.59, "step": 101125, "token_acc": 0.8532818532818532, "train_speed(iter/s)": 1.461306 }, { "epoch": 4.332719249389486, "grad_norm": 1.9303268194198608, "learning_rate": 4.330577287579079e-06, "loss": 0.2089614152908325, "memory(GiB)": 77.59, "step": 101130, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.461307 }, { "epoch": 4.332933464718735, "grad_norm": 3.67143177986145, "learning_rate": 4.327838084037372e-06, "loss": 0.34692814350128176, "memory(GiB)": 77.59, "step": 101135, "token_acc": 0.9307359307359307, "train_speed(iter/s)": 1.461318 }, { "epoch": 4.3331476800479845, "grad_norm": 7.086150646209717, "learning_rate": 4.325099707889724e-06, "loss": 0.4419133186340332, "memory(GiB)": 77.59, "step": 101140, "token_acc": 0.9209621993127147, "train_speed(iter/s)": 1.461323 }, { "epoch": 4.333361895377233, "grad_norm": 4.5439453125, "learning_rate": 4.322362159185739e-06, "loss": 0.3322859764099121, "memory(GiB)": 77.59, "step": 101145, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.461326 }, { "epoch": 4.333576110706482, "grad_norm": 3.9867403507232666, "learning_rate": 4.319625437975006e-06, "loss": 0.3996554136276245, "memory(GiB)": 77.59, "step": 101150, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.461329 }, { "epoch": 4.333790326035731, "grad_norm": 2.2138020992279053, "learning_rate": 4.3168895443071124e-06, "loss": 0.26236352920532224, "memory(GiB)": 77.59, "step": 101155, "token_acc": 0.9327485380116959, "train_speed(iter/s)": 1.461334 }, { "epoch": 4.33400454136498, "grad_norm": 2.011296510696411, "learning_rate": 4.314154478231619e-06, "loss": 0.29014291763305666, "memory(GiB)": 77.59, "step": 101160, "token_acc": 0.9238410596026491, "train_speed(iter/s)": 1.461336 }, { "epoch": 4.334218756694229, "grad_norm": 3.845211982727051, "learning_rate": 4.3114202397980755e-06, "loss": 0.42386903762817385, "memory(GiB)": 77.59, "step": 101165, "token_acc": 0.9026548672566371, "train_speed(iter/s)": 1.461334 }, { "epoch": 4.334432972023478, "grad_norm": 4.447188377380371, "learning_rate": 4.308686829056019e-06, "loss": 0.6177269458770752, "memory(GiB)": 77.59, "step": 101170, "token_acc": 0.8529411764705882, "train_speed(iter/s)": 1.461341 }, { "epoch": 4.334647187352727, "grad_norm": 4.200685501098633, "learning_rate": 4.305954246054961e-06, "loss": 0.6228216171264649, "memory(GiB)": 77.59, "step": 101175, "token_acc": 0.8704225352112676, "train_speed(iter/s)": 1.461342 }, { "epoch": 4.334861402681976, "grad_norm": 3.989079713821411, "learning_rate": 4.303222490844405e-06, "loss": 0.13359297513961793, "memory(GiB)": 77.59, "step": 101180, "token_acc": 0.9662162162162162, "train_speed(iter/s)": 1.461348 }, { "epoch": 4.335075618011225, "grad_norm": 2.9821667671203613, "learning_rate": 4.30049156347383e-06, "loss": 0.4132709503173828, "memory(GiB)": 77.59, "step": 101185, "token_acc": 0.909967845659164, "train_speed(iter/s)": 1.461351 }, { "epoch": 4.3352898333404735, "grad_norm": 2.7293243408203125, "learning_rate": 4.297761463992733e-06, "loss": 0.1794515371322632, "memory(GiB)": 77.59, "step": 101190, "token_acc": 0.9610027855153204, "train_speed(iter/s)": 1.461355 }, { "epoch": 4.335504048669723, "grad_norm": 4.043247222900391, "learning_rate": 4.29503219245056e-06, "loss": 0.30770220756530764, "memory(GiB)": 77.59, "step": 101195, "token_acc": 0.9323076923076923, "train_speed(iter/s)": 1.461353 }, { "epoch": 4.335718263998972, "grad_norm": 0.9807121753692627, "learning_rate": 4.292303748896748e-06, "loss": 0.17382413148880005, "memory(GiB)": 77.59, "step": 101200, "token_acc": 0.9546925566343042, "train_speed(iter/s)": 1.461356 }, { "epoch": 4.33593247932822, "grad_norm": 8.940947532653809, "learning_rate": 4.289576133380735e-06, "loss": 0.20041978359222412, "memory(GiB)": 77.59, "step": 101205, "token_acc": 0.9629629629629629, "train_speed(iter/s)": 1.461351 }, { "epoch": 4.33614669465747, "grad_norm": 3.3888962268829346, "learning_rate": 4.286849345951932e-06, "loss": 0.43379878997802734, "memory(GiB)": 77.59, "step": 101210, "token_acc": 0.9214876033057852, "train_speed(iter/s)": 1.461351 }, { "epoch": 4.336360909986719, "grad_norm": 1.418843150138855, "learning_rate": 4.284123386659722e-06, "loss": 0.5246121406555175, "memory(GiB)": 77.59, "step": 101215, "token_acc": 0.865979381443299, "train_speed(iter/s)": 1.461352 }, { "epoch": 4.336575125315967, "grad_norm": 1.1188188791275024, "learning_rate": 4.281398255553515e-06, "loss": 0.25378539562225344, "memory(GiB)": 77.59, "step": 101220, "token_acc": 0.936, "train_speed(iter/s)": 1.46135 }, { "epoch": 4.3367893406452165, "grad_norm": 3.548412799835205, "learning_rate": 4.278673952682655e-06, "loss": 0.3517481327056885, "memory(GiB)": 77.59, "step": 101225, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.461353 }, { "epoch": 4.337003555974466, "grad_norm": 3.4636685848236084, "learning_rate": 4.275950478096513e-06, "loss": 0.2799858570098877, "memory(GiB)": 77.59, "step": 101230, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.461354 }, { "epoch": 4.337217771303714, "grad_norm": 4.226729393005371, "learning_rate": 4.273227831844423e-06, "loss": 0.2188413619995117, "memory(GiB)": 77.59, "step": 101235, "token_acc": 0.9490616621983914, "train_speed(iter/s)": 1.461354 }, { "epoch": 4.337431986632963, "grad_norm": 2.3312482833862305, "learning_rate": 4.2705060139757054e-06, "loss": 0.3043022632598877, "memory(GiB)": 77.59, "step": 101240, "token_acc": 0.9319526627218935, "train_speed(iter/s)": 1.461352 }, { "epoch": 4.337646201962213, "grad_norm": 3.1193666458129883, "learning_rate": 4.267785024539661e-06, "loss": 0.26630239486694335, "memory(GiB)": 77.59, "step": 101245, "token_acc": 0.9488188976377953, "train_speed(iter/s)": 1.461361 }, { "epoch": 4.337860417291461, "grad_norm": 2.0089504718780518, "learning_rate": 4.2650648635856e-06, "loss": 0.38605732917785646, "memory(GiB)": 77.59, "step": 101250, "token_acc": 0.9135446685878963, "train_speed(iter/s)": 1.461364 }, { "epoch": 4.33807463262071, "grad_norm": 0.5213606953620911, "learning_rate": 4.262345531162792e-06, "loss": 0.3323110818862915, "memory(GiB)": 77.59, "step": 101255, "token_acc": 0.9293286219081273, "train_speed(iter/s)": 1.461367 }, { "epoch": 4.3382888479499595, "grad_norm": 2.395845413208008, "learning_rate": 4.259627027320501e-06, "loss": 0.4846632957458496, "memory(GiB)": 77.59, "step": 101260, "token_acc": 0.8972431077694235, "train_speed(iter/s)": 1.461366 }, { "epoch": 4.338503063279208, "grad_norm": 5.6761064529418945, "learning_rate": 4.256909352107969e-06, "loss": 0.2949413299560547, "memory(GiB)": 77.59, "step": 101265, "token_acc": 0.9424460431654677, "train_speed(iter/s)": 1.461365 }, { "epoch": 4.338717278608457, "grad_norm": 6.103508949279785, "learning_rate": 4.254192505574433e-06, "loss": 0.5198281764984131, "memory(GiB)": 77.59, "step": 101270, "token_acc": 0.8916967509025271, "train_speed(iter/s)": 1.461364 }, { "epoch": 4.338931493937706, "grad_norm": 2.2528350353240967, "learning_rate": 4.2514764877691184e-06, "loss": 0.3870755910873413, "memory(GiB)": 77.59, "step": 101275, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.461364 }, { "epoch": 4.339145709266955, "grad_norm": 3.3554835319519043, "learning_rate": 4.2487612987412215e-06, "loss": 0.19294670820236207, "memory(GiB)": 77.59, "step": 101280, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.461367 }, { "epoch": 4.339359924596204, "grad_norm": 1.0437710285186768, "learning_rate": 4.246046938539933e-06, "loss": 0.2146972417831421, "memory(GiB)": 77.59, "step": 101285, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.461373 }, { "epoch": 4.339574139925453, "grad_norm": 0.9715976119041443, "learning_rate": 4.243333407214422e-06, "loss": 0.43425164222717283, "memory(GiB)": 77.59, "step": 101290, "token_acc": 0.9090909090909091, "train_speed(iter/s)": 1.461384 }, { "epoch": 4.339788355254702, "grad_norm": 3.6721549034118652, "learning_rate": 4.240620704813847e-06, "loss": 0.39865272045135497, "memory(GiB)": 77.59, "step": 101295, "token_acc": 0.9183006535947712, "train_speed(iter/s)": 1.461392 }, { "epoch": 4.340002570583951, "grad_norm": 3.5606722831726074, "learning_rate": 4.23790883138736e-06, "loss": 0.6541102886199951, "memory(GiB)": 77.59, "step": 101300, "token_acc": 0.8739255014326648, "train_speed(iter/s)": 1.461395 }, { "epoch": 4.3402167859132, "grad_norm": 2.5170204639434814, "learning_rate": 4.235197786984074e-06, "loss": 0.26635901927947997, "memory(GiB)": 77.59, "step": 101305, "token_acc": 0.9427710843373494, "train_speed(iter/s)": 1.461393 }, { "epoch": 4.3404310012424485, "grad_norm": 4.7524919509887695, "learning_rate": 4.232487571653115e-06, "loss": 0.3303654670715332, "memory(GiB)": 77.59, "step": 101310, "token_acc": 0.9405204460966543, "train_speed(iter/s)": 1.461394 }, { "epoch": 4.340645216571698, "grad_norm": 1.6354237794876099, "learning_rate": 4.229778185443583e-06, "loss": 0.23347349166870118, "memory(GiB)": 77.59, "step": 101315, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.461397 }, { "epoch": 4.340859431900947, "grad_norm": 3.2632179260253906, "learning_rate": 4.227069628404551e-06, "loss": 0.20184047222137452, "memory(GiB)": 77.59, "step": 101320, "token_acc": 0.933852140077821, "train_speed(iter/s)": 1.461408 }, { "epoch": 4.341073647230195, "grad_norm": 2.6580874919891357, "learning_rate": 4.224361900585094e-06, "loss": 0.2702394962310791, "memory(GiB)": 77.59, "step": 101325, "token_acc": 0.9464882943143813, "train_speed(iter/s)": 1.461419 }, { "epoch": 4.341287862559445, "grad_norm": 2.306924343109131, "learning_rate": 4.221655002034253e-06, "loss": 0.1778348207473755, "memory(GiB)": 77.59, "step": 101330, "token_acc": 0.9632107023411371, "train_speed(iter/s)": 1.461425 }, { "epoch": 4.341502077888694, "grad_norm": 3.6551356315612793, "learning_rate": 4.218948932801087e-06, "loss": 0.44995684623718263, "memory(GiB)": 77.59, "step": 101335, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.46143 }, { "epoch": 4.341716293217942, "grad_norm": 2.0845158100128174, "learning_rate": 4.216243692934602e-06, "loss": 0.3438123226165771, "memory(GiB)": 77.59, "step": 101340, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.461437 }, { "epoch": 4.3419305085471915, "grad_norm": 2.3494389057159424, "learning_rate": 4.213539282483814e-06, "loss": 0.348982048034668, "memory(GiB)": 77.59, "step": 101345, "token_acc": 0.9337979094076655, "train_speed(iter/s)": 1.461441 }, { "epoch": 4.342144723876441, "grad_norm": 2.746291160583496, "learning_rate": 4.210835701497712e-06, "loss": 0.47578206062316897, "memory(GiB)": 77.59, "step": 101350, "token_acc": 0.8773584905660378, "train_speed(iter/s)": 1.461443 }, { "epoch": 4.342358939205689, "grad_norm": 4.35871696472168, "learning_rate": 4.208132950025273e-06, "loss": 0.32745516300201416, "memory(GiB)": 77.59, "step": 101355, "token_acc": 0.93125, "train_speed(iter/s)": 1.461444 }, { "epoch": 4.342573154534938, "grad_norm": 5.073497295379639, "learning_rate": 4.205431028115458e-06, "loss": 0.4944461345672607, "memory(GiB)": 77.59, "step": 101360, "token_acc": 0.8870967741935484, "train_speed(iter/s)": 1.461451 }, { "epoch": 4.342787369864188, "grad_norm": 4.333206653594971, "learning_rate": 4.202729935817224e-06, "loss": 0.35316293239593505, "memory(GiB)": 77.59, "step": 101365, "token_acc": 0.9118773946360154, "train_speed(iter/s)": 1.461453 }, { "epoch": 4.343001585193436, "grad_norm": 4.060690879821777, "learning_rate": 4.20002967317949e-06, "loss": 0.19557468891143798, "memory(GiB)": 77.59, "step": 101370, "token_acc": 0.9545454545454546, "train_speed(iter/s)": 1.461454 }, { "epoch": 4.343215800522685, "grad_norm": 3.908449649810791, "learning_rate": 4.19733024025119e-06, "loss": 0.5469189643859863, "memory(GiB)": 77.59, "step": 101375, "token_acc": 0.8975265017667845, "train_speed(iter/s)": 1.461458 }, { "epoch": 4.3434300158519346, "grad_norm": 2.7415523529052734, "learning_rate": 4.19463163708122e-06, "loss": 0.290436315536499, "memory(GiB)": 77.59, "step": 101380, "token_acc": 0.9212827988338192, "train_speed(iter/s)": 1.461462 }, { "epoch": 4.343644231181183, "grad_norm": 3.295381546020508, "learning_rate": 4.191933863718467e-06, "loss": 0.4873295783996582, "memory(GiB)": 77.59, "step": 101385, "token_acc": 0.8949152542372881, "train_speed(iter/s)": 1.461468 }, { "epoch": 4.343858446510432, "grad_norm": 5.028715133666992, "learning_rate": 4.189236920211798e-06, "loss": 0.4217081069946289, "memory(GiB)": 77.59, "step": 101390, "token_acc": 0.8984962406015038, "train_speed(iter/s)": 1.461461 }, { "epoch": 4.344072661839681, "grad_norm": 1.433548092842102, "learning_rate": 4.1865408066100785e-06, "loss": 0.2519207239151001, "memory(GiB)": 77.59, "step": 101395, "token_acc": 0.9259259259259259, "train_speed(iter/s)": 1.461477 }, { "epoch": 4.34428687716893, "grad_norm": 2.958332061767578, "learning_rate": 4.183845522962154e-06, "loss": 0.4255256175994873, "memory(GiB)": 77.59, "step": 101400, "token_acc": 0.8968253968253969, "train_speed(iter/s)": 1.461482 }, { "epoch": 4.344501092498179, "grad_norm": 3.6648411750793457, "learning_rate": 4.181151069316846e-06, "loss": 0.301509428024292, "memory(GiB)": 77.59, "step": 101405, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.461488 }, { "epoch": 4.344715307827428, "grad_norm": 5.1755852699279785, "learning_rate": 4.178457445722967e-06, "loss": 0.5078950881958008, "memory(GiB)": 77.59, "step": 101410, "token_acc": 0.89419795221843, "train_speed(iter/s)": 1.461488 }, { "epoch": 4.344929523156677, "grad_norm": 2.537118673324585, "learning_rate": 4.175764652229308e-06, "loss": 0.2582973003387451, "memory(GiB)": 77.59, "step": 101415, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.461493 }, { "epoch": 4.345143738485926, "grad_norm": 0.3854601979255676, "learning_rate": 4.173072688884666e-06, "loss": 0.1556634783744812, "memory(GiB)": 77.59, "step": 101420, "token_acc": 0.9733333333333334, "train_speed(iter/s)": 1.461494 }, { "epoch": 4.345357953815175, "grad_norm": 4.314749240875244, "learning_rate": 4.170381555737801e-06, "loss": 0.5195286750793457, "memory(GiB)": 77.59, "step": 101425, "token_acc": 0.8916967509025271, "train_speed(iter/s)": 1.461505 }, { "epoch": 4.345572169144424, "grad_norm": 1.997642159461975, "learning_rate": 4.167691252837463e-06, "loss": 0.5672763824462891, "memory(GiB)": 77.59, "step": 101430, "token_acc": 0.8990825688073395, "train_speed(iter/s)": 1.461521 }, { "epoch": 4.345786384473673, "grad_norm": 2.2253904342651367, "learning_rate": 4.165001780232397e-06, "loss": 0.5855161190032959, "memory(GiB)": 77.59, "step": 101435, "token_acc": 0.8475836431226765, "train_speed(iter/s)": 1.461531 }, { "epoch": 4.346000599802922, "grad_norm": 3.5504400730133057, "learning_rate": 4.1623131379713186e-06, "loss": 0.2901404142379761, "memory(GiB)": 77.59, "step": 101440, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.461535 }, { "epoch": 4.3462148151321705, "grad_norm": 4.408580780029297, "learning_rate": 4.159625326102934e-06, "loss": 0.35861949920654296, "memory(GiB)": 77.59, "step": 101445, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.461537 }, { "epoch": 4.34642903046142, "grad_norm": 5.500342845916748, "learning_rate": 4.156938344675932e-06, "loss": 0.2484222650527954, "memory(GiB)": 77.59, "step": 101450, "token_acc": 0.95703125, "train_speed(iter/s)": 1.461538 }, { "epoch": 4.346643245790669, "grad_norm": 3.1795592308044434, "learning_rate": 4.154252193739006e-06, "loss": 0.20497457981109618, "memory(GiB)": 77.59, "step": 101455, "token_acc": 0.9511400651465798, "train_speed(iter/s)": 1.461539 }, { "epoch": 4.346857461119917, "grad_norm": 4.240200519561768, "learning_rate": 4.151566873340806e-06, "loss": 0.25002501010894773, "memory(GiB)": 77.59, "step": 101460, "token_acc": 0.9335664335664335, "train_speed(iter/s)": 1.46154 }, { "epoch": 4.347071676449167, "grad_norm": 2.6414706707000732, "learning_rate": 4.148882383529984e-06, "loss": 0.21623754501342773, "memory(GiB)": 77.59, "step": 101465, "token_acc": 0.9421768707482994, "train_speed(iter/s)": 1.461539 }, { "epoch": 4.347285891778416, "grad_norm": 0.17687052488327026, "learning_rate": 4.146198724355171e-06, "loss": 0.10814529657363892, "memory(GiB)": 77.59, "step": 101470, "token_acc": 0.9811320754716981, "train_speed(iter/s)": 1.461542 }, { "epoch": 4.347500107107664, "grad_norm": 7.23358154296875, "learning_rate": 4.143515895864969e-06, "loss": 0.47806873321533205, "memory(GiB)": 77.59, "step": 101475, "token_acc": 0.922077922077922, "train_speed(iter/s)": 1.461545 }, { "epoch": 4.3477143224369135, "grad_norm": 8.373373031616211, "learning_rate": 4.1408338981080096e-06, "loss": 0.5743863105773925, "memory(GiB)": 77.59, "step": 101480, "token_acc": 0.8884615384615384, "train_speed(iter/s)": 1.461548 }, { "epoch": 4.347928537766163, "grad_norm": 3.1259751319885254, "learning_rate": 4.138152731132855e-06, "loss": 0.5261325359344482, "memory(GiB)": 77.59, "step": 101485, "token_acc": 0.9007633587786259, "train_speed(iter/s)": 1.461547 }, { "epoch": 4.348142753095411, "grad_norm": 2.3701705932617188, "learning_rate": 4.135472394988094e-06, "loss": 0.23999366760253907, "memory(GiB)": 77.59, "step": 101490, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.46155 }, { "epoch": 4.34835696842466, "grad_norm": 0.13746701180934906, "learning_rate": 4.1327928897222705e-06, "loss": 0.43122406005859376, "memory(GiB)": 77.59, "step": 101495, "token_acc": 0.9014598540145985, "train_speed(iter/s)": 1.461548 }, { "epoch": 4.34857118375391, "grad_norm": 2.305772066116333, "learning_rate": 4.130114215383929e-06, "loss": 0.6200616359710693, "memory(GiB)": 77.59, "step": 101500, "token_acc": 0.863481228668942, "train_speed(iter/s)": 1.46156 }, { "epoch": 4.34857118375391, "eval_loss": 2.506633996963501, "eval_runtime": 11.258, "eval_samples_per_second": 8.883, "eval_steps_per_second": 8.883, "eval_token_acc": 0.4837758112094395, "step": 101500 }, { "epoch": 4.348785399083158, "grad_norm": 5.085952281951904, "learning_rate": 4.1274363720215924e-06, "loss": 0.45489091873168946, "memory(GiB)": 77.59, "step": 101505, "token_acc": 0.6146694214876033, "train_speed(iter/s)": 1.461314 }, { "epoch": 4.348999614412407, "grad_norm": 5.083634376525879, "learning_rate": 4.124759359683783e-06, "loss": 0.17827329635620118, "memory(GiB)": 77.59, "step": 101510, "token_acc": 0.946969696969697, "train_speed(iter/s)": 1.46132 }, { "epoch": 4.3492138297416565, "grad_norm": 3.8113133907318115, "learning_rate": 4.122083178418995e-06, "loss": 0.4809504508972168, "memory(GiB)": 77.59, "step": 101515, "token_acc": 0.8795620437956204, "train_speed(iter/s)": 1.461326 }, { "epoch": 4.349428045070905, "grad_norm": 3.360507011413574, "learning_rate": 4.119407828275696e-06, "loss": 0.3850550174713135, "memory(GiB)": 77.59, "step": 101520, "token_acc": 0.8963210702341137, "train_speed(iter/s)": 1.461338 }, { "epoch": 4.349642260400154, "grad_norm": 5.453113555908203, "learning_rate": 4.116733309302373e-06, "loss": 0.37569262981414797, "memory(GiB)": 77.59, "step": 101525, "token_acc": 0.9153846153846154, "train_speed(iter/s)": 1.461338 }, { "epoch": 4.349856475729403, "grad_norm": 3.754560708999634, "learning_rate": 4.114059621547472e-06, "loss": 0.36787662506103513, "memory(GiB)": 77.59, "step": 101530, "token_acc": 0.8976377952755905, "train_speed(iter/s)": 1.461342 }, { "epoch": 4.350070691058653, "grad_norm": 2.7229349613189697, "learning_rate": 4.111386765059416e-06, "loss": 0.43071622848510743, "memory(GiB)": 77.59, "step": 101535, "token_acc": 0.9074074074074074, "train_speed(iter/s)": 1.461343 }, { "epoch": 4.350284906387901, "grad_norm": 4.448983192443848, "learning_rate": 4.108714739886638e-06, "loss": 0.44573450088500977, "memory(GiB)": 77.59, "step": 101540, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.461346 }, { "epoch": 4.35049912171715, "grad_norm": 0.8008148670196533, "learning_rate": 4.106043546077548e-06, "loss": 0.443975830078125, "memory(GiB)": 77.59, "step": 101545, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.461347 }, { "epoch": 4.350713337046399, "grad_norm": 0.96487957239151, "learning_rate": 4.103373183680531e-06, "loss": 0.3173358917236328, "memory(GiB)": 77.59, "step": 101550, "token_acc": 0.9214285714285714, "train_speed(iter/s)": 1.461353 }, { "epoch": 4.350927552375648, "grad_norm": 3.4435150623321533, "learning_rate": 4.100703652743959e-06, "loss": 0.24202301502227783, "memory(GiB)": 77.59, "step": 101555, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.461345 }, { "epoch": 4.351141767704897, "grad_norm": 2.16888165473938, "learning_rate": 4.098034953316194e-06, "loss": 0.3640062570571899, "memory(GiB)": 77.59, "step": 101560, "token_acc": 0.9256756756756757, "train_speed(iter/s)": 1.461347 }, { "epoch": 4.351355983034146, "grad_norm": 1.9777668714523315, "learning_rate": 4.09536708544559e-06, "loss": 0.3093660593032837, "memory(GiB)": 77.59, "step": 101565, "token_acc": 0.9356913183279743, "train_speed(iter/s)": 1.461347 }, { "epoch": 4.351570198363395, "grad_norm": 7.504862308502197, "learning_rate": 4.092700049180476e-06, "loss": 0.3904132843017578, "memory(GiB)": 77.59, "step": 101570, "token_acc": 0.9118541033434651, "train_speed(iter/s)": 1.461358 }, { "epoch": 4.351784413692644, "grad_norm": 1.3879785537719727, "learning_rate": 4.090033844569163e-06, "loss": 0.36031601428985593, "memory(GiB)": 77.59, "step": 101575, "token_acc": 0.9083969465648855, "train_speed(iter/s)": 1.46136 }, { "epoch": 4.351998629021892, "grad_norm": 4.1796650886535645, "learning_rate": 4.087368471659958e-06, "loss": 0.11495392322540283, "memory(GiB)": 77.59, "step": 101580, "token_acc": 0.9696048632218845, "train_speed(iter/s)": 1.46136 }, { "epoch": 4.352212844351142, "grad_norm": 1.552212119102478, "learning_rate": 4.084703930501138e-06, "loss": 0.436847448348999, "memory(GiB)": 77.59, "step": 101585, "token_acc": 0.9033232628398792, "train_speed(iter/s)": 1.461367 }, { "epoch": 4.352427059680391, "grad_norm": 3.2079758644104004, "learning_rate": 4.08204022114097e-06, "loss": 0.20231549739837645, "memory(GiB)": 77.59, "step": 101590, "token_acc": 0.9529780564263323, "train_speed(iter/s)": 1.461366 }, { "epoch": 4.35264127500964, "grad_norm": 4.1708831787109375, "learning_rate": 4.079377343627722e-06, "loss": 0.4809995174407959, "memory(GiB)": 77.59, "step": 101595, "token_acc": 0.8779527559055118, "train_speed(iter/s)": 1.461378 }, { "epoch": 4.3528554903388885, "grad_norm": 2.4106249809265137, "learning_rate": 4.076715298009637e-06, "loss": 0.3124258995056152, "memory(GiB)": 77.59, "step": 101600, "token_acc": 0.9192982456140351, "train_speed(iter/s)": 1.461378 }, { "epoch": 4.353069705668138, "grad_norm": 4.7676682472229, "learning_rate": 4.0740540843349325e-06, "loss": 0.4539023399353027, "memory(GiB)": 77.59, "step": 101605, "token_acc": 0.8916083916083916, "train_speed(iter/s)": 1.46138 }, { "epoch": 4.353283920997386, "grad_norm": 2.1221179962158203, "learning_rate": 4.071393702651821e-06, "loss": 0.3694819450378418, "memory(GiB)": 77.59, "step": 101610, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.461377 }, { "epoch": 4.353498136326635, "grad_norm": 2.5100057125091553, "learning_rate": 4.068734153008497e-06, "loss": 0.3317371368408203, "memory(GiB)": 77.59, "step": 101615, "token_acc": 0.937037037037037, "train_speed(iter/s)": 1.461378 }, { "epoch": 4.353712351655885, "grad_norm": 3.826294183731079, "learning_rate": 4.066075435453132e-06, "loss": 0.4388144016265869, "memory(GiB)": 77.59, "step": 101620, "token_acc": 0.9113924050632911, "train_speed(iter/s)": 1.461379 }, { "epoch": 4.353926566985134, "grad_norm": 4.711773872375488, "learning_rate": 4.06341755003391e-06, "loss": 0.34729585647583006, "memory(GiB)": 77.59, "step": 101625, "token_acc": 0.9256505576208178, "train_speed(iter/s)": 1.461381 }, { "epoch": 4.354140782314382, "grad_norm": 3.6096105575561523, "learning_rate": 4.060760496798971e-06, "loss": 0.3896145820617676, "memory(GiB)": 77.59, "step": 101630, "token_acc": 0.9321533923303835, "train_speed(iter/s)": 1.461381 }, { "epoch": 4.3543549976436315, "grad_norm": 3.1951346397399902, "learning_rate": 4.058104275796449e-06, "loss": 0.5807657718658448, "memory(GiB)": 77.59, "step": 101635, "token_acc": 0.8885245901639345, "train_speed(iter/s)": 1.461385 }, { "epoch": 4.35456921297288, "grad_norm": 3.1045563220977783, "learning_rate": 4.0554488870744645e-06, "loss": 0.29293136596679686, "memory(GiB)": 77.59, "step": 101640, "token_acc": 0.9328358208955224, "train_speed(iter/s)": 1.461393 }, { "epoch": 4.354783428302129, "grad_norm": 4.620826721191406, "learning_rate": 4.052794330681125e-06, "loss": 0.39889357089996336, "memory(GiB)": 77.59, "step": 101645, "token_acc": 0.9136690647482014, "train_speed(iter/s)": 1.461393 }, { "epoch": 4.354997643631378, "grad_norm": 1.5822038650512695, "learning_rate": 4.050140606664505e-06, "loss": 0.22763361930847167, "memory(GiB)": 77.59, "step": 101650, "token_acc": 0.9512987012987013, "train_speed(iter/s)": 1.461395 }, { "epoch": 4.355211858960628, "grad_norm": 1.0518516302108765, "learning_rate": 4.047487715072706e-06, "loss": 0.14997791051864623, "memory(GiB)": 77.59, "step": 101655, "token_acc": 0.9475806451612904, "train_speed(iter/s)": 1.461404 }, { "epoch": 4.355426074289876, "grad_norm": 4.930720329284668, "learning_rate": 4.044835655953772e-06, "loss": 0.6427118301391601, "memory(GiB)": 77.59, "step": 101660, "token_acc": 0.8766666666666667, "train_speed(iter/s)": 1.461414 }, { "epoch": 4.355640289619125, "grad_norm": 3.692026138305664, "learning_rate": 4.042184429355739e-06, "loss": 0.39156394004821776, "memory(GiB)": 77.59, "step": 101665, "token_acc": 0.9066265060240963, "train_speed(iter/s)": 1.461416 }, { "epoch": 4.355854504948374, "grad_norm": 4.707671642303467, "learning_rate": 4.039534035326659e-06, "loss": 0.26752915382385256, "memory(GiB)": 77.59, "step": 101670, "token_acc": 0.9341692789968652, "train_speed(iter/s)": 1.461416 }, { "epoch": 4.356068720277623, "grad_norm": 12.353384971618652, "learning_rate": 4.0368844739145315e-06, "loss": 0.4316694736480713, "memory(GiB)": 77.59, "step": 101675, "token_acc": 0.9131652661064426, "train_speed(iter/s)": 1.461421 }, { "epoch": 4.356282935606872, "grad_norm": 2.585183620452881, "learning_rate": 4.034235745167353e-06, "loss": 0.29858429431915284, "memory(GiB)": 77.59, "step": 101680, "token_acc": 0.9390681003584229, "train_speed(iter/s)": 1.461423 }, { "epoch": 4.356497150936121, "grad_norm": 2.819673776626587, "learning_rate": 4.031587849133123e-06, "loss": 0.31615643501281737, "memory(GiB)": 77.59, "step": 101685, "token_acc": 0.9191616766467066, "train_speed(iter/s)": 1.461435 }, { "epoch": 4.35671136626537, "grad_norm": 4.353437900543213, "learning_rate": 4.0289407858597974e-06, "loss": 0.38602185249328613, "memory(GiB)": 77.59, "step": 101690, "token_acc": 0.9288888888888889, "train_speed(iter/s)": 1.461447 }, { "epoch": 4.356925581594619, "grad_norm": 2.121406316757202, "learning_rate": 4.026294555395332e-06, "loss": 0.3733814239501953, "memory(GiB)": 77.59, "step": 101695, "token_acc": 0.9185185185185185, "train_speed(iter/s)": 1.46145 }, { "epoch": 4.357139796923867, "grad_norm": 3.1849825382232666, "learning_rate": 4.023649157787673e-06, "loss": 0.3230432510375977, "memory(GiB)": 77.59, "step": 101700, "token_acc": 0.9379310344827586, "train_speed(iter/s)": 1.461454 }, { "epoch": 4.357354012253117, "grad_norm": 3.0462048053741455, "learning_rate": 4.021004593084726e-06, "loss": 0.5849740028381347, "memory(GiB)": 77.59, "step": 101705, "token_acc": 0.8784722222222222, "train_speed(iter/s)": 1.461457 }, { "epoch": 4.357568227582366, "grad_norm": 0.6724727153778076, "learning_rate": 4.0183608613344244e-06, "loss": 0.16537686586380004, "memory(GiB)": 77.59, "step": 101710, "token_acc": 0.9644970414201184, "train_speed(iter/s)": 1.461458 }, { "epoch": 4.357782442911615, "grad_norm": 0.6352661848068237, "learning_rate": 4.015717962584648e-06, "loss": 0.4375617504119873, "memory(GiB)": 77.59, "step": 101715, "token_acc": 0.9, "train_speed(iter/s)": 1.461455 }, { "epoch": 4.357996658240864, "grad_norm": 5.958620548248291, "learning_rate": 4.013075896883278e-06, "loss": 0.4161695003509521, "memory(GiB)": 77.59, "step": 101720, "token_acc": 0.9193548387096774, "train_speed(iter/s)": 1.461467 }, { "epoch": 4.358210873570113, "grad_norm": 4.76558780670166, "learning_rate": 4.0104346642781785e-06, "loss": 0.4688542366027832, "memory(GiB)": 77.59, "step": 101725, "token_acc": 0.901840490797546, "train_speed(iter/s)": 1.461465 }, { "epoch": 4.358425088899361, "grad_norm": 2.2870266437530518, "learning_rate": 4.007794264817194e-06, "loss": 0.2702461004257202, "memory(GiB)": 77.59, "step": 101730, "token_acc": 0.9290322580645162, "train_speed(iter/s)": 1.46147 }, { "epoch": 4.3586393042286105, "grad_norm": 1.5194388628005981, "learning_rate": 4.005154698548152e-06, "loss": 0.2819650888442993, "memory(GiB)": 77.59, "step": 101735, "token_acc": 0.9420731707317073, "train_speed(iter/s)": 1.461471 }, { "epoch": 4.35885351955786, "grad_norm": 4.873916149139404, "learning_rate": 4.002515965518883e-06, "loss": 0.6392566204071045, "memory(GiB)": 77.59, "step": 101740, "token_acc": 0.863481228668942, "train_speed(iter/s)": 1.461479 }, { "epoch": 4.359067734887109, "grad_norm": 5.42298698425293, "learning_rate": 3.999878065777191e-06, "loss": 0.2769233226776123, "memory(GiB)": 77.59, "step": 101745, "token_acc": 0.932475884244373, "train_speed(iter/s)": 1.461479 }, { "epoch": 4.359281950216357, "grad_norm": 4.195010662078857, "learning_rate": 3.9972409993708605e-06, "loss": 0.25090622901916504, "memory(GiB)": 77.59, "step": 101750, "token_acc": 0.9415807560137457, "train_speed(iter/s)": 1.461485 }, { "epoch": 4.359496165545607, "grad_norm": 1.2870301008224487, "learning_rate": 3.994604766347665e-06, "loss": 0.3398630380630493, "memory(GiB)": 77.59, "step": 101755, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.461488 }, { "epoch": 4.359710380874855, "grad_norm": 1.982466220855713, "learning_rate": 3.99196936675536e-06, "loss": 0.2951848268508911, "memory(GiB)": 77.59, "step": 101760, "token_acc": 0.9593220338983051, "train_speed(iter/s)": 1.461491 }, { "epoch": 4.359924596204104, "grad_norm": 2.5590946674346924, "learning_rate": 3.989334800641681e-06, "loss": 0.33265113830566406, "memory(GiB)": 77.59, "step": 101765, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.461491 }, { "epoch": 4.3601388115333535, "grad_norm": 3.271059989929199, "learning_rate": 3.986701068054371e-06, "loss": 0.40577449798583987, "memory(GiB)": 77.59, "step": 101770, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.461492 }, { "epoch": 4.360353026862603, "grad_norm": 5.383742809295654, "learning_rate": 3.9840681690411384e-06, "loss": 0.4425179958343506, "memory(GiB)": 77.59, "step": 101775, "token_acc": 0.9016393442622951, "train_speed(iter/s)": 1.461493 }, { "epoch": 4.360567242191851, "grad_norm": 2.7403297424316406, "learning_rate": 3.981436103649672e-06, "loss": 0.513450288772583, "memory(GiB)": 77.59, "step": 101780, "token_acc": 0.9087591240875912, "train_speed(iter/s)": 1.461507 }, { "epoch": 4.3607814575211, "grad_norm": 2.7280635833740234, "learning_rate": 3.978804871927661e-06, "loss": 0.30595424175262453, "memory(GiB)": 77.59, "step": 101785, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.46151 }, { "epoch": 4.360995672850349, "grad_norm": 1.5567563772201538, "learning_rate": 3.976174473922772e-06, "loss": 0.12131597995758056, "memory(GiB)": 77.59, "step": 101790, "token_acc": 0.9707112970711297, "train_speed(iter/s)": 1.461515 }, { "epoch": 4.361209888179598, "grad_norm": 1.8040705919265747, "learning_rate": 3.973544909682647e-06, "loss": 0.20887744426727295, "memory(GiB)": 77.59, "step": 101795, "token_acc": 0.9622641509433962, "train_speed(iter/s)": 1.461519 }, { "epoch": 4.361424103508847, "grad_norm": 4.644045829772949, "learning_rate": 3.970916179254941e-06, "loss": 0.3904045820236206, "memory(GiB)": 77.59, "step": 101800, "token_acc": 0.9183673469387755, "train_speed(iter/s)": 1.46152 }, { "epoch": 4.3616383188380965, "grad_norm": 4.144597053527832, "learning_rate": 3.968288282687272e-06, "loss": 0.40198655128479005, "memory(GiB)": 77.59, "step": 101805, "token_acc": 0.9125475285171103, "train_speed(iter/s)": 1.461523 }, { "epoch": 4.361852534167345, "grad_norm": 1.540069580078125, "learning_rate": 3.965661220027233e-06, "loss": 0.43882951736450193, "memory(GiB)": 77.59, "step": 101810, "token_acc": 0.8908045977011494, "train_speed(iter/s)": 1.46153 }, { "epoch": 4.362066749496594, "grad_norm": 3.891554117202759, "learning_rate": 3.963034991322423e-06, "loss": 0.3708931922912598, "memory(GiB)": 77.59, "step": 101815, "token_acc": 0.8868778280542986, "train_speed(iter/s)": 1.461543 }, { "epoch": 4.3622809648258425, "grad_norm": 2.04250168800354, "learning_rate": 3.9604095966204295e-06, "loss": 0.2787394285202026, "memory(GiB)": 77.59, "step": 101820, "token_acc": 0.930921052631579, "train_speed(iter/s)": 1.461545 }, { "epoch": 4.362495180155092, "grad_norm": 6.846263885498047, "learning_rate": 3.957785035968792e-06, "loss": 0.4568460464477539, "memory(GiB)": 77.59, "step": 101825, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.461545 }, { "epoch": 4.362709395484341, "grad_norm": 1.3973699808120728, "learning_rate": 3.955161309415078e-06, "loss": 0.3947919845581055, "memory(GiB)": 77.59, "step": 101830, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.461544 }, { "epoch": 4.36292361081359, "grad_norm": 1.2139196395874023, "learning_rate": 3.9525384170068125e-06, "loss": 0.35262291431427, "memory(GiB)": 77.59, "step": 101835, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.461549 }, { "epoch": 4.363137826142839, "grad_norm": 2.5453853607177734, "learning_rate": 3.949916358791511e-06, "loss": 0.15375845432281493, "memory(GiB)": 77.59, "step": 101840, "token_acc": 0.9693486590038314, "train_speed(iter/s)": 1.461561 }, { "epoch": 4.363352041472088, "grad_norm": 2.2748613357543945, "learning_rate": 3.94729513481667e-06, "loss": 0.2566936731338501, "memory(GiB)": 77.59, "step": 101845, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.461562 }, { "epoch": 4.363566256801336, "grad_norm": 5.071557521820068, "learning_rate": 3.944674745129773e-06, "loss": 0.39788341522216797, "memory(GiB)": 77.59, "step": 101850, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.461568 }, { "epoch": 4.3637804721305855, "grad_norm": 2.5756349563598633, "learning_rate": 3.942055189778305e-06, "loss": 0.4479671001434326, "memory(GiB)": 77.59, "step": 101855, "token_acc": 0.903125, "train_speed(iter/s)": 1.461569 }, { "epoch": 4.363994687459835, "grad_norm": 1.647615671157837, "learning_rate": 3.93943646880971e-06, "loss": 0.2498990774154663, "memory(GiB)": 77.59, "step": 101860, "token_acc": 0.9532163742690059, "train_speed(iter/s)": 1.461574 }, { "epoch": 4.364208902789084, "grad_norm": 0.8044785261154175, "learning_rate": 3.936818582271428e-06, "loss": 0.33151845932006835, "memory(GiB)": 77.59, "step": 101865, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.461575 }, { "epoch": 4.364423118118332, "grad_norm": 7.6158552169799805, "learning_rate": 3.934201530210896e-06, "loss": 0.2685936450958252, "memory(GiB)": 77.59, "step": 101870, "token_acc": 0.9427609427609428, "train_speed(iter/s)": 1.461578 }, { "epoch": 4.364637333447582, "grad_norm": 5.383625030517578, "learning_rate": 3.931585312675512e-06, "loss": 0.3904134750366211, "memory(GiB)": 77.59, "step": 101875, "token_acc": 0.9099378881987578, "train_speed(iter/s)": 1.461589 }, { "epoch": 4.36485154877683, "grad_norm": 1.8554459810256958, "learning_rate": 3.928969929712673e-06, "loss": 0.2305997848510742, "memory(GiB)": 77.59, "step": 101880, "token_acc": 0.9457364341085271, "train_speed(iter/s)": 1.461587 }, { "epoch": 4.365065764106079, "grad_norm": 4.815619945526123, "learning_rate": 3.926355381369762e-06, "loss": 0.38931310176849365, "memory(GiB)": 77.59, "step": 101885, "token_acc": 0.9245283018867925, "train_speed(iter/s)": 1.461593 }, { "epoch": 4.3652799794353285, "grad_norm": 3.7353384494781494, "learning_rate": 3.923741667694142e-06, "loss": 0.4138832092285156, "memory(GiB)": 77.59, "step": 101890, "token_acc": 0.9059561128526645, "train_speed(iter/s)": 1.461602 }, { "epoch": 4.365494194764578, "grad_norm": 0.3971637487411499, "learning_rate": 3.921128788733169e-06, "loss": 0.1401178479194641, "memory(GiB)": 77.59, "step": 101895, "token_acc": 0.9742647058823529, "train_speed(iter/s)": 1.461616 }, { "epoch": 4.365708410093826, "grad_norm": 4.533706188201904, "learning_rate": 3.918516744534178e-06, "loss": 0.3286141395568848, "memory(GiB)": 77.59, "step": 101900, "token_acc": 0.924812030075188, "train_speed(iter/s)": 1.461625 }, { "epoch": 4.365922625423075, "grad_norm": 1.7424513101577759, "learning_rate": 3.915905535144482e-06, "loss": 0.35734717845916747, "memory(GiB)": 77.59, "step": 101905, "token_acc": 0.9368770764119602, "train_speed(iter/s)": 1.46163 }, { "epoch": 4.366136840752324, "grad_norm": 1.8957140445709229, "learning_rate": 3.9132951606113745e-06, "loss": 0.1937070369720459, "memory(GiB)": 77.59, "step": 101910, "token_acc": 0.9484848484848485, "train_speed(iter/s)": 1.461642 }, { "epoch": 4.366351056081573, "grad_norm": 0.40331918001174927, "learning_rate": 3.910685620982174e-06, "loss": 0.1674961566925049, "memory(GiB)": 77.59, "step": 101915, "token_acc": 0.9525547445255474, "train_speed(iter/s)": 1.461648 }, { "epoch": 4.366565271410822, "grad_norm": 0.3208419382572174, "learning_rate": 3.908076916304132e-06, "loss": 0.27214226722717283, "memory(GiB)": 77.59, "step": 101920, "token_acc": 0.9460916442048517, "train_speed(iter/s)": 1.461659 }, { "epoch": 4.3667794867400715, "grad_norm": 4.089642524719238, "learning_rate": 3.9054690466245215e-06, "loss": 0.22491135597229003, "memory(GiB)": 77.59, "step": 101925, "token_acc": 0.9340277777777778, "train_speed(iter/s)": 1.461658 }, { "epoch": 4.36699370206932, "grad_norm": 5.370346546173096, "learning_rate": 3.902862011990571e-06, "loss": 0.7127554893493653, "memory(GiB)": 77.59, "step": 101930, "token_acc": 0.8476190476190476, "train_speed(iter/s)": 1.461659 }, { "epoch": 4.367207917398569, "grad_norm": 14.827346801757812, "learning_rate": 3.900255812449527e-06, "loss": 0.4085966110229492, "memory(GiB)": 77.59, "step": 101935, "token_acc": 0.9019607843137255, "train_speed(iter/s)": 1.461663 }, { "epoch": 4.3674221327278175, "grad_norm": 2.4403719902038574, "learning_rate": 3.897650448048579e-06, "loss": 0.2543109655380249, "memory(GiB)": 77.59, "step": 101940, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.461665 }, { "epoch": 4.367636348057067, "grad_norm": 5.025214195251465, "learning_rate": 3.89504591883495e-06, "loss": 0.39317007064819337, "memory(GiB)": 77.59, "step": 101945, "token_acc": 0.9123376623376623, "train_speed(iter/s)": 1.46167 }, { "epoch": 4.367850563386316, "grad_norm": 2.606553316116333, "learning_rate": 3.892442224855813e-06, "loss": 0.5154307365417481, "memory(GiB)": 77.59, "step": 101950, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.46167 }, { "epoch": 4.368064778715565, "grad_norm": 3.939901828765869, "learning_rate": 3.889839366158343e-06, "loss": 0.3221132755279541, "memory(GiB)": 77.59, "step": 101955, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.461672 }, { "epoch": 4.368278994044814, "grad_norm": 3.749378204345703, "learning_rate": 3.8872373427896735e-06, "loss": 0.22126731872558594, "memory(GiB)": 77.59, "step": 101960, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.461674 }, { "epoch": 4.368493209374063, "grad_norm": 2.326178789138794, "learning_rate": 3.884636154796967e-06, "loss": 0.20393543243408202, "memory(GiB)": 77.59, "step": 101965, "token_acc": 0.9591836734693877, "train_speed(iter/s)": 1.461673 }, { "epoch": 4.368707424703311, "grad_norm": 3.9329895973205566, "learning_rate": 3.88203580222733e-06, "loss": 0.44348831176757814, "memory(GiB)": 77.59, "step": 101970, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.461674 }, { "epoch": 4.368921640032561, "grad_norm": 1.024096131324768, "learning_rate": 3.879436285127886e-06, "loss": 0.19640852212905885, "memory(GiB)": 77.59, "step": 101975, "token_acc": 0.9614147909967846, "train_speed(iter/s)": 1.461677 }, { "epoch": 4.36913585536181, "grad_norm": 5.384939670562744, "learning_rate": 3.876837603545713e-06, "loss": 0.24894320964813232, "memory(GiB)": 77.59, "step": 101980, "token_acc": 0.9470198675496688, "train_speed(iter/s)": 1.461683 }, { "epoch": 4.369350070691059, "grad_norm": 6.022385597229004, "learning_rate": 3.874239757527897e-06, "loss": 0.3017905235290527, "memory(GiB)": 77.59, "step": 101985, "token_acc": 0.9180887372013652, "train_speed(iter/s)": 1.461682 }, { "epoch": 4.369564286020307, "grad_norm": 2.108247756958008, "learning_rate": 3.8716427471214955e-06, "loss": 0.49773530960083007, "memory(GiB)": 77.59, "step": 101990, "token_acc": 0.8931750741839762, "train_speed(iter/s)": 1.461685 }, { "epoch": 4.369778501349557, "grad_norm": 3.4949393272399902, "learning_rate": 3.869046572373552e-06, "loss": 0.27514450550079345, "memory(GiB)": 77.59, "step": 101995, "token_acc": 0.9554655870445344, "train_speed(iter/s)": 1.461687 }, { "epoch": 4.369992716678805, "grad_norm": 3.176802635192871, "learning_rate": 3.866451233331108e-06, "loss": 0.31603026390075684, "memory(GiB)": 77.59, "step": 102000, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.461697 }, { "epoch": 4.369992716678805, "eval_loss": 2.4641153812408447, "eval_runtime": 11.5087, "eval_samples_per_second": 8.689, "eval_steps_per_second": 8.689, "eval_token_acc": 0.4796044499381953, "step": 102000 }, { "epoch": 4.370206932008054, "grad_norm": 4.355025291442871, "learning_rate": 3.863856730041182e-06, "loss": 0.3146980285644531, "memory(GiB)": 77.59, "step": 102005, "token_acc": 0.5923862581244197, "train_speed(iter/s)": 1.461438 }, { "epoch": 4.370421147337304, "grad_norm": 2.5635275840759277, "learning_rate": 3.861263062550769e-06, "loss": 0.2021177053451538, "memory(GiB)": 77.59, "step": 102010, "token_acc": 0.9551820728291317, "train_speed(iter/s)": 1.461441 }, { "epoch": 4.370635362666553, "grad_norm": 2.5807762145996094, "learning_rate": 3.858670230906852e-06, "loss": 0.27856030464172366, "memory(GiB)": 77.59, "step": 102015, "token_acc": 0.9335548172757475, "train_speed(iter/s)": 1.461444 }, { "epoch": 4.370849577995801, "grad_norm": 4.27592134475708, "learning_rate": 3.856078235156413e-06, "loss": 0.2849113464355469, "memory(GiB)": 77.59, "step": 102020, "token_acc": 0.948339483394834, "train_speed(iter/s)": 1.461444 }, { "epoch": 4.3710637933250505, "grad_norm": 3.182215690612793, "learning_rate": 3.853487075346396e-06, "loss": 0.33952698707580564, "memory(GiB)": 77.59, "step": 102025, "token_acc": 0.9180327868852459, "train_speed(iter/s)": 1.461456 }, { "epoch": 4.371278008654299, "grad_norm": 2.8106400966644287, "learning_rate": 3.850896751523758e-06, "loss": 0.2638895511627197, "memory(GiB)": 77.59, "step": 102030, "token_acc": 0.948051948051948, "train_speed(iter/s)": 1.461458 }, { "epoch": 4.371492223983548, "grad_norm": 1.0813570022583008, "learning_rate": 3.8483072637354065e-06, "loss": 0.2164360761642456, "memory(GiB)": 77.59, "step": 102035, "token_acc": 0.9449275362318841, "train_speed(iter/s)": 1.461459 }, { "epoch": 4.371706439312797, "grad_norm": 0.6909071207046509, "learning_rate": 3.845718612028271e-06, "loss": 0.3621103286743164, "memory(GiB)": 77.59, "step": 102040, "token_acc": 0.940983606557377, "train_speed(iter/s)": 1.461461 }, { "epoch": 4.371920654642047, "grad_norm": 0.13093999028205872, "learning_rate": 3.843130796449235e-06, "loss": 0.23465840816497802, "memory(GiB)": 77.59, "step": 102045, "token_acc": 0.9353846153846154, "train_speed(iter/s)": 1.461466 }, { "epoch": 4.372134869971295, "grad_norm": 3.0784857273101807, "learning_rate": 3.840543817045189e-06, "loss": 0.32766172885894773, "memory(GiB)": 77.59, "step": 102050, "token_acc": 0.9486166007905138, "train_speed(iter/s)": 1.46147 }, { "epoch": 4.372349085300544, "grad_norm": 2.2034151554107666, "learning_rate": 3.8379576738629856e-06, "loss": 0.4245858669281006, "memory(GiB)": 77.59, "step": 102055, "token_acc": 0.9154411764705882, "train_speed(iter/s)": 1.46147 }, { "epoch": 4.3725633006297935, "grad_norm": 2.6525607109069824, "learning_rate": 3.8353723669494855e-06, "loss": 0.2801202774047852, "memory(GiB)": 77.59, "step": 102060, "token_acc": 0.9317507418397626, "train_speed(iter/s)": 1.461481 }, { "epoch": 4.372777515959042, "grad_norm": 5.438822269439697, "learning_rate": 3.8327878963515256e-06, "loss": 0.2730985641479492, "memory(GiB)": 77.59, "step": 102065, "token_acc": 0.9392712550607287, "train_speed(iter/s)": 1.461483 }, { "epoch": 4.372991731288291, "grad_norm": 2.350632429122925, "learning_rate": 3.830204262115916e-06, "loss": 0.5251811981201172, "memory(GiB)": 77.59, "step": 102070, "token_acc": 0.8758169934640523, "train_speed(iter/s)": 1.461488 }, { "epoch": 4.37320594661754, "grad_norm": 3.361271858215332, "learning_rate": 3.827621464289471e-06, "loss": 0.4681502342224121, "memory(GiB)": 77.59, "step": 102075, "token_acc": 0.8914728682170543, "train_speed(iter/s)": 1.461489 }, { "epoch": 4.373420161946789, "grad_norm": 5.3332695960998535, "learning_rate": 3.825039502918965e-06, "loss": 0.30282676219940186, "memory(GiB)": 77.59, "step": 102080, "token_acc": 0.926056338028169, "train_speed(iter/s)": 1.46149 }, { "epoch": 4.373634377276038, "grad_norm": 1.5249265432357788, "learning_rate": 3.822458378051197e-06, "loss": 0.31638669967651367, "memory(GiB)": 77.59, "step": 102085, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.461493 }, { "epoch": 4.373848592605287, "grad_norm": 2.708928108215332, "learning_rate": 3.819878089732909e-06, "loss": 0.21602962017059327, "memory(GiB)": 77.59, "step": 102090, "token_acc": 0.9352750809061489, "train_speed(iter/s)": 1.461492 }, { "epoch": 4.374062807934536, "grad_norm": 3.3237192630767822, "learning_rate": 3.817298638010852e-06, "loss": 0.2138505458831787, "memory(GiB)": 77.59, "step": 102095, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.461493 }, { "epoch": 4.374277023263785, "grad_norm": 1.8579589128494263, "learning_rate": 3.814720022931756e-06, "loss": 0.27676384449005126, "memory(GiB)": 77.59, "step": 102100, "token_acc": 0.9541666666666667, "train_speed(iter/s)": 1.461493 }, { "epoch": 4.374491238593034, "grad_norm": 3.4839696884155273, "learning_rate": 3.812142244542327e-06, "loss": 0.4044055938720703, "memory(GiB)": 77.59, "step": 102105, "token_acc": 0.905511811023622, "train_speed(iter/s)": 1.461494 }, { "epoch": 4.3747054539222825, "grad_norm": 0.04992932081222534, "learning_rate": 3.809565302889262e-06, "loss": 0.5011977672576904, "memory(GiB)": 77.59, "step": 102110, "token_acc": 0.9124579124579124, "train_speed(iter/s)": 1.461503 }, { "epoch": 4.374919669251532, "grad_norm": 1.586935043334961, "learning_rate": 3.8069891980192507e-06, "loss": 0.1625565767288208, "memory(GiB)": 77.59, "step": 102115, "token_acc": 0.954954954954955, "train_speed(iter/s)": 1.461514 }, { "epoch": 4.375133884580781, "grad_norm": 3.315066337585449, "learning_rate": 3.8044139299789727e-06, "loss": 0.4399867057800293, "memory(GiB)": 77.59, "step": 102120, "token_acc": 0.9064516129032258, "train_speed(iter/s)": 1.461523 }, { "epoch": 4.375348099910029, "grad_norm": 1.9830241203308105, "learning_rate": 3.801839498815074e-06, "loss": 0.36542794704437254, "memory(GiB)": 77.59, "step": 102125, "token_acc": 0.9261992619926199, "train_speed(iter/s)": 1.461532 }, { "epoch": 4.375562315239279, "grad_norm": 3.123833179473877, "learning_rate": 3.799265904574184e-06, "loss": 0.2062601089477539, "memory(GiB)": 77.59, "step": 102130, "token_acc": 0.9529411764705882, "train_speed(iter/s)": 1.461547 }, { "epoch": 4.375776530568528, "grad_norm": 2.792893886566162, "learning_rate": 3.7966931473029378e-06, "loss": 0.32528746128082275, "memory(GiB)": 77.59, "step": 102135, "token_acc": 0.9173913043478261, "train_speed(iter/s)": 1.461549 }, { "epoch": 4.375990745897776, "grad_norm": 3.5598254203796387, "learning_rate": 3.7941212270479254e-06, "loss": 0.4253571033477783, "memory(GiB)": 77.59, "step": 102140, "token_acc": 0.90625, "train_speed(iter/s)": 1.461552 }, { "epoch": 4.3762049612270255, "grad_norm": 1.6678985357284546, "learning_rate": 3.79155014385576e-06, "loss": 0.4398487567901611, "memory(GiB)": 77.59, "step": 102145, "token_acc": 0.9029850746268657, "train_speed(iter/s)": 1.461557 }, { "epoch": 4.376419176556275, "grad_norm": 3.3122050762176514, "learning_rate": 3.7889798977730153e-06, "loss": 0.3324385643005371, "memory(GiB)": 77.59, "step": 102150, "token_acc": 0.9306569343065694, "train_speed(iter/s)": 1.461566 }, { "epoch": 4.376633391885523, "grad_norm": 0.09385988861322403, "learning_rate": 3.7864104888462425e-06, "loss": 0.33106389045715334, "memory(GiB)": 77.59, "step": 102155, "token_acc": 0.9469964664310954, "train_speed(iter/s)": 1.461571 }, { "epoch": 4.376847607214772, "grad_norm": 2.6181530952453613, "learning_rate": 3.783841917122e-06, "loss": 0.46223154067993166, "memory(GiB)": 77.59, "step": 102160, "token_acc": 0.9153605015673981, "train_speed(iter/s)": 1.46158 }, { "epoch": 4.377061822544022, "grad_norm": 2.7445573806762695, "learning_rate": 3.781274182646816e-06, "loss": 0.5343538284301758, "memory(GiB)": 77.59, "step": 102165, "token_acc": 0.8923076923076924, "train_speed(iter/s)": 1.46158 }, { "epoch": 4.37727603787327, "grad_norm": 2.206979513168335, "learning_rate": 3.778707285467198e-06, "loss": 0.38041462898254397, "memory(GiB)": 77.59, "step": 102170, "token_acc": 0.9037037037037037, "train_speed(iter/s)": 1.461592 }, { "epoch": 4.377490253202519, "grad_norm": 3.403134822845459, "learning_rate": 3.776141225629659e-06, "loss": 0.44957818984985354, "memory(GiB)": 77.59, "step": 102175, "token_acc": 0.9042145593869731, "train_speed(iter/s)": 1.461595 }, { "epoch": 4.3777044685317685, "grad_norm": 2.288938283920288, "learning_rate": 3.77357600318069e-06, "loss": 0.3415544509887695, "memory(GiB)": 77.59, "step": 102180, "token_acc": 0.9361702127659575, "train_speed(iter/s)": 1.461597 }, { "epoch": 4.377918683861017, "grad_norm": 1.7407089471817017, "learning_rate": 3.7710116181667422e-06, "loss": 0.23670122623443604, "memory(GiB)": 77.59, "step": 102185, "token_acc": 0.96, "train_speed(iter/s)": 1.461604 }, { "epoch": 4.378132899190266, "grad_norm": 5.118200778961182, "learning_rate": 3.7684480706343005e-06, "loss": 0.2710652589797974, "memory(GiB)": 77.59, "step": 102190, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.461604 }, { "epoch": 4.378347114519515, "grad_norm": 2.9111225605010986, "learning_rate": 3.765885360629784e-06, "loss": 0.39656968116760255, "memory(GiB)": 77.59, "step": 102195, "token_acc": 0.8962536023054755, "train_speed(iter/s)": 1.461613 }, { "epoch": 4.378561329848764, "grad_norm": 3.7352962493896484, "learning_rate": 3.7633234881996216e-06, "loss": 0.4090108871459961, "memory(GiB)": 77.59, "step": 102200, "token_acc": 0.9064748201438849, "train_speed(iter/s)": 1.461615 }, { "epoch": 4.378775545178013, "grad_norm": 5.277361869812012, "learning_rate": 3.760762453390232e-06, "loss": 0.5274491786956788, "memory(GiB)": 77.59, "step": 102205, "token_acc": 0.8618181818181818, "train_speed(iter/s)": 1.461624 }, { "epoch": 4.378989760507262, "grad_norm": 3.0570034980773926, "learning_rate": 3.7582022562480112e-06, "loss": 0.22461531162261963, "memory(GiB)": 77.59, "step": 102210, "token_acc": 0.9391634980988594, "train_speed(iter/s)": 1.461626 }, { "epoch": 4.379203975836511, "grad_norm": 3.483116865158081, "learning_rate": 3.7556428968193335e-06, "loss": 0.31414659023284913, "memory(GiB)": 77.59, "step": 102215, "token_acc": 0.9124087591240876, "train_speed(iter/s)": 1.461636 }, { "epoch": 4.37941819116576, "grad_norm": 1.7668254375457764, "learning_rate": 3.7530843751505617e-06, "loss": 0.3265772581100464, "memory(GiB)": 77.59, "step": 102220, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.461638 }, { "epoch": 4.379632406495009, "grad_norm": 7.588313579559326, "learning_rate": 3.7505266912880423e-06, "loss": 0.4738770484924316, "memory(GiB)": 77.59, "step": 102225, "token_acc": 0.9125475285171103, "train_speed(iter/s)": 1.461639 }, { "epoch": 4.3798466218242575, "grad_norm": 3.910344362258911, "learning_rate": 3.7479698452781264e-06, "loss": 0.3542048692703247, "memory(GiB)": 77.59, "step": 102230, "token_acc": 0.9190283400809717, "train_speed(iter/s)": 1.461639 }, { "epoch": 4.380060837153507, "grad_norm": 4.197340488433838, "learning_rate": 3.7454138371671275e-06, "loss": 0.3768757343292236, "memory(GiB)": 77.59, "step": 102235, "token_acc": 0.9121813031161473, "train_speed(iter/s)": 1.461647 }, { "epoch": 4.380275052482756, "grad_norm": 2.8908355236053467, "learning_rate": 3.742858667001342e-06, "loss": 0.33346781730651853, "memory(GiB)": 77.59, "step": 102240, "token_acc": 0.9399141630901288, "train_speed(iter/s)": 1.461649 }, { "epoch": 4.380489267812004, "grad_norm": 2.5546483993530273, "learning_rate": 3.7403043348270716e-06, "loss": 0.38528740406036377, "memory(GiB)": 77.59, "step": 102245, "token_acc": 0.9142011834319527, "train_speed(iter/s)": 1.461652 }, { "epoch": 4.380703483141254, "grad_norm": 0.21504853665828705, "learning_rate": 3.737750840690579e-06, "loss": 0.2248504638671875, "memory(GiB)": 77.59, "step": 102250, "token_acc": 0.9469387755102041, "train_speed(iter/s)": 1.461657 }, { "epoch": 4.380917698470503, "grad_norm": 3.3261919021606445, "learning_rate": 3.735198184638117e-06, "loss": 0.29215245246887206, "memory(GiB)": 77.59, "step": 102255, "token_acc": 0.9368421052631579, "train_speed(iter/s)": 1.461662 }, { "epoch": 4.381131913799751, "grad_norm": 2.1280808448791504, "learning_rate": 3.732646366715942e-06, "loss": 0.19314701557159425, "memory(GiB)": 77.59, "step": 102260, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.46166 }, { "epoch": 4.381346129129001, "grad_norm": 4.255247116088867, "learning_rate": 3.7300953869702894e-06, "loss": 0.2570141315460205, "memory(GiB)": 77.59, "step": 102265, "token_acc": 0.9490196078431372, "train_speed(iter/s)": 1.461661 }, { "epoch": 4.38156034445825, "grad_norm": 5.97975492477417, "learning_rate": 3.7275452454473613e-06, "loss": 0.6265383243560791, "memory(GiB)": 77.59, "step": 102270, "token_acc": 0.8812260536398467, "train_speed(iter/s)": 1.461662 }, { "epoch": 4.381774559787498, "grad_norm": 4.606752872467041, "learning_rate": 3.72499594219336e-06, "loss": 0.32615563869476316, "memory(GiB)": 77.59, "step": 102275, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.461667 }, { "epoch": 4.3819887751167474, "grad_norm": 8.472163200378418, "learning_rate": 3.7224474772544706e-06, "loss": 0.25363898277282715, "memory(GiB)": 77.59, "step": 102280, "token_acc": 0.9336099585062241, "train_speed(iter/s)": 1.461667 }, { "epoch": 4.382202990445997, "grad_norm": 5.281009674072266, "learning_rate": 3.7198998506768444e-06, "loss": 0.32066295146942136, "memory(GiB)": 77.59, "step": 102285, "token_acc": 0.9321533923303835, "train_speed(iter/s)": 1.461668 }, { "epoch": 4.382417205775245, "grad_norm": 4.07383394241333, "learning_rate": 3.717353062506651e-06, "loss": 0.3818878412246704, "memory(GiB)": 77.59, "step": 102290, "token_acc": 0.9049079754601227, "train_speed(iter/s)": 1.46167 }, { "epoch": 4.382631421104494, "grad_norm": 4.343846797943115, "learning_rate": 3.714807112790031e-06, "loss": 0.5380324363708496, "memory(GiB)": 77.59, "step": 102295, "token_acc": 0.8942598187311178, "train_speed(iter/s)": 1.461673 }, { "epoch": 4.382845636433744, "grad_norm": 1.8611286878585815, "learning_rate": 3.7122620015730967e-06, "loss": 0.1895733118057251, "memory(GiB)": 77.59, "step": 102300, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.461677 }, { "epoch": 4.383059851762992, "grad_norm": 0.7110298275947571, "learning_rate": 3.709717728901957e-06, "loss": 0.28327958583831786, "memory(GiB)": 77.59, "step": 102305, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.461682 }, { "epoch": 4.383274067092241, "grad_norm": 5.127408981323242, "learning_rate": 3.707174294822702e-06, "loss": 0.35574390888214114, "memory(GiB)": 77.59, "step": 102310, "token_acc": 0.9277566539923955, "train_speed(iter/s)": 1.461685 }, { "epoch": 4.3834882824214905, "grad_norm": 2.908517360687256, "learning_rate": 3.7046316993814058e-06, "loss": 0.30460076332092284, "memory(GiB)": 77.59, "step": 102315, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.461684 }, { "epoch": 4.383702497750739, "grad_norm": 2.281994342803955, "learning_rate": 3.702089942624143e-06, "loss": 0.2784027338027954, "memory(GiB)": 77.59, "step": 102320, "token_acc": 0.9311377245508982, "train_speed(iter/s)": 1.461695 }, { "epoch": 4.383916713079988, "grad_norm": 2.508723020553589, "learning_rate": 3.699549024596949e-06, "loss": 0.34990129470825193, "memory(GiB)": 77.59, "step": 102325, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.461693 }, { "epoch": 4.384130928409237, "grad_norm": 3.4977517127990723, "learning_rate": 3.6970089453458534e-06, "loss": 0.25562572479248047, "memory(GiB)": 77.59, "step": 102330, "token_acc": 0.9494949494949495, "train_speed(iter/s)": 1.461703 }, { "epoch": 4.384345143738486, "grad_norm": 2.948971748352051, "learning_rate": 3.6944697049168807e-06, "loss": 0.5267543315887451, "memory(GiB)": 77.59, "step": 102335, "token_acc": 0.8648648648648649, "train_speed(iter/s)": 1.461704 }, { "epoch": 4.384559359067735, "grad_norm": 0.2734977900981903, "learning_rate": 3.6919313033560278e-06, "loss": 0.1963382840156555, "memory(GiB)": 77.59, "step": 102340, "token_acc": 0.9551282051282052, "train_speed(iter/s)": 1.461706 }, { "epoch": 4.384773574396984, "grad_norm": 3.365936756134033, "learning_rate": 3.689393740709274e-06, "loss": 0.4224372863769531, "memory(GiB)": 77.59, "step": 102345, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.461707 }, { "epoch": 4.384987789726233, "grad_norm": 2.464747428894043, "learning_rate": 3.686857017022605e-06, "loss": 0.3576180458068848, "memory(GiB)": 77.59, "step": 102350, "token_acc": 0.9096209912536443, "train_speed(iter/s)": 1.461705 }, { "epoch": 4.385202005055482, "grad_norm": 5.809988021850586, "learning_rate": 3.684321132341961e-06, "loss": 0.4668510913848877, "memory(GiB)": 77.59, "step": 102355, "token_acc": 0.896797153024911, "train_speed(iter/s)": 1.461702 }, { "epoch": 4.385416220384731, "grad_norm": 5.292524814605713, "learning_rate": 3.681786086713296e-06, "loss": 0.3473869800567627, "memory(GiB)": 77.59, "step": 102360, "token_acc": 0.9262295081967213, "train_speed(iter/s)": 1.461705 }, { "epoch": 4.3856304357139795, "grad_norm": 3.1446986198425293, "learning_rate": 3.679251880182516e-06, "loss": 0.1258687973022461, "memory(GiB)": 77.59, "step": 102365, "token_acc": 0.9693877551020408, "train_speed(iter/s)": 1.461712 }, { "epoch": 4.385844651043229, "grad_norm": 4.462905406951904, "learning_rate": 3.6767185127955407e-06, "loss": 0.2912299633026123, "memory(GiB)": 77.59, "step": 102370, "token_acc": 0.932258064516129, "train_speed(iter/s)": 1.461718 }, { "epoch": 4.386058866372478, "grad_norm": 4.395513534545898, "learning_rate": 3.674185984598266e-06, "loss": 0.3679506301879883, "memory(GiB)": 77.59, "step": 102375, "token_acc": 0.9251968503937008, "train_speed(iter/s)": 1.461719 }, { "epoch": 4.386273081701726, "grad_norm": 3.4851160049438477, "learning_rate": 3.671654295636573e-06, "loss": 0.5018876552581787, "memory(GiB)": 77.59, "step": 102380, "token_acc": 0.8850574712643678, "train_speed(iter/s)": 1.461719 }, { "epoch": 4.386487297030976, "grad_norm": 3.5747616291046143, "learning_rate": 3.669123445956324e-06, "loss": 0.46192479133605957, "memory(GiB)": 77.59, "step": 102385, "token_acc": 0.9181494661921709, "train_speed(iter/s)": 1.461721 }, { "epoch": 4.386701512360225, "grad_norm": 4.371147632598877, "learning_rate": 3.6665934356033604e-06, "loss": 0.39331004619598386, "memory(GiB)": 77.59, "step": 102390, "token_acc": 0.926984126984127, "train_speed(iter/s)": 1.461727 }, { "epoch": 4.386915727689473, "grad_norm": 3.4904890060424805, "learning_rate": 3.6640642646235236e-06, "loss": 0.2118804931640625, "memory(GiB)": 77.59, "step": 102395, "token_acc": 0.9558823529411765, "train_speed(iter/s)": 1.461738 }, { "epoch": 4.3871299430187225, "grad_norm": 4.386895656585693, "learning_rate": 3.661535933062621e-06, "loss": 0.17630693912506104, "memory(GiB)": 77.59, "step": 102400, "token_acc": 0.9547169811320755, "train_speed(iter/s)": 1.461749 }, { "epoch": 4.387344158347972, "grad_norm": 3.317685842514038, "learning_rate": 3.6590084409664605e-06, "loss": 0.4723381996154785, "memory(GiB)": 77.59, "step": 102405, "token_acc": 0.8910505836575876, "train_speed(iter/s)": 1.461761 }, { "epoch": 4.38755837367722, "grad_norm": 3.6400418281555176, "learning_rate": 3.65648178838085e-06, "loss": 0.255877161026001, "memory(GiB)": 77.59, "step": 102410, "token_acc": 0.9476744186046512, "train_speed(iter/s)": 1.461759 }, { "epoch": 4.387772589006469, "grad_norm": 3.5361557006835938, "learning_rate": 3.653955975351536e-06, "loss": 0.43622617721557616, "memory(GiB)": 77.59, "step": 102415, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.461756 }, { "epoch": 4.387986804335719, "grad_norm": 4.188658714294434, "learning_rate": 3.651431001924294e-06, "loss": 0.46804094314575195, "memory(GiB)": 77.59, "step": 102420, "token_acc": 0.891566265060241, "train_speed(iter/s)": 1.461759 }, { "epoch": 4.388201019664967, "grad_norm": 3.9986352920532227, "learning_rate": 3.6489068681448522e-06, "loss": 0.30974640846252444, "memory(GiB)": 77.59, "step": 102425, "token_acc": 0.9318181818181818, "train_speed(iter/s)": 1.461764 }, { "epoch": 4.388415234994216, "grad_norm": 3.182584047317505, "learning_rate": 3.646383574058937e-06, "loss": 0.36996827125549314, "memory(GiB)": 77.59, "step": 102430, "token_acc": 0.926530612244898, "train_speed(iter/s)": 1.461771 }, { "epoch": 4.3886294503234655, "grad_norm": 2.0164737701416016, "learning_rate": 3.6438611197122773e-06, "loss": 0.1990032434463501, "memory(GiB)": 77.59, "step": 102435, "token_acc": 0.9568106312292359, "train_speed(iter/s)": 1.461778 }, { "epoch": 4.388843665652714, "grad_norm": 0.38420239090919495, "learning_rate": 3.641339505150554e-06, "loss": 0.2876487970352173, "memory(GiB)": 77.59, "step": 102440, "token_acc": 0.9326923076923077, "train_speed(iter/s)": 1.461788 }, { "epoch": 4.389057880981963, "grad_norm": 2.108255386352539, "learning_rate": 3.6388187304194577e-06, "loss": 0.3158128023147583, "memory(GiB)": 77.59, "step": 102445, "token_acc": 0.9151515151515152, "train_speed(iter/s)": 1.461796 }, { "epoch": 4.389272096311212, "grad_norm": 2.201242446899414, "learning_rate": 3.6362987955646468e-06, "loss": 0.2009979248046875, "memory(GiB)": 77.59, "step": 102450, "token_acc": 0.9550561797752809, "train_speed(iter/s)": 1.4618 }, { "epoch": 4.389486311640461, "grad_norm": 2.6465301513671875, "learning_rate": 3.6337797006317785e-06, "loss": 0.2753817081451416, "memory(GiB)": 77.59, "step": 102455, "token_acc": 0.9303030303030303, "train_speed(iter/s)": 1.461804 }, { "epoch": 4.38970052696971, "grad_norm": 0.2991377115249634, "learning_rate": 3.6312614456664782e-06, "loss": 0.24895410537719725, "memory(GiB)": 77.59, "step": 102460, "token_acc": 0.9413919413919414, "train_speed(iter/s)": 1.461804 }, { "epoch": 4.389914742298959, "grad_norm": 6.315512180328369, "learning_rate": 3.6287440307143816e-06, "loss": 0.6218793869018555, "memory(GiB)": 77.59, "step": 102465, "token_acc": 0.856, "train_speed(iter/s)": 1.461812 }, { "epoch": 4.390128957628208, "grad_norm": 4.499445915222168, "learning_rate": 3.6262274558210852e-06, "loss": 0.28575022220611573, "memory(GiB)": 77.59, "step": 102470, "token_acc": 0.9420289855072463, "train_speed(iter/s)": 1.461815 }, { "epoch": 4.390343172957457, "grad_norm": 3.677151918411255, "learning_rate": 3.6237117210321802e-06, "loss": 0.4551736831665039, "memory(GiB)": 77.59, "step": 102475, "token_acc": 0.9006410256410257, "train_speed(iter/s)": 1.461824 }, { "epoch": 4.390557388286706, "grad_norm": 3.969635486602783, "learning_rate": 3.621196826393236e-06, "loss": 0.4643622875213623, "memory(GiB)": 77.59, "step": 102480, "token_acc": 0.8952702702702703, "train_speed(iter/s)": 1.461829 }, { "epoch": 4.3907716036159545, "grad_norm": 1.7899168729782104, "learning_rate": 3.6186827719498272e-06, "loss": 0.2406458616256714, "memory(GiB)": 77.59, "step": 102485, "token_acc": 0.9367469879518072, "train_speed(iter/s)": 1.461832 }, { "epoch": 4.390985818945204, "grad_norm": 4.503034591674805, "learning_rate": 3.6161695577474784e-06, "loss": 0.23692913055419923, "memory(GiB)": 77.59, "step": 102490, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.461835 }, { "epoch": 4.391200034274453, "grad_norm": 2.3433620929718018, "learning_rate": 3.613657183831737e-06, "loss": 0.2792028903961182, "memory(GiB)": 77.59, "step": 102495, "token_acc": 0.9422492401215805, "train_speed(iter/s)": 1.461837 }, { "epoch": 4.391414249603701, "grad_norm": 0.7329298257827759, "learning_rate": 3.61114565024811e-06, "loss": 0.24666173458099366, "memory(GiB)": 77.59, "step": 102500, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.461839 }, { "epoch": 4.391414249603701, "eval_loss": 2.391770839691162, "eval_runtime": 11.8328, "eval_samples_per_second": 8.451, "eval_steps_per_second": 8.451, "eval_token_acc": 0.48493543758967, "step": 102500 }, { "epoch": 4.391628464932951, "grad_norm": 2.048659563064575, "learning_rate": 3.608634957042095e-06, "loss": 0.16136748790740968, "memory(GiB)": 77.59, "step": 102505, "token_acc": 0.6058002148227712, "train_speed(iter/s)": 1.461572 }, { "epoch": 4.3918426802622, "grad_norm": 2.333728551864624, "learning_rate": 3.6061251042591837e-06, "loss": 0.33780612945556643, "memory(GiB)": 77.59, "step": 102510, "token_acc": 0.9146757679180887, "train_speed(iter/s)": 1.461584 }, { "epoch": 4.392056895591448, "grad_norm": 1.9740017652511597, "learning_rate": 3.6036160919448226e-06, "loss": 0.32034764289855955, "memory(GiB)": 77.59, "step": 102515, "token_acc": 0.8913043478260869, "train_speed(iter/s)": 1.461582 }, { "epoch": 4.3922711109206976, "grad_norm": 7.204108715057373, "learning_rate": 3.6011079201444865e-06, "loss": 0.32154016494750975, "memory(GiB)": 77.59, "step": 102520, "token_acc": 0.9359430604982206, "train_speed(iter/s)": 1.461584 }, { "epoch": 4.392485326249947, "grad_norm": 6.103866100311279, "learning_rate": 3.598600588903611e-06, "loss": 0.3159912109375, "memory(GiB)": 77.59, "step": 102525, "token_acc": 0.9103448275862069, "train_speed(iter/s)": 1.461595 }, { "epoch": 4.392699541579195, "grad_norm": 1.4474844932556152, "learning_rate": 3.5960940982676105e-06, "loss": 0.2742654800415039, "memory(GiB)": 77.59, "step": 102530, "token_acc": 0.9214876033057852, "train_speed(iter/s)": 1.461604 }, { "epoch": 4.392913756908444, "grad_norm": 2.4413344860076904, "learning_rate": 3.593588448281893e-06, "loss": 0.26729907989501955, "memory(GiB)": 77.59, "step": 102535, "token_acc": 0.9409937888198758, "train_speed(iter/s)": 1.461605 }, { "epoch": 4.393127972237694, "grad_norm": 0.45717892050743103, "learning_rate": 3.59108363899186e-06, "loss": 0.23000712394714357, "memory(GiB)": 77.59, "step": 102540, "token_acc": 0.9546925566343042, "train_speed(iter/s)": 1.461607 }, { "epoch": 4.393342187566942, "grad_norm": 3.25498104095459, "learning_rate": 3.5885796704428707e-06, "loss": 0.4622795581817627, "memory(GiB)": 77.59, "step": 102545, "token_acc": 0.8637602179836512, "train_speed(iter/s)": 1.461608 }, { "epoch": 4.393556402896191, "grad_norm": 5.0804877281188965, "learning_rate": 3.5860765426803056e-06, "loss": 0.4416989326477051, "memory(GiB)": 77.59, "step": 102550, "token_acc": 0.919732441471572, "train_speed(iter/s)": 1.461613 }, { "epoch": 4.393770618225441, "grad_norm": 1.5724871158599854, "learning_rate": 3.5835742557494943e-06, "loss": 0.19895944595336915, "memory(GiB)": 77.59, "step": 102555, "token_acc": 0.9498525073746312, "train_speed(iter/s)": 1.461616 }, { "epoch": 4.393984833554689, "grad_norm": 2.2991015911102295, "learning_rate": 3.58107280969579e-06, "loss": 0.23524737358093262, "memory(GiB)": 77.59, "step": 102560, "token_acc": 0.9237668161434978, "train_speed(iter/s)": 1.461626 }, { "epoch": 4.394199048883938, "grad_norm": 3.0760507583618164, "learning_rate": 3.5785722045644898e-06, "loss": 0.3175687313079834, "memory(GiB)": 77.59, "step": 102565, "token_acc": 0.9174603174603174, "train_speed(iter/s)": 1.461622 }, { "epoch": 4.3944132642131875, "grad_norm": 1.947813868522644, "learning_rate": 3.576072440400907e-06, "loss": 0.30246145725250245, "memory(GiB)": 77.59, "step": 102570, "token_acc": 0.9383561643835616, "train_speed(iter/s)": 1.461622 }, { "epoch": 4.394627479542436, "grad_norm": 2.195233106613159, "learning_rate": 3.5735735172503117e-06, "loss": 0.3405130624771118, "memory(GiB)": 77.59, "step": 102575, "token_acc": 0.9246031746031746, "train_speed(iter/s)": 1.46163 }, { "epoch": 4.394841694871685, "grad_norm": 5.573729515075684, "learning_rate": 3.571075435157989e-06, "loss": 0.52567138671875, "memory(GiB)": 77.59, "step": 102580, "token_acc": 0.8847583643122676, "train_speed(iter/s)": 1.46164 }, { "epoch": 4.395055910200934, "grad_norm": 5.073814868927002, "learning_rate": 3.568578194169192e-06, "loss": 0.48814144134521487, "memory(GiB)": 77.59, "step": 102585, "token_acc": 0.8959731543624161, "train_speed(iter/s)": 1.461641 }, { "epoch": 4.395270125530183, "grad_norm": 1.3312805891036987, "learning_rate": 3.566081794329157e-06, "loss": 0.1289984703063965, "memory(GiB)": 77.59, "step": 102590, "token_acc": 0.9673202614379085, "train_speed(iter/s)": 1.461638 }, { "epoch": 4.395484340859432, "grad_norm": 1.4388551712036133, "learning_rate": 3.563586235683103e-06, "loss": 0.3014682769775391, "memory(GiB)": 77.59, "step": 102595, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.461644 }, { "epoch": 4.395698556188681, "grad_norm": 0.7966964840888977, "learning_rate": 3.5610915182762492e-06, "loss": 0.23057498931884765, "memory(GiB)": 77.59, "step": 102600, "token_acc": 0.9456521739130435, "train_speed(iter/s)": 1.461644 }, { "epoch": 4.39591277151793, "grad_norm": 2.314239501953125, "learning_rate": 3.5585976421537825e-06, "loss": 0.21758484840393066, "memory(GiB)": 77.59, "step": 102605, "token_acc": 0.9509433962264151, "train_speed(iter/s)": 1.461647 }, { "epoch": 4.396126986847179, "grad_norm": 0.8191334009170532, "learning_rate": 3.556104607360888e-06, "loss": 0.3145124912261963, "memory(GiB)": 77.59, "step": 102610, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.461648 }, { "epoch": 4.396341202176428, "grad_norm": 3.5111074447631836, "learning_rate": 3.553612413942725e-06, "loss": 0.38841824531555175, "memory(GiB)": 77.59, "step": 102615, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.461648 }, { "epoch": 4.3965554175056765, "grad_norm": 2.1641955375671387, "learning_rate": 3.551121061944446e-06, "loss": 0.2914820671081543, "memory(GiB)": 77.59, "step": 102620, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.46165 }, { "epoch": 4.396769632834926, "grad_norm": 1.7196301221847534, "learning_rate": 3.5486305514111805e-06, "loss": 0.521583890914917, "memory(GiB)": 77.59, "step": 102625, "token_acc": 0.8884758364312267, "train_speed(iter/s)": 1.461658 }, { "epoch": 4.396983848164175, "grad_norm": 2.36441969871521, "learning_rate": 3.5461408823880437e-06, "loss": 0.32347052097320556, "memory(GiB)": 77.59, "step": 102630, "token_acc": 0.9288025889967637, "train_speed(iter/s)": 1.461665 }, { "epoch": 4.397198063493423, "grad_norm": 0.19503332674503326, "learning_rate": 3.543652054920138e-06, "loss": 0.18135302066802977, "memory(GiB)": 77.59, "step": 102635, "token_acc": 0.9592592592592593, "train_speed(iter/s)": 1.46167 }, { "epoch": 4.397412278822673, "grad_norm": 5.698501110076904, "learning_rate": 3.5411640690525606e-06, "loss": 0.41170196533203124, "memory(GiB)": 77.59, "step": 102640, "token_acc": 0.9227941176470589, "train_speed(iter/s)": 1.461688 }, { "epoch": 4.397626494151922, "grad_norm": 0.5087948441505432, "learning_rate": 3.538676924830381e-06, "loss": 0.39943063259124756, "memory(GiB)": 77.59, "step": 102645, "token_acc": 0.9297752808988764, "train_speed(iter/s)": 1.461695 }, { "epoch": 4.39784070948117, "grad_norm": 4.089905261993408, "learning_rate": 3.5361906222986517e-06, "loss": 0.4969949722290039, "memory(GiB)": 77.59, "step": 102650, "token_acc": 0.8931034482758621, "train_speed(iter/s)": 1.461697 }, { "epoch": 4.3980549248104195, "grad_norm": 3.7008562088012695, "learning_rate": 3.5337051615024207e-06, "loss": 0.44532098770141604, "memory(GiB)": 77.59, "step": 102655, "token_acc": 0.9114754098360656, "train_speed(iter/s)": 1.461717 }, { "epoch": 4.398269140139669, "grad_norm": 4.344244003295898, "learning_rate": 3.531220542486696e-06, "loss": 0.27587854862213135, "memory(GiB)": 77.59, "step": 102660, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.46172 }, { "epoch": 4.398483355468917, "grad_norm": 4.399242877960205, "learning_rate": 3.528736765296514e-06, "loss": 0.37762112617492677, "memory(GiB)": 77.59, "step": 102665, "token_acc": 0.9150326797385621, "train_speed(iter/s)": 1.46172 }, { "epoch": 4.398697570798166, "grad_norm": 5.2246551513671875, "learning_rate": 3.5262538299768553e-06, "loss": 0.46307907104492185, "memory(GiB)": 77.59, "step": 102670, "token_acc": 0.8933333333333333, "train_speed(iter/s)": 1.461717 }, { "epoch": 4.398911786127416, "grad_norm": 2.062452554702759, "learning_rate": 3.5237717365727065e-06, "loss": 0.35749657154083253, "memory(GiB)": 77.59, "step": 102675, "token_acc": 0.9230769230769231, "train_speed(iter/s)": 1.461726 }, { "epoch": 4.399126001456664, "grad_norm": 2.472275972366333, "learning_rate": 3.521290485129031e-06, "loss": 0.3750325918197632, "memory(GiB)": 77.59, "step": 102680, "token_acc": 0.9076923076923077, "train_speed(iter/s)": 1.461725 }, { "epoch": 4.399340216785913, "grad_norm": 1.8292391300201416, "learning_rate": 3.5188100756907815e-06, "loss": 0.20286123752593993, "memory(GiB)": 77.59, "step": 102685, "token_acc": 0.9392265193370166, "train_speed(iter/s)": 1.461727 }, { "epoch": 4.3995544321151625, "grad_norm": 3.6729907989501953, "learning_rate": 3.5163305083028785e-06, "loss": 0.2172644853591919, "memory(GiB)": 77.59, "step": 102690, "token_acc": 0.950381679389313, "train_speed(iter/s)": 1.461737 }, { "epoch": 4.399768647444411, "grad_norm": 3.2817442417144775, "learning_rate": 3.5138517830102635e-06, "loss": 0.2413226842880249, "memory(GiB)": 77.59, "step": 102695, "token_acc": 0.9525316455696202, "train_speed(iter/s)": 1.46174 }, { "epoch": 4.39998286277366, "grad_norm": 8.16904354095459, "learning_rate": 3.5113738998578227e-06, "loss": 0.3455121278762817, "memory(GiB)": 77.59, "step": 102700, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.461736 }, { "epoch": 4.400197078102909, "grad_norm": 0.735561728477478, "learning_rate": 3.5088968588904593e-06, "loss": 0.16072510480880736, "memory(GiB)": 77.59, "step": 102705, "token_acc": 0.95625, "train_speed(iter/s)": 1.46174 }, { "epoch": 4.400411293432158, "grad_norm": 3.969001054763794, "learning_rate": 3.5064206601530423e-06, "loss": 0.44601898193359374, "memory(GiB)": 77.59, "step": 102710, "token_acc": 0.9, "train_speed(iter/s)": 1.461742 }, { "epoch": 4.400625508761407, "grad_norm": 10.589555740356445, "learning_rate": 3.503945303690431e-06, "loss": 0.38328895568847654, "memory(GiB)": 77.59, "step": 102715, "token_acc": 0.9107806691449815, "train_speed(iter/s)": 1.461741 }, { "epoch": 4.400839724090656, "grad_norm": 1.2499158382415771, "learning_rate": 3.501470789547462e-06, "loss": 0.2574322700500488, "memory(GiB)": 77.59, "step": 102720, "token_acc": 0.9442622950819672, "train_speed(iter/s)": 1.461745 }, { "epoch": 4.401053939419905, "grad_norm": 2.580305337905884, "learning_rate": 3.4989971177689707e-06, "loss": 0.2892350196838379, "memory(GiB)": 77.59, "step": 102725, "token_acc": 0.9480968858131488, "train_speed(iter/s)": 1.461743 }, { "epoch": 4.401268154749154, "grad_norm": 3.7144367694854736, "learning_rate": 3.496524288399772e-06, "loss": 0.29452626705169677, "memory(GiB)": 77.59, "step": 102730, "token_acc": 0.9243986254295533, "train_speed(iter/s)": 1.461745 }, { "epoch": 4.401482370078403, "grad_norm": 1.9714329242706299, "learning_rate": 3.4940523014846572e-06, "loss": 0.2516180515289307, "memory(GiB)": 77.59, "step": 102735, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.461744 }, { "epoch": 4.4016965854076515, "grad_norm": 2.675178050994873, "learning_rate": 3.491581157068413e-06, "loss": 0.2721111536026001, "memory(GiB)": 77.59, "step": 102740, "token_acc": 0.9301470588235294, "train_speed(iter/s)": 1.461741 }, { "epoch": 4.401910800736901, "grad_norm": 2.230192184448242, "learning_rate": 3.4891108551958042e-06, "loss": 0.3359892129898071, "memory(GiB)": 77.59, "step": 102745, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.461741 }, { "epoch": 4.40212501606615, "grad_norm": 7.714084148406982, "learning_rate": 3.4866413959115774e-06, "loss": 0.4552027702331543, "memory(GiB)": 77.59, "step": 102750, "token_acc": 0.9291338582677166, "train_speed(iter/s)": 1.461742 }, { "epoch": 4.402339231395398, "grad_norm": 5.7099432945251465, "learning_rate": 3.4841727792604805e-06, "loss": 0.5711403846740722, "memory(GiB)": 77.59, "step": 102755, "token_acc": 0.8896321070234113, "train_speed(iter/s)": 1.461748 }, { "epoch": 4.402553446724648, "grad_norm": 4.2626471519470215, "learning_rate": 3.481705005287228e-06, "loss": 0.42557363510131835, "memory(GiB)": 77.59, "step": 102760, "token_acc": 0.9020979020979021, "train_speed(iter/s)": 1.46175 }, { "epoch": 4.402767662053897, "grad_norm": 2.5050432682037354, "learning_rate": 3.479238074036528e-06, "loss": 0.4330791473388672, "memory(GiB)": 77.59, "step": 102765, "token_acc": 0.9182156133828996, "train_speed(iter/s)": 1.461753 }, { "epoch": 4.402981877383145, "grad_norm": 2.3109946250915527, "learning_rate": 3.476771985553068e-06, "loss": 0.21892199516296387, "memory(GiB)": 77.59, "step": 102770, "token_acc": 0.9387096774193548, "train_speed(iter/s)": 1.461759 }, { "epoch": 4.4031960927123945, "grad_norm": 3.184223175048828, "learning_rate": 3.474306739881522e-06, "loss": 0.265102219581604, "memory(GiB)": 77.59, "step": 102775, "token_acc": 0.9360902255639098, "train_speed(iter/s)": 1.461757 }, { "epoch": 4.403410308041644, "grad_norm": 6.278067588806152, "learning_rate": 3.47184233706655e-06, "loss": 0.46890668869018554, "memory(GiB)": 77.59, "step": 102780, "token_acc": 0.9104938271604939, "train_speed(iter/s)": 1.461764 }, { "epoch": 4.403624523370892, "grad_norm": 1.0404518842697144, "learning_rate": 3.469378777152815e-06, "loss": 0.5796900749206543, "memory(GiB)": 77.59, "step": 102785, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.461764 }, { "epoch": 4.403838738700141, "grad_norm": 3.568281650543213, "learning_rate": 3.4669160601849272e-06, "loss": 0.3837564945220947, "memory(GiB)": 77.59, "step": 102790, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.461777 }, { "epoch": 4.404052954029391, "grad_norm": 3.9995932579040527, "learning_rate": 3.4644541862075054e-06, "loss": 0.26158952713012695, "memory(GiB)": 77.59, "step": 102795, "token_acc": 0.9538461538461539, "train_speed(iter/s)": 1.461775 }, { "epoch": 4.404267169358639, "grad_norm": 2.556391477584839, "learning_rate": 3.461993155265153e-06, "loss": 0.4434248447418213, "memory(GiB)": 77.59, "step": 102800, "token_acc": 0.8849840255591054, "train_speed(iter/s)": 1.461778 }, { "epoch": 4.404481384687888, "grad_norm": 5.3142266273498535, "learning_rate": 3.4595329674024456e-06, "loss": 0.5525129795074463, "memory(GiB)": 77.59, "step": 102805, "token_acc": 0.8643533123028391, "train_speed(iter/s)": 1.461776 }, { "epoch": 4.404695600017138, "grad_norm": 3.071697473526001, "learning_rate": 3.4570736226639588e-06, "loss": 0.527379035949707, "memory(GiB)": 77.59, "step": 102810, "token_acc": 0.8947368421052632, "train_speed(iter/s)": 1.461782 }, { "epoch": 4.404909815346386, "grad_norm": 4.741899490356445, "learning_rate": 3.454615121094251e-06, "loss": 0.37799170017242434, "memory(GiB)": 77.59, "step": 102815, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.461785 }, { "epoch": 4.405124030675635, "grad_norm": 5.575362205505371, "learning_rate": 3.4521574627378485e-06, "loss": 0.366666841506958, "memory(GiB)": 77.59, "step": 102820, "token_acc": 0.9426229508196722, "train_speed(iter/s)": 1.461796 }, { "epoch": 4.405338246004884, "grad_norm": 3.4542908668518066, "learning_rate": 3.4497006476392813e-06, "loss": 0.12348350286483764, "memory(GiB)": 77.59, "step": 102825, "token_acc": 0.967032967032967, "train_speed(iter/s)": 1.461794 }, { "epoch": 4.405552461334133, "grad_norm": 0.19625428318977356, "learning_rate": 3.4472446758430533e-06, "loss": 0.13078209161758422, "memory(GiB)": 77.59, "step": 102830, "token_acc": 0.9535714285714286, "train_speed(iter/s)": 1.461797 }, { "epoch": 4.405766676663382, "grad_norm": 2.357938289642334, "learning_rate": 3.4447895473936453e-06, "loss": 0.221488356590271, "memory(GiB)": 77.59, "step": 102835, "token_acc": 0.9513677811550152, "train_speed(iter/s)": 1.461801 }, { "epoch": 4.405980891992631, "grad_norm": 1.2171576023101807, "learning_rate": 3.442335262335561e-06, "loss": 0.21684799194335938, "memory(GiB)": 77.59, "step": 102840, "token_acc": 0.9526813880126183, "train_speed(iter/s)": 1.461799 }, { "epoch": 4.40619510732188, "grad_norm": 3.8824527263641357, "learning_rate": 3.439881820713248e-06, "loss": 0.5155314922332763, "memory(GiB)": 77.59, "step": 102845, "token_acc": 0.9025270758122743, "train_speed(iter/s)": 1.461796 }, { "epoch": 4.406409322651129, "grad_norm": 7.070133209228516, "learning_rate": 3.4374292225711423e-06, "loss": 0.48640899658203124, "memory(GiB)": 77.59, "step": 102850, "token_acc": 0.9175257731958762, "train_speed(iter/s)": 1.46179 }, { "epoch": 4.406623537980378, "grad_norm": 3.22446870803833, "learning_rate": 3.434977467953693e-06, "loss": 0.3248149394989014, "memory(GiB)": 77.59, "step": 102855, "token_acc": 0.935374149659864, "train_speed(iter/s)": 1.461792 }, { "epoch": 4.406837753309627, "grad_norm": 2.880044460296631, "learning_rate": 3.4325265569053134e-06, "loss": 0.2435028076171875, "memory(GiB)": 77.59, "step": 102860, "token_acc": 0.9566929133858267, "train_speed(iter/s)": 1.461796 }, { "epoch": 4.407051968638876, "grad_norm": 4.4944658279418945, "learning_rate": 3.430076489470385e-06, "loss": 0.1662289619445801, "memory(GiB)": 77.59, "step": 102865, "token_acc": 0.9552845528455285, "train_speed(iter/s)": 1.4618 }, { "epoch": 4.407266183968125, "grad_norm": 4.581160545349121, "learning_rate": 3.4276272656933174e-06, "loss": 0.4981378078460693, "memory(GiB)": 77.59, "step": 102870, "token_acc": 0.8928571428571429, "train_speed(iter/s)": 1.461803 }, { "epoch": 4.4074803992973735, "grad_norm": 6.126906871795654, "learning_rate": 3.425178885618474e-06, "loss": 0.31872031688690183, "memory(GiB)": 77.59, "step": 102875, "token_acc": 0.9328621908127208, "train_speed(iter/s)": 1.461811 }, { "epoch": 4.407694614626623, "grad_norm": 4.083532333374023, "learning_rate": 3.422731349290198e-06, "loss": 0.4504429340362549, "memory(GiB)": 77.59, "step": 102880, "token_acc": 0.8993288590604027, "train_speed(iter/s)": 1.461816 }, { "epoch": 4.407908829955872, "grad_norm": 5.509454250335693, "learning_rate": 3.4202846567528424e-06, "loss": 0.3151030302047729, "memory(GiB)": 77.59, "step": 102885, "token_acc": 0.935361216730038, "train_speed(iter/s)": 1.461823 }, { "epoch": 4.40812304528512, "grad_norm": 0.6741588115692139, "learning_rate": 3.4178388080507274e-06, "loss": 0.408967924118042, "memory(GiB)": 77.59, "step": 102890, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.46183 }, { "epoch": 4.40833726061437, "grad_norm": 4.3186492919921875, "learning_rate": 3.4153938032281454e-06, "loss": 0.34622702598571775, "memory(GiB)": 77.59, "step": 102895, "token_acc": 0.915129151291513, "train_speed(iter/s)": 1.461848 }, { "epoch": 4.408551475943619, "grad_norm": 3.7934188842773438, "learning_rate": 3.412949642329416e-06, "loss": 0.3713581085205078, "memory(GiB)": 77.59, "step": 102900, "token_acc": 0.9228187919463087, "train_speed(iter/s)": 1.461851 }, { "epoch": 4.408765691272867, "grad_norm": 2.926438570022583, "learning_rate": 3.4105063253988045e-06, "loss": 0.5453434944152832, "memory(GiB)": 77.59, "step": 102905, "token_acc": 0.8857142857142857, "train_speed(iter/s)": 1.461852 }, { "epoch": 4.4089799066021165, "grad_norm": 1.5377345085144043, "learning_rate": 3.408063852480581e-06, "loss": 0.29931910037994386, "memory(GiB)": 77.59, "step": 102910, "token_acc": 0.950354609929078, "train_speed(iter/s)": 1.461861 }, { "epoch": 4.409194121931366, "grad_norm": 1.7039809226989746, "learning_rate": 3.4056222236189815e-06, "loss": 0.2642936706542969, "memory(GiB)": 77.59, "step": 102915, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.461863 }, { "epoch": 4.409408337260614, "grad_norm": 3.1970646381378174, "learning_rate": 3.4031814388582384e-06, "loss": 0.3521300792694092, "memory(GiB)": 77.59, "step": 102920, "token_acc": 0.93, "train_speed(iter/s)": 1.461874 }, { "epoch": 4.409622552589863, "grad_norm": 0.3906015455722809, "learning_rate": 3.4007414982425766e-06, "loss": 0.2929401397705078, "memory(GiB)": 77.59, "step": 102925, "token_acc": 0.9304029304029304, "train_speed(iter/s)": 1.461881 }, { "epoch": 4.409836767919113, "grad_norm": 3.964409828186035, "learning_rate": 3.3983024018162004e-06, "loss": 0.2177652359008789, "memory(GiB)": 77.59, "step": 102930, "token_acc": 0.9392857142857143, "train_speed(iter/s)": 1.461883 }, { "epoch": 4.410050983248361, "grad_norm": 1.3202191591262817, "learning_rate": 3.395864149623296e-06, "loss": 0.1795314908027649, "memory(GiB)": 77.59, "step": 102935, "token_acc": 0.9663299663299664, "train_speed(iter/s)": 1.461888 }, { "epoch": 4.41026519857761, "grad_norm": 4.120957374572754, "learning_rate": 3.393426741708028e-06, "loss": 0.3244185447692871, "memory(GiB)": 77.59, "step": 102940, "token_acc": 0.915057915057915, "train_speed(iter/s)": 1.461895 }, { "epoch": 4.4104794139068595, "grad_norm": 4.7011027336120605, "learning_rate": 3.390990178114556e-06, "loss": 0.23629226684570312, "memory(GiB)": 77.59, "step": 102945, "token_acc": 0.9501915708812261, "train_speed(iter/s)": 1.461896 }, { "epoch": 4.410693629236108, "grad_norm": 1.2399704456329346, "learning_rate": 3.388554458887011e-06, "loss": 0.24199190139770507, "memory(GiB)": 77.59, "step": 102950, "token_acc": 0.9515418502202643, "train_speed(iter/s)": 1.461894 }, { "epoch": 4.410907844565357, "grad_norm": 3.435046672821045, "learning_rate": 3.386119584069536e-06, "loss": 0.39194607734680176, "memory(GiB)": 77.59, "step": 102955, "token_acc": 0.9292929292929293, "train_speed(iter/s)": 1.461895 }, { "epoch": 4.411122059894606, "grad_norm": 2.5668387413024902, "learning_rate": 3.383685553706234e-06, "loss": 0.41835894584655764, "memory(GiB)": 77.59, "step": 102960, "token_acc": 0.9185667752442996, "train_speed(iter/s)": 1.461895 }, { "epoch": 4.411336275223855, "grad_norm": 1.9600189924240112, "learning_rate": 3.381252367841192e-06, "loss": 0.2585024356842041, "memory(GiB)": 77.59, "step": 102965, "token_acc": 0.9370860927152318, "train_speed(iter/s)": 1.461898 }, { "epoch": 4.411550490553104, "grad_norm": 0.07739300280809402, "learning_rate": 3.3788200265185033e-06, "loss": 0.18992601633071898, "memory(GiB)": 77.59, "step": 102970, "token_acc": 0.9501557632398754, "train_speed(iter/s)": 1.461898 }, { "epoch": 4.411764705882353, "grad_norm": 1.6399861574172974, "learning_rate": 3.376388529782215e-06, "loss": 0.3478264570236206, "memory(GiB)": 77.59, "step": 102975, "token_acc": 0.9236363636363636, "train_speed(iter/s)": 1.461904 }, { "epoch": 4.411978921211602, "grad_norm": 1.9929097890853882, "learning_rate": 3.3739578776763814e-06, "loss": 0.48834476470947263, "memory(GiB)": 77.59, "step": 102980, "token_acc": 0.9052631578947369, "train_speed(iter/s)": 1.461908 }, { "epoch": 4.412193136540851, "grad_norm": 2.4972476959228516, "learning_rate": 3.3715280702450502e-06, "loss": 0.27957706451416015, "memory(GiB)": 77.59, "step": 102985, "token_acc": 0.95, "train_speed(iter/s)": 1.461905 }, { "epoch": 4.4124073518701, "grad_norm": 3.6444742679595947, "learning_rate": 3.3690991075322254e-06, "loss": 0.3111088752746582, "memory(GiB)": 77.59, "step": 102990, "token_acc": 0.9289940828402367, "train_speed(iter/s)": 1.461914 }, { "epoch": 4.4126215671993485, "grad_norm": 2.000309705734253, "learning_rate": 3.3666709895819104e-06, "loss": 0.3030554294586182, "memory(GiB)": 77.59, "step": 102995, "token_acc": 0.9239130434782609, "train_speed(iter/s)": 1.461915 }, { "epoch": 4.412835782528598, "grad_norm": 3.3823699951171875, "learning_rate": 3.364243716438098e-06, "loss": 0.3360884189605713, "memory(GiB)": 77.59, "step": 103000, "token_acc": 0.9384615384615385, "train_speed(iter/s)": 1.461916 }, { "epoch": 4.412835782528598, "eval_loss": 2.456714391708374, "eval_runtime": 11.4923, "eval_samples_per_second": 8.701, "eval_steps_per_second": 8.701, "eval_token_acc": 0.4541607898448519, "step": 103000 }, { "epoch": 4.413049997857847, "grad_norm": 2.750042200088501, "learning_rate": 3.361817288144764e-06, "loss": 0.48351149559020995, "memory(GiB)": 77.59, "step": 103005, "token_acc": 0.5966958211856171, "train_speed(iter/s)": 1.461659 }, { "epoch": 4.413264213187095, "grad_norm": 1.9552905559539795, "learning_rate": 3.3593917047458455e-06, "loss": 0.2918405532836914, "memory(GiB)": 77.59, "step": 103010, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.46166 }, { "epoch": 4.413478428516345, "grad_norm": 1.5616830587387085, "learning_rate": 3.3569669662853133e-06, "loss": 0.5538567066192627, "memory(GiB)": 77.59, "step": 103015, "token_acc": 0.8825622775800712, "train_speed(iter/s)": 1.461664 }, { "epoch": 4.413692643845594, "grad_norm": 2.331825017929077, "learning_rate": 3.354543072807076e-06, "loss": 0.28505094051361085, "memory(GiB)": 77.59, "step": 103020, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.461664 }, { "epoch": 4.413906859174842, "grad_norm": 0.25465670228004456, "learning_rate": 3.352120024355049e-06, "loss": 0.1870538830757141, "memory(GiB)": 77.59, "step": 103025, "token_acc": 0.9682539682539683, "train_speed(iter/s)": 1.461665 }, { "epoch": 4.4141210745040915, "grad_norm": 7.156242370605469, "learning_rate": 3.3496978209731244e-06, "loss": 0.4227184295654297, "memory(GiB)": 77.59, "step": 103030, "token_acc": 0.9163179916317992, "train_speed(iter/s)": 1.461667 }, { "epoch": 4.414335289833341, "grad_norm": 3.8174479007720947, "learning_rate": 3.3472764627051847e-06, "loss": 0.17930145263671876, "memory(GiB)": 77.59, "step": 103035, "token_acc": 0.9401709401709402, "train_speed(iter/s)": 1.461665 }, { "epoch": 4.414549505162589, "grad_norm": 4.825937271118164, "learning_rate": 3.3448559495950937e-06, "loss": 0.46810102462768555, "memory(GiB)": 77.59, "step": 103040, "token_acc": 0.8929889298892989, "train_speed(iter/s)": 1.461681 }, { "epoch": 4.414763720491838, "grad_norm": 2.7684621810913086, "learning_rate": 3.342436281686706e-06, "loss": 0.3575248718261719, "memory(GiB)": 77.59, "step": 103045, "token_acc": 0.908675799086758, "train_speed(iter/s)": 1.461685 }, { "epoch": 4.414977935821088, "grad_norm": 3.8071165084838867, "learning_rate": 3.340017459023853e-06, "loss": 0.4811088085174561, "memory(GiB)": 77.59, "step": 103050, "token_acc": 0.8895522388059701, "train_speed(iter/s)": 1.461696 }, { "epoch": 4.415192151150336, "grad_norm": 0.20457826554775238, "learning_rate": 3.3375994816503554e-06, "loss": 0.2759571552276611, "memory(GiB)": 77.59, "step": 103055, "token_acc": 0.9479553903345725, "train_speed(iter/s)": 1.461702 }, { "epoch": 4.415406366479585, "grad_norm": 3.9022748470306396, "learning_rate": 3.3351823496100167e-06, "loss": 0.34725394248962405, "memory(GiB)": 77.59, "step": 103060, "token_acc": 0.9342105263157895, "train_speed(iter/s)": 1.461703 }, { "epoch": 4.4156205818088345, "grad_norm": 1.9842668771743774, "learning_rate": 3.3327660629466186e-06, "loss": 0.3267650127410889, "memory(GiB)": 77.59, "step": 103065, "token_acc": 0.9108910891089109, "train_speed(iter/s)": 1.461703 }, { "epoch": 4.415834797138083, "grad_norm": 3.5295913219451904, "learning_rate": 3.330350621703937e-06, "loss": 0.5012635707855224, "memory(GiB)": 77.59, "step": 103070, "token_acc": 0.8927444794952681, "train_speed(iter/s)": 1.461705 }, { "epoch": 4.416049012467332, "grad_norm": 5.332789421081543, "learning_rate": 3.3279360259257486e-06, "loss": 0.4054579257965088, "memory(GiB)": 77.59, "step": 103075, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.461709 }, { "epoch": 4.416263227796581, "grad_norm": 0.11123693734407425, "learning_rate": 3.3255222756557735e-06, "loss": 0.0900408685207367, "memory(GiB)": 77.59, "step": 103080, "token_acc": 0.9747634069400631, "train_speed(iter/s)": 1.461712 }, { "epoch": 4.41647744312583, "grad_norm": 4.847392559051514, "learning_rate": 3.3231093709377492e-06, "loss": 0.36736140251159666, "memory(GiB)": 77.59, "step": 103085, "token_acc": 0.916083916083916, "train_speed(iter/s)": 1.461717 }, { "epoch": 4.416691658455079, "grad_norm": 4.227273464202881, "learning_rate": 3.320697311815385e-06, "loss": 0.7284860610961914, "memory(GiB)": 77.59, "step": 103090, "token_acc": 0.8487972508591065, "train_speed(iter/s)": 1.461722 }, { "epoch": 4.416905873784328, "grad_norm": 5.11418342590332, "learning_rate": 3.3182860983323683e-06, "loss": 0.4237488269805908, "memory(GiB)": 77.59, "step": 103095, "token_acc": 0.8903654485049833, "train_speed(iter/s)": 1.461724 }, { "epoch": 4.417120089113577, "grad_norm": 1.8223402500152588, "learning_rate": 3.3158757305324027e-06, "loss": 0.2600691556930542, "memory(GiB)": 77.59, "step": 103100, "token_acc": 0.9169435215946844, "train_speed(iter/s)": 1.461728 }, { "epoch": 4.417334304442826, "grad_norm": 2.9524195194244385, "learning_rate": 3.313466208459137e-06, "loss": 0.6160240173339844, "memory(GiB)": 77.59, "step": 103105, "token_acc": 0.8440860215053764, "train_speed(iter/s)": 1.46174 }, { "epoch": 4.417548519772075, "grad_norm": 6.942122459411621, "learning_rate": 3.3110575321562255e-06, "loss": 0.34253511428833006, "memory(GiB)": 77.59, "step": 103110, "token_acc": 0.9150326797385621, "train_speed(iter/s)": 1.461753 }, { "epoch": 4.417762735101324, "grad_norm": 3.44897198677063, "learning_rate": 3.30864970166731e-06, "loss": 0.35530576705932615, "memory(GiB)": 77.59, "step": 103115, "token_acc": 0.9359756097560976, "train_speed(iter/s)": 1.461761 }, { "epoch": 4.417976950430573, "grad_norm": 2.5697202682495117, "learning_rate": 3.306242717036001e-06, "loss": 0.15310176610946655, "memory(GiB)": 77.59, "step": 103120, "token_acc": 0.9604743083003953, "train_speed(iter/s)": 1.461767 }, { "epoch": 4.418191165759822, "grad_norm": 3.460143566131592, "learning_rate": 3.303836578305902e-06, "loss": 0.3412087202072144, "memory(GiB)": 77.59, "step": 103125, "token_acc": 0.9329446064139941, "train_speed(iter/s)": 1.461767 }, { "epoch": 4.41840538108907, "grad_norm": 4.589786529541016, "learning_rate": 3.301431285520612e-06, "loss": 0.3131126880645752, "memory(GiB)": 77.59, "step": 103130, "token_acc": 0.9247311827956989, "train_speed(iter/s)": 1.461778 }, { "epoch": 4.41861959641832, "grad_norm": 3.967670202255249, "learning_rate": 3.2990268387237065e-06, "loss": 0.3825406074523926, "memory(GiB)": 77.59, "step": 103135, "token_acc": 0.9337539432176656, "train_speed(iter/s)": 1.461779 }, { "epoch": 4.418833811747569, "grad_norm": 2.20798659324646, "learning_rate": 3.296623237958735e-06, "loss": 0.3335462808609009, "memory(GiB)": 77.59, "step": 103140, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.461778 }, { "epoch": 4.419048027076817, "grad_norm": 2.5412328243255615, "learning_rate": 3.2942204832692337e-06, "loss": 0.47416396141052247, "memory(GiB)": 77.59, "step": 103145, "token_acc": 0.9153846153846154, "train_speed(iter/s)": 1.461791 }, { "epoch": 4.419262242406067, "grad_norm": 1.7965368032455444, "learning_rate": 3.2918185746987575e-06, "loss": 0.12065953016281128, "memory(GiB)": 77.59, "step": 103150, "token_acc": 0.9713261648745519, "train_speed(iter/s)": 1.461796 }, { "epoch": 4.419476457735316, "grad_norm": 3.9120986461639404, "learning_rate": 3.2894175122907934e-06, "loss": 0.2731020927429199, "memory(GiB)": 77.59, "step": 103155, "token_acc": 0.9409937888198758, "train_speed(iter/s)": 1.461809 }, { "epoch": 4.419690673064564, "grad_norm": 3.1678531169891357, "learning_rate": 3.2870172960888513e-06, "loss": 0.3742526054382324, "memory(GiB)": 77.59, "step": 103160, "token_acc": 0.9075907590759076, "train_speed(iter/s)": 1.461812 }, { "epoch": 4.4199048883938135, "grad_norm": 3.4194207191467285, "learning_rate": 3.2846179261364183e-06, "loss": 0.43302407264709475, "memory(GiB)": 77.59, "step": 103165, "token_acc": 0.8939393939393939, "train_speed(iter/s)": 1.461813 }, { "epoch": 4.420119103723063, "grad_norm": 4.8843278884887695, "learning_rate": 3.2822194024769494e-06, "loss": 0.33716323375701907, "memory(GiB)": 77.59, "step": 103170, "token_acc": 0.9401197604790419, "train_speed(iter/s)": 1.461814 }, { "epoch": 4.420333319052311, "grad_norm": 3.5005407333374023, "learning_rate": 3.279821725153903e-06, "loss": 0.4775230884552002, "memory(GiB)": 77.59, "step": 103175, "token_acc": 0.8711864406779661, "train_speed(iter/s)": 1.461824 }, { "epoch": 4.42054753438156, "grad_norm": 4.522106647491455, "learning_rate": 3.277424894210712e-06, "loss": 0.4631487846374512, "memory(GiB)": 77.59, "step": 103180, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.46183 }, { "epoch": 4.42076174971081, "grad_norm": 2.5332560539245605, "learning_rate": 3.2750289096907915e-06, "loss": 0.4596975803375244, "memory(GiB)": 77.59, "step": 103185, "token_acc": 0.9188191881918819, "train_speed(iter/s)": 1.46183 }, { "epoch": 4.420975965040058, "grad_norm": 2.7328364849090576, "learning_rate": 3.272633771637562e-06, "loss": 0.4595158576965332, "memory(GiB)": 77.59, "step": 103190, "token_acc": 0.8797653958944281, "train_speed(iter/s)": 1.461826 }, { "epoch": 4.421190180369307, "grad_norm": 4.422942638397217, "learning_rate": 3.2702394800944057e-06, "loss": 0.2671933174133301, "memory(GiB)": 77.59, "step": 103195, "token_acc": 0.9358490566037736, "train_speed(iter/s)": 1.461828 }, { "epoch": 4.4214043956985565, "grad_norm": 2.3593122959136963, "learning_rate": 3.2678460351046993e-06, "loss": 0.3127322196960449, "memory(GiB)": 77.59, "step": 103200, "token_acc": 0.953125, "train_speed(iter/s)": 1.461829 }, { "epoch": 4.421618611027805, "grad_norm": 0.560939610004425, "learning_rate": 3.265453436711796e-06, "loss": 0.10716387033462524, "memory(GiB)": 77.59, "step": 103205, "token_acc": 0.972972972972973, "train_speed(iter/s)": 1.461834 }, { "epoch": 4.421832826357054, "grad_norm": 1.4624922275543213, "learning_rate": 3.26306168495904e-06, "loss": 0.170185124874115, "memory(GiB)": 77.59, "step": 103210, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.461841 }, { "epoch": 4.422047041686303, "grad_norm": 3.310572862625122, "learning_rate": 3.2606707798897683e-06, "loss": 0.40680460929870604, "memory(GiB)": 77.59, "step": 103215, "token_acc": 0.9041095890410958, "train_speed(iter/s)": 1.461849 }, { "epoch": 4.422261257015552, "grad_norm": 2.100799083709717, "learning_rate": 3.2582807215472854e-06, "loss": 0.5426620006561279, "memory(GiB)": 77.59, "step": 103220, "token_acc": 0.9084507042253521, "train_speed(iter/s)": 1.461859 }, { "epoch": 4.422475472344801, "grad_norm": 4.230330944061279, "learning_rate": 3.255891509974901e-06, "loss": 0.3129417896270752, "memory(GiB)": 77.59, "step": 103225, "token_acc": 0.9274924471299094, "train_speed(iter/s)": 1.461855 }, { "epoch": 4.42268968767405, "grad_norm": 2.23575758934021, "learning_rate": 3.2535031452158915e-06, "loss": 0.3187024354934692, "memory(GiB)": 77.59, "step": 103230, "token_acc": 0.9285714285714286, "train_speed(iter/s)": 1.46186 }, { "epoch": 4.422903903003299, "grad_norm": 0.39342567324638367, "learning_rate": 3.251115627313528e-06, "loss": 0.39848291873931885, "memory(GiB)": 77.59, "step": 103235, "token_acc": 0.9065420560747663, "train_speed(iter/s)": 1.461862 }, { "epoch": 4.423118118332548, "grad_norm": 8.054346084594727, "learning_rate": 3.2487289563110478e-06, "loss": 0.24693546295166016, "memory(GiB)": 77.59, "step": 103240, "token_acc": 0.9493243243243243, "train_speed(iter/s)": 1.461866 }, { "epoch": 4.423332333661797, "grad_norm": 0.9236149787902832, "learning_rate": 3.246343132251706e-06, "loss": 0.37824788093566897, "memory(GiB)": 77.59, "step": 103245, "token_acc": 0.8960573476702509, "train_speed(iter/s)": 1.46187 }, { "epoch": 4.4235465489910455, "grad_norm": 1.846612572669983, "learning_rate": 3.243958155178717e-06, "loss": 0.3475534439086914, "memory(GiB)": 77.59, "step": 103250, "token_acc": 0.9261538461538461, "train_speed(iter/s)": 1.461869 }, { "epoch": 4.423760764320295, "grad_norm": 3.1997241973876953, "learning_rate": 3.241574025135291e-06, "loss": 0.4025541305541992, "memory(GiB)": 77.59, "step": 103255, "token_acc": 0.9033333333333333, "train_speed(iter/s)": 1.46188 }, { "epoch": 4.423974979649544, "grad_norm": 2.1515402793884277, "learning_rate": 3.2391907421646106e-06, "loss": 0.20888133049011232, "memory(GiB)": 77.59, "step": 103260, "token_acc": 0.9487179487179487, "train_speed(iter/s)": 1.461886 }, { "epoch": 4.424189194978792, "grad_norm": 2.6274847984313965, "learning_rate": 3.2368083063098518e-06, "loss": 0.28545031547546384, "memory(GiB)": 77.59, "step": 103265, "token_acc": 0.9389067524115756, "train_speed(iter/s)": 1.461889 }, { "epoch": 4.424403410308042, "grad_norm": 1.5669865608215332, "learning_rate": 3.2344267176141753e-06, "loss": 0.2645861148834229, "memory(GiB)": 77.59, "step": 103270, "token_acc": 0.9563636363636364, "train_speed(iter/s)": 1.4619 }, { "epoch": 4.424617625637291, "grad_norm": 1.931093692779541, "learning_rate": 3.2320459761207344e-06, "loss": 0.2734061717987061, "memory(GiB)": 77.59, "step": 103275, "token_acc": 0.9448275862068966, "train_speed(iter/s)": 1.461901 }, { "epoch": 4.424831840966539, "grad_norm": 0.41422218084335327, "learning_rate": 3.229666081872651e-06, "loss": 0.09996446967124939, "memory(GiB)": 77.59, "step": 103280, "token_acc": 0.9714285714285714, "train_speed(iter/s)": 1.461904 }, { "epoch": 4.4250460562957885, "grad_norm": 5.92359733581543, "learning_rate": 3.227287034913046e-06, "loss": 0.34167981147766113, "memory(GiB)": 77.59, "step": 103285, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.461904 }, { "epoch": 4.425260271625038, "grad_norm": 3.989736795425415, "learning_rate": 3.2249088352850065e-06, "loss": 0.514876127243042, "memory(GiB)": 77.59, "step": 103290, "token_acc": 0.8907849829351536, "train_speed(iter/s)": 1.461905 }, { "epoch": 4.425474486954286, "grad_norm": 4.301228046417236, "learning_rate": 3.2225314830316155e-06, "loss": 0.3475778579711914, "memory(GiB)": 77.59, "step": 103295, "token_acc": 0.927170868347339, "train_speed(iter/s)": 1.461908 }, { "epoch": 4.425688702283535, "grad_norm": 1.6362693309783936, "learning_rate": 3.2201549781959493e-06, "loss": 0.1930663466453552, "memory(GiB)": 77.59, "step": 103300, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.461917 }, { "epoch": 4.425902917612785, "grad_norm": 1.7874459028244019, "learning_rate": 3.2177793208210683e-06, "loss": 0.30401756763458254, "memory(GiB)": 77.59, "step": 103305, "token_acc": 0.9276729559748428, "train_speed(iter/s)": 1.461921 }, { "epoch": 4.426117132942033, "grad_norm": 1.0285001993179321, "learning_rate": 3.2154045109499985e-06, "loss": 0.43850555419921877, "memory(GiB)": 77.59, "step": 103310, "token_acc": 0.9039735099337748, "train_speed(iter/s)": 1.461932 }, { "epoch": 4.426331348271282, "grad_norm": 3.086801052093506, "learning_rate": 3.2130305486257615e-06, "loss": 0.3705465316772461, "memory(GiB)": 77.59, "step": 103315, "token_acc": 0.898876404494382, "train_speed(iter/s)": 1.461935 }, { "epoch": 4.4265455636005315, "grad_norm": 4.629354476928711, "learning_rate": 3.2106574338913676e-06, "loss": 0.19763879776000975, "memory(GiB)": 77.59, "step": 103320, "token_acc": 0.9508196721311475, "train_speed(iter/s)": 1.461942 }, { "epoch": 4.42675977892978, "grad_norm": 5.570720195770264, "learning_rate": 3.2082851667897985e-06, "loss": 0.4327519416809082, "memory(GiB)": 77.59, "step": 103325, "token_acc": 0.8735632183908046, "train_speed(iter/s)": 1.461937 }, { "epoch": 4.426973994259029, "grad_norm": 2.396592617034912, "learning_rate": 3.2059137473640476e-06, "loss": 0.3152249097824097, "memory(GiB)": 77.59, "step": 103330, "token_acc": 0.9130434782608695, "train_speed(iter/s)": 1.461936 }, { "epoch": 4.427188209588278, "grad_norm": 3.1082398891448975, "learning_rate": 3.203543175657059e-06, "loss": 0.2515141010284424, "memory(GiB)": 77.59, "step": 103335, "token_acc": 0.9550173010380623, "train_speed(iter/s)": 1.461936 }, { "epoch": 4.427402424917527, "grad_norm": 3.4484128952026367, "learning_rate": 3.2011734517117917e-06, "loss": 0.3230747699737549, "memory(GiB)": 77.59, "step": 103340, "token_acc": 0.929368029739777, "train_speed(iter/s)": 1.461939 }, { "epoch": 4.427616640246776, "grad_norm": 2.2643048763275146, "learning_rate": 3.1988045755711627e-06, "loss": 0.2492429256439209, "memory(GiB)": 77.59, "step": 103345, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.461939 }, { "epoch": 4.427830855576025, "grad_norm": 2.867974042892456, "learning_rate": 3.1964365472780976e-06, "loss": 0.3325268983840942, "memory(GiB)": 77.59, "step": 103350, "token_acc": 0.912621359223301, "train_speed(iter/s)": 1.461943 }, { "epoch": 4.428045070905274, "grad_norm": 1.6494451761245728, "learning_rate": 3.194069366875474e-06, "loss": 0.2580333471298218, "memory(GiB)": 77.59, "step": 103355, "token_acc": 0.9526813880126183, "train_speed(iter/s)": 1.461944 }, { "epoch": 4.428259286234523, "grad_norm": 4.45923376083374, "learning_rate": 3.191703034406207e-06, "loss": 0.26104063987731935, "memory(GiB)": 77.59, "step": 103360, "token_acc": 0.9314079422382672, "train_speed(iter/s)": 1.461947 }, { "epoch": 4.428473501563772, "grad_norm": 3.1531105041503906, "learning_rate": 3.1893375499131352e-06, "loss": 0.29505965709686277, "memory(GiB)": 77.59, "step": 103365, "token_acc": 0.930379746835443, "train_speed(iter/s)": 1.461951 }, { "epoch": 4.4286877168930205, "grad_norm": 3.3110220432281494, "learning_rate": 3.18697291343914e-06, "loss": 0.331296968460083, "memory(GiB)": 77.59, "step": 103370, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.461957 }, { "epoch": 4.42890193222227, "grad_norm": 3.9018750190734863, "learning_rate": 3.184609125027038e-06, "loss": 0.3432819366455078, "memory(GiB)": 77.59, "step": 103375, "token_acc": 0.9458483754512635, "train_speed(iter/s)": 1.461959 }, { "epoch": 4.429116147551519, "grad_norm": 2.5722451210021973, "learning_rate": 3.1822461847196617e-06, "loss": 0.2446269989013672, "memory(GiB)": 77.59, "step": 103380, "token_acc": 0.9492753623188406, "train_speed(iter/s)": 1.461958 }, { "epoch": 4.429330362880767, "grad_norm": 2.687818765640259, "learning_rate": 3.1798840925598093e-06, "loss": 0.2037205457687378, "memory(GiB)": 77.59, "step": 103385, "token_acc": 0.9484536082474226, "train_speed(iter/s)": 1.461964 }, { "epoch": 4.429544578210017, "grad_norm": 1.8327643871307373, "learning_rate": 3.1775228485902807e-06, "loss": 0.30791563987731935, "memory(GiB)": 77.59, "step": 103390, "token_acc": 0.9365671641791045, "train_speed(iter/s)": 1.461974 }, { "epoch": 4.429758793539266, "grad_norm": 3.4761388301849365, "learning_rate": 3.175162452853847e-06, "loss": 0.4797371864318848, "memory(GiB)": 77.59, "step": 103395, "token_acc": 0.9011406844106464, "train_speed(iter/s)": 1.461976 }, { "epoch": 4.429973008868514, "grad_norm": 2.5207390785217285, "learning_rate": 3.1728029053932738e-06, "loss": 0.400342321395874, "memory(GiB)": 77.59, "step": 103400, "token_acc": 0.9235880398671097, "train_speed(iter/s)": 1.46198 }, { "epoch": 4.430187224197764, "grad_norm": 3.627235174179077, "learning_rate": 3.170444206251305e-06, "loss": 0.24672114849090576, "memory(GiB)": 77.59, "step": 103405, "token_acc": 0.9465648854961832, "train_speed(iter/s)": 1.461984 }, { "epoch": 4.430401439527013, "grad_norm": 2.871299982070923, "learning_rate": 3.1680863554706674e-06, "loss": 0.6503304958343505, "memory(GiB)": 77.59, "step": 103410, "token_acc": 0.8786127167630058, "train_speed(iter/s)": 1.461993 }, { "epoch": 4.430615654856261, "grad_norm": 3.312309741973877, "learning_rate": 3.165729353094071e-06, "loss": 0.2302065372467041, "memory(GiB)": 77.59, "step": 103415, "token_acc": 0.9535603715170279, "train_speed(iter/s)": 1.461996 }, { "epoch": 4.43082987018551, "grad_norm": 4.472128391265869, "learning_rate": 3.163373199164227e-06, "loss": 0.32524111270904543, "memory(GiB)": 77.59, "step": 103420, "token_acc": 0.9368770764119602, "train_speed(iter/s)": 1.461999 }, { "epoch": 4.43104408551476, "grad_norm": 3.4238386154174805, "learning_rate": 3.1610178937238167e-06, "loss": 0.38225612640380857, "memory(GiB)": 77.59, "step": 103425, "token_acc": 0.8929765886287625, "train_speed(iter/s)": 1.462001 }, { "epoch": 4.431258300844008, "grad_norm": 3.4204864501953125, "learning_rate": 3.158663436815501e-06, "loss": 0.4560101509094238, "memory(GiB)": 77.59, "step": 103430, "token_acc": 0.8869047619047619, "train_speed(iter/s)": 1.462011 }, { "epoch": 4.431472516173257, "grad_norm": 2.5659210681915283, "learning_rate": 3.1563098284819404e-06, "loss": 0.3113293409347534, "memory(GiB)": 77.59, "step": 103435, "token_acc": 0.9350180505415162, "train_speed(iter/s)": 1.462013 }, { "epoch": 4.431686731502507, "grad_norm": 1.5642167329788208, "learning_rate": 3.1539570687657617e-06, "loss": 0.32257034778594973, "memory(GiB)": 77.59, "step": 103440, "token_acc": 0.9416666666666667, "train_speed(iter/s)": 1.462015 }, { "epoch": 4.431900946831755, "grad_norm": 4.900313854217529, "learning_rate": 3.151605157709597e-06, "loss": 0.594711446762085, "memory(GiB)": 77.59, "step": 103445, "token_acc": 0.8711656441717791, "train_speed(iter/s)": 1.462026 }, { "epoch": 4.432115162161004, "grad_norm": 0.4648725688457489, "learning_rate": 3.1492540953560635e-06, "loss": 0.21371254920959473, "memory(GiB)": 77.59, "step": 103450, "token_acc": 0.9467084639498433, "train_speed(iter/s)": 1.462033 }, { "epoch": 4.4323293774902535, "grad_norm": 6.807082176208496, "learning_rate": 3.146903881747737e-06, "loss": 0.3337381839752197, "memory(GiB)": 77.59, "step": 103455, "token_acc": 0.9302325581395349, "train_speed(iter/s)": 1.462037 }, { "epoch": 4.432543592819502, "grad_norm": 7.3048415184021, "learning_rate": 3.1445545169272005e-06, "loss": 0.22827112674713135, "memory(GiB)": 77.59, "step": 103460, "token_acc": 0.9457364341085271, "train_speed(iter/s)": 1.462038 }, { "epoch": 4.432757808148751, "grad_norm": 3.433241367340088, "learning_rate": 3.142206000937009e-06, "loss": 0.15065722465515136, "memory(GiB)": 77.59, "step": 103465, "token_acc": 0.9627118644067797, "train_speed(iter/s)": 1.46204 }, { "epoch": 4.432972023478, "grad_norm": 3.0735764503479004, "learning_rate": 3.139858333819706e-06, "loss": 0.4933948516845703, "memory(GiB)": 77.59, "step": 103470, "token_acc": 0.8819875776397516, "train_speed(iter/s)": 1.462045 }, { "epoch": 4.433186238807249, "grad_norm": 2.8886311054229736, "learning_rate": 3.137511515617836e-06, "loss": 0.3463951349258423, "memory(GiB)": 77.59, "step": 103475, "token_acc": 0.9375, "train_speed(iter/s)": 1.462045 }, { "epoch": 4.433400454136498, "grad_norm": 1.883233666419983, "learning_rate": 3.135165546373908e-06, "loss": 0.3482660293579102, "memory(GiB)": 77.59, "step": 103480, "token_acc": 0.9373040752351097, "train_speed(iter/s)": 1.462056 }, { "epoch": 4.433614669465747, "grad_norm": 3.411223888397217, "learning_rate": 3.1328204261304173e-06, "loss": 0.40571889877319334, "memory(GiB)": 77.59, "step": 103485, "token_acc": 0.9066666666666666, "train_speed(iter/s)": 1.462058 }, { "epoch": 4.433828884794996, "grad_norm": 2.8614702224731445, "learning_rate": 3.1304761549298457e-06, "loss": 0.33812694549560546, "memory(GiB)": 77.59, "step": 103490, "token_acc": 0.9246031746031746, "train_speed(iter/s)": 1.462056 }, { "epoch": 4.434043100124245, "grad_norm": 4.40010404586792, "learning_rate": 3.1281327328146703e-06, "loss": 0.6715186595916748, "memory(GiB)": 77.59, "step": 103495, "token_acc": 0.8440677966101695, "train_speed(iter/s)": 1.462055 }, { "epoch": 4.434257315453494, "grad_norm": 3.3244526386260986, "learning_rate": 3.12579015982733e-06, "loss": 0.48398194313049314, "memory(GiB)": 77.59, "step": 103500, "token_acc": 0.8885350318471338, "train_speed(iter/s)": 1.462071 }, { "epoch": 4.434257315453494, "eval_loss": 2.3080263137817383, "eval_runtime": 11.8878, "eval_samples_per_second": 8.412, "eval_steps_per_second": 8.412, "eval_token_acc": 0.44878706199460916, "step": 103500 }, { "epoch": 4.4344715307827425, "grad_norm": 3.521118402481079, "learning_rate": 3.1234484360102846e-06, "loss": 0.42543864250183105, "memory(GiB)": 77.59, "step": 103505, "token_acc": 0.578383641674781, "train_speed(iter/s)": 1.461815 }, { "epoch": 4.434685746111992, "grad_norm": 6.264172554016113, "learning_rate": 3.121107561405939e-06, "loss": 0.48735966682434084, "memory(GiB)": 77.59, "step": 103510, "token_acc": 0.9004329004329005, "train_speed(iter/s)": 1.461818 }, { "epoch": 4.434899961441241, "grad_norm": 0.6039650440216064, "learning_rate": 3.118767536056699e-06, "loss": 0.30980448722839354, "memory(GiB)": 77.59, "step": 103515, "token_acc": 0.9290540540540541, "train_speed(iter/s)": 1.461821 }, { "epoch": 4.435114176770489, "grad_norm": 3.632610321044922, "learning_rate": 3.116428360004975e-06, "loss": 0.4036895751953125, "memory(GiB)": 77.59, "step": 103520, "token_acc": 0.9172413793103448, "train_speed(iter/s)": 1.461824 }, { "epoch": 4.435328392099739, "grad_norm": 4.846930980682373, "learning_rate": 3.1140900332931265e-06, "loss": 0.4085724353790283, "memory(GiB)": 77.59, "step": 103525, "token_acc": 0.9208211143695014, "train_speed(iter/s)": 1.461828 }, { "epoch": 4.435542607428988, "grad_norm": 2.9868342876434326, "learning_rate": 3.111752555963515e-06, "loss": 0.46550421714782714, "memory(GiB)": 77.59, "step": 103530, "token_acc": 0.8976109215017065, "train_speed(iter/s)": 1.461826 }, { "epoch": 4.435756822758236, "grad_norm": 3.587447166442871, "learning_rate": 3.1094159280584957e-06, "loss": 0.4632739067077637, "memory(GiB)": 77.59, "step": 103535, "token_acc": 0.8922155688622755, "train_speed(iter/s)": 1.461825 }, { "epoch": 4.4359710380874855, "grad_norm": 6.615805149078369, "learning_rate": 3.107080149620395e-06, "loss": 0.43155202865600584, "memory(GiB)": 77.59, "step": 103540, "token_acc": 0.8790035587188612, "train_speed(iter/s)": 1.461823 }, { "epoch": 4.436185253416735, "grad_norm": 3.605518341064453, "learning_rate": 3.1047452206915294e-06, "loss": 0.3560297727584839, "memory(GiB)": 77.59, "step": 103545, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.461823 }, { "epoch": 4.436399468745983, "grad_norm": 1.7487066984176636, "learning_rate": 3.102411141314193e-06, "loss": 0.2856872081756592, "memory(GiB)": 77.59, "step": 103550, "token_acc": 0.933933933933934, "train_speed(iter/s)": 1.46183 }, { "epoch": 4.436613684075232, "grad_norm": 1.0966392755508423, "learning_rate": 3.100077911530669e-06, "loss": 0.29052441120147704, "memory(GiB)": 77.59, "step": 103555, "token_acc": 0.9119496855345912, "train_speed(iter/s)": 1.461831 }, { "epoch": 4.436827899404482, "grad_norm": 3.9684741497039795, "learning_rate": 3.0977455313832283e-06, "loss": 0.11269888877868653, "memory(GiB)": 77.59, "step": 103560, "token_acc": 0.9689655172413794, "train_speed(iter/s)": 1.46183 }, { "epoch": 4.43704211473373, "grad_norm": 1.2867422103881836, "learning_rate": 3.0954140009141264e-06, "loss": 0.34957849979400635, "memory(GiB)": 77.59, "step": 103565, "token_acc": 0.9262820512820513, "train_speed(iter/s)": 1.461827 }, { "epoch": 4.437256330062979, "grad_norm": 2.5052058696746826, "learning_rate": 3.093083320165602e-06, "loss": 0.321109676361084, "memory(GiB)": 77.59, "step": 103570, "token_acc": 0.9315589353612167, "train_speed(iter/s)": 1.461835 }, { "epoch": 4.4374705453922285, "grad_norm": 3.2795236110687256, "learning_rate": 3.090753489179876e-06, "loss": 0.46605892181396485, "memory(GiB)": 77.59, "step": 103575, "token_acc": 0.8909657320872274, "train_speed(iter/s)": 1.461835 }, { "epoch": 4.437684760721477, "grad_norm": 1.5864722728729248, "learning_rate": 3.08842450799916e-06, "loss": 0.14984904527664183, "memory(GiB)": 77.59, "step": 103580, "token_acc": 0.9705014749262537, "train_speed(iter/s)": 1.461838 }, { "epoch": 4.437898976050726, "grad_norm": 3.9298689365386963, "learning_rate": 3.086096376665626e-06, "loss": 0.41051645278930665, "memory(GiB)": 77.59, "step": 103585, "token_acc": 0.9113149847094801, "train_speed(iter/s)": 1.461839 }, { "epoch": 4.438113191379975, "grad_norm": 3.438173294067383, "learning_rate": 3.083769095221467e-06, "loss": 0.36142854690551757, "memory(GiB)": 77.59, "step": 103590, "token_acc": 0.9294871794871795, "train_speed(iter/s)": 1.461839 }, { "epoch": 4.438327406709224, "grad_norm": 2.9612483978271484, "learning_rate": 3.0814426637088502e-06, "loss": 0.4547117710113525, "memory(GiB)": 77.59, "step": 103595, "token_acc": 0.9024390243902439, "train_speed(iter/s)": 1.461836 }, { "epoch": 4.438541622038473, "grad_norm": 4.868314266204834, "learning_rate": 3.0791170821699133e-06, "loss": 0.4561042308807373, "memory(GiB)": 77.59, "step": 103600, "token_acc": 0.9047619047619048, "train_speed(iter/s)": 1.461839 }, { "epoch": 4.438755837367722, "grad_norm": 1.7544817924499512, "learning_rate": 3.0767923506467845e-06, "loss": 0.24493327140808105, "memory(GiB)": 77.59, "step": 103605, "token_acc": 0.9527027027027027, "train_speed(iter/s)": 1.461837 }, { "epoch": 4.438970052696971, "grad_norm": 3.7345998287200928, "learning_rate": 3.0744684691815795e-06, "loss": 0.2505201816558838, "memory(GiB)": 77.59, "step": 103610, "token_acc": 0.9421221864951769, "train_speed(iter/s)": 1.461841 }, { "epoch": 4.43918426802622, "grad_norm": 0.9798698425292969, "learning_rate": 3.0721454378163925e-06, "loss": 0.2516953945159912, "memory(GiB)": 77.59, "step": 103615, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.461846 }, { "epoch": 4.439398483355469, "grad_norm": 2.796919822692871, "learning_rate": 3.069823256593318e-06, "loss": 0.42614455223083497, "memory(GiB)": 77.59, "step": 103620, "token_acc": 0.898989898989899, "train_speed(iter/s)": 1.461851 }, { "epoch": 4.4396126986847175, "grad_norm": 2.565415859222412, "learning_rate": 3.0675019255544224e-06, "loss": 0.3041412353515625, "memory(GiB)": 77.59, "step": 103625, "token_acc": 0.9454545454545454, "train_speed(iter/s)": 1.46185 }, { "epoch": 4.439826914013967, "grad_norm": 2.9791953563690186, "learning_rate": 3.0651814447417492e-06, "loss": 0.46234827041625975, "memory(GiB)": 77.59, "step": 103630, "token_acc": 0.9257142857142857, "train_speed(iter/s)": 1.461858 }, { "epoch": 4.440041129343216, "grad_norm": 0.986024796962738, "learning_rate": 3.0628618141973486e-06, "loss": 0.3662412643432617, "memory(GiB)": 77.59, "step": 103635, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.461858 }, { "epoch": 4.440255344672464, "grad_norm": 3.3220276832580566, "learning_rate": 3.060543033963237e-06, "loss": 0.4137455940246582, "memory(GiB)": 77.59, "step": 103640, "token_acc": 0.9288256227758007, "train_speed(iter/s)": 1.461861 }, { "epoch": 4.440469560001714, "grad_norm": 3.5659449100494385, "learning_rate": 3.058225104081408e-06, "loss": 0.2972689390182495, "memory(GiB)": 77.59, "step": 103645, "token_acc": 0.9344827586206896, "train_speed(iter/s)": 1.461863 }, { "epoch": 4.440683775330963, "grad_norm": 0.8507516384124756, "learning_rate": 3.0559080245938787e-06, "loss": 0.26842517852783204, "memory(GiB)": 77.59, "step": 103650, "token_acc": 0.9297124600638977, "train_speed(iter/s)": 1.461862 }, { "epoch": 4.440897990660211, "grad_norm": 2.877809524536133, "learning_rate": 3.0535917955426096e-06, "loss": 0.1375652551651001, "memory(GiB)": 77.59, "step": 103655, "token_acc": 0.967479674796748, "train_speed(iter/s)": 1.461863 }, { "epoch": 4.4411122059894605, "grad_norm": 3.245547294616699, "learning_rate": 3.0512764169695617e-06, "loss": 0.20500223636627196, "memory(GiB)": 77.59, "step": 103660, "token_acc": 0.9440993788819876, "train_speed(iter/s)": 1.461862 }, { "epoch": 4.44132642131871, "grad_norm": 0.47662588953971863, "learning_rate": 3.048961888916685e-06, "loss": 0.34370925426483157, "memory(GiB)": 77.59, "step": 103665, "token_acc": 0.9352226720647774, "train_speed(iter/s)": 1.461863 }, { "epoch": 4.441540636647958, "grad_norm": 5.722512722015381, "learning_rate": 3.046648211425912e-06, "loss": 0.46373300552368163, "memory(GiB)": 77.59, "step": 103670, "token_acc": 0.88, "train_speed(iter/s)": 1.461864 }, { "epoch": 4.441754851977207, "grad_norm": 2.766950845718384, "learning_rate": 3.044335384539143e-06, "loss": 0.47388515472412107, "memory(GiB)": 77.59, "step": 103675, "token_acc": 0.9180887372013652, "train_speed(iter/s)": 1.461869 }, { "epoch": 4.441969067306457, "grad_norm": 6.95186185836792, "learning_rate": 3.042023408298289e-06, "loss": 0.19003700017929076, "memory(GiB)": 77.59, "step": 103680, "token_acc": 0.9493243243243243, "train_speed(iter/s)": 1.461873 }, { "epoch": 4.442183282635705, "grad_norm": 5.667239665985107, "learning_rate": 3.0397122827452384e-06, "loss": 0.3535437107086182, "memory(GiB)": 77.59, "step": 103685, "token_acc": 0.9306930693069307, "train_speed(iter/s)": 1.461875 }, { "epoch": 4.442397497964954, "grad_norm": 3.147437810897827, "learning_rate": 3.037402007921847e-06, "loss": 0.5448683738708496, "memory(GiB)": 77.59, "step": 103690, "token_acc": 0.8985074626865671, "train_speed(iter/s)": 1.461896 }, { "epoch": 4.442611713294204, "grad_norm": 3.7887625694274902, "learning_rate": 3.035092583869975e-06, "loss": 0.415986156463623, "memory(GiB)": 77.59, "step": 103695, "token_acc": 0.9163498098859315, "train_speed(iter/s)": 1.461911 }, { "epoch": 4.442825928623452, "grad_norm": 3.685539484024048, "learning_rate": 3.032784010631462e-06, "loss": 0.6331137657165528, "memory(GiB)": 77.59, "step": 103700, "token_acc": 0.878125, "train_speed(iter/s)": 1.461914 }, { "epoch": 4.443040143952701, "grad_norm": 3.69897723197937, "learning_rate": 3.030476288248113e-06, "loss": 0.2859755516052246, "memory(GiB)": 77.59, "step": 103705, "token_acc": 0.9240506329113924, "train_speed(iter/s)": 1.461916 }, { "epoch": 4.4432543592819504, "grad_norm": 3.979091167449951, "learning_rate": 3.028169416761756e-06, "loss": 0.3622141361236572, "memory(GiB)": 77.59, "step": 103710, "token_acc": 0.9027777777777778, "train_speed(iter/s)": 1.461916 }, { "epoch": 4.443468574611199, "grad_norm": 1.8736516237258911, "learning_rate": 3.025863396214168e-06, "loss": 0.24821386337280274, "memory(GiB)": 77.59, "step": 103715, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.461923 }, { "epoch": 4.443682789940448, "grad_norm": 0.8934811353683472, "learning_rate": 3.0235582266471328e-06, "loss": 0.13963167667388915, "memory(GiB)": 77.59, "step": 103720, "token_acc": 0.9659090909090909, "train_speed(iter/s)": 1.461929 }, { "epoch": 4.443897005269697, "grad_norm": 1.8157098293304443, "learning_rate": 3.0212539081024117e-06, "loss": 0.2852010250091553, "memory(GiB)": 77.59, "step": 103725, "token_acc": 0.9380530973451328, "train_speed(iter/s)": 1.461928 }, { "epoch": 4.444111220598946, "grad_norm": 4.783232688903809, "learning_rate": 3.0189504406217375e-06, "loss": 0.4079730033874512, "memory(GiB)": 77.59, "step": 103730, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.461935 }, { "epoch": 4.444325435928195, "grad_norm": 3.219186782836914, "learning_rate": 3.016647824246843e-06, "loss": 0.5422533512115478, "memory(GiB)": 77.59, "step": 103735, "token_acc": 0.8691588785046729, "train_speed(iter/s)": 1.461931 }, { "epoch": 4.444539651257444, "grad_norm": 3.1164963245391846, "learning_rate": 3.014346059019457e-06, "loss": 0.44658384323120115, "memory(GiB)": 77.59, "step": 103740, "token_acc": 0.9037800687285223, "train_speed(iter/s)": 1.461932 }, { "epoch": 4.444753866586693, "grad_norm": 0.46805432438850403, "learning_rate": 3.012045144981268e-06, "loss": 0.12640044689178467, "memory(GiB)": 77.59, "step": 103745, "token_acc": 0.9711191335740073, "train_speed(iter/s)": 1.461937 }, { "epoch": 4.444968081915942, "grad_norm": 0.31671109795570374, "learning_rate": 3.009745082173965e-06, "loss": 0.38141779899597167, "memory(GiB)": 77.59, "step": 103750, "token_acc": 0.9386503067484663, "train_speed(iter/s)": 1.461932 }, { "epoch": 4.445182297245191, "grad_norm": 5.956330299377441, "learning_rate": 3.0074458706392028e-06, "loss": 0.2580071687698364, "memory(GiB)": 77.59, "step": 103755, "token_acc": 0.9311475409836065, "train_speed(iter/s)": 1.461938 }, { "epoch": 4.4453965125744395, "grad_norm": 3.287702798843384, "learning_rate": 3.0051475104186376e-06, "loss": 0.3291463375091553, "memory(GiB)": 77.59, "step": 103760, "token_acc": 0.9392523364485982, "train_speed(iter/s)": 1.461951 }, { "epoch": 4.445610727903689, "grad_norm": 2.6603033542633057, "learning_rate": 3.002850001553914e-06, "loss": 0.5501738548278808, "memory(GiB)": 77.59, "step": 103765, "token_acc": 0.8966565349544073, "train_speed(iter/s)": 1.461959 }, { "epoch": 4.445824943232938, "grad_norm": 1.9153841733932495, "learning_rate": 3.000553344086654e-06, "loss": 0.37224178314208983, "memory(GiB)": 77.59, "step": 103770, "token_acc": 0.9151943462897526, "train_speed(iter/s)": 1.461962 }, { "epoch": 4.446039158562186, "grad_norm": 3.670908212661743, "learning_rate": 2.9982575380584576e-06, "loss": 0.2423185348510742, "memory(GiB)": 77.59, "step": 103775, "token_acc": 0.9434628975265018, "train_speed(iter/s)": 1.461963 }, { "epoch": 4.446253373891436, "grad_norm": 0.6295031309127808, "learning_rate": 2.995962583510914e-06, "loss": 0.17504230737686158, "memory(GiB)": 77.59, "step": 103780, "token_acc": 0.9565217391304348, "train_speed(iter/s)": 1.46197 }, { "epoch": 4.446467589220685, "grad_norm": 1.8963632583618164, "learning_rate": 2.9936684804856064e-06, "loss": 0.2372905731201172, "memory(GiB)": 77.59, "step": 103785, "token_acc": 0.9437086092715232, "train_speed(iter/s)": 1.461978 }, { "epoch": 4.446681804549933, "grad_norm": 2.994945526123047, "learning_rate": 2.991375229024079e-06, "loss": 0.3874929428100586, "memory(GiB)": 77.59, "step": 103790, "token_acc": 0.9244186046511628, "train_speed(iter/s)": 1.461979 }, { "epoch": 4.4468960198791825, "grad_norm": 5.186663627624512, "learning_rate": 2.989082829167894e-06, "loss": 0.3472152233123779, "memory(GiB)": 77.59, "step": 103795, "token_acc": 0.9354838709677419, "train_speed(iter/s)": 1.461982 }, { "epoch": 4.447110235208432, "grad_norm": 2.8713810443878174, "learning_rate": 2.9867912809585784e-06, "loss": 0.2722845792770386, "memory(GiB)": 77.59, "step": 103800, "token_acc": 0.9520295202952029, "train_speed(iter/s)": 1.461986 }, { "epoch": 4.44732445053768, "grad_norm": 4.916960716247559, "learning_rate": 2.9845005844376327e-06, "loss": 0.4909525871276855, "memory(GiB)": 77.59, "step": 103805, "token_acc": 0.9093851132686084, "train_speed(iter/s)": 1.461988 }, { "epoch": 4.447538665866929, "grad_norm": 2.84844708442688, "learning_rate": 2.982210739646563e-06, "loss": 0.2424246311187744, "memory(GiB)": 77.59, "step": 103810, "token_acc": 0.9471830985915493, "train_speed(iter/s)": 1.46199 }, { "epoch": 4.447752881196179, "grad_norm": 6.381822109222412, "learning_rate": 2.979921746626857e-06, "loss": 0.6195128917694092, "memory(GiB)": 77.59, "step": 103815, "token_acc": 0.8738170347003155, "train_speed(iter/s)": 1.461999 }, { "epoch": 4.447967096525427, "grad_norm": 0.7420321106910706, "learning_rate": 2.977633605419966e-06, "loss": 0.2596641778945923, "memory(GiB)": 77.59, "step": 103820, "token_acc": 0.9074626865671642, "train_speed(iter/s)": 1.461998 }, { "epoch": 4.448181311854676, "grad_norm": 8.23372745513916, "learning_rate": 2.9753463160673624e-06, "loss": 0.3930727481842041, "memory(GiB)": 77.59, "step": 103825, "token_acc": 0.9131832797427653, "train_speed(iter/s)": 1.462012 }, { "epoch": 4.4483955271839255, "grad_norm": 4.086063861846924, "learning_rate": 2.9730598786104735e-06, "loss": 0.404171085357666, "memory(GiB)": 77.59, "step": 103830, "token_acc": 0.9261744966442953, "train_speed(iter/s)": 1.462012 }, { "epoch": 4.448609742513174, "grad_norm": 1.3492401838302612, "learning_rate": 2.970774293090722e-06, "loss": 0.42120137214660647, "memory(GiB)": 77.59, "step": 103835, "token_acc": 0.9059233449477352, "train_speed(iter/s)": 1.462018 }, { "epoch": 4.448823957842423, "grad_norm": 2.7264513969421387, "learning_rate": 2.9684895595495088e-06, "loss": 0.41034865379333496, "memory(GiB)": 77.59, "step": 103840, "token_acc": 0.8900343642611683, "train_speed(iter/s)": 1.462019 }, { "epoch": 4.449038173171672, "grad_norm": 2.630734920501709, "learning_rate": 2.9662056780282275e-06, "loss": 0.24592857360839843, "memory(GiB)": 77.59, "step": 103845, "token_acc": 0.9333333333333333, "train_speed(iter/s)": 1.462018 }, { "epoch": 4.449252388500921, "grad_norm": 1.8049893379211426, "learning_rate": 2.9639226485682457e-06, "loss": 0.3202247381210327, "memory(GiB)": 77.59, "step": 103850, "token_acc": 0.9358108108108109, "train_speed(iter/s)": 1.462019 }, { "epoch": 4.44946660383017, "grad_norm": 2.4887492656707764, "learning_rate": 2.9616404712109304e-06, "loss": 0.2486950159072876, "memory(GiB)": 77.59, "step": 103855, "token_acc": 0.9364548494983278, "train_speed(iter/s)": 1.462029 }, { "epoch": 4.449680819159419, "grad_norm": 2.516275405883789, "learning_rate": 2.9593591459976256e-06, "loss": 0.3666654586791992, "memory(GiB)": 77.59, "step": 103860, "token_acc": 0.9221556886227545, "train_speed(iter/s)": 1.462033 }, { "epoch": 4.449895034488668, "grad_norm": 4.523772239685059, "learning_rate": 2.9570786729696598e-06, "loss": 0.29479331970214845, "memory(GiB)": 77.59, "step": 103865, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.462037 }, { "epoch": 4.450109249817917, "grad_norm": 3.851304531097412, "learning_rate": 2.9547990521683444e-06, "loss": 0.4711584568023682, "memory(GiB)": 77.59, "step": 103870, "token_acc": 0.9077490774907749, "train_speed(iter/s)": 1.46204 }, { "epoch": 4.450323465147166, "grad_norm": 7.960512638092041, "learning_rate": 2.9525202836349686e-06, "loss": 0.29643001556396487, "memory(GiB)": 77.59, "step": 103875, "token_acc": 0.9394904458598726, "train_speed(iter/s)": 1.462039 }, { "epoch": 4.4505376804764145, "grad_norm": 2.696281909942627, "learning_rate": 2.950242367410827e-06, "loss": 0.3579278469085693, "memory(GiB)": 77.59, "step": 103880, "token_acc": 0.9135338345864662, "train_speed(iter/s)": 1.462038 }, { "epoch": 4.450751895805664, "grad_norm": 3.691392421722412, "learning_rate": 2.94796530353717e-06, "loss": 0.2633510589599609, "memory(GiB)": 77.59, "step": 103885, "token_acc": 0.9512195121951219, "train_speed(iter/s)": 1.462038 }, { "epoch": 4.450966111134913, "grad_norm": 2.3631882667541504, "learning_rate": 2.945689092055276e-06, "loss": 0.19654561281204225, "memory(GiB)": 77.59, "step": 103890, "token_acc": 0.9602888086642599, "train_speed(iter/s)": 1.462042 }, { "epoch": 4.451180326464161, "grad_norm": 6.8982110023498535, "learning_rate": 2.9434137330063614e-06, "loss": 0.3567115545272827, "memory(GiB)": 77.59, "step": 103895, "token_acc": 0.9254237288135593, "train_speed(iter/s)": 1.462047 }, { "epoch": 4.451394541793411, "grad_norm": 5.190340995788574, "learning_rate": 2.941139226431644e-06, "loss": 0.4026597499847412, "memory(GiB)": 77.59, "step": 103900, "token_acc": 0.8923076923076924, "train_speed(iter/s)": 1.462047 }, { "epoch": 4.45160875712266, "grad_norm": 3.6093668937683105, "learning_rate": 2.9388655723723345e-06, "loss": 0.20901720523834227, "memory(GiB)": 77.59, "step": 103905, "token_acc": 0.9450867052023122, "train_speed(iter/s)": 1.462049 }, { "epoch": 4.451822972451908, "grad_norm": 4.3944902420043945, "learning_rate": 2.936592770869628e-06, "loss": 0.47823801040649416, "memory(GiB)": 77.59, "step": 103910, "token_acc": 0.8986013986013986, "train_speed(iter/s)": 1.462053 }, { "epoch": 4.4520371877811575, "grad_norm": 2.2793631553649902, "learning_rate": 2.9343208219646867e-06, "loss": 0.3184140920639038, "memory(GiB)": 77.59, "step": 103915, "token_acc": 0.932258064516129, "train_speed(iter/s)": 1.462054 }, { "epoch": 4.452251403110407, "grad_norm": 3.4915707111358643, "learning_rate": 2.9320497256986822e-06, "loss": 0.43050398826599123, "memory(GiB)": 77.59, "step": 103920, "token_acc": 0.8954703832752613, "train_speed(iter/s)": 1.462055 }, { "epoch": 4.452465618439655, "grad_norm": 3.6220383644104004, "learning_rate": 2.9297794821127488e-06, "loss": 0.44118757247924806, "memory(GiB)": 77.59, "step": 103925, "token_acc": 0.9116465863453815, "train_speed(iter/s)": 1.462052 }, { "epoch": 4.452679833768904, "grad_norm": 1.7128651142120361, "learning_rate": 2.927510091248009e-06, "loss": 0.2299304723739624, "memory(GiB)": 77.59, "step": 103930, "token_acc": 0.9326599326599326, "train_speed(iter/s)": 1.462064 }, { "epoch": 4.452894049098154, "grad_norm": 4.882075309753418, "learning_rate": 2.9252415531455802e-06, "loss": 0.4392275333404541, "memory(GiB)": 77.59, "step": 103935, "token_acc": 0.9172185430463576, "train_speed(iter/s)": 1.462064 }, { "epoch": 4.453108264427402, "grad_norm": 1.7327702045440674, "learning_rate": 2.922973867846568e-06, "loss": 0.1479145646095276, "memory(GiB)": 77.59, "step": 103940, "token_acc": 0.9628378378378378, "train_speed(iter/s)": 1.462067 }, { "epoch": 4.453322479756651, "grad_norm": 1.2407264709472656, "learning_rate": 2.920707035392045e-06, "loss": 0.12585527896881105, "memory(GiB)": 77.59, "step": 103945, "token_acc": 0.9655172413793104, "train_speed(iter/s)": 1.46207 }, { "epoch": 4.4535366950859006, "grad_norm": 4.454718589782715, "learning_rate": 2.918441055823079e-06, "loss": 0.2791996955871582, "memory(GiB)": 77.59, "step": 103950, "token_acc": 0.9246575342465754, "train_speed(iter/s)": 1.462072 }, { "epoch": 4.453750910415149, "grad_norm": 0.307831346988678, "learning_rate": 2.9161759291807144e-06, "loss": 0.20160927772521972, "memory(GiB)": 77.59, "step": 103955, "token_acc": 0.9572953736654805, "train_speed(iter/s)": 1.46208 }, { "epoch": 4.453965125744398, "grad_norm": 4.740945339202881, "learning_rate": 2.913911655505991e-06, "loss": 0.5867267608642578, "memory(GiB)": 77.59, "step": 103960, "token_acc": 0.8801261829652997, "train_speed(iter/s)": 1.462082 }, { "epoch": 4.454179341073647, "grad_norm": 5.031988143920898, "learning_rate": 2.91164823483992e-06, "loss": 0.19093526601791383, "memory(GiB)": 77.59, "step": 103965, "token_acc": 0.9663865546218487, "train_speed(iter/s)": 1.462083 }, { "epoch": 4.454393556402896, "grad_norm": 1.8320974111557007, "learning_rate": 2.9093856672235297e-06, "loss": 0.43674869537353517, "memory(GiB)": 77.59, "step": 103970, "token_acc": 0.8955696202531646, "train_speed(iter/s)": 1.462082 }, { "epoch": 4.454607771732145, "grad_norm": 2.5716617107391357, "learning_rate": 2.907123952697788e-06, "loss": 0.4851999759674072, "memory(GiB)": 77.59, "step": 103975, "token_acc": 0.9099099099099099, "train_speed(iter/s)": 1.462082 }, { "epoch": 4.454821987061394, "grad_norm": 2.240943670272827, "learning_rate": 2.9048630913036723e-06, "loss": 0.3591834306716919, "memory(GiB)": 77.59, "step": 103980, "token_acc": 0.9111111111111111, "train_speed(iter/s)": 1.462078 }, { "epoch": 4.455036202390643, "grad_norm": 2.60434627532959, "learning_rate": 2.9026030830821396e-06, "loss": 0.2318873643875122, "memory(GiB)": 77.59, "step": 103985, "token_acc": 0.9423728813559322, "train_speed(iter/s)": 1.462079 }, { "epoch": 4.455250417719892, "grad_norm": 3.8199777603149414, "learning_rate": 2.900343928074134e-06, "loss": 0.24772143363952637, "memory(GiB)": 77.59, "step": 103990, "token_acc": 0.956, "train_speed(iter/s)": 1.462083 }, { "epoch": 4.455464633049141, "grad_norm": 1.678515911102295, "learning_rate": 2.8980856263205736e-06, "loss": 0.3344932794570923, "memory(GiB)": 77.59, "step": 103995, "token_acc": 0.9384057971014492, "train_speed(iter/s)": 1.462083 }, { "epoch": 4.45567884837839, "grad_norm": 3.2960526943206787, "learning_rate": 2.895828177862386e-06, "loss": 0.18020739555358886, "memory(GiB)": 77.59, "step": 104000, "token_acc": 0.9619771863117871, "train_speed(iter/s)": 1.462081 }, { "epoch": 4.45567884837839, "eval_loss": 2.3643109798431396, "eval_runtime": 11.7086, "eval_samples_per_second": 8.541, "eval_steps_per_second": 8.541, "eval_token_acc": 0.43757292882147025, "step": 104000 }, { "epoch": 4.455893063707639, "grad_norm": 4.603425979614258, "learning_rate": 2.893571582740462e-06, "loss": 0.46727876663208007, "memory(GiB)": 77.59, "step": 104005, "token_acc": 0.5390909090909091, "train_speed(iter/s)": 1.461818 }, { "epoch": 4.456107279036888, "grad_norm": 2.8062376976013184, "learning_rate": 2.8913158409956733e-06, "loss": 0.3506906032562256, "memory(GiB)": 77.59, "step": 104010, "token_acc": 0.931740614334471, "train_speed(iter/s)": 1.461822 }, { "epoch": 4.4563214943661364, "grad_norm": 1.980615496635437, "learning_rate": 2.8890609526688873e-06, "loss": 0.31967933177948, "memory(GiB)": 77.59, "step": 104015, "token_acc": 0.9221311475409836, "train_speed(iter/s)": 1.461838 }, { "epoch": 4.456535709695386, "grad_norm": 4.316009044647217, "learning_rate": 2.886806917800955e-06, "loss": 0.5385609149932862, "memory(GiB)": 77.59, "step": 104020, "token_acc": 0.8892857142857142, "train_speed(iter/s)": 1.46184 }, { "epoch": 4.456749925024635, "grad_norm": 4.923282146453857, "learning_rate": 2.8845537364327104e-06, "loss": 0.3915215015411377, "memory(GiB)": 77.59, "step": 104025, "token_acc": 0.9236111111111112, "train_speed(iter/s)": 1.46184 }, { "epoch": 4.456964140353883, "grad_norm": 2.4715723991394043, "learning_rate": 2.8823014086049704e-06, "loss": 0.4422764778137207, "memory(GiB)": 77.59, "step": 104030, "token_acc": 0.9139784946236559, "train_speed(iter/s)": 1.461845 }, { "epoch": 4.457178355683133, "grad_norm": 1.9842675924301147, "learning_rate": 2.8800499343585474e-06, "loss": 0.4127791881561279, "memory(GiB)": 77.59, "step": 104035, "token_acc": 0.8921933085501859, "train_speed(iter/s)": 1.461853 }, { "epoch": 4.457392571012382, "grad_norm": 3.441930055618286, "learning_rate": 2.8777993137342197e-06, "loss": 0.3377416133880615, "memory(GiB)": 77.59, "step": 104040, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.461857 }, { "epoch": 4.45760678634163, "grad_norm": 2.3403170108795166, "learning_rate": 2.8755495467727657e-06, "loss": 0.5872798919677734, "memory(GiB)": 77.59, "step": 104045, "token_acc": 0.8842443729903537, "train_speed(iter/s)": 1.461858 }, { "epoch": 4.4578210016708795, "grad_norm": 2.1897499561309814, "learning_rate": 2.8733006335149305e-06, "loss": 0.48862438201904296, "memory(GiB)": 77.59, "step": 104050, "token_acc": 0.8952380952380953, "train_speed(iter/s)": 1.461871 }, { "epoch": 4.458035217000129, "grad_norm": 1.2177646160125732, "learning_rate": 2.871052574001465e-06, "loss": 0.2994356870651245, "memory(GiB)": 77.59, "step": 104055, "token_acc": 0.9496644295302014, "train_speed(iter/s)": 1.461869 }, { "epoch": 4.458249432329377, "grad_norm": 1.7037659883499146, "learning_rate": 2.8688053682730973e-06, "loss": 0.4464524269104004, "memory(GiB)": 77.59, "step": 104060, "token_acc": 0.8670886075949367, "train_speed(iter/s)": 1.461879 }, { "epoch": 4.458463647658626, "grad_norm": 3.2803070545196533, "learning_rate": 2.8665590163705293e-06, "loss": 0.20868985652923583, "memory(GiB)": 77.59, "step": 104065, "token_acc": 0.9477351916376306, "train_speed(iter/s)": 1.461882 }, { "epoch": 4.458677862987876, "grad_norm": 2.2172904014587402, "learning_rate": 2.864313518334466e-06, "loss": 0.3425022602081299, "memory(GiB)": 77.59, "step": 104070, "token_acc": 0.9196141479099679, "train_speed(iter/s)": 1.461887 }, { "epoch": 4.458892078317124, "grad_norm": 1.2031384706497192, "learning_rate": 2.862068874205576e-06, "loss": 0.18543314933776855, "memory(GiB)": 77.59, "step": 104075, "token_acc": 0.9572953736654805, "train_speed(iter/s)": 1.461887 }, { "epoch": 4.459106293646373, "grad_norm": 6.712139129638672, "learning_rate": 2.85982508402452e-06, "loss": 0.5009604454040527, "memory(GiB)": 77.59, "step": 104080, "token_acc": 0.8688524590163934, "train_speed(iter/s)": 1.461888 }, { "epoch": 4.4593205089756225, "grad_norm": 2.502268075942993, "learning_rate": 2.8575821478319554e-06, "loss": 0.22759225368499755, "memory(GiB)": 77.59, "step": 104085, "token_acc": 0.9468085106382979, "train_speed(iter/s)": 1.461889 }, { "epoch": 4.459534724304871, "grad_norm": 1.603842854499817, "learning_rate": 2.8553400656685216e-06, "loss": 0.39304242134094236, "memory(GiB)": 77.59, "step": 104090, "token_acc": 0.9125, "train_speed(iter/s)": 1.46189 }, { "epoch": 4.45974893963412, "grad_norm": 2.121516227722168, "learning_rate": 2.8530988375748247e-06, "loss": 0.2605544328689575, "memory(GiB)": 77.59, "step": 104095, "token_acc": 0.9436619718309859, "train_speed(iter/s)": 1.461895 }, { "epoch": 4.459963154963369, "grad_norm": 3.39841890335083, "learning_rate": 2.8508584635914715e-06, "loss": 0.35621330738067625, "memory(GiB)": 77.59, "step": 104100, "token_acc": 0.9324324324324325, "train_speed(iter/s)": 1.461898 }, { "epoch": 4.460177370292618, "grad_norm": 1.714925765991211, "learning_rate": 2.848618943759035e-06, "loss": 0.2812263250350952, "memory(GiB)": 77.59, "step": 104105, "token_acc": 0.952112676056338, "train_speed(iter/s)": 1.461901 }, { "epoch": 4.460391585621867, "grad_norm": 0.24342931807041168, "learning_rate": 2.846380278118099e-06, "loss": 0.3150906801223755, "memory(GiB)": 77.59, "step": 104110, "token_acc": 0.9247648902821317, "train_speed(iter/s)": 1.461904 }, { "epoch": 4.460605800951116, "grad_norm": 3.1562435626983643, "learning_rate": 2.844142466709232e-06, "loss": 0.2434466600418091, "memory(GiB)": 77.59, "step": 104115, "token_acc": 0.9326241134751773, "train_speed(iter/s)": 1.461909 }, { "epoch": 4.460820016280365, "grad_norm": 3.653442144393921, "learning_rate": 2.8419055095729507e-06, "loss": 0.5157814979553222, "memory(GiB)": 77.59, "step": 104120, "token_acc": 0.8764705882352941, "train_speed(iter/s)": 1.46191 }, { "epoch": 4.461034231609614, "grad_norm": 1.4022047519683838, "learning_rate": 2.839669406749795e-06, "loss": 0.2872201681137085, "memory(GiB)": 77.59, "step": 104125, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.461914 }, { "epoch": 4.461248446938863, "grad_norm": 4.542088031768799, "learning_rate": 2.8374341582802664e-06, "loss": 0.160847270488739, "memory(GiB)": 77.59, "step": 104130, "token_acc": 0.9562043795620438, "train_speed(iter/s)": 1.461916 }, { "epoch": 4.4614626622681115, "grad_norm": 3.3934848308563232, "learning_rate": 2.83519976420486e-06, "loss": 0.24818286895751954, "memory(GiB)": 77.59, "step": 104135, "token_acc": 0.9477611940298507, "train_speed(iter/s)": 1.461922 }, { "epoch": 4.461676877597361, "grad_norm": 6.510392665863037, "learning_rate": 2.8329662245640487e-06, "loss": 0.5267648696899414, "memory(GiB)": 77.59, "step": 104140, "token_acc": 0.8871595330739299, "train_speed(iter/s)": 1.461923 }, { "epoch": 4.46189109292661, "grad_norm": 7.723862648010254, "learning_rate": 2.830733539398306e-06, "loss": 0.4167473316192627, "memory(GiB)": 77.59, "step": 104145, "token_acc": 0.9, "train_speed(iter/s)": 1.461922 }, { "epoch": 4.462105308255858, "grad_norm": 5.567760467529297, "learning_rate": 2.8285017087480715e-06, "loss": 0.3350059032440186, "memory(GiB)": 77.59, "step": 104150, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.461921 }, { "epoch": 4.462319523585108, "grad_norm": 3.1194891929626465, "learning_rate": 2.8262707326537795e-06, "loss": 0.6842159271240235, "memory(GiB)": 77.59, "step": 104155, "token_acc": 0.8798701298701299, "train_speed(iter/s)": 1.461918 }, { "epoch": 4.462533738914357, "grad_norm": 4.6910481452941895, "learning_rate": 2.824040611155848e-06, "loss": 0.24832522869110107, "memory(GiB)": 77.59, "step": 104160, "token_acc": 0.9498069498069498, "train_speed(iter/s)": 1.461928 }, { "epoch": 4.462747954243605, "grad_norm": 1.0220218896865845, "learning_rate": 2.821811344294667e-06, "loss": 0.30269436836242675, "memory(GiB)": 77.59, "step": 104165, "token_acc": 0.9422382671480144, "train_speed(iter/s)": 1.461929 }, { "epoch": 4.4629621695728545, "grad_norm": 2.968899965286255, "learning_rate": 2.819582932110637e-06, "loss": 0.35331144332885744, "memory(GiB)": 77.59, "step": 104170, "token_acc": 0.9276315789473685, "train_speed(iter/s)": 1.461932 }, { "epoch": 4.463176384902104, "grad_norm": 3.7761919498443604, "learning_rate": 2.8173553746441207e-06, "loss": 0.4803877830505371, "memory(GiB)": 77.59, "step": 104175, "token_acc": 0.8948948948948949, "train_speed(iter/s)": 1.461935 }, { "epoch": 4.463390600231352, "grad_norm": 3.0407752990722656, "learning_rate": 2.8151286719354686e-06, "loss": 0.2137500286102295, "memory(GiB)": 77.59, "step": 104180, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.461939 }, { "epoch": 4.463604815560601, "grad_norm": 3.515486001968384, "learning_rate": 2.8129028240250267e-06, "loss": 0.37090210914611815, "memory(GiB)": 77.59, "step": 104185, "token_acc": 0.9245901639344263, "train_speed(iter/s)": 1.46194 }, { "epoch": 4.463819030889851, "grad_norm": 4.4620161056518555, "learning_rate": 2.8106778309531178e-06, "loss": 0.24850037097930908, "memory(GiB)": 77.59, "step": 104190, "token_acc": 0.9322033898305084, "train_speed(iter/s)": 1.46195 }, { "epoch": 4.464033246219099, "grad_norm": 5.180398464202881, "learning_rate": 2.808453692760038e-06, "loss": 0.34868898391723635, "memory(GiB)": 77.59, "step": 104195, "token_acc": 0.9082278481012658, "train_speed(iter/s)": 1.461947 }, { "epoch": 4.464247461548348, "grad_norm": 0.1826580911874771, "learning_rate": 2.806230409486099e-06, "loss": 0.22156155109405518, "memory(GiB)": 77.59, "step": 104200, "token_acc": 0.96, "train_speed(iter/s)": 1.461948 }, { "epoch": 4.4644616768775975, "grad_norm": 6.396040439605713, "learning_rate": 2.8040079811715638e-06, "loss": 0.3268155336380005, "memory(GiB)": 77.59, "step": 104205, "token_acc": 0.928082191780822, "train_speed(iter/s)": 1.46196 }, { "epoch": 4.464675892206846, "grad_norm": 3.8500068187713623, "learning_rate": 2.8017864078566992e-06, "loss": 0.5183325290679932, "memory(GiB)": 77.59, "step": 104210, "token_acc": 0.8854166666666666, "train_speed(iter/s)": 1.461971 }, { "epoch": 4.464890107536095, "grad_norm": 1.8573366403579712, "learning_rate": 2.7995656895817513e-06, "loss": 0.19465553760528564, "memory(GiB)": 77.59, "step": 104215, "token_acc": 0.9560117302052786, "train_speed(iter/s)": 1.461971 }, { "epoch": 4.465104322865344, "grad_norm": 4.343760013580322, "learning_rate": 2.797345826386949e-06, "loss": 0.3389659643173218, "memory(GiB)": 77.59, "step": 104220, "token_acc": 0.9325842696629213, "train_speed(iter/s)": 1.461974 }, { "epoch": 4.465318538194593, "grad_norm": 2.1519696712493896, "learning_rate": 2.7951268183124935e-06, "loss": 0.23233304023742676, "memory(GiB)": 77.59, "step": 104225, "token_acc": 0.9420849420849421, "train_speed(iter/s)": 1.461977 }, { "epoch": 4.465532753523842, "grad_norm": 0.8873899579048157, "learning_rate": 2.792908665398608e-06, "loss": 0.15356011390686036, "memory(GiB)": 77.59, "step": 104230, "token_acc": 0.9681978798586572, "train_speed(iter/s)": 1.461988 }, { "epoch": 4.465746968853091, "grad_norm": 5.068668842315674, "learning_rate": 2.7906913676854664e-06, "loss": 0.5286267757415771, "memory(GiB)": 77.59, "step": 104235, "token_acc": 0.9018987341772152, "train_speed(iter/s)": 1.461991 }, { "epoch": 4.465961184182341, "grad_norm": 3.5429487228393555, "learning_rate": 2.788474925213236e-06, "loss": 0.5030910968780518, "memory(GiB)": 77.59, "step": 104240, "token_acc": 0.9111842105263158, "train_speed(iter/s)": 1.461995 }, { "epoch": 4.466175399511589, "grad_norm": 1.962922215461731, "learning_rate": 2.7862593380220737e-06, "loss": 0.3764235734939575, "memory(GiB)": 77.59, "step": 104245, "token_acc": 0.8986486486486487, "train_speed(iter/s)": 1.461999 }, { "epoch": 4.466389614840838, "grad_norm": 2.6914689540863037, "learning_rate": 2.784044606152103e-06, "loss": 0.24869742393493652, "memory(GiB)": 77.59, "step": 104250, "token_acc": 0.9290780141843972, "train_speed(iter/s)": 1.462005 }, { "epoch": 4.4666038301700866, "grad_norm": 2.855651617050171, "learning_rate": 2.7818307296434585e-06, "loss": 0.30635900497436525, "memory(GiB)": 77.59, "step": 104255, "token_acc": 0.9525423728813559, "train_speed(iter/s)": 1.462006 }, { "epoch": 4.466818045499336, "grad_norm": 0.2388760894536972, "learning_rate": 2.7796177085362476e-06, "loss": 0.28422491550445556, "memory(GiB)": 77.59, "step": 104260, "token_acc": 0.9338235294117647, "train_speed(iter/s)": 1.462012 }, { "epoch": 4.467032260828585, "grad_norm": 2.5575077533721924, "learning_rate": 2.777405542870559e-06, "loss": 0.6528433322906494, "memory(GiB)": 77.59, "step": 104265, "token_acc": 0.8695652173913043, "train_speed(iter/s)": 1.462021 }, { "epoch": 4.467246476157834, "grad_norm": 4.237175464630127, "learning_rate": 2.775194232686473e-06, "loss": 0.22542085647583007, "memory(GiB)": 77.59, "step": 104270, "token_acc": 0.9578947368421052, "train_speed(iter/s)": 1.462022 }, { "epoch": 4.467460691487083, "grad_norm": 3.9007139205932617, "learning_rate": 2.772983778024035e-06, "loss": 0.38319859504699705, "memory(GiB)": 77.59, "step": 104275, "token_acc": 0.9250936329588015, "train_speed(iter/s)": 1.462023 }, { "epoch": 4.467674906816332, "grad_norm": 2.2109458446502686, "learning_rate": 2.7707741789233078e-06, "loss": 0.13011841773986815, "memory(GiB)": 77.59, "step": 104280, "token_acc": 0.9709090909090909, "train_speed(iter/s)": 1.462027 }, { "epoch": 4.46788912214558, "grad_norm": 2.5778379440307617, "learning_rate": 2.768565435424297e-06, "loss": 0.32842378616333007, "memory(GiB)": 77.59, "step": 104285, "token_acc": 0.9136690647482014, "train_speed(iter/s)": 1.462037 }, { "epoch": 4.46810333747483, "grad_norm": 2.647041082382202, "learning_rate": 2.7663575475670387e-06, "loss": 0.5767872333526611, "memory(GiB)": 77.59, "step": 104290, "token_acc": 0.8870967741935484, "train_speed(iter/s)": 1.462048 }, { "epoch": 4.468317552804079, "grad_norm": 4.047182559967041, "learning_rate": 2.7641505153915227e-06, "loss": 0.20273053646087646, "memory(GiB)": 77.59, "step": 104295, "token_acc": 0.9536679536679536, "train_speed(iter/s)": 1.462049 }, { "epoch": 4.468531768133328, "grad_norm": 3.1739583015441895, "learning_rate": 2.761944338937733e-06, "loss": 0.3710179805755615, "memory(GiB)": 77.59, "step": 104300, "token_acc": 0.9280575539568345, "train_speed(iter/s)": 1.462061 }, { "epoch": 4.4687459834625765, "grad_norm": 4.518255233764648, "learning_rate": 2.7597390182456327e-06, "loss": 0.5511866092681885, "memory(GiB)": 77.59, "step": 104305, "token_acc": 0.8992805755395683, "train_speed(iter/s)": 1.46208 }, { "epoch": 4.468960198791826, "grad_norm": 3.0287868976593018, "learning_rate": 2.757534553355168e-06, "loss": 0.30171923637390136, "memory(GiB)": 77.59, "step": 104310, "token_acc": 0.9345454545454546, "train_speed(iter/s)": 1.462081 }, { "epoch": 4.469174414121074, "grad_norm": 4.694368839263916, "learning_rate": 2.755330944306289e-06, "loss": 0.16708019971847535, "memory(GiB)": 77.59, "step": 104315, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.462083 }, { "epoch": 4.469388629450323, "grad_norm": 5.063000202178955, "learning_rate": 2.7531281911389094e-06, "loss": 0.46062164306640624, "memory(GiB)": 77.59, "step": 104320, "token_acc": 0.8819672131147541, "train_speed(iter/s)": 1.462093 }, { "epoch": 4.469602844779573, "grad_norm": 8.115987777709961, "learning_rate": 2.750926293892925e-06, "loss": 0.2606045246124268, "memory(GiB)": 77.59, "step": 104325, "token_acc": 0.9311594202898551, "train_speed(iter/s)": 1.462092 }, { "epoch": 4.469817060108822, "grad_norm": 2.72080659866333, "learning_rate": 2.748725252608242e-06, "loss": 0.118782377243042, "memory(GiB)": 77.59, "step": 104330, "token_acc": 0.9684542586750788, "train_speed(iter/s)": 1.462095 }, { "epoch": 4.47003127543807, "grad_norm": 3.953704833984375, "learning_rate": 2.746525067324729e-06, "loss": 0.38954098224639894, "memory(GiB)": 77.59, "step": 104335, "token_acc": 0.925, "train_speed(iter/s)": 1.462095 }, { "epoch": 4.4702454907673195, "grad_norm": 3.629615306854248, "learning_rate": 2.7443257380822318e-06, "loss": 0.34512326717376707, "memory(GiB)": 77.59, "step": 104340, "token_acc": 0.9222222222222223, "train_speed(iter/s)": 1.462097 }, { "epoch": 4.470459706096568, "grad_norm": 2.324547290802002, "learning_rate": 2.7421272649206077e-06, "loss": 0.27842040061950685, "memory(GiB)": 77.59, "step": 104345, "token_acc": 0.9258160237388724, "train_speed(iter/s)": 1.462099 }, { "epoch": 4.470673921425817, "grad_norm": 1.4305052757263184, "learning_rate": 2.7399296478796855e-06, "loss": 0.2247305393218994, "memory(GiB)": 77.59, "step": 104350, "token_acc": 0.9475524475524476, "train_speed(iter/s)": 1.4621 }, { "epoch": 4.470888136755066, "grad_norm": 3.4940414428710938, "learning_rate": 2.737732886999267e-06, "loss": 0.3040785551071167, "memory(GiB)": 77.59, "step": 104355, "token_acc": 0.9243027888446215, "train_speed(iter/s)": 1.4621 }, { "epoch": 4.471102352084316, "grad_norm": 1.8433101177215576, "learning_rate": 2.7355369823191536e-06, "loss": 0.22303056716918945, "memory(GiB)": 77.59, "step": 104360, "token_acc": 0.9340659340659341, "train_speed(iter/s)": 1.4621 }, { "epoch": 4.471316567413564, "grad_norm": 3.691006898880005, "learning_rate": 2.73334193387913e-06, "loss": 0.3218871593475342, "memory(GiB)": 77.59, "step": 104365, "token_acc": 0.9221311475409836, "train_speed(iter/s)": 1.46211 }, { "epoch": 4.471530782742813, "grad_norm": 5.14331579208374, "learning_rate": 2.7311477417189424e-06, "loss": 0.5045305728912354, "memory(GiB)": 77.59, "step": 104370, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.462111 }, { "epoch": 4.471744998072062, "grad_norm": 2.6271374225616455, "learning_rate": 2.72895440587837e-06, "loss": 0.2138070583343506, "memory(GiB)": 77.59, "step": 104375, "token_acc": 0.9446366782006921, "train_speed(iter/s)": 1.462116 }, { "epoch": 4.471959213401311, "grad_norm": 5.151856899261475, "learning_rate": 2.7267619263971255e-06, "loss": 0.2834265470504761, "memory(GiB)": 77.59, "step": 104380, "token_acc": 0.931899641577061, "train_speed(iter/s)": 1.462114 }, { "epoch": 4.47217342873056, "grad_norm": 3.136428117752075, "learning_rate": 2.7245703033149384e-06, "loss": 0.4546207427978516, "memory(GiB)": 77.59, "step": 104385, "token_acc": 0.8834688346883469, "train_speed(iter/s)": 1.462111 }, { "epoch": 4.472387644059809, "grad_norm": 3.7679545879364014, "learning_rate": 2.7223795366715098e-06, "loss": 0.35036420822143555, "memory(GiB)": 77.59, "step": 104390, "token_acc": 0.9109195402298851, "train_speed(iter/s)": 1.462114 }, { "epoch": 4.472601859389058, "grad_norm": 5.448635101318359, "learning_rate": 2.720189626506514e-06, "loss": 0.33427183628082274, "memory(GiB)": 77.59, "step": 104395, "token_acc": 0.9254901960784314, "train_speed(iter/s)": 1.462116 }, { "epoch": 4.472816074718307, "grad_norm": 2.1271212100982666, "learning_rate": 2.718000572859636e-06, "loss": 0.41745333671569823, "memory(GiB)": 77.59, "step": 104400, "token_acc": 0.9100346020761245, "train_speed(iter/s)": 1.462119 }, { "epoch": 4.473030290047555, "grad_norm": 0.14517374336719513, "learning_rate": 2.7158123757705433e-06, "loss": 0.621551513671875, "memory(GiB)": 77.59, "step": 104405, "token_acc": 0.8901515151515151, "train_speed(iter/s)": 1.462121 }, { "epoch": 4.473244505376805, "grad_norm": 3.528852701187134, "learning_rate": 2.7136250352788606e-06, "loss": 0.4067336082458496, "memory(GiB)": 77.59, "step": 104410, "token_acc": 0.9210526315789473, "train_speed(iter/s)": 1.462129 }, { "epoch": 4.473458720706054, "grad_norm": 0.5702508091926575, "learning_rate": 2.7114385514242222e-06, "loss": 0.17033780813217164, "memory(GiB)": 77.59, "step": 104415, "token_acc": 0.9656357388316151, "train_speed(iter/s)": 1.462134 }, { "epoch": 4.473672936035303, "grad_norm": 4.690647602081299, "learning_rate": 2.7092529242462307e-06, "loss": 0.4608471870422363, "memory(GiB)": 77.59, "step": 104420, "token_acc": 0.8813559322033898, "train_speed(iter/s)": 1.462134 }, { "epoch": 4.4738871513645515, "grad_norm": 1.3680111169815063, "learning_rate": 2.707068153784481e-06, "loss": 0.21597895622253419, "memory(GiB)": 77.59, "step": 104425, "token_acc": 0.9597069597069597, "train_speed(iter/s)": 1.462135 }, { "epoch": 4.474101366693801, "grad_norm": 3.1835711002349854, "learning_rate": 2.704884240078559e-06, "loss": 0.4265460968017578, "memory(GiB)": 77.59, "step": 104430, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.462146 }, { "epoch": 4.474315582023049, "grad_norm": 5.836519718170166, "learning_rate": 2.702701183168033e-06, "loss": 0.3601263761520386, "memory(GiB)": 77.59, "step": 104435, "token_acc": 0.924, "train_speed(iter/s)": 1.46215 }, { "epoch": 4.474529797352298, "grad_norm": 8.130420684814453, "learning_rate": 2.7005189830924373e-06, "loss": 0.6074734210968018, "memory(GiB)": 77.59, "step": 104440, "token_acc": 0.8731117824773413, "train_speed(iter/s)": 1.462153 }, { "epoch": 4.474744012681548, "grad_norm": 6.734340667724609, "learning_rate": 2.6983376398913128e-06, "loss": 0.4426288604736328, "memory(GiB)": 77.59, "step": 104445, "token_acc": 0.9084507042253521, "train_speed(iter/s)": 1.462157 }, { "epoch": 4.474958228010797, "grad_norm": 3.0957424640655518, "learning_rate": 2.696157153604173e-06, "loss": 0.3255736827850342, "memory(GiB)": 77.59, "step": 104450, "token_acc": 0.9239543726235742, "train_speed(iter/s)": 1.462169 }, { "epoch": 4.475172443340045, "grad_norm": 3.116875648498535, "learning_rate": 2.6939775242705133e-06, "loss": 0.3627466201782227, "memory(GiB)": 77.59, "step": 104455, "token_acc": 0.9163636363636364, "train_speed(iter/s)": 1.462174 }, { "epoch": 4.4753866586692945, "grad_norm": 3.29233980178833, "learning_rate": 2.6917987519298415e-06, "loss": 0.18958239555358886, "memory(GiB)": 77.59, "step": 104460, "token_acc": 0.9634146341463414, "train_speed(iter/s)": 1.462178 }, { "epoch": 4.475600873998543, "grad_norm": 2.448718547821045, "learning_rate": 2.6896208366216034e-06, "loss": 0.3165303707122803, "memory(GiB)": 77.59, "step": 104465, "token_acc": 0.9159891598915989, "train_speed(iter/s)": 1.462174 }, { "epoch": 4.475815089327792, "grad_norm": 6.8517584800720215, "learning_rate": 2.6874437783852734e-06, "loss": 0.3896085023880005, "memory(GiB)": 77.59, "step": 104470, "token_acc": 0.9161073825503355, "train_speed(iter/s)": 1.462174 }, { "epoch": 4.476029304657041, "grad_norm": 3.183243751525879, "learning_rate": 2.68526757726027e-06, "loss": 0.69489426612854, "memory(GiB)": 77.59, "step": 104475, "token_acc": 0.8461538461538461, "train_speed(iter/s)": 1.462179 }, { "epoch": 4.476243519986291, "grad_norm": 1.4630553722381592, "learning_rate": 2.6830922332860387e-06, "loss": 0.28382437229156493, "memory(GiB)": 77.59, "step": 104480, "token_acc": 0.9272727272727272, "train_speed(iter/s)": 1.46218 }, { "epoch": 4.476457735315539, "grad_norm": 4.48967981338501, "learning_rate": 2.6809177465019706e-06, "loss": 0.5157535552978516, "memory(GiB)": 77.59, "step": 104485, "token_acc": 0.87890625, "train_speed(iter/s)": 1.462183 }, { "epoch": 4.476671950644788, "grad_norm": 5.0270233154296875, "learning_rate": 2.6787441169474735e-06, "loss": 0.5557469367980957, "memory(GiB)": 77.59, "step": 104490, "token_acc": 0.872093023255814, "train_speed(iter/s)": 1.462198 }, { "epoch": 4.476886165974037, "grad_norm": 4.902878761291504, "learning_rate": 2.6765713446619157e-06, "loss": 0.4524225234985352, "memory(GiB)": 77.59, "step": 104495, "token_acc": 0.8858131487889274, "train_speed(iter/s)": 1.462208 }, { "epoch": 4.477100381303286, "grad_norm": 3.622185707092285, "learning_rate": 2.6743994296846653e-06, "loss": 0.2930592060089111, "memory(GiB)": 77.59, "step": 104500, "token_acc": 0.9270833333333334, "train_speed(iter/s)": 1.462217 }, { "epoch": 4.477100381303286, "eval_loss": 2.5453319549560547, "eval_runtime": 11.3188, "eval_samples_per_second": 8.835, "eval_steps_per_second": 8.835, "eval_token_acc": 0.43478260869565216, "step": 104500 }, { "epoch": 4.477314596632535, "grad_norm": 0.4037240445613861, "learning_rate": 2.6722283720550635e-06, "loss": 0.4509110927581787, "memory(GiB)": 77.59, "step": 104505, "token_acc": 0.5692007797270955, "train_speed(iter/s)": 1.461967 }, { "epoch": 4.477528811961784, "grad_norm": 4.867579460144043, "learning_rate": 2.6700581718124398e-06, "loss": 0.1601899027824402, "memory(GiB)": 77.59, "step": 104510, "token_acc": 0.9523809523809523, "train_speed(iter/s)": 1.461971 }, { "epoch": 4.477743027291033, "grad_norm": 5.898445129394531, "learning_rate": 2.6678888289961015e-06, "loss": 0.312380313873291, "memory(GiB)": 77.59, "step": 104515, "token_acc": 0.9509803921568627, "train_speed(iter/s)": 1.461982 }, { "epoch": 4.477957242620282, "grad_norm": 0.4780559837818146, "learning_rate": 2.665720343645367e-06, "loss": 0.3627758979797363, "memory(GiB)": 77.59, "step": 104520, "token_acc": 0.9258064516129032, "train_speed(iter/s)": 1.461989 }, { "epoch": 4.47817145794953, "grad_norm": 1.0337575674057007, "learning_rate": 2.6635527157995164e-06, "loss": 0.37789335250854494, "memory(GiB)": 77.59, "step": 104525, "token_acc": 0.9248554913294798, "train_speed(iter/s)": 1.46199 }, { "epoch": 4.47838567327878, "grad_norm": 1.9221382141113281, "learning_rate": 2.6613859454978074e-06, "loss": 0.3250014543533325, "memory(GiB)": 77.59, "step": 104530, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.461993 }, { "epoch": 4.478599888608029, "grad_norm": 2.588247060775757, "learning_rate": 2.6592200327795025e-06, "loss": 0.6629988670349121, "memory(GiB)": 77.59, "step": 104535, "token_acc": 0.8942731277533039, "train_speed(iter/s)": 1.461999 }, { "epoch": 4.478814103937278, "grad_norm": 5.560445308685303, "learning_rate": 2.6570549776838317e-06, "loss": 0.44315176010131835, "memory(GiB)": 77.59, "step": 104540, "token_acc": 0.919093851132686, "train_speed(iter/s)": 1.462003 }, { "epoch": 4.479028319266527, "grad_norm": 3.2393672466278076, "learning_rate": 2.6548907802500244e-06, "loss": 0.19492902755737304, "memory(GiB)": 77.59, "step": 104545, "token_acc": 0.9572368421052632, "train_speed(iter/s)": 1.462006 }, { "epoch": 4.479242534595776, "grad_norm": 2.326303005218506, "learning_rate": 2.6527274405172776e-06, "loss": 0.2674574613571167, "memory(GiB)": 77.59, "step": 104550, "token_acc": 0.9358490566037736, "train_speed(iter/s)": 1.462013 }, { "epoch": 4.479456749925024, "grad_norm": 6.62896203994751, "learning_rate": 2.6505649585247926e-06, "loss": 0.2877995491027832, "memory(GiB)": 77.59, "step": 104555, "token_acc": 0.9377162629757786, "train_speed(iter/s)": 1.462017 }, { "epoch": 4.479670965254273, "grad_norm": 2.1637861728668213, "learning_rate": 2.648403334311744e-06, "loss": 0.20482144355773926, "memory(GiB)": 77.59, "step": 104560, "token_acc": 0.9506172839506173, "train_speed(iter/s)": 1.462018 }, { "epoch": 4.479885180583523, "grad_norm": 4.7797064781188965, "learning_rate": 2.6462425679172898e-06, "loss": 0.27253189086914065, "memory(GiB)": 77.59, "step": 104565, "token_acc": 0.93646408839779, "train_speed(iter/s)": 1.462019 }, { "epoch": 4.480099395912772, "grad_norm": 3.543604850769043, "learning_rate": 2.6440826593805644e-06, "loss": 0.3429630517959595, "memory(GiB)": 77.59, "step": 104570, "token_acc": 0.93359375, "train_speed(iter/s)": 1.462022 }, { "epoch": 4.48031361124202, "grad_norm": 2.4458677768707275, "learning_rate": 2.6419236087407097e-06, "loss": 0.27058844566345214, "memory(GiB)": 77.59, "step": 104575, "token_acc": 0.935064935064935, "train_speed(iter/s)": 1.46202 }, { "epoch": 4.48052782657127, "grad_norm": 1.6775965690612793, "learning_rate": 2.639765416036838e-06, "loss": 0.24362030029296874, "memory(GiB)": 77.59, "step": 104580, "token_acc": 0.9274809160305344, "train_speed(iter/s)": 1.462022 }, { "epoch": 4.480742041900518, "grad_norm": 0.6044834852218628, "learning_rate": 2.6376080813080404e-06, "loss": 0.20301527976989747, "memory(GiB)": 77.59, "step": 104585, "token_acc": 0.9548872180451128, "train_speed(iter/s)": 1.462022 }, { "epoch": 4.480956257229767, "grad_norm": 3.424661874771118, "learning_rate": 2.635451604593403e-06, "loss": 0.4279609203338623, "memory(GiB)": 77.59, "step": 104590, "token_acc": 0.9003322259136213, "train_speed(iter/s)": 1.462024 }, { "epoch": 4.4811704725590165, "grad_norm": 2.3886566162109375, "learning_rate": 2.6332959859319882e-06, "loss": 0.2540069341659546, "memory(GiB)": 77.59, "step": 104595, "token_acc": 0.9303135888501742, "train_speed(iter/s)": 1.462025 }, { "epoch": 4.481384687888266, "grad_norm": 1.6402649879455566, "learning_rate": 2.6311412253628486e-06, "loss": 0.3968193531036377, "memory(GiB)": 77.59, "step": 104600, "token_acc": 0.8974358974358975, "train_speed(iter/s)": 1.462026 }, { "epoch": 4.481598903217514, "grad_norm": 1.6401630640029907, "learning_rate": 2.628987322925025e-06, "loss": 0.3250824689865112, "memory(GiB)": 77.59, "step": 104605, "token_acc": 0.939297124600639, "train_speed(iter/s)": 1.462033 }, { "epoch": 4.481813118546763, "grad_norm": 8.30751895904541, "learning_rate": 2.626834278657536e-06, "loss": 0.2290559768676758, "memory(GiB)": 77.59, "step": 104610, "token_acc": 0.9540983606557377, "train_speed(iter/s)": 1.462037 }, { "epoch": 4.482027333876012, "grad_norm": 3.8581039905548096, "learning_rate": 2.624682092599384e-06, "loss": 0.43761777877807617, "memory(GiB)": 77.59, "step": 104615, "token_acc": 0.9198606271777003, "train_speed(iter/s)": 1.462044 }, { "epoch": 4.482241549205261, "grad_norm": 5.138589859008789, "learning_rate": 2.6225307647895547e-06, "loss": 0.33149290084838867, "memory(GiB)": 77.59, "step": 104620, "token_acc": 0.9283154121863799, "train_speed(iter/s)": 1.462051 }, { "epoch": 4.48245576453451, "grad_norm": 5.267580509185791, "learning_rate": 2.6203802952670163e-06, "loss": 0.5468525886535645, "memory(GiB)": 77.59, "step": 104625, "token_acc": 0.8996282527881041, "train_speed(iter/s)": 1.462053 }, { "epoch": 4.4826699798637595, "grad_norm": 4.4731550216674805, "learning_rate": 2.6182306840707384e-06, "loss": 0.4014463424682617, "memory(GiB)": 77.59, "step": 104630, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.462055 }, { "epoch": 4.482884195193008, "grad_norm": 2.647324562072754, "learning_rate": 2.6160819312396612e-06, "loss": 0.18777674436569214, "memory(GiB)": 77.59, "step": 104635, "token_acc": 0.9496402877697842, "train_speed(iter/s)": 1.462061 }, { "epoch": 4.483098410522257, "grad_norm": 3.4485745429992676, "learning_rate": 2.6139340368127153e-06, "loss": 0.34331235885620115, "memory(GiB)": 77.59, "step": 104640, "token_acc": 0.9010989010989011, "train_speed(iter/s)": 1.462071 }, { "epoch": 4.4833126258515055, "grad_norm": 0.9505252838134766, "learning_rate": 2.6117870008288027e-06, "loss": 0.3303847074508667, "memory(GiB)": 77.59, "step": 104645, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.462082 }, { "epoch": 4.483526841180755, "grad_norm": 4.752738952636719, "learning_rate": 2.6096408233268255e-06, "loss": 0.6066962718963623, "memory(GiB)": 77.59, "step": 104650, "token_acc": 0.8522727272727273, "train_speed(iter/s)": 1.462092 }, { "epoch": 4.483741056510004, "grad_norm": 1.338132619857788, "learning_rate": 2.6074955043456583e-06, "loss": 0.2690101623535156, "memory(GiB)": 77.59, "step": 104655, "token_acc": 0.9310344827586207, "train_speed(iter/s)": 1.462099 }, { "epoch": 4.483955271839253, "grad_norm": 5.078920364379883, "learning_rate": 2.605351043924159e-06, "loss": 0.3210531949996948, "memory(GiB)": 77.59, "step": 104660, "token_acc": 0.9266409266409267, "train_speed(iter/s)": 1.46211 }, { "epoch": 4.484169487168502, "grad_norm": 2.7474937438964844, "learning_rate": 2.6032074421011963e-06, "loss": 0.1937403917312622, "memory(GiB)": 77.59, "step": 104665, "token_acc": 0.9566563467492261, "train_speed(iter/s)": 1.46212 }, { "epoch": 4.484383702497751, "grad_norm": 0.5618628263473511, "learning_rate": 2.6010646989155897e-06, "loss": 0.20769202709197998, "memory(GiB)": 77.59, "step": 104670, "token_acc": 0.9444444444444444, "train_speed(iter/s)": 1.46213 }, { "epoch": 4.484597917826999, "grad_norm": 2.664790391921997, "learning_rate": 2.5989228144061626e-06, "loss": 0.1570068597793579, "memory(GiB)": 77.59, "step": 104675, "token_acc": 0.9627118644067797, "train_speed(iter/s)": 1.462135 }, { "epoch": 4.4848121331562485, "grad_norm": 1.4165313243865967, "learning_rate": 2.5967817886117076e-06, "loss": 0.34678354263305666, "memory(GiB)": 77.59, "step": 104680, "token_acc": 0.9228395061728395, "train_speed(iter/s)": 1.462136 }, { "epoch": 4.485026348485498, "grad_norm": 3.4819347858428955, "learning_rate": 2.5946416215710203e-06, "loss": 0.44358625411987307, "memory(GiB)": 77.59, "step": 104685, "token_acc": 0.9042553191489362, "train_speed(iter/s)": 1.462137 }, { "epoch": 4.485240563814747, "grad_norm": 1.2791210412979126, "learning_rate": 2.59250231332287e-06, "loss": 0.16553033590316774, "memory(GiB)": 77.59, "step": 104690, "token_acc": 0.9642857142857143, "train_speed(iter/s)": 1.462138 }, { "epoch": 4.485454779143995, "grad_norm": 3.5528600215911865, "learning_rate": 2.590363863906009e-06, "loss": 0.7120731353759766, "memory(GiB)": 77.59, "step": 104695, "token_acc": 0.8639240506329114, "train_speed(iter/s)": 1.462142 }, { "epoch": 4.485668994473245, "grad_norm": 2.4374892711639404, "learning_rate": 2.5882262733591843e-06, "loss": 0.37239530086517336, "memory(GiB)": 77.59, "step": 104700, "token_acc": 0.9144736842105263, "train_speed(iter/s)": 1.462149 }, { "epoch": 4.485883209802493, "grad_norm": 4.6604814529418945, "learning_rate": 2.5860895417211204e-06, "loss": 0.44893684387207033, "memory(GiB)": 77.59, "step": 104705, "token_acc": 0.9175627240143369, "train_speed(iter/s)": 1.462156 }, { "epoch": 4.486097425131742, "grad_norm": 1.7574204206466675, "learning_rate": 2.58395366903052e-06, "loss": 0.34003894329071044, "memory(GiB)": 77.59, "step": 104710, "token_acc": 0.9233333333333333, "train_speed(iter/s)": 1.46216 }, { "epoch": 4.4863116404609915, "grad_norm": 1.40464186668396, "learning_rate": 2.581818655326074e-06, "loss": 0.3592484951019287, "memory(GiB)": 77.59, "step": 104715, "token_acc": 0.9203821656050956, "train_speed(iter/s)": 1.462171 }, { "epoch": 4.486525855790241, "grad_norm": 3.664510726928711, "learning_rate": 2.5796845006464676e-06, "loss": 0.30114691257476806, "memory(GiB)": 77.59, "step": 104720, "token_acc": 0.9323308270676691, "train_speed(iter/s)": 1.46217 }, { "epoch": 4.486740071119489, "grad_norm": 1.4582439661026, "learning_rate": 2.5775512050303653e-06, "loss": 0.27582478523254395, "memory(GiB)": 77.59, "step": 104725, "token_acc": 0.9418181818181818, "train_speed(iter/s)": 1.462172 }, { "epoch": 4.486954286448738, "grad_norm": 2.3839848041534424, "learning_rate": 2.575418768516408e-06, "loss": 0.29588608741760253, "memory(GiB)": 77.59, "step": 104730, "token_acc": 0.9388646288209607, "train_speed(iter/s)": 1.462183 }, { "epoch": 4.487168501777987, "grad_norm": 5.160384654998779, "learning_rate": 2.5732871911432254e-06, "loss": 0.5423030376434326, "memory(GiB)": 77.59, "step": 104735, "token_acc": 0.8810408921933085, "train_speed(iter/s)": 1.462185 }, { "epoch": 4.487382717107236, "grad_norm": 2.0651729106903076, "learning_rate": 2.571156472949432e-06, "loss": 0.40363278388977053, "memory(GiB)": 77.59, "step": 104740, "token_acc": 0.9198717948717948, "train_speed(iter/s)": 1.462188 }, { "epoch": 4.487596932436485, "grad_norm": 2.5870184898376465, "learning_rate": 2.56902661397363e-06, "loss": 0.108272385597229, "memory(GiB)": 77.59, "step": 104745, "token_acc": 0.9823943661971831, "train_speed(iter/s)": 1.46219 }, { "epoch": 4.4878111477657345, "grad_norm": 5.766993522644043, "learning_rate": 2.5668976142544097e-06, "loss": 0.17481012344360353, "memory(GiB)": 77.59, "step": 104750, "token_acc": 0.952755905511811, "train_speed(iter/s)": 1.462195 }, { "epoch": 4.488025363094983, "grad_norm": 4.933694362640381, "learning_rate": 2.564769473830331e-06, "loss": 0.5537356376647949, "memory(GiB)": 77.59, "step": 104755, "token_acc": 0.8837209302325582, "train_speed(iter/s)": 1.462208 }, { "epoch": 4.488239578424232, "grad_norm": 2.9799304008483887, "learning_rate": 2.5626421927399557e-06, "loss": 0.7006792545318603, "memory(GiB)": 77.59, "step": 104760, "token_acc": 0.8432203389830508, "train_speed(iter/s)": 1.462218 }, { "epoch": 4.488453793753481, "grad_norm": 5.157886028289795, "learning_rate": 2.5605157710218154e-06, "loss": 0.23203129768371583, "memory(GiB)": 77.59, "step": 104765, "token_acc": 0.939799331103679, "train_speed(iter/s)": 1.462218 }, { "epoch": 4.48866800908273, "grad_norm": 3.867652416229248, "learning_rate": 2.5583902087144287e-06, "loss": 0.3028172492980957, "memory(GiB)": 77.59, "step": 104770, "token_acc": 0.9344262295081968, "train_speed(iter/s)": 1.462221 }, { "epoch": 4.488882224411979, "grad_norm": 1.314069390296936, "learning_rate": 2.556265505856309e-06, "loss": 0.22112681865692138, "memory(GiB)": 77.59, "step": 104775, "token_acc": 0.9562289562289562, "train_speed(iter/s)": 1.462222 }, { "epoch": 4.489096439741228, "grad_norm": 3.144914388656616, "learning_rate": 2.5541416624859483e-06, "loss": 0.39819185733795165, "memory(GiB)": 77.59, "step": 104780, "token_acc": 0.9060150375939849, "train_speed(iter/s)": 1.462225 }, { "epoch": 4.489310655070477, "grad_norm": 1.69915771484375, "learning_rate": 2.552018678641821e-06, "loss": 0.1493169903755188, "memory(GiB)": 77.59, "step": 104785, "token_acc": 0.9534883720930233, "train_speed(iter/s)": 1.462231 }, { "epoch": 4.489524870399726, "grad_norm": 0.7083025574684143, "learning_rate": 2.5498965543623797e-06, "loss": 0.5117555141448975, "memory(GiB)": 77.59, "step": 104790, "token_acc": 0.8782051282051282, "train_speed(iter/s)": 1.462243 }, { "epoch": 4.489739085728975, "grad_norm": 0.6776761412620544, "learning_rate": 2.5477752896860827e-06, "loss": 0.089849191904068, "memory(GiB)": 77.59, "step": 104795, "token_acc": 0.9739776951672863, "train_speed(iter/s)": 1.462244 }, { "epoch": 4.4899533010582235, "grad_norm": 2.829087972640991, "learning_rate": 2.545654884651344e-06, "loss": 0.4179716110229492, "memory(GiB)": 77.59, "step": 104800, "token_acc": 0.9072847682119205, "train_speed(iter/s)": 1.462246 }, { "epoch": 4.490167516387473, "grad_norm": 4.003014087677002, "learning_rate": 2.5435353392965766e-06, "loss": 0.3543220281600952, "memory(GiB)": 77.59, "step": 104805, "token_acc": 0.8951048951048951, "train_speed(iter/s)": 1.462255 }, { "epoch": 4.490381731716722, "grad_norm": 4.6902618408203125, "learning_rate": 2.5414166536601945e-06, "loss": 0.4155571937561035, "memory(GiB)": 77.59, "step": 104810, "token_acc": 0.9159420289855073, "train_speed(iter/s)": 1.462268 }, { "epoch": 4.49059594704597, "grad_norm": 0.23225389420986176, "learning_rate": 2.539298827780567e-06, "loss": 0.4748037815093994, "memory(GiB)": 77.59, "step": 104815, "token_acc": 0.9057239057239057, "train_speed(iter/s)": 1.462268 }, { "epoch": 4.49081016237522, "grad_norm": 3.6793179512023926, "learning_rate": 2.537181861696064e-06, "loss": 0.2890788078308105, "memory(GiB)": 77.59, "step": 104820, "token_acc": 0.9227467811158798, "train_speed(iter/s)": 1.462271 }, { "epoch": 4.491024377704469, "grad_norm": 1.844957947731018, "learning_rate": 2.5350657554450375e-06, "loss": 0.3167715072631836, "memory(GiB)": 77.59, "step": 104825, "token_acc": 0.9365079365079365, "train_speed(iter/s)": 1.462281 }, { "epoch": 4.491238593033717, "grad_norm": 7.125828266143799, "learning_rate": 2.532950509065807e-06, "loss": 0.3856193065643311, "memory(GiB)": 77.59, "step": 104830, "token_acc": 0.926829268292683, "train_speed(iter/s)": 1.462281 }, { "epoch": 4.491452808362967, "grad_norm": 3.4054388999938965, "learning_rate": 2.5308361225967194e-06, "loss": 0.39865286350250245, "memory(GiB)": 77.59, "step": 104835, "token_acc": 0.9122257053291536, "train_speed(iter/s)": 1.462281 }, { "epoch": 4.491667023692216, "grad_norm": 9.562030792236328, "learning_rate": 2.528722596076061e-06, "loss": 0.6527802467346191, "memory(GiB)": 77.59, "step": 104840, "token_acc": 0.8787878787878788, "train_speed(iter/s)": 1.46228 }, { "epoch": 4.491881239021464, "grad_norm": 2.4806160926818848, "learning_rate": 2.5266099295421176e-06, "loss": 0.45832152366638185, "memory(GiB)": 77.59, "step": 104845, "token_acc": 0.9006622516556292, "train_speed(iter/s)": 1.462292 }, { "epoch": 4.4920954543507134, "grad_norm": 3.5654544830322266, "learning_rate": 2.524498123033181e-06, "loss": 0.36927802562713624, "memory(GiB)": 77.59, "step": 104850, "token_acc": 0.9146341463414634, "train_speed(iter/s)": 1.462293 }, { "epoch": 4.492309669679963, "grad_norm": 3.3418381214141846, "learning_rate": 2.522387176587493e-06, "loss": 0.35835421085357666, "memory(GiB)": 77.59, "step": 104855, "token_acc": 0.9258064516129032, "train_speed(iter/s)": 1.462295 }, { "epoch": 4.492523885009211, "grad_norm": 7.528841495513916, "learning_rate": 2.5202770902432894e-06, "loss": 0.4169743537902832, "memory(GiB)": 77.59, "step": 104860, "token_acc": 0.8961937716262975, "train_speed(iter/s)": 1.462307 }, { "epoch": 4.49273810033846, "grad_norm": 3.3427395820617676, "learning_rate": 2.518167864038817e-06, "loss": 0.27575292587280276, "memory(GiB)": 77.59, "step": 104865, "token_acc": 0.9442231075697212, "train_speed(iter/s)": 1.462309 }, { "epoch": 4.49295231566771, "grad_norm": 0.29462310671806335, "learning_rate": 2.5160594980122686e-06, "loss": 0.2844125986099243, "memory(GiB)": 77.59, "step": 104870, "token_acc": 0.9379562043795621, "train_speed(iter/s)": 1.462307 }, { "epoch": 4.493166530996958, "grad_norm": 2.121760845184326, "learning_rate": 2.5139519922018517e-06, "loss": 0.42444615364074706, "memory(GiB)": 77.59, "step": 104875, "token_acc": 0.9104938271604939, "train_speed(iter/s)": 1.462308 }, { "epoch": 4.493380746326207, "grad_norm": 1.497458815574646, "learning_rate": 2.5118453466457413e-06, "loss": 0.5046183109283447, "memory(GiB)": 77.59, "step": 104880, "token_acc": 0.896551724137931, "train_speed(iter/s)": 1.46231 }, { "epoch": 4.4935949616554565, "grad_norm": 1.1618131399154663, "learning_rate": 2.5097395613820963e-06, "loss": 0.28025977611541747, "memory(GiB)": 77.59, "step": 104885, "token_acc": 0.9423076923076923, "train_speed(iter/s)": 1.462312 }, { "epoch": 4.493809176984705, "grad_norm": 5.009873867034912, "learning_rate": 2.507634636449063e-06, "loss": 0.47494802474975584, "memory(GiB)": 77.59, "step": 104890, "token_acc": 0.9063670411985019, "train_speed(iter/s)": 1.462318 }, { "epoch": 4.494023392313954, "grad_norm": 3.132240056991577, "learning_rate": 2.50553057188479e-06, "loss": 0.17101367712020873, "memory(GiB)": 77.59, "step": 104895, "token_acc": 0.9535714285714286, "train_speed(iter/s)": 1.462323 }, { "epoch": 4.494237607643203, "grad_norm": 2.861384391784668, "learning_rate": 2.503427367727379e-06, "loss": 0.38792421817779543, "memory(GiB)": 77.59, "step": 104900, "token_acc": 0.9260273972602739, "train_speed(iter/s)": 1.462326 }, { "epoch": 4.494451822972452, "grad_norm": 1.1836953163146973, "learning_rate": 2.501325024014939e-06, "loss": 0.28874051570892334, "memory(GiB)": 77.59, "step": 104905, "token_acc": 0.9377289377289377, "train_speed(iter/s)": 1.462326 }, { "epoch": 4.494666038301701, "grad_norm": 3.984290599822998, "learning_rate": 2.49922354078555e-06, "loss": 0.28478331565856935, "memory(GiB)": 77.59, "step": 104910, "token_acc": 0.9419795221843004, "train_speed(iter/s)": 1.462324 }, { "epoch": 4.49488025363095, "grad_norm": 7.0936784744262695, "learning_rate": 2.4971229180772826e-06, "loss": 0.3952314376831055, "memory(GiB)": 77.59, "step": 104915, "token_acc": 0.9321533923303835, "train_speed(iter/s)": 1.462324 }, { "epoch": 4.495094468960199, "grad_norm": 2.4497792720794678, "learning_rate": 2.4950231559281944e-06, "loss": 0.3059451341629028, "memory(GiB)": 77.59, "step": 104920, "token_acc": 0.9385113268608414, "train_speed(iter/s)": 1.462332 }, { "epoch": 4.495308684289448, "grad_norm": 2.1571309566497803, "learning_rate": 2.4929242543763333e-06, "loss": 0.3011560201644897, "memory(GiB)": 77.59, "step": 104925, "token_acc": 0.9388379204892966, "train_speed(iter/s)": 1.462334 }, { "epoch": 4.495522899618697, "grad_norm": 6.507726192474365, "learning_rate": 2.4908262134597128e-06, "loss": 0.3277434825897217, "memory(GiB)": 77.59, "step": 104930, "token_acc": 0.9222972972972973, "train_speed(iter/s)": 1.462341 }, { "epoch": 4.4957371149479455, "grad_norm": 2.832367420196533, "learning_rate": 2.4887290332163417e-06, "loss": 0.4149008274078369, "memory(GiB)": 77.59, "step": 104935, "token_acc": 0.9166666666666666, "train_speed(iter/s)": 1.462338 }, { "epoch": 4.495951330277195, "grad_norm": 2.3492918014526367, "learning_rate": 2.486632713684217e-06, "loss": 0.5532340049743653, "memory(GiB)": 77.59, "step": 104940, "token_acc": 0.8688524590163934, "train_speed(iter/s)": 1.462337 }, { "epoch": 4.496165545606444, "grad_norm": 1.3638993501663208, "learning_rate": 2.484537254901309e-06, "loss": 0.23478689193725585, "memory(GiB)": 77.59, "step": 104945, "token_acc": 0.9508771929824561, "train_speed(iter/s)": 1.462346 }, { "epoch": 4.496379760935692, "grad_norm": 3.669424295425415, "learning_rate": 2.482442656905576e-06, "loss": 0.4921121597290039, "memory(GiB)": 77.59, "step": 104950, "token_acc": 0.8919753086419753, "train_speed(iter/s)": 1.462351 }, { "epoch": 4.496593976264942, "grad_norm": 2.1831440925598145, "learning_rate": 2.4803489197349704e-06, "loss": 0.5344000816345215, "memory(GiB)": 77.59, "step": 104955, "token_acc": 0.9037267080745341, "train_speed(iter/s)": 1.462352 }, { "epoch": 4.496808191594191, "grad_norm": 3.1845033168792725, "learning_rate": 2.4782560434274293e-06, "loss": 0.6546217918395996, "memory(GiB)": 77.59, "step": 104960, "token_acc": 0.872852233676976, "train_speed(iter/s)": 1.462358 }, { "epoch": 4.497022406923439, "grad_norm": 4.434353351593018, "learning_rate": 2.4761640280208553e-06, "loss": 0.4660048484802246, "memory(GiB)": 77.59, "step": 104965, "token_acc": 0.8975903614457831, "train_speed(iter/s)": 1.462359 }, { "epoch": 4.4972366222526885, "grad_norm": 4.021943092346191, "learning_rate": 2.4740728735531514e-06, "loss": 0.45668849945068357, "memory(GiB)": 77.59, "step": 104970, "token_acc": 0.8913857677902621, "train_speed(iter/s)": 1.462368 }, { "epoch": 4.497450837581938, "grad_norm": 2.3068621158599854, "learning_rate": 2.471982580062193e-06, "loss": 0.16317453384399414, "memory(GiB)": 77.59, "step": 104975, "token_acc": 0.9608540925266904, "train_speed(iter/s)": 1.462373 }, { "epoch": 4.497665052911186, "grad_norm": 3.5629255771636963, "learning_rate": 2.469893147585861e-06, "loss": 0.21102843284606934, "memory(GiB)": 77.59, "step": 104980, "token_acc": 0.9550173010380623, "train_speed(iter/s)": 1.462373 }, { "epoch": 4.497879268240435, "grad_norm": 7.45986795425415, "learning_rate": 2.4678045761620027e-06, "loss": 0.37987265586853025, "memory(GiB)": 77.59, "step": 104985, "token_acc": 0.9182879377431906, "train_speed(iter/s)": 1.462382 }, { "epoch": 4.498093483569685, "grad_norm": 2.08439040184021, "learning_rate": 2.4657168658284546e-06, "loss": 0.43216261863708494, "memory(GiB)": 77.59, "step": 104990, "token_acc": 0.8961424332344213, "train_speed(iter/s)": 1.462381 }, { "epoch": 4.498307698898933, "grad_norm": 5.191148281097412, "learning_rate": 2.463630016623031e-06, "loss": 0.38461663722991946, "memory(GiB)": 77.59, "step": 104995, "token_acc": 0.9139072847682119, "train_speed(iter/s)": 1.462381 }, { "epoch": 4.498521914228182, "grad_norm": 4.614174842834473, "learning_rate": 2.4615440285835457e-06, "loss": 0.2816993236541748, "memory(GiB)": 77.59, "step": 105000, "token_acc": 0.9393939393939394, "train_speed(iter/s)": 1.462381 }, { "epoch": 4.498521914228182, "eval_loss": 2.202039957046509, "eval_runtime": 11.8644, "eval_samples_per_second": 8.429, "eval_steps_per_second": 8.429, "eval_token_acc": 0.4740834386852086, "step": 105000 } ], "logging_steps": 5, "max_steps": 116705, "num_input_tokens_seen": 0, "num_train_epochs": 5, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 3.4428799879420923e+18, "train_batch_size": 8, "trial_name": null, "trial_params": null }