| { |
| "best_global_step": 7450, |
| "best_metric": 1.0779017210006714, |
| "best_model_checkpoint": "./CanonBert/checkpoint-7450", |
| "epoch": 10.0, |
| "eval_steps": 500, |
| "global_step": 7450, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.001342957864696995, |
| "grad_norm": 31.485458374023438, |
| "learning_rate": 0.0, |
| "loss": 32.58583068847656, |
| "num_input_tokens_seen": 327680, |
| "step": 1, |
| "train_runtime": 8.0777, |
| "train_tokens_per_second": 40566.209 |
| }, |
| { |
| "epoch": 0.06714789323484975, |
| "grad_norm": 29.44062614440918, |
| "learning_rate": 1.0961968680089486e-05, |
| "loss": 32.123365752551024, |
| "num_input_tokens_seen": 16384000, |
| "step": 50, |
| "train_runtime": 382.1108, |
| "train_tokens_per_second": 42877.616 |
| }, |
| { |
| "epoch": 0.1342957864696995, |
| "grad_norm": 30.11578369140625, |
| "learning_rate": 2.2147651006711412e-05, |
| "loss": 31.29671875, |
| "num_input_tokens_seen": 32768000, |
| "step": 100, |
| "train_runtime": 764.7247, |
| "train_tokens_per_second": 42849.407 |
| }, |
| { |
| "epoch": 0.20144367970454927, |
| "grad_norm": 31.309894561767578, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 30.311025390625, |
| "num_input_tokens_seen": 49152000, |
| "step": 150, |
| "train_runtime": 1146.9476, |
| "train_tokens_per_second": 42854.617 |
| }, |
| { |
| "epoch": 0.268591572939399, |
| "grad_norm": 34.5892448425293, |
| "learning_rate": 4.451901565995526e-05, |
| "loss": 29.528154296875, |
| "num_input_tokens_seen": 65536000, |
| "step": 200, |
| "train_runtime": 1528.6613, |
| "train_tokens_per_second": 42871.499 |
| }, |
| { |
| "epoch": 0.33573946617424877, |
| "grad_norm": 32.915374755859375, |
| "learning_rate": 5.570469798657718e-05, |
| "loss": 28.81987060546875, |
| "num_input_tokens_seen": 81920000, |
| "step": 250, |
| "train_runtime": 1910.4025, |
| "train_tokens_per_second": 42881.016 |
| }, |
| { |
| "epoch": 0.40288735940909853, |
| "grad_norm": 33.54853057861328, |
| "learning_rate": 6.689038031319911e-05, |
| "loss": 28.16813232421875, |
| "num_input_tokens_seen": 98304000, |
| "step": 300, |
| "train_runtime": 2292.7173, |
| "train_tokens_per_second": 42876.633 |
| }, |
| { |
| "epoch": 0.4700352526439483, |
| "grad_norm": 32.1806526184082, |
| "learning_rate": 7.807606263982104e-05, |
| "loss": 27.5452734375, |
| "num_input_tokens_seen": 114688000, |
| "step": 350, |
| "train_runtime": 2675.3376, |
| "train_tokens_per_second": 42868.608 |
| }, |
| { |
| "epoch": 0.537183145878798, |
| "grad_norm": 31.719738006591797, |
| "learning_rate": 8.926174496644296e-05, |
| "loss": 27.00053955078125, |
| "num_input_tokens_seen": 131072000, |
| "step": 400, |
| "train_runtime": 3057.9475, |
| "train_tokens_per_second": 42862.738 |
| }, |
| { |
| "epoch": 0.6043310391136478, |
| "grad_norm": 32.894378662109375, |
| "learning_rate": 9.997144081108098e-05, |
| "loss": 26.5186083984375, |
| "num_input_tokens_seen": 147456000, |
| "step": 450, |
| "train_runtime": 3440.035, |
| "train_tokens_per_second": 42864.681 |
| }, |
| { |
| "epoch": 0.6714789323484975, |
| "grad_norm": 31.206104278564453, |
| "learning_rate": 9.925746108810509e-05, |
| "loss": 25.94174072265625, |
| "num_input_tokens_seen": 163840000, |
| "step": 500, |
| "train_runtime": 3822.0237, |
| "train_tokens_per_second": 42867.343 |
| }, |
| { |
| "epoch": 0.6714789323484975, |
| "eval_loss": 1.5885820388793945, |
| "eval_masked_accuracy": 0.680313674457695, |
| "eval_runtime": 18.1032, |
| "eval_samples_per_second": 217.254, |
| "eval_steps_per_second": 3.425, |
| "num_input_tokens_seen": 163840000, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.7386268255833474, |
| "grad_norm": 28.808670043945312, |
| "learning_rate": 9.854348136512923e-05, |
| "loss": 25.40889404296875, |
| "num_input_tokens_seen": 180224000, |
| "step": 550, |
| "train_runtime": 4224.9229, |
| "train_tokens_per_second": 42657.347 |
| }, |
| { |
| "epoch": 0.8057747188181971, |
| "grad_norm": 27.676965713500977, |
| "learning_rate": 9.782950164215337e-05, |
| "loss": 25.1195703125, |
| "num_input_tokens_seen": 196608000, |
| "step": 600, |
| "train_runtime": 4606.8011, |
| "train_tokens_per_second": 42677.771 |
| }, |
| { |
| "epoch": 0.8729226120530469, |
| "grad_norm": 26.387683868408203, |
| "learning_rate": 9.71155219191775e-05, |
| "loss": 24.66832275390625, |
| "num_input_tokens_seen": 212992000, |
| "step": 650, |
| "train_runtime": 4988.9391, |
| "train_tokens_per_second": 42692.844 |
| }, |
| { |
| "epoch": 0.9400705052878966, |
| "grad_norm": 26.794294357299805, |
| "learning_rate": 9.640154219620163e-05, |
| "loss": 24.388662109375, |
| "num_input_tokens_seen": 229376000, |
| "step": 700, |
| "train_runtime": 5371.3616, |
| "train_tokens_per_second": 42703.511 |
| }, |
| { |
| "epoch": 1.006714789323485, |
| "grad_norm": 26.925813674926758, |
| "learning_rate": 9.568756247322577e-05, |
| "loss": 23.8973291015625, |
| "num_input_tokens_seen": 245617664, |
| "step": 750, |
| "train_runtime": 5750.9923, |
| "train_tokens_per_second": 42708.745 |
| }, |
| { |
| "epoch": 1.0738626825583348, |
| "grad_norm": 28.082378387451172, |
| "learning_rate": 9.49735827502499e-05, |
| "loss": 23.6507275390625, |
| "num_input_tokens_seen": 262001664, |
| "step": 800, |
| "train_runtime": 6133.6157, |
| "train_tokens_per_second": 42715.696 |
| }, |
| { |
| "epoch": 1.1410105757931845, |
| "grad_norm": 29.90462303161621, |
| "learning_rate": 9.425960302727404e-05, |
| "loss": 23.39791015625, |
| "num_input_tokens_seen": 278385664, |
| "step": 850, |
| "train_runtime": 6516.0776, |
| "train_tokens_per_second": 42722.889 |
| }, |
| { |
| "epoch": 1.2081584690280343, |
| "grad_norm": 26.34969139099121, |
| "learning_rate": 9.354562330429816e-05, |
| "loss": 23.130556640625, |
| "num_input_tokens_seen": 294769664, |
| "step": 900, |
| "train_runtime": 6898.4036, |
| "train_tokens_per_second": 42730.127 |
| }, |
| { |
| "epoch": 1.275306362262884, |
| "grad_norm": 40.12674331665039, |
| "learning_rate": 9.283164358132229e-05, |
| "loss": 23.02666015625, |
| "num_input_tokens_seen": 311153664, |
| "step": 950, |
| "train_runtime": 7280.3114, |
| "train_tokens_per_second": 42739.06 |
| }, |
| { |
| "epoch": 1.3424542554977337, |
| "grad_norm": 26.551250457763672, |
| "learning_rate": 9.211766385834643e-05, |
| "loss": 22.74733642578125, |
| "num_input_tokens_seen": 327537664, |
| "step": 1000, |
| "train_runtime": 7662.2497, |
| "train_tokens_per_second": 42746.932 |
| }, |
| { |
| "epoch": 1.3424542554977337, |
| "eval_loss": 1.4059845209121704, |
| "eval_masked_accuracy": 0.7082938621956478, |
| "eval_runtime": 18.0953, |
| "eval_samples_per_second": 217.349, |
| "eval_steps_per_second": 3.426, |
| "num_input_tokens_seen": 327537664, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.4096021487325836, |
| "grad_norm": 26.1070556640625, |
| "learning_rate": 9.140368413537057e-05, |
| "loss": 22.5372021484375, |
| "num_input_tokens_seen": 343921664, |
| "step": 1050, |
| "train_runtime": 8065.2674, |
| "train_tokens_per_second": 42642.314 |
| }, |
| { |
| "epoch": 1.4767500419674333, |
| "grad_norm": 26.403989791870117, |
| "learning_rate": 9.068970441239469e-05, |
| "loss": 22.54834228515625, |
| "num_input_tokens_seen": 360305664, |
| "step": 1100, |
| "train_runtime": 8447.5369, |
| "train_tokens_per_second": 42652.157 |
| }, |
| { |
| "epoch": 1.543897935202283, |
| "grad_norm": 27.94965362548828, |
| "learning_rate": 8.997572468941883e-05, |
| "loss": 22.2678564453125, |
| "num_input_tokens_seen": 376689664, |
| "step": 1150, |
| "train_runtime": 8829.7028, |
| "train_tokens_per_second": 42661.647 |
| }, |
| { |
| "epoch": 1.6110458284371327, |
| "grad_norm": 28.19874382019043, |
| "learning_rate": 8.926174496644296e-05, |
| "loss": 22.1563134765625, |
| "num_input_tokens_seen": 393073664, |
| "step": 1200, |
| "train_runtime": 9212.0375, |
| "train_tokens_per_second": 42669.568 |
| }, |
| { |
| "epoch": 1.6781937216719824, |
| "grad_norm": 23.451824188232422, |
| "learning_rate": 8.854776524346708e-05, |
| "loss": 21.9630419921875, |
| "num_input_tokens_seen": 409457664, |
| "step": 1250, |
| "train_runtime": 9594.0382, |
| "train_tokens_per_second": 42678.344 |
| }, |
| { |
| "epoch": 1.7453416149068324, |
| "grad_norm": 29.39751434326172, |
| "learning_rate": 8.783378552049122e-05, |
| "loss": 21.88580322265625, |
| "num_input_tokens_seen": 425841664, |
| "step": 1300, |
| "train_runtime": 9976.9812, |
| "train_tokens_per_second": 42682.416 |
| }, |
| { |
| "epoch": 1.812489508141682, |
| "grad_norm": 24.062496185302734, |
| "learning_rate": 8.711980579751535e-05, |
| "loss": 21.69278564453125, |
| "num_input_tokens_seen": 442225664, |
| "step": 1350, |
| "train_runtime": 10359.0967, |
| "train_tokens_per_second": 42689.597 |
| }, |
| { |
| "epoch": 1.8796374013765318, |
| "grad_norm": 31.466777801513672, |
| "learning_rate": 8.640582607453949e-05, |
| "loss": 21.459140625, |
| "num_input_tokens_seen": 458609664, |
| "step": 1400, |
| "train_runtime": 10740.7818, |
| "train_tokens_per_second": 42697.978 |
| }, |
| { |
| "epoch": 1.9467852946113817, |
| "grad_norm": 26.98602294921875, |
| "learning_rate": 8.569184635156362e-05, |
| "loss": 21.45286376953125, |
| "num_input_tokens_seen": 474993664, |
| "step": 1450, |
| "train_runtime": 11123.0693, |
| "train_tokens_per_second": 42703.471 |
| }, |
| { |
| "epoch": 2.01342957864697, |
| "grad_norm": 28.346364974975586, |
| "learning_rate": 8.497786662858775e-05, |
| "loss": 21.163310546875, |
| "num_input_tokens_seen": 491235328, |
| "step": 1500, |
| "train_runtime": 11502.4827, |
| "train_tokens_per_second": 42706.896 |
| }, |
| { |
| "epoch": 2.01342957864697, |
| "eval_loss": 1.322235345840454, |
| "eval_masked_accuracy": 0.7215658834545166, |
| "eval_runtime": 18.098, |
| "eval_samples_per_second": 217.316, |
| "eval_steps_per_second": 3.426, |
| "num_input_tokens_seen": 491235328, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.0805774718818197, |
| "grad_norm": 24.22477149963379, |
| "learning_rate": 8.426388690561189e-05, |
| "loss": 21.14776123046875, |
| "num_input_tokens_seen": 507619328, |
| "step": 1550, |
| "train_runtime": 11904.9842, |
| "train_tokens_per_second": 42639.227 |
| }, |
| { |
| "epoch": 2.1477253651166697, |
| "grad_norm": 26.27544593811035, |
| "learning_rate": 8.354990718263603e-05, |
| "loss": 20.9605078125, |
| "num_input_tokens_seen": 524003328, |
| "step": 1600, |
| "train_runtime": 12286.9253, |
| "train_tokens_per_second": 42647.23 |
| }, |
| { |
| "epoch": 2.214873258351519, |
| "grad_norm": 28.20328712463379, |
| "learning_rate": 8.283592745966014e-05, |
| "loss": 20.92798095703125, |
| "num_input_tokens_seen": 540387328, |
| "step": 1650, |
| "train_runtime": 12669.3045, |
| "train_tokens_per_second": 42653.275 |
| }, |
| { |
| "epoch": 2.282021151586369, |
| "grad_norm": 23.897136688232422, |
| "learning_rate": 8.212194773668428e-05, |
| "loss": 20.790810546875, |
| "num_input_tokens_seen": 556771328, |
| "step": 1700, |
| "train_runtime": 13051.1941, |
| "train_tokens_per_second": 42660.566 |
| }, |
| { |
| "epoch": 2.3491690448212186, |
| "grad_norm": 27.00596809387207, |
| "learning_rate": 8.140796801370842e-05, |
| "loss": 20.76756591796875, |
| "num_input_tokens_seen": 573155328, |
| "step": 1750, |
| "train_runtime": 13433.2554, |
| "train_tokens_per_second": 42666.897 |
| }, |
| { |
| "epoch": 2.4163169380560685, |
| "grad_norm": 34.40134811401367, |
| "learning_rate": 8.069398829073254e-05, |
| "loss": 20.6015283203125, |
| "num_input_tokens_seen": 589539328, |
| "step": 1800, |
| "train_runtime": 13814.9757, |
| "train_tokens_per_second": 42673.932 |
| }, |
| { |
| "epoch": 2.4834648312909184, |
| "grad_norm": 31.715700149536133, |
| "learning_rate": 7.998000856775668e-05, |
| "loss": 20.6743505859375, |
| "num_input_tokens_seen": 605923328, |
| "step": 1850, |
| "train_runtime": 14197.1991, |
| "train_tokens_per_second": 42679.075 |
| }, |
| { |
| "epoch": 2.550612724525768, |
| "grad_norm": 23.027774810791016, |
| "learning_rate": 7.926602884478082e-05, |
| "loss": 20.44476318359375, |
| "num_input_tokens_seen": 622307328, |
| "step": 1900, |
| "train_runtime": 14579.226, |
| "train_tokens_per_second": 42684.525 |
| }, |
| { |
| "epoch": 2.617760617760618, |
| "grad_norm": 30.85573959350586, |
| "learning_rate": 7.855204912180495e-05, |
| "loss": 20.362332763671876, |
| "num_input_tokens_seen": 638691328, |
| "step": 1950, |
| "train_runtime": 14961.5926, |
| "train_tokens_per_second": 42688.726 |
| }, |
| { |
| "epoch": 2.6849085109954673, |
| "grad_norm": 24.106107711791992, |
| "learning_rate": 7.783806939882907e-05, |
| "loss": 20.340782470703125, |
| "num_input_tokens_seen": 655075328, |
| "step": 2000, |
| "train_runtime": 15343.7785, |
| "train_tokens_per_second": 42693.221 |
| }, |
| { |
| "epoch": 2.6849085109954673, |
| "eval_loss": 1.2741047143936157, |
| "eval_masked_accuracy": 0.7301424206751543, |
| "eval_runtime": 18.0943, |
| "eval_samples_per_second": 217.361, |
| "eval_steps_per_second": 3.426, |
| "num_input_tokens_seen": 655075328, |
| "step": 2000 |
| }, |
| { |
| "epoch": 2.7520564042303173, |
| "grad_norm": 20.974048614501953, |
| "learning_rate": 7.712408967585321e-05, |
| "loss": 20.1228564453125, |
| "num_input_tokens_seen": 671459328, |
| "step": 2050, |
| "train_runtime": 15745.9366, |
| "train_tokens_per_second": 42643.34 |
| }, |
| { |
| "epoch": 2.819204297465167, |
| "grad_norm": 30.808757781982422, |
| "learning_rate": 7.641010995287734e-05, |
| "loss": 20.217064208984375, |
| "num_input_tokens_seen": 687843328, |
| "step": 2100, |
| "train_runtime": 16128.0983, |
| "train_tokens_per_second": 42648.756 |
| }, |
| { |
| "epoch": 2.8863521907000167, |
| "grad_norm": 23.832176208496094, |
| "learning_rate": 7.569613022990148e-05, |
| "loss": 20.1458984375, |
| "num_input_tokens_seen": 704227328, |
| "step": 2150, |
| "train_runtime": 16510.4238, |
| "train_tokens_per_second": 42653.498 |
| }, |
| { |
| "epoch": 2.9535000839348666, |
| "grad_norm": 88.81837463378906, |
| "learning_rate": 7.49821505069256e-05, |
| "loss": 19.993797607421875, |
| "num_input_tokens_seen": 720611328, |
| "step": 2200, |
| "train_runtime": 16893.09, |
| "train_tokens_per_second": 42657.165 |
| }, |
| { |
| "epoch": 3.0201443679704547, |
| "grad_norm": 25.890085220336914, |
| "learning_rate": 7.426817078394974e-05, |
| "loss": 19.76076171875, |
| "num_input_tokens_seen": 736852992, |
| "step": 2250, |
| "train_runtime": 17272.5849, |
| "train_tokens_per_second": 42660.262 |
| }, |
| { |
| "epoch": 3.0872922612053046, |
| "grad_norm": 21.964996337890625, |
| "learning_rate": 7.355419106097388e-05, |
| "loss": 19.6687451171875, |
| "num_input_tokens_seen": 753236992, |
| "step": 2300, |
| "train_runtime": 17654.8634, |
| "train_tokens_per_second": 42664.561 |
| }, |
| { |
| "epoch": 3.1544401544401546, |
| "grad_norm": 25.932247161865234, |
| "learning_rate": 7.2840211337998e-05, |
| "loss": 19.772432861328124, |
| "num_input_tokens_seen": 769620992, |
| "step": 2350, |
| "train_runtime": 18036.9121, |
| "train_tokens_per_second": 42669.221 |
| }, |
| { |
| "epoch": 3.221588047675004, |
| "grad_norm": 21.372344970703125, |
| "learning_rate": 7.212623161502213e-05, |
| "loss": 19.579937744140626, |
| "num_input_tokens_seen": 786004992, |
| "step": 2400, |
| "train_runtime": 18418.9135, |
| "train_tokens_per_second": 42673.798 |
| }, |
| { |
| "epoch": 3.288735940909854, |
| "grad_norm": 41.207489013671875, |
| "learning_rate": 7.141225189204627e-05, |
| "loss": 19.558551025390624, |
| "num_input_tokens_seen": 802388992, |
| "step": 2450, |
| "train_runtime": 18801.1902, |
| "train_tokens_per_second": 42677.564 |
| }, |
| { |
| "epoch": 3.355883834144704, |
| "grad_norm": 20.699298858642578, |
| "learning_rate": 7.06982721690704e-05, |
| "loss": 19.5032666015625, |
| "num_input_tokens_seen": 818772992, |
| "step": 2500, |
| "train_runtime": 19183.4239, |
| "train_tokens_per_second": 42681.275 |
| }, |
| { |
| "epoch": 3.355883834144704, |
| "eval_loss": 1.23164963722229, |
| "eval_masked_accuracy": 0.7365415684330173, |
| "eval_runtime": 18.1012, |
| "eval_samples_per_second": 217.278, |
| "eval_steps_per_second": 3.425, |
| "num_input_tokens_seen": 818772992, |
| "step": 2500 |
| }, |
| { |
| "epoch": 3.4230317273795534, |
| "grad_norm": 35.69466781616211, |
| "learning_rate": 6.998429244609453e-05, |
| "loss": 19.529766845703126, |
| "num_input_tokens_seen": 835156992, |
| "step": 2550, |
| "train_runtime": 19586.4251, |
| "train_tokens_per_second": 42639.583 |
| }, |
| { |
| "epoch": 3.4901796206144033, |
| "grad_norm": 22.10000991821289, |
| "learning_rate": 6.927031272311867e-05, |
| "loss": 19.36791015625, |
| "num_input_tokens_seen": 851540992, |
| "step": 2600, |
| "train_runtime": 19967.8987, |
| "train_tokens_per_second": 42645.498 |
| }, |
| { |
| "epoch": 3.557327513849253, |
| "grad_norm": 22.041030883789062, |
| "learning_rate": 6.85563330001428e-05, |
| "loss": 19.318353271484376, |
| "num_input_tokens_seen": 867924992, |
| "step": 2650, |
| "train_runtime": 20350.1026, |
| "train_tokens_per_second": 42649.662 |
| }, |
| { |
| "epoch": 3.6244754070841028, |
| "grad_norm": 23.8006591796875, |
| "learning_rate": 6.784235327716694e-05, |
| "loss": 19.310184326171875, |
| "num_input_tokens_seen": 884308992, |
| "step": 2700, |
| "train_runtime": 20732.93, |
| "train_tokens_per_second": 42652.389 |
| }, |
| { |
| "epoch": 3.6916233003189527, |
| "grad_norm": 21.48434066772461, |
| "learning_rate": 6.712837355419106e-05, |
| "loss": 19.189322509765624, |
| "num_input_tokens_seen": 900692992, |
| "step": 2750, |
| "train_runtime": 21115.2157, |
| "train_tokens_per_second": 42656.111 |
| }, |
| { |
| "epoch": 3.758771193553802, |
| "grad_norm": 27.336259841918945, |
| "learning_rate": 6.641439383121519e-05, |
| "loss": 19.158172607421875, |
| "num_input_tokens_seen": 917076992, |
| "step": 2800, |
| "train_runtime": 21497.1386, |
| "train_tokens_per_second": 42660.421 |
| }, |
| { |
| "epoch": 3.825919086788652, |
| "grad_norm": 20.084529876708984, |
| "learning_rate": 6.570041410823933e-05, |
| "loss": 19.0932373046875, |
| "num_input_tokens_seen": 933460992, |
| "step": 2850, |
| "train_runtime": 21879.1616, |
| "train_tokens_per_second": 42664.386 |
| }, |
| { |
| "epoch": 3.8930669800235016, |
| "grad_norm": 23.61543846130371, |
| "learning_rate": 6.498643438526347e-05, |
| "loss": 19.03164794921875, |
| "num_input_tokens_seen": 949844992, |
| "step": 2900, |
| "train_runtime": 22261.824, |
| "train_tokens_per_second": 42666.989 |
| }, |
| { |
| "epoch": 3.9602148732583515, |
| "grad_norm": 24.03072738647461, |
| "learning_rate": 6.427245466228759e-05, |
| "loss": 19.039158935546876, |
| "num_input_tokens_seen": 966228992, |
| "step": 2950, |
| "train_runtime": 22643.7405, |
| "train_tokens_per_second": 42670.909 |
| }, |
| { |
| "epoch": 4.02685915729394, |
| "grad_norm": 19.51736068725586, |
| "learning_rate": 6.355847493931173e-05, |
| "loss": 18.874622802734375, |
| "num_input_tokens_seen": 982470656, |
| "step": 3000, |
| "train_runtime": 23023.9109, |
| "train_tokens_per_second": 42671.754 |
| }, |
| { |
| "epoch": 4.02685915729394, |
| "eval_loss": 1.1917675733566284, |
| "eval_masked_accuracy": 0.7423570262408483, |
| "eval_runtime": 18.1046, |
| "eval_samples_per_second": 217.238, |
| "eval_steps_per_second": 3.425, |
| "num_input_tokens_seen": 982470656, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.0940070505287895, |
| "grad_norm": 22.008718490600586, |
| "learning_rate": 6.284449521633586e-05, |
| "loss": 18.817777099609376, |
| "num_input_tokens_seen": 998854656, |
| "step": 3050, |
| "train_runtime": 23426.7943, |
| "train_tokens_per_second": 42637.274 |
| }, |
| { |
| "epoch": 4.1611549437636395, |
| "grad_norm": 25.215585708618164, |
| "learning_rate": 6.213051549335998e-05, |
| "loss": 18.789967041015625, |
| "num_input_tokens_seen": 1015238656, |
| "step": 3100, |
| "train_runtime": 23809.1884, |
| "train_tokens_per_second": 42640.624 |
| }, |
| { |
| "epoch": 4.228302836998489, |
| "grad_norm": 24.252426147460938, |
| "learning_rate": 6.141653577038412e-05, |
| "loss": 18.7561767578125, |
| "num_input_tokens_seen": 1031622656, |
| "step": 3150, |
| "train_runtime": 24191.755, |
| "train_tokens_per_second": 42643.564 |
| }, |
| { |
| "epoch": 4.295450730233339, |
| "grad_norm": 21.702259063720703, |
| "learning_rate": 6.0702556047408255e-05, |
| "loss": 18.75292724609375, |
| "num_input_tokens_seen": 1048006656, |
| "step": 3200, |
| "train_runtime": 24573.2637, |
| "train_tokens_per_second": 42648.248 |
| }, |
| { |
| "epoch": 4.362598623468188, |
| "grad_norm": 95.9570541381836, |
| "learning_rate": 5.998857632443239e-05, |
| "loss": 18.597052001953124, |
| "num_input_tokens_seen": 1064390656, |
| "step": 3250, |
| "train_runtime": 24955.5038, |
| "train_tokens_per_second": 42651.54 |
| }, |
| { |
| "epoch": 4.429746516703038, |
| "grad_norm": 20.15200424194336, |
| "learning_rate": 5.927459660145652e-05, |
| "loss": 18.606531982421874, |
| "num_input_tokens_seen": 1080774656, |
| "step": 3300, |
| "train_runtime": 25337.6615, |
| "train_tokens_per_second": 42654.87 |
| }, |
| { |
| "epoch": 4.496894409937888, |
| "grad_norm": 20.575227737426758, |
| "learning_rate": 5.856061687848066e-05, |
| "loss": 18.473587646484376, |
| "num_input_tokens_seen": 1097158656, |
| "step": 3350, |
| "train_runtime": 25719.2745, |
| "train_tokens_per_second": 42659.005 |
| }, |
| { |
| "epoch": 4.564042303172738, |
| "grad_norm": 20.6302433013916, |
| "learning_rate": 5.784663715550479e-05, |
| "loss": 18.556181640625, |
| "num_input_tokens_seen": 1113542656, |
| "step": 3400, |
| "train_runtime": 26100.9704, |
| "train_tokens_per_second": 42662.883 |
| }, |
| { |
| "epoch": 4.631190196407587, |
| "grad_norm": 21.177248001098633, |
| "learning_rate": 5.713265743252892e-05, |
| "loss": 18.51771484375, |
| "num_input_tokens_seen": 1129926656, |
| "step": 3450, |
| "train_runtime": 26483.1367, |
| "train_tokens_per_second": 42665.892 |
| }, |
| { |
| "epoch": 4.698338089642437, |
| "grad_norm": 20.200031280517578, |
| "learning_rate": 5.641867770955305e-05, |
| "loss": 18.4009619140625, |
| "num_input_tokens_seen": 1146310656, |
| "step": 3500, |
| "train_runtime": 26865.5394, |
| "train_tokens_per_second": 42668.44 |
| }, |
| { |
| "epoch": 4.698338089642437, |
| "eval_loss": 1.1745275259017944, |
| "eval_masked_accuracy": 0.7461218274111675, |
| "eval_runtime": 18.093, |
| "eval_samples_per_second": 217.376, |
| "eval_steps_per_second": 3.427, |
| "num_input_tokens_seen": 1146310656, |
| "step": 3500 |
| }, |
| { |
| "epoch": 4.765485982877287, |
| "grad_norm": 22.780622482299805, |
| "learning_rate": 5.570469798657718e-05, |
| "loss": 18.44864013671875, |
| "num_input_tokens_seen": 1162694656, |
| "step": 3550, |
| "train_runtime": 27267.9686, |
| "train_tokens_per_second": 42639.577 |
| }, |
| { |
| "epoch": 4.832633876112137, |
| "grad_norm": 22.299123764038086, |
| "learning_rate": 5.499071826360131e-05, |
| "loss": 18.365985107421874, |
| "num_input_tokens_seen": 1179078656, |
| "step": 3600, |
| "train_runtime": 27649.4052, |
| "train_tokens_per_second": 42643.907 |
| }, |
| { |
| "epoch": 4.899781769346987, |
| "grad_norm": 20.132230758666992, |
| "learning_rate": 5.427673854062545e-05, |
| "loss": 18.296614990234374, |
| "num_input_tokens_seen": 1195462656, |
| "step": 3650, |
| "train_runtime": 28031.8042, |
| "train_tokens_per_second": 42646.654 |
| }, |
| { |
| "epoch": 4.966929662581837, |
| "grad_norm": 22.012845993041992, |
| "learning_rate": 5.3562758817649584e-05, |
| "loss": 18.1852392578125, |
| "num_input_tokens_seen": 1211846656, |
| "step": 3700, |
| "train_runtime": 28413.9026, |
| "train_tokens_per_second": 42649.779 |
| }, |
| { |
| "epoch": 5.033573946617425, |
| "grad_norm": 19.76921844482422, |
| "learning_rate": 5.284877909467372e-05, |
| "loss": 17.99932373046875, |
| "num_input_tokens_seen": 1228088320, |
| "step": 3750, |
| "train_runtime": 28793.0555, |
| "train_tokens_per_second": 42652.24 |
| }, |
| { |
| "epoch": 5.100721839852275, |
| "grad_norm": 19.71321678161621, |
| "learning_rate": 5.213479937169785e-05, |
| "loss": 18.186424560546875, |
| "num_input_tokens_seen": 1244472320, |
| "step": 3800, |
| "train_runtime": 29175.5764, |
| "train_tokens_per_second": 42654.592 |
| }, |
| { |
| "epoch": 5.167869733087125, |
| "grad_norm": 19.083816528320312, |
| "learning_rate": 5.1420819648721974e-05, |
| "loss": 18.081846923828124, |
| "num_input_tokens_seen": 1260856320, |
| "step": 3850, |
| "train_runtime": 29557.9804, |
| "train_tokens_per_second": 42657.052 |
| }, |
| { |
| "epoch": 5.235017626321974, |
| "grad_norm": 21.77515983581543, |
| "learning_rate": 5.070683992574611e-05, |
| "loss": 18.0148828125, |
| "num_input_tokens_seen": 1277240320, |
| "step": 3900, |
| "train_runtime": 29940.6282, |
| "train_tokens_per_second": 42659.102 |
| }, |
| { |
| "epoch": 5.302165519556824, |
| "grad_norm": 25.122718811035156, |
| "learning_rate": 4.999286020277024e-05, |
| "loss": 17.95366943359375, |
| "num_input_tokens_seen": 1293624320, |
| "step": 3950, |
| "train_runtime": 30322.7025, |
| "train_tokens_per_second": 42661.907 |
| }, |
| { |
| "epoch": 5.369313412791674, |
| "grad_norm": 20.051774978637695, |
| "learning_rate": 4.927888047979438e-05, |
| "loss": 17.99229736328125, |
| "num_input_tokens_seen": 1310008320, |
| "step": 4000, |
| "train_runtime": 30705.0003, |
| "train_tokens_per_second": 42664.332 |
| }, |
| { |
| "epoch": 5.369313412791674, |
| "eval_loss": 1.1476529836654663, |
| "eval_masked_accuracy": 0.7508973219435078, |
| "eval_runtime": 18.0946, |
| "eval_samples_per_second": 217.358, |
| "eval_steps_per_second": 3.426, |
| "num_input_tokens_seen": 1310008320, |
| "step": 4000 |
| }, |
| { |
| "epoch": 5.436461306026524, |
| "grad_norm": 22.737159729003906, |
| "learning_rate": 4.856490075681851e-05, |
| "loss": 17.99468017578125, |
| "num_input_tokens_seen": 1326392320, |
| "step": 4050, |
| "train_runtime": 31107.2176, |
| "train_tokens_per_second": 42639.375 |
| }, |
| { |
| "epoch": 5.503609199261373, |
| "grad_norm": 19.653364181518555, |
| "learning_rate": 4.785092103384264e-05, |
| "loss": 17.9478125, |
| "num_input_tokens_seen": 1342776320, |
| "step": 4100, |
| "train_runtime": 31489.8056, |
| "train_tokens_per_second": 42641.62 |
| }, |
| { |
| "epoch": 5.570757092496223, |
| "grad_norm": 21.537843704223633, |
| "learning_rate": 4.7136941310866775e-05, |
| "loss": 17.9237060546875, |
| "num_input_tokens_seen": 1359160320, |
| "step": 4150, |
| "train_runtime": 31871.5203, |
| "train_tokens_per_second": 42644.979 |
| }, |
| { |
| "epoch": 5.637904985731073, |
| "grad_norm": 23.525545120239258, |
| "learning_rate": 4.642296158789091e-05, |
| "loss": 17.881396484375, |
| "num_input_tokens_seen": 1375544320, |
| "step": 4200, |
| "train_runtime": 32253.1239, |
| "train_tokens_per_second": 42648.406 |
| }, |
| { |
| "epoch": 5.7050528789659225, |
| "grad_norm": 27.282806396484375, |
| "learning_rate": 4.570898186491504e-05, |
| "loss": 17.892135009765624, |
| "num_input_tokens_seen": 1391928320, |
| "step": 4250, |
| "train_runtime": 32635.0417, |
| "train_tokens_per_second": 42651.342 |
| }, |
| { |
| "epoch": 5.772200772200772, |
| "grad_norm": 20.510400772094727, |
| "learning_rate": 4.499500214193917e-05, |
| "loss": 17.907117919921873, |
| "num_input_tokens_seen": 1408312320, |
| "step": 4300, |
| "train_runtime": 33017.8689, |
| "train_tokens_per_second": 42653.035 |
| }, |
| { |
| "epoch": 5.839348665435622, |
| "grad_norm": 19.230182647705078, |
| "learning_rate": 4.4281022418963304e-05, |
| "loss": 17.78746826171875, |
| "num_input_tokens_seen": 1424696320, |
| "step": 4350, |
| "train_runtime": 33400.5094, |
| "train_tokens_per_second": 42654.928 |
| }, |
| { |
| "epoch": 5.906496558670471, |
| "grad_norm": 24.074186325073242, |
| "learning_rate": 4.3567042695987436e-05, |
| "loss": 17.6533203125, |
| "num_input_tokens_seen": 1441080320, |
| "step": 4400, |
| "train_runtime": 33783.0047, |
| "train_tokens_per_second": 42656.961 |
| }, |
| { |
| "epoch": 5.973644451905321, |
| "grad_norm": 19.16803741455078, |
| "learning_rate": 4.285306297301157e-05, |
| "loss": 17.756685791015624, |
| "num_input_tokens_seen": 1457464320, |
| "step": 4450, |
| "train_runtime": 34165.0509, |
| "train_tokens_per_second": 42659.51 |
| }, |
| { |
| "epoch": 6.040288735940909, |
| "grad_norm": 20.361373901367188, |
| "learning_rate": 4.21390832500357e-05, |
| "loss": 17.543341064453124, |
| "num_input_tokens_seen": 1473705984, |
| "step": 4500, |
| "train_runtime": 34544.159, |
| "train_tokens_per_second": 42661.51 |
| }, |
| { |
| "epoch": 6.040288735940909, |
| "eval_loss": 1.1311087608337402, |
| "eval_masked_accuracy": 0.7534060731038085, |
| "eval_runtime": 18.1085, |
| "eval_samples_per_second": 217.191, |
| "eval_steps_per_second": 3.424, |
| "num_input_tokens_seen": 1473705984, |
| "step": 4500 |
| }, |
| { |
| "epoch": 6.107436629175759, |
| "grad_norm": 22.37014389038086, |
| "learning_rate": 4.142510352705983e-05, |
| "loss": 17.574459228515625, |
| "num_input_tokens_seen": 1490089984, |
| "step": 4550, |
| "train_runtime": 34947.0679, |
| "train_tokens_per_second": 42638.484 |
| }, |
| { |
| "epoch": 6.174584522410609, |
| "grad_norm": 40.5280647277832, |
| "learning_rate": 4.0711123804083966e-05, |
| "loss": 17.67244140625, |
| "num_input_tokens_seen": 1506473984, |
| "step": 4600, |
| "train_runtime": 35329.4518, |
| "train_tokens_per_second": 42640.74 |
| }, |
| { |
| "epoch": 6.241732415645459, |
| "grad_norm": 20.747886657714844, |
| "learning_rate": 3.99971440811081e-05, |
| "loss": 17.596712646484374, |
| "num_input_tokens_seen": 1522857984, |
| "step": 4650, |
| "train_runtime": 35711.6986, |
| "train_tokens_per_second": 42643.113 |
| }, |
| { |
| "epoch": 6.308880308880309, |
| "grad_norm": 24.38255500793457, |
| "learning_rate": 3.928316435813223e-05, |
| "loss": 17.52099853515625, |
| "num_input_tokens_seen": 1539241984, |
| "step": 4700, |
| "train_runtime": 36093.7209, |
| "train_tokens_per_second": 42645.7 |
| }, |
| { |
| "epoch": 6.376028202115158, |
| "grad_norm": 19.407743453979492, |
| "learning_rate": 3.856918463515636e-05, |
| "loss": 17.53602783203125, |
| "num_input_tokens_seen": 1555625984, |
| "step": 4750, |
| "train_runtime": 36475.9888, |
| "train_tokens_per_second": 42647.946 |
| }, |
| { |
| "epoch": 6.443176095350008, |
| "grad_norm": 20.32953453063965, |
| "learning_rate": 3.7855204912180495e-05, |
| "loss": 17.43239990234375, |
| "num_input_tokens_seen": 1572009984, |
| "step": 4800, |
| "train_runtime": 36858.3509, |
| "train_tokens_per_second": 42650.036 |
| }, |
| { |
| "epoch": 6.510323988584858, |
| "grad_norm": 19.538949966430664, |
| "learning_rate": 3.714122518920463e-05, |
| "loss": 17.46402099609375, |
| "num_input_tokens_seen": 1588393984, |
| "step": 4850, |
| "train_runtime": 37240.3757, |
| "train_tokens_per_second": 42652.469 |
| }, |
| { |
| "epoch": 6.577471881819708, |
| "grad_norm": 27.391788482666016, |
| "learning_rate": 3.642724546622876e-05, |
| "loss": 17.46167724609375, |
| "num_input_tokens_seen": 1604777984, |
| "step": 4900, |
| "train_runtime": 37622.395, |
| "train_tokens_per_second": 42654.86 |
| }, |
| { |
| "epoch": 6.644619775054558, |
| "grad_norm": 24.001676559448242, |
| "learning_rate": 3.571326574325289e-05, |
| "loss": 17.38479736328125, |
| "num_input_tokens_seen": 1621161984, |
| "step": 4950, |
| "train_runtime": 38004.335, |
| "train_tokens_per_second": 42657.291 |
| }, |
| { |
| "epoch": 6.711767668289408, |
| "grad_norm": 22.344423294067383, |
| "learning_rate": 3.499928602027703e-05, |
| "loss": 17.3615673828125, |
| "num_input_tokens_seen": 1637545984, |
| "step": 5000, |
| "train_runtime": 38386.7223, |
| "train_tokens_per_second": 42659.177 |
| }, |
| { |
| "epoch": 6.711767668289408, |
| "eval_loss": 1.1254918575286865, |
| "eval_masked_accuracy": 0.7547708276205275, |
| "eval_runtime": 18.0962, |
| "eval_samples_per_second": 217.338, |
| "eval_steps_per_second": 3.426, |
| "num_input_tokens_seen": 1637545984, |
| "step": 5000 |
| }, |
| { |
| "epoch": 6.778915561524257, |
| "grad_norm": 20.349462509155273, |
| "learning_rate": 3.4285306297301156e-05, |
| "loss": 17.385616455078125, |
| "num_input_tokens_seen": 1653929984, |
| "step": 5050, |
| "train_runtime": 38789.5078, |
| "train_tokens_per_second": 42638.592 |
| }, |
| { |
| "epoch": 6.846063454759107, |
| "grad_norm": 19.716110229492188, |
| "learning_rate": 3.357132657432529e-05, |
| "loss": 17.39263671875, |
| "num_input_tokens_seen": 1670313984, |
| "step": 5100, |
| "train_runtime": 39171.8181, |
| "train_tokens_per_second": 42640.706 |
| }, |
| { |
| "epoch": 6.913211347993957, |
| "grad_norm": 23.193069458007812, |
| "learning_rate": 3.285734685134942e-05, |
| "loss": 17.322354736328126, |
| "num_input_tokens_seen": 1686697984, |
| "step": 5150, |
| "train_runtime": 39553.757, |
| "train_tokens_per_second": 42643.18 |
| }, |
| { |
| "epoch": 6.980359241228807, |
| "grad_norm": 19.377727508544922, |
| "learning_rate": 3.214336712837356e-05, |
| "loss": 17.24675048828125, |
| "num_input_tokens_seen": 1703081984, |
| "step": 5200, |
| "train_runtime": 39936.046, |
| "train_tokens_per_second": 42645.233 |
| }, |
| { |
| "epoch": 7.047003525264395, |
| "grad_norm": 19.60344696044922, |
| "learning_rate": 3.1429387405397685e-05, |
| "loss": 17.098548583984375, |
| "num_input_tokens_seen": 1719323648, |
| "step": 5250, |
| "train_runtime": 40315.6311, |
| "train_tokens_per_second": 42646.577 |
| }, |
| { |
| "epoch": 7.114151418499245, |
| "grad_norm": 22.405685424804688, |
| "learning_rate": 3.071540768242182e-05, |
| "loss": 17.122021484375, |
| "num_input_tokens_seen": 1735707648, |
| "step": 5300, |
| "train_runtime": 40697.747, |
| "train_tokens_per_second": 42648.74 |
| }, |
| { |
| "epoch": 7.181299311734095, |
| "grad_norm": 19.352375030517578, |
| "learning_rate": 3.0001427959445953e-05, |
| "loss": 17.219647216796876, |
| "num_input_tokens_seen": 1752091648, |
| "step": 5350, |
| "train_runtime": 41080.3111, |
| "train_tokens_per_second": 42650.399 |
| }, |
| { |
| "epoch": 7.248447204968944, |
| "grad_norm": 19.293392181396484, |
| "learning_rate": 2.9287448236470082e-05, |
| "loss": 17.0911181640625, |
| "num_input_tokens_seen": 1768475648, |
| "step": 5400, |
| "train_runtime": 41463.0281, |
| "train_tokens_per_second": 42651.869 |
| }, |
| { |
| "epoch": 7.315595098203794, |
| "grad_norm": 20.061555862426758, |
| "learning_rate": 2.8573468513494218e-05, |
| "loss": 17.16145751953125, |
| "num_input_tokens_seen": 1784859648, |
| "step": 5450, |
| "train_runtime": 41845.4451, |
| "train_tokens_per_second": 42653.618 |
| }, |
| { |
| "epoch": 7.3827429914386435, |
| "grad_norm": 19.52706527709961, |
| "learning_rate": 2.785948879051835e-05, |
| "loss": 17.160897216796876, |
| "num_input_tokens_seen": 1801243648, |
| "step": 5500, |
| "train_runtime": 42228.0537, |
| "train_tokens_per_second": 42655.143 |
| }, |
| { |
| "epoch": 7.3827429914386435, |
| "eval_loss": 1.1056840419769287, |
| "eval_masked_accuracy": 0.7587028345843005, |
| "eval_runtime": 18.1241, |
| "eval_samples_per_second": 217.004, |
| "eval_steps_per_second": 3.421, |
| "num_input_tokens_seen": 1801243648, |
| "step": 5500 |
| }, |
| { |
| "epoch": 7.4498908846734935, |
| "grad_norm": 18.919979095458984, |
| "learning_rate": 2.7145509067542486e-05, |
| "loss": 17.10321044921875, |
| "num_input_tokens_seen": 1817627648, |
| "step": 5550, |
| "train_runtime": 42630.7742, |
| "train_tokens_per_second": 42636.515 |
| }, |
| { |
| "epoch": 7.517038777908343, |
| "grad_norm": 19.892620086669922, |
| "learning_rate": 2.6431529344566615e-05, |
| "loss": 17.095526123046874, |
| "num_input_tokens_seen": 1834011648, |
| "step": 5600, |
| "train_runtime": 43013.4145, |
| "train_tokens_per_second": 42638.132 |
| }, |
| { |
| "epoch": 7.584186671143193, |
| "grad_norm": 20.655860900878906, |
| "learning_rate": 2.5717549621590747e-05, |
| "loss": 17.05685302734375, |
| "num_input_tokens_seen": 1850395648, |
| "step": 5650, |
| "train_runtime": 43395.7235, |
| "train_tokens_per_second": 42640.046 |
| }, |
| { |
| "epoch": 7.651334564378042, |
| "grad_norm": 19.32682228088379, |
| "learning_rate": 2.5003569898614883e-05, |
| "loss": 16.990732421875, |
| "num_input_tokens_seen": 1866779648, |
| "step": 5700, |
| "train_runtime": 43777.9257, |
| "train_tokens_per_second": 42642.031 |
| }, |
| { |
| "epoch": 7.718482457612892, |
| "grad_norm": 20.431968688964844, |
| "learning_rate": 2.4289590175639012e-05, |
| "loss": 17.013017578125, |
| "num_input_tokens_seen": 1883163648, |
| "step": 5750, |
| "train_runtime": 44160.1064, |
| "train_tokens_per_second": 42644.002 |
| }, |
| { |
| "epoch": 7.785630350847742, |
| "grad_norm": 24.700050354003906, |
| "learning_rate": 2.3575610452663147e-05, |
| "loss": 16.923485107421875, |
| "num_input_tokens_seen": 1899547648, |
| "step": 5800, |
| "train_runtime": 44542.2389, |
| "train_tokens_per_second": 42645.985 |
| }, |
| { |
| "epoch": 7.852778244082592, |
| "grad_norm": 26.109970092773438, |
| "learning_rate": 2.2861630729687276e-05, |
| "loss": 16.96089111328125, |
| "num_input_tokens_seen": 1915931648, |
| "step": 5850, |
| "train_runtime": 44924.6097, |
| "train_tokens_per_second": 42647.708 |
| }, |
| { |
| "epoch": 7.919926137317441, |
| "grad_norm": 40.166683197021484, |
| "learning_rate": 2.2147651006711412e-05, |
| "loss": 16.97099365234375, |
| "num_input_tokens_seen": 1932315648, |
| "step": 5900, |
| "train_runtime": 45306.7483, |
| "train_tokens_per_second": 42649.621 |
| }, |
| { |
| "epoch": 7.987074030552291, |
| "grad_norm": 20.37265968322754, |
| "learning_rate": 2.1433671283735544e-05, |
| "loss": 16.91640625, |
| "num_input_tokens_seen": 1948699648, |
| "step": 5950, |
| "train_runtime": 45688.7131, |
| "train_tokens_per_second": 42651.664 |
| }, |
| { |
| "epoch": 8.05371831458788, |
| "grad_norm": 20.146041870117188, |
| "learning_rate": 2.0719691560759673e-05, |
| "loss": 16.76911865234375, |
| "num_input_tokens_seen": 1964941312, |
| "step": 6000, |
| "train_runtime": 46067.8327, |
| "train_tokens_per_second": 42653.218 |
| }, |
| { |
| "epoch": 8.05371831458788, |
| "eval_loss": 1.096535086631775, |
| "eval_masked_accuracy": 0.7597164266773505, |
| "eval_runtime": 18.1153, |
| "eval_samples_per_second": 217.109, |
| "eval_steps_per_second": 3.423, |
| "num_input_tokens_seen": 1964941312, |
| "step": 6000 |
| }, |
| { |
| "epoch": 8.12086620782273, |
| "grad_norm": 19.625225067138672, |
| "learning_rate": 2.000571183778381e-05, |
| "loss": 16.9940234375, |
| "num_input_tokens_seen": 1981325312, |
| "step": 6050, |
| "train_runtime": 46469.6541, |
| "train_tokens_per_second": 42636.971 |
| }, |
| { |
| "epoch": 8.188014101057579, |
| "grad_norm": 18.703025817871094, |
| "learning_rate": 1.9291732114807938e-05, |
| "loss": 16.83005126953125, |
| "num_input_tokens_seen": 1997709312, |
| "step": 6100, |
| "train_runtime": 46851.8703, |
| "train_tokens_per_second": 42638.838 |
| }, |
| { |
| "epoch": 8.255161994292429, |
| "grad_norm": 20.429426193237305, |
| "learning_rate": 1.8577752391832074e-05, |
| "loss": 16.810699462890625, |
| "num_input_tokens_seen": 2014093312, |
| "step": 6150, |
| "train_runtime": 47233.5489, |
| "train_tokens_per_second": 42641.16 |
| }, |
| { |
| "epoch": 8.322309887527279, |
| "grad_norm": 21.28598976135254, |
| "learning_rate": 1.7863772668856206e-05, |
| "loss": 16.7576171875, |
| "num_input_tokens_seen": 2030477312, |
| "step": 6200, |
| "train_runtime": 47616.4549, |
| "train_tokens_per_second": 42642.345 |
| }, |
| { |
| "epoch": 8.389457780762129, |
| "grad_norm": 21.31905174255371, |
| "learning_rate": 1.7149792945880338e-05, |
| "loss": 16.805338134765623, |
| "num_input_tokens_seen": 2046861312, |
| "step": 6250, |
| "train_runtime": 47999.4484, |
| "train_tokens_per_second": 42643.434 |
| }, |
| { |
| "epoch": 8.456605673996979, |
| "grad_norm": 18.618684768676758, |
| "learning_rate": 1.643581322290447e-05, |
| "loss": 16.7966064453125, |
| "num_input_tokens_seen": 2063245312, |
| "step": 6300, |
| "train_runtime": 48381.6311, |
| "train_tokens_per_second": 42645.22 |
| }, |
| { |
| "epoch": 8.523753567231829, |
| "grad_norm": 21.598894119262695, |
| "learning_rate": 1.5721833499928603e-05, |
| "loss": 16.784290771484375, |
| "num_input_tokens_seen": 2079629312, |
| "step": 6350, |
| "train_runtime": 48763.1337, |
| "train_tokens_per_second": 42647.573 |
| }, |
| { |
| "epoch": 8.590901460466679, |
| "grad_norm": 30.116657257080078, |
| "learning_rate": 1.5007853776952735e-05, |
| "loss": 16.74375, |
| "num_input_tokens_seen": 2096013312, |
| "step": 6400, |
| "train_runtime": 49145.7118, |
| "train_tokens_per_second": 42648.956 |
| }, |
| { |
| "epoch": 8.658049353701527, |
| "grad_norm": 60.88018035888672, |
| "learning_rate": 1.4293874053976869e-05, |
| "loss": 16.743026123046874, |
| "num_input_tokens_seen": 2112397312, |
| "step": 6450, |
| "train_runtime": 49527.861, |
| "train_tokens_per_second": 42650.687 |
| }, |
| { |
| "epoch": 8.725197246936377, |
| "grad_norm": 79.8419189453125, |
| "learning_rate": 1.3579894331001e-05, |
| "loss": 16.7031982421875, |
| "num_input_tokens_seen": 2128781312, |
| "step": 6500, |
| "train_runtime": 49910.3805, |
| "train_tokens_per_second": 42652.075 |
| }, |
| { |
| "epoch": 8.725197246936377, |
| "eval_loss": 1.0834989547729492, |
| "eval_masked_accuracy": 0.7624798575184463, |
| "eval_runtime": 18.0968, |
| "eval_samples_per_second": 217.331, |
| "eval_steps_per_second": 3.426, |
| "num_input_tokens_seen": 2128781312, |
| "step": 6500 |
| }, |
| { |
| "epoch": 8.792345140171227, |
| "grad_norm": 24.531356811523438, |
| "learning_rate": 1.2865914608025132e-05, |
| "loss": 16.679058837890626, |
| "num_input_tokens_seen": 2145165312, |
| "step": 6550, |
| "train_runtime": 50313.0537, |
| "train_tokens_per_second": 42636.357 |
| }, |
| { |
| "epoch": 8.859493033406077, |
| "grad_norm": 19.91719627380371, |
| "learning_rate": 1.2151934885049266e-05, |
| "loss": 16.6896630859375, |
| "num_input_tokens_seen": 2161549312, |
| "step": 6600, |
| "train_runtime": 50694.9327, |
| "train_tokens_per_second": 42638.37 |
| }, |
| { |
| "epoch": 8.926640926640927, |
| "grad_norm": 85.14508056640625, |
| "learning_rate": 1.1437955162073398e-05, |
| "loss": 16.694083251953124, |
| "num_input_tokens_seen": 2177933312, |
| "step": 6650, |
| "train_runtime": 51077.1242, |
| "train_tokens_per_second": 42640.093 |
| }, |
| { |
| "epoch": 8.993788819875776, |
| "grad_norm": 19.09105682373047, |
| "learning_rate": 1.072397543909753e-05, |
| "loss": 16.690823974609376, |
| "num_input_tokens_seen": 2194317312, |
| "step": 6700, |
| "train_runtime": 51459.4495, |
| "train_tokens_per_second": 42641.679 |
| }, |
| { |
| "epoch": 9.060433103911365, |
| "grad_norm": 20.286251068115234, |
| "learning_rate": 1.0009995716121663e-05, |
| "loss": 16.543135986328124, |
| "num_input_tokens_seen": 2210558976, |
| "step": 6750, |
| "train_runtime": 51838.5574, |
| "train_tokens_per_second": 42643.142 |
| }, |
| { |
| "epoch": 9.127580997146215, |
| "grad_norm": 19.63930892944336, |
| "learning_rate": 9.296015993145795e-06, |
| "loss": 16.60925048828125, |
| "num_input_tokens_seen": 2226942976, |
| "step": 6800, |
| "train_runtime": 52221.018, |
| "train_tokens_per_second": 42644.572 |
| }, |
| { |
| "epoch": 9.194728890381064, |
| "grad_norm": 20.226930618286133, |
| "learning_rate": 8.582036270169927e-06, |
| "loss": 16.68048828125, |
| "num_input_tokens_seen": 2243326976, |
| "step": 6850, |
| "train_runtime": 52603.1483, |
| "train_tokens_per_second": 42646.249 |
| }, |
| { |
| "epoch": 9.261876783615914, |
| "grad_norm": 18.24042320251465, |
| "learning_rate": 7.86805654719406e-06, |
| "loss": 16.55905517578125, |
| "num_input_tokens_seen": 2259710976, |
| "step": 6900, |
| "train_runtime": 52985.937, |
| "train_tokens_per_second": 42647.372 |
| }, |
| { |
| "epoch": 9.329024676850764, |
| "grad_norm": 22.2435245513916, |
| "learning_rate": 7.154076824218192e-06, |
| "loss": 16.5089697265625, |
| "num_input_tokens_seen": 2276094976, |
| "step": 6950, |
| "train_runtime": 53367.9926, |
| "train_tokens_per_second": 42649.065 |
| }, |
| { |
| "epoch": 9.396172570085614, |
| "grad_norm": 20.44721031188965, |
| "learning_rate": 6.440097101242325e-06, |
| "loss": 16.580111083984374, |
| "num_input_tokens_seen": 2292478976, |
| "step": 7000, |
| "train_runtime": 53749.9804, |
| "train_tokens_per_second": 42650.787 |
| }, |
| { |
| "epoch": 9.396172570085614, |
| "eval_loss": 1.0845201015472412, |
| "eval_masked_accuracy": 0.7618803372351338, |
| "eval_runtime": 18.1075, |
| "eval_samples_per_second": 217.202, |
| "eval_steps_per_second": 3.424, |
| "num_input_tokens_seen": 2292478976, |
| "step": 7000 |
| }, |
| { |
| "epoch": 9.463320463320464, |
| "grad_norm": 18.416465759277344, |
| "learning_rate": 5.726117378266457e-06, |
| "loss": 16.533978271484376, |
| "num_input_tokens_seen": 2308862976, |
| "step": 7050, |
| "train_runtime": 54152.7328, |
| "train_tokens_per_second": 42636.13 |
| }, |
| { |
| "epoch": 9.530468356555312, |
| "grad_norm": 19.039291381835938, |
| "learning_rate": 5.01213765529059e-06, |
| "loss": 16.6065771484375, |
| "num_input_tokens_seen": 2325246976, |
| "step": 7100, |
| "train_runtime": 54534.7391, |
| "train_tokens_per_second": 42637.904 |
| }, |
| { |
| "epoch": 9.597616249790162, |
| "grad_norm": 19.21045684814453, |
| "learning_rate": 4.298157932314723e-06, |
| "loss": 16.554276123046876, |
| "num_input_tokens_seen": 2341630976, |
| "step": 7150, |
| "train_runtime": 54916.6532, |
| "train_tokens_per_second": 42639.725 |
| }, |
| { |
| "epoch": 9.664764143025012, |
| "grad_norm": 21.09920310974121, |
| "learning_rate": 3.584178209338855e-06, |
| "loss": 16.55759033203125, |
| "num_input_tokens_seen": 2358014976, |
| "step": 7200, |
| "train_runtime": 55298.628, |
| "train_tokens_per_second": 42641.473 |
| }, |
| { |
| "epoch": 9.731912036259862, |
| "grad_norm": 30.43587875366211, |
| "learning_rate": 2.8701984863629874e-06, |
| "loss": 16.526624755859373, |
| "num_input_tokens_seen": 2374398976, |
| "step": 7250, |
| "train_runtime": 55680.03, |
| "train_tokens_per_second": 42643.637 |
| }, |
| { |
| "epoch": 9.799059929494712, |
| "grad_norm": 25.07560157775879, |
| "learning_rate": 2.15621876338712e-06, |
| "loss": 16.5307763671875, |
| "num_input_tokens_seen": 2390782976, |
| "step": 7300, |
| "train_runtime": 56062.5396, |
| "train_tokens_per_second": 42644.928 |
| }, |
| { |
| "epoch": 9.866207822729562, |
| "grad_norm": 24.447111129760742, |
| "learning_rate": 1.4422390404112524e-06, |
| "loss": 16.479052734375, |
| "num_input_tokens_seen": 2407166976, |
| "step": 7350, |
| "train_runtime": 56444.4228, |
| "train_tokens_per_second": 42646.675 |
| }, |
| { |
| "epoch": 9.933355715964412, |
| "grad_norm": 23.812557220458984, |
| "learning_rate": 7.282593174353849e-07, |
| "loss": 16.499632568359374, |
| "num_input_tokens_seen": 2423550976, |
| "step": 7400, |
| "train_runtime": 56826.7541, |
| "train_tokens_per_second": 42648.063 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 27.18109130859375, |
| "learning_rate": 1.427959445951735e-08, |
| "loss": 16.4600341796875, |
| "num_input_tokens_seen": 2439792640, |
| "step": 7450, |
| "train_runtime": 57205.5292, |
| "train_tokens_per_second": 42649.595 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 1.0779017210006714, |
| "eval_masked_accuracy": 0.763301870184039, |
| "eval_runtime": 18.0983, |
| "eval_samples_per_second": 217.313, |
| "eval_steps_per_second": 3.426, |
| "num_input_tokens_seen": 2439792640, |
| "step": 7450 |
| }, |
| { |
| "epoch": 10.0, |
| "num_input_tokens_seen": 2439792640, |
| "step": 7450, |
| "total_flos": 8.192069135329812e+18, |
| "train_loss": 19.39006050468291, |
| "train_runtime": 57225.9994, |
| "train_samples_per_second": 83.27, |
| "train_steps_per_second": 0.13 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 7450, |
| "num_input_tokens_seen": 2439792640, |
| "num_train_epochs": 10, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 5, |
| "early_stopping_threshold": 0.0001 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 0 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 8.192069135329812e+18, |
| "train_batch_size": 40, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
