| { |
| "best_global_step": 741, |
| "best_metric": 0.012937678955495358, |
| "best_model_checkpoint": "saves_stability/prefix-tuning/llama-3-8b-instruct/train_cb_1757081469/checkpoint-741", |
| "epoch": 10.0, |
| "eval_steps": 57, |
| "global_step": 1130, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.04424778761061947, |
| "grad_norm": 172.90432739257812, |
| "learning_rate": 1.7699115044247788e-06, |
| "loss": 9.3038, |
| "num_input_tokens_seen": 1520, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.08849557522123894, |
| "grad_norm": 162.92501831054688, |
| "learning_rate": 3.982300884955752e-06, |
| "loss": 7.8011, |
| "num_input_tokens_seen": 2976, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.13274336283185842, |
| "grad_norm": 118.36426544189453, |
| "learning_rate": 6.194690265486726e-06, |
| "loss": 5.8089, |
| "num_input_tokens_seen": 4688, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.17699115044247787, |
| "grad_norm": 119.76227569580078, |
| "learning_rate": 8.407079646017701e-06, |
| "loss": 3.2162, |
| "num_input_tokens_seen": 5776, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.22123893805309736, |
| "grad_norm": 63.143497467041016, |
| "learning_rate": 1.0619469026548673e-05, |
| "loss": 1.0524, |
| "num_input_tokens_seen": 6992, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.26548672566371684, |
| "grad_norm": 64.27384185791016, |
| "learning_rate": 1.2831858407079647e-05, |
| "loss": 1.2224, |
| "num_input_tokens_seen": 8320, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.30973451327433627, |
| "grad_norm": 54.031402587890625, |
| "learning_rate": 1.504424778761062e-05, |
| "loss": 0.959, |
| "num_input_tokens_seen": 9536, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.35398230088495575, |
| "grad_norm": 58.8524284362793, |
| "learning_rate": 1.7256637168141594e-05, |
| "loss": 0.3716, |
| "num_input_tokens_seen": 11072, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.39823008849557523, |
| "grad_norm": 48.3114128112793, |
| "learning_rate": 1.946902654867257e-05, |
| "loss": 1.0354, |
| "num_input_tokens_seen": 12320, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.4424778761061947, |
| "grad_norm": 48.01361083984375, |
| "learning_rate": 2.1681415929203542e-05, |
| "loss": 0.3628, |
| "num_input_tokens_seen": 13616, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.48672566371681414, |
| "grad_norm": 65.13843536376953, |
| "learning_rate": 2.3893805309734516e-05, |
| "loss": 0.8643, |
| "num_input_tokens_seen": 15104, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.504424778761062, |
| "eval_loss": 0.4547373652458191, |
| "eval_runtime": 0.7799, |
| "eval_samples_per_second": 32.053, |
| "eval_steps_per_second": 16.668, |
| "num_input_tokens_seen": 15664, |
| "step": 57 |
| }, |
| { |
| "epoch": 0.5309734513274337, |
| "grad_norm": 16.273540496826172, |
| "learning_rate": 2.610619469026549e-05, |
| "loss": 0.3507, |
| "num_input_tokens_seen": 16352, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.5752212389380531, |
| "grad_norm": 6.617971897125244, |
| "learning_rate": 2.831858407079646e-05, |
| "loss": 0.3175, |
| "num_input_tokens_seen": 17888, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.6194690265486725, |
| "grad_norm": 56.716156005859375, |
| "learning_rate": 3.0530973451327434e-05, |
| "loss": 0.7932, |
| "num_input_tokens_seen": 19008, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.6637168141592921, |
| "grad_norm": 6.397486686706543, |
| "learning_rate": 3.274336283185841e-05, |
| "loss": 0.2819, |
| "num_input_tokens_seen": 20480, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.7079646017699115, |
| "grad_norm": 49.061546325683594, |
| "learning_rate": 3.495575221238938e-05, |
| "loss": 0.5588, |
| "num_input_tokens_seen": 22000, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.7522123893805309, |
| "grad_norm": 101.45230102539062, |
| "learning_rate": 3.716814159292036e-05, |
| "loss": 0.303, |
| "num_input_tokens_seen": 23456, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.7964601769911505, |
| "grad_norm": 7.564614772796631, |
| "learning_rate": 3.938053097345133e-05, |
| "loss": 0.2319, |
| "num_input_tokens_seen": 25280, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.8407079646017699, |
| "grad_norm": 7.837498188018799, |
| "learning_rate": 4.15929203539823e-05, |
| "loss": 0.1757, |
| "num_input_tokens_seen": 26496, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.8849557522123894, |
| "grad_norm": 11.753522872924805, |
| "learning_rate": 4.380530973451328e-05, |
| "loss": 0.1669, |
| "num_input_tokens_seen": 28016, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.9292035398230089, |
| "grad_norm": 17.86298179626465, |
| "learning_rate": 4.601769911504425e-05, |
| "loss": 0.538, |
| "num_input_tokens_seen": 29344, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.9734513274336283, |
| "grad_norm": 15.845224380493164, |
| "learning_rate": 4.823008849557522e-05, |
| "loss": 0.2715, |
| "num_input_tokens_seen": 30496, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.008849557522124, |
| "eval_loss": 0.5620186924934387, |
| "eval_runtime": 0.7758, |
| "eval_samples_per_second": 32.224, |
| "eval_steps_per_second": 16.756, |
| "num_input_tokens_seen": 31304, |
| "step": 114 |
| }, |
| { |
| "epoch": 1.0176991150442478, |
| "grad_norm": 20.063274383544922, |
| "learning_rate": 4.9999880720033856e-05, |
| "loss": 0.2842, |
| "num_input_tokens_seen": 31560, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.0619469026548674, |
| "grad_norm": 4.028425693511963, |
| "learning_rate": 4.999570604073014e-05, |
| "loss": 0.9137, |
| "num_input_tokens_seen": 32744, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.1061946902654867, |
| "grad_norm": 11.262805938720703, |
| "learning_rate": 4.998556850128691e-05, |
| "loss": 0.4827, |
| "num_input_tokens_seen": 34440, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.1504424778761062, |
| "grad_norm": 229.07241821289062, |
| "learning_rate": 4.996947052006874e-05, |
| "loss": 0.4721, |
| "num_input_tokens_seen": 35992, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.1946902654867257, |
| "grad_norm": 33.70645523071289, |
| "learning_rate": 4.9947415937335635e-05, |
| "loss": 0.6239, |
| "num_input_tokens_seen": 37480, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.238938053097345, |
| "grad_norm": 20.447160720825195, |
| "learning_rate": 4.9919410014326955e-05, |
| "loss": 0.3049, |
| "num_input_tokens_seen": 38776, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.2831858407079646, |
| "grad_norm": 11.072810173034668, |
| "learning_rate": 4.98854594320063e-05, |
| "loss": 0.2219, |
| "num_input_tokens_seen": 40408, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.3274336283185841, |
| "grad_norm": 5.849891185760498, |
| "learning_rate": 4.98455722894677e-05, |
| "loss": 0.9584, |
| "num_input_tokens_seen": 41944, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.3716814159292037, |
| "grad_norm": 3.3164310455322266, |
| "learning_rate": 4.979975810200359e-05, |
| "loss": 0.2229, |
| "num_input_tokens_seen": 43576, |
| "step": 155 |
| }, |
| { |
| "epoch": 1.415929203539823, |
| "grad_norm": 149.41354370117188, |
| "learning_rate": 4.974802779883484e-05, |
| "loss": 0.5863, |
| "num_input_tokens_seen": 44984, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.4601769911504425, |
| "grad_norm": 9.543984413146973, |
| "learning_rate": 4.969039372050356e-05, |
| "loss": 0.5308, |
| "num_input_tokens_seen": 46424, |
| "step": 165 |
| }, |
| { |
| "epoch": 1.504424778761062, |
| "grad_norm": 14.939043045043945, |
| "learning_rate": 4.9626869615929175e-05, |
| "loss": 0.3501, |
| "num_input_tokens_seen": 47800, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.5132743362831858, |
| "eval_loss": 0.2233014702796936, |
| "eval_runtime": 0.7784, |
| "eval_samples_per_second": 32.116, |
| "eval_steps_per_second": 16.7, |
| "num_input_tokens_seen": 48008, |
| "step": 171 |
| }, |
| { |
| "epoch": 1.5486725663716814, |
| "grad_norm": 2.0158536434173584, |
| "learning_rate": 4.9557470639128554e-05, |
| "loss": 0.4255, |
| "num_input_tokens_seen": 48776, |
| "step": 175 |
| }, |
| { |
| "epoch": 1.592920353982301, |
| "grad_norm": 4.7409796714782715, |
| "learning_rate": 4.9482213345600936e-05, |
| "loss": 0.1645, |
| "num_input_tokens_seen": 49960, |
| "step": 180 |
| }, |
| { |
| "epoch": 1.6371681415929205, |
| "grad_norm": 0.18256613612174988, |
| "learning_rate": 4.940111568837854e-05, |
| "loss": 0.0174, |
| "num_input_tokens_seen": 51624, |
| "step": 185 |
| }, |
| { |
| "epoch": 1.6814159292035398, |
| "grad_norm": 16.183256149291992, |
| "learning_rate": 4.931419701374377e-05, |
| "loss": 0.2941, |
| "num_input_tokens_seen": 53048, |
| "step": 190 |
| }, |
| { |
| "epoch": 1.7256637168141593, |
| "grad_norm": 20.320951461791992, |
| "learning_rate": 4.922147805661402e-05, |
| "loss": 1.1045, |
| "num_input_tokens_seen": 54552, |
| "step": 195 |
| }, |
| { |
| "epoch": 1.7699115044247788, |
| "grad_norm": 12.029666900634766, |
| "learning_rate": 4.91229809355953e-05, |
| "loss": 0.4722, |
| "num_input_tokens_seen": 55656, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.8141592920353982, |
| "grad_norm": 0.27077895402908325, |
| "learning_rate": 4.901872914770569e-05, |
| "loss": 0.3305, |
| "num_input_tokens_seen": 56984, |
| "step": 205 |
| }, |
| { |
| "epoch": 1.8584070796460177, |
| "grad_norm": 3.5023787021636963, |
| "learning_rate": 4.8908747562769995e-05, |
| "loss": 0.5475, |
| "num_input_tokens_seen": 58200, |
| "step": 210 |
| }, |
| { |
| "epoch": 1.9026548672566372, |
| "grad_norm": 6.9777631759643555, |
| "learning_rate": 4.8793062417486976e-05, |
| "loss": 0.1367, |
| "num_input_tokens_seen": 59544, |
| "step": 215 |
| }, |
| { |
| "epoch": 1.9469026548672566, |
| "grad_norm": 9.807963371276855, |
| "learning_rate": 4.867170130917034e-05, |
| "loss": 0.5382, |
| "num_input_tokens_seen": 60968, |
| "step": 220 |
| }, |
| { |
| "epoch": 1.991150442477876, |
| "grad_norm": 10.069539070129395, |
| "learning_rate": 4.8544693189165324e-05, |
| "loss": 0.5262, |
| "num_input_tokens_seen": 62216, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.017699115044248, |
| "eval_loss": 0.32860472798347473, |
| "eval_runtime": 0.777, |
| "eval_samples_per_second": 32.176, |
| "eval_steps_per_second": 16.731, |
| "num_input_tokens_seen": 63040, |
| "step": 228 |
| }, |
| { |
| "epoch": 2.0353982300884956, |
| "grad_norm": 14.066644668579102, |
| "learning_rate": 4.841206835594222e-05, |
| "loss": 0.2754, |
| "num_input_tokens_seen": 63568, |
| "step": 230 |
| }, |
| { |
| "epoch": 2.079646017699115, |
| "grad_norm": 9.898775100708008, |
| "learning_rate": 4.8273858447868424e-05, |
| "loss": 0.1062, |
| "num_input_tokens_seen": 64800, |
| "step": 235 |
| }, |
| { |
| "epoch": 2.1238938053097347, |
| "grad_norm": 0.7710229158401489, |
| "learning_rate": 4.813009643566101e-05, |
| "loss": 0.4179, |
| "num_input_tokens_seen": 66144, |
| "step": 240 |
| }, |
| { |
| "epoch": 2.168141592920354, |
| "grad_norm": 0.04852516949176788, |
| "learning_rate": 4.798081661452135e-05, |
| "loss": 0.047, |
| "num_input_tokens_seen": 67600, |
| "step": 245 |
| }, |
| { |
| "epoch": 2.2123893805309733, |
| "grad_norm": 14.3930082321167, |
| "learning_rate": 4.7826054595953815e-05, |
| "loss": 0.3323, |
| "num_input_tokens_seen": 69056, |
| "step": 250 |
| }, |
| { |
| "epoch": 2.256637168141593, |
| "grad_norm": 2.9247560501098633, |
| "learning_rate": 4.766584729927049e-05, |
| "loss": 0.361, |
| "num_input_tokens_seen": 70304, |
| "step": 255 |
| }, |
| { |
| "epoch": 2.3008849557522124, |
| "grad_norm": 0.6192066073417664, |
| "learning_rate": 4.750023294278381e-05, |
| "loss": 0.1337, |
| "num_input_tokens_seen": 71872, |
| "step": 260 |
| }, |
| { |
| "epoch": 2.3451327433628317, |
| "grad_norm": 6.6043701171875, |
| "learning_rate": 4.732925103468944e-05, |
| "loss": 0.2764, |
| "num_input_tokens_seen": 73520, |
| "step": 265 |
| }, |
| { |
| "epoch": 2.3893805309734515, |
| "grad_norm": 0.0575706772506237, |
| "learning_rate": 4.715294236364135e-05, |
| "loss": 0.0771, |
| "num_input_tokens_seen": 75184, |
| "step": 270 |
| }, |
| { |
| "epoch": 2.433628318584071, |
| "grad_norm": 9.62030029296875, |
| "learning_rate": 4.6971348989021446e-05, |
| "loss": 0.0644, |
| "num_input_tokens_seen": 76592, |
| "step": 275 |
| }, |
| { |
| "epoch": 2.47787610619469, |
| "grad_norm": 11.746428489685059, |
| "learning_rate": 4.6784514230906146e-05, |
| "loss": 0.7766, |
| "num_input_tokens_seen": 77632, |
| "step": 280 |
| }, |
| { |
| "epoch": 2.52212389380531, |
| "grad_norm": 0.6659058928489685, |
| "learning_rate": 4.659248265973205e-05, |
| "loss": 0.2992, |
| "num_input_tokens_seen": 79104, |
| "step": 285 |
| }, |
| { |
| "epoch": 2.52212389380531, |
| "eval_loss": 0.13774029910564423, |
| "eval_runtime": 0.7715, |
| "eval_samples_per_second": 32.404, |
| "eval_steps_per_second": 16.85, |
| "num_input_tokens_seen": 79104, |
| "step": 285 |
| }, |
| { |
| "epoch": 2.566371681415929, |
| "grad_norm": 7.603933334350586, |
| "learning_rate": 4.639530008566349e-05, |
| "loss": 0.0878, |
| "num_input_tokens_seen": 80432, |
| "step": 290 |
| }, |
| { |
| "epoch": 2.6106194690265485, |
| "grad_norm": 12.166590690612793, |
| "learning_rate": 4.6193013547664255e-05, |
| "loss": 0.5536, |
| "num_input_tokens_seen": 81744, |
| "step": 295 |
| }, |
| { |
| "epoch": 2.6548672566371683, |
| "grad_norm": 20.62239646911621, |
| "learning_rate": 4.5985671302276165e-05, |
| "loss": 0.6957, |
| "num_input_tokens_seen": 83104, |
| "step": 300 |
| }, |
| { |
| "epoch": 2.6991150442477876, |
| "grad_norm": 7.500833034515381, |
| "learning_rate": 4.577332281210727e-05, |
| "loss": 0.3277, |
| "num_input_tokens_seen": 84336, |
| "step": 305 |
| }, |
| { |
| "epoch": 2.7433628318584073, |
| "grad_norm": 4.258105278015137, |
| "learning_rate": 4.55560187340322e-05, |
| "loss": 0.4949, |
| "num_input_tokens_seen": 85936, |
| "step": 310 |
| }, |
| { |
| "epoch": 2.7876106194690267, |
| "grad_norm": 1.3956164121627808, |
| "learning_rate": 4.533381090710776e-05, |
| "loss": 0.1694, |
| "num_input_tokens_seen": 87232, |
| "step": 315 |
| }, |
| { |
| "epoch": 2.831858407079646, |
| "grad_norm": 0.49883967638015747, |
| "learning_rate": 4.5106752340206435e-05, |
| "loss": 0.0296, |
| "num_input_tokens_seen": 88336, |
| "step": 320 |
| }, |
| { |
| "epoch": 2.8761061946902657, |
| "grad_norm": 4.606618881225586, |
| "learning_rate": 4.4874897199370805e-05, |
| "loss": 0.3007, |
| "num_input_tokens_seen": 89584, |
| "step": 325 |
| }, |
| { |
| "epoch": 2.920353982300885, |
| "grad_norm": 8.798304557800293, |
| "learning_rate": 4.463830079489196e-05, |
| "loss": 0.1205, |
| "num_input_tokens_seen": 91008, |
| "step": 330 |
| }, |
| { |
| "epoch": 2.9646017699115044, |
| "grad_norm": 0.30231973528862, |
| "learning_rate": 4.4397019568114987e-05, |
| "loss": 0.3998, |
| "num_input_tokens_seen": 92288, |
| "step": 335 |
| }, |
| { |
| "epoch": 3.0088495575221237, |
| "grad_norm": 1.7744728326797485, |
| "learning_rate": 4.415111107797445e-05, |
| "loss": 0.0962, |
| "num_input_tokens_seen": 93456, |
| "step": 340 |
| }, |
| { |
| "epoch": 3.0265486725663715, |
| "eval_loss": 0.15844351053237915, |
| "eval_runtime": 0.7752, |
| "eval_samples_per_second": 32.251, |
| "eval_steps_per_second": 16.77, |
| "num_input_tokens_seen": 93888, |
| "step": 342 |
| }, |
| { |
| "epoch": 3.0530973451327434, |
| "grad_norm": 0.12336965650320053, |
| "learning_rate": 4.390063398726356e-05, |
| "loss": 0.0034, |
| "num_input_tokens_seen": 94752, |
| "step": 345 |
| }, |
| { |
| "epoch": 3.0973451327433628, |
| "grad_norm": 0.4525299072265625, |
| "learning_rate": 4.3645648048639734e-05, |
| "loss": 0.3721, |
| "num_input_tokens_seen": 96000, |
| "step": 350 |
| }, |
| { |
| "epoch": 3.1415929203539825, |
| "grad_norm": 2.180138111114502, |
| "learning_rate": 4.338621409037031e-05, |
| "loss": 0.0544, |
| "num_input_tokens_seen": 97568, |
| "step": 355 |
| }, |
| { |
| "epoch": 3.185840707964602, |
| "grad_norm": 15.457188606262207, |
| "learning_rate": 4.312239400182166e-05, |
| "loss": 0.2974, |
| "num_input_tokens_seen": 98896, |
| "step": 360 |
| }, |
| { |
| "epoch": 3.230088495575221, |
| "grad_norm": 0.034761518239974976, |
| "learning_rate": 4.285425071869511e-05, |
| "loss": 0.2165, |
| "num_input_tokens_seen": 100336, |
| "step": 365 |
| }, |
| { |
| "epoch": 3.274336283185841, |
| "grad_norm": 8.344590187072754, |
| "learning_rate": 4.258184820801331e-05, |
| "loss": 0.3636, |
| "num_input_tokens_seen": 101888, |
| "step": 370 |
| }, |
| { |
| "epoch": 3.3185840707964602, |
| "grad_norm": 13.406426429748535, |
| "learning_rate": 4.230525145286057e-05, |
| "loss": 0.4511, |
| "num_input_tokens_seen": 103184, |
| "step": 375 |
| }, |
| { |
| "epoch": 3.3628318584070795, |
| "grad_norm": 0.5580623745918274, |
| "learning_rate": 4.2024526436880785e-05, |
| "loss": 0.0776, |
| "num_input_tokens_seen": 104672, |
| "step": 380 |
| }, |
| { |
| "epoch": 3.4070796460176993, |
| "grad_norm": 0.5934048891067505, |
| "learning_rate": 4.173974012853665e-05, |
| "loss": 0.2344, |
| "num_input_tokens_seen": 105968, |
| "step": 385 |
| }, |
| { |
| "epoch": 3.4513274336283186, |
| "grad_norm": 2.0349247455596924, |
| "learning_rate": 4.1450960465134025e-05, |
| "loss": 0.0518, |
| "num_input_tokens_seen": 107472, |
| "step": 390 |
| }, |
| { |
| "epoch": 3.495575221238938, |
| "grad_norm": 0.06618373095989227, |
| "learning_rate": 4.1158256336615096e-05, |
| "loss": 0.0753, |
| "num_input_tokens_seen": 108688, |
| "step": 395 |
| }, |
| { |
| "epoch": 3.5309734513274336, |
| "eval_loss": 0.3117164373397827, |
| "eval_runtime": 0.7709, |
| "eval_samples_per_second": 32.432, |
| "eval_steps_per_second": 16.864, |
| "num_input_tokens_seen": 109952, |
| "step": 399 |
| }, |
| { |
| "epoch": 3.5398230088495577, |
| "grad_norm": 5.4218974113464355, |
| "learning_rate": 4.086169756912431e-05, |
| "loss": 0.5723, |
| "num_input_tokens_seen": 110320, |
| "step": 400 |
| }, |
| { |
| "epoch": 3.584070796460177, |
| "grad_norm": 2.7802650928497314, |
| "learning_rate": 4.056135490835098e-05, |
| "loss": 0.3036, |
| "num_input_tokens_seen": 111952, |
| "step": 405 |
| }, |
| { |
| "epoch": 3.6283185840707963, |
| "grad_norm": 3.5373213291168213, |
| "learning_rate": 4.025730000265251e-05, |
| "loss": 0.1427, |
| "num_input_tokens_seen": 113152, |
| "step": 410 |
| }, |
| { |
| "epoch": 3.672566371681416, |
| "grad_norm": 9.953773498535156, |
| "learning_rate": 3.994960538596232e-05, |
| "loss": 0.1798, |
| "num_input_tokens_seen": 114592, |
| "step": 415 |
| }, |
| { |
| "epoch": 3.7168141592920354, |
| "grad_norm": 0.10559792071580887, |
| "learning_rate": 3.963834446048644e-05, |
| "loss": 0.3672, |
| "num_input_tokens_seen": 115936, |
| "step": 420 |
| }, |
| { |
| "epoch": 3.7610619469026547, |
| "grad_norm": 0.500779926776886, |
| "learning_rate": 3.9323591479193074e-05, |
| "loss": 0.1044, |
| "num_input_tokens_seen": 117232, |
| "step": 425 |
| }, |
| { |
| "epoch": 3.8053097345132745, |
| "grad_norm": 1.8762531280517578, |
| "learning_rate": 3.90054215280991e-05, |
| "loss": 0.2184, |
| "num_input_tokens_seen": 118448, |
| "step": 430 |
| }, |
| { |
| "epoch": 3.849557522123894, |
| "grad_norm": 2.281862735748291, |
| "learning_rate": 3.8683910508357934e-05, |
| "loss": 0.7422, |
| "num_input_tokens_seen": 119760, |
| "step": 435 |
| }, |
| { |
| "epoch": 3.893805309734513, |
| "grad_norm": 0.25059983134269714, |
| "learning_rate": 3.83591351181529e-05, |
| "loss": 0.0681, |
| "num_input_tokens_seen": 121488, |
| "step": 440 |
| }, |
| { |
| "epoch": 3.938053097345133, |
| "grad_norm": 5.358190059661865, |
| "learning_rate": 3.803117283440044e-05, |
| "loss": 0.3385, |
| "num_input_tokens_seen": 122816, |
| "step": 445 |
| }, |
| { |
| "epoch": 3.982300884955752, |
| "grad_norm": 3.208345413208008, |
| "learning_rate": 3.7700101894267615e-05, |
| "loss": 0.0398, |
| "num_input_tokens_seen": 124352, |
| "step": 450 |
| }, |
| { |
| "epoch": 4.0265486725663715, |
| "grad_norm": 0.5785738825798035, |
| "learning_rate": 3.736600127650817e-05, |
| "loss": 0.4213, |
| "num_input_tokens_seen": 125432, |
| "step": 455 |
| }, |
| { |
| "epoch": 4.035398230088496, |
| "eval_loss": 0.1032278910279274, |
| "eval_runtime": 0.7761, |
| "eval_samples_per_second": 32.214, |
| "eval_steps_per_second": 16.751, |
| "num_input_tokens_seen": 125688, |
| "step": 456 |
| }, |
| { |
| "epoch": 4.070796460176991, |
| "grad_norm": 1.3466124534606934, |
| "learning_rate": 3.702895068262177e-05, |
| "loss": 0.0458, |
| "num_input_tokens_seen": 126744, |
| "step": 460 |
| }, |
| { |
| "epoch": 4.115044247787611, |
| "grad_norm": 9.679712295532227, |
| "learning_rate": 3.66890305178407e-05, |
| "loss": 0.0737, |
| "num_input_tokens_seen": 128424, |
| "step": 465 |
| }, |
| { |
| "epoch": 4.15929203539823, |
| "grad_norm": 0.07868790626525879, |
| "learning_rate": 3.63463218719489e-05, |
| "loss": 0.1408, |
| "num_input_tokens_seen": 129640, |
| "step": 470 |
| }, |
| { |
| "epoch": 4.20353982300885, |
| "grad_norm": 0.5945039391517639, |
| "learning_rate": 3.600090649993741e-05, |
| "loss": 0.1903, |
| "num_input_tokens_seen": 131016, |
| "step": 475 |
| }, |
| { |
| "epoch": 4.247787610619469, |
| "grad_norm": 8.945996284484863, |
| "learning_rate": 3.565286680250138e-05, |
| "loss": 0.0682, |
| "num_input_tokens_seen": 132552, |
| "step": 480 |
| }, |
| { |
| "epoch": 4.292035398230088, |
| "grad_norm": 1.1376537084579468, |
| "learning_rate": 3.530228580638291e-05, |
| "loss": 0.2208, |
| "num_input_tokens_seen": 133624, |
| "step": 485 |
| }, |
| { |
| "epoch": 4.336283185840708, |
| "grad_norm": 0.1638191044330597, |
| "learning_rate": 3.494924714456454e-05, |
| "loss": 0.0085, |
| "num_input_tokens_seen": 135464, |
| "step": 490 |
| }, |
| { |
| "epoch": 4.380530973451328, |
| "grad_norm": 0.43489018082618713, |
| "learning_rate": 3.459383503631823e-05, |
| "loss": 0.1116, |
| "num_input_tokens_seen": 136664, |
| "step": 495 |
| }, |
| { |
| "epoch": 4.424778761061947, |
| "grad_norm": 14.490530967712402, |
| "learning_rate": 3.423613426711431e-05, |
| "loss": 0.4935, |
| "num_input_tokens_seen": 137752, |
| "step": 500 |
| }, |
| { |
| "epoch": 4.469026548672566, |
| "grad_norm": 0.006479791831225157, |
| "learning_rate": 3.3876230168395525e-05, |
| "loss": 0.0561, |
| "num_input_tokens_seen": 139000, |
| "step": 505 |
| }, |
| { |
| "epoch": 4.513274336283186, |
| "grad_norm": 0.9910786747932434, |
| "learning_rate": 3.3514208597220705e-05, |
| "loss": 0.0041, |
| "num_input_tokens_seen": 140184, |
| "step": 510 |
| }, |
| { |
| "epoch": 4.539823008849558, |
| "eval_loss": 0.10921850800514221, |
| "eval_runtime": 0.7726, |
| "eval_samples_per_second": 32.357, |
| "eval_steps_per_second": 16.826, |
| "num_input_tokens_seen": 141000, |
| "step": 513 |
| }, |
| { |
| "epoch": 4.557522123893805, |
| "grad_norm": 0.04926896467804909, |
| "learning_rate": 3.315015591578314e-05, |
| "loss": 0.129, |
| "num_input_tokens_seen": 141512, |
| "step": 515 |
| }, |
| { |
| "epoch": 4.601769911504425, |
| "grad_norm": 3.648982286453247, |
| "learning_rate": 3.278415897080839e-05, |
| "loss": 0.2731, |
| "num_input_tokens_seen": 142888, |
| "step": 520 |
| }, |
| { |
| "epoch": 4.646017699115045, |
| "grad_norm": 0.5655894875526428, |
| "learning_rate": 3.2416305072836557e-05, |
| "loss": 0.1353, |
| "num_input_tokens_seen": 144312, |
| "step": 525 |
| }, |
| { |
| "epoch": 4.6902654867256635, |
| "grad_norm": 12.462039947509766, |
| "learning_rate": 3.204668197539385e-05, |
| "loss": 0.3391, |
| "num_input_tokens_seen": 146008, |
| "step": 530 |
| }, |
| { |
| "epoch": 4.734513274336283, |
| "grad_norm": 14.530315399169922, |
| "learning_rate": 3.167537785405854e-05, |
| "loss": 0.3179, |
| "num_input_tokens_seen": 147448, |
| "step": 535 |
| }, |
| { |
| "epoch": 4.778761061946903, |
| "grad_norm": 0.20855922996997833, |
| "learning_rate": 3.13024812854262e-05, |
| "loss": 0.0135, |
| "num_input_tokens_seen": 148952, |
| "step": 540 |
| }, |
| { |
| "epoch": 4.823008849557522, |
| "grad_norm": 0.016964532434940338, |
| "learning_rate": 3.092808122597924e-05, |
| "loss": 0.1094, |
| "num_input_tokens_seen": 150472, |
| "step": 545 |
| }, |
| { |
| "epoch": 4.867256637168142, |
| "grad_norm": 17.008296966552734, |
| "learning_rate": 3.0552266990865934e-05, |
| "loss": 0.8798, |
| "num_input_tokens_seen": 151880, |
| "step": 550 |
| }, |
| { |
| "epoch": 4.911504424778761, |
| "grad_norm": 1.905720829963684, |
| "learning_rate": 3.017512823259373e-05, |
| "loss": 0.1146, |
| "num_input_tokens_seen": 153336, |
| "step": 555 |
| }, |
| { |
| "epoch": 4.95575221238938, |
| "grad_norm": 0.22706276178359985, |
| "learning_rate": 2.9796754919642228e-05, |
| "loss": 0.0376, |
| "num_input_tokens_seen": 154584, |
| "step": 560 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.08005864918231964, |
| "learning_rate": 2.941723731500068e-05, |
| "loss": 0.1895, |
| "num_input_tokens_seen": 155672, |
| "step": 565 |
| }, |
| { |
| "epoch": 5.04424778761062, |
| "grad_norm": 2.539489984512329, |
| "learning_rate": 2.9036665954635266e-05, |
| "loss": 0.2002, |
| "num_input_tokens_seen": 156920, |
| "step": 570 |
| }, |
| { |
| "epoch": 5.04424778761062, |
| "eval_loss": 0.1037435531616211, |
| "eval_runtime": 0.7749, |
| "eval_samples_per_second": 32.262, |
| "eval_steps_per_second": 16.776, |
| "num_input_tokens_seen": 156920, |
| "step": 570 |
| }, |
| { |
| "epoch": 5.088495575221239, |
| "grad_norm": 1.3146986961364746, |
| "learning_rate": 2.8655131625891274e-05, |
| "loss": 0.0115, |
| "num_input_tokens_seen": 158872, |
| "step": 575 |
| }, |
| { |
| "epoch": 5.132743362831858, |
| "grad_norm": 0.059930458664894104, |
| "learning_rate": 2.8272725345835278e-05, |
| "loss": 0.002, |
| "num_input_tokens_seen": 159960, |
| "step": 580 |
| }, |
| { |
| "epoch": 5.176991150442478, |
| "grad_norm": 0.26709067821502686, |
| "learning_rate": 2.7889538339542527e-05, |
| "loss": 0.0781, |
| "num_input_tokens_seen": 161112, |
| "step": 585 |
| }, |
| { |
| "epoch": 5.221238938053097, |
| "grad_norm": 21.705875396728516, |
| "learning_rate": 2.7505662018334698e-05, |
| "loss": 0.0835, |
| "num_input_tokens_seen": 162472, |
| "step": 590 |
| }, |
| { |
| "epoch": 5.265486725663717, |
| "grad_norm": 12.45948314666748, |
| "learning_rate": 2.712118795797329e-05, |
| "loss": 0.3655, |
| "num_input_tokens_seen": 164216, |
| "step": 595 |
| }, |
| { |
| "epoch": 5.3097345132743365, |
| "grad_norm": 3.4451708793640137, |
| "learning_rate": 2.6736207876813646e-05, |
| "loss": 0.0235, |
| "num_input_tokens_seen": 165832, |
| "step": 600 |
| }, |
| { |
| "epoch": 5.353982300884955, |
| "grad_norm": 0.08687157928943634, |
| "learning_rate": 2.635081361392513e-05, |
| "loss": 0.0756, |
| "num_input_tokens_seen": 167000, |
| "step": 605 |
| }, |
| { |
| "epoch": 5.398230088495575, |
| "grad_norm": 12.41464900970459, |
| "learning_rate": 2.596509710718235e-05, |
| "loss": 0.0677, |
| "num_input_tokens_seen": 168456, |
| "step": 610 |
| }, |
| { |
| "epoch": 5.442477876106195, |
| "grad_norm": 0.03950650617480278, |
| "learning_rate": 2.5579150371332956e-05, |
| "loss": 0.002, |
| "num_input_tokens_seen": 169624, |
| "step": 615 |
| }, |
| { |
| "epoch": 5.486725663716814, |
| "grad_norm": 1.2885109186172485, |
| "learning_rate": 2.5193065476046955e-05, |
| "loss": 0.0027, |
| "num_input_tokens_seen": 170888, |
| "step": 620 |
| }, |
| { |
| "epoch": 5.530973451327434, |
| "grad_norm": 1.0878580808639526, |
| "learning_rate": 2.480693452395305e-05, |
| "loss": 0.0287, |
| "num_input_tokens_seen": 172056, |
| "step": 625 |
| }, |
| { |
| "epoch": 5.548672566371682, |
| "eval_loss": 0.08361060917377472, |
| "eval_runtime": 0.7732, |
| "eval_samples_per_second": 32.335, |
| "eval_steps_per_second": 16.814, |
| "num_input_tokens_seen": 172504, |
| "step": 627 |
| }, |
| { |
| "epoch": 5.575221238938053, |
| "grad_norm": 21.944652557373047, |
| "learning_rate": 2.4420849628667046e-05, |
| "loss": 0.2663, |
| "num_input_tokens_seen": 173448, |
| "step": 630 |
| }, |
| { |
| "epoch": 5.619469026548672, |
| "grad_norm": 0.16154882311820984, |
| "learning_rate": 2.4034902892817648e-05, |
| "loss": 0.1753, |
| "num_input_tokens_seen": 174872, |
| "step": 635 |
| }, |
| { |
| "epoch": 5.663716814159292, |
| "grad_norm": 0.21520113945007324, |
| "learning_rate": 2.3649186386074872e-05, |
| "loss": 0.0021, |
| "num_input_tokens_seen": 176488, |
| "step": 640 |
| }, |
| { |
| "epoch": 5.707964601769912, |
| "grad_norm": 0.9986706376075745, |
| "learning_rate": 2.3263792123186353e-05, |
| "loss": 0.0118, |
| "num_input_tokens_seen": 177816, |
| "step": 645 |
| }, |
| { |
| "epoch": 5.752212389380531, |
| "grad_norm": 14.675823211669922, |
| "learning_rate": 2.2878812042026713e-05, |
| "loss": 0.0679, |
| "num_input_tokens_seen": 179352, |
| "step": 650 |
| }, |
| { |
| "epoch": 5.79646017699115, |
| "grad_norm": 0.008065628819167614, |
| "learning_rate": 2.2494337981665304e-05, |
| "loss": 0.0226, |
| "num_input_tokens_seen": 180504, |
| "step": 655 |
| }, |
| { |
| "epoch": 5.84070796460177, |
| "grad_norm": 17.46634864807129, |
| "learning_rate": 2.2110461660457482e-05, |
| "loss": 0.161, |
| "num_input_tokens_seen": 181976, |
| "step": 660 |
| }, |
| { |
| "epoch": 5.88495575221239, |
| "grad_norm": 1.4565507173538208, |
| "learning_rate": 2.1727274654164728e-05, |
| "loss": 0.1769, |
| "num_input_tokens_seen": 183000, |
| "step": 665 |
| }, |
| { |
| "epoch": 5.929203539823009, |
| "grad_norm": 0.0984095111489296, |
| "learning_rate": 2.1344868374108732e-05, |
| "loss": 0.0023, |
| "num_input_tokens_seen": 184344, |
| "step": 670 |
| }, |
| { |
| "epoch": 5.9734513274336285, |
| "grad_norm": 9.466740608215332, |
| "learning_rate": 2.096333404536474e-05, |
| "loss": 0.0631, |
| "num_input_tokens_seen": 185976, |
| "step": 675 |
| }, |
| { |
| "epoch": 6.017699115044247, |
| "grad_norm": 0.01006875280290842, |
| "learning_rate": 2.0582762684999327e-05, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 187312, |
| "step": 680 |
| }, |
| { |
| "epoch": 6.053097345132743, |
| "eval_loss": 0.051821138709783554, |
| "eval_runtime": 0.7739, |
| "eval_samples_per_second": 32.302, |
| "eval_steps_per_second": 16.797, |
| "num_input_tokens_seen": 188160, |
| "step": 684 |
| }, |
| { |
| "epoch": 6.061946902654867, |
| "grad_norm": 11.252352714538574, |
| "learning_rate": 2.0203245080357775e-05, |
| "loss": 0.1977, |
| "num_input_tokens_seen": 188512, |
| "step": 685 |
| }, |
| { |
| "epoch": 6.106194690265487, |
| "grad_norm": 0.029097890481352806, |
| "learning_rate": 1.982487176740627e-05, |
| "loss": 0.009, |
| "num_input_tokens_seen": 189616, |
| "step": 690 |
| }, |
| { |
| "epoch": 6.150442477876107, |
| "grad_norm": 0.006874327547848225, |
| "learning_rate": 1.944773300913407e-05, |
| "loss": 0.0074, |
| "num_input_tokens_seen": 191264, |
| "step": 695 |
| }, |
| { |
| "epoch": 6.1946902654867255, |
| "grad_norm": 0.07622929662466049, |
| "learning_rate": 1.907191877402076e-05, |
| "loss": 0.0018, |
| "num_input_tokens_seen": 192576, |
| "step": 700 |
| }, |
| { |
| "epoch": 6.238938053097345, |
| "grad_norm": 0.07754877954721451, |
| "learning_rate": 1.8697518714573807e-05, |
| "loss": 0.0152, |
| "num_input_tokens_seen": 193920, |
| "step": 705 |
| }, |
| { |
| "epoch": 6.283185840707965, |
| "grad_norm": 0.005654957611113787, |
| "learning_rate": 1.832462214594146e-05, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 195216, |
| "step": 710 |
| }, |
| { |
| "epoch": 6.327433628318584, |
| "grad_norm": 0.14916832745075226, |
| "learning_rate": 1.7953318024606157e-05, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 196640, |
| "step": 715 |
| }, |
| { |
| "epoch": 6.371681415929204, |
| "grad_norm": 0.03626057133078575, |
| "learning_rate": 1.7583694927163453e-05, |
| "loss": 0.0401, |
| "num_input_tokens_seen": 197696, |
| "step": 720 |
| }, |
| { |
| "epoch": 6.415929203539823, |
| "grad_norm": 0.057499129325151443, |
| "learning_rate": 1.7215841029191616e-05, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 199040, |
| "step": 725 |
| }, |
| { |
| "epoch": 6.460176991150442, |
| "grad_norm": 0.2001536786556244, |
| "learning_rate": 1.6849844084216874e-05, |
| "loss": 0.0061, |
| "num_input_tokens_seen": 200928, |
| "step": 730 |
| }, |
| { |
| "epoch": 6.504424778761062, |
| "grad_norm": 0.045186158269643784, |
| "learning_rate": 1.648579140277931e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 202784, |
| "step": 735 |
| }, |
| { |
| "epoch": 6.548672566371682, |
| "grad_norm": 0.09090372174978256, |
| "learning_rate": 1.612376983160449e-05, |
| "loss": 0.0071, |
| "num_input_tokens_seen": 204080, |
| "step": 740 |
| }, |
| { |
| "epoch": 6.557522123893805, |
| "eval_loss": 0.012937678955495358, |
| "eval_runtime": 0.7734, |
| "eval_samples_per_second": 32.325, |
| "eval_steps_per_second": 16.809, |
| "num_input_tokens_seen": 204240, |
| "step": 741 |
| }, |
| { |
| "epoch": 6.592920353982301, |
| "grad_norm": 0.22665704786777496, |
| "learning_rate": 1.576386573288569e-05, |
| "loss": 0.0012, |
| "num_input_tokens_seen": 205296, |
| "step": 745 |
| }, |
| { |
| "epoch": 6.6371681415929205, |
| "grad_norm": 0.053931985050439835, |
| "learning_rate": 1.5406164963681774e-05, |
| "loss": 0.0009, |
| "num_input_tokens_seen": 206768, |
| "step": 750 |
| }, |
| { |
| "epoch": 6.68141592920354, |
| "grad_norm": 0.282073050737381, |
| "learning_rate": 1.5050752855435457e-05, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 208048, |
| "step": 755 |
| }, |
| { |
| "epoch": 6.725663716814159, |
| "grad_norm": 10.507343292236328, |
| "learning_rate": 1.4697714193617097e-05, |
| "loss": 0.0132, |
| "num_input_tokens_seen": 209200, |
| "step": 760 |
| }, |
| { |
| "epoch": 6.769911504424779, |
| "grad_norm": 0.005959831643849611, |
| "learning_rate": 1.4347133197498619e-05, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 210448, |
| "step": 765 |
| }, |
| { |
| "epoch": 6.814159292035399, |
| "grad_norm": 30.997230529785156, |
| "learning_rate": 1.3999093500062587e-05, |
| "loss": 0.2387, |
| "num_input_tokens_seen": 211984, |
| "step": 770 |
| }, |
| { |
| "epoch": 6.8584070796460175, |
| "grad_norm": 0.04386006295681, |
| "learning_rate": 1.3653678128051106e-05, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 213328, |
| "step": 775 |
| }, |
| { |
| "epoch": 6.902654867256637, |
| "grad_norm": 7.10620641708374, |
| "learning_rate": 1.3310969482159297e-05, |
| "loss": 0.0646, |
| "num_input_tokens_seen": 214400, |
| "step": 780 |
| }, |
| { |
| "epoch": 6.946902654867257, |
| "grad_norm": 0.07594583928585052, |
| "learning_rate": 1.2971049317378237e-05, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 215888, |
| "step": 785 |
| }, |
| { |
| "epoch": 6.991150442477876, |
| "grad_norm": 0.006478244438767433, |
| "learning_rate": 1.263399872349183e-05, |
| "loss": 0.003, |
| "num_input_tokens_seen": 217664, |
| "step": 790 |
| }, |
| { |
| "epoch": 7.035398230088496, |
| "grad_norm": 0.1744823157787323, |
| "learning_rate": 1.2299898105732385e-05, |
| "loss": 0.0039, |
| "num_input_tokens_seen": 218840, |
| "step": 795 |
| }, |
| { |
| "epoch": 7.061946902654867, |
| "eval_loss": 0.022299928590655327, |
| "eval_runtime": 0.777, |
| "eval_samples_per_second": 32.173, |
| "eval_steps_per_second": 16.73, |
| "num_input_tokens_seen": 219512, |
| "step": 798 |
| }, |
| { |
| "epoch": 7.079646017699115, |
| "grad_norm": 0.02160412073135376, |
| "learning_rate": 1.1968827165599564e-05, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 219992, |
| "step": 800 |
| }, |
| { |
| "epoch": 7.123893805309734, |
| "grad_norm": 0.010718903504312038, |
| "learning_rate": 1.1640864881847105e-05, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 221160, |
| "step": 805 |
| }, |
| { |
| "epoch": 7.168141592920354, |
| "grad_norm": 0.011407556012272835, |
| "learning_rate": 1.1316089491642076e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 222696, |
| "step": 810 |
| }, |
| { |
| "epoch": 7.212389380530974, |
| "grad_norm": 0.014685764908790588, |
| "learning_rate": 1.099457847190091e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 224088, |
| "step": 815 |
| }, |
| { |
| "epoch": 7.256637168141593, |
| "grad_norm": 0.02897009812295437, |
| "learning_rate": 1.067640852080693e-05, |
| "loss": 0.0605, |
| "num_input_tokens_seen": 225432, |
| "step": 820 |
| }, |
| { |
| "epoch": 7.300884955752212, |
| "grad_norm": 0.0066994777880609035, |
| "learning_rate": 1.0361655539513565e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 226696, |
| "step": 825 |
| }, |
| { |
| "epoch": 7.345132743362832, |
| "grad_norm": 0.007768039591610432, |
| "learning_rate": 1.0050394614037687e-05, |
| "loss": 0.001, |
| "num_input_tokens_seen": 228168, |
| "step": 830 |
| }, |
| { |
| "epoch": 7.389380530973451, |
| "grad_norm": 0.07913458347320557, |
| "learning_rate": 9.742699997347498e-06, |
| "loss": 0.0009, |
| "num_input_tokens_seen": 229528, |
| "step": 835 |
| }, |
| { |
| "epoch": 7.433628318584071, |
| "grad_norm": 0.011403764598071575, |
| "learning_rate": 9.438645091649029e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 231208, |
| "step": 840 |
| }, |
| { |
| "epoch": 7.477876106194691, |
| "grad_norm": 0.030279316008090973, |
| "learning_rate": 9.138302430875695e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 232392, |
| "step": 845 |
| }, |
| { |
| "epoch": 7.522123893805309, |
| "grad_norm": 0.01019184198230505, |
| "learning_rate": 8.841743663384913e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 233480, |
| "step": 850 |
| }, |
| { |
| "epoch": 7.566371681415929, |
| "grad_norm": 0.01505549531430006, |
| "learning_rate": 8.549039534865979e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 235256, |
| "step": 855 |
| }, |
| { |
| "epoch": 7.566371681415929, |
| "eval_loss": 0.021708844229578972, |
| "eval_runtime": 0.7819, |
| "eval_samples_per_second": 31.972, |
| "eval_steps_per_second": 16.625, |
| "num_input_tokens_seen": 235256, |
| "step": 855 |
| }, |
| { |
| "epoch": 7.610619469026549, |
| "grad_norm": 0.03689683601260185, |
| "learning_rate": 8.260259871463352e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 236744, |
| "step": 860 |
| }, |
| { |
| "epoch": 7.654867256637168, |
| "grad_norm": 0.0034927541855722666, |
| "learning_rate": 7.975473563119216e-06, |
| "loss": 0.0009, |
| "num_input_tokens_seen": 237880, |
| "step": 865 |
| }, |
| { |
| "epoch": 7.699115044247788, |
| "grad_norm": 0.6410696506500244, |
| "learning_rate": 7.69474854713943e-06, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 239544, |
| "step": 870 |
| }, |
| { |
| "epoch": 7.743362831858407, |
| "grad_norm": 0.01861269399523735, |
| "learning_rate": 7.4181517919866914e-06, |
| "loss": 0.0182, |
| "num_input_tokens_seen": 241048, |
| "step": 875 |
| }, |
| { |
| "epoch": 7.787610619469026, |
| "grad_norm": 0.003941065166145563, |
| "learning_rate": 7.1457492813049e-06, |
| "loss": 0.0022, |
| "num_input_tokens_seen": 242408, |
| "step": 880 |
| }, |
| { |
| "epoch": 7.831858407079646, |
| "grad_norm": 0.10631576925516129, |
| "learning_rate": 6.877605998178344e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 243688, |
| "step": 885 |
| }, |
| { |
| "epoch": 7.876106194690266, |
| "grad_norm": 0.20832419395446777, |
| "learning_rate": 6.613785909629686e-06, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 245320, |
| "step": 890 |
| }, |
| { |
| "epoch": 7.920353982300885, |
| "grad_norm": 0.004488417878746986, |
| "learning_rate": 6.354351951360268e-06, |
| "loss": 0.0023, |
| "num_input_tokens_seen": 246584, |
| "step": 895 |
| }, |
| { |
| "epoch": 7.964601769911504, |
| "grad_norm": 0.11372503638267517, |
| "learning_rate": 6.099366012736438e-06, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 248024, |
| "step": 900 |
| }, |
| { |
| "epoch": 8.008849557522124, |
| "grad_norm": 0.05825456231832504, |
| "learning_rate": 5.848888922025553e-06, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 249024, |
| "step": 905 |
| }, |
| { |
| "epoch": 8.053097345132743, |
| "grad_norm": 0.051966357976198196, |
| "learning_rate": 5.602980431885024e-06, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 250672, |
| "step": 910 |
| }, |
| { |
| "epoch": 8.070796460176991, |
| "eval_loss": 0.013118009082973003, |
| "eval_runtime": 0.7874, |
| "eval_samples_per_second": 31.752, |
| "eval_steps_per_second": 16.511, |
| "num_input_tokens_seen": 251392, |
| "step": 912 |
| }, |
| { |
| "epoch": 8.097345132743364, |
| "grad_norm": 0.005551676731556654, |
| "learning_rate": 5.361699205108042e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 252368, |
| "step": 915 |
| }, |
| { |
| "epoch": 8.141592920353983, |
| "grad_norm": 0.006533629726618528, |
| "learning_rate": 5.125102800629203e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 253920, |
| "step": 920 |
| }, |
| { |
| "epoch": 8.185840707964601, |
| "grad_norm": 0.007273109629750252, |
| "learning_rate": 4.893247659793568e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 255248, |
| "step": 925 |
| }, |
| { |
| "epoch": 8.230088495575222, |
| "grad_norm": 0.010464943014085293, |
| "learning_rate": 4.666189092892245e-06, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 256336, |
| "step": 930 |
| }, |
| { |
| "epoch": 8.274336283185841, |
| "grad_norm": 0.004114061594009399, |
| "learning_rate": 4.443981265967806e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 257648, |
| "step": 935 |
| }, |
| { |
| "epoch": 8.31858407079646, |
| "grad_norm": 0.09662500768899918, |
| "learning_rate": 4.226677187892739e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 258752, |
| "step": 940 |
| }, |
| { |
| "epoch": 8.36283185840708, |
| "grad_norm": 0.08983103185892105, |
| "learning_rate": 4.014328697723835e-06, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 260144, |
| "step": 945 |
| }, |
| { |
| "epoch": 8.4070796460177, |
| "grad_norm": 0.012256860733032227, |
| "learning_rate": 3.806986452335748e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 261328, |
| "step": 950 |
| }, |
| { |
| "epoch": 8.451327433628318, |
| "grad_norm": 0.022380976006388664, |
| "learning_rate": 3.604699914336515e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 262784, |
| "step": 955 |
| }, |
| { |
| "epoch": 8.495575221238939, |
| "grad_norm": 0.00660680141299963, |
| "learning_rate": 3.4075173402679574e-06, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 264448, |
| "step": 960 |
| }, |
| { |
| "epoch": 8.539823008849558, |
| "grad_norm": 0.9003037214279175, |
| "learning_rate": 3.215485769093862e-06, |
| "loss": 0.0022, |
| "num_input_tokens_seen": 265920, |
| "step": 965 |
| }, |
| { |
| "epoch": 8.575221238938052, |
| "eval_loss": 0.014585654251277447, |
| "eval_runtime": 0.7764, |
| "eval_samples_per_second": 32.201, |
| "eval_steps_per_second": 16.744, |
| "num_input_tokens_seen": 266784, |
| "step": 969 |
| }, |
| { |
| "epoch": 8.584070796460177, |
| "grad_norm": 0.025870338082313538, |
| "learning_rate": 3.028651010978556e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 266992, |
| "step": 970 |
| }, |
| { |
| "epoch": 8.628318584070797, |
| "grad_norm": 0.011923308484256268, |
| "learning_rate": 2.8470576363586633e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 268688, |
| "step": 975 |
| }, |
| { |
| "epoch": 8.672566371681416, |
| "grad_norm": 0.01803506724536419, |
| "learning_rate": 2.67074896531056e-06, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 270000, |
| "step": 980 |
| }, |
| { |
| "epoch": 8.716814159292035, |
| "grad_norm": 0.00940409954637289, |
| "learning_rate": 2.499767057216193e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 271056, |
| "step": 985 |
| }, |
| { |
| "epoch": 8.761061946902656, |
| "grad_norm": 0.028705783188343048, |
| "learning_rate": 2.334152700729511e-06, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 272608, |
| "step": 990 |
| }, |
| { |
| "epoch": 8.805309734513274, |
| "grad_norm": 0.013868856243789196, |
| "learning_rate": 2.1739454040461794e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 274000, |
| "step": 995 |
| }, |
| { |
| "epoch": 8.849557522123893, |
| "grad_norm": 0.015428607352077961, |
| "learning_rate": 2.019183385478654e-06, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 275200, |
| "step": 1000 |
| }, |
| { |
| "epoch": 8.893805309734514, |
| "grad_norm": 1.5666464567184448, |
| "learning_rate": 1.8699035643389928e-06, |
| "loss": 0.0031, |
| "num_input_tokens_seen": 276352, |
| "step": 1005 |
| }, |
| { |
| "epoch": 8.938053097345133, |
| "grad_norm": 0.10010976344347, |
| "learning_rate": 1.7261415521315798e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 277952, |
| "step": 1010 |
| }, |
| { |
| "epoch": 8.982300884955752, |
| "grad_norm": 0.005919489078223705, |
| "learning_rate": 1.587931644057783e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 279328, |
| "step": 1015 |
| }, |
| { |
| "epoch": 9.026548672566372, |
| "grad_norm": 0.00307848840020597, |
| "learning_rate": 1.455306810834678e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 280456, |
| "step": 1020 |
| }, |
| { |
| "epoch": 9.070796460176991, |
| "grad_norm": 0.015283796936273575, |
| "learning_rate": 1.3282986908296713e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 281752, |
| "step": 1025 |
| }, |
| { |
| "epoch": 9.079646017699115, |
| "eval_loss": 0.015441279858350754, |
| "eval_runtime": 0.7746, |
| "eval_samples_per_second": 32.275, |
| "eval_steps_per_second": 16.783, |
| "num_input_tokens_seen": 281976, |
| "step": 1026 |
| }, |
| { |
| "epoch": 9.11504424778761, |
| "grad_norm": 0.0041525596752762794, |
| "learning_rate": 1.2069375825130292e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 283464, |
| "step": 1030 |
| }, |
| { |
| "epoch": 9.15929203539823, |
| "grad_norm": 0.016132457181811333, |
| "learning_rate": 1.091252437230003e-06, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 284760, |
| "step": 1035 |
| }, |
| { |
| "epoch": 9.20353982300885, |
| "grad_norm": 0.01380851399153471, |
| "learning_rate": 9.812708522943164e-07, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 286104, |
| "step": 1040 |
| }, |
| { |
| "epoch": 9.247787610619469, |
| "grad_norm": 0.005584151484072208, |
| "learning_rate": 8.770190644047077e-07, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 287672, |
| "step": 1045 |
| }, |
| { |
| "epoch": 9.29203539823009, |
| "grad_norm": 0.09154026955366135, |
| "learning_rate": 7.785219433859847e-07, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 288920, |
| "step": 1050 |
| }, |
| { |
| "epoch": 9.336283185840708, |
| "grad_norm": 0.002845821203663945, |
| "learning_rate": 6.858029862562343e-07, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 289960, |
| "step": 1055 |
| }, |
| { |
| "epoch": 9.380530973451327, |
| "grad_norm": 0.013985905796289444, |
| "learning_rate": 5.988843116214616e-07, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 291544, |
| "step": 1060 |
| }, |
| { |
| "epoch": 9.424778761061948, |
| "grad_norm": 0.029856637120246887, |
| "learning_rate": 5.177866543990689e-07, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 292808, |
| "step": 1065 |
| }, |
| { |
| "epoch": 9.469026548672566, |
| "grad_norm": 0.05611226707696915, |
| "learning_rate": 4.425293608714548e-07, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 293912, |
| "step": 1070 |
| }, |
| { |
| "epoch": 9.513274336283185, |
| "grad_norm": 0.046232786029577255, |
| "learning_rate": 3.7313038407083003e-07, |
| "loss": 0.0023, |
| "num_input_tokens_seen": 295416, |
| "step": 1075 |
| }, |
| { |
| "epoch": 9.557522123893806, |
| "grad_norm": 0.007815594784915447, |
| "learning_rate": 3.0960627949644107e-07, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 297016, |
| "step": 1080 |
| }, |
| { |
| "epoch": 9.584070796460177, |
| "eval_loss": 0.016118336468935013, |
| "eval_runtime": 0.7771, |
| "eval_samples_per_second": 32.172, |
| "eval_steps_per_second": 16.729, |
| "num_input_tokens_seen": 297688, |
| "step": 1083 |
| }, |
| { |
| "epoch": 9.601769911504425, |
| "grad_norm": 0.003943083342164755, |
| "learning_rate": 2.5197220116515987e-07, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 298168, |
| "step": 1085 |
| }, |
| { |
| "epoch": 9.646017699115044, |
| "grad_norm": 0.03541100025177002, |
| "learning_rate": 2.0024189799641435e-07, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 299672, |
| "step": 1090 |
| }, |
| { |
| "epoch": 9.690265486725664, |
| "grad_norm": 0.011666457168757915, |
| "learning_rate": 1.5442771053230665e-07, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 301080, |
| "step": 1095 |
| }, |
| { |
| "epoch": 9.734513274336283, |
| "grad_norm": 0.012163754552602768, |
| "learning_rate": 1.1454056799370672e-07, |
| "loss": 0.0008, |
| "num_input_tokens_seen": 302472, |
| "step": 1100 |
| }, |
| { |
| "epoch": 9.778761061946902, |
| "grad_norm": 0.01618136093020439, |
| "learning_rate": 8.058998567304633e-08, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 303592, |
| "step": 1105 |
| }, |
| { |
| "epoch": 9.823008849557523, |
| "grad_norm": 0.0019596414640545845, |
| "learning_rate": 5.2584062664368105e-08, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 305224, |
| "step": 1110 |
| }, |
| { |
| "epoch": 9.867256637168142, |
| "grad_norm": 0.026996349915862083, |
| "learning_rate": 3.0529479931265535e-08, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 306200, |
| "step": 1115 |
| }, |
| { |
| "epoch": 9.91150442477876, |
| "grad_norm": 0.019850771874189377, |
| "learning_rate": 1.443149871309113e-08, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 307880, |
| "step": 1120 |
| }, |
| { |
| "epoch": 9.955752212389381, |
| "grad_norm": 0.00854608416557312, |
| "learning_rate": 4.293959269863201e-09, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 309240, |
| "step": 1125 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.010417921468615532, |
| "learning_rate": 1.192799661514643e-10, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 310504, |
| "step": 1130 |
| }, |
| { |
| "epoch": 10.0, |
| "num_input_tokens_seen": 310504, |
| "step": 1130, |
| "total_flos": 1.3981854782128128e+16, |
| "train_loss": 0.2885503941710855, |
| "train_runtime": 156.0359, |
| "train_samples_per_second": 14.42, |
| "train_steps_per_second": 7.242 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 1130, |
| "num_input_tokens_seen": 310504, |
| "num_train_epochs": 10, |
| "save_steps": 57, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.3981854782128128e+16, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
