| { |
| "best_metric": 3.749399185180664, |
| "best_model_checkpoint": "models/GPT2_bigram_function_within_boundary_53/checkpoint-64200", |
| "epoch": 10.0, |
| "global_step": 64200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.16, |
| "learning_rate": 1e-05, |
| "loss": 7.8094, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 2e-05, |
| "loss": 6.0912, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.47, |
| "learning_rate": 3e-05, |
| "loss": 5.6446, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 4e-05, |
| "loss": 5.3458, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.78, |
| "learning_rate": 5e-05, |
| "loss": 5.1168, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.93, |
| "learning_rate": 6e-05, |
| "loss": 4.9279, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.26604163886293875, |
| "eval_loss": 4.7261810302734375, |
| "eval_runtime": 1.8805, |
| "eval_samples_per_second": 627.498, |
| "eval_steps_per_second": 5.318, |
| "step": 6420 |
| }, |
| { |
| "epoch": 1.09, |
| "learning_rate": 7e-05, |
| "loss": 4.7693, |
| "step": 7000 |
| }, |
| { |
| "epoch": 1.25, |
| "learning_rate": 8e-05, |
| "loss": 4.645, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.4, |
| "learning_rate": 9e-05, |
| "loss": 4.549, |
| "step": 9000 |
| }, |
| { |
| "epoch": 1.56, |
| "learning_rate": 0.0001, |
| "loss": 4.4691, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.71, |
| "learning_rate": 9.81568265682657e-05, |
| "loss": 4.4002, |
| "step": 11000 |
| }, |
| { |
| "epoch": 1.87, |
| "learning_rate": 9.631180811808118e-05, |
| "loss": 4.3403, |
| "step": 12000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.31247164019751766, |
| "eval_loss": 4.217563152313232, |
| "eval_runtime": 1.9076, |
| "eval_samples_per_second": 618.593, |
| "eval_steps_per_second": 5.242, |
| "step": 12840 |
| }, |
| { |
| "epoch": 2.02, |
| "learning_rate": 9.446863468634687e-05, |
| "loss": 4.284, |
| "step": 13000 |
| }, |
| { |
| "epoch": 2.18, |
| "learning_rate": 9.262361623616237e-05, |
| "loss": 4.214, |
| "step": 14000 |
| }, |
| { |
| "epoch": 2.34, |
| "learning_rate": 9.078044280442805e-05, |
| "loss": 4.186, |
| "step": 15000 |
| }, |
| { |
| "epoch": 2.49, |
| "learning_rate": 8.893542435424355e-05, |
| "loss": 4.1596, |
| "step": 16000 |
| }, |
| { |
| "epoch": 2.65, |
| "learning_rate": 8.709225092250924e-05, |
| "loss": 4.1315, |
| "step": 17000 |
| }, |
| { |
| "epoch": 2.8, |
| "learning_rate": 8.524723247232472e-05, |
| "loss": 4.1101, |
| "step": 18000 |
| }, |
| { |
| "epoch": 2.96, |
| "learning_rate": 8.340405904059041e-05, |
| "loss": 4.0904, |
| "step": 19000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.33630722007206726, |
| "eval_loss": 4.028186321258545, |
| "eval_runtime": 1.8998, |
| "eval_samples_per_second": 621.118, |
| "eval_steps_per_second": 5.264, |
| "step": 19260 |
| }, |
| { |
| "epoch": 3.12, |
| "learning_rate": 8.155904059040591e-05, |
| "loss": 4.0288, |
| "step": 20000 |
| }, |
| { |
| "epoch": 3.27, |
| "learning_rate": 7.971586715867159e-05, |
| "loss": 4.0076, |
| "step": 21000 |
| }, |
| { |
| "epoch": 3.43, |
| "learning_rate": 7.787084870848709e-05, |
| "loss": 3.9975, |
| "step": 22000 |
| }, |
| { |
| "epoch": 3.58, |
| "learning_rate": 7.602952029520296e-05, |
| "loss": 3.9856, |
| "step": 23000 |
| }, |
| { |
| "epoch": 3.74, |
| "learning_rate": 7.418450184501846e-05, |
| "loss": 3.9768, |
| "step": 24000 |
| }, |
| { |
| "epoch": 3.89, |
| "learning_rate": 7.233948339483395e-05, |
| "loss": 3.9677, |
| "step": 25000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.3478379821166422, |
| "eval_loss": 3.927978515625, |
| "eval_runtime": 1.8804, |
| "eval_samples_per_second": 627.52, |
| "eval_steps_per_second": 5.318, |
| "step": 25680 |
| }, |
| { |
| "epoch": 4.05, |
| "learning_rate": 7.049446494464945e-05, |
| "loss": 3.9334, |
| "step": 26000 |
| }, |
| { |
| "epoch": 4.21, |
| "learning_rate": 6.865129151291513e-05, |
| "loss": 3.8891, |
| "step": 27000 |
| }, |
| { |
| "epoch": 4.36, |
| "learning_rate": 6.680627306273063e-05, |
| "loss": 3.8918, |
| "step": 28000 |
| }, |
| { |
| "epoch": 4.52, |
| "learning_rate": 6.496309963099632e-05, |
| "loss": 3.8852, |
| "step": 29000 |
| }, |
| { |
| "epoch": 4.67, |
| "learning_rate": 6.3119926199262e-05, |
| "loss": 3.8829, |
| "step": 30000 |
| }, |
| { |
| "epoch": 4.83, |
| "learning_rate": 6.12749077490775e-05, |
| "loss": 3.8762, |
| "step": 31000 |
| }, |
| { |
| "epoch": 4.98, |
| "learning_rate": 5.942988929889299e-05, |
| "loss": 3.8705, |
| "step": 32000 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.35631255838782866, |
| "eval_loss": 3.868091583251953, |
| "eval_runtime": 1.8984, |
| "eval_samples_per_second": 621.581, |
| "eval_steps_per_second": 5.268, |
| "step": 32100 |
| }, |
| { |
| "epoch": 5.14, |
| "learning_rate": 5.7584870848708496e-05, |
| "loss": 3.8072, |
| "step": 33000 |
| }, |
| { |
| "epoch": 5.3, |
| "learning_rate": 5.574169741697417e-05, |
| "loss": 3.8096, |
| "step": 34000 |
| }, |
| { |
| "epoch": 5.45, |
| "learning_rate": 5.389667896678967e-05, |
| "loss": 3.8074, |
| "step": 35000 |
| }, |
| { |
| "epoch": 5.61, |
| "learning_rate": 5.205350553505536e-05, |
| "loss": 3.8056, |
| "step": 36000 |
| }, |
| { |
| "epoch": 5.76, |
| "learning_rate": 5.020848708487085e-05, |
| "loss": 3.805, |
| "step": 37000 |
| }, |
| { |
| "epoch": 5.92, |
| "learning_rate": 4.836531365313653e-05, |
| "loss": 3.8011, |
| "step": 38000 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.36181769651674894, |
| "eval_loss": 3.8269381523132324, |
| "eval_runtime": 1.8825, |
| "eval_samples_per_second": 626.843, |
| "eval_steps_per_second": 5.312, |
| "step": 38520 |
| }, |
| { |
| "epoch": 6.07, |
| "learning_rate": 4.652029520295203e-05, |
| "loss": 3.7663, |
| "step": 39000 |
| }, |
| { |
| "epoch": 6.23, |
| "learning_rate": 4.4677121771217715e-05, |
| "loss": 3.739, |
| "step": 40000 |
| }, |
| { |
| "epoch": 6.39, |
| "learning_rate": 4.283210332103321e-05, |
| "loss": 3.7415, |
| "step": 41000 |
| }, |
| { |
| "epoch": 6.54, |
| "learning_rate": 4.099077490774908e-05, |
| "loss": 3.7448, |
| "step": 42000 |
| }, |
| { |
| "epoch": 6.7, |
| "learning_rate": 3.9145756457564576e-05, |
| "loss": 3.7444, |
| "step": 43000 |
| }, |
| { |
| "epoch": 6.85, |
| "learning_rate": 3.7300738007380073e-05, |
| "loss": 3.7431, |
| "step": 44000 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.3668423862271453, |
| "eval_loss": 3.7945444583892822, |
| "eval_runtime": 1.8941, |
| "eval_samples_per_second": 622.99, |
| "eval_steps_per_second": 5.28, |
| "step": 44940 |
| }, |
| { |
| "epoch": 7.01, |
| "learning_rate": 3.545571955719557e-05, |
| "loss": 3.737, |
| "step": 45000 |
| }, |
| { |
| "epoch": 7.17, |
| "learning_rate": 3.3612546125461256e-05, |
| "loss": 3.6807, |
| "step": 46000 |
| }, |
| { |
| "epoch": 7.32, |
| "learning_rate": 3.1767527675276754e-05, |
| "loss": 3.6881, |
| "step": 47000 |
| }, |
| { |
| "epoch": 7.48, |
| "learning_rate": 2.9924354243542435e-05, |
| "loss": 3.6913, |
| "step": 48000 |
| }, |
| { |
| "epoch": 7.63, |
| "learning_rate": 2.8079335793357936e-05, |
| "loss": 3.6903, |
| "step": 49000 |
| }, |
| { |
| "epoch": 7.79, |
| "learning_rate": 2.6236162361623618e-05, |
| "loss": 3.6902, |
| "step": 50000 |
| }, |
| { |
| "epoch": 7.94, |
| "learning_rate": 2.4391143911439115e-05, |
| "loss": 3.6902, |
| "step": 51000 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.37069931936474043, |
| "eval_loss": 3.7720694541931152, |
| "eval_runtime": 1.8866, |
| "eval_samples_per_second": 625.451, |
| "eval_steps_per_second": 5.3, |
| "step": 51360 |
| }, |
| { |
| "epoch": 8.1, |
| "learning_rate": 2.2547970479704797e-05, |
| "loss": 3.655, |
| "step": 52000 |
| }, |
| { |
| "epoch": 8.26, |
| "learning_rate": 2.0702952029520294e-05, |
| "loss": 3.6396, |
| "step": 53000 |
| }, |
| { |
| "epoch": 8.41, |
| "learning_rate": 1.885977859778598e-05, |
| "loss": 3.6443, |
| "step": 54000 |
| }, |
| { |
| "epoch": 8.57, |
| "learning_rate": 1.7014760147601477e-05, |
| "loss": 3.6434, |
| "step": 55000 |
| }, |
| { |
| "epoch": 8.72, |
| "learning_rate": 1.5171586715867158e-05, |
| "loss": 3.6438, |
| "step": 56000 |
| }, |
| { |
| "epoch": 8.88, |
| "learning_rate": 1.3326568265682656e-05, |
| "loss": 3.644, |
| "step": 57000 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.37262778593353796, |
| "eval_loss": 3.7567532062530518, |
| "eval_runtime": 1.8785, |
| "eval_samples_per_second": 628.152, |
| "eval_steps_per_second": 5.323, |
| "step": 57780 |
| }, |
| { |
| "epoch": 9.03, |
| "learning_rate": 1.148339483394834e-05, |
| "loss": 3.634, |
| "step": 58000 |
| }, |
| { |
| "epoch": 9.19, |
| "learning_rate": 9.638376383763839e-06, |
| "loss": 3.6046, |
| "step": 59000 |
| }, |
| { |
| "epoch": 9.35, |
| "learning_rate": 7.79520295202952e-06, |
| "loss": 3.6045, |
| "step": 60000 |
| }, |
| { |
| "epoch": 9.5, |
| "learning_rate": 5.9501845018450185e-06, |
| "loss": 3.6066, |
| "step": 61000 |
| }, |
| { |
| "epoch": 9.66, |
| "learning_rate": 4.107011070110702e-06, |
| "loss": 3.6051, |
| "step": 62000 |
| }, |
| { |
| "epoch": 9.81, |
| "learning_rate": 2.2619926199261997e-06, |
| "loss": 3.6038, |
| "step": 63000 |
| }, |
| { |
| "epoch": 9.97, |
| "learning_rate": 4.1881918819188193e-07, |
| "loss": 3.6026, |
| "step": 64000 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.373848925663953, |
| "eval_loss": 3.749399185180664, |
| "eval_runtime": 1.9384, |
| "eval_samples_per_second": 608.743, |
| "eval_steps_per_second": 5.159, |
| "step": 64200 |
| }, |
| { |
| "epoch": 10.0, |
| "step": 64200, |
| "total_flos": 5.3679573762048e+17, |
| "train_loss": 4.073395565425124, |
| "train_runtime": 28330.1372, |
| "train_samples_per_second": 290.064, |
| "train_steps_per_second": 2.266 |
| } |
| ], |
| "max_steps": 64200, |
| "num_train_epochs": 10, |
| "total_flos": 5.3679573762048e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
