| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 10.0, |
| "global_step": 3950, |
| "is_hyper_param_search": true, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0, |
| "learning_rate": 1.2658227848101266e-07, |
| "loss": 0.6546, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.63, |
| "learning_rate": 3.1518987341772155e-05, |
| "loss": 0.4879, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.10857142857142857, |
| "eval_f1": 0.07811852465810196, |
| "eval_loss": 0.40317419171333313, |
| "eval_precision": 0.17927631578947367, |
| "eval_recall": 0.08782679738562092, |
| "eval_runtime": 0.3221, |
| "eval_samples_per_second": 543.295, |
| "eval_steps_per_second": 68.3, |
| "step": 395 |
| }, |
| { |
| "epoch": 1.27, |
| "learning_rate": 4.9894490513601396e-05, |
| "loss": 0.4006, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.9, |
| "learning_rate": 4.878663257984936e-05, |
| "loss": 0.3759, |
| "step": 750 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.17714285714285713, |
| "eval_f1": 0.26697819927843125, |
| "eval_loss": 0.37731701135635376, |
| "eval_precision": 0.5218837535014006, |
| "eval_recall": 0.2022453723084196, |
| "eval_runtime": 0.286, |
| "eval_samples_per_second": 611.901, |
| "eval_steps_per_second": 76.925, |
| "step": 790 |
| }, |
| { |
| "epoch": 2.53, |
| "learning_rate": 4.652248579391458e-05, |
| "loss": 0.2948, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.24571428571428572, |
| "eval_f1": 0.36607677500846103, |
| "eval_loss": 0.39780309796333313, |
| "eval_precision": 0.5472431077694235, |
| "eval_recall": 0.3120779820710713, |
| "eval_runtime": 0.3071, |
| "eval_samples_per_second": 569.892, |
| "eval_steps_per_second": 71.644, |
| "step": 1185 |
| }, |
| { |
| "epoch": 3.16, |
| "learning_rate": 4.3212112295929505e-05, |
| "loss": 0.2649, |
| "step": 1250 |
| }, |
| { |
| "epoch": 3.8, |
| "learning_rate": 3.901643221443403e-05, |
| "loss": 0.1864, |
| "step": 1500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.2742857142857143, |
| "eval_f1": 0.4266986709033833, |
| "eval_loss": 0.4376567602157593, |
| "eval_precision": 0.5016284584009042, |
| "eval_recall": 0.38560897951242346, |
| "eval_runtime": 0.2998, |
| "eval_samples_per_second": 583.659, |
| "eval_steps_per_second": 73.374, |
| "step": 1580 |
| }, |
| { |
| "epoch": 4.43, |
| "learning_rate": 3.413940119927783e-05, |
| "loss": 0.1384, |
| "step": 1750 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.2342857142857143, |
| "eval_f1": 0.39448414058744574, |
| "eval_loss": 0.5389729738235474, |
| "eval_precision": 0.48297759387155675, |
| "eval_recall": 0.37081969717487395, |
| "eval_runtime": 0.3757, |
| "eval_samples_per_second": 465.771, |
| "eval_steps_per_second": 58.554, |
| "step": 1975 |
| }, |
| { |
| "epoch": 5.06, |
| "learning_rate": 2.8818095960484663e-05, |
| "loss": 0.111, |
| "step": 2000 |
| }, |
| { |
| "epoch": 5.7, |
| "learning_rate": 2.3311189763638598e-05, |
| "loss": 0.0735, |
| "step": 2250 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.24571428571428572, |
| "eval_f1": 0.46370146148440794, |
| "eval_loss": 0.5818756222724915, |
| "eval_precision": 0.5038962453668336, |
| "eval_recall": 0.4432562275600327, |
| "eval_runtime": 0.2959, |
| "eval_samples_per_second": 591.513, |
| "eval_steps_per_second": 74.362, |
| "step": 2370 |
| }, |
| { |
| "epoch": 6.33, |
| "learning_rate": 1.788637809796257e-05, |
| "loss": 0.0593, |
| "step": 2500 |
| }, |
| { |
| "epoch": 6.96, |
| "learning_rate": 1.2807365766276206e-05, |
| "loss": 0.0417, |
| "step": 2750 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.2571428571428571, |
| "eval_f1": 0.45537043929039545, |
| "eval_loss": 0.6109325289726257, |
| "eval_precision": 0.5327423959044721, |
| "eval_recall": 0.41917933979084293, |
| "eval_runtime": 0.2925, |
| "eval_samples_per_second": 598.344, |
| "eval_steps_per_second": 75.22, |
| "step": 2765 |
| }, |
| { |
| "epoch": 7.59, |
| "learning_rate": 8.321047965682896e-06, |
| "loss": 0.0247, |
| "step": 3000 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.26857142857142857, |
| "eval_f1": 0.44079821839093025, |
| "eval_loss": 0.6337146162986755, |
| "eval_precision": 0.4776951058201059, |
| "eval_recall": 0.42734534657817325, |
| "eval_runtime": 0.3404, |
| "eval_samples_per_second": 514.096, |
| "eval_steps_per_second": 64.629, |
| "step": 3160 |
| }, |
| { |
| "epoch": 8.23, |
| "learning_rate": 4.645508497778378e-06, |
| "loss": 0.0209, |
| "step": 3250 |
| }, |
| { |
| "epoch": 8.86, |
| "learning_rate": 1.9594185257981066e-06, |
| "loss": 0.0159, |
| "step": 3500 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.2342857142857143, |
| "eval_f1": 0.4276706487492863, |
| "eval_loss": 0.6485006809234619, |
| "eval_precision": 0.4811818562937489, |
| "eval_recall": 0.40231404333907983, |
| "eval_runtime": 0.3206, |
| "eval_samples_per_second": 545.875, |
| "eval_steps_per_second": 68.624, |
| "step": 3555 |
| }, |
| { |
| "epoch": 9.49, |
| "learning_rate": 3.933512143182022e-07, |
| "loss": 0.0147, |
| "step": 3750 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.2571428571428571, |
| "eval_f1": 0.4350893581328566, |
| "eval_loss": 0.6474436521530151, |
| "eval_precision": 0.49936246351687524, |
| "eval_recall": 0.40231404333907983, |
| "eval_runtime": 0.3074, |
| "eval_samples_per_second": 569.268, |
| "eval_steps_per_second": 71.565, |
| "step": 3950 |
| }, |
| { |
| "epoch": 10.0, |
| "step": 3950, |
| "total_flos": 200561037492224.0, |
| "train_loss": 0.0, |
| "train_runtime": 0.0071, |
| "train_samples_per_second": 4470373.653, |
| "train_steps_per_second": 559504.941 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.2571428571428571, |
| "eval_f1": 0.4350893581328566, |
| "eval_loss": 0.6474436521530151, |
| "eval_precision": 0.49936246351687524, |
| "eval_recall": 0.40231404333907983, |
| "eval_runtime": 0.3645, |
| "eval_samples_per_second": 480.122, |
| "eval_steps_per_second": 60.358, |
| "step": 3950 |
| } |
| ], |
| "max_steps": 3950, |
| "num_train_epochs": 10, |
| "total_flos": 200561037492224.0, |
| "trial_name": null, |
| "trial_params": { |
| "learning_rate": 5e-05 |
| } |
| } |
|
|
