| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.46153846153846156, | |
| "eval_steps": 10, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006153846153846154, | |
| "eval_loss": 1.489196538925171, | |
| "eval_runtime": 1.2747, | |
| "eval_samples_per_second": 89.432, | |
| "eval_steps_per_second": 4.707, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03076923076923077, | |
| "grad_norm": 17.32219123840332, | |
| "learning_rate": 8.771929824561404e-07, | |
| "loss": 1.5722, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03076923076923077, | |
| "eval_loss": 1.3865292072296143, | |
| "eval_runtime": 1.1749, | |
| "eval_samples_per_second": 97.029, | |
| "eval_steps_per_second": 5.107, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06153846153846154, | |
| "grad_norm": 21.88077735900879, | |
| "learning_rate": 1.7543859649122807e-06, | |
| "loss": 1.3935, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06153846153846154, | |
| "eval_loss": 1.0996264219284058, | |
| "eval_runtime": 1.1706, | |
| "eval_samples_per_second": 97.386, | |
| "eval_steps_per_second": 5.126, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09230769230769231, | |
| "grad_norm": 17.395177841186523, | |
| "learning_rate": 2.631578947368421e-06, | |
| "loss": 1.0664, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.09230769230769231, | |
| "eval_loss": 0.8006665706634521, | |
| "eval_runtime": 1.1747, | |
| "eval_samples_per_second": 97.044, | |
| "eval_steps_per_second": 5.108, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12307692307692308, | |
| "grad_norm": 6.869284152984619, | |
| "learning_rate": 3.5087719298245615e-06, | |
| "loss": 0.7994, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.12307692307692308, | |
| "eval_loss": 0.708875298500061, | |
| "eval_runtime": 1.1742, | |
| "eval_samples_per_second": 97.087, | |
| "eval_steps_per_second": 5.11, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 6.4737162590026855, | |
| "learning_rate": 4.385964912280702e-06, | |
| "loss": 0.7751, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "eval_loss": 0.6778659224510193, | |
| "eval_runtime": 1.1722, | |
| "eval_samples_per_second": 97.257, | |
| "eval_steps_per_second": 5.119, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18461538461538463, | |
| "grad_norm": 5.418182373046875, | |
| "learning_rate": 5.263157894736842e-06, | |
| "loss": 0.6203, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18461538461538463, | |
| "eval_loss": 0.6695793271064758, | |
| "eval_runtime": 1.1744, | |
| "eval_samples_per_second": 97.071, | |
| "eval_steps_per_second": 5.109, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2153846153846154, | |
| "grad_norm": 2.7423934936523438, | |
| "learning_rate": 6.140350877192983e-06, | |
| "loss": 0.767, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2153846153846154, | |
| "eval_loss": 0.6650205850601196, | |
| "eval_runtime": 1.1671, | |
| "eval_samples_per_second": 97.675, | |
| "eval_steps_per_second": 5.141, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24615384615384617, | |
| "grad_norm": 2.572129726409912, | |
| "learning_rate": 7.017543859649123e-06, | |
| "loss": 0.6336, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.24615384615384617, | |
| "eval_loss": 0.6613836884498596, | |
| "eval_runtime": 1.1671, | |
| "eval_samples_per_second": 97.676, | |
| "eval_steps_per_second": 5.141, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.27692307692307694, | |
| "grad_norm": 4.622325420379639, | |
| "learning_rate": 7.894736842105265e-06, | |
| "loss": 0.631, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.27692307692307694, | |
| "eval_loss": 0.6617783904075623, | |
| "eval_runtime": 1.1744, | |
| "eval_samples_per_second": 97.071, | |
| "eval_steps_per_second": 5.109, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 3.24354887008667, | |
| "learning_rate": 8.771929824561405e-06, | |
| "loss": 0.6086, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "eval_loss": 0.662232518196106, | |
| "eval_runtime": 1.172, | |
| "eval_samples_per_second": 97.27, | |
| "eval_steps_per_second": 5.119, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3384615384615385, | |
| "grad_norm": 3.5868513584136963, | |
| "learning_rate": 9.649122807017545e-06, | |
| "loss": 0.7057, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3384615384615385, | |
| "eval_loss": 0.6619779467582703, | |
| "eval_runtime": 1.165, | |
| "eval_samples_per_second": 97.856, | |
| "eval_steps_per_second": 5.15, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.36923076923076925, | |
| "grad_norm": 3.4894511699676514, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.7385, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.36923076923076925, | |
| "eval_loss": 0.6673641800880432, | |
| "eval_runtime": 1.1682, | |
| "eval_samples_per_second": 97.588, | |
| "eval_steps_per_second": 5.136, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 5.117406845092773, | |
| "learning_rate": 1.1403508771929826e-05, | |
| "loss": 0.6533, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.6705042123794556, | |
| "eval_runtime": 1.1726, | |
| "eval_samples_per_second": 97.217, | |
| "eval_steps_per_second": 5.117, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4307692307692308, | |
| "grad_norm": 3.9341437816619873, | |
| "learning_rate": 1.2280701754385966e-05, | |
| "loss": 0.7066, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4307692307692308, | |
| "eval_loss": 0.6691470146179199, | |
| "eval_runtime": 1.1702, | |
| "eval_samples_per_second": 97.418, | |
| "eval_steps_per_second": 5.127, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 3.4131579399108887, | |
| "learning_rate": 1.3157894736842108e-05, | |
| "loss": 0.6065, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "eval_loss": 0.6724759340286255, | |
| "eval_runtime": 1.1746, | |
| "eval_samples_per_second": 97.05, | |
| "eval_steps_per_second": 5.108, | |
| "step": 150 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2275, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 50, | |
| "total_flos": 9451377658953728.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |