| { | |
| "best_metric": 0.8307240704500979, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-21/checkpoint-6894", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 7660, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.5445048809051514, | |
| "learning_rate": 0.0001354005618154073, | |
| "loss": 0.4902, | |
| "step": 766 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7710371819960861, | |
| "eval_f1": 0.7943760984182776, | |
| "eval_loss": 0.4425993263721466, | |
| "eval_precision": 0.7208931419457735, | |
| "eval_recall": 0.8845401174168297, | |
| "eval_runtime": 133.7206, | |
| "eval_samples_per_second": 7.643, | |
| "eval_steps_per_second": 1.914, | |
| "step": 766 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.882767677307129, | |
| "learning_rate": 0.00012035605494702872, | |
| "loss": 0.4409, | |
| "step": 1532 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8111545988258317, | |
| "eval_f1": 0.8360237892948172, | |
| "eval_loss": 0.41539597511291504, | |
| "eval_precision": 0.7387387387387387, | |
| "eval_recall": 0.9628180039138943, | |
| "eval_runtime": 133.4678, | |
| "eval_samples_per_second": 7.657, | |
| "eval_steps_per_second": 1.918, | |
| "step": 1532 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.518984794616699, | |
| "learning_rate": 0.00010531154807865013, | |
| "loss": 0.4251, | |
| "step": 2298 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8180039138943248, | |
| "eval_f1": 0.8264925373134328, | |
| "eval_loss": 0.42979857325553894, | |
| "eval_precision": 0.7896613190730838, | |
| "eval_recall": 0.8669275929549902, | |
| "eval_runtime": 132.8888, | |
| "eval_samples_per_second": 7.691, | |
| "eval_steps_per_second": 1.926, | |
| "step": 2298 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.3853329420089722, | |
| "learning_rate": 9.026704121027153e-05, | |
| "loss": 0.4192, | |
| "step": 3064 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8189823874755382, | |
| "eval_f1": 0.8269410664172124, | |
| "eval_loss": 0.4213884174823761, | |
| "eval_precision": 0.7921146953405018, | |
| "eval_recall": 0.8649706457925636, | |
| "eval_runtime": 133.0333, | |
| "eval_samples_per_second": 7.682, | |
| "eval_steps_per_second": 1.924, | |
| "step": 3064 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 16.2620792388916, | |
| "learning_rate": 7.522253434189295e-05, | |
| "loss": 0.4053, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8228962818003914, | |
| "eval_f1": 0.8300469483568074, | |
| "eval_loss": 0.4430813789367676, | |
| "eval_precision": 0.7978339350180506, | |
| "eval_recall": 0.8649706457925636, | |
| "eval_runtime": 133.6623, | |
| "eval_samples_per_second": 7.646, | |
| "eval_steps_per_second": 1.915, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.759765386581421, | |
| "learning_rate": 6.017802747351436e-05, | |
| "loss": 0.4021, | |
| "step": 4596 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8189823874755382, | |
| "eval_f1": 0.8316651501364878, | |
| "eval_loss": 0.43012145161628723, | |
| "eval_precision": 0.7772108843537415, | |
| "eval_recall": 0.8943248532289628, | |
| "eval_runtime": 133.3962, | |
| "eval_samples_per_second": 7.661, | |
| "eval_steps_per_second": 1.919, | |
| "step": 4596 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.6848570108413696, | |
| "learning_rate": 4.5133520605135766e-05, | |
| "loss": 0.3933, | |
| "step": 5362 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8199608610567515, | |
| "eval_f1": 0.8362989323843416, | |
| "eval_loss": 0.4251769185066223, | |
| "eval_precision": 0.766721044045677, | |
| "eval_recall": 0.9197651663405088, | |
| "eval_runtime": 133.0713, | |
| "eval_samples_per_second": 7.68, | |
| "eval_steps_per_second": 1.924, | |
| "step": 5362 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.3459274768829346, | |
| "learning_rate": 3.008901373675718e-05, | |
| "loss": 0.3864, | |
| "step": 6128 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8199608610567515, | |
| "eval_f1": 0.8351254480286738, | |
| "eval_loss": 0.42666804790496826, | |
| "eval_precision": 0.7702479338842976, | |
| "eval_recall": 0.9119373776908023, | |
| "eval_runtime": 131.53, | |
| "eval_samples_per_second": 7.77, | |
| "eval_steps_per_second": 1.946, | |
| "step": 6128 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.43761202692985535, | |
| "learning_rate": 1.504450686837859e-05, | |
| "loss": 0.3865, | |
| "step": 6894 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8307240704500979, | |
| "eval_f1": 0.840258541089566, | |
| "eval_loss": 0.4156372845172882, | |
| "eval_precision": 0.7954545454545454, | |
| "eval_recall": 0.8904109589041096, | |
| "eval_runtime": 131.4696, | |
| "eval_samples_per_second": 7.774, | |
| "eval_steps_per_second": 1.947, | |
| "step": 6894 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.9364714622497559, | |
| "learning_rate": 0.0, | |
| "loss": 0.3784, | |
| "step": 7660 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8277886497064579, | |
| "eval_f1": 0.8391224862888481, | |
| "eval_loss": 0.4185905158519745, | |
| "eval_precision": 0.7873070325900514, | |
| "eval_recall": 0.898238747553816, | |
| "eval_runtime": 131.6081, | |
| "eval_samples_per_second": 7.765, | |
| "eval_steps_per_second": 1.945, | |
| "step": 7660 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7660, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2356951972800.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7711149722361785, | |
| "learning_rate": 0.0001504450686837859, | |
| "num_train_epochs": 10, | |
| "temperature": 20 | |
| } | |
| } | |