| { | |
| "best_metric": 0.5148514851485149, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-176", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 1760, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.5431386232376099, | |
| "learning_rate": 0.00012088388613301416, | |
| "loss": 0.2133, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5148514851485149, | |
| "eval_f1": 0.14035087719298245, | |
| "eval_loss": 0.20053359866142273, | |
| "eval_precision": 0.6060606060606061, | |
| "eval_recall": 0.07936507936507936, | |
| "eval_runtime": 15.0376, | |
| "eval_samples_per_second": 33.582, | |
| "eval_steps_per_second": 1.064, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.2198910117149353, | |
| "learning_rate": 0.00010745234322934593, | |
| "loss": 0.2028, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5069306930693069, | |
| "eval_f1": 0.1263157894736842, | |
| "eval_loss": 0.20091117918491364, | |
| "eval_precision": 0.5454545454545454, | |
| "eval_recall": 0.07142857142857142, | |
| "eval_runtime": 14.9354, | |
| "eval_samples_per_second": 33.812, | |
| "eval_steps_per_second": 1.071, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.5269190669059753, | |
| "learning_rate": 9.402080032567768e-05, | |
| "loss": 0.2015, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5128712871287129, | |
| "eval_f1": 0.15172413793103445, | |
| "eval_loss": 0.20115335285663605, | |
| "eval_precision": 0.5789473684210527, | |
| "eval_recall": 0.0873015873015873, | |
| "eval_runtime": 15.3006, | |
| "eval_samples_per_second": 33.005, | |
| "eval_steps_per_second": 1.046, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6160932183265686, | |
| "learning_rate": 8.058925742200945e-05, | |
| "loss": 0.2007, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5128712871287129, | |
| "eval_f1": 0.15753424657534246, | |
| "eval_loss": 0.20070397853851318, | |
| "eval_precision": 0.575, | |
| "eval_recall": 0.09126984126984126, | |
| "eval_runtime": 15.3305, | |
| "eval_samples_per_second": 32.941, | |
| "eval_steps_per_second": 1.044, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.17378416657447815, | |
| "learning_rate": 6.71577145183412e-05, | |
| "loss": 0.2004, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5089108910891089, | |
| "eval_f1": 0.1267605633802817, | |
| "eval_loss": 0.19792087376117706, | |
| "eval_precision": 0.5625, | |
| "eval_recall": 0.07142857142857142, | |
| "eval_runtime": 15.059, | |
| "eval_samples_per_second": 33.535, | |
| "eval_steps_per_second": 1.062, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.18278618156909943, | |
| "learning_rate": 5.3726171614672966e-05, | |
| "loss": 0.1992, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.504950495049505, | |
| "eval_f1": 0.13194444444444445, | |
| "eval_loss": 0.19790016114711761, | |
| "eval_precision": 0.5277777777777778, | |
| "eval_recall": 0.07539682539682539, | |
| "eval_runtime": 14.8776, | |
| "eval_samples_per_second": 33.944, | |
| "eval_steps_per_second": 1.075, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.32294225692749023, | |
| "learning_rate": 4.029462871100472e-05, | |
| "loss": 0.1985, | |
| "step": 1232 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5108910891089109, | |
| "eval_f1": 0.13937282229965156, | |
| "eval_loss": 0.19708825647830963, | |
| "eval_precision": 0.5714285714285714, | |
| "eval_recall": 0.07936507936507936, | |
| "eval_runtime": 15.1516, | |
| "eval_samples_per_second": 33.33, | |
| "eval_steps_per_second": 1.056, | |
| "step": 1232 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.24637407064437866, | |
| "learning_rate": 2.6863085807336483e-05, | |
| "loss": 0.1984, | |
| "step": 1408 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5108910891089109, | |
| "eval_f1": 0.13937282229965156, | |
| "eval_loss": 0.19653446972370148, | |
| "eval_precision": 0.5714285714285714, | |
| "eval_recall": 0.07936507936507936, | |
| "eval_runtime": 14.8734, | |
| "eval_samples_per_second": 33.953, | |
| "eval_steps_per_second": 1.076, | |
| "step": 1408 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.2087116688489914, | |
| "learning_rate": 1.3431542903668242e-05, | |
| "loss": 0.198, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5108910891089109, | |
| "eval_f1": 0.13937282229965156, | |
| "eval_loss": 0.19648496806621552, | |
| "eval_precision": 0.5714285714285714, | |
| "eval_recall": 0.07936507936507936, | |
| "eval_runtime": 15.3381, | |
| "eval_samples_per_second": 32.925, | |
| "eval_steps_per_second": 1.043, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.36322712898254395, | |
| "learning_rate": 0.0, | |
| "loss": 0.1975, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5108910891089109, | |
| "eval_f1": 0.13937282229965156, | |
| "eval_loss": 0.19634994864463806, | |
| "eval_precision": 0.5714285714285714, | |
| "eval_recall": 0.07936507936507936, | |
| "eval_runtime": 14.8929, | |
| "eval_samples_per_second": 33.909, | |
| "eval_steps_per_second": 1.074, | |
| "step": 1760 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1760, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 7778432831400.0, | |
| "train_batch_size": 54, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.239256796645401, | |
| "learning_rate": 0.0001343154290366824, | |
| "num_train_epochs": 10, | |
| "per_device_train_batch_size": 54, | |
| "temperature": 13 | |
| } | |
| } | |