| { | |
| "best_metric": 0.8256880733944955, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-527", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 2635, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002646598360783595, | |
| "loss": 0.4864, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8256880733944955, | |
| "eval_loss": 0.5460890531539917, | |
| "eval_runtime": 6.1659, | |
| "eval_samples_per_second": 141.424, | |
| "eval_steps_per_second": 1.135, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00019849487705876962, | |
| "loss": 0.2623, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8188073394495413, | |
| "eval_loss": 0.7069737315177917, | |
| "eval_runtime": 7.2395, | |
| "eval_samples_per_second": 120.451, | |
| "eval_steps_per_second": 0.967, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00013232991803917974, | |
| "loss": 0.1902, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.823394495412844, | |
| "eval_loss": 0.7423160076141357, | |
| "eval_runtime": 7.2214, | |
| "eval_samples_per_second": 120.753, | |
| "eval_steps_per_second": 0.969, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 6.616495901958987e-05, | |
| "loss": 0.1515, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.819954128440367, | |
| "eval_loss": 0.7632457613945007, | |
| "eval_runtime": 7.0986, | |
| "eval_samples_per_second": 122.841, | |
| "eval_steps_per_second": 0.986, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1242, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8222477064220184, | |
| "eval_loss": 0.8318789601325989, | |
| "eval_runtime": 7.1337, | |
| "eval_samples_per_second": 122.236, | |
| "eval_steps_per_second": 0.981, | |
| "step": 2635 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2635, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 40436714110320.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9299267464979704, | |
| "learning_rate": 0.00033082479509794936, | |
| "num_train_epochs": 5, | |
| "temperature": 11 | |
| } | |
| } | |