| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.861635220125786, | |
| "eval_steps": 500, | |
| "global_step": 2500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.584516129032258, | |
| "eval_loss": 0.19766665995121002, | |
| "eval_runtime": 1.3839, | |
| "eval_samples_per_second": 2239.966, | |
| "eval_steps_per_second": 46.967, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.5089908242225647, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.318, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8293548387096774, | |
| "eval_loss": 0.0966305062174797, | |
| "eval_runtime": 1.3724, | |
| "eval_samples_per_second": 2258.896, | |
| "eval_steps_per_second": 47.364, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8806451612903226, | |
| "eval_loss": 0.0646040290594101, | |
| "eval_runtime": 1.3887, | |
| "eval_samples_per_second": 2232.348, | |
| "eval_steps_per_second": 46.807, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.47452715039253235, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1118, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8938709677419355, | |
| "eval_loss": 0.04966060072183609, | |
| "eval_runtime": 1.4052, | |
| "eval_samples_per_second": 2206.033, | |
| "eval_steps_per_second": 46.256, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.31503432989120483, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.0711, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9077419354838709, | |
| "eval_loss": 0.04069723188877106, | |
| "eval_runtime": 1.3704, | |
| "eval_samples_per_second": 2262.072, | |
| "eval_steps_per_second": 47.431, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9148387096774193, | |
| "eval_loss": 0.035783424973487854, | |
| "eval_runtime": 1.5987, | |
| "eval_samples_per_second": 1939.088, | |
| "eval_steps_per_second": 40.658, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.2500377297401428, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.0561, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9193548387096774, | |
| "eval_loss": 0.033154334872961044, | |
| "eval_runtime": 1.3708, | |
| "eval_samples_per_second": 2261.503, | |
| "eval_steps_per_second": 47.419, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.861635220125786, | |
| "grad_norm": 0.22538858652114868, | |
| "learning_rate": 2.5296995108315863e-06, | |
| "loss": 0.0492, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 651155886807636.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.18633073638587916, | |
| "num_train_epochs": 9, | |
| "temperature": 12 | |
| } | |
| } | |