| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.716981132075472, | |
| "eval_steps": 500, | |
| "global_step": 1500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.21216703951358795, | |
| "eval_runtime": 5.8214, | |
| "eval_samples_per_second": 532.514, | |
| "eval_steps_per_second": 11.166, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.5472351908683777, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.3411, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8267741935483871, | |
| "eval_loss": 0.10011853277683258, | |
| "eval_runtime": 6.0665, | |
| "eval_samples_per_second": 511.002, | |
| "eval_steps_per_second": 10.715, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8829032258064516, | |
| "eval_loss": 0.06541892141103745, | |
| "eval_runtime": 5.7333, | |
| "eval_samples_per_second": 540.702, | |
| "eval_steps_per_second": 11.337, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.4667431116104126, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1165, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9003225806451612, | |
| "eval_loss": 0.04985565319657326, | |
| "eval_runtime": 5.7103, | |
| "eval_samples_per_second": 542.881, | |
| "eval_steps_per_second": 11.383, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.31565672159194946, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.073, | |
| "step": 1500 | |
| } | |
| ], | |
| "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": 474373276737108.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.46921361817552565, | |
| "num_train_epochs": 9, | |
| "temperature": 7 | |
| } | |
| } | |