| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00984009840098401, | |
| "eval_steps": 9, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005466721333880005, | |
| "eval_loss": 0.947159469127655, | |
| "eval_runtime": 422.6377, | |
| "eval_samples_per_second": 7.29, | |
| "eval_steps_per_second": 0.913, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016400164001640015, | |
| "grad_norm": 0.4066868722438812, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.9585, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.003280032800328003, | |
| "grad_norm": 0.5946331024169922, | |
| "learning_rate": 3e-05, | |
| "loss": 0.904, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004920049200492005, | |
| "grad_norm": 0.6447503566741943, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.9863, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004920049200492005, | |
| "eval_loss": 0.9110166430473328, | |
| "eval_runtime": 426.9481, | |
| "eval_samples_per_second": 7.216, | |
| "eval_steps_per_second": 0.904, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006560065600656006, | |
| "grad_norm": 0.4532524645328522, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.9468, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008200082000820008, | |
| "grad_norm": 0.45407116413116455, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.7672, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00984009840098401, | |
| "grad_norm": 0.31059959530830383, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.7584, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00984009840098401, | |
| "eval_loss": 0.7376940250396729, | |
| "eval_runtime": 426.7475, | |
| "eval_samples_per_second": 7.22, | |
| "eval_steps_per_second": 0.905, | |
| "step": 18 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.693768229879808e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |