| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02702702702702703, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003003003003003003, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 2.0249, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003003003003003003, | |
| "eval_loss": 2.2815349102020264, | |
| "eval_runtime": 5.7285, | |
| "eval_samples_per_second": 12.394, | |
| "eval_steps_per_second": 12.394, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006006006006006006, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 2.8148, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.009009009009009009, | |
| "grad_norm": 8.639544486999512, | |
| "learning_rate": 2e-05, | |
| "loss": 2.3406, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.009009009009009009, | |
| "eval_loss": 2.2815349102020264, | |
| "eval_runtime": 5.1823, | |
| "eval_samples_per_second": 13.701, | |
| "eval_steps_per_second": 13.701, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.012012012012012012, | |
| "grad_norm": 3.1249139308929443, | |
| "learning_rate": 4e-05, | |
| "loss": 2.0188, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.015015015015015015, | |
| "grad_norm": 2.0706639289855957, | |
| "learning_rate": 6e-05, | |
| "loss": 1.4195, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.018018018018018018, | |
| "grad_norm": 4.249810218811035, | |
| "learning_rate": 8e-05, | |
| "loss": 2.6119, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.018018018018018018, | |
| "eval_loss": 2.254021644592285, | |
| "eval_runtime": 5.358, | |
| "eval_samples_per_second": 13.251, | |
| "eval_steps_per_second": 13.251, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.021021021021021023, | |
| "grad_norm": 5.296385765075684, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1297, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.024024024024024024, | |
| "grad_norm": 3.105656623840332, | |
| "learning_rate": 0.00012, | |
| "loss": 2.0236, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02702702702702703, | |
| "grad_norm": 3.085301399230957, | |
| "learning_rate": 0.00014, | |
| "loss": 1.7983, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02702702702702703, | |
| "eval_loss": 2.1258544921875, | |
| "eval_runtime": 5.0108, | |
| "eval_samples_per_second": 14.169, | |
| "eval_steps_per_second": 14.169, | |
| "step": 9 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1664607823331328.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |