| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.002832321141643291, | |
| "eval_steps": 10, | |
| "global_step": 26, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00010893542852474196, | |
| "eval_loss": 4.965454578399658, | |
| "eval_runtime": 146.1989, | |
| "eval_samples_per_second": 26.443, | |
| "eval_steps_per_second": 13.222, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005446771426237097, | |
| "grad_norm": 2.617262125015259, | |
| "learning_rate": 5e-05, | |
| "loss": 4.4074, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0010893542852474195, | |
| "grad_norm": 3.2852272987365723, | |
| "learning_rate": 0.0001, | |
| "loss": 4.7221, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0010893542852474195, | |
| "eval_loss": 4.596639633178711, | |
| "eval_runtime": 146.8778, | |
| "eval_samples_per_second": 26.321, | |
| "eval_steps_per_second": 13.161, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0016340314278711294, | |
| "grad_norm": 1.9587860107421875, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 4.1211, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.002178708570494839, | |
| "grad_norm": 3.960905075073242, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 4.1157, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.002178708570494839, | |
| "eval_loss": 3.9367244243621826, | |
| "eval_runtime": 147.1567, | |
| "eval_samples_per_second": 26.271, | |
| "eval_steps_per_second": 13.136, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.002723385713118549, | |
| "grad_norm": 2.2079458236694336, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 3.8274, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2132126528962560.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |