| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 63, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 0.5804110765457153, | |
| "learning_rate": 0.00019357168190404936, | |
| "loss": 2.8315, | |
| "memory/device_reserved (GiB)": 21.65, | |
| "memory/max_active (GiB)": 19.67, | |
| "memory/max_allocated (GiB)": 19.67, | |
| "step": 10, | |
| "tokens_per_second_per_gpu": 1089.23 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 0.5418248772621155, | |
| "learning_rate": 0.00016405931786981755, | |
| "loss": 2.5331, | |
| "memory/device_reserved (GiB)": 21.65, | |
| "memory/max_active (GiB)": 17.12, | |
| "memory/max_allocated (GiB)": 17.12, | |
| "step": 20, | |
| "tokens_per_second_per_gpu": 1399.77 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.926287829875946, | |
| "learning_rate": 0.00011792807588107357, | |
| "loss": 2.2908, | |
| "memory/device_reserved (GiB)": 21.66, | |
| "memory/max_active (GiB)": 19.67, | |
| "memory/max_allocated (GiB)": 19.67, | |
| "step": 30, | |
| "tokens_per_second_per_gpu": 1503.62 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.741166353225708, | |
| "learning_rate": 6.714576180891654e-05, | |
| "loss": 2.3613, | |
| "memory/device_reserved (GiB)": 21.66, | |
| "memory/max_active (GiB)": 13.69, | |
| "memory/max_allocated (GiB)": 13.69, | |
| "step": 40, | |
| "tokens_per_second_per_gpu": 1292.51 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.8406546115875244, | |
| "learning_rate": 2.4886806912948035e-05, | |
| "loss": 2.1182, | |
| "memory/device_reserved (GiB)": 21.66, | |
| "memory/max_active (GiB)": 19.67, | |
| "memory/max_allocated (GiB)": 19.67, | |
| "step": 50, | |
| "tokens_per_second_per_gpu": 1380.03 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.7622830271720886, | |
| "learning_rate": 2.1144314904642195e-06, | |
| "loss": 2.1277, | |
| "memory/device_reserved (GiB)": 21.66, | |
| "memory/max_active (GiB)": 17.48, | |
| "memory/max_allocated (GiB)": 17.48, | |
| "step": 60, | |
| "tokens_per_second_per_gpu": 1322.61 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 63, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.2466390650454016e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |