| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0011595696708110545, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001288410745345616, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 3.2623, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001288410745345616, | |
| "eval_loss": 3.1917941570281982, | |
| "eval_runtime": 49.1154, | |
| "eval_samples_per_second": 266.149, | |
| "eval_steps_per_second": 33.269, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002576821490691232, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 2.9077, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00038652322360368483, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 2.9139, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00038652322360368483, | |
| "eval_loss": 3.1917941570281982, | |
| "eval_runtime": 48.8794, | |
| "eval_samples_per_second": 267.434, | |
| "eval_steps_per_second": 33.429, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005153642981382464, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 2.6969, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0006442053726728081, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 2.7296, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0007730464472073697, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 3.0787, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0007730464472073697, | |
| "eval_loss": 3.1917941570281982, | |
| "eval_runtime": 48.3147, | |
| "eval_samples_per_second": 270.559, | |
| "eval_steps_per_second": 33.82, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0009018875217419313, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 2.8677, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0010307285962764929, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 2.7432, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0011595696708110545, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 3.1043, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0011595696708110545, | |
| "eval_loss": 3.1917941570281982, | |
| "eval_runtime": 48.3935, | |
| "eval_samples_per_second": 270.119, | |
| "eval_steps_per_second": 33.765, | |
| "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": 999630619803648.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |