| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.010436234606553955, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0002087246921310791, | |
| "eval_loss": 3.6245229244232178, | |
| "eval_runtime": 17.5376, | |
| "eval_samples_per_second": 57.533, | |
| "eval_steps_per_second": 28.795, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002087246921310791, | |
| "grad_norm": 4.278122425079346, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 3.091, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0027134209977040286, | |
| "eval_loss": 3.0272154808044434, | |
| "eval_runtime": 16.3987, | |
| "eval_samples_per_second": 61.529, | |
| "eval_steps_per_second": 30.795, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.004174493842621582, | |
| "grad_norm": 1.7829546928405762, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.9497, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.005426841995408057, | |
| "eval_loss": 2.953101396560669, | |
| "eval_runtime": 16.483, | |
| "eval_samples_per_second": 61.215, | |
| "eval_steps_per_second": 30.638, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.006261740763932373, | |
| "grad_norm": 2.3199422359466553, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 2.8382, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.008140262993112084, | |
| "eval_loss": 2.917341470718384, | |
| "eval_runtime": 16.5153, | |
| "eval_samples_per_second": 61.095, | |
| "eval_steps_per_second": 30.578, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.008348987685243165, | |
| "grad_norm": 2.5687716007232666, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 2.8883, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.010436234606553955, | |
| "grad_norm": 4.335829257965088, | |
| "learning_rate": 0.0, | |
| "loss": 2.9597, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 238816112148480.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |