| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0014025319207499338, | |
| "eval_steps": 25, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.610127682999735e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22644.069, | |
| "eval_samples_per_second": 2.799, | |
| "eval_steps_per_second": 2.799, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00028050638414998675, | |
| "grad_norm": 8.75216007232666, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 8.2406, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0005610127682999735, | |
| "grad_norm": 6.417360305786133, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 4.3853, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0008415191524499603, | |
| "grad_norm": 7.757654190063477, | |
| "learning_rate": 8.296729075500344e-05, | |
| "loss": 2.1783, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.001122025536599947, | |
| "grad_norm": 4.7035722732543945, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.887, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0014025319207499338, | |
| "grad_norm": 3.3673248291015625, | |
| "learning_rate": 5.327015646150716e-05, | |
| "loss": 1.5397, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0014025319207499338, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22696.0747, | |
| "eval_samples_per_second": 2.792, | |
| "eval_steps_per_second": 2.792, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.41887283560448e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |