| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10526315789473684, | |
| "eval_steps": 3, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010526315789473684, | |
| "grad_norm": 36.76760482788086, | |
| "learning_rate": 2e-05, | |
| "loss": 5.8338, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010526315789473684, | |
| "eval_loss": 5.785942554473877, | |
| "eval_runtime": 0.9354, | |
| "eval_samples_per_second": 42.762, | |
| "eval_steps_per_second": 21.381, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.021052631578947368, | |
| "grad_norm": 45.108951568603516, | |
| "learning_rate": 4e-05, | |
| "loss": 5.5921, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.031578947368421054, | |
| "grad_norm": 20.835744857788086, | |
| "learning_rate": 6e-05, | |
| "loss": 5.4559, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.031578947368421054, | |
| "eval_loss": 5.616115570068359, | |
| "eval_runtime": 0.5881, | |
| "eval_samples_per_second": 68.016, | |
| "eval_steps_per_second": 34.008, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.042105263157894736, | |
| "grad_norm": 18.85835075378418, | |
| "learning_rate": 8e-05, | |
| "loss": 5.7312, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.05263157894736842, | |
| "grad_norm": 9.160707473754883, | |
| "learning_rate": 0.0001, | |
| "loss": 5.3818, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06315789473684211, | |
| "grad_norm": 6.502281188964844, | |
| "learning_rate": 0.00012, | |
| "loss": 5.3188, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.06315789473684211, | |
| "eval_loss": 5.2194647789001465, | |
| "eval_runtime": 0.574, | |
| "eval_samples_per_second": 69.686, | |
| "eval_steps_per_second": 34.843, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07368421052631578, | |
| "grad_norm": 5.528709411621094, | |
| "learning_rate": 0.00014, | |
| "loss": 4.6828, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.08421052631578947, | |
| "grad_norm": 8.433736801147461, | |
| "learning_rate": 0.00016, | |
| "loss": 4.4533, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.09473684210526316, | |
| "grad_norm": 8.037420272827148, | |
| "learning_rate": 0.00018, | |
| "loss": 4.3143, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09473684210526316, | |
| "eval_loss": 4.349830627441406, | |
| "eval_runtime": 0.5587, | |
| "eval_samples_per_second": 71.589, | |
| "eval_steps_per_second": 35.794, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10526315789473684, | |
| "grad_norm": 6.39585018157959, | |
| "learning_rate": 0.0002, | |
| "loss": 4.29, | |
| "step": 10 | |
| } | |
| ], | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 255791978250240.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |