| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.17777777777777778, | |
| "eval_steps": 8, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005925925925925926, | |
| "eval_loss": 15.121915817260742, | |
| "eval_runtime": 9.8278, | |
| "eval_samples_per_second": 7.224, | |
| "eval_steps_per_second": 3.663, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.017777777777777778, | |
| "grad_norm": 13.326007843017578, | |
| "learning_rate": 6e-05, | |
| "loss": 15.0493, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.035555555555555556, | |
| "grad_norm": 23.55678939819336, | |
| "learning_rate": 0.00012, | |
| "loss": 13.5722, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.047407407407407405, | |
| "eval_loss": 7.749727725982666, | |
| "eval_runtime": 9.0799, | |
| "eval_samples_per_second": 7.819, | |
| "eval_steps_per_second": 3.965, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.05333333333333334, | |
| "grad_norm": 26.799129486083984, | |
| "learning_rate": 0.00018, | |
| "loss": 9.4622, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07111111111111111, | |
| "grad_norm": 9.160043716430664, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 5.7567, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 6.911787509918213, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 4.5565, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09481481481481481, | |
| "eval_loss": 4.17349910736084, | |
| "eval_runtime": 9.0339, | |
| "eval_samples_per_second": 7.859, | |
| "eval_steps_per_second": 3.985, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.10666666666666667, | |
| "grad_norm": 5.684746265411377, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 4.1979, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.12444444444444444, | |
| "grad_norm": 7.674033164978027, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 4.2213, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.14222222222222222, | |
| "grad_norm": 7.858697414398193, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 4.0256, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.14222222222222222, | |
| "eval_loss": 3.9667611122131348, | |
| "eval_runtime": 9.1522, | |
| "eval_samples_per_second": 7.758, | |
| "eval_steps_per_second": 3.933, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 6.60933256149292, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 3.9633, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 6.702687740325928, | |
| "learning_rate": 0.0, | |
| "loss": 3.9846, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 70, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.112830925340672e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |