| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0016437490121700648, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.21474105684632e-05, | |
| "eval_loss": 0.8874483108520508, | |
| "eval_runtime": 1530.7819, | |
| "eval_samples_per_second": 6.526, | |
| "eval_steps_per_second": 3.263, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.000210737052842316, | |
| "grad_norm": 4.626171112060547, | |
| "learning_rate": 5e-05, | |
| "loss": 3.5018, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.000421474105684632, | |
| "grad_norm": 3.7195966243743896, | |
| "learning_rate": 0.0001, | |
| "loss": 3.3865, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.000421474105684632, | |
| "eval_loss": 0.7994943261146545, | |
| "eval_runtime": 1558.7542, | |
| "eval_samples_per_second": 6.409, | |
| "eval_steps_per_second": 3.204, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.000632211158526948, | |
| "grad_norm": 5.0703887939453125, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.0742, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.000842948211369264, | |
| "grad_norm": 3.5031182765960693, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.9059, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.000842948211369264, | |
| "eval_loss": 0.7205164432525635, | |
| "eval_runtime": 1556.9817, | |
| "eval_samples_per_second": 6.416, | |
| "eval_steps_per_second": 3.208, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00105368526421158, | |
| "grad_norm": 4.231757640838623, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.7684, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.001264422317053896, | |
| "grad_norm": 3.9624576568603516, | |
| "learning_rate": 5e-05, | |
| "loss": 2.8031, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.001264422317053896, | |
| "eval_loss": 0.6921516060829163, | |
| "eval_runtime": 1561.0748, | |
| "eval_samples_per_second": 6.399, | |
| "eval_steps_per_second": 3.2, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.001475159369896212, | |
| "grad_norm": 4.666100978851318, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 2.7865, | |
| "step": 35 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2137252209557504e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |