| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00045106564258059665, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.011840473117741e-05, | |
| "grad_norm": 0.43347492814064026, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9834, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 5.011840473117741e-05, | |
| "eval_loss": 1.196916937828064, | |
| "eval_runtime": 583.5761, | |
| "eval_samples_per_second": 14.397, | |
| "eval_steps_per_second": 7.199, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00010023680946235482, | |
| "grad_norm": 0.5679823756217957, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2146, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0001503552141935322, | |
| "grad_norm": 0.6408512592315674, | |
| "learning_rate": 6e-05, | |
| "loss": 1.266, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0001503552141935322, | |
| "eval_loss": 1.1941131353378296, | |
| "eval_runtime": 541.3168, | |
| "eval_samples_per_second": 15.521, | |
| "eval_steps_per_second": 7.761, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00020047361892470964, | |
| "grad_norm": 0.5734588503837585, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2434, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00025059202365588704, | |
| "grad_norm": 0.5008823275566101, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3256, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0003007104283870644, | |
| "grad_norm": 0.5050337910652161, | |
| "learning_rate": 0.00012, | |
| "loss": 1.0693, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0003007104283870644, | |
| "eval_loss": 1.1512657403945923, | |
| "eval_runtime": 407.5488, | |
| "eval_samples_per_second": 20.616, | |
| "eval_steps_per_second": 10.308, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00035082883311824184, | |
| "grad_norm": 0.506304919719696, | |
| "learning_rate": 0.00014, | |
| "loss": 0.9069, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0004009472378494193, | |
| "grad_norm": 0.7822049260139465, | |
| "learning_rate": 0.00016, | |
| "loss": 1.3244, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00045106564258059665, | |
| "grad_norm": 0.65694260597229, | |
| "learning_rate": 0.00018, | |
| "loss": 1.1036, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00045106564258059665, | |
| "eval_loss": 1.054183006286621, | |
| "eval_runtime": 407.7644, | |
| "eval_samples_per_second": 20.605, | |
| "eval_steps_per_second": 10.303, | |
| "step": 9 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2681868159811584.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |