| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.0, | |
| "eval_steps": 2000, | |
| "global_step": 4494, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.778816199376947, | |
| "grad_norm": 34.75956726074219, | |
| "learning_rate": 1.777481085892301e-05, | |
| "loss": 3.2057, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.557632398753894, | |
| "grad_norm": 32.37627029418945, | |
| "learning_rate": 1.5549621717846018e-05, | |
| "loss": 2.0499, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.336448598130841, | |
| "grad_norm": 49.43497085571289, | |
| "learning_rate": 1.3324432576769027e-05, | |
| "loss": 1.6891, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.115264797507788, | |
| "grad_norm": 37.82111358642578, | |
| "learning_rate": 1.1099243435692035e-05, | |
| "loss": 1.4529, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.115264797507788, | |
| "eval_runtime": 237.0246, | |
| "eval_samples_per_second": 16.606, | |
| "eval_steps_per_second": 1.038, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.8940809968847354, | |
| "grad_norm": 43.13986587524414, | |
| "learning_rate": 8.874054294615043e-06, | |
| "loss": 1.2499, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.672897196261682, | |
| "grad_norm": 43.04648208618164, | |
| "learning_rate": 6.648865153538051e-06, | |
| "loss": 1.089, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.451713395638629, | |
| "grad_norm": 44.60111999511719, | |
| "learning_rate": 4.42367601246106e-06, | |
| "loss": 0.9941, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.230529595015576, | |
| "grad_norm": 35.048545837402344, | |
| "learning_rate": 2.1984868713840677e-06, | |
| "loss": 0.9121, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.230529595015576, | |
| "eval_runtime": 226.635, | |
| "eval_samples_per_second": 17.367, | |
| "eval_steps_per_second": 1.085, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4494, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 2000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.28609863122688e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |