| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.002203492535669035, | |
| "eval_steps": 3, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00022034925356690353, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 5.3398, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00022034925356690353, | |
| "eval_loss": NaN, | |
| "eval_runtime": 65.6349, | |
| "eval_samples_per_second": 29.116, | |
| "eval_steps_per_second": 14.565, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00044069850713380706, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 6.4798, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0006610477607007106, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 2.4963, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0006610477607007106, | |
| "eval_loss": NaN, | |
| "eval_runtime": 63.6588, | |
| "eval_samples_per_second": 30.019, | |
| "eval_steps_per_second": 15.018, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0008813970142676141, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 5.6806, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0011017462678345176, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 4.5203, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0013220955214014213, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 1.1433, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0013220955214014213, | |
| "eval_loss": NaN, | |
| "eval_runtime": 63.7691, | |
| "eval_samples_per_second": 29.967, | |
| "eval_steps_per_second": 14.992, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0015424447749683248, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 2.9261, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0017627940285352283, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0019831432821021317, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 6.3788, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0019831432821021317, | |
| "eval_loss": NaN, | |
| "eval_runtime": 63.6972, | |
| "eval_samples_per_second": 30.001, | |
| "eval_steps_per_second": 15.009, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002203492535669035, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5175, | |
| "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": 2221860958765056.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |