| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00019995001249687578, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.221666805520842e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 2.0674, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.221666805520842e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 528.8764, | |
| "eval_samples_per_second": 35.836, | |
| "eval_steps_per_second": 17.919, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 4.443333611041684e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 6.665000416562526e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 1.027, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 6.665000416562526e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 526.6934, | |
| "eval_samples_per_second": 35.985, | |
| "eval_steps_per_second": 17.993, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 8.886667222083368e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0001110833402760421, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00013330000833125053, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00013330000833125053, | |
| "eval_loss": NaN, | |
| "eval_runtime": 528.2439, | |
| "eval_samples_per_second": 35.879, | |
| "eval_steps_per_second": 17.941, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00015551667638645893, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00017773334444166737, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00019995001249687578, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00019995001249687578, | |
| "eval_loss": NaN, | |
| "eval_runtime": 528.1465, | |
| "eval_samples_per_second": 35.886, | |
| "eval_steps_per_second": 17.944, | |
| "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": 1568372441481216.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |