| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00031038309032923884, | |
| "eval_steps": 3, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.103830903292389e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 3.103830903292389e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1042.6018, | |
| "eval_samples_per_second": 13.012, | |
| "eval_steps_per_second": 6.506, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 6.207661806584778e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 12.2387, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 9.311492709877166e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 9.311492709877166e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1041.4288, | |
| "eval_samples_per_second": 13.026, | |
| "eval_steps_per_second": 6.513, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00012415323613169555, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00015519154516461942, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00018622985419754331, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00018622985419754331, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1041.2817, | |
| "eval_samples_per_second": 13.028, | |
| "eval_steps_per_second": 6.514, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0002172681632304672, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0002483064722633911, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00027934478129631497, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00027934478129631497, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1041.3251, | |
| "eval_samples_per_second": 13.028, | |
| "eval_steps_per_second": 6.514, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00031038309032923884, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "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": 4899183840460800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |