| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06299212598425197, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0012598425196850393, | |
| "eval_loss": 3.4401309490203857, | |
| "eval_runtime": 7.0083, | |
| "eval_samples_per_second": 47.8, | |
| "eval_steps_per_second": 23.971, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.012598425196850394, | |
| "grad_norm": 0.8125695586204529, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 3.2232, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.016377952755905513, | |
| "eval_loss": 3.1041011810302734, | |
| "eval_runtime": 6.1943, | |
| "eval_samples_per_second": 54.082, | |
| "eval_steps_per_second": 27.121, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.025196850393700787, | |
| "grad_norm": 0.6350086331367493, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 3.1128, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.032755905511811026, | |
| "eval_loss": 3.009904623031616, | |
| "eval_runtime": 6.1583, | |
| "eval_samples_per_second": 54.398, | |
| "eval_steps_per_second": 27.28, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03779527559055118, | |
| "grad_norm": 0.7571959495544434, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 2.9715, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04913385826771653, | |
| "eval_loss": 2.9804084300994873, | |
| "eval_runtime": 6.0644, | |
| "eval_samples_per_second": 55.241, | |
| "eval_steps_per_second": 27.703, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.050393700787401574, | |
| "grad_norm": 0.7553784847259521, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 3.0366, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06299212598425197, | |
| "grad_norm": 0.8436616659164429, | |
| "learning_rate": 0.0, | |
| "loss": 2.932, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 383619013017600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |