| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.057770075101097634, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011554015020219526, | |
| "eval_loss": 6.479506969451904, | |
| "eval_runtime": 205.9246, | |
| "eval_samples_per_second": 28.316, | |
| "eval_steps_per_second": 3.54, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011554015020219527, | |
| "grad_norm": 2.6358954906463623, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 5.4872, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.011554015020219527, | |
| "eval_loss": 4.352705955505371, | |
| "eval_runtime": 204.1824, | |
| "eval_samples_per_second": 28.558, | |
| "eval_steps_per_second": 3.57, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.023108030040439053, | |
| "grad_norm": 2.1231181621551514, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.8224, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.023108030040439053, | |
| "eval_loss": 1.9355292320251465, | |
| "eval_runtime": 203.7468, | |
| "eval_samples_per_second": 28.619, | |
| "eval_steps_per_second": 3.578, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03466204506065858, | |
| "grad_norm": 1.285149097442627, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 1.6201, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03466204506065858, | |
| "eval_loss": 1.2765699625015259, | |
| "eval_runtime": 203.7198, | |
| "eval_samples_per_second": 28.623, | |
| "eval_steps_per_second": 3.578, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.046216060080878106, | |
| "grad_norm": 1.075261116027832, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 1.1615, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.046216060080878106, | |
| "eval_loss": 1.1079412698745728, | |
| "eval_runtime": 204.389, | |
| "eval_samples_per_second": 28.529, | |
| "eval_steps_per_second": 3.567, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.057770075101097634, | |
| "grad_norm": 1.0194032192230225, | |
| "learning_rate": 0.0, | |
| "loss": 1.0724, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.057770075101097634, | |
| "eval_loss": 1.0836020708084106, | |
| "eval_runtime": 204.0597, | |
| "eval_samples_per_second": 28.575, | |
| "eval_steps_per_second": 3.572, | |
| "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": 4.776991368255898e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |