| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.8385744234800838, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.016771488469601678, | |
| "eval_loss": 10.19597053527832, | |
| "eval_runtime": 22.5864, | |
| "eval_samples_per_second": 17.798, | |
| "eval_steps_per_second": 2.258, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.16771488469601678, | |
| "grad_norm": 8.230109214782715, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 9.3317, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16771488469601678, | |
| "eval_loss": 6.821496963500977, | |
| "eval_runtime": 22.9033, | |
| "eval_samples_per_second": 17.552, | |
| "eval_steps_per_second": 2.227, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.33542976939203356, | |
| "grad_norm": 4.020260810852051, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 5.2304, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.33542976939203356, | |
| "eval_loss": 3.8652100563049316, | |
| "eval_runtime": 22.696, | |
| "eval_samples_per_second": 17.712, | |
| "eval_steps_per_second": 2.247, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5031446540880503, | |
| "grad_norm": 2.6368815898895264, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 3.5338, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5031446540880503, | |
| "eval_loss": 3.3421852588653564, | |
| "eval_runtime": 22.6281, | |
| "eval_samples_per_second": 17.766, | |
| "eval_steps_per_second": 2.254, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6708595387840671, | |
| "grad_norm": 3.0625898838043213, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 3.1788, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6708595387840671, | |
| "eval_loss": 3.1706862449645996, | |
| "eval_runtime": 22.6332, | |
| "eval_samples_per_second": 17.762, | |
| "eval_steps_per_second": 2.253, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8385744234800838, | |
| "grad_norm": 2.183499813079834, | |
| "learning_rate": 0.0, | |
| "loss": 3.1481, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8385744234800838, | |
| "eval_loss": 3.1483066082000732, | |
| "eval_runtime": 22.6609, | |
| "eval_samples_per_second": 17.74, | |
| "eval_steps_per_second": 2.251, | |
| "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": 1.0755878286262272e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |