|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.00718562874251497, |
|
"eval_steps": 3, |
|
"global_step": 9, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0007984031936127744, |
|
"grad_norm": 4.354898452758789, |
|
"learning_rate": 2e-05, |
|
"loss": 3.8702, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.0007984031936127744, |
|
"eval_loss": 4.836042881011963, |
|
"eval_runtime": 21.0899, |
|
"eval_samples_per_second": 25.036, |
|
"eval_steps_per_second": 12.518, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.0015968063872255488, |
|
"grad_norm": 2.6686439514160156, |
|
"learning_rate": 4e-05, |
|
"loss": 2.9897, |
|
"step": 2 |
|
}, |
|
{ |
|
"epoch": 0.0023952095808383233, |
|
"grad_norm": 5.189035892486572, |
|
"learning_rate": 6e-05, |
|
"loss": 3.8363, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.0023952095808383233, |
|
"eval_loss": 4.815651893615723, |
|
"eval_runtime": 20.3656, |
|
"eval_samples_per_second": 25.926, |
|
"eval_steps_per_second": 12.963, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.0031936127744510976, |
|
"grad_norm": 4.79802131652832, |
|
"learning_rate": 8e-05, |
|
"loss": 3.6042, |
|
"step": 4 |
|
}, |
|
{ |
|
"epoch": 0.003992015968063872, |
|
"grad_norm": 7.123534679412842, |
|
"learning_rate": 0.0001, |
|
"loss": 4.5736, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.004790419161676647, |
|
"grad_norm": 10.06795883178711, |
|
"learning_rate": 0.00012, |
|
"loss": 5.7979, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.004790419161676647, |
|
"eval_loss": 4.422317028045654, |
|
"eval_runtime": 20.7254, |
|
"eval_samples_per_second": 25.476, |
|
"eval_steps_per_second": 12.738, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.005588822355289421, |
|
"grad_norm": 3.140864133834839, |
|
"learning_rate": 0.00014, |
|
"loss": 3.073, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 0.006387225548902195, |
|
"grad_norm": 7.965623378753662, |
|
"learning_rate": 0.00016, |
|
"loss": 3.2922, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 0.00718562874251497, |
|
"grad_norm": 5.314204216003418, |
|
"learning_rate": 0.00018, |
|
"loss": 2.7362, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 0.00718562874251497, |
|
"eval_loss": 3.3656184673309326, |
|
"eval_runtime": 20.677, |
|
"eval_samples_per_second": 25.536, |
|
"eval_steps_per_second": 12.768, |
|
"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": 1900964437032960.0, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|