|
{ |
|
"best_metric": 1.3716470003128052, |
|
"best_model_checkpoint": "training/mcq/checkpoint-9344", |
|
"epoch": 0.8000684990153266, |
|
"eval_steps": 2336, |
|
"global_step": 9344, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.050004281188457914, |
|
"grad_norm": 33.87300109863281, |
|
"learning_rate": 2.5e-06, |
|
"loss": 1.9187, |
|
"step": 584 |
|
}, |
|
{ |
|
"epoch": 0.10000856237691583, |
|
"grad_norm": 21.070466995239258, |
|
"learning_rate": 5e-06, |
|
"loss": 1.4832, |
|
"step": 1168 |
|
}, |
|
{ |
|
"epoch": 0.15001284356537375, |
|
"grad_norm": 26.92688751220703, |
|
"learning_rate": 4.962012173713227e-06, |
|
"loss": 1.4456, |
|
"step": 1752 |
|
}, |
|
{ |
|
"epoch": 0.20001712475383165, |
|
"grad_norm": 23.34733009338379, |
|
"learning_rate": 4.849203154809702e-06, |
|
"loss": 1.4308, |
|
"step": 2336 |
|
}, |
|
{ |
|
"epoch": 0.20001712475383165, |
|
"eval_loss": 1.4146325588226318, |
|
"eval_runtime": 45.195, |
|
"eval_samples_per_second": 22.126, |
|
"eval_steps_per_second": 22.126, |
|
"step": 2336 |
|
}, |
|
{ |
|
"epoch": 0.25002140594228955, |
|
"grad_norm": 24.534526824951172, |
|
"learning_rate": 4.665001238820376e-06, |
|
"loss": 1.438, |
|
"step": 2920 |
|
}, |
|
{ |
|
"epoch": 0.3000256871307475, |
|
"grad_norm": 17.830188751220703, |
|
"learning_rate": 4.4150043700542834e-06, |
|
"loss": 1.4406, |
|
"step": 3504 |
|
}, |
|
{ |
|
"epoch": 0.3500299683192054, |
|
"grad_norm": 23.592960357666016, |
|
"learning_rate": 4.106810018609764e-06, |
|
"loss": 1.3984, |
|
"step": 4088 |
|
}, |
|
{ |
|
"epoch": 0.4000342495076633, |
|
"grad_norm": 26.053688049316406, |
|
"learning_rate": 3.7497842912750083e-06, |
|
"loss": 1.4088, |
|
"step": 4672 |
|
}, |
|
{ |
|
"epoch": 0.4000342495076633, |
|
"eval_loss": 1.3911218643188477, |
|
"eval_runtime": 45.1415, |
|
"eval_samples_per_second": 22.153, |
|
"eval_steps_per_second": 22.153, |
|
"step": 4672 |
|
}, |
|
{ |
|
"epoch": 0.45003853069612126, |
|
"grad_norm": 18.178329467773438, |
|
"learning_rate": 3.3547772930979383e-06, |
|
"loss": 1.4211, |
|
"step": 5256 |
|
}, |
|
{ |
|
"epoch": 0.5000428118845791, |
|
"grad_norm": 24.71056365966797, |
|
"learning_rate": 2.9337933898616017e-06, |
|
"loss": 1.3645, |
|
"step": 5840 |
|
}, |
|
{ |
|
"epoch": 0.5500470930730371, |
|
"grad_norm": 20.96603012084961, |
|
"learning_rate": 2.499626392274534e-06, |
|
"loss": 1.3794, |
|
"step": 6424 |
|
}, |
|
{ |
|
"epoch": 0.600051374261495, |
|
"grad_norm": 21.262020111083984, |
|
"learning_rate": 2.0654707487237674e-06, |
|
"loss": 1.3988, |
|
"step": 7008 |
|
}, |
|
{ |
|
"epoch": 0.600051374261495, |
|
"eval_loss": 1.3772270679473877, |
|
"eval_runtime": 44.5512, |
|
"eval_samples_per_second": 22.446, |
|
"eval_steps_per_second": 22.446, |
|
"step": 7008 |
|
}, |
|
{ |
|
"epoch": 0.6500556554499529, |
|
"grad_norm": 33.738304138183594, |
|
"learning_rate": 1.6445205625442019e-06, |
|
"loss": 1.3821, |
|
"step": 7592 |
|
}, |
|
{ |
|
"epoch": 0.7000599366384108, |
|
"grad_norm": 25.255008697509766, |
|
"learning_rate": 1.2495686197742195e-06, |
|
"loss": 1.3797, |
|
"step": 8176 |
|
}, |
|
{ |
|
"epoch": 0.7500642178268687, |
|
"grad_norm": 25.995363235473633, |
|
"learning_rate": 8.926176130486741e-07, |
|
"loss": 1.3834, |
|
"step": 8760 |
|
}, |
|
{ |
|
"epoch": 0.8000684990153266, |
|
"grad_norm": 23.964170455932617, |
|
"learning_rate": 5.845153766366707e-07, |
|
"loss": 1.378, |
|
"step": 9344 |
|
}, |
|
{ |
|
"epoch": 0.8000684990153266, |
|
"eval_loss": 1.3716470003128052, |
|
"eval_runtime": 44.1994, |
|
"eval_samples_per_second": 22.625, |
|
"eval_steps_per_second": 22.625, |
|
"step": 9344 |
|
} |
|
], |
|
"logging_steps": 584, |
|
"max_steps": 11679, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 2336, |
|
"total_flos": 0.0, |
|
"train_batch_size": 1, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|