|
{ |
|
"best_metric": 0.9542460441589355, |
|
"best_model_checkpoint": "./outputs/checkpoint-1600", |
|
"epoch": 2.1333333333333333, |
|
"eval_steps": 100, |
|
"global_step": 1600, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 0.0002, |
|
"loss": 1.0003, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"eval_loss": 1.09341299533844, |
|
"eval_runtime": 882.5981, |
|
"eval_samples_per_second": 2.187, |
|
"eval_steps_per_second": 0.274, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 0.0002, |
|
"loss": 0.8971, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"eval_loss": 1.0693918466567993, |
|
"eval_runtime": 888.706, |
|
"eval_samples_per_second": 2.172, |
|
"eval_steps_per_second": 0.272, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 0.0002, |
|
"loss": 0.8816, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"eval_loss": 1.0537036657333374, |
|
"eval_runtime": 878.8304, |
|
"eval_samples_per_second": 2.196, |
|
"eval_steps_per_second": 0.275, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"learning_rate": 0.0002, |
|
"loss": 0.8696, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"eval_loss": 1.0404157638549805, |
|
"eval_runtime": 880.6616, |
|
"eval_samples_per_second": 2.192, |
|
"eval_steps_per_second": 0.275, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"learning_rate": 0.0002, |
|
"loss": 0.8552, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"eval_loss": 1.0278141498565674, |
|
"eval_runtime": 910.8505, |
|
"eval_samples_per_second": 2.119, |
|
"eval_steps_per_second": 0.266, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 0.0002, |
|
"loss": 0.8496, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"eval_loss": 1.0164580345153809, |
|
"eval_runtime": 919.0357, |
|
"eval_samples_per_second": 2.1, |
|
"eval_steps_per_second": 0.263, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 0.0002, |
|
"loss": 0.8447, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"eval_loss": 1.011406660079956, |
|
"eval_runtime": 917.8526, |
|
"eval_samples_per_second": 2.103, |
|
"eval_steps_per_second": 0.264, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 1.07, |
|
"learning_rate": 0.0002, |
|
"loss": 0.8277, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 1.07, |
|
"eval_loss": 1.0011341571807861, |
|
"eval_runtime": 915.3225, |
|
"eval_samples_per_second": 2.109, |
|
"eval_steps_per_second": 0.264, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"learning_rate": 0.0002, |
|
"loss": 0.8139, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"eval_loss": 0.9942707419395447, |
|
"eval_runtime": 917.4155, |
|
"eval_samples_per_second": 2.104, |
|
"eval_steps_per_second": 0.264, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"learning_rate": 0.0002, |
|
"loss": 0.8085, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"eval_loss": 0.9867916107177734, |
|
"eval_runtime": 916.7878, |
|
"eval_samples_per_second": 2.105, |
|
"eval_steps_per_second": 0.264, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"learning_rate": 0.0002, |
|
"loss": 0.8146, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"eval_loss": 0.98065584897995, |
|
"eval_runtime": 731.037, |
|
"eval_samples_per_second": 2.64, |
|
"eval_steps_per_second": 0.331, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"learning_rate": 0.0002, |
|
"loss": 0.805, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"eval_loss": 0.9752777218818665, |
|
"eval_runtime": 730.7507, |
|
"eval_samples_per_second": 2.641, |
|
"eval_steps_per_second": 0.331, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 1.73, |
|
"learning_rate": 0.0002, |
|
"loss": 0.7946, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 1.73, |
|
"eval_loss": 0.9679266810417175, |
|
"eval_runtime": 730.5439, |
|
"eval_samples_per_second": 2.642, |
|
"eval_steps_per_second": 0.331, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 1.87, |
|
"learning_rate": 0.0002, |
|
"loss": 0.7937, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 1.87, |
|
"eval_loss": 0.9621365666389465, |
|
"eval_runtime": 729.9199, |
|
"eval_samples_per_second": 2.644, |
|
"eval_steps_per_second": 0.332, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 0.0002, |
|
"loss": 0.7865, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 0.9568483829498291, |
|
"eval_runtime": 730.7929, |
|
"eval_samples_per_second": 2.641, |
|
"eval_steps_per_second": 0.331, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 2.13, |
|
"learning_rate": 0.0002, |
|
"loss": 0.7657, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 2.13, |
|
"eval_loss": 0.9542460441589355, |
|
"eval_runtime": 730.7608, |
|
"eval_samples_per_second": 2.641, |
|
"eval_steps_per_second": 0.331, |
|
"step": 1600 |
|
} |
|
], |
|
"logging_steps": 100, |
|
"max_steps": 2250, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 100, |
|
"total_flos": 4.14342522878976e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|