hashtag / summary /checkpoint-2000 /trainer_state.json
jjae's picture
Hash-tag Producer
67d097e verified
raw
history blame
1.86 kB
{
"best_metric": 0.37720897793769836,
"best_model_checkpoint": "C:/Users/user/Documents/EWHA/log/log/modelling/summary/checkpoint-1000",
"epoch": 5.698005698005698,
"eval_steps": 500,
"global_step": 2000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.42,
"learning_rate": 5.5350724637681156e-05,
"loss": 1.4776,
"step": 500
},
{
"epoch": 1.42,
"eval_loss": 0.3854794204235077,
"eval_runtime": 3.9836,
"eval_samples_per_second": 78.572,
"eval_steps_per_second": 10.041,
"step": 500
},
{
"epoch": 2.85,
"learning_rate": 5.372753623188405e-05,
"loss": 0.2469,
"step": 1000
},
{
"epoch": 2.85,
"eval_loss": 0.37720897793769836,
"eval_runtime": 3.9849,
"eval_samples_per_second": 78.547,
"eval_steps_per_second": 10.038,
"step": 1000
},
{
"epoch": 4.27,
"learning_rate": 5.210434782608695e-05,
"loss": 0.1303,
"step": 1500
},
{
"epoch": 4.27,
"eval_loss": 0.4201067090034485,
"eval_runtime": 3.9854,
"eval_samples_per_second": 78.536,
"eval_steps_per_second": 10.037,
"step": 1500
},
{
"epoch": 5.7,
"learning_rate": 5.0481159420289856e-05,
"loss": 0.0722,
"step": 2000
},
{
"epoch": 5.7,
"eval_loss": 0.4348262846469879,
"eval_runtime": 3.9835,
"eval_samples_per_second": 78.574,
"eval_steps_per_second": 10.041,
"step": 2000
}
],
"logging_steps": 500,
"max_steps": 17550,
"num_input_tokens_seen": 0,
"num_train_epochs": 50,
"save_steps": 1000,
"total_flos": 4877891665920000.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}