ppo-sft-rm / bert_output4 /trainer_state.json
lxylxy's picture
Upload 20 files
da9830e verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 9.795918367346939,
"eval_steps": 500,
"global_step": 210,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.98,
"eval_f1": 0.7436216552582452,
"eval_loss": 0.6690771579742432,
"eval_runtime": 61.1056,
"eval_samples_per_second": 157.792,
"eval_steps_per_second": 2.471,
"step": 21
},
{
"epoch": 1.96,
"eval_f1": 0.7696535988384152,
"eval_loss": 0.5603442192077637,
"eval_runtime": 61.0312,
"eval_samples_per_second": 157.985,
"eval_steps_per_second": 2.474,
"step": 42
},
{
"epoch": 2.99,
"eval_f1": 0.7833437046255962,
"eval_loss": 0.5153456926345825,
"eval_runtime": 60.9434,
"eval_samples_per_second": 158.212,
"eval_steps_per_second": 2.478,
"step": 64
},
{
"epoch": 3.97,
"eval_f1": 0.7964115328769965,
"eval_loss": 0.5128823518753052,
"eval_runtime": 61.1456,
"eval_samples_per_second": 157.689,
"eval_steps_per_second": 2.47,
"step": 85
},
{
"epoch": 4.99,
"eval_f1": 0.8009749014727235,
"eval_loss": 0.5235405564308167,
"eval_runtime": 61.309,
"eval_samples_per_second": 157.269,
"eval_steps_per_second": 2.463,
"step": 107
},
{
"epoch": 5.97,
"eval_f1": 0.808442231902095,
"eval_loss": 0.5221107602119446,
"eval_runtime": 61.1652,
"eval_samples_per_second": 157.639,
"eval_steps_per_second": 2.469,
"step": 128
},
{
"epoch": 7.0,
"eval_f1": 0.8134204521883428,
"eval_loss": 0.529384195804596,
"eval_runtime": 61.8448,
"eval_samples_per_second": 155.906,
"eval_steps_per_second": 2.442,
"step": 150
},
{
"epoch": 7.98,
"eval_f1": 0.8148724331051649,
"eval_loss": 0.532220721244812,
"eval_runtime": 61.0347,
"eval_samples_per_second": 157.976,
"eval_steps_per_second": 2.474,
"step": 171
},
{
"epoch": 8.96,
"eval_f1": 0.8171541174030285,
"eval_loss": 0.535590648651123,
"eval_runtime": 61.0479,
"eval_samples_per_second": 157.942,
"eval_steps_per_second": 2.473,
"step": 192
},
{
"epoch": 9.8,
"eval_f1": 0.8157021364862063,
"eval_loss": 0.5402965545654297,
"eval_runtime": 61.4418,
"eval_samples_per_second": 156.929,
"eval_steps_per_second": 2.458,
"step": 210
},
{
"epoch": 9.8,
"step": 210,
"total_flos": 2.0036545414026394e+17,
"train_loss": 0.484508296421596,
"train_runtime": 7951.5871,
"train_samples_per_second": 55.203,
"train_steps_per_second": 0.026
}
],
"logging_steps": 500,
"max_steps": 210,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"total_flos": 2.0036545414026394e+17,
"train_batch_size": 64,
"trial_name": null,
"trial_params": null
}