mamba_0_5_sft / eval_results.json
Junxiong Wang
add models
74d28bb
{
"epoch": 1.0,
"eval_loss": 0.5660967230796814,
"eval_runtime": 17.7951,
"eval_samples": 4096,
"eval_samples_per_second": 48.103,
"eval_steps_per_second": 0.787
}