mamba_0_5_distill / eval_results.json
Junxiong Wang
add models
083ba79
raw
history blame contribute delete
187 Bytes
{
"epoch": 1.0,
"eval_loss": 14.704147338867188,
"eval_runtime": 16.6393,
"eval_samples": 4096,
"eval_samples_per_second": 51.445,
"eval_steps_per_second": 0.841
}