Qwen2-0.5B-Reward-DR-HH-Seed0 / all_results.json
mamba413's picture
Model save
236719b verified
{
"epoch": 0,
"eval_accuracy": 0.4,
"eval_loss": 0.6474097967147827,
"eval_runtime": 0.497,
"eval_samples_per_second": 10.06,
"eval_steps_per_second": 6.036
}