Qwen2.5-1.5B-Open-R1-Distill / train_results.json
wenyl's picture
Model save
d1513fd verified
raw
history blame contribute delete
250 Bytes
{
"epoch": 0.999259807549963,
"total_flos": 5.2170668949005926e+17,
"train_loss": 0.7709741959748445,
"train_runtime": 4715.7656,
"train_samples": 16610,
"train_samples_per_second": 4.583,
"train_steps_per_second": 0.143
}