Qwen2.5-0.5B-Open-R1-Distill / train_results.json
Qucy's picture
Model save
e5ea3de verified
raw
history blame contribute delete
251 Bytes
{
"epoch": 0.9994447529150472,
"total_flos": 1.6624354892709888e+17,
"train_loss": 0.9221899901496039,
"train_runtime": 5284.4545,
"train_samples": 16610,
"train_samples_per_second": 4.089,
"train_steps_per_second": 0.128
}