Qwen2.5-1.5B-Open-R1-Distill / train_results.json
davidkim205's picture
Model save
1f241f6 verified
raw
history blame contribute delete
249 Bytes
{
"epoch": 0.9994447529150472,
"total_flos": 6.955833048956928e+17,
"train_loss": 0.820948981002525,
"train_runtime": 8242.4634,
"train_samples": 16610,
"train_samples_per_second": 2.622,
"train_steps_per_second": 0.164
}