Qwen2.5-1.5B-Open-R1-Distill / train_results.json
amrn's picture
Model save
78d4a52 verified
raw
history blame
230 Bytes
{
"epoch": 1.0,
"total_flos": 76916824473600.0,
"train_loss": 0.8026671296746067,
"train_runtime": 1041.4464,
"train_samples": 16610,
"train_samples_per_second": 20.75,
"train_steps_per_second": 0.162
}