Qwen2.5-1.5B-Open-R1-Distill / train_results.json
a-F1's picture
Model save
ecee1e3 verified
raw
history blame
214 Bytes
{
"total_flos": 7.883277455484518e+16,
"train_loss": 1.23619465266957,
"train_runtime": 1783.7305,
"train_samples": 1000,
"train_samples_per_second": 1.374,
"train_steps_per_second": 0.172
}