Qwen2.5-Math-7B-s1k / train_results.json
flyingbugs's picture
Model save
504d30d verified
raw
history blame contribute delete
216 Bytes
{
"total_flos": 8.363833783064986e+16,
"train_loss": 1.3066247133981614,
"train_runtime": 6202.6685,
"train_samples": 1000,
"train_samples_per_second": 0.266,
"train_steps_per_second": 0.017
}