Qwen2.5-1.5B-hl-baseline-v27 / train_results.json
Muennighoff's picture
Model save
a3457e7 verified
{
"total_flos": 0.0,
"train_loss": 1264545902427.3843,
"train_runtime": 60559.9674,
"train_samples": 1200000,
"train_samples_per_second": 23.672,
"train_steps_per_second": 0.026
}