Qwen2.5-1.5B-hl-baseline-v22 / all_results.json
Muennighoff's picture
Model save
5c65324 verified
raw
history blame contribute delete
202 Bytes
{
"total_flos": 0.0,
"train_loss": 69.64168917144616,
"train_runtime": 28121.2965,
"train_samples": 1200000,
"train_samples_per_second": 50.979,
"train_steps_per_second": 0.057
}