Qwen2.5-1.5B-hl-false-v7 / all_results.json
Muennighoff's picture
Model save
dbedc48 verified
{
"total_flos": 0.0,
"train_loss": -0.017879978650705047,
"train_runtime": 58472.4999,
"train_samples": 12000,
"train_samples_per_second": 24.518,
"train_steps_per_second": 0.027
}