Qwen2.5-1.5B-hl-baseline-v4 / all_results.json
Muennighoff's picture
Model save
0ae06ea verified
{
"total_flos": 0.0,
"train_loss": 13.440545084960068,
"train_runtime": 65092.8727,
"train_samples": 12000,
"train_samples_per_second": 22.024,
"train_steps_per_second": 0.025
}