Qwen2.5-1.5B-hl-baseline / all_results.json
Muennighoff's picture
Model save
1a6a37a verified
{
"total_flos": 0.0,
"train_loss": -0.03323389737907974,
"train_runtime": 206541.0301,
"train_samples": 12000,
"train_samples_per_second": 6.941,
"train_steps_per_second": 0.008
}