Qwen2.5-1.5B-Open-R1-Distill / train_results.json
zyl2023's picture
Model save
32111d0 verified
raw
history blame
215 Bytes
{
"total_flos": 2444245755494400.0,
"train_loss": 0.42360519842668015,
"train_runtime": 82605.1286,
"train_samples": 93733,
"train_samples_per_second": 2.077,
"train_steps_per_second": 0.13
}