Qwen3-14B-30K-sft-nopacking / train_results.json
yueqis's picture
Upload train_results.json with huggingface_hub
f9d3c2c verified
raw
history blame contribute delete
219 Bytes
{
"epoch": 0.9968186638388123,
"total_flos": 412229836996608.0,
"train_loss": 0.3199658251823263,
"train_runtime": 20135.6606,
"train_samples_per_second": 1.498,
"train_steps_per_second": 0.012
}