Qwen-2.5-7B / train_results.json
Dongwei's picture
Model save
4c0d020 verified
raw
history blame contribute delete
201 Bytes
{
"total_flos": 0.0,
"train_loss": 0.018138731037502857,
"train_runtime": 11014.2043,
"train_samples": 7500,
"train_samples_per_second": 0.681,
"train_steps_per_second": 0.006
}