Qwen-sft-la-v0.1 / train_results.json
sci-m-wang's picture
Upload 7 files
9b7fea5 verified
raw
history blame
223 Bytes
{
"epoch": 9.996258885147775,
"total_flos": 1.2597949543307674e+18,
"train_loss": 0.7376079930516775,
"train_runtime": 29450.1993,
"train_samples_per_second": 1.815,
"train_steps_per_second": 0.113
}