qwen2.5-1.5b-sft3-25-4 / train_results.json
hZzy's picture
Model save
45b0bcd verified
{
"epoch": 10.0,
"total_flos": 7.930133386100736e+16,
"train_loss": 2.201316048024775,
"train_runtime": 48928.7509,
"train_samples": 118004,
"train_samples_per_second": 5.948,
"train_steps_per_second": 0.019
}