zephyr-7b-dpo-lora / train_results.json
LaoRay's picture
Model save
1fa9dbe verified
{
"epoch": 19.84,
"total_flos": 0.0,
"train_loss": 0.13093096087141606,
"train_runtime": 15450.2645,
"train_samples": 1000,
"train_samples_per_second": 1.294,
"train_steps_per_second": 0.08
}