PEFT
Safetensors
qwen2
alignment-handbook
trl
dpo
Generated from Trainer
khongtrunght's picture
Model save
e965a13 verified
raw
history blame
No virus
232 Bytes
{
"epoch": 0.9995419147961521,
"total_flos": 0.0,
"train_loss": 0.4306906612229719,
"train_runtime": 8689.0202,
"train_samples": 34924,
"train_samples_per_second": 4.019,
"train_steps_per_second": 0.126
}