PEFT
Safetensors
mistral
alignment-handbook
trl
dpo
Generated from Trainer
File size: 194 Bytes
f9e6f19
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
{
    "epoch": 1.0,
    "train_loss": 0.150745850255092,
    "train_runtime": 48820.8334,
    "train_samples": 38447,
    "train_samples_per_second": 0.788,
    "train_steps_per_second": 0.012
}