llama-8b-instruct-rdpo-full / train_results.json
Sean13's picture
Model save
3802569 verified
raw
history blame contribute delete
216 Bytes
{
"epoch": 1.0,
"total_flos": 0.0,
"train_loss": 0.386704023147368,
"train_runtime": 9257.6719,
"train_samples": 59876,
"train_samples_per_second": 6.468,
"train_steps_per_second": 0.101
}