llama3_orpo_best_entropy / train_results.json
yakazimir's picture
Model save
42a6f26 verified
raw
history blame contribute delete
232 Bytes
{
"epoch": 0.9989071038251366,
"total_flos": 0.0,
"train_loss": 3.4314852449513107,
"train_runtime": 5934.8281,
"train_samples": 58558,
"train_samples_per_second": 9.867,
"train_steps_per_second": 0.077
}