DrGRPO / all_results.json
LLucass's picture
Model save
136a08c verified
raw
history blame contribute delete
199 Bytes
{
"total_flos": 0.0,
"train_loss": 0.03577750400174409,
"train_runtime": 8397.3301,
"train_samples": 7000,
"train_samples_per_second": 1.143,
"train_steps_per_second": 0.012
}