GRM-1.5b / all_results.json
DedeProGames's picture
Import open-thoughts/OpenThinker3-1.5B as GRM-1.5b
b185feb verified
raw
history blame contribute delete
211 Bytes
{
"epoch": 7.0,
"total_flos": 1.1955108328583987e+17,
"train_loss": 0.9723419804781719,
"train_runtime": 594145.3794,
"train_samples_per_second": 14.138,
"train_steps_per_second": 0.055
}