tulu-3-sft-mixture / train_results.json
sedrickkeh's picture
End of training
12e4f15 verified
raw
history blame contribute delete
206 Bytes
{
"epoch": 3.0,
"total_flos": 2883576618024960.0,
"train_loss": 0.5825082723410426,
"train_runtime": 26219.2762,
"train_samples_per_second": 33.609,
"train_steps_per_second": 0.066
}