Files
OLMoE-1B-7B-0125-sft-math7k…/train_results.json

8 lines
210 B
JSON
Raw Normal View History

{
"total_flos": 6812915466240.0,
"train_loss": 0.3203352055983779,
"train_runtime": 2102.8589,
"train_samples": 6851,
"train_samples_per_second": 9.774,
"train_steps_per_second": 0.077
}