Files
mistral-7b-grok/train_results.json

8 lines
195 B
JSON
Raw Normal View History

{
"epoch": 1.0,
"train_loss": 0.9725383741046311,
"train_runtime": 5277.8235,
"train_samples": 211055,
"train_samples_per_second": 26.46,
"train_steps_per_second": 0.103
}