Gemma-2-9B-It-SFT / all_results.json
chchen's picture
End of training
17ba8fd verified
raw
history blame contribute delete
362 Bytes
{
"epoch": 2.986666666666667,
"eval_loss": 0.14741086959838867,
"eval_runtime": 6.1172,
"eval_samples_per_second": 16.347,
"eval_steps_per_second": 8.174,
"total_flos": 1.7300235104796672e+16,
"train_loss": 0.8548009863921574,
"train_runtime": 612.7985,
"train_samples_per_second": 4.406,
"train_steps_per_second": 0.274
}