mistral_llama_2_code_math_0_full / train_results.json
CharlesLi's picture
Model save
24db9d2 verified
raw
history blame contribute delete
222 Bytes
{
"epoch": 1.0,
"total_flos": 261724569600.0,
"train_loss": 1.1633186340332031,
"train_runtime": 18.98,
"train_samples": 480,
"train_samples_per_second": 4.689,
"train_steps_per_second": 0.158
}