mistral_llama_2_code_math_3_full / train_results.json
CharlesLi's picture
Model save
12ad0d4 verified
raw
history blame contribute delete
242 Bytes
{
"epoch": 0.9777777777777777,
"total_flos": 2250831298560.0,
"train_loss": 0.6727288040247831,
"train_runtime": 125.1337,
"train_samples": 3980,
"train_samples_per_second": 5.682,
"train_steps_per_second": 0.176
}