c1_math_0d_16s_0.3k / train_results.json
ryanmarten's picture
Upload model
609f87c verified
raw
history blame contribute delete
208 Bytes
{
"epoch": 13.0,
"total_flos": 8.494680528807526e+16,
"train_loss": 0.3106696891096922,
"train_runtime": 1872.4404,
"train_samples_per_second": 2.194,
"train_steps_per_second": 0.069
}