c1_math_0d_4s_3k / train_results.json
ryanmarten's picture
Upload model
7f66460 verified
raw
history blame contribute delete
206 Bytes
{
"epoch": 7.0,
"total_flos": 7.881049861301207e+17,
"train_loss": 0.3857915287281012,
"train_runtime": 8985.043,
"train_samples_per_second": 2.462,
"train_steps_per_second": 0.026
}