DeepScaleR-1.5B-L36-RollE6 / all_results.json
ZMC2019's picture
Model save
c3a7aed verified
{
"total_flos": 2.5527473493612954e+17,
"train_loss": 0.3212856304745714,
"train_runtime": 7845.5794,
"train_samples": 93733,
"train_samples_per_second": 1.451,
"train_steps_per_second": 0.091
}