SmollerLM2-10M-sft / all_results.json
mehmetkeremturkcan's picture
Model save
8767dd1 verified
raw
history blame contribute delete
251 Bytes
{
"epoch": 1.9998815423190064,
"total_flos": 1.83007472173056e+16,
"train_loss": 3.3040246668631688,
"train_runtime": 19562.1005,
"train_samples": 460142,
"train_samples_per_second": 5.178,
"train_steps_per_second": 1.295
}