LIMO-R1-Distill-Qwen-7B / adapter /all_results.json
RDson's picture
update files
fde772f
raw
history blame contribute delete
222 Bytes
{
"epoch": 0.9987760097919217,
"total_flos": 1.7231028658783027e+17,
"train_loss": 0.4889225305295458,
"train_runtime": 5417.273,
"train_samples_per_second": 0.151,
"train_steps_per_second": 0.038
}