Qwen2.5-0.5B-Open-R1-Distill / train_results.json
Blancy's picture
Model save
8661df7 verified
raw
history blame
209 Bytes
{
"total_flos": 4079631728640.0,
"train_loss": 1.5182941075029044,
"train_runtime": 127.2004,
"train_samples": 1000,
"train_samples_per_second": 72.728,
"train_steps_per_second": 1.14
}