Qwen2.5-0.5B-Open-R1-Distill / train_results.json
jdqqjr's picture
Model save
b18b40a verified
raw
history blame contribute delete
244 Bytes
{
"epoch": 0.999537251272559,
"total_flos": 37979261239296.0,
"train_loss": 0.9636553647783067,
"train_runtime": 8704.1469,
"train_samples": 16610,
"train_samples_per_second": 4.965,
"train_steps_per_second": 0.103
}