Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
Dongwei's picture
Model save
c18e96e verified
raw
history blame contribute delete
200 Bytes
{
"total_flos": 0.0,
"train_loss": 0.0030148602681540288,
"train_runtime": 6716.3318,
"train_samples": 7500,
"train_samples_per_second": 1.117,
"train_steps_per_second": 0.01
}