32b_pymethods2test-large_Qwen3-32B / generation_config.json
atutej's picture
Best train reward at step 17 (reward=0.580, pass@8=0.625). Uploading nearest HF export at step 20 (reward=0.465). FSDP2 checkpoint at step 18 also available but requires consolidation.
5573186 verified
raw
history blame contribute delete
214 Bytes
{
"bos_token_id": 151643,
"do_sample": true,
"eos_token_id": [
151645,
151643
],
"pad_token_id": 151643,
"temperature": 0.6,
"top_k": 20,
"top_p": 0.95,
"transformers_version": "4.57.6"
}