qwen3-4b-structeval-v6-w4a16 / generation_config.json
miitarou's picture
v6: Qwen3-4B-Instruct W4A16 quantized via llm-compressor
8ee6f41 verified
raw
history blame contribute delete
213 Bytes
{
"bos_token_id": 151643,
"do_sample": true,
"eos_token_id": [
151645,
151643
],
"pad_token_id": 151643,
"temperature": 0.7,
"top_k": 20,
"top_p": 0.8,
"transformers_version": "4.57.6"
}