gemma-4-31b-heretic-fp8 / local_settings.json
dev-willbird1936's picture
Extend chat context and response limits
b953c04 verified
Raw History Blame Contribute Delete
343 Bytes
{
"model_id": "0ppxnhximxr/gemma-4-31B-it-heretic-FP8-Dynamic",
"system_prompt": "You are a precise and helpful assistant. State uncertainty clearly.",
"thinking": false,
"max_new_tokens": 1024,
"temperature": 1.0,
"top_p": 0.95,
"top_k": 64,
"repetition_penalty": 1.05,
"max_input_tokens": 16384,
"history_messages": 16
}