darkmaniac7's picture
v3: KL-distilled 0.6B from Qwen3-8B (10K samples, KL=0.339, +41% uplift on SM8850)
1f8c190 verified
raw
history blame contribute delete
211 Bytes
{
"llm_model": "llm.mnn",
"llm_weight": "llm.mnn.weight",
"backend_type": "cpu",
"thread_num": 2,
"precision": "low",
"memory": "low",
"sampler_type": "greedy",
"power": "high"
}