=== Qwen3.6-711-RFA ===
arch: qwen3_5
codebook: [-127, -104, -83, -65, -49, -35, -22, -10, 1, 13, 25, 38, 53, 69, 89, 113]
rotation: hadamard16

[integrity] shards=6 missing=none
[integrity] index total_size=31.35GB  on-disk safetensors sum=31.35GB
[integrity] tensors: index=1577 headers=1577 OK

[coverage] quantized expert-mods=0 linear-mods=189 fp16-protected=0 copied=1007
[coverage] missing/bad: 3  e.g. model.language_model.layers.0.mlp.down_proj

[bpw] quant-target params=17.11B  rfa bytes=9.48GB  -> 4.431 bits/weight
[size] total=31.35GB = rfa 9.48GB + copied(bf16/other) 21.87GB

[fidelity] dequant vs source (relFro / SQNR):
   model.language_model.layers.32.mlp.down_proj  relFro=6.337%  SQNR=24.0 dB
   model.language_model.layers.32.mlp.gate_proj  relFro=6.327%  SQNR=24.0 dB
   model.language_model.layers.32.mlp.up_proj  relFro=6.325%  SQNR=24.0 dB
   model.language_model.layers.63.mlp.down_proj  relFro=6.321%  SQNR=24.0 dB
   model.language_model.layers.63.mlp.gate_proj  relFro=6.334%  SQNR=24.0 dB
   model.language_model.layers.63.mlp.up_proj  relFro=6.335%  SQNR=24.0 dB

DONE