Jeesup/Llama-3.2-1B-Instruct-safetyquant-c1

Llama-3.2-1B-Instruct — configuration c1 (W8A8)

Quantized variant used in the Safety-Aware Configuration-Conditioned LoRA study.

field	value
scheme	W8A8
method	smoothquant+gptq
simulated	no — native kernel
weight bits	8
activation bits	8
group / block size	None
symmetric	True
calibration	512 × 2048 tok, C4
calibration sha256	b5736155d5b24bf5eb15faae0eb4a742aea4a9d0532db9684085a49750606e19

The calibration set is byte-identical across every configuration so that calibration data cannot be confounded with quantization method. c2 vs c3 isolates method (GPTQ vs AWQ) at equal bit-width, and is the study's headline control: 0.0276 vs 0.0661 pooled ASR with non-overlapping 95% CIs.

Metrics, per-prompt judge verdicts and the full report: Jeesup/safety-quant-phase0.