Correct the Sigourney Weaver resemblance figure

RESEARCH.md

@@ -128,7 +128,7 @@ At nynxz's sampler settings (CFG 2.5, no SLG/TSR; relative to SFT at the same se
 
 ## 6. Beyond MOSS: the same lesson on a character LoRA
 
-The same question — keep a concept when asked for, change nothing else — appears with character LoRAs ([DeTangle](https://matlowai.github.io/DeTangle/)). Its "A-side surgery" refits one LoRA factor from captured activations, the same family as the Gram solve, and shows the same weakness. On two of eight test scenes (a sci-fi "space station" and a fantasy "wizard workshop", the only prompts ending in "Photorealistic live-action science-fiction/fantasy film still.") the character's likeness collapses after surgery and the face falls back to the base model's own default for the scene; for the space station that default resembles the model's own rendering of Sigourney Weaver (face match 0.50; the model's named-actress renders match each other at 0.9).
+The same question — keep a concept when asked for, change nothing else — appears with character LoRAs ([DeTangle](https://matlowai.github.io/DeTangle/)). Its "A-side surgery" refits one LoRA factor from captured activations, the same family as the Gram solve, and shows the same weakness. On two of eight test scenes (a sci-fi "space station" and a fantasy "wizard workshop", the only prompts ending in "Photorealistic live-action science-fiction/fantasy film still.") the character's likeness collapses after surgery and the face falls back to the base model's own default for the scene; for the space station that default face looks Ripley-like, but measured against the model's concept of Sigourney Weaver (her studio portraits rendered by name) it is a different person: 0.27 for the base model's default woman and 0.33 after surgery, against 0.75–0.82 for her own renders. The base model's default woman leans slightly toward Weaver in every scene (beach and arcade too), so there is no sci-fi-specific Sigourney association; surgery simply hands the face back to the model's generic woman, and styling does the rest.
 
 Probe results (424 renders, same protocol as the DeTangle study):
 - Only transformer blocks 8–15 matter: restoring the original LoRA there brings likeness back (0.36–0.55); every other block window, including the first, does nothing.