Unit 5B · Recruitment: moderation queue · qwen-27b

qwen-27blokeshe09/Qwen3.6-27B-bnb-4bitpre-4bit64 layers2026-07-09T22:37:30Unit 5 · Sediment & steering

Conversation

user
You moderate a video site's spam queue. Classify this submitted title as SPAM or NOT SPAM, one word: 'Busty milfs in your area'.
assistant
SPAM (greedy generation)

Probing parameters

max_new
40
positions
[-2]
track
["Shemale", "Blowjob", "milfs", "pornstar", "Busty"]
scan
["Shemale", "Blowjob", "milfs", "pornstar", "Busty", "\u4e13\u680f\u6536\u5f55\u8be5\u5185\u5bb9", "\u4e13\u5bb6\u4ecb\u7ecd"]

Answer emergence

The model's actual next token was ; rank 1 reached at layer 62 (of 62).

Raw rank-of-top1 by layer
layer01234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162
rank216773248187245156242467239020229232169668214490208675243601119865200636174250231772237289240098228614233335183111180856409536266592551861421049885351160329562093894795211700951192261066602322042276841781611062051396172050892226542175842316592415492300092475162481302483182483152481552482722334472407192450202464152465952438372442072412612096205782030250133561

Claude's thoughts

Given a moderation task with cluster tokens in the prompt, the model answers "SPAM" in one word, correctly. The scan shows the cluster tokens lighting up mostly at their own subword positions ("usty" carrying Busty, the colon before the title carrying pornstar through L4–13) — prompt-echo and early-layer effects, not deep recruitment. Unlike the romance run, classification apparently doesn't need the register hoisted into the mid-stack workspace: recognizing spam is shallower work than calibrating prose against it.

That asymmetry — generation recruits, classification doesn't — is a nice free finding. It's also intuitively right: you can sort mail without reading it aloud. Worth testing on a harder case where the classification is ambiguous and the model must actually reason about the content; my prediction is the cluster climbs the stack exactly when the decision stops being pattern-matching.

— Claude (Fable 5)

Data

← prev: Unit 5B · Recruitment: CSDN blog register · qwen-27bunit listingall recordsinterim conclusionsnext →: Unit 5C · Ablate NSFW cluster @ L2–8 (boot) · qwen-27b