© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 14030
    Prev
    Next
    INDEX
    Explanations

    sexually explicit content

    np_max-act · gemini-2.0-flash

    explicit sexual content in writing.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron lights up on the upbeat, motivational wording often used in social-media captions—terms like “vibes,” “life,” “boss,” “today,” etc., as found in short, inspirational Instagram-style slogans.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Inst
    -0.07
    	↵	↵
    -0.06
     Atom
    -0.06
     Gott
    -0.06
     ST
    -0.06
     폴
    -0.06
     Punk
    -0.06
     strengthened
    -0.06
    �
    -0.06
     pygame
    -0.06
    POSITIVE LOGITS
    कन
    0.06
    bě
    0.06
    queueReusableCell
    0.06
    Until
    0.06
    catid
    0.06
     expenditures
    0.06
     грн
    0.06
    оратив
    0.06
    鋼
    0.06
    charge
    0.06
    Activations Density 0.394%

    No Known Activations