Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 130528
    Prev
    Next
    INDEX
    Explanations

    This neuron responds to explicit mentions of incest, particularly sexual contexts involving a mother and her son.

    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
     Elli
    -0.07
    (field
    -0.06
    .Url
    -0.06
     Kum
    -0.06
    ове
    -0.06
     StringUtils
    -0.06
     SPORT
    -0.06
    _Part
    -0.06
     Companion
    -0.06
     относ
    -0.06
    POSITIVE LOGITS
    Plans
    0.07
    unts
    0.06
    native
    0.06
    _opacity
    0.06
    ώσεις
    0.06
     Geological
    0.06
     tion
    0.06
    ATING
    0.06
    ньої
    0.06
    '^$',
    0.06
    Activations Density 0.034%

    No Known Activations