© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 68625
    Prev
    Next
    INDEX
    Explanations

    period

    np_max-act · gemini-2.0-flash

    This neuron flags vulgar or profane tokens (e.g. “fuck,” “slut,” “bullshit”).

    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
    appropriate
    -0.07
    -fashioned
    -0.06
     Κά
    -0.06
     advantage
    -0.06
    _waiting
    -0.06
    cala
    -0.06
    udem
    -0.06
     "-",
    -0.06
     Valley
    -0.06
    -to
    -0.06
    POSITIVE LOGITS
    ("{\"
    0.07
    %</
    0.06
    _excel
    0.06
    }"↵
    0.06
     المست
    0.06
     condu
    0.06
     batter
    0.06
     mücadele
    0.06
    ?
    0.06
    olicitud
    0.06
    Activations Density 0.221%

    No Known Activations