© 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. 41251
    Prev
    Next
    INDEX
    Explanations

    movie scenes

    np_max-act · gemini-2.0-flash

    The neuron is most strongly activated by words referring to kissing or tongue contact.

    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
    .option
    -0.07
     hotspot
    -0.07
    目的
    -0.07
     Ninth
    -0.06
     stepped
    -0.06
    尽管
    -0.06
     slander
    -0.06
     Akt
    -0.06
    975
    -0.06
    ंधन
    -0.06
    POSITIVE LOGITS
    $input
    0.07
    .Maximum
    0.06
    ;background
    0.06
    ۲۱
    0.06
    ARENT
    0.06
     теор
    0.06
    _Items
    0.06
     QIcon
    0.06
    ACCEPT
    0.06
    ：<
    0.06
    Activations Density 0.036%

    No Known Activations