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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 58541
    Prev
    Next
    INDEX
    Explanations

    private messages

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of private messaging (e.g. “PM,” “private message,” “send a PM”).

    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
    gons
    -0.07
    .props
    -0.07
     />,↵
    -0.06
     //*
    -0.06
    attle
    -0.06
     Σύ
    -0.06
    /Library
    -0.06
    ельзя
    -0.06
     порушення
    -0.06
    uell
    -0.06
    POSITIVE LOGITS
     buc
    0.07
     każ
    0.06
    -position
    0.06
    coordinate
    0.06
     tanto
    0.06
    DM
    0.06
     treasury
    0.06
    ازد
    0.06
     ettik
    0.06
    _analysis
    0.06
    Activations Density 0.008%

    No Known Activations