© 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 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. 56903
    Prev
    Next
    INDEX
    Explanations

    Personal information requests

    np_max-act · gemini-2.0-flash

    This neuron fires on tokens where the agent is requesting or verifying the customer’s personal or account information (e.g., account number, phone number, email, address).

    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
     halt
    -0.07
     contraception
    -0.07
    γα
    -0.06
     Tv
    -0.06
     práci
    -0.06
    rende
    -0.06
     قدم
    -0.06
     progressDialog
    -0.06
     ttl
    -0.06
     Mang
    -0.06
    POSITIVE LOGITS
    getParameter
    0.09
     allied
    0.08
     slab
    0.07
    社会
    0.07
    顔
    0.07
    FromFile
    0.06
    {}↵↵
    0.06
    StyleSheet
    0.06
    leanor
    0.06
    _valor
    0.06
    Activations Density 0.019%

    No Known Activations