© 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. 58380
    Prev
    Next
    INDEX
    Explanations

    person

    np_max-act · gemini-2.0-flash

    The neuron detects references to “individual” (and by extension “group”/“people”)—i.e. mentions of individuals or groups in disclaimer-style statements.

    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
     Fransa
    -0.07
    iltere
    -0.07
    CLU
    -0.07
     Fransız
    -0.06
    ๋
    -0.06
     Ren
    -0.06
    rimon
    -0.06
     金
    -0.06
    Nota
    -0.06
    affles
    -0.06
    POSITIVE LOGITS
     ValidationError
    0.06
    ζα
    0.06
    Nov
    0.06
     dermat
    0.06
     asthma
    0.06
     airl
    0.06
    operand
    0.06
     forget
    0.06
     accessed
    0.06
     tolerated
    0.06
    Activations Density 0.013%

    No Known Activations