© 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. 57600
    Prev
    Next
    INDEX
    Explanations

    Introductions

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens involved in the assistant’s self‐identification—especially the model’s name (“Vicuna”) and the header markers around that introduction.

    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
     Тому
    -0.07
     Aralık
    -0.07
    _relation
    -0.07
    (dc
    -0.06
    监听页面
    -0.06
     uc
    -0.06
     cassette
    -0.06
     asserts
    -0.06
    cak
    -0.06
    ‘
    -0.06
    POSITIVE LOGITS
     braking
    0.07
    ZERO
    0.06
    .isEmpty
    0.06
     wholly
    0.06
     bracket
    0.06
     legacy
    0.06
    (pre
    0.06
     départ
    0.06
     QWidget
    0.06
    '))↵↵
    0.06
    Activations Density 0.012%

    No Known Activations