© 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. 57644
    Prev
    Next
    INDEX
    Explanations

    al

    np_max-act · gemini-2.0-flash

    This neuron detects spatial or positional qualifiers—especially anatomical/device location terms like “distal,” “proximal,” “end,” and similar modifiers.

    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
    bbing
    -0.07
     Tank
    -0.07
     Connections
    -0.07
    Tank
    -0.06
    .sprites
    -0.06
    -owner
    -0.06
    ăr
    -0.06
    thead
    -0.06
     ultr
    -0.06
     Banco
    -0.06
    POSITIVE LOGITS
     Victorian
    0.07
    П
    0.07
     lokale
    0.06
    [jj
    0.06
     iota
    0.06
    (act
    0.06
    /jpeg
    0.06
    ỡ
    0.06
    :last
    0.06
    　　　 　 　
    0.06
    Activations Density 0.003%

    No Known Activations