© 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. 57149
    Prev
    Next
    INDEX
    Explanations

    Sec

    np_max-act · gemini-2.0-flash

    This neuron activates on section or subsection references (e.g., “Sec 1.484375,” numeric identifiers in legal or technical citations).

    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
    중
    -0.06
    repeat
    -0.06
    .layouts
    -0.06
    增加
    -0.06
     評
    -0.06
     );
↵
↵
    -0.06
    ?“↵↵
    -0.06
    umericUpDown
    -0.06
     수정
    -0.06
    とは
    -0.06
    POSITIVE LOGITS
    Char
    0.07
     Descriptor
    0.07
     ka
    0.06
     Vari
    0.06
    SCREEN
    0.06
    estation
    0.06
    emap
    0.06
    -find
    0.06
    /auth
    0.06
    \Catalog
    0.06
    Activations Density 0.079%

    No Known Activations