© 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. 58450
    Prev
    Next
    INDEX
    Explanations

    S

    np_max-act · gemini-2.0-flash

    This neuron flags numeric data tokens (especially decimal numbers) in scientific‐style text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    tokens that signal structured technical text—such as code, logs, tables, and camelCase/class/protocol identifiers with brackets, braces, or column delimiters.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
    游
    -0.07
    .ball
    -0.07
    英雄
    -0.07
     lạc
    -0.07
    ']",
    -0.07
    柱
    -0.06
    анд
    -0.06
    nego
    -0.06
     debugger
    -0.06
    _THREAD
    -0.06
    POSITIVE LOGITS
     ओवर
    0.07
    urable
    0.07
    .Username
    0.07
     maliyet
    0.06
    prs
    0.06
     dab
    0.06
     datos
    0.06
    ABS
    0.06
    OFFSET
    0.06
    QUIRES
    0.06
    Activations Density 8.319%

    No Known Activations