© 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. 57591
    Prev
    Next
    INDEX
    Explanations

    npm package metadata

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric version strings (e.g. “0.4.0”, “2.4.13”, etc.), effectively spotting semantic-version patterns.

    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
     causing
    -0.06
     garnered
    -0.06
    .xaml
    -0.05
    itched
    -0.05
     emitting
    -0.05
    _dt
    -0.05
     derivative
    -0.05
    ial
    -0.05
     machining
    -0.05
     wording
    -0.05
    POSITIVE LOGITS
     исследования
    0.09
     zad
    0.07
     Serbia
    0.07
    (ListNode
    0.07
     Gew
    0.07
    кова
    0.07
    Council
    0.07
    Biz
    0.06
    coli
    0.06
     Denmark
    0.06
    Activations Density 0.007%

    No Known Activations