© 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. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128243
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on numeric literal tokens (especially single‐digit numbers) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    mathematical expressions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    knapp
    -1.02
    kreuz
    -1.00
     합니다
    -0.95
    متن
    -0.95
    翻攝
    -0.93
    Effec
    -0.91
    Spons
    -0.90
     nuevamente
    -0.90
    AGEM
    -0.89
     Yeni
    -0.89
    POSITIVE LOGITS
     even
    1.11
     silnika
    1.08
     filós
    1.04
    cologist
    1.02
    strator
    0.99
     vrijdag
    0.99
    ͑
    0.99
    Facebook
    0.98
     buy
    0.98
    thalam
    0.97
    Activations Density 0.052%

    No Known Activations