© 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. 33619
    Prev
    Next
    INDEX
    Explanations

    book chapter or episode numbers

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron detects reference and citation markers—tokens like book titles, section or chapter numbers (Arabic and Roman numerals), and similar document‐structure identifiers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     pasaran
    -1.03
    ײ
    -1.02
    ∛
    -0.97
    रीदारी
    -0.96
     who
    -0.95
    หมือน
    -0.95
    .
    -0.94
    ScrollBar
    -0.94
     legislat
    -0.93
     preexisting
    -0.92
    POSITIVE LOGITS
    說的
    1.16
     apropiado
    1.08
    atics
    1.06
     yeux
    1.05
    ーい
    1.04
     rác
    1.03
     شخصية
    1.02
    籲
    1.02
    onians
    0.98
    不喜欢
    0.97
    Activations Density 0.013%

    No Known Activations