© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 6991
    Prev
    Next
    INDEX
    Explanations

    I wonder if

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

    This neuron fires strongly on the word “if” (and other basic copulas like “is”/“was”), i.e. it detects conditional or interrogative “if” clauses.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    k
    -2.23
    c
    -2.06
    l
    -2.03
    m
    -1.95
    g
    -1.91
    te
    -1.89
    ns
    -1.88
    r
    -1.85
    b
    -1.85
    se
    -1.84
    POSITIVE LOGITS
     in
    1.87
     as
    1.64
    Ⅳ
    1.63
    adget
    1.53
     للمعارف
    1.52
     also
    1.51
    趼
    1.47
    летка
    1.46
    　（
    1.45
     vienas
    1.45
    Activations Density 0.009%

    No Known Activations