© 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. 34-GEMMASCOPE-RES-131K
    7. 118712
    Prev
    Next
    INDEX
    Explanations

    operand type

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

    The neuron selectively responds to tokens that begin with an uppercase letter (proper names or section‐heading words).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     недель
    -0.70
     raised
    -0.70
     Loading
    -0.69
    parole
    -0.68
    isesti
    -0.68
    틱
    -0.67
    منبع
    -0.67
    OC
    -0.66
    Branch
    -0.64
     якості
    -0.63
    POSITIVE LOGITS
    gons
    0.68
    外し
    0.68
    分け
    0.68
     crustace
    0.67
    tät
    0.65
     cranberries
    0.65
     flaky
    0.64
    श्न
    0.63
    piele
    0.63
    ている
    0.63
    Activations Density 0.067%

    No Known Activations