© 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. 127573
    Prev
    Next
    INDEX
    Explanations

    javascript variable declarations

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

    The neuron fires on numeric literals (and their surrounding assignment or indexing tokens) in code.

    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
    ou
    -1.37
    -
    -1.34
    uc
    -1.30
    u
    -1.30
    na
    -1.28
     would
    -1.27
     riuscito
    -1.27
     based
    -1.25
     one
    -1.25
     p
    -1.25
    POSITIVE LOGITS
    thof
    1.58
     jät
    1.55
     dôle
    1.52
     täh
    1.52
     sitten
    1.50
     geschenk
    1.48
     hvilke
    1.48
     Chapitre
    1.47
    itization
    1.46
    にとっては
    1.46
    Activations Density 0.060%

    No Known Activations