© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 73074
    Prev
    Next
    INDEX
    Explanations

    JavaScript prototype and function definitions

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

    The neuron fires on code‐definition tokens—class/type names, extensions, prototypes, etc.—marking places where types or object prototypes are declared or referenced.

    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
     XXXVI
    -1.38
    MYSQL
    -1.38
    гают
    -1.29
    ·
    -1.27
     LXXX
    -1.27
     LEY
    -1.27
     XXIX
    -1.27
     XXVI
    -1.26
     XXVIII
    -1.25
     XXXIII
    -1.25
    POSITIVE LOGITS
     to
    2.50
     at
    1.25
     упаковка
    1.24
     into
    1.22
     vielfäl
    1.20
     time
    1.16
     applies
    1.16
     you
    1.15
     new
    1.14
     uses
    1.14
    Activations Density 0.030%

    No Known Activations