© 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. 33684
    Prev
    Next
    INDEX
    Explanations

    programming related keywords and syntactic structures

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

    The neuron fires most strongly on standalone identifiers immediately before an opening parenthesis (i.e. function or macro names 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_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
     into
    -1.45
     for
    -1.42
    ="">
    -1.04
     from
    -1.02
     against
    -0.94
    勉
    -0.91
     Deutschlands
    -0.90
     about
    -0.90
    了一会儿
    -0.90
     investigación
    -0.90
    POSITIVE LOGITS
     you
    1.24
    said
    1.16
     ciment
    1.10
     peluche
    1.09
     iny
    1.05
    錞
    1.05
     вам
    1.03
     mètre
    1.02
     Valentín
    1.00
    attet
    1.00
    Activations Density 0.000%

    No Known Activations