© 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. 33483
    Prev
    Next
    INDEX
    Explanations

    Q: question marker

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

    The neuron fires on code identifier tokens—i.e. variable, function, and class names (especially in camelCase or PascalCase).

    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
     (
    -1.56
     from
    -1.38
     in
    -1.37
     från
    -1.34
    july
    -1.34
     emplois
    -1.30
    .
    -1.23
     his
    -1.21
    ‘
    -1.20
     februari
    -1.19
    POSITIVE LOGITS
    Ilustra
    1.66
    Ogni
    1.61
    しかし
    1.57
    rivial
    1.57
    تفسير
    1.54
    Nasc
    1.54
     figliu
    1.54
    Scopri
    1.53
    Legenda
    1.52
    ὴν
    1.52
    Activations Density 0.020%

    No Known Activations