© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 80809
    Prev
    Next
    INDEX
    Explanations

    numeric type conversions

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

    The neuron responds to mentions of programming types—particularly the words “type” and related type‐declaration keywords (e.g. “typedef,” “primitive type”) in code contexts.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    versite
    -0.77
    阳台
    -0.75
     tahap
    -0.72
    ără
    -0.70
    liani
    -0.68
    Sche
    -0.68
    粮
    -0.68
    涧
    -0.68
     Pana
    -0.68
    били
    -0.67
    POSITIVE LOGITS
     numeric
    1.38
    Integral
    1.20
     Integral
    1.15
     integral
    1.14
     Numeric
    1.03
    numeric
    1.03
    INTEGR
    1.03
     numerical
    1.02
    Numeric
    1.02
    numerical
    0.99
    Activations Density 0.077%

    No Known Activations