© 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. 91017
    Prev
    Next
    INDEX
    Explanations

    Microsoft Surface, Jan Op, Analysis

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

    The neuron fires on C/C++ code syntax tokens—especially include directives, JNI‐related identifiers, scope operators (::), angle brackets (< >), underscores, semicolons, and other punctuation used in header and template declarations.

    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
    DTD
    -0.83
     Marken
    -0.74
    Behaviour
    -0.74
     agrup
    -0.73
     clazz
    -0.71
    behaviour
    -0.71
     memor
    -0.70
    Фак
    -0.69
     Goodman
    -0.69
     prípade
    -0.69
    POSITIVE LOGITS
    graphe
    0.81
     isl
    0.80
     Norwalk
    0.79
    ЕТ
    0.78
    беле
    0.78
     Loan
    0.77
     kiosk
    0.76
     Naira
    0.75
     Lucknow
    0.75
     Condens
    0.75
    Activations Density 0.002%

    No Known Activations