© 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. 89930
    Prev
    Next
    INDEX
    Explanations

    java class and types

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

    The neuron is specialized in spotting programming‐language punctuation and structural tokens (e.g. colons, parentheses, braces, comparison operators, and similar syntax markers).

    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
     APPELLATE
    -0.82
     Vorsitzende
    -0.75
     Dostupné
    -0.73
    策略
    -0.72
     TokenType
    -0.71
    typeof
    -0.71
     potuto
    -0.71
    ênero
    -0.71
     Snell
    -0.71
     Verlauf
    -0.70
    POSITIVE LOGITS
     Vector
    0.93
    Vector
    0.93
    String
    0.91
     short
    0.85
     String
    0.85
     vector
    0.84
    boolean
    0.83
    short
    0.82
     boolean
    0.82
    vectors
    0.82
    Activations Density 0.018%

    No Known Activations