© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 116341
    Prev
    Next
    INDEX
    Explanations

    skating boarding surfing tricks

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

    The neuron fires most strongly on words (and the surrounding punctuation) that are framed as highlighted terms—i.e. things put in quotation marks or joined by hyphens.

    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
    figer
    -0.80
     Determine
    -0.79
     kura
    -0.78
     memiliki
    -0.78
    ドラム
    -0.77
    ică
    -0.77
    рисо
    -0.77
    StringWriter
    -0.77
    VPWR
    -0.76
    誨
    -0.76
    POSITIVE LOGITS
     tricks
    1.05
     carve
    0.98
     estilo
    0.87
     carving
    0.83
     turns
    0.82
     style
    0.79
     stalled
    0.79
     technique
    0.78
     edge
    0.77
     aerial
    0.76
    Activations Density 0.013%

    No Known Activations