© 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. 50314
    Prev
    Next
    INDEX
    Explanations

    dice and card games

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

    This neuron selectively activates on technical identifiers and numeric constants—e.g. code variables, function names, and numbers.

    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
     symbolized
    -0.76
    ecirc
    -0.75
    tila
    -0.71
     resume
    -0.71
     цвета
    -0.71
     Architektur
    -0.71
     люд
    -0.70
    หมาย
    -0.69
     אחרי
    -0.68
     кусо
    -0.68
    POSITIVE LOGITS
     sena
    0.77
     quads
    0.76
    Flush
    0.75
     Four
    0.75
     Yacht
    0.74
    Trips
    0.73
    Four
    0.71
     same
    0.71
     Flush
    0.70
     jawa
    0.70
    Activations Density 0.010%

    No Known Activations