© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 78212
    Prev
    Next
    INDEX
    Explanations

    International Jamaican

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

    The neuron specializes in spotting tokens composed chiefly of digits (e.g. numeric IDs, version numbers, or numeric parts of filenames) within code comment blocks.

    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
    assé
    -0.74
    SuppressMessage
    -0.72
     Metropol
    -0.72
    乾
    -0.71
     nghiệp
    -0.71
    如意
    -0.70
    作曲
    -0.69
     seas
    -0.69
     dorm
    -0.69
     capacitor
    -0.68
    POSITIVE LOGITS
     TIM
    0.93
    ophila
    0.88
    TIM
    0.79
    rists
    0.73
    eless
    0.71
    Hindu
    0.71
     Kail
    0.71
     Kristina
    0.71
    (")
    0.70
    aians
    0.70
    Activations Density 0.001%

    No Known Activations