© 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. 57732
    Prev
    Next
    INDEX
    Explanations

    collegiate sports

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

    The neuron strongly activates on numeric tokens—especially digits (and hyphens) used in dates, years, seasons, or scores.

    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
     chłop
    -0.78
    鳃
    -0.77
     Undang
    -0.74
    chufe
    -0.72
     schoolboy
    -0.71
    ஞ்ச
    -0.71
     clientèle
    -0.71
    платье
    -0.71
    小学
    -0.69
     Redskins
    -0.69
    POSITIVE LOGITS
     NA
    1.42
     campus
    1.03
     collegiate
    1.00
     γυναι
    0.98
     universidad
    0.97
     college
    0.95
     Viter
    0.94
    RV
    0.93
     NCC
    0.92
    NA
    0.92
    Activations Density 0.068%

    No Known Activations