© 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. 106943
    Prev
    Next
    INDEX
    Explanations

    average points per game

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

    The neuron fires on numeric tokens—especially multi‐digit statistics (e.g. point totals, rebounds, averages)—in the text.

    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
    ˎ
    -0.83
    olella
    -0.82
     four
    -0.80
    Biblia
    -0.79
     lebaran
    -0.79
     mẽ
    -0.76
     poignant
    -0.76
     jenis
    -0.76
    ｖ
    -0.76
    中には
    -0.75
    POSITIVE LOGITS
    Avg
    1.41
     per
    1.35
     averaging
    1.28
     averaged
    1.27
     Avg
    1.23
     averages
    1.21
     average
    1.21
     avg
    1.16
    平均
    1.16
    avg
    1.16
    Activations Density 0.005%

    No Known Activations