© 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. 10-GEMMASCOPE-RES-131K
    7. 45330
    Prev
    Next
    INDEX
    Explanations

    points and assists

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

    The neuron responds to numeric game statistics—especially point and score figures in sports recaps.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     that
    -1.70
    opies
    -1.48
     meninos
    -1.45
     blessures
    -1.45
    dotti
    -1.42
     renk
    -1.42
    ldquo
    -1.42
    ⾞
    -1.42
    阗
    -1.39
    ctable
    -1.38
    POSITIVE LOGITS
     at
    1.91
     of
    1.70
     then
    1.50
     assist
    1.45
     help
    1.41
     helps
    1.41
     kept
    1.33
     aid
    1.33
     helped
    1.31
     alone
    1.30
    Activations Density 0.024%

    No Known Activations