© 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. 22-GEMMASCOPE-RES-131K
    7. 36629
    Prev
    Next
    INDEX
    Explanations

    Red Sox

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

    The neuron strongly activates on tokens that include non-alphabetic characters—e.g. digits, hyphens, periods, or apostrophes—such as “high-profile,” “16th,” “Fla.,” or “NFL’s.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     XXVIII
    -1.17
     dora
    -1.07
     rigue
    -1.04
     :{
    -0.94
     rafael
    -0.94
    RRRR
    -0.93
     prada
    -0.92
     boute
    -0.91
     colonnes
    -0.91
     venice
    -0.90
    POSITIVE LOGITS
     if
    1.09
     from
    1.02
    ご了承
    0.96
    AN
    0.94
    比起
    0.93
    اشة
    0.90
    fits
    0.89
     następ
    0.88
    Welcome
    0.86
    증
    0.86
    Activations Density 0.002%

    No Known Activations