© 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. 22-GEMMASCOPE-RES-131K
    7. 15463
    Prev
    Next
    INDEX
    Explanations

    endings or extremes

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

    The neuron specializes in detecting numeric tokens, i.e. occurrences of numbers or number‐related strings.

    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
     until
    -0.93
    女性の
    -0.82
    topp
    -0.82
    billing
    -0.78
     intersecting
    -0.78
    "}}>
    -0.78
    (),"
    -0.77
    ulação
    -0.77
     all
    -0.77
     tüm
    -0.77
    POSITIVE LOGITS
     extremes
    1.34
     extremo
    1.02
    Ends
    1.02
    Extreme
    0.98
     ends
    0.98
     extrema
    0.97
     endpoints
    0.95
     extreme
    0.94
     extremities
    0.94
    extreme
    0.88
    Activations Density 0.103%

    No Known Activations