© 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. 68335
    Prev
    Next
    INDEX
    Explanations

    v followed by R, I, 0, 1, or out

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

    This neuron detects numeric tokens and related punctuation (digits, decimal points, minus signs), i.e. number or version-style 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_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
     a
    -2.48
    ).
    -2.27
    ‘
    -2.11
    at
    -2.09
    in
    -2.09
     thee
    -1.90
    3
    -1.85
     \
    -1.85
    r
    -1.84
    B
    -1.80
    POSITIVE LOGITS
    
    2.11
    熥
    2.06
    桫
    2.03
    ᯾
    1.95
     でも
    1.94
    妡
    1.92
    ᆞ
    1.91
    甃
    1.91
    魆
    1.91
    ㋛
    1.88
    Activations Density 0.091%

    No Known Activations