© 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. 2672
    Prev
    Next
    INDEX
    Explanations

    technical or international content

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

    This neuron activates on tokens that include numeric strings (e.g. version numbers, hex-style revisions, CSS numeric values), i.e. it’s looking for digits or digit-heavy identifiers.

    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
    testCase
    -0.73
     contraband
    -0.73
     palate
    -0.71
    Vegeta
    -0.68
    perten
    -0.68
    ＃
    -0.67
     MIB
    -0.67
     recorrer
    -0.67
    igin
    -0.66
    izhou
    -0.65
    POSITIVE LOGITS
    エ
    0.69
    फ
    0.69
     shield
    0.69
    Lab
    0.69
    境的
    0.68
    เอ
    0.67
    Lambda
    0.67
     Unfall
    0.66
     Novak
    0.65
    Cmd
    0.65
    Activations Density 0.055%

    No Known Activations