© 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. 30335
    Prev
    Next
    INDEX
    Explanations

    `ifndef` or `/style.css`

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

    This neuron detects tokens containing numerals (digits) such as version numbers, numeric constants, or hyphenated numbers.

    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
     Oc
    -0.73
     Hä
    -0.72
     Graph
    -0.72
     Mali
    -0.71
     рис
    -0.69
    Operating
    -0.68
     Octopus
    -0.68
     Mont
    -0.68
     Bicycle
    -0.68
     Fortune
    -0.67
    POSITIVE LOGITS
    adne
    0.76
     Renal
    0.73
    pielt
    0.73
    Renal
    0.71
    ポール
    0.68
    鄧
    0.68
    Schritt
    0.67
     Arroyo
    0.66
    Simone
    0.66
    Ultimo
    0.66
    Activations Density 0.025%

    No Known Activations