© 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. 107179
    Prev
    Next
    INDEX
    Explanations

    COPYRIGHT or originated

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

    This neuron is tuned to spot numeric tokens or sequences of digits (e.g. years, IDs, counts).

    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
    CW
    -0.84
    pixel
    -0.79
    dl
    -0.74
     Boundary
    -0.74
    Pixel
    -0.73
    Wiz
    -0.71
    附加
    -0.70
    vac
    -0.70
    Broadcast
    -0.70
    Odd
    -0.69
    POSITIVE LOGITS
    %@",
    0.77
     gärna
    0.75
     devaient
    0.74
     bénévoles
    0.71
     Zeta
    0.67
    :「
    0.67
    collage
    0.67
    presidente
    0.66
     Erik
    0.65
     PILOT
    0.65
    Activations Density 0.048%

    No Known Activations