© 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. 3053
    Prev
    Next
    INDEX
    Explanations

    technical errors and data analysis

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

    The neuron consistently lights up on numeric tokens (digits and numbers) in the text.

    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
    centralwidget
    -0.74
    しても
    -0.72
    씩
    -0.71
     Catalog
    -0.70
    celana
    -0.69
     роки
    -0.69
     અ
    -0.69
     noqa
    -0.69
     Passat
    -0.68
     souten
    -0.67
    POSITIVE LOGITS
    Levels
    1.02
    nearly
    0.93
    levels
    0.87
     APPROPRI
    0.82
    imar
    0.79
     ört
    0.79
    Calls
    0.79
     Levels
    0.78
    zium
    0.77
    weer
    0.77
    Activations Density 0.027%

    No Known Activations