© 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. 22-GEMMASCOPE-RES-131K
    7. 128406
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively fires on standalone numeric tokens (individual digits).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    publication details, metadata labels

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    Distribuzione
    -1.29
    Beskrivning
    -1.17
    Étymologie
    -1.16
    KURZ
    -1.13
    gili
    -1.12
    maar
    -1.11
    насе
    -1.09
    前は
    -1.07
     教程
    -1.07
    Răsp
    -1.07
    POSITIVE LOGITS
    .}
    1.17
     who
    1.09
    專輯
    1.03
     　
    1.02
    o
    1.00
    ↵
    1.00
    .}\
    0.98
    Y
    0.97
    May
    0.96
     ссылка
    0.96
    Activations Density 0.033%

    No Known Activations