© 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. 129473
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric tokens (digits and numbers) in the text.

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

    `$$` and math formulas

    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
    ização
    -0.93
    </tbody>
    -0.90
    naire
    -0.89
    SourceChecksum
    -0.86
    ДИ
    -0.83
    ional
    -0.83
    atized
    -0.81
    chaos
    -0.81
     sempre
    -0.81
     milhões
    -0.81
    POSITIVE LOGITS
    Răspunde
    0.93
    してみると
    0.92
    具体的な
    0.90
     Tongue
    0.90
    ｹ
    0.88
    Новые
    0.88
    散策
    0.85
     conocemos
    0.85
    exprimer
    0.84
     джере
    0.84
    Activations Density 0.041%

    No Known Activations