© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 73897
    Prev
    Next
    INDEX
    Explanations

    punctuation followed by capitalization

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

    The neuron is highly responsive to non‐word tokens—especially commas and numeric tokens—i.e. punctuation and 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_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
     nya
    -1.05
     ANAL
    -1.00
     oktober
    -0.97
     Objek
    -0.95
     jura
    -0.94
     adem
    -0.92
     empat
    -0.91
    ками
    -0.90
     deng
    -0.90
     svin
    -0.90
    POSITIVE LOGITS
     acelerar
    1.05
     maravilh
    1.02
     cápsulas
    1.02
     vorigen
    1.02
     asumir
    1.00
     sustituir
    0.95
     invitaciones
    0.94
     declaró
    0.93
     berço
    0.93
     rebajas
    0.93
    Activations Density 0.019%

    No Known Activations