© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130496
    Prev
    Next
    INDEX
    Explanations

    The neuron detects tokens involved in parenthetical or enumerative asides—numbers, parentheses, and related punctuation used for in-text annotations.

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

    specific pairings

    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
     angst
    -1.03
     choc
    -0.90
    tiera
    -0.90
    caria
    -0.89
    рода
    -0.88
    dian
    -0.88
     машину
    -0.87
     Sade
    -0.86
    quila
    -0.85
     bard
    -0.85
    POSITIVE LOGITS
     menyediakan
    1.05
    ところに
    1.02
     specialized
    1.01
     children
    0.97
     peculiar
    0.94
     berguna
    0.93
     immediate
    0.93
     under
    0.93
     makeshift
    0.92
     voedsel
    0.91
    Activations Density 0.006%

    No Known Activations