© 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. 34-GEMMASCOPE-RES-131K
    7. 97967
    Prev
    Next
    INDEX
    Explanations

    questions about word usage

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

    The neuron is tuned to punctuation tokens—especially quotes, apostrophes, parentheses, and similar punctuation marks.

    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
    ilités
    -0.97
    filtered
    -0.82
    undet
    -0.81
    چسب
    -0.79
    elaar
    -0.78
    uropa
    -0.76
    nagem
    -0.76
    ازی
    -0.76
    новид
    -0.75
     נע
    -0.75
    POSITIVE LOGITS
    Affect
    1.02
     affect
    1.02
     its
    0.92
    affect
    0.89
    其
    0.87
    hopefully
    0.86
     whom
    0.84
    svr
    0.83
     isShow
    0.81
     their
    0.81
    Activations Density 0.020%

    No Known Activations