© 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. 66324
    Prev
    Next
    INDEX
    Explanations

    invasion, denied citations, drug, Christ

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

    The neuron fires strongly on the curly apostrophe character (’), i.e. it detects the typographic apostrophe.

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

    apostrophes in possessive forms and contractions.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    formal proper nouns and titles—especially organization names, institutional/legal terms, and named entities.

    oai_token-act-pair · gpt-5Triggered 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
    名单
    -0.78
    タイム
    -0.73
    劳
    -0.71
    富士
    -0.71
     Release
    -0.71
    release
    -0.71
     CLICK
    -0.71
    Мария
    -0.69
     release
    -0.69
    Release
    -0.69
    POSITIVE LOGITS
    Armstrong
    0.78
     Armstrong
    0.78
    esar
    0.75
    Mess
    0.68
    Dop
    0.68
     obě
    0.67
    Lucas
    0.67
    dop
    0.65
    НОГО
    0.65
     Dimensiones
    0.64
    Activations Density 0.033%

    No Known Activations