© 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. 10-GEMMASCOPE-RES-131K
    7. 112470
    Prev
    Next
    INDEX
    Explanations

    all or not

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

    The neuron fires on short, common “function” words or symbols (e.g. negations, articles, tiny particles, even things like “#!”) regardless of language.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    (
    -2.42
    >
    -2.19
    any
    -2.06
    也會
    -2.00
    謌
    -1.92
    懨
    -1.88
     you
    -1.86
    end
    -1.85
    なんでも
    -1.84
    搁
    -1.84
    POSITIVE LOGITS
    激烈
    2.52
    愉快
    2.06
    咳
    2.02
    糖尿病
    1.97
    攫
    1.94
    反抗
    1.91
    逡
    1.91
    消防
    1.91
    迸
    1.86
    潜水
    1.81
    Activations Density 0.002%

    No Known Activations