© 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. 22-GEMMASCOPE-RES-131K
    7. 33197
    Prev
    Next
    INDEX
    Explanations

    verbs reporting speech or action

    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
     even
    -0.93
    andinav
    -0.92
     Adverti
    -0.91
     also
    -0.85
     now
    -0.82
     бать
    -0.81
     sekarang
    -0.79
    圏
    -0.79
     chairman
    -0.78
     from
    -0.76
    POSITIVE LOGITS
     iar
    0.91
    kosi
    0.90
     whereupon
    0.89
    に対し
    0.89
    彇
    0.84
    有一些
    0.83
    something
    0.82
    woch
    0.80
     пыта
    0.80
    醤
    0.80
    Activations Density 0.030%

    No Known Activations