© 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. 65609
    Prev
    Next
    INDEX
    Explanations

    actor, actress, son, England

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

    This neuron identifies biography‐style metadata terms and headings (e.g. “born,” dates in parentheses, occupations like “actress,” and section titles such as “Personal life” or “Early life”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     this
    -1.68
     should
    -1.06
     will
    -1.00
     strategy
    -0.97
     our
    -0.96
     new
    -0.94
     ]);
    -0.94
     parseFrom
    -0.93
     desta
    -0.93
     data
    -0.91
    POSITIVE LOGITS
    ごちそう
    1.23
    mainly
    1.20
    颊
    1.16
    having
    1.15
     myr
    1.14
    apprendre
    1.12
    ppled
    1.09
    officially
    1.07
    pled
    1.06
    however
    1.06
    Activations Density 0.113%

    No Known Activations