© 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. 72475
    Prev
    Next
    INDEX
    Explanations

    not just / not only / not even

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

    The neuron detects occurrences of the negation “not” (in any casing).

    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
     as
    -1.86
     usually
    -1.84
    礎
    -1.81
     ALWAYS
    -1.80
     and
    -1.78
     new
    -1.78
     ultra
    -1.73
     two
    -1.73
     more
    -1.64
     enticing
    -1.60
    POSITIVE LOGITS
     ส์
    1.97
     consid
    1.84
    鮰
    1.83
    訁
    1.80
     flexibel
    1.69
    genodigd
    1.66
     extravag
    1.66
     florales
    1.63
    遇到了
    1.62
     specif
    1.61
    Activations Density 0.025%

    No Known Activations