© 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. 113265
    Prev
    Next
    INDEX
    Explanations

    not followed by other words

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

    The neuron strongly activates on the negation “not.”

    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
     and
    -2.59
     by
    -2.20
     &
    -1.84
     in
    -1.84
    瑧
    -1.63
     there
    -1.61
    adí
    -1.58
    ↵↵
    -1.58
     også
    -1.56
    there
    -1.53
    POSITIVE LOGITS
    킁
    2.16
     ACTUALLY
    1.94
     zoude
    1.81
    ౩
    1.81
    摀
    1.79
    actéristi
    1.74
     ส์
    1.73
    神情
    1.73
     таких
    1.73
    Actually
    1.73
    Activations Density 0.012%

    No Known Activations