© 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. 84588
    Prev
    Next
    INDEX
    Explanations

    definitely not

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

    The neuron specifically activates on occurrences of the negation “not,” especially in emphatic negative phrases (e.g. “definitely not,” “certainly 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
    锨
    -2.34
     แต่
    -2.27
    咙
    -2.23
    唏
    -2.11
     cependant
    -2.03
    eded
    -2.00
    милия
    -2.00
     Однако
    -1.95
    ᛬
    -1.92
     toutefois
    -1.90
    POSITIVE LOGITS
     under
    1.98
    3
    1.91
     for
    1.83
    5
    1.80
     during
    1.80
    With
    1.76
     the
    1.74
     in
    1.74
     like
    1.72
     while
    1.69
    Activations Density 0.009%

    No Known Activations