© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 1662
    Prev
    Next
    INDEX
    Explanations

    not followed by qualifier

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

    This neuron fires on negation constructions (e.g. “is not,” “do not,” “does not,” “not available,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    т
    0.95
    誑
    0.68
    丝毫
    0.67
     siquiera
    0.66
    н
    0.64
    檛
    0.64
    з
    0.64
    詆
    0.63
    ख
    0.63
     hề
    0.63
    POSITIVE LOGITS
     reminiscent
    0.63
    aday
    0.62
     designed
    0.61
     denominada
    0.57
    ००
    0.57
     delimit
    0.56
    েরও
    0.55
     bewildered
    0.54
    ments
    0.54
     versus
    0.54
    Activations Density 0.000%

    No Known Activations