© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 32964
    Prev
    Next
    INDEX
    Explanations

    not or without

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

    The neuron strongly activates on negation words (especially “not” and related negating forms).

    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
     حتی
    -1.03
     навіть
    -1.02
     даже
    -1.01
     nawet
    -0.98
     sodio
    -0.93
    भग
    -0.90
    创建一个
    -0.88
     fré
    -0.86
     schützen
    -0.85
     bahkan
    -0.85
    POSITIVE LOGITS
    否定
    1.35
    isDirectory
    0.91
    ñadir
    0.90
    insley
    0.90
     doel
    0.90
     abolition
    0.89
    fruta
    0.88
    нули
    0.87
     loss
    0.87
    шому
    0.86
    Activations Density 0.027%

    No Known Activations