© 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. 24944
    Prev
    Next
    INDEX
    Explanations

    negation and doubt

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

    The neuron strongly activates on verbs and auxiliaries that express negation or modality/stance (e.g. not, cannot, will, need, believe).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     célè
    -1.11
    いろんな
    -1.10
    細かい
    -1.07
     alcuni
    -1.07
     capuz
    -1.05
     хотят
    -1.03
    וכר
    -1.03
    (^_^
    -1.03
     también
    -1.03
     biraz
    -1.03
    POSITIVE LOGITS
     etc
    1.15
     any
    1.14
     "
    1.11
     quelconque
    1.09
     special
    1.07
    特
    1.05
    千
    1.02
    The
    1.02
     necessarily
    1.01
    新
    1.01
    Activations Density 0.046%

    No Known Activations