© 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. 16654
    Prev
    Next
    INDEX
    Explanations

    forces, fighters, disasters

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

    The neuron strongly activates on adjectives and adverbs that express degree, magnitude, or modality (e.g. “possible,” “easy,” “enough,” “large,” “only”).

    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
     intitulé
    -1.25
     chrétien
    -1.22
     cabine
    -1.20
     chaussure
    -1.17
     gimnas
    -1.16
     monté
    -1.14
     notori
    -1.13
     arkiv
    -1.10
     inova
    -1.09
     pandemi
    -1.09
    POSITIVE LOGITS
     or
    1.13
     and
    1.09
    ↵
    0.95
     samtidig
    0.93
    仍然
    0.93
     respetu
    0.91
    bairro
    0.90
    仍
    0.89
    _{\
    0.87
    mika
    0.87
    Activations Density 0.028%

    No Known Activations