© 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. 9571
    Prev
    Next
    INDEX
    Explanations

    exclamations of disgust or pain

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

    The neuron flags exclamatory interjections expressing disgust or annoyance (e.g., “ugh,” “urrrgh”).

    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
    ኧ
    -2.28
    豨
    -2.25
     gleichen
    -2.20
     umfass
    -2.11
     obtiene
    -2.09
     benö
    -2.09
    Sì
    -2.06
     Ersten
    -2.06
    ➬
    -2.02
    
    -2.02
    POSITIVE LOGITS
    s
    2.84
    or
    2.45
    .
    2.28
    蟊
    2.23
    та
    2.23
    埽
    2.22
    𝐒
    2.05
    к
    2.03
    犽
    2.02
     transfé
    1.91
    Activations Density 0.003%

    No Known Activations