© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 9641
    Prev
    Next
    INDEX
    Explanations

    refused to participate or do

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

    The neuron fires on infinitive constructions expressing refusal (e.g. “refused to …”).

    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
     dwindling
    -2.05
     crafty
    -1.94
    先日
    -1.85
    ar
    -1.85
     –
    -1.80
    遨
    -1.80
     lengthy
    -1.80
     enormous
    -1.79
    そして
    -1.76
     dozens
    -1.72
    POSITIVE LOGITS
    🫃
    2.05
     horloge
    1.91
    腘
    1.88
    脔
    1.81
     anormal
    1.74
     Signalez
    1.69
    乀
    1.68
    );
    1.68
     micrófono
    1.65
     herv
    1.62
    Activations Density 0.014%

    No Known Activations