© 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. 34-GEMMASCOPE-RES-131K
    7. 60835
    Prev
    Next
    INDEX
    Explanations

    driving drunk

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

    Explanation of neuron 4 behavior: the main thing this neuron does is find mentions of driving under the influence (DUI/DWI) incidents.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    
    -0.88
     Delivered
    -0.78
    伤害
    -0.77
     existi
    -0.73
    无语
    -0.72
    Vivo
    -0.70
     Imagen
    -0.68
     vect
    -0.68
     Infe
    -0.67
    hara
    -0.67
    POSITIVE LOGITS
     drunk
    2.22
     intoxicated
    1.81
     influence
    1.77
     intoxication
    1.76
     alcohol
    1.64
     influenced
    1.56
     influences
    1.55
     drugs
    1.54
    drunk
    1.53
     Drunk
    1.44
    Activations Density 0.025%

    No Known Activations