© 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. 23713
    Prev
    Next
    INDEX
    Explanations

    drunk driving

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

    The neuron activates on words and variants referring to intoxication (e.g., “drunk,” “drunken,” etc.).

    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
    outdir
    -0.82
    före
    -0.78
    ie
    -0.77
     ベン
    -0.77
     płyn
    -0.76
     considerando
    -0.75
     capire
    -0.75
     フィット
    -0.74
     alchemist
    -0.73
    ?.
    -0.73
    POSITIVE LOGITS
     drunk
    1.42
    Drunk
    1.23
     Drunk
    1.16
    drunk
    1.09
     stup
    1.08
    ards
    1.03
    ARDS
    0.95
    сете
    0.88
     disorderly
    0.87
    mados
    0.84
    Activations Density 0.015%

    No Known Activations