© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 59439
    Prev
    Next
    INDEX
    Explanations

    murder trials and legal cases

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

    The neuron is a proper-name detector—it lights up on people’s and places’ names (e.g. “New,” “George,” “Zimmerman,” “Sanford”) and similarly formatted dateline/place tokens in legal-news text.

    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
    |_{\
    -0.94
     ire
    -0.82
    Toss
    -0.81
    avras
    -0.77
     neuter
    -0.77
    circa
    -0.76
    unnitel
    -0.75
     スイーツ
    -0.75
     delito
    -0.75
    Siren
    -0.75
    POSITIVE LOGITS
    ancier
    0.99
     convicted
    0.97
    Serial
    0.95
     OJ
    0.95
     kasus
    0.93
     Jian
    0.90
     footballer
    0.88
     murder
    0.88
    SERIAL
    0.88
     former
    0.88
    Activations Density 0.148%

    No Known Activations