© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 11612
    Prev
    Next
    INDEX
    Explanations

    legal parties

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

    The neuron detects high numerical tokens that mark legal citations or docket/reference numbers (e.g. the big numbers following party names or footnotes) in court‐opinion text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    보
    0.78
     রাখতে
    0.71
    で
    0.71
    علام
    0.69
    おり
    0.68
    要
    0.67
     any
    0.66
     doit
    0.66
    到来
    0.66
    με
    0.66
    POSITIVE LOGITS
    们的
    0.79
    들의
    0.75
    ๗
    0.71
     wú
    0.68
     villes
    0.66
     herein
    0.66
    ucker
    0.66
    들은
    0.66
     organizaciones
    0.65
     سازمان
    0.65
    Activations Density 0.009%

    No Known Activations