© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 16531
    Prev
    Next
    INDEX
    Explanations

    legal defense, appeals, and charges

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

    The neuron detects mentions of legal defense—terms like “defense,” “defended,” “entrapment,” “alibi,” and related lawyer/attorney language.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     zodia
    -1.04
     maille
    -0.97
    ↑↑
    -0.97
     bross
    -0.96
     tibur
    -0.96
    <<<<
    -0.95
     sodio
    -0.94
    ဋ
    -0.94
     palab
    -0.93
     VIOL
    -0.92
    POSITIVE LOGITS
     defense
    1.36
     against
    1.16
     defenses
    1.14
    はこちらから
    1.13
     challenge
    1.05
    argmin
    1.02
     before
    0.98
     argument
    0.95
     defence
    0.94
     défense
    0.94
    Activations Density 0.055%

    No Known Activations