© 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. 108242
    Prev
    Next
    INDEX
    Explanations

    legal opinions or advice

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

    The neuron detects passages full of formal legal jargon—especially terms naming legal roles (e.g. “chief counsel,” “legal officer”) and procedural or definitional language typical of legal documents.

    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
    AutoSize
    -0.79
    OPHER
    -0.77
    tilde
    -0.76
     диван
    -0.74
    voye
    -0.74
    сия
    -0.73
    目安
    -0.73
    izacin
    -0.72
    生姜
    -0.71
    Fau
    -0.70
    POSITIVE LOGITS
     legal
    6.50
    legal
    5.50
    Legal
    5.06
     Legal
    4.72
    LEGAL
    4.28
     LEGAL
    4.28
     legally
    3.98
    法律
    3.73
     юриди
    3.63
     légal
    3.33
    Activations Density 0.078%

    No Known Activations