© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21297
    Prev
    Next
    INDEX
    Explanations

    provided by law

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

    The neuron detects the boilerplate legal phrase “provided by” (often in statutes or rules).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    formal legal language indicating provisions or rule-stating clauses in judicial or statutory texts.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    the word "provided" followed by "by" or "that" in legal or formal documents.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     jener
    -1.71
    ing
    -1.70
    }
    -1.70
    幛
    -1.65
    凩
    -1.63
    ୌ
    -1.55
    VERTISEMENT
    -1.54
    weeney
    -1.53
    齜
    -1.52
    жете
    -1.52
    POSITIVE LOGITS
     is
    2.28
     หรือ
    1.79
     или
    1.78
     тис
    1.73
     pertinentes
    1.62
     Another
    1.55
     условия
    1.54
    豋
    1.52
     üblichen
    1.50
    壑
    1.48
    Activations Density 0.028%

    No Known Activations