© 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. 10-GEMMASCOPE-RES-131K
    7. 129402
    Prev
    Next
    INDEX
    Explanations

    compliance versus violation

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

    The neuron fires on legal‐style compliance or violation language, spotting words like “compliance” and “violation” in regulatory or legal contexts.

    oai_token-act-pair · o4-miniTriggered 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
     in
    -2.33
     by
    -2.16
     through
    -2.00
     will
    -1.95
     give
    -1.87
     distribute
    -1.86
     deliver
    -1.81
     get
    -1.78
     at
    -1.77
     …
    -1.77
    POSITIVE LOGITS
     花纹
    2.03
    袛
    1.93
     niedersachsen
    1.91
     ſou
    1.84
    鶫
    1.84
    ‪.
    1.77
    ...»
    1.74
     手绘
    1.73
    begle
    1.73
     パッド
    1.73
    Activations Density 0.026%

    No Known Activations