© 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. 22-GEMMASCOPE-RES-131K
    7. 64922
    Prev
    Next
    INDEX
    Explanations

    understanding terms and conditions

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

    The neuron detects formal legal or terms-and-conditions and user-consent phrasing (e.g. license agreement and “you agree that…” statements).

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     messen
    -0.91
     what
    -0.90
     usually
    -0.87
     alér
    -0.84
    restre
    -0.83
     reconsider
    -0.82
     people
    -0.82
    jualan
    -0.81
    ◛
    -0.81
     obviously
    -0.78
    POSITIVE LOGITS
    RIGHT
    0.95
    know
    0.95
     وكل
    0.93
    ight
    0.92
    FX
    0.91
    ono
    0.91
     FX
    0.91
    っていて
    0.90
     oike
    0.90
     октя
    0.88
    Activations Density 0.014%

    No Known Activations