© 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. 34-GEMMASCOPE-RES-131K
    7. 64194
    Prev
    Next
    INDEX
    Explanations

    or agreed

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

    The neuron flags the repeated boilerplate clause “Unless required by applicable law or agreed to in writing,” specifically firing on the words “law” and “or” in that license text.

    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
    kelen
    -0.82
    Besitos
    -0.79
    XHR
    -0.77
    RENDER
    -0.73
     Danilo
    -0.72
    enses
    -0.70
     börjar
    -0.68
     Persia
    -0.66
     CHAR
    -0.66
     embar
    -0.66
    POSITIVE LOGITS
     integral
    0.66
    僚
    0.66
     explained
    0.64
    COMMENT
    0.64
     competir
    0.63
    ノリ
    0.63
    logits
    0.62
     социальной
    0.62
     comment
    0.59
    溯
    0.59
    Activations Density 0.080%

    No Known Activations