© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 59966
    Prev
    Next
    INDEX
    Explanations

    agreed to in writing

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

    The neuron selectively fires on the word “writing,” especially in license‐header contexts (“agreed to in writing”).

    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
     كام
    -0.75
    WebService
    -0.73
     Wie
    -0.70
     benutzen
    -0.69
    ól
    -0.69
     Tweets
    -0.68
     мона
    -0.67
    tey
    -0.67
     choice
    -0.66
    Wie
    -0.65
    POSITIVE LOGITS
    IIII
    0.85
    uche
    0.82
     horno
    0.75
    Search
    0.71
    Soft
    0.70
     מרו
    0.69
     Soft
    0.68
     Connectez
    0.67
    ícia
    0.64
    ਿਆ
    0.64
    Activations Density 0.061%

    No Known Activations