© 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. 33085
    Prev
    Next
    INDEX
    Explanations

    start of title or section of text

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

    This neuron fires on formal “field-of-invention” or scope-specifying phrases (e.g. “relates to the field of…,” “more particularly to systems and…,” “method, system, and computer program for…”).

    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
     
    -1.17
    jinal
    -1.13
    您
    -1.09
     Помо
    -0.97
    rabaya
    -0.96
     You
    -0.96
     mukaan
    -0.95
     Conseils
    -0.95
     juillet
    -0.94
     sisters
    -0.94
    POSITIVE LOGITS
    خخ
    1.54
    اااا
    1.21
     trattano
    1.07
    的東西
    1.02
    اا
    1.00
    랴
    1.00
    imetry
    1.00
     superintend
    0.99
     用品
    0.98
     marts
    0.97
    Activations Density 0.000%

    No Known Activations