© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 40096
    Prev
    Next
    INDEX
    Explanations

    copyright and license terms

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

    This neuron fires on text strings that serve as labels, names, or other standalone terms in code or markup (e.g. UI strings inside `<source>` tags, identifiers like “Arduino,” “Bluetooth Devices Paired,” comment‐header words such as “program” or “useful”).

    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
     artesanía
    -1.92
     geene
    -1.89
     impresora
    -1.78
    nữ
    -1.78
     gastronomía
    -1.70
     stück
    -1.69
     Aktual
    -1.67
     италья
    -1.66
     dát
    -1.66
    koala
    -1.64
    POSITIVE LOGITS
     although
    2.13
     Although
    2.02
     Even
    1.95
     Though
    1.94
     while
    1.90
     with
    1.88
     Despite
    1.84
     While
    1.77
     chrétien
    1.77
     less
    1.76
    Activations Density 0.003%

    No Known Activations