© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 15483
    Prev
    Next
    INDEX
    Explanations

    android package names com

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

    The neuron activates on numeric tokens—version numbers, resource IDs, and other digits (often with decimal points) embedded in XML or code.

    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
    actéristi
    -1.88
     ludicrous
    -1.79
     Dinas
    -1.76
    
    -1.73
     astounding
    -1.72
    Ꮅ
    -1.69
     !!!!
    -1.68
    
    -1.64
    ჱ
    -1.64
     😁
    -1.63
    POSITIVE LOGITS
     which
    2.31
    Even
    2.22
     Despite
    2.17
    Despite
    2.13
     Even
    2.13
     Although
    2.11
     still
    1.98
     Maybe
    1.92
    Although
    1.92
     Though
    1.87
    Activations Density 0.002%

    No Known Activations