© 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. 10-GEMMASCOPE-RES-131K
    7. 113327
    Prev
    Next
    INDEX
    Explanations

    voice assistant, biometric ID, virtualization

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

    technical product names and abbreviations related to technology and computing.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron strongly activates on proprietary technology or product feature names (e.g. Bluetooth, WiDi, True Tone, AirDrop, Hyper-V, Touch ID, Intel Centrino, etc.).

    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
     there
    -1.84
     these
    -1.57
     Aufwand
    -1.49
    就會
    -1.48
     giudi
    -1.47
     that
    -1.38
    襤
    -1.37
    銥
    -1.37
     /\.(
    -1.36
    蕯
    -1.36
    POSITIVE LOGITS
    ="
    1.39
     somente
    1.36
     REALLY
    1.35
     molto
    1.30
     chè
    1.27
    ση
    1.27
     très
    1.27
    蒾
    1.27
    IGraphics
    1.27
     tà
    1.25
    Activations Density 0.116%

    No Known Activations