© 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. 50335
    Prev
    Next
    INDEX
    Explanations

    computer input devices

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

    This neuron specializes in spotting code‐style identifiers and section headers—uppercase configuration directives, macro names, and numbered labels characteristic of technical or patent‐style documents.

    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
     Chromecast
    -0.97
     軽量
    -0.94
     Smartphones
    -0.94
     MPEG
    -0.94
     geolocation
    -0.93
     GoPro
    -0.93
    ˒
    -0.90
     yds
    -0.89
     gopro
    -0.89
     смарт
    -0.88
    POSITIVE LOGITS
     keyboard
    1.51
     mouse
    1.47
     keyboards
    1.35
     typing
    1.33
     mice
    1.33
     Keyboard
    1.27
     Logitech
    1.27
    Typing
    1.27
     Mice
    1.25
    keyboard
    1.19
    Activations Density 0.031%

    No Known Activations