© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 93778
    Prev
    Next
    INDEX
    Explanations

    on and off

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

    This neuron detects occurrences of the words “on” and “off,” especially when they appear together in toggling contexts (e.g., “on and off”).

    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
    روط
    -0.82
     bern
    -0.81
     Rude
    -0.81
    vecs
    -0.79
     Dacă
    -0.75
    toarele
    -0.74
    movi
    -0.74
     Drones
    -0.73
     <--
    -0.73
     Yog
    -0.73
    POSITIVE LOGITS
     off
    2.39
    off
    1.69
     Off
    1.47
    Off
    1.45
     OFF
    1.44
    offs
    1.23
    OFF
    1.19
    オフ
    1.10
     offline
    1.09
     away
    1.05
    Activations Density 0.013%

    No Known Activations