© 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. 118167
    Prev
    Next
    INDEX
    Explanations

    off- compound words

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

    This neuron detects occurrences of “off” (especially as the prefix in hyphenated compounds like off-site, off-label, off-pump, off-topic, 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_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
     də
    -0.90
    RIQUE
    -0.76
    viers
    -0.75
    γέν
    -0.71
    تها
    -0.69
    ➙
    -0.69
     Nein
    -0.69
    шей
    -0.69
    Asu
    -0.68
     Directives
    -0.68
    POSITIVE LOGITS
     campus
    1.48
    campus
    1.03
    TOPIC
    1.02
     topic
    0.94
     camera
    0.92
     Off
    0.91
     off
    0.88
    Off
    0.86
    screen
    0.84
    site
    0.84
    Activations Density 0.037%

    No Known Activations