© 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. 59402
    Prev
    Next
    INDEX
    Explanations

    breathe / wreaks havoc

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

    The neuron is activated by forms of the verb “breathe,” such as “breathe,” “breathed,” 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
     neuen
    -2.72
    鸲
    -2.70
    
    -2.63
     für
    -2.59
    嫫
    -2.58
     Somewhat
    -2.52
     জন্য
    -2.50
     untuk
    -2.48
    Ꭳ
    -2.47
     Currently
    -2.45
    POSITIVE LOGITS
    of
    2.75
    $
    2.67
     unſ
    2.59
    硨
    2.47
    te
    2.47
    while
    2.42
    蹰
    2.39
     {
    2.39
    甗
    2.30
    *
    2.28
    Activations Density 0.001%

    No Known Activations