© 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. 100209
    Prev
    Next
    INDEX
    Explanations

    smirk, smug, irk,bumps

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

    The neuron fires on words that describe emotional or physical reactions (especially facial expressions or visceral responses).

    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
    訁
    -2.64
     cherish
    -2.61
    行って
    -2.50
    姈
    -2.47
    ᠌
    -2.31
    The
    -2.27
    もちろん
    -2.27
     molte
    -2.06
    said
    -2.05
     immerse
    -1.98
    POSITIVE LOGITS
    '
    2.38
    毉
    2.27
     dejen
    2.16
    鬭
    2.03
    tutup
    2.03
    臜
    1.99
    𖥦
    1.95
     个
    1.95
    Alp
    1.90
    她说
    1.88
    Activations Density 0.001%

    No Known Activations