© 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. 10-GEMMASCOPE-RES-131K
    4. 111003
    Prev
    Next
    INDEX
    Explanations

    doing an action

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

    The neuron detects mentions of a facial expression—phrases describing someone’s “look” of an emotion (e.g. “a confused look of mine”).

    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
     to
    -1.68
    both
    -1.39
     perchè
    -1.37
    demic
    -1.37
    though
    -1.34
    3
    -1.33
     leſs
    -1.32
     isShow
    -1.29
    8
    -1.28
     azzurro
    -1.27
    POSITIVE LOGITS
     pediu
    1.23
    我們的
    1.23
    präch
    1.19
    过来了
    1.19
    してない
    1.19
    Ár
    1.18
     perdeu
    1.14
     iria
    1.14
    られています
    1.13
    できない
    1.13
    Activations Density 0.003%

    No Known Activations