© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8658
    Prev
    Next
    INDEX
    Explanations

    The neuron detects mentions of faces or facial-related terms (e.g. “facial recognition,” “face masks,” “frowning”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     کمر
    0.66
    腰
    0.64
     衣
    0.62
     Carpet
    0.62
     chlorophyll
    0.60
     腰
    0.60
     bod
    0.58
     simb
    0.57
     limb
    0.57
     Flooring
    0.56
    POSITIVE LOGITS
     facial
    2.41
     Facial
    2.34
     얼굴
    2.19
    脸
    2.05
    顔
    2.05
    臉
    2.05
    facial
    2.03
     face
    1.95
    Face
    1.95
     Face
    1.93
    Activations Density 0.470%

    No Known Activations