© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 74034
    Prev
    Next
    INDEX
    Explanations

    face recognition, perception, images

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

    This neuron activates strongly on occurrences of the word “face” (and its close variants) in the text.

    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
    .
    -2.75
     vorigen
    -2.69
     gän
    -2.38
     bleib
    -2.06
     vollständige
    -2.05
    ༀ
    -2.03
    もはや
    -2.02
     Veränder
    -2.02
    考えて
    -2.00
    ATION
    -1.97
    POSITIVE LOGITS
    桄
    2.31
    𖣘
    2.28
     dahil
    2.22
     lakini
    2.16
     nasıl
    2.09
     brasileño
    2.08
    e
    2.08
    遢
    2.03
    اتب
    2.02
     kwenye
    1.99
    Activations Density 0.009%

    No Known Activations