© 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. 66703
    Prev
    Next
    INDEX
    Explanations

    face and associated phrases

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

    The neuron activates on occurrences of the substring “face” (e.g., in “face,” “face-saving,” 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
    .”
    -2.45
    gencias
    -1.66
    ”
    -1.65
    ,’’
    -1.64
    ulico
    -1.61
    なのかな
    -1.59
    ifts
    -1.59
    ódio
    -1.57
     builds
    -1.57
     Feuchtigkeit
    -1.55
    POSITIVE LOGITS
     --->
    1.72
    灕
    1.65
     ........
    1.61
    ations
    1.49
     ----
    1.45
    乐观
    1.45
    Sabes
    1.45
    羮
    1.45
    ........
    1.44
     -"
    1.43
    Activations Density 0.002%

    No Known Activations