© 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. 34-GEMMASCOPE-RES-131K
    7. 62989
    Prev
    Next
    INDEX
    Explanations

    images and photos

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

    This neuron detects UI labels and section headings for images or photos—especially Japanese terms like “写真,” “その他の,” “関連,” and English words like “Images” or “Photos.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    tellungs
    -0.93
     pasaran
    -0.79
     larga
    -0.78
    чих
    -0.78
     legger
    -0.77
    بيرة
    -0.75
     meticulous
    -0.75
     this
    -0.75
    部份
    -0.75
     those
    -0.74
    POSITIVE LOGITS
     Images
    1.29
     images
    1.23
     imágenes
    1.14
     εικό
    1.09
     immagini
    1.09
     image
    1.05
    Images
    1.03
    Autoritní
    0.97
    images
    0.94
    image
    0.94
    Activations Density 0.002%

    No Known Activations