© 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-3-27B-IT
    5. 40-GEMMASCOPE-2-TRANSCODER-262K
    7. 17346
    Prev
    Next
    INDEX
    Explanations

    peeling, delamination, separation, gaps

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_40_width_262k_l0_small_affine
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    缚
    0.37
    大使
    0.35
     دقت
    0.35
    seekbar
    0.35
    burned
    0.34
    entities
    0.34
     छिपा
    0.33
    镐
    0.33
     burned
    0.33
    Stylish
    0.33
    POSITIVE LOGITS
     peeling
    0.94
     peel
    0.82
     peels
    0.82
     peeled
    0.80
     del
    0.77
     exfol
    0.74
    剥
    0.73
     Peel
    0.71
     bubbling
    0.70
    剝
    0.70
    Activations Density 0.056%

    No Known Activations