© 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
    5. 31-GEMMASCOPE-2-RES-262K
    7. 7292
    Prev
    Next
    INDEX
    Explanations

    cut and fold

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

    fold, cut, tear, roll

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_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
     injustice
    0.44
     üzerinden
    0.44
     racism
    0.44
     hust
    0.43
     racially
    0.43
    嗌
    0.42
     campuran
    0.41
     legit
    0.41
     militant
    0.41
     freaking
    0.40
    POSITIVE LOGITS
    Fold
    0.68
    FOLD
    0.64
    折り
    0.63
    Folding
    0.63
     Fold
    0.62
    折
    0.62
     folds
    0.61
    folds
    0.59
     foldable
    0.58
     折
    0.57
    Activations Density 0.015%

    No Known Activations