© 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. 243711
    Prev
    Next
    INDEX
    Explanations

    words through holes

    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
    \{
    0.52
     knack
    0.45
     condem
    0.45
     deliberate
    0.44
     meta
    0.44
     rejection
    0.43
     lapse
    0.43
     conduit
    0.42
     up
    0.42
     fond
    0.42
    POSITIVE LOGITS
     യു
    0.57
     ଗ
    0.54
     സ
    0.50
     ಯು
    0.50
     війни
    0.50
    iectasis
    0.48
     σει
    0.48
     Ε
    0.47
    iei
    0.47
     ड्राइव
    0.46
    Activations Density 0.000%

    No Known Activations