© 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. 31-GEMMASCOPE-2-RES-262K
    7. 21256
    Prev
    Next
    INDEX
    Explanations

    redefining rogue actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    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
     nowego
    0.45
     эф
    0.45
     nieuwe
    0.43
     μας
    0.42
     нового
    0.42
    icoes
    0.42
     emoz
    0.42
    줬
    0.42
    الغ
    0.41
    gerät
    0.41
    POSITIVE LOGITS
     usual
    0.43
     threatening
    0.43
     conclusive
    0.43
     Interpret
    0.42
     domine
    0.42
     proliferation
    0.41
    ↵
    0.40
     Traditionally
    0.40
     redefined
    0.39
    ׂ
    0.39
    Activations Density 0.001%

    No Known Activations