© 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. 41-GEMMASCOPE-2-RES-262K
    7. 25856
    Prev
    Next
    INDEX
    Explanations

    intended meaning or purpose

    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_all/layer_41_width_262k_l0_big
    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
     rarer
    0.41
    ्रेस
    0.39
     devoting
    0.38
    ẹn
    0.38
    ņ
    0.37
    잖아
    0.37
    取材
    0.37
    䡋
    0.37
    ))){
    0.37
    ással
    0.37
    POSITIVE LOGITS
     intended
    0.85
     meant
    0.79
    intended
    0.69
     bedo
    0.68
     intend
    0.59
    ToBe
    0.57
     предназна
    0.52
     tarko
    0.52
     dimaksud
    0.50
     предназначен
    0.48
    Activations Density 0.000%

    No Known Activations