© 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. 55359
    Prev
    Next
    INDEX
    Explanations

    avoid processing politics lost visa write align

    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
     confines
    0.44
    초
    0.43
    ون
    0.41
    引导
    0.39
    צ
    0.39
     sightseeing
    0.39
    for
    0.38
     experimentar
    0.38
     utiliser
    0.38
    ….
    0.38
    POSITIVE LOGITS
     聞い
    0.47
     ओलंप
    0.44
    рона
    0.43
    RENDER
    0.43
     Тру
    0.42
    судар
    0.42
     metabolism
    0.41
     mutil
    0.41
     वोल्ट
    0.40
    ୂ
    0.40
    Activations Density 0.001%

    No Known Activations