© 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. 102867
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     weaken
    0.48
     वक़्त
    0.43
    ኸ
    0.40
     weak
    0.40
     filling
    0.39
    atzeko
    0.38
     weakens
    0.37
     weakened
    0.37
     hydride
    0.37
    withTrashed
    0.37
    POSITIVE LOGITS
    ريط
    0.41
    рім
    0.38
    лізу
    0.38
    ㄥ
    0.38
    بيرة
    0.37
    নিয়া
    0.37
    臧
    0.37
    ونية
    0.37
    谬
    0.37
    োর্ড
    0.36
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.