© 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. 101886
    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
     нам
    0.46
     pentru
    0.45
     для
    0.43
     মন্ত
    0.43
    竣
    0.43
    สำหรับ
    0.41
     प्रश्‍न
    0.41
    给我们
    0.40
     өчен
    0.40
     үшін
    0.39
    POSITIVE LOGITS
     Parkinson
    0.40
     highly
    0.39
    ∑</
    0.39
     leverage
    0.38
     zároveň
    0.38
    Surprisingly
    0.38
     glioma
    0.38
     advantages
    0.37
     peacock
    0.37
     cubs
    0.37
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.