© 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. 136985
    Prev
    Next
    INDEX
    Explanations

    resume and summary optimization

    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
    ي
    0.70
    i
    0.63
    ن
    0.60
    ла
    0.58
    ра
    0.53
    ме
    0.51
    غیر
    0.51
    หรือ
    0.50
    त
    0.50
    ан
    0.50
    POSITIVE LOGITS
     daqu
    0.50
     مخطط
    0.42
    getKeyCode
    0.42
    ﺴ
    0.42
    <0x0E>
    0.41
     prism
    0.41
    ﻅ
    0.40
     bulund
    0.40
    EEK
    0.39
     padr
    0.39
    Activations Density 0.002%

    No Known Activations