© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2871
    Prev
    Next
    INDEX
    Explanations

    crucial importance

    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_16k_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
    이지만
    0.45
     यूँ
    0.44
    ｛
    0.44
    ンの
    0.44
     subdued
    0.43
    oflu
    0.43
     fearing
    0.43
    であるが
    0.42
    🫰
    0.42
    imagens
    0.41
    POSITIVE LOGITS
     Seriously
    0.57
     огромное
    0.56
     absolutamente
    0.54
     Absolutely
    0.54
     absolutely
    0.54
     Crazy
    0.53
    لي
    0.51
    9
    0.51
    se
    0.51
     PLEASE
    0.50
    Activations Density 0.331%

    No Known Activations