© 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. 55352
    Prev
    Next
    INDEX
    Explanations

    catch errors and identify issues

    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
    good
    0.58
    using
    0.54
    r
    0.52
    article
    0.51
    container
    0.44
    须
    0.44
    cooling
    0.43
    announce
    0.43
    注
    0.43
    内
    0.43
    POSITIVE LOGITS
    izam
    0.57
     हु
    0.52
    sse
    0.48
     frequência
    0.48
     xóa
    0.48
     usp
    0.47
     게
    0.47
     mandar
    0.47
     مکان
    0.46
     ف
    0.46
    Activations Density 0.003%

    No Known Activations