© 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-262K
    7. 82753
    Prev
    Next
    INDEX
    Explanations

    engineer is, May Weather, color based

    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
    按照
    0.48
     contaba
    0.47
     поддержи
    0.45
     decía
    0.45
    ﻀ
    0.45
    individuals
    0.44
    ანი
    0.44
    CA
    0.43
    (
    0.43
    iesel
    0.43
    POSITIVE LOGITS
     yet
    0.45
     can
    0.43
     Recogn
    0.43
    column
    0.42
     Ransom
    0.42
     apasion
    0.40
     *.
    0.39
    रोवर
    0.39
     Provident
    0.38
     Committed
    0.38
    Activations Density 0.000%

    No Known Activations