© 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. 2-GEMMASCOPE-2-TRANSCODER-262K
    7. 107617
    Prev
    Next
    INDEX
    Explanations

    degauss | degas | asses | as | assing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_2_width_262k_l0_small_affine
    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
     doit
    1.30
     എന്നാൽ
    1.16
     on
    1.12
     hvilket
    1.12
    و
    1.12
     specialize
    1.10
     принима
    1.09
     mesmer
    1.09
     traitement
    1.08
     clé
    1.08
    POSITIVE LOGITS
    ل
    1.50
     It
    1.30
    ون
    1.29
    ä
    1.21
    ina
    1.20
     I
    1.19
    et
    1.17
    ha
    1.15
     Imped
    1.09
    ל
    1.08
    Activations Density 0.000%

    No Known Activations