© 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. 41-GEMMASCOPE-2-RES-262K
    7. 57863
    Prev
    Next
    INDEX
    Explanations

    tempered glass

    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_all/layer_41_width_262k_l0_big
    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
    дами
    0.42
     Dant
    0.41
     negated
    0.40
     Bloomington
    0.39
    㭂
    0.39
    Grain
    0.39
     Grain
    0.38
     forts
    0.37
    actone
    0.37
    itamins
    0.37
    POSITIVE LOGITS
    ϳ
    0.43
    older
    0.38
     benefici
    0.38
    benef
    0.38
     Benef
    0.38
     panic
    0.38
     W
    0.37
    Bola
    0.37
    WINDOW
    0.37
    W
    0.37
    Activations Density 0.000%

    No Known Activations