© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 740
    Prev
    Next
    INDEX
    Explanations

    contrast and caveats

    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
     或者
    0.32
     perfeito
    0.30
     каком
    0.30
     whatnot
    0.30
    ভূতি
    0.29
     इत्या
    0.29
     ಏನು
    0.29
    𝗴
    0.29
    恋爱
    0.29
     เงี้ย
    0.29
    POSITIVE LOGITS
    However
    0.34
    although
    0.33
     however
    0.33
     However
    0.32
    NOTE
    0.31
    Although
    0.30
     cependant
    0.30
    however
    0.30
     NOTE
    0.30
    ※
    0.29
    Activations Density 0.780%

    No Known Activations