© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50282
    Prev
    Next
    INDEX
    Explanations

    is this a question about

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æij¸ç´¢
    -0.10
    yon
    -0.08
     Seat
    -0.08
    åĨµ
    -0.08
    å±İ
    -0.08
    çĽ¾
    -0.08
    å¾Īå°ı
    -0.08
    楯
    -0.08
     engr
    -0.08
    æĺİæĺİ
    -0.08
    POSITIVE LOGITS
    åĮį
    0.10
    æĢ¿
    0.09
    åĴĮå®ĮåĸĦ
    0.09
    è¿ĺç®Ĺ
    0.08
    áo
    0.08
    å§ĭäºİ
    0.08
    妥
    0.08
    HeaderCode
    0.08
    æľĿ
    0.08
    جة
    0.08
    Activations Density 0.261%

    No Known Activations