© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 35794
    Prev
    Next
    INDEX
    Explanations

    Manhattan

    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
    yan
    -0.11
    CAPE
    -0.10
    å¥
    -0.10
    ä¸İåıijå±ķ
    -0.09
    ìŀ¥
    -0.09
    lessly
    -0.09
    家乡
    -0.09
    æ¶ŀ
    -0.09
    æIJŀ好
    -0.09
    åĴĮåıijå±ķ
    -0.09
    POSITIVE LOGITS
    tron
    0.10
    åIJĦåĮº
    0.10
    ization
    0.10
    redo
    0.09
    è¯Ŀ说
    0.09
    æĺİçıł
    0.09
    IZER
    0.09
    æŁĦ
    0.09
    cern
    0.09
    orte
    0.09
    Activations Density 0.003%

    No Known Activations