© 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. 36074
    Prev
    Next
    INDEX
    Explanations

    beating, battered, ransacked, excoriates, trashing

    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
    åľ¨æĪij
    -0.11
    鸳
    -0.11
    isy
    -0.10
    flix
    -0.10
    wis
    -0.09
    æĸĩæĺĮ
    -0.09
    fram
    -0.09
    overs
    -0.08
    opers
    -0.08
    æ¨Ĭ
    -0.08
    POSITIVE LOGITS
    ä¸Ģçķª
    0.10
    åĵº
    0.09
    èĻIJ
    0.09
    çIJĨå·¥
    0.09
    笺
    0.09
    è¹Ĥ
    0.09
    åij¨åĽ´çļĦ
    0.08
    ven
    0.08
    æij§æ¯ģ
    0.08
    æĬ¨
    0.08
    Activations Density 0.037%

    No Known Activations