© 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. 62031
    Prev
    Next
    INDEX
    Explanations

    plural nouns as agents or targets

    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
    æĺ¯ä¸Ģä½į
    -0.11
    æĪIJ为ä¸ĢåIJį
    -0.11
    éĤ£åıª
    -0.11
    åĥı个
    -0.11
    æĪIJ为ä¸Ģ个
    -0.10
    åĴĮå®ĮåĸĦ
    -0.10
    è¿Ļåıª
    -0.10
    ä¼ģä¸ļåĴĮ
    -0.10
    æĺ¯ä¸ª
    -0.10
    å®ĥçļĦ
    -0.09
    POSITIVE LOGITS
     themselves
    0.21
     yourselves
    0.12
     Ñģами
    0.12
    åIJĦèĩª
    0.12
     selves
    0.11
     ourselves
    0.10
    åIJĦèĩªçļĦ
    0.10
     oldukları
    0.10
     olduk
    0.10
    ä½Ĩä»ĸ们
    0.09
    Activations Density 0.163%

    No Known Activations