© 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. 3388
    Prev
    Next
    INDEX
    Explanations

    no, not, non, without

    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
    è¿ĺä¸įå¤Ł
    -0.10
    ä¸įå°ij人
    -0.10
    è¿ĺæĺ¯
    -0.10
    没èĥ½
    -0.10
    lte
    -0.09
    åIJĮèĥŀ
    -0.09
    ç´ł
    -0.09
    å¾Īå¤ļæľĭåıĭ
    -0.09
    åıĭåĸĦ
    -0.09
    并ä¸įæĺ¯å¾Ī
    -0.08
    POSITIVE LOGITS
    æľīä»»ä½ķ
    0.12
    ä¸Ŀ毫
    0.10
    çĤŃ
    0.10
    让人
    0.09
    ä»»ä½ķå½¢å¼ı
    0.09
    ä»»ä½ķ
    0.09
    ä¸įåĥı
    0.09
    ä¸įåIJĮäºİ
    0.09
    对人ä½ĵ
    0.09
    èĵĸ
    0.09
    Activations Density 0.091%

    No Known Activations