© 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. 10345
    Prev
    Next
    INDEX
    Explanations

    not ... significant, use, die, affectedwithout ... die, co, fancyshow ... significant

    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
    æĸ°å¨ĺ
    -0.10
    å¾ĭå¸ĪäºĭåĬ¡
    -0.10
    åĥıæĺ¯
    -0.10
    grunt
    -0.09
    ä¸įåĥı
    -0.09
    .bl
    -0.09
    ä¸įèĩ³äºİ
    -0.09
    éĵĬ
    -0.09
    uddenly
    -0.08
    åıªæĢķ
    -0.08
    POSITIVE LOGITS
     вообÑīе
    0.15
    æł¹æľ¬ä¸į
    0.13
    è¿Ļä¹Īåģļ
    0.11
     altogether
    0.10
    没æľīä»»ä½ķ
    0.10
    ä»»ä½ķ
    0.09
    æł¹æľ¬å°±ä¸į
    0.09
    epad
    0.09
    èĢĮæĺ¯
    0.09
    éªĬ
    0.09
    Activations Density 0.258%

    No Known Activations