© 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. 2770
    Prev
    Next
    INDEX
    Explanations

    as followed by comparison or state

    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¸įæĺ¯
    -0.11
    磫
    -0.09
    ä¸Ģå¦Ĥ
    -0.09
    ëĭ¹
    -0.08
    çĿ¦
    -0.08
    ä¸Ģéģĵ
    -0.08
    ä½ľä¸ºä¸ĢåIJį
    -0.08
    ä¸Ńä»ĭ
    -0.08
    ault
    -0.08
    åħ³æ³¨
    -0.08
    POSITIVE LOGITS
    帮ä»ĸ
    0.10
    笼
    0.10
    è¿ĻéĩĮçļĦ
    0.10
    .createTextNode
    0.09
    èī¯
    0.09
    å°ıå°ıçļĦ
    0.09
    æİ¥éĢģ
    0.09
    æĭīçĿĢ
    0.09
    mı
    0.09
    å²³
    0.08
    Activations Density 0.061%

    No Known Activations