© 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. 63675
    Prev
    Next
    INDEX
    Explanations

    Bennett's fragment

    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
    第ä¸Ģéĥ¨
    -0.10
    çľ¼çĿĽ
    -0.10
    æĩĤäºĭ
    -0.09
    ä¸Ģçīĩ
    -0.09
    éĴŀ
    -0.09
    zd
    -0.09
    ç§ī
    -0.09
    OSH
    -0.08
    ä¸Ģéĥ¨
    -0.08
    大äºĭ
    -0.08
    POSITIVE LOGITS
    å¥Ĺè·¯
    0.11
    ç»ıåħ¸
    0.11
    ç»ıåħ¸çļĦ
    0.11
    模æĿ¿
    0.10
    éļ¾åº¦
    0.10
    ç®Ģåįķ
    0.10
    æ¯Ķè¾ĥç®Ģåįķ
    0.08
    éħįæĸ¹
    0.08
    ç®Ĺæ³ķ
    0.08
    æ¼Ķåıĺ
    0.08
    Activations Density 0.201%

    No Known Activations