© 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. 11109
    Prev
    Next
    INDEX
    Explanations

    but / however followed by contrast

    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
     Grip
    -0.09
    æģŃåĸľ
    -0.09
    æĶ¯
    -0.09
    èĥª
    -0.09
    åºĨ幸
    -0.09
    éļıæľº
    -0.09
    åĢĴæĺ¯
    -0.09
    shiv
    -0.08
    è¿ĻçĤ¹
    -0.08
    æľĢç¾İ
    -0.08
    POSITIVE LOGITS
    åīįæıIJæĺ¯
    0.14
    离ä¸įå¼Ģ
    0.13
    æģ¶
    0.10
    åīįæıIJ
    0.10
    ä¿Ŀåį«
    0.09
    çļĦåīįæıIJ
    0.09
    éĻIJå®ļ
    0.09
    éķķ
    0.09
    è·
    0.09
    çļĦåīįæıIJä¸ĭ
    0.09
    Activations Density 0.151%

    No Known Activations