© 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. 50720
    Prev
    Next
    INDEX
    Explanations

    I cannot tell

    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
     zwar
    -0.15
    èĻ½çĦ¶
    -0.13
    èĻ½è¯´
    -0.12
    éĽĸçĦ¶
    -0.12
    åıįåĢĴ
    -0.11
    ä¸įä½Ĩ
    -0.11
    åħĪæĺ¯
    -0.11
    èĢĮä¸į
    -0.11
    æ¯Ķå¦Ĥ说
    -0.11
     thus
    -0.11
    POSITIVE LOGITS
    ä¹Łä¸įèĥ½
    0.11
    ä¹Łæľī
    0.11
    ä¾ĿçĦ¶
    0.10
    ละ
    0.10
    åIJĮæĹ¶ä¹Ł
    0.10
    åį»
    0.10
    ä¹Łåı¯ä»¥
    0.10
    ÑģÑı
    0.10
     nonetheless
    0.10
    بÙĤÙī
    0.10
    Activations Density 0.246%

    No Known Activations