© 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. 17231
    Prev
    Next
    INDEX
    Explanations

    rush hour, morning, shift, time

    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
    头éĥ¨
    -0.10
    å¢Ļä¸Ĭ
    -0.09
    edia
    -0.09
    é»ijçĻ½
    -0.09
    ç´¯
    -0.09
    æľ¨è´¨
    -0.09
    _BLACK
    -0.09
    вал
    -0.09
    å¢ĻéĿ¢
    -0.09
     black
    -0.08
    POSITIVE LOGITS
    ä¸ĭçıŃ
    0.13
    ä¸ĬçıŃ
    0.12
    é«ĺå³°
    0.10
    æĶ¾åѦ
    0.10
    é»Ħæĺı
    0.10
    ç«ĸ
    0.09
    åįĪ
    0.09
    交éĢļæŀ¢çº½
    0.09
    çıŃ车
    0.09
    æĹ¶æ®µ
    0.09
    Activations Density 0.185%

    No Known Activations