© 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. 27626
    Prev
    Next
    INDEX
    Explanations

    site or source

    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¿Ļå¥Ĺ
    -0.10
    ä¹Łæĺ¯å¦ĤæѤ
    -0.09
    ç²²
    -0.08
    æĤ¦
    -0.08
    OTES
    -0.08
     Jeh
    -0.08
    æľīæ¯Ĵ
    -0.08
    æĸĻ
    -0.08
    esty
    -0.08
    rolling
    -0.08
    POSITIVE LOGITS
    ä¸Ģç«Ļå¼ı
    0.14
    æĿĥå¨ģ
    0.11
    çĶ³
    0.10
    åħĪéĶĭ
    0.09
    éŨæĪ·
    0.09
    åľ¨çº¿
    0.09
    çIJIJ
    0.09
    åı¯éĿłçļĦ
    0.09
    大åħ¨
    0.09
    çªŁ
    0.09
    Activations Density 0.109%

    No Known Activations