© 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. 3824
    Prev
    Next
    INDEX
    Explanations

    unsettling, disturbing, frightening, confusing

    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
    çľ¼æ³ª
    -0.10
    pure
    -0.10
     издели
    -0.09
     ÑĢеÑĪа
    -0.09
    åĬ¨çĶŁæĪIJ
    -0.09
    éĢĤ
    -0.08
    æĪijå¸ĮæľĽ
    -0.08
    anson
    -0.08
    ched
    -0.08
    heet
    -0.08
    POSITIVE LOGITS
    ly
    0.20
    LY
    0.12
    ä¸Ģå¹ķ
    0.11
    èİ«è¿ĩäºİ
    0.10
    ãģĮãĤī
    0.09
    ãģªãĤĤãģ®
    0.09
    çģ«çĥ§
    0.09
    ãģª
    0.09
    oose
    0.09
    å¾Ĺå¤ļ
    0.09
    Activations Density 0.078%

    No Known Activations