© 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. 62285
    Prev
    Next
    INDEX
    Explanations

    provoke, provoked, provocative

    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
    cape
    -0.09
    iring
    -0.09
    ï¸
    -0.09
    åĪĨ辨
    -0.09
    owie
    -0.09
    robe
    -0.09
    rics
    -0.09
    lap
    -0.09
    itted
    -0.08
    ì¦Ī
    -0.08
    POSITIVE LOGITS
    ively
    0.12
    åįģ足
    0.11
    æĢ§å¼º
    0.10
    ä¸įäºĨ
    0.10
    èĵ¼
    0.10
    åľ¨çº¿åĴ¨è¯¢
    0.10
    é£Łæ¬²
    0.10
    ../../../
    0.10
    /ext
    0.09
    ä¹ĭæĥħ
    0.09
    Activations Density 0.005%

    No Known Activations