© 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. 3283
    Prev
    Next
    INDEX
    Explanations

    behind the curtain

    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
    åĴķ
    -0.09
    稳妥
    -0.08
    åĮ»çĸĹæľįåĬ¡
    -0.08
     ÑģкаÑĩ
    -0.08
    迦
    -0.08
    ãĤµãĥ¼
    -0.08
    åľ°çĤ¹
    -0.08
     näm
    -0.08
    æĪij说
    -0.08
    æĬ¥åIJį
    -0.08
    POSITIVE LOGITS
    å¹ķåIJİ
    0.23
     behind
    0.18
    èĥĮåIJİ
    0.16
     backstage
    0.15
    èĥĮåIJİçļĦ
    0.15
     Behind
    0.15
    Behind
    0.14
    æıŃç§ĺ
    0.14
    çļĦèĥĮåIJİ
    0.14
    åIJİåı°
    0.12
    Activations Density 0.179%

    No Known Activations