© 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. 18493
    Prev
    Next
    INDEX
    Explanations

    step up

    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
    éºĴ
    -0.10
    tant
    -0.09
    ä¸ĭæīĭ
    -0.09
    tain
    -0.09
    æ¡ĮåŃIJä¸Ĭ
    -0.08
     démarch
    -0.08
    tems
    -0.08
    ä¸ĵåĪ©
    -0.08
    çŃĶåºĶ
    -0.08
    ç´¯äºĨ
    -0.08
    POSITIVE LOGITS
    ä¹ĭ称
    0.09
    rub
    0.09
    ç§ĭåĨ¬
    0.09
    è¾Ī
    0.09
    atif
    0.09
    YPE
    0.09
    人们对
    0.08
    åĵī
    0.08
    æľī幸
    0.08
    æĪĴ
    0.08
    Activations Density 0.002%

    No Known Activations