© 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. 29882
    Prev
    Next
    INDEX
    Explanations

    We followed by verb

    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
    ä¸Ģåī¯
    -0.09
    way
    -0.09
    ä¸įæĸŃåıijå±ķ
    -0.09
    pired
    -0.08
    ç¾İèªī
    -0.08
    ients
    -0.08
    èĻ
    -0.08
    è¯ķçĤ¹å·¥ä½ľ
    -0.08
    æݾ
    -0.08
    æĦ§
    -0.08
    POSITIVE LOGITS
    orem
    0.12
    oretical
    0.10
    imb
    0.09
    éªĨ
    0.09
    iming
    0.09
     hereby
    0.09
    åĶ®
    0.09
     Lv
    0.09
    íıIJ
    0.09
    PM
    0.08
    Activations Density 0.093%

    No Known Activations