© 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. 32075
    Prev
    Next
    INDEX
    Explanations

    as directed by

    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
    iou
    -0.10
    çĽ¸å¯¹æĿ¥è¯´
    -0.09
    å¿ĺè®°
    -0.09
    Ã¤ÃŁ
    -0.09
    æīĢåľ¨çļĦ
    -0.09
    为代表
    -0.09
    为代表çļĦ
    -0.09
    åºł
    -0.09
    代表
    -0.09
    æĸ¹éĿ¢çļĦ
    -0.08
    POSITIVE LOGITS
     manner
    0.11
    æŶéĹ´åĨħ
    0.10
    okie
    0.10
    essian
    0.09
     ease
    0.09
    df
    0.09
    并ä¸İ
    0.09
    dden
    0.09
     Spo
    0.08
    ease
    0.08
    Activations Density 0.061%

    No Known Activations