© 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. 23725
    Prev
    Next
    INDEX
    Explanations

    plain or downright

    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
    åı¯ä¸įæĺ¯
    -0.11
    æģIJ
    -0.09
    alyzed
    -0.09
    èĮĢ
    -0.09
    第ä¸Ģ
    -0.08
    æĹ¢æĺ¯
    -0.08
    ãģªãģĭãģªãģĭ
    -0.08
    ä¸Ģ次次
    -0.08
    å²Ī
    -0.08
    禧
    -0.08
    POSITIVE LOGITS
     outright
    0.11
     downright
    0.10
    elt
    0.10
    好çļĦ
    0.09
    æŶéĹ´åĨħ
    0.09
    CustomAttributes
    0.09
    ascus
    0.09
    å¾Ģå¤ĸ
    0.09
    ä½ľèĢħæľ¬äºº
    0.08
    åıijå±ķ壮大
    0.08
    Activations Density 0.004%

    No Known Activations