© 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. 48053
    Prev
    Next
    INDEX
    Explanations

    or

    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
    管çIJĨåĴĮ
    -0.10
     EITHER
    -0.09
    åĴĮå®ĮåĸĦ
    -0.09
    retch
    -0.09
     either
    -0.09
    è¯Ŀ说
    -0.09
    ãģĦãģļãĤĮ
    -0.08
    è¦ģä¹Ī
    -0.08
    驳
    -0.08
     apl
    -0.08
    POSITIVE LOGITS
    SIGN
    0.10
    å¸ĤåľºéľĢæ±Ĥ
    0.09
    äºĶè¡Į
    0.09
    æ¸İ
    0.09
    æĹºåŃ£
    0.09
    ç¥ĸ
    0.09
    群
    0.09
    åĪĨéĴŁåIJİ
    0.09
    å¼Ĭ
    0.09
    æĺ¯æĪij们
    0.09
    Activations Density 0.025%

    No Known Activations