© 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. 48649
    Prev
    Next
    INDEX
    Explanations

    most of the

    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
    AspectRatio
    -0.09
    chw
    -0.08
    -router
    -0.08
    åĮ¡
    -0.08
    æĹĮ
    -0.08
    Spo
    -0.08
    Ø·ÙĦ
    -0.08
    Router
    -0.08
     Cousins
    -0.08
    خب
    -0.08
    POSITIVE LOGITS
    æŁ¬
    0.09
     вÑģего
    0.09
     all
    0.08
    éĨĴäºĨ
    0.08
    ä¸ĭæīĭ
    0.08
    bez
    0.08
    æĹłå½¢
    0.08
    åѦä¼ļäºĨ
    0.08
    ç»
    0.08
    eters
    0.08
    Activations Density 0.026%

    No Known Activations