© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7169
    Prev
    Next
    INDEX
    Explanations

    indicated below or pictured here

    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
     Kapoor
    -0.09
    _PRIV
    -0.09
    è½
    -0.09
    æĶ¸
    -0.09
    秣
    -0.09
    æĭĺ
    -0.09
    Äĥm
    -0.08
    (decoded
    -0.08
    æ¯ķç«Łæĺ¯
    -0.08
     ÑĥÑģи
    -0.08
    POSITIVE LOGITS
    æħ¨
    0.09
    ä¸Ģèµ·åİ»
    0.09
    æĢ»ä¹ĭ
    0.09
    æ¦Ĥ
    0.09
    oro
    0.09
    mos
    0.08
    姬
    0.08
     NotImplemented
    0.08
    FUN
    0.08
    rtle
    0.08
    Activations Density 0.042%

    No Known Activations