© 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. 35065
    Prev
    Next
    INDEX
    Explanations

    release, out, escaped, leakage

    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
    åŀ¢
    -0.11
    å¾Ģä¸ĭ
    -0.10
    ä¸įåĬ¨
    -0.10
    å¿įåıĹ
    -0.10
    å¿į
    -0.10
    ropy
    -0.10
    æµĵ
    -0.09
    åŁł
    -0.09
    å¾ĢæĿ¥
    -0.09
     ÑģоÑħ
    -0.09
    POSITIVE LOGITS
    åĩºåİ»
    0.10
    ouver
    0.10
    éĢ¸
    0.09
    æł¡å¤ĸ
    0.09
    æ¼ı
    0.09
    æĹ·
    0.09
    ä½ļ
    0.09
    å¥Ķ
    0.09
    nell
    0.09
    [out
    0.09
    Activations Density 0.183%

    No Known Activations