© 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. 2661
    Prev
    Next
    INDEX
    Explanations

    use of cookies

    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
    æ¶Ī
    -0.10
    /preferences
    -0.09
    èµİ
    -0.09
    æ¯Ĵ
    -0.09
    çįIJ
    -0.08
    .Sm
    -0.08
    积
    -0.08
    _lex
    -0.08
    èĢĮæĺ¯
    -0.08
    è°ģçŁ¥
    -0.08
    POSITIVE LOGITS
    æѤåIJİ
    0.10
    åIJ¦
    0.10
    为åħ¶
    0.09
    对åħ¶è¿Ľè¡Į
    0.09
     Howell
    0.09
    ä¸įåĩº
    0.08
    Lean
    0.08
    æŃ¤é¡¹
    0.08
    ãĤĴ使çĶ¨
    0.08
    ä¿Ł
    0.08
    Activations Density 0.029%

    No Known Activations