© 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. 9754
    Prev
    Next
    INDEX
    Explanations

    almost always or all

    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
    æ³¼
    -0.10
    å¾Īé«ĺåħ´
    -0.09
    å¾Ī好
    -0.09
    妾
    -0.09
    足以
    -0.08
    ific
    -0.08
    ervo
    -0.08
    ãĥĸãĥ©
    -0.08
    è¿ĩå¾Ĺ
    -0.08
    ROL
    -0.08
    POSITIVE LOGITS
    åł¤
    0.10
    ãĥ¡ãĥ³
    0.09
    delegate
    0.08
    ìłģìľ¼ë¡ľ
    0.08
    ива
    0.08
    æŃ¤åŁºç¡Ģä¸Ĭ
    0.08
    æľ
    0.08
    otty
    0.08
    é³Į
    0.08
    é«ĺæĸ°æĬĢæľ¯
    0.08
    Activations Density 0.088%

    No Known Activations