© 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. 52657
    Prev
    Next
    INDEX
    Explanations

    second derivative of

    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
    çļĦåıijçĶŁ
    -0.10
    èĢģçīĮ
    -0.10
    ä¸įèĩ³äºİ
    -0.09
    ä¾¿å®ľ
    -0.09
    æĭ¼
    -0.09
    è¿ĩæĿ¥
    -0.09
    æĸij
    -0.09
    pone
    -0.08
    å°ıå¾®
    -0.08
    贯穿
    -0.08
    POSITIVE LOGITS
    NBC
    0.10
    ANNER
    0.10
    anity
    0.09
    iss
    0.09
    缴å±ŀ
    0.09
    jal
    0.09
    รà¸Ńà¸ļ
    0.08
    èĭ¥æľī
    0.08
    åIJij社ä¼ļ
    0.08
     Bắc
    0.08
    Activations Density 0.264%

    No Known Activations