© 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. 54751
    Prev
    Next
    INDEX
    Explanations

    for noun/pronoun

    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
    è¿Ļæĺ¯
    -0.08
    تÙĤ
    -0.08
    è¿Ļæĺ¯ä¸Ģç§į
    -0.08
    ä¸Ģèµ·åİ»
    -0.08
     Stub
    -0.08
    DL
    -0.08
    æĿĵ
    -0.08
    å¥ī
    -0.08
    è¿Ļæĺ¯ä¸Ģ个
    -0.08
    _DL
    -0.07
    POSITIVE LOGITS
    ä¸įåı¯ç¼ºå°ij
    0.10
    unya
    0.10
    æĽ´æĺ¯
    0.10
    èĢĮéĿŀ
    0.10
    åIJĦ家
    0.09
    ç«¥
    0.09
    _INS
    0.09
    ingly
    0.09
    éĶĭ
    0.09
     reb
    0.09
    Activations Density 0.199%

    No Known Activations