© 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. 33776
    Prev
    Next
    INDEX
    Explanations

    a or an followed by a word

    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
    Courtesy
    -0.10
    幡
    -0.10
    ãĤ«ãĥ«
    -0.09
     Vand
    -0.09
     nét
    -0.09
    éĿŀ常å¤ļ
    -0.08
    æ£Ĵ
    -0.08
     scrambling
    -0.08
     Courtesy
    -0.08
    otent
    -0.08
    POSITIVE LOGITS
    ä»»ä½ķå½¢å¼ı
    0.13
     eventual
    0.12
    åĪ°æŶåĢĻ
    0.09
    åĪ°æĿ¥
    0.09
    ä¸įåıĬ
    0.09
    ä¸ĢæŦ
    0.09
    å±ĬæŶ
    0.09
    ä¸ĵ人
    0.09
    SEE
    0.09
    解åĨ³æĸ¹æ¡Ī
    0.08
    Activations Density 0.119%

    No Known Activations