© 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. 1592
    Prev
    Next
    INDEX
    Explanations

    as Moscow

    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
    æĿĥ
    -0.08
    æ®ī
    -0.08
    è§Ĵ度çľĭ
    -0.08
     afs
    -0.08
    isters
    -0.08
    ustr
    -0.08
    yr
    -0.08
     Magic
    -0.07
    为é¦ĸ
    -0.07
     Vel
    -0.07
    POSITIVE LOGITS
    çļĦåIJĮæŶ
    0.18
    ä¸İæѤåIJĮæŶ
    0.14
    while
    0.12
    åIJĮæĻĤãģ«
    0.12
    åIJĮæŶ
    0.11
     while
    0.11
    ä¼´éļıçĿĢ
    0.10
     ëıĻìĭľìĹIJ
    0.09
    	while
    0.09
    ä¸ĢéĿ¢
    0.09
    Activations Density 0.062%

    No Known Activations