© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 60604
    Prev
    Next
    INDEX
    Explanations

    astronauts and air force officers

    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
    Negative Logits
    é£İæ°´
    -0.11
    常æĢģ
    -0.10
    è¦ģ说
    -0.10
    æĤ¬å¿µ
    -0.10
    bourne
    -0.10
    fuck
    -0.09
    ä¸ĵ人
    -0.09
    åıĺè¿ģ
    -0.09
    UTO
    -0.09
    è¿Ļæĺ¯ä¸Ģç§į
    -0.09
    POSITIVE LOGITS
    们
    0.12
    innen
    0.10
    æīĭåĨĮ
    0.09
    ÑĪÑĥ
    0.09
    son
    0.09
    ÑĪа
    0.08
    ä¹ĭ家
    0.08
    sWith
    0.08
    åı²è¯Ĺ
    0.08
    unes
    0.08
    Activations Density 0.010%

    No Known Activations