© 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. 47671
    Prev
    Next
    INDEX
    Explanations

    ironically, paradoxically, curiously, surprisingly

    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
    红线
    -0.09
     Bald
    -0.09
    èĻ¹
    -0.09
    èĻ¹æ¡¥
    -0.09
    лев
    -0.08
    king
    -0.08
    å±Ĥ次
    -0.08
     ÑģоÑĨиалÑĮно
    -0.08
    ewolf
    -0.08
    anter
    -0.08
    POSITIVE LOGITS
    åģıåģı
    0.13
    å±ħçĦ¶
    0.12
    æģ°æģ°
    0.12
    åıįèĢĮ
    0.12
    åıįåĢĴ
    0.11
    ç«Ł
    0.10
    è¶Ĭ大
    0.10
    æľĢ容æĺĵ
    0.10
     ironically
    0.10
    ç«ŁçĦ¶
    0.10
    Activations Density 0.065%

    No Known Activations