© 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. 26210
    Prev
    Next
    INDEX
    Explanations

    a stark illustration

    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
    æĪijåĸľæ¬¢
    -0.10
    ä»Ģä¹Īä¸ľè¥¿
    -0.10
    æľīåĩłä¸ª
    -0.09
    ä»Ģä¹Īåij¢
    -0.09
    æĢ¥åī§
    -0.09
    تصÙģ
    -0.08
    ä½ľæ¡Ī
    -0.08
    ongyang
    -0.08
    æľīçĽĬ
    -0.08
    æĪijåıijçİ°
    -0.08
    POSITIVE LOGITS
     testament
    0.19
    ä½ĵçİ°
    0.13
    ä½ĵçİ°äºĨ
    0.13
    表æĺİ
    0.11
    æıIJéĨĴ
    0.11
    è¯ģæĺİ
    0.10
     indication
    0.10
     testimony
    0.10
    è§ģè¯ģ
    0.10
    æĦŁåı¹
    0.10
    Activations Density 0.025%

    No Known Activations