© 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. 35113
    Prev
    Next
    INDEX
    Explanations

    too good to be true

    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
    iso
    -0.09
    ä¸įè¿ľ
    -0.09
    æģĤ
    -0.09
    æĬĬè¿Ļ个
    -0.08
    ä»»ä½ķå½¢å¼ı
    -0.08
    hausen
    -0.08
     Ferr
    -0.08
    arak
    -0.08
    ä»»æĦı
    -0.08
    é¢ĦåĪ¶
    -0.08
    POSITIVE LOGITS
    è¿ĩäºİ
    0.15
    太è¿ĩ
    0.13
     TOO
    0.11
    å®ŀåľ¨æĺ¯å¤ª
    0.11
     too
    0.11
    太
    0.11
    æį·
    0.10
    太快
    0.10
    æĺ¯çľŁçļĦ
    0.10
    å®ŀåľ¨å¤ª
    0.10
    Activations Density 0.088%

    No Known Activations