© 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. 11013
    Prev
    Next
    INDEX
    Explanations

    incident involving

    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
    娶
    -0.09
     Rut
    -0.09
    anc
    -0.08
    ä¸ĸ代
    -0.08
    äºĭä¸ļ
    -0.08
    ake
    -0.08
    ä¿Ł
    -0.08
    æĮĩåIJij
    -0.08
    侩
    -0.08
    setting
    -0.07
    POSITIVE LOGITS
    éĤ£æ¬¡
    0.13
    ä¸Ģçķª
    0.12
    åıijçĶŁåľ¨
    0.11
    è¿ĩçļĦ
    0.10
    ä¸Ģ次
    0.10
    äºĨä¸Ģåľº
    0.10
    ä»ĸæĽ¾
    0.10
    è¿Ļ段
    0.10
    éļ¾å¿ĺ
    0.10
    interaction
    0.10
    Activations Density 0.107%

    No Known Activations