© 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. 63473
    Prev
    Next
    INDEX
    Explanations

    inappropriate language

    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
    èľľ
    -0.09
     Sour
    -0.09
    .tf
    -0.09
    reuse
    -0.08
    SION
    -0.08
    ILA
    -0.08
    @brief
    -0.08
    ables
    -0.08
    æĻĭ
    -0.08
    æŀ¢
    -0.08
    POSITIVE LOGITS
    åİĨåı²æĢ§
    0.12
    åīįçŀ»
    0.10
    yms
    0.10
    æĸ°é¢ĸ
    0.09
    ock
    0.09
    åĩ¡äºĭ
    0.09
    ä¼ļç»Ļ
    0.09
    oga
    0.09
    ä¼ļ让
    0.08
    举缣
    0.08
    Activations Density 0.015%

    No Known Activations