© 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. 34892
    Prev
    Next
    INDEX
    Explanations

    dirty sexy money

    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
    ç«¿
    -0.09
    itti
    -0.09
    owa
    -0.09
    è¾ķ
    -0.09
    é¹ŀ
    -0.09
    ÑĢин
    -0.08
    Parsed
    -0.08
    Slash
    -0.08
    itto
    -0.08
    æ¯ħ
    -0.08
    POSITIVE LOGITS
    dirty
    0.17
    èĦı
    0.17
     dirty
    0.16
    _dirty
    0.14
    Dirty
    0.14
    污
    0.14
    ç²Ĺç³Ļ
    0.14
    èĤ®
    0.13
     Dirty
    0.13
    æ±ļ
    0.13
    Activations Density 0.033%

    No Known Activations