© 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. 56148
    Prev
    Next
    INDEX
    Explanations

    illegal and black market activities

    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
    éķ¶
    -0.10
    诰
    -0.10
    群éĩĮ
    -0.10
    ä¸ĢéĶ®
    -0.09
    èĤ¥
    -0.09
     Vid
    -0.09
    ç¾ģ
    -0.09
    æĭĻ
    -0.09
    è¥Ł
    -0.09
    åģĪ
    -0.09
    POSITIVE LOGITS
    åģ·åģ·
    0.24
    åģ·
    0.22
    ç§ģä¸ĭ
    0.21
    éĿŀæ³ķ
    0.21
    ç§ģèĩª
    0.21
    è¿Ŀè§Ħ
    0.21
     unofficial
    0.19
    è¿Ŀæ³ķ
    0.19
    æĵħèĩª
    0.18
     illegal
    0.17
    Activations Density 0.208%

    No Known Activations