© 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. 58507
    Prev
    Next
    INDEX
    Explanations

    abuse of position or power

    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
    æĤ¸
    -0.09
    ä¸ĬæĬ¥
    -0.09
    èĻ¢
    -0.09
    ALLE
    -0.09
    ermo
    -0.09
    pir
    -0.09
    纶
    -0.09
     Dak
    -0.08
    елÑĮ
    -0.08
    .bel
    -0.08
    POSITIVE LOGITS
    èģĮæĿĥ
    0.13
    滥çĶ¨
    0.10
    OMET
    0.09
    UTH
    0.09
    è¯Ī
    0.09
     Tart
    0.09
     Freedom
    0.08
    ingen
    0.08
     abuse
    0.08
     Abuse
    0.08
    Activations Density 0.016%

    No Known Activations