© 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. 2690
    Prev
    Next
    INDEX
    Explanations

    privilege or pleasure

    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
    ye
    -0.10
    èĨ³é£Ł
    -0.09
    åģĩæľŁ
    -0.09
    诱
    -0.09
     Consent
    -0.09
    æŀķ
    -0.09
    åĶ¤éĨĴ
    -0.09
    éªĭ
    -0.09
    æĶ¾åģĩ
    -0.09
    ä¹°çļĦ
    -0.09
    POSITIVE LOGITS
    èį£å¹¸
    0.12
    ä¸Ģè·¯ä¸Ĭ
    0.10
    æĪijè¿ĺ
    0.09
     seeing
    0.09
    çIJĨå·¥
    0.09
    纯åĩĢ
    0.09
    æĪijéĥ½
    0.09
    æĬĵ好
    0.09
    è§ģåĪ°
    0.09
    缮çĿ¹
    0.09
    Activations Density 0.055%

    No Known Activations