© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50849
    Prev
    Next
    INDEX
    Explanations

    historical context and connections

    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
    éļ¾æĢª
    -0.10
    家åĸ»æĪ·æĻĵ
    -0.09
    ä¸ļåĨħ人士
    -0.09
    æľīå¾ħ
    -0.09
    æľīçļĦçݩ家
    -0.08
    å¯IJ
    -0.08
     GANG
    -0.08
    ãĢĿ
    -0.08
    ä¸Ŀ毫
    -0.08
    LIKELY
    -0.08
    POSITIVE LOGITS
    fuck
    0.08
    çŃĿ
    0.08
     titular
    0.08
    æ¶Ĥ
    0.08
     ÑģобÑģÑĤвенно
    0.08
     fuck
    0.07
     viet
    0.07
    WithValue
    0.07
    Dot
    0.07
    æģ©
    0.07
    Activations Density 1.058%

    No Known Activations