© 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. 37745
    Prev
    Next
    INDEX
    Explanations

    superpower or hyperpower

    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
    æĥħæĢĢ
    -0.09
    å¸ı
    -0.09
    ename
    -0.09
    ä¿¡å¿ĥ
    -0.08
    æľīçļĦ
    -0.08
    Ùĥر
    -0.08
    æĭĽ
    -0.08
    è¿Ķ
    -0.08
    åīįä»»
    -0.08
    queeze
    -0.08
    POSITIVE LOGITS
     ÑĦÑĥÑĤ
    0.09
    /engine
    0.08
     samt
    0.08
    å·¥ç¨ĭå¸Ī
    0.08
     Ling
    0.08
    æľī趣çļĦ
    0.08
    å±ł
    0.08
     fascinating
    0.08
    è´º
    0.08
    ç²¾å¯Ĩ
    0.08
    Activations Density 0.298%

    No Known Activations