© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8641
    Prev
    Next
    INDEX
    Explanations

    moving backward or reducing

    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
    èĬĹ
    -0.10
    acho
    -0.10
    ä¾ij
    -0.09
    æľīçĶ¨çļĦ
    -0.09
    å°ļ书
    -0.09
    ä¸ĭè¾¾
    -0.09
    ussion
    -0.09
    éĢĨ
    -0.09
    飧æĢ§
    -0.09
    è°ĥçļ®
    -0.09
    POSITIVE LOGITS
    éĢĢ
    0.14
    éĢĢåĩº
    0.11
    æĭīå¼Ģ
    0.11
    æĴ¤ç¦»
    0.11
    离å¼Ģ
    0.11
    å¼ĢæĿ¥
    0.10
    è¿ľç¦»
    0.10
    æĭĶ
    0.10
     retreat
    0.10
     withdrew
    0.10
    Activations Density 0.060%

    No Known Activations