© 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. 20934
    Prev
    Next
    INDEX
    Explanations

    Format Version

    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
    äºĶæĺŁ
    -0.08
    _modes
    -0.08
    azio
    -0.08
    asio
    -0.08
    顺
    -0.08
    éŃģ
    -0.08
    пан
    -0.08
    éĻ¨
    -0.08
    pch
    -0.07
    å°±ç®Ĺæĺ¯
    -0.07
    POSITIVE LOGITS
    ä¸ĢåĪĩéĥ½
    0.09
     бÑĭ
    0.09
    ä¸įç»ıæĦı
    0.09
    é²µ
    0.09
    dance
    0.09
     Kiss
    0.08
    æŃ¥æŃ¥
    0.08
    æĥķ
    0.08
    ught
    0.08
    æ¸Ĭ
    0.08
    Activations Density 1.316%

    No Known Activations