© 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. 5192
    Prev
    Next
    INDEX
    Explanations

    That's or is leading to result

    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
     Carpet
    -0.10
    æĭĴ
    -0.09
    ÐļÑĥÑĢÑģ
    -0.09
    istine
    -0.09
    åįģåĽĽäºĶ
    -0.09
     оÑĤз
    -0.09
    à¸Ĺำà¹Ħม
    -0.09
    abeled
    -0.09
    Stamp
    -0.09
    æĸ°æĺ¥
    -0.09
    POSITIVE LOGITS
    缸å½ĵäºİ
    0.09
    ëĭ¨
    0.08
    ä¹Łåıªèĥ½
    0.08
    çķ¢ç«Ł
    0.08
    åĬŃ
    0.08
    æ¯ķç«Ł
    0.08
    .collect
    0.08
    æģ¸
    0.07
    -d
    0.07
    ä¸Ģå¼ł
    0.07
    Activations Density 0.024%

    No Known Activations