© 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. 3297
    Prev
    Next
    INDEX
    Explanations

    verdict or impression

    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
    enson
    -0.10
    iec
    -0.09
    缸继
    -0.09
    å¿Į
    -0.08
    æĻ¨æĬ¥
    -0.08
     chall
    -0.08
    itionally
    -0.08
    onth
    -0.08
    anky
    -0.08
    Ñĩи
    -0.08
    POSITIVE LOGITS
    æĢ»ä½ĵ
    0.11
    åŀĽ
    0.10
     verdict
    0.10
    Overall
    0.09
    è¯Ħ为
    0.09
     overall
    0.09
    æķ´ä½ĵ
    0.09
    æĴ
    0.09
    èī¯
    0.09
    ä¸įç¡®å®ļæĢ§
    0.09
    Activations Density 0.017%

    No Known Activations