© 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. 21722
    Prev
    Next
    INDEX
    Explanations

    working with

    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
    讲课
    -0.09
     ÑĥÑģпеÑĪ
    -0.09
    æ¦Ń
    -0.09
    jh
    -0.09
    åŁĭ
    -0.09
    ä¹ıåĬĽ
    -0.08
     disc
    -0.08
     Saúde
    -0.08
     lessons
    -0.08
    disk
    -0.08
    POSITIVE LOGITS
    pmat
    0.09
     бÑĥмаг
    0.09
    驯
    0.09
    æīĵ交éģĵ
    0.09
     pis
    0.08
    ä¾®
    0.08
    å¼Ħ
    0.08
     Colonial
    0.08
    段æŶéĹ´
    0.08
    earer
    0.08
    Activations Density 0.162%

    No Known Activations