© 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. 58680
    Prev
    Next
    INDEX
    Explanations

    hyacinths, hygiene, hyponatremia, hypospadias

    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
     Glory
    -0.10
    便äºİ
    -0.09
    é³Į
    -0.09
    ych
    -0.09
    piece
    -0.09
    ê¸ī
    -0.09
    çģ°å°ĺ
    -0.08
    èĪIJ
    -0.08
    庾
    -0.08
    /or
    -0.08
    POSITIVE LOGITS
    gens
    0.11
    gger
    0.10
    æīĢå¾Ĺ
    0.09
    çŃĸ
    0.09
    åĽ½åľŁ
    0.09
    zem
    0.09
    ä¸įéĹ´
    0.09
    iggs
    0.09
    icode
    0.08
    å¾Ĺ天
    0.08
    Activations Density 0.037%

    No Known Activations