© 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. 27922
    Prev
    Next
    INDEX
    Explanations

    lucky enough to

    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
    inte
    -0.11
    éķ¿åŁİ
    -0.10
    好æĦŁ
    -0.10
    ç»ıåħ¸çļĦ
    -0.10
    uela
    -0.09
    è¿Ļ个è¡Įä¸ļ
    -0.09
    æĸ¹å¼ıè¿Ľè¡Į
    -0.09
    /Base
    -0.09
    ç«£
    -0.09
    å¾Īéĩįè¦ģ
    -0.09
    POSITIVE LOGITS
    产åĩº
    0.09
    èĥ½åľ¨
    0.09
    èĥ½æī¾åĪ°
    0.09
    ahl
    0.08
    eger
    0.08
    7
    0.08
    5
    0.08
    æĴŀ
    0.08
    0
    0.08
    å¸ļ
    0.08
    Activations Density 0.010%

    No Known Activations