© 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. 27587
    Prev
    Next
    INDEX
    Explanations

    sent, loan, bought, save

    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
    æŃ¤ä¸¾
    -0.10
    mens
    -0.10
    巨大çļĦ
    -0.09
    iente
    -0.08
    è±¹
    -0.08
    Dou
    -0.08
    ä¸įå®ļ
    -0.08
    æı£
    -0.08
    ä»ĸæĺ¯
    -0.08
    声åĵį
    -0.08
    POSITIVE LOGITS
     us
    0.13
     вам
    0.13
     ÑĤебе
    0.12
    æ¯ıä½į
    0.12
     емÑĥ
    0.12
    人æ°ijæľįåĬ¡
    0.11
    人们
    0.11
     ÐĴам
    0.11
    客æĪ·æıIJä¾Ľ
    0.11
     мне
    0.11
    Activations Density 0.112%

    No Known Activations