© 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. 6640
    Prev
    Next
    INDEX
    Explanations

    prepositions indicating location or time

    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
    ä½ľä¸ºä¸ĢåIJį
    -0.11
    人们çļĦ
    -0.10
    è¿Ļ个人
    -0.10
    éĽ
    -0.10
    éĤ£äºº
    -0.09
    ä¸ĢåįĬ
    -0.09
    人们
    -0.09
    åħ¬æĬ¥
    -0.09
    身为
    -0.09
    åľ¨åĵª
    -0.09
    POSITIVE LOGITS
     whom
    0.12
    路边
    0.10
    ê²IJ
    0.09
    æ¿Ĵ临
    0.09
     ÑģеÑĢвиÑģ
    0.08
    à¹ģ
    0.08
     whose
    0.08
     اÙĦØ°ÙĬÙĨ
    0.08
    ç´łè´¨æķĻèĤ²
    0.08
     who
    0.08
    Activations Density 0.377%

    No Known Activations