© 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. 18046
    Prev
    Next
    INDEX
    Explanations

    verbs followed by objects/pronouns

    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
    çĶ¨æĿ¥
    -0.11
    ainers
    -0.10
    è¦ģ注æĦı
    -0.10
    usk
    -0.10
     gonna
    -0.10
    era
    -0.10
     gotta
    -0.09
    imon
    -0.09
    ppard
    -0.09
    è¦ģåıĬæŶ
    -0.09
    POSITIVE LOGITS
     ×ķ׾
    0.11
     ×ķ׾×Ķ
    0.10
    åĴĮåıijå±ķ
    0.09
     oneself
    0.08
    åĴĮæĶ¯æĮģ
    0.08
    ä¸ĭä¸ĢæŃ¥
    0.08
    çĽ¸ä¿¡
    0.08
    ä»ĬåIJİ
    0.08
    /be
    0.08
    èĬĻ
    0.08
    Activations Density 1.188%

    No Known Activations