© 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. 17007
    Prev
    Next
    INDEX
    Explanations

    man, gentleman, or woman followed by whom/who/they

    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
    æĮ¤
    -0.10
    éĩįä¸Ń
    -0.10
    æ¾Ī
    -0.09
    çݰ身
    -0.09
    èIJ¥åķĨ
    -0.09
    æĤł
    -0.09
    ä¸ĭæīĭ
    -0.09
    æķ¬è¯·
    -0.09
    ç»ıå¼Ģ
    -0.09
    çĶ¨åĬĽ
    -0.08
    POSITIVE LOGITS
    èµĸ以çĶŁåŃĺ
    0.10
     whose
    0.09
    ose
    0.09
     اÙĦØ°ÙĬ
    0.08
     اÙĦتÙĬ
    0.08
     beste
    0.08
     closest
    0.08
     ç¬
    0.08
    whose
    0.08
    京津
    0.08
    Activations Density 0.206%

    No Known Activations