© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36724
    Prev
    Next
    INDEX
    Explanations

    be, am, were, holds, through, not, someone

    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
    åĪ¸
    -0.09
    åĢĺ
    -0.09
    uala
    -0.09
    å¿łå®ŀ
    -0.09
    ç§ĭåĨ¬
    -0.08
    éģĩéļ¾
    -0.08
    éļIJéļIJ
    -0.08
    ddy
    -0.08
    apper
    -0.08
    ä¸īåĽĽ
    -0.08
    POSITIVE LOGITS
    éĺħåİĨ
    0.08
    /or
    0.08
    éĩį
    0.08
     Ashe
    0.08
    æĪį
    0.08
    责任人
    0.08
     ChÃŃnh
    0.08
    è¿Ľåº¦
    0.08
    Kelly
    0.08
    acular
    0.08
    Activations Density 0.102%

    No Known Activations