© 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. 52156
    Prev
    Next
    INDEX
    Explanations

    adverbs modifying nouns or adjectives

    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
    hev
    -0.09
    persons
    -0.09
     Zheng
    -0.09
    æĭħ
    -0.08
    soon
    -0.08
    tec
    -0.08
    ctp
    -0.08
    éĩį大
    -0.08
    apt
    -0.08
     Ä°ki
    -0.08
    POSITIVE LOGITS
    æīĴ
    0.10
    追éļı
    0.10
    çĭIJ
    0.10
    ä¹IJè§Ĥ
    0.09
    ä¸ĢåĽ¢
    0.09
    æķ´é¡¿
    0.09
    çļĦåħ³æ³¨
    0.09
    æĬĵ好
    0.09
    ç®Ģ缴
    0.09
    åĵij
    0.09
    Activations Density 0.027%

    No Known Activations