© 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. 32662
    Prev
    Next
    INDEX
    Explanations

    quite followed by adjective/adverb

    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
    渤
    -0.10
    姹
    -0.09
     дело
    -0.09
    éĵĥ
    -0.08
    anst
    -0.08
    åĴĮå®ĮåĸĦ
    -0.08
     Rudd
    -0.08
     Serg
    -0.08
    ä¸į对
    -0.08
     Nicolas
    -0.08
    POSITIVE LOGITS
    æĹ©å°±
    0.10
    ibaba
    0.10
    åĪ°æľĢåIJİ
    0.09
    èħ¼
    0.09
    å¾Īé«ĺ
    0.08
    çļĦåĬĽéĩı
    0.08
    athom
    0.08
    ayah
    0.08
    gage
    0.08
    -LAST
    0.08
    Activations Density 0.018%

    No Known Activations