© 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. 56775
    Prev
    Next
    INDEX
    Explanations

    after 'will' or 'can'

    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
     however
    -0.13
    ä¹ĭæīĢ以
    -0.11
     поÑĤомÑĥ
    -0.10
     же
    -0.10
    åIJ§
    -0.10
    åĽłä¸º
    -0.10
    åĽłä¸ºåľ¨
    -0.10
     porém
    -0.10
    æĺ¯ä¸įæĺ¯
    -0.09
    åĩ¡æĺ¯
    -0.09
    POSITIVE LOGITS
    å®ŀåľ¨æĺ¯å¤ª
    0.11
    ä¸ŃéĢĶ
    0.10
    å®ŀåľ¨å¤ª
    0.09
     przecież
    0.09
    åľ¨æѤä¹ĭåīį
    0.09
    èĢĮçİ°åľ¨
    0.08
    æ¯ķç«Ł
    0.08
    éĤ£ä¹Ī容æĺĵ
    0.08
    obia
    0.08
    æ¯ķç«Łæĺ¯
    0.08
    Activations Density 0.377%

    No Known Activations