© 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. 53550
    Prev
    Next
    INDEX
    Explanations

    all of them begin with words like 'all', 'both', 'whom'

    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
    ä¸įåĨį
    -0.10
    æīĢæľī
    -0.10
    æīĢæľīçļĦ
    -0.10
    ä¸įæĥ³
    -0.10
    åĪ°å¤Ħ
    -0.10
    ä¸ĢåĪĩéĥ½
    -0.10
    éĥ¨åĪĨåľ°åĮº
    -0.10
    æĺ¯æľĢ好çļĦ
    -0.10
    ãģĵãģĿ
    -0.10
    ä¸ĢåĪĩ
    -0.09
    POSITIVE LOGITS
    æĪĸå¤ļæĪĸå°ij
    0.14
    ä¸įåIJĮç¨ĭ度
    0.14
    åıĹåĪ°å½±åĵį
    0.12
    æĪĸå¤ļ
    0.11
    ä¸Ģå®ļç¨ĭ度
    0.10
     except
    0.10
    åıĤä¸İåĪ°
    0.10
    Except
    0.10
    tes
    0.10
    æĪĸå¤ļæĪĸ
    0.09
    Activations Density 0.091%

    No Known Activations