© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36058
    Prev
    Next
    INDEX
    Explanations

    the sequence "the" followed by a word

    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
    æŨ
    -0.10
    éĺ¿å§¨
    -0.09
    çıł
    -0.09
    æĪ¾
    -0.09
    /cop
    -0.09
    忽
    -0.09
    æĦı大
    -0.09
    æīĢ说
    -0.09
    åī¯æł¡éķ¿
    -0.08
    åī¯å¸Ĥéķ¿
    -0.08
    POSITIVE LOGITS
    é¢Ģ
    0.10
    å¹³åĿĩ
    0.09
    åĨħå¤ĸ
    0.09
    çļĦä¸ĢåĪĩ
    0.09
    åŁºå±Ĥ
    0.08
    åĨħåľ¨
    0.08
    æĦŁè§īåĪ°
    0.08
    кÑĢо
    0.08
    èĻļæĭŁ
    0.08
    ä¸ĵ线
    0.08
    Activations Density 0.023%

    No Known Activations