© 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. 57794
    Prev
    Next
    INDEX
    Explanations

    want to / know what

    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
    åı£åij³
    -0.10
    æ±Ŀ
    -0.10
    ields
    -0.09
     СеÑĢ
    -0.09
    zman
    -0.09
    çĶŁæ´»ä¹łæĥ¯
    -0.09
    çĶŁæ´»æ°´å¹³
    -0.09
    èģĶåIJĪ
    -0.09
    å¿ĹæĦ¿
    -0.09
    æĹ¥å¸¸çĶŁæ´»
    -0.08
    POSITIVE LOGITS
     itself
    0.14
    è¯ĨåĪ«
    0.11
    åIJIJ
    0.10
    å¸®ä½ł
    0.10
    çĽį
    0.10
    åıĹçIJĨ
    0.10
    çĪ¬ä¸Ĭ
    0.09
    èĩªèº«
    0.09
    é©®
    0.09
    éĢĬ
    0.09
    Activations Density 0.548%

    No Known Activations