© 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. 33930
    Prev
    Next
    INDEX
    Explanations

    start of events

    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¸ĩä¸Ģ
    -0.10
    ç¿ĺ
    -0.09
     менее
    -0.09
    软
    -0.09
     WW
    -0.08
    ion
    -0.08
    (generator
    -0.08
    ativ
    -0.08
     gener
    -0.08
    why
    -0.08
    POSITIVE LOGITS
    认è¯ĨåĪ°
    0.09
    compat
    0.09
    æĪIJç«ĭ以æĿ¥
    0.09
    ç»IJ
    0.08
    ķĮ
    0.08
    对ä»ĸ们
    0.08
    éĤ£æ¬¡
    0.08
    æĦıè¯ĨåĪ°
    0.08
     contr
    0.08
    æݩ饰
    0.08
    Activations Density 0.018%

    No Known Activations