© 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. 38712
    Prev
    Next
    INDEX
    Explanations

    economics suggests | winds caused | politician did | ask not | metric splitting | embedded tumor

    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
    ç»Ŀ大å¤ļæķ°
    -0.10
    å°¤åħ¶æĺ¯åľ¨
    -0.09
    åį³æŶ
    -0.09
    æľĢ容æĺĵ
    -0.08
    otti
    -0.08
    ç»Ŀ大éĥ¨åĪĨ
    -0.08
    å°¤åħ¶æĺ¯
    -0.08
    åĶ¿
    -0.08
    2
    -0.08
    ä»»ä½ķä¸Ģ个
    -0.08
    POSITIVE LOGITS
    çŀ°
    0.09
    罨
    0.09
    幸好
    0.09
     вмеÑģÑĤ
    0.08
    楼ä¸ĭ
    0.08
    [--
    0.08
    产ä¸ļéĽĨèģļ
    0.08
    åIJĽ
    0.08
    æİ£
    0.08
    èݽ
    0.08
    Activations Density 0.026%

    No Known Activations