© 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. 56849
    Prev
    Next
    INDEX
    Explanations

    conjunctions and contrasting words

    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
    æł¹æľ¬ä¸į
    -0.09
    ilar
    -0.09
    èIJĥ
    -0.09
    æł¹æľ¬æ²¡æľī
    -0.09
    ä¸Ģ举
    -0.09
    ä¿®åħ»
    -0.08
    æIJ¶
    -0.08
    èĶ·
    -0.08
    æ±Łå±±
    -0.08
     ведÑĮ
    -0.08
    POSITIVE LOGITS
    éĢļ常æĺ¯
    0.10
    ä¸į失
    0.10
    deer
    0.09
    æĽĻ
    0.09
    æĪĸå¤ļ
    0.08
    å°ıåŃIJ
    0.08
    /OR
    0.08
    æļ´é£İ
    0.08
    åģ¥åº·çļĦ
    0.08
    å¾Ģå¾Ģæĺ¯
    0.08
    Activations Density 0.106%

    No Known Activations