© 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. 19332
    Prev
    Next
    INDEX
    Explanations

    the reason for

    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
    ighth
    -0.11
    æĦıåij³çĿĢ
    -0.10
    ¾ç¤º
    -0.10
    ä»Ģä¹ĪæĦıæĢĿ
    -0.10
    åĽŀåIJĪ
    -0.09
    äºĭå®ľ
    -0.09
    è¿Ļ件äºĭæĥħ
    -0.09
    ç¨Ĺ
    -0.09
    é¦Ĵ头
    -0.09
    ä¸Ģçŀ¬éĹ´
    -0.09
    POSITIVE LOGITS
    å¹¿æ³Ľåħ³æ³¨
    0.08
     caution
    0.08
    coverage
    0.08
    ä¹ĭæīĢ以
    0.08
    ä¾£
    0.08
     having
    0.08
    æĭħ忧
    0.08
     каж
    0.08
    åľ¨éĤ£éĩĮ
    0.08
     needing
    0.08
    Activations Density 0.039%

    No Known Activations