© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 3060
    Prev
    Next
    INDEX
    Explanations

    programming and mathematical comparisons

    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
     dile
    -0.09
    çļĦä¿¡å¿ĥ
    -0.09
    æĢİèĥ½
    -0.09
    æŀľæĸŃ
    -0.08
    ä¸įéĻIJ
    -0.08
    æ£ĺ
    -0.08
    BJECT
    -0.08
    ä½ľèĢħæľ¬äºº
    -0.08
    å¼ĥ
    -0.08
    çĭłæĬĵ
    -0.08
    POSITIVE LOGITS
     overrun
    0.09
    èĦĪ
    0.09
    CORD
    0.09
    çĵľ
    0.09
    ä½ıäºĨ
    0.09
    è·Ŀ离
    0.08
    ä¸Ģèĩ´
    0.08
    eken
    0.08
    legalArgumentException
    0.08
    缸è¯Ĩ
    0.08
    Activations Density 0.004%

    No Known Activations