© 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. 48390
    Prev
    Next
    INDEX
    Explanations

    abolishing or ending things

    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
    佩
    -0.09
    å¿ĥæĦı
    -0.09
    åīįæĸ¹
    -0.09
    uche
    -0.09
    èĪī
    -0.09
    çļĦéģĵçIJĨ
    -0.09
    eced
    -0.08
    河西
    -0.08
    ients
    -0.08
    çİ°å®ŀ
    -0.08
    POSITIVE LOGITS
     altogether
    0.12
     abol
    0.11
     вообÑīе
    0.10
    åŃĺåľ¨çļĦ
    0.10
    ç³Ł
    0.10
     abolished
    0.09
    ä¸Ģ次æĢ§
    0.09
    .exist
    0.09
    çļĦåŃĺåľ¨
    0.09
     existence
    0.09
    Activations Density 0.064%

    No Known Activations