© 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. 21337
    Prev
    Next
    INDEX
    Explanations

    experienced people and professionals

    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
    è´¨æĬ¼
    -0.10
    åİĨæŶ
    -0.10
    çĿ£
    -0.10
    ä¸Ĭ游
    -0.09
    ä¸ŃåĽ½æ¢¦
    -0.09
    éĶµ
    -0.09
    ç£Ĭ
    -0.09
    èĩªèIJ¥
    -0.09
    ä¸įä»ħä»ħæĺ¯
    -0.09
    宪
    -0.09
    POSITIVE LOGITS
    å°±åľ¨
    0.09
    æľį
    0.08
    æIJŃéħį
    0.08
    cy
    0.08
    交代
    0.08
    cwd
    0.08
     WN
    0.08
     forms
    0.08
     unde
    0.08
    åĬŁåĬĽ
    0.07
    Activations Density 0.004%

    No Known Activations