© 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. 55105
    Prev
    Next
    INDEX
    Explanations

    slight imperfections or no major defects

    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
    缤纷
    -0.10
    çķĻä¸ĭæĿ¥
    -0.10
    è¯ŀ
    -0.09
    æī¹å¤į
    -0.09
    çķĻç»Ļ
    -0.09
    па
    -0.09
    åij½è¿IJ
    -0.09
    éħĬ
    -0.09
    éħ£
    -0.09
    æѦæŀĹ
    -0.09
    POSITIVE LOGITS
    轻微
    0.10
    èĤī
    0.09
    æĸ¯é¡¿
    0.08
    arian
    0.08
    æļ´åĬĽ
    0.08
    è¢Ī
    0.08
    交æµģåIJĪä½ľ
    0.08
     moderate
    0.08
    ç£ķ
    0.08
    8
    0.08
    Activations Density 0.042%

    No Known Activations