© 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. 12064
    Prev
    Next
    INDEX
    Explanations

    business practices and techniques

    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
    ä¸įåIJĮç±»åŀĭ
    -0.09
    .tc
    -0.09
    getExtension
    -0.08
    voke
    -0.08
    rapper
    -0.08
    享æľī
    -0.08
     horn
    -0.08
     Ðļа
    -0.08
     ÑįкÑģ
    -0.08
    æĤ£
    -0.08
    POSITIVE LOGITS
     practices
    0.23
     Practices
    0.19
     techniques
    0.19
    åģļæ³ķ
    0.17
     methods
    0.16
    actices
    0.15
     Techniques
    0.14
     processes
    0.13
     tactics
    0.13
    æĬĢå·§
    0.13
    Activations Density 0.158%

    No Known Activations