© 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. 5134
    Prev
    Next
    INDEX
    Explanations

    hearing or listening to testimony

    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
    cil
    -0.09
    è¾Ī
    -0.09
    æįī
    -0.08
    家åĸ»æĪ·æĻĵ
    -0.08
    æľºåĬ¨
    -0.08
    çIJĨæĥ³çļĦ
    -0.08
    对åѦçĶŁ
    -0.07
    让åѦçĶŁ
    -0.07
     beau
    -0.07
    åıī
    -0.07
    POSITIVE LOGITS
    åIJ¬åıĸ
    0.17
    åIJ¬å®Į
    0.15
    åIJ¬äºĨ
    0.13
    åIJ¬åIJ¬
    0.12
    äºĨ解
    0.11
     hear
    0.11
    çŁ¥éģĵäºĨ
    0.11
    åIJ¬è¿ĩ
    0.11
    èģ½
    0.10
    åIJ¬
    0.10
    Activations Density 0.070%

    No Known Activations