© 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. 62038
    Prev
    Next
    INDEX
    Explanations

    part of collections

    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
    _MAXIMUM
    -0.09
    æĭĽåķĨ
    -0.09
    åIJĦèĩª
    -0.08
    _SUFFIX
    -0.08
    åIJĦ家
    -0.08
    èĽĢ
    -0.08
    _rsp
    -0.08
    åıĭæĥħ
    -0.08
    æĭĵ
    -0.08
     Pand
    -0.08
    POSITIVE LOGITS
    ä»ĸçļĦ
    0.24
    äºĨä»ĸçļĦ
    0.22
    ä»ĸåľ¨
    0.19
    æĺ¯ä»ĸ
    0.18
     his
    0.18
    çľĭä»ĸ
    0.17
    ä»ĸ对
    0.16
    åĴĮä»ĸçļĦ
    0.16
    åľ¨ä»ĸçļĦ
    0.16
    his
    0.16
    Activations Density 0.216%

    No Known Activations