© 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. 61390
    Prev
    Next
    INDEX
    Explanations

    composed of / made up of

    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¸Ģ段
    -0.11
    ä¸ĢæĿ¡é¾Ļ
    -0.09
    Ãło
    -0.09
    ç¬Ļ
    -0.09
    åĻľ
    -0.09
    大åĨĽ
    -0.09
    imeline
    -0.09
    gang
    -0.09
    æĹı群
    -0.09
    触åıĬ
    -0.09
    POSITIVE LOGITS
    æĪIJåijĺ
    0.17
     gá»ĵm
    0.14
    ç»ĦæĪIJ
    0.14
     ÑģоÑģÑĤоиÑĤ
    0.12
     members
    0.11
     consists
    0.11
     consisting
    0.10
    ç»ĦæĪIJéĥ¨åĪĨ
    0.10
     consisted
    0.10
     consist
    0.10
    Activations Density 0.136%

    No Known Activations