© 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. 32646
    Prev
    Next
    INDEX
    Explanations

    actions that involve participation or creation

    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
    èĻĶ
    -0.08
    æĹ¶ç©º
    -0.08
    è¾Ļ
    -0.07
    èĪģ
    -0.07
    antic
    -0.07
     Bundy
    -0.07
     sân
    -0.07
    éģĩ
    -0.07
    uger
    -0.07
    åIJ
    -0.07
    POSITIVE LOGITS
    读åIJİ
    0.10
    为åħ¶
    0.09
    åıªä¼ļ
    0.09
    èĬľ
    0.09
    æĸ°å¨ĺ
    0.09
    åĪĿæģĭ
    0.09
    趾
    0.09
    åħ¥åѦ
    0.09
    çĦ¶æĺ¯
    0.09
    èħĬ
    0.09
    Activations Density 0.067%

    No Known Activations