© 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. 57683
    Prev
    Next
    INDEX
    Explanations

    bowing or kneeling

    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
    åĸľæĤ¦
    -0.11
    æĺ¼
    -0.10
    é«ĺ强度
    -0.10
    opa
    -0.10
    é«ĺåİŁ
    -0.09
    æµģ转
    -0.09
    顶éĥ¨
    -0.09
    檫
    -0.09
    缧
    -0.09
    åŁ¤
    -0.09
    POSITIVE LOGITS
     cam
    0.10
    ä¸Ĭå¸Ŀ
    0.09
    躬
    0.09
    ç¾ļ
    0.08
    æģIJ
    0.08
    éļıå¤Ħ
    0.08
    EqualityComparer
    0.08
    éĻĮçĶŁäºº
    0.08
    ä¸ĭæĿ¥
    0.08
    ion
    0.08
    Activations Density 0.004%

    No Known Activations