© 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. 58183
    Prev
    Next
    INDEX
    Explanations

    download a copy

    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æŶéĹ´
    -0.10
    è¸ı
    -0.10
    iring
    -0.09
    çĶ¨å·¥
    -0.09
    URN
    -0.09
    åİŁåŀĭ
    -0.09
    orns
    -0.09
    è§Ĵèī²
    -0.09
    涮
    -0.09
    /generated
    -0.08
    POSITIVE LOGITS
    ãģ¾ãģĦ
    0.10
    æĪĸå¤ļ
    0.10
    /IP
    0.09
    åĬĽæ±Ĥ
    0.09
    æĮ¹
    0.09
    份
    0.08
    åºı
    0.08
    人人éĥ½
    0.08
    大éģĵ
    0.08
    ಮ
    0.08
    Activations Density 0.042%

    No Known Activations