© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10570
    Prev
    Next
    INDEX
    Explanations

    This followed by file type

    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
    è¿Ļ两个
    -0.11
    è¿Ļæł·çļĦ
    -0.10
    /OR
    -0.09
    个å°ıæŶ
    -0.09
    è¿ĻäºĽéĹ®é¢ĺ
    -0.08
    è¿Ļåĩłä¸ª
    -0.08
    ervers
    -0.08
    秫
    -0.08
     Weiss
    -0.08
    aira
    -0.08
    POSITIVE LOGITS
    /th
    0.14
    å¹´é¾Ħ段
    0.14
    ترÙĨت
    0.12
     же
    0.12
    ìłĢ
    0.12
    ðŁijĨ
    0.11
    iner
    0.10
    bez
    0.10
     kind
    0.10
    çİ©æĦı
    0.10
    Activations Density 0.455%

    No Known Activations