© 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. 27338
    Prev
    Next
    INDEX
    Explanations

    you will get

    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¶Ĭ
    -0.10
    lear
    -0.10
    ointment
    -0.09
    çĿ¿
    -0.09
    æĬ±
    -0.09
    SplitOptions
    -0.09
    erk
    -0.08
    stroy
    -0.08
    ELS
    -0.08
    lya
    -0.08
    POSITIVE LOGITS
    åı¯ä»¥è¯´æĺ¯
    0.13
    åı¯è°ĵæĺ¯
    0.12
    è¿Ļä¹Ī说
    0.11
    ä¸Ģåı¥è¯Ŀ
    0.11
    åı¯ä»¥è¯´
    0.11
    วล
    0.10
    æį¢åı¥è¯Ŀ说
    0.10
    说æĺ¯
    0.10
    åı¯è°ĵ
    0.09
    æĦı大
    0.09
    Activations Density 0.055%

    No Known Activations