© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37208
    Prev
    Next
    INDEX
    Explanations

    to enable or facilitate actions

    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
    åĨĪ
    -0.10
    abad
    -0.09
     anth
    -0.08
    åĽ«
    -0.08
    aload
    -0.08
    æĪª
    -0.08
    ä¸Ģå®ļç¨ĭ度
    -0.08
    -import
    -0.08
    KER
    -0.08
    çļĦä¸ĢéĿ¢
    -0.08
    POSITIVE LOGITS
     sake
    0.11
    弥补
    0.11
    æĽ´å¥½åľ°
    0.11
    æĽ´å¥½çļĦ
    0.11
    æĽ´å¥½
    0.11
    便äºİ
    0.10
    æĸ¹ä¾¿
    0.10
    iej
    0.09
    ä¸ĭçıŃ
    0.09
     better
    0.09
    Activations Density 0.303%

    No Known Activations