© 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. 6314
    Prev
    Next
    INDEX
    Explanations

    include

    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
    çī¯
    -0.10
    ç«ĭæ¡Ī
    -0.10
    æĻ·
    -0.09
    å¾Ĺä¸įåĪ°
    -0.09
    trak
    -0.09
    éļıæŶéļıåľ°
    -0.09
    erie
    -0.09
    为空
    -0.09
    æĶ¾å¼Ģ
    -0.09
    sheets
    -0.09
    POSITIVE LOGITS
    åľ¨åĨħçļĦ
    0.11
    /ex
    0.10
     ÑģебÑı
    0.10
    oop
    0.10
    æĸ¹å¼ıè¿Ľè¡Į
    0.09
    à¹ĥà¸Ļà¸ģาร
    0.09
    /include
    0.09
    LAG
    0.08
    è¶³å¤ŁçļĦ
    0.08
     ocas
    0.08
    Activations Density 0.103%

    No Known Activations