© 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. 29891
    Prev
    Next
    INDEX
    Explanations

    upon Thames

    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
    uffer
    -0.09
     Roose
    -0.09
    alore
    -0.09
    parator
    -0.09
    çķ´
    -0.09
    екÑģ
    -0.09
    ETCH
    -0.08
    etch
    -0.08
    å¾Ń
    -0.08
    èĮ´
    -0.08
    POSITIVE LOGITS
    hart
    0.08
     dotyc
    0.08
    å±ĢéĿ¢
    0.08
    驾é©Ń
    0.08
    thon
    0.08
    个å°ıæŶ
    0.08
    对该
    0.08
    æ¿Ģ
    0.08
    人æ°ijåĮ»éĻ¢
    0.08
    æĬ¼
    0.08
    Activations Density 0.001%

    No Known Activations