© 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. 19283
    Prev
    Next
    INDEX
    Explanations

    proof that, start button, vehicle thereby, isolated exercise

    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
    æĨĭ
    -0.09
    action
    -0.09
    ught
    -0.09
    igo
    -0.08
    UGHT
    -0.08
    åĬ¿
    -0.08
    ackle
    -0.08
    åħ¨éķ¿
    -0.08
    æ²¹
    -0.08
    ForObject
    -0.08
    POSITIVE LOGITS
    篷
    0.11
    iese
    0.09
    åIJİå¤ĩ
    0.08
    è¿Ŀæ³ķçĬ¯ç½ª
    0.08
    zens
    0.08
    åIJ¸
    0.08
    æŁĺ
    0.08
    æĹĮ
    0.08
    åĿļå®ļä¸į移
    0.08
    burgh
    0.08
    Activations Density 0.027%

    No Known Activations