© 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. 48655
    Prev
    Next
    INDEX
    Explanations

    conference proceedings

    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¿Ļåĩłä¸ª
    -0.08
    åĹ¡
    -0.08
     kè
    -0.08
    裸
    -0.08
    oyer
    -0.08
    åīįèĢħ
    -0.07
    æ´ĭ
    -0.07
     вн
    -0.07
    .Style
    -0.07
    Ñģел
    -0.07
    POSITIVE LOGITS
    (pp
    0.10
     proceedings
    0.09
    举è¡ĮçļĦ
    0.09
    _rg
    0.08
    ç»Ħå§Ķä¼ļ
    0.08
    æĬĽå¼ĥ
    0.08
     Proceedings
    0.08
    ä¸ĸçķĮåIJĦåĽ½
    0.08
    带åĽŀ
    0.08
    çĨ¹
    0.08
    Activations Density 0.050%

    No Known Activations