© 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. 12022
    Prev
    Next
    INDEX
    Explanations

    section specifically

    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
    Ùħرا
    -0.10
    /xhtml
    -0.10
     Hubb
    -0.10
    antan
    -0.10
    ëł¹
    -0.09
    æĸŃ
    -0.09
    uttle
    -0.09
    DialogContent
    -0.09
     luáºŃt
    -0.09
    ENTITY
    -0.09
    POSITIVE LOGITS
    æłı
    0.15
    æłı缮
    0.15
     section
    0.14
    æ¬Ħ
    0.13
     category
    0.12
     ÑĢаздел
    0.12
    æĿ¿åĿĹ
    0.12
     sections
    0.12
    æ·±å¤Ħ
    0.11
    ä¸ĵåĮº
    0.10
    Activations Density 0.092%

    No Known Activations