© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9525
    Prev
    Next
    INDEX
    Explanations

    section start

    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
    åªļ
    -0.09
    æ¯ķç«Ł
    -0.08
    foot
    -0.08
    æ±Ĭ
    -0.08
    orate
    -0.08
    å¦Ĥæľīä¾µæĿĥ
    -0.08
    æĻ´
    -0.07
    ç»ĻæĤ¨
    -0.07
    è¡·å¿ĥ
    -0.07
    ç¼
    -0.07
    POSITIVE LOGITS
    alytics
    0.10
    auc
    0.09
    FIRST
    0.09
    以çĤº
    0.08
    ä¸ĢåĢĭ人
    0.08
     oslo
    0.08
    webElement
    0.08
    WITHOUT
    0.08
    à¹Ĥà¸Ľà¸£à¹ģ
    0.08
     bulbs
    0.08
    Activations Density 8.982%

    No Known Activations