© 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. 23143
    Prev
    Next
    INDEX
    Explanations

    compromise or strain

    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
    iness
    -0.09
    ĺ
    -0.09
    ying
    -0.08
    ä¼ļåIJĮ
    -0.08
    thing
    -0.08
    çı
    -0.08
    åľ¨è¿Ļç§į
    -0.08
    åĪĿåĪĽ
    -0.08
    ways
    -0.08
    /column
    -0.08
    POSITIVE LOGITS
    ety
    0.13
    ably
    0.11
    nable
    0.10
    è¡·
    0.09
    chw
    0.09
    人人éĥ½
    0.09
     Latter
    0.08
    ä¸įå°ijäºİ
    0.08
    sut
    0.08
    ç»Ĵ
    0.08
    Activations Density 0.011%

    No Known Activations