© 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. 54842
    Prev
    Next
    INDEX
    Explanations

    bert takes

    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
    ijľ
    -0.11
    建ç«ĭåģ¥åħ¨
    -0.11
    容å¿į
    -0.09
    å¸ĥæĭī
    -0.09
    æłĭ
    -0.08
    oron
    -0.08
    èĥ½åĬĽåĴĮ
    -0.08
    缼
    -0.08
    uela
    -0.08
    Margins
    -0.08
    POSITIVE LOGITS
    urge
    0.10
    é¢į
    0.09
    ÃľR
    0.09
    å¿ĥæĻº
    0.09
    åıĪæĺ¯
    0.09
    举
    0.09
    æĭ³å¤´
    0.09
    ocha
    0.08
    çĤĻ
    0.08
    .increment
    0.08
    Activations Density 0.020%

    No Known Activations