© 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. 55404
    Prev
    Next
    INDEX
    Explanations

    packing essentials

    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
    åIJ¦
    -0.11
    å«ģç»Ļ
    -0.10
    æ¦Ĥ
    -0.09
    æĮİ
    -0.09
    ç©¿ä¸Ĭ
    -0.09
    离
    -0.09
    飫
    -0.09
    ogle
    -0.09
    ç»´å¥ĩ
    -0.09
    ibil
    -0.09
    POSITIVE LOGITS
    åīĥ
    0.10
    ä¸Ģ次
    0.09
    çĶ¨åĵģ
    0.09
    ode
    0.09
    æľĶ
    0.09
    ä¸Ģæŀ¶
    0.08
    py
    0.08
    åį«çĶŁåģ¥åº·
    0.08
    oa
    0.08
    æ³Ĭ
    0.08
    Activations Density 0.021%

    No Known Activations