© 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. 39036
    Prev
    Next
    INDEX
    Explanations

    bi-, di-, tri-, quad- prefixes

    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
     divis
    -0.10
    å¤ļåħĥåĮĸ
    -0.10
    ../
    -0.09
    iard
    -0.09
    ç©·
    -0.09
     Ratings
    -0.09
    èĥ½å¾ĹåĪ°
    -0.09
    ople
    -0.08
    ä¸Ĭæĸ¹
    -0.08
    ç¼Ĺ
    -0.08
    POSITIVE LOGITS
    æĪĸå¤ļ
    0.26
    /th
    0.11
    ,double
    0.11
     dual
    0.10
    åıĮ
    0.10
    -two
    0.10
    /tr
    0.10
    ä¸Ģä½ĵ
    0.09
    åıĮéĩį
    0.09
    éĽĻ
    0.09
    Activations Density 0.260%

    No Known Activations