© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20374
    Prev
    Next
    INDEX
    Explanations

    Lego sets and figures

    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
    èĢģ乡
    -0.10
    åĬĽè¿ĺæĺ¯
    -0.09
    jer
    -0.09
    çĭ¼
    -0.08
    iÄĩ
    -0.08
     sut
    -0.08
    åľ¨å®¶
    -0.08
    .Formatter
    -0.08
    å®ļçĤ¹
    -0.08
    çĨ¬
    -0.08
    POSITIVE LOGITS
    ansson
    0.11
    eft
    0.10
    é¼ł
    0.09
    æĢ«
    0.09
    callback
    0.09
    jang
    0.09
    榫
    0.09
     Lego
    0.09
    èĤ¢
    0.09
    .callback
    0.09
    Activations Density 0.095%

    No Known Activations