© 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. 5558
    Prev
    Next
    INDEX
    Explanations

    twice, or ice

    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
    ../
    -0.11
    eer
    -0.10
    ../../../
    -0.09
    stud
    -0.09
    atics
    -0.09
    swer
    -0.09
    nine
    -0.08
    ding
    -0.08
    ÃĩÃĥO
    -0.08
    wealth
    -0.08
    POSITIVE LOGITS
    ä¸Ģ个æĺ¯
    0.12
    -thirds
    0.11
     dozen
    0.10
    inkle
    0.10
    ächt
    0.09
    /th
    0.09
    åݦ
    0.09
    eking
    0.09
    .parseDouble
    0.09
    /random
    0.09
    Activations Density 0.008%

    No Known Activations