© 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. 39825
    Prev
    Next
    INDEX
    Explanations

    arrietty trailer, Yeezus, Schweitzer, RLgraph authors, Atzmon, melioidosis, Derek, cummings, Gladwell, Johnson

    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
    èĮĦ
    -0.11
    鸢
    -0.10
    Sad
    -0.10
    erin
    -0.10
    .edu
    -0.09
    erde
    -0.09
    ä¼ŀ
    -0.09
     Ñĥдален
    -0.09
     ÑĢеÑĪа
    -0.09
    æĺĬ
    -0.09
    POSITIVE LOGITS
    ported
    0.09
     stalk
    0.08
    stalk
    0.08
    åĺ±
    0.08
    istrate
    0.08
     propriet
    0.08
    åģļé¥Ń
    0.08
    åıijè´¢
    0.07
    æĸ°åŁİ
    0.07
    çĺ¦
    0.07
    Activations Density 0.066%

    No Known Activations