© 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. 54621
    Prev
    Next
    INDEX
    Explanations

    events and activities

    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
    交æİ¥
    -0.10
    侬
    -0.09
    奢
    -0.09
    åģ¥åº·çļĦ
    -0.09
    nob
    -0.08
    è´±
    -0.08
    社ä¼ļ稳å®ļ
    -0.08
    lings
    -0.08
    åłĨ
    -0.08
    èĢģå©Ĩ
    -0.08
    POSITIVE LOGITS
    mony
    0.09
    é¢IJ
    0.08
    寻æ±Ĥ
    0.08
    FER
    0.08
    çļĦåIJĮåѦ
    0.08
    åįĹæŀģ
    0.08
    颧
    0.08
    blas
    0.08
    .Publish
    0.08
    åįĸç»Ļ
    0.08
    Activations Density 0.044%

    No Known Activations