© 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. 50072
    Prev
    Next
    INDEX
    Explanations

    not understand / alter / get / have / consent

    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
    oling
    -0.10
    å®ĮäºĨ
    -0.09
    æ¯ıä¸Ģ个人
    -0.08
     اÙĦÙħÙĨا
    -0.08
    .Cmd
    -0.08
    phabet
    -0.08
    alach
    -0.08
    æľĢåħĪè¿ĽçļĦ
    -0.08
    lek
    -0.08
    ol
    -0.08
    POSITIVE LOGITS
    htags
    0.10
    lah
    0.10
    å©¿
    0.09
    æĢ
    0.09
    æĥħå¢ĥ
    0.09
    inks
    0.09
    /single
    0.08
    esis
    0.08
    bole
    0.08
    ÑĪей
    0.08
    Activations Density 0.063%

    No Known Activations