© 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. 39241
    Prev
    Next
    INDEX
    Explanations

    Borough of

    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
    æĶ¯éĺŁ
    -0.09
    orious
    -0.09
    apult
    -0.09
    é¬ĵ
    -0.09
    urus
    -0.09
    omics
    -0.08
    ÑĢай
    -0.08
    жд
    -0.08
    çĸij
    -0.08
    ernel
    -0.08
    POSITIVE LOGITS
    æĦıæĢĿ
    0.10
    anson
    0.09
    ä¸Ģç³»åĪĹ
    0.09
    åŃIJéĩĮ
    0.09
    åİ»åģļ
    0.09
    conti
    0.08
    åıijå±ķ模å¼ı
    0.08
    quette
    0.08
    clud
    0.08
    ipples
    0.08
    Activations Density 0.006%

    No Known Activations