© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52830
    Prev
    Next
    INDEX
    Explanations

    virulent, virality, virgins, virile

    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
    ики
    -0.10
    ãģĮãĤī
    -0.10
    emed
    -0.09
    è¯Ŀè¯ŃæĿĥ
    -0.09
    è¯Ŀ说
    -0.09
    lint
    -0.09
    è¾ĩ
    -0.09
    emat
    -0.09
     HOLDER
    -0.09
    lse
    -0.09
    POSITIVE LOGITS
    aceous
    0.09
    ulence
    0.09
    iting
    0.09
    ulent
    0.08
    ç«ŀ
    0.08
    ty
    0.08
    enci
    0.08
    teenth
    0.08
    ä¸įåĩº
    0.08
    篱
    0.08
    Activations Density 0.020%

    No Known Activations