© 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. 62747
    Prev
    Next
    INDEX
    Explanations

    categories and classification

    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
    ðŁı»
    -0.11
    代表大ä¼ļ
    -0.10
    æł·çļĦ
    -0.09
    ¼
    -0.09
    ðŁı¼
    -0.09
    åĬ¨çĶŁæĪIJ
    -0.09
    ly
    -0.09
    iable
    -0.09
    fully
    -0.09
    igkeit
    -0.09
    POSITIVE LOGITS
    åĨħçļĦ
    0.11
    ical
    0.11
    als
    0.10
    obra
    0.10
    人士
    0.10
     quo
    0.10
    lid
    0.10
    nement
    0.09
    ¡
    0.09
    cz
    0.09
    Activations Density 0.047%

    No Known Activations