© 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. 32103
    Prev
    Next
    INDEX
    Explanations

    elegant descriptors

    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
    itive
    -0.11
    åŁ
    -0.09
    ame
    -0.09
    amp
    -0.09
    ammen
    -0.09
    íģ¬
    -0.09
    èĩ¾
    -0.09
    çĶŁäº§æĢ»åĢ¼
    -0.09
    æ´¼
    -0.08
    oj
    -0.08
    POSITIVE LOGITS
    udies
    0.10
    é¦ĸ饰
    0.10
    转身
    0.10
    ury
    0.10
    anus
    0.10
    rics
    0.09
    ude
    0.09
    ulous
    0.09
    icians
    0.09
    -minded
    0.09
    Activations Density 0.008%

    No Known Activations