© 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. 4124
    Prev
    Next
    INDEX
    Explanations

    access

    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
    亲
    -0.11
    åIJ¬åIJ¬
    -0.09
    åģļ大
    -0.08
    mente
    -0.08
    ì§Ī
    -0.08
     bald
    -0.08
    atch
    -0.08
    éķĺ
    -0.08
    hea
    -0.08
    éļ³
    -0.08
    POSITIVE LOGITS
    ç¼ī
    0.10
    ToEnd
    0.10
    PerPixel
    0.09
    ucher
    0.09
    ç³ĸå°¿
    0.09
    ãĤ¨ãĥ«
    0.09
     Guth
    0.09
    rpc
    0.08
    uits
    0.08
    她说
    0.08
    Activations Density 0.001%

    No Known Activations