© 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. 12053
    Prev
    Next
    INDEX
    Explanations

    replace with instruction

    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
    主åĬ¨æĢ§
    -0.10
    ä¸į足以
    -0.10
    æİĬ
    -0.10
    åĽ¾æĸĩ
    -0.09
    .Library
    -0.09
    isko
    -0.09
    æĿ¾
    -0.09
    agged
    -0.08
     Hib
    -0.08
    üs
    -0.08
    POSITIVE LOGITS
    æį¢æĪIJ
    0.12
    éĢĤ
    0.11
    æĶ¹æĪIJ
    0.11
     actual
    0.10
    æĽ¿
    0.10
    æĽ¿æį¢
    0.09
    éģ©
    0.09
    å®ŀéĻħ
    0.09
    ä½łèĩªå·±
    0.09
    (actual
    0.08
    Activations Density 0.236%

    No Known Activations