© 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. 63463
    Prev
    Next
    INDEX
    Explanations

    attachment type or using attachments

    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
    ally
    -0.09
    ä¸įå®ļ
    -0.08
    ä¸įæŃ»
    -0.08
    åįĥ
    -0.08
     Ones
    -0.08
    ewriter
    -0.08
    odem
    -0.08
     antib
    -0.08
    uck
    -0.08
    markers
    -0.08
    POSITIVE LOGITS
    人家
    0.11
    åįļ主
    0.09
    tern
    0.09
    梳
    0.09
    aje
    0.09
    æĪŁ
    0.08
    æľ¬ç½ijç«Ļ
    0.08
    æ®Ĭ
    0.08
    ToLocal
    0.08
    çĸ¡
    0.08
    Activations Density 0.010%

    No Known Activations