© 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. 23307
    Prev
    Next
    INDEX
    Explanations

    add until level

    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
    åij±
    -0.09
    讳
    -0.09
    èIJ½æĪ·
    -0.08
    åĪĹ表
    -0.08
     Gund
    -0.08
    åĸ·
    -0.08
    å¾Ī满æĦı
    -0.08
    纵深
    -0.08
    æĭ¥
    -0.08
    ovich
    -0.08
    POSITIVE LOGITS
    è°ĥæķ´
    0.11
    調æķ´
    0.10
    è°ĥèĬĤ
    0.09
     adjust
    0.09
    -adjust
    0.09
    ester
    0.09
     Adjustment
    0.08
     adjustment
    0.08
     ajust
    0.08
    Adjust
    0.08
    Activations Density 0.013%

    No Known Activations