© 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. 31909
    Prev
    Next
    INDEX
    Explanations

    tools and related software contexts

    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
    Doctors
    -0.09
     tas
    -0.09
    illac
    -0.09
    ikan
    -0.09
     Reconstruction
    -0.08
    ä¸Ģ个éĩįè¦ģ
    -0.08
    /her
    -0.08
    plans
    -0.08
    樵
    -0.08
    奴
    -0.08
    POSITIVE LOGITS
    kits
    0.15
    ç¨ļ
    0.12
    æĢ§å¼º
    0.12
    åĵĩ
    0.11
    chain
    0.11
    set
    0.10
    ç«ĭãģ¦
    0.10
    spot
    0.10
    anted
    0.10
    ç¾¤å²Ľ
    0.10
    Activations Density 0.054%

    No Known Activations