© 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. 48349
    Prev
    Next
    INDEX
    Explanations

    San Diego

    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
    igg
    -0.11
    еж
    -0.09
    vely
    -0.09
    mind
    -0.09
    ValueCollection
    -0.09
    nels
    -0.08
    nel
    -0.08
    幸ç¦ıæĦŁ
    -0.08
    enden
    -0.08
    æŀģå°ij
    -0.08
    POSITIVE LOGITS
    wards
    0.11
    OrCreate
    0.09
    å·ŀ
    0.09
     summers
    0.08
     outskirts
    0.08
    æĴ¼
    0.08
    /out
    0.08
    explo
    0.08
    ovel
    0.08
    anter
    0.08
    Activations Density 0.002%

    No Known Activations