© 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. 16972
    Prev
    Next
    INDEX
    Explanations

    article followed by text

    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
    å³Ļ
    -0.10
    etu
    -0.10
    obil
    -0.10
    oeff
    -0.10
    erten
    -0.09
    RelativeTo
    -0.09
    æĿ¼
    -0.09
    ãĤĮ
    -0.08
    ycz
    -0.08
    å·¥ç¨ĭè´¨éĩı
    -0.08
    POSITIVE LOGITS
    aceous
    0.09
    ç¾Ķ
    0.09
    apers
    0.09
    ize
    0.09
    umped
    0.08
    Äĥn
    0.08
    insk
    0.08
    orges
    0.08
    tte
    0.08
    inds
    0.08
    Activations Density 0.054%

    No Known Activations