© 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. 56169
    Prev
    Next
    INDEX
    Explanations

    prepare for or preparing and

    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
    uate
    -0.10
    -0.10
    Ń
    -0.09
    æĬĵ好
    -0.09
    ry
    -0.08
    漪
    -0.08
    oser
    -0.08
    ways
    -0.08
    佬
    -0.08
    trap
    -0.08
    POSITIVE LOGITS
    ForSegue
    0.13
    centage
    0.11
    liament
    0.10
    ми
    0.10
    ÑĪив
    0.09
    à¸ķà¸Ļ
    0.09
    WindowState
    0.09
    ctime
    0.09
    quoi
    0.09
    SENT
    0.09
    Activations Density 0.029%

    No Known Activations