© 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. 53744
    Prev
    Next
    INDEX
    Explanations

    tremendous impact or opportunity

    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
    athon
    -0.09
    -fashioned
    -0.08
    ward
    -0.08
    oot
    -0.08
    thon
    -0.08
     Hilton
    -0.08
    ift
    -0.08
    spots
    -0.08
    itudes
    -0.08
     Booth
    -0.07
    POSITIVE LOGITS
    naire
    0.14
    æĦıä¹īä¸Ĭ
    0.10
     amounts
    0.10
    .metamodel
    0.10
    elves
    0.10
    orry
    0.10
     занÑıÑĤ
    0.10
     обÑĢазом
    0.10
    หมาย
    0.10
    ely
    0.09
    Activations Density 0.006%

    No Known Activations