© 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. 4991
    Prev
    Next
    INDEX
    Explanations

    names and titles

    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
    nya
    -0.12
    shire
    -0.10
    leri
    -0.10
    à¹Į
    -0.10
    ski
    -0.10
    gram
    -0.10
    èĮ´
    -0.09
    mente
    -0.09
    gn
    -0.09
    αι
    -0.09
    POSITIVE LOGITS
    Ùij
    0.12
    0.11
    èĩ¾
    0.11
    ht
    0.10
    aters
    0.10
    resh
    0.09
    è¾Ī
    0.09
    ehler
    0.09
    asu
    0.09
    WARD
    0.09
    Activations Density 0.227%

    No Known Activations