© 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. 27634
    Prev
    Next
    INDEX
    Explanations

    hank, shank, ankh

    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
    ally
    -0.16
    teenth
    -0.15
    ../../../
    -0.12
    phans
    -0.12
    teen
    -0.12
    rens
    -0.10
    aceous
    -0.10
    -0.10
    icity
    -0.09
    .amazon
    -0.09
    POSITIVE LOGITS
    weed
    0.10
    inous
    0.09
    sgiving
    0.09
    ungan
    0.09
    StringBuilder
    0.09
    yy
    0.09
    IJ
    0.09
    éļĨ
    0.08
    溪
    0.08
    ús
    0.08
    Activations Density 0.022%

    No Known Activations