© 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. 58542
    Prev
    Next
    INDEX
    Explanations

    MIT license or Mozilla Public License

    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
    åĨķ
    -0.11
     manner
    -0.11
    mark
    -0.11
    صر
    -0.10
    om
    -0.10
    Ùħا
    -0.10
    itto
    -0.10
    marks
    -0.10
    IJ
    -0.10
    ĺ
    -0.09
    POSITIVE LOGITS
     Mondays
    0.17
    ellaneous
    0.17
    intosh
    0.16
     Madness
    0.16
    ioned
    0.16
    IMUM
    0.15
    riad
    0.15
     mắn
    0.15
     mẽ
    0.15
    Carthy
    0.14
    Activations Density 3.205%

    No Known Activations