© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 847
    Prev
    Next
    INDEX
    Explanations

    Names like Jill, Erin, and Joan

    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
    Negative Logits
    oot
    -0.10
    äºĭåIJİ
    -0.09
    essed
    -0.09
    entine
    -0.08
     Ñģобой
    -0.08
     knots
    -0.08
    /Framework
    -0.08
    ÑģÑĤан
    -0.08
     вÑģем
    -0.08
    itos
    -0.08
    POSITIVE LOGITS
    ingly
    0.15
    fully
    0.11
    rary
    0.10
    fulness
    0.10
    ful
    0.10
    abilities
    0.10
    sst
    0.09
    wick
    0.09
     gái
    0.09
    RAP
    0.09
    Activations Density 0.003%

    No Known Activations