© 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. 47697
    Prev
    Next
    INDEX
    Explanations

    Real Madrid

    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
    yard
    -0.10
    kills
    -0.08
    rong
    -0.08
    ellow
    -0.08
    æ¸ħé¦Ļ
    -0.08
    orton
    -0.08
    irth
    -0.08
    yla
    -0.08
    lem
    -0.08
    Master
    -0.08
    POSITIVE LOGITS
    ../../../
    0.11
    å£ħ
    0.10
    fully
    0.10
    iness
    0.09
    æ£Ĵ
    0.09
    .ham
    0.09
    Äĥng
    0.09
    mapped
    0.09
    коÑģ
    0.09
    zeug
    0.09
    Activations Density 0.008%

    No Known Activations