© 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. 18985
    Prev
    Next
    INDEX
    Explanations

    do it

    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
    人士
    -0.10
    åķ·
    -0.10
    atan
    -0.10
    äºĭçī©
    -0.10
    tran
    -0.09
    çļĦåij½è¿IJ
    -0.09
    forth
    -0.09
    hips
    -0.09
    ivol
    -0.08
    ries
    -0.08
    POSITIVE LOGITS
    utt
    0.10
    regor
    0.09
    á»ħ
    0.09
    è¿Ŀç«ł
    0.08
    éĿł
    0.08
    å¢ŀåĬłåĢ¼
    0.08
    ÙĪا
    0.08
    unding
    0.08
     Samar
    0.08
    xygen
    0.08
    Activations Density 0.037%

    No Known Activations