© 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. 17463
    Prev
    Next
    INDEX
    Explanations

    typical noun

    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
    å¡ĺ
    -0.09
    manship
    -0.09
    ¼
    -0.09
    empl
    -0.08
    åŀł
    -0.08
    edImage
    -0.08
    APE
    -0.08
    ese
    -0.08
    imed
    -0.08
    tá
    -0.08
    POSITIVE LOGITS
    ity
    0.15
     xuyên
    0.15
    åħ¸åŀĭçļĦ
    0.12
    åħ¸åŀĭ
    0.12
    ITY
    0.11
    æĥħåĨµä¸ĭ
    0.11
    -sized
    0.10
    ahead
    0.10
    çļĦåģļæ³ķ
    0.10
    åģļæ³ķ
    0.10
    Activations Density 0.057%

    No Known Activations