© 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. 23443
    Prev
    Next
    INDEX
    Explanations

    enter or giveaway contest

    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
     Cer
    -0.08
    çIJĨå·¥
    -0.08
     probes
    -0.08
    éļĶéŁ³
    -0.08
    UnderTest
    -0.08
    使åij½
    -0.08
     пÑĢож
    -0.08
    çĦ¯
    -0.08
    CTest
    -0.08
    çªģ
    -0.08
    POSITIVE LOGITS
    æĬ½å¥ĸ
    0.14
    æĬ½åıĸ
    0.13
    幸è¿IJ
    0.13
    æĬ½
    0.11
    éĴĿ
    0.10
     entr
    0.10
    èµ¢
    0.10
     giveaway
    0.10
    æĬ½åĩº
    0.10
    èİ·å¥ĸ
    0.10
    Activations Density 0.091%

    No Known Activations