© 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. 4219
    Prev
    Next
    INDEX
    Explanations

    radioactive substances

    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
    饱满
    -0.11
    横åIJij
    -0.09
    渲
    -0.09
    峪
    -0.09
    æľīçĽĬ
    -0.09
     Sentinel
    -0.09
    ä¸Ģ对ä¸Ģ
    -0.09
    éĻIJå®ļ
    -0.08
    ä¼łè¾¾
    -0.08
    èij£äºĭ
    -0.08
    POSITIVE LOGITS
    æĶ¾å°Ħ
    0.14
     radioactive
    0.11
     radi
    0.09
    è¡°
    0.09
     phóng
    0.09
     decay
    0.09
    éĴĭ
    0.09
    èĮ±
    0.09
    代è¨Ģ人
    0.09
    ulers
    0.09
    Activations Density 0.130%

    No Known Activations