© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36892
    Prev
    Next
    INDEX
    Explanations

    lost or losing

    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
    etting
    -0.10
    emic
    -0.10
    æ²¾
    -0.09
    /from
    -0.09
    西路
    -0.09
    åĤĴ
    -0.09
    æĪIJ绩
    -0.09
    rovers
    -0.09
    éŀħ
    -0.08
    LOWER
    -0.08
    POSITIVE LOGITS
    æĥ¨
    0.11
     عربÙĬ
    0.09
    ãģı
    0.09
    æĿĸ
    0.09
    åĴĮæľįåĬ¡
    0.08
    主人åħ¬
    0.08
    æĬ¤çħ§
    0.08
    è¯Ĵ
    0.08
    第ä¸Ģ书记
    0.08
    çĽĹ
    0.08
    Activations Density 0.018%

    No Known Activations