© 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. 8138
    Prev
    Next
    INDEX
    Explanations

    add, include, provided

    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
    ç¾
    -0.10
    ä¸ĢæŃ¥æŃ¥
    -0.09
    ë°
    -0.09
    çıłä¸īè§Ĵ
    -0.09
    xCA
    -0.09
    æľīèī²éĩijå±ŀ
    -0.09
    .unknown
    -0.08
     Rew
    -0.08
    å±ķå¼Ģ
    -0.08
    æĭīçĿĢ
    -0.08
    POSITIVE LOGITS
     inclusion
    0.14
    纳åħ¥
    0.13
     included
    0.13
    åħ¥ãĤĮ
    0.13
    åĬłåħ¥
    0.13
     include
    0.12
     Include
    0.12
    åĪĹåħ¥
    0.12
    è¿Ľåİ»
    0.11
    /include
    0.11
    Activations Density 0.247%

    No Known Activations