© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7116
    Prev
    Next
    INDEX
    Explanations

    progress or completion

    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
    _LAT
    -0.10
    íħIJ
    -0.09
    æŃ£å¸¸ä½¿çĶ¨
    -0.09
    lean
    -0.09
    leh
    -0.09
    æĬ¢
    -0.09
    BUR
    -0.09
     phá»iji
    -0.09
    åīįæıIJæĺ¯
    -0.08
    æĪijåİ»
    -0.08
    POSITIVE LOGITS
    è¿Ľåº¦
    0.12
    ä¸Ģ段
    0.10
    é¢Ĩ
    0.10
    éłĺ
    0.10
     progress
    0.10
    åıªåī©ä¸ĭ
    0.10
    èĿĮ
    0.09
    缮åīįå·²ç»ı
    0.09
     part
    0.09
    é¢Ĩä¼ļ
    0.09
    Activations Density 0.111%

    No Known Activations