© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-8B
    5. 18-RESID-BATCHTOPK-65K__L0-80
    7. 3065
    Prev
    Next
    INDEX
    Explanations

    crucial to understanding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-8b-saes/saes_Qwen_Qwen3-8B_batch_top_k/resid_post_layer_18
    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
     duly
    -0.23
     mostly
    -0.23
     vigorously
    -0.22
     bitterly
    -0.22
     crazy
    -0.22
     freely
    -0.22
     leaked
    -0.22
    orted
    -0.22
     emptied
    -0.21
    ä¼¼çļĦ
    -0.21
    POSITIVE LOGITS
    ç¡®ä¿Ŀ
    0.34
     understanding
    0.32
    å®ŀçİ°
    0.31
    çIJĨ解åĴĮ
    0.30
    æݨåĬ¨
    0.28
    ä¿Ŀè¯ģ
    0.27
     ensure
    0.27
    æĪIJå°±
    0.27
    loh
    0.26
    ç«ŀäºīåĬĽ
    0.26
    Activations Density 0.165%

    No Known Activations