© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 15492
    Prev
    Next
    INDEX
    Explanations

    maintain, preserve, protect, continue

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    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
    GLOBALS
    -0.19
    åįķ车
    -0.19
    /bin
    -0.19
    ensored
    -0.19
    och
    -0.18
     smear
    -0.18
    åħ¨å¥Ĺ
    -0.17
    adata
    -0.17
    å®ŀå®ŀåľ¨åľ¨
    -0.17
    å¼ķçĶ¨
    -0.17
    POSITIVE LOGITS
    ç»´æĬ¤
    0.28
    ç»´æĮģ
    0.26
    ä¿Ŀ
    0.25
    ä¿Ŀä½ı
    0.25
     preservation
    0.23
     ìľłì§Ģ
    0.22
    ä¿Ŀåħ»
    0.22
     Preserve
    0.21
    ä½ıäºĨ
    0.21
    ç»´
    0.21
    Activations Density 0.068%

    No Known Activations