© 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. 12911
    Prev
    Next
    INDEX
    Explanations

    before leaving instructions

    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
    ä¸Ģè¾Ĩ
    -0.21
     STILL
    -0.19
    çªģå¦Ĥåħ¶æĿ¥
    -0.19
    æ¤Ĵ
    -0.19
    Shock
    -0.18
    .AutoSize
    -0.18
    æĦŁåĬ¨
    -0.18
    éķ¿å¤§
    -0.18
    头åıij
    -0.18
    ä¸Ģåı°
    -0.18
    POSITIVE LOGITS
     when
    0.31
    å½ĵä»ĸ
    0.28
    when
    0.28
     whenever
    0.27
    rather
    0.27
    å½ĵä½ł
    0.27
    èĢĮä¸įæĺ¯
    0.26
     before
    0.26
     after
    0.25
    	when
    0.24
    Activations Density 0.174%

    No Known Activations