© 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. 16373
    Prev
    Next
    INDEX
    Explanations

    for the full

    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
    é«ĺ楼
    -0.24
    ä¸ĢæŃ¥ä¸ĢæŃ¥
    -0.20
    并äºİ
    -0.20
    ãģĻãģIJãģ«
    -0.19
    æĹĭ
    -0.19
    åĤ§
    -0.19
    好象
    -0.19
    -0.19
    好似
    -0.19
    iki
    -0.18
    POSITIVE LOGITS
    æĿ¥è¯´
    0.38
    çļĦ人æĿ¥è¯´
    0.34
    èĢĮè¨Ģ
    0.33
    çݩ家æĿ¥è¯´
    0.33
    æĿ¥è®²
    0.31
    ä¾Ĩ說
    0.29
    æĪĺ士æĿ¥è¯´
    0.26
     purposes
    0.26
     sake
    0.26
    unately
    0.25
    Activations Density 0.048%

    No Known Activations