© 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. 15480
    Prev
    Next
    INDEX
    Explanations

    expressed, offered, used, provided

    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
    说æĪij
    -0.22
    æŃĮ声
    -0.20
    ä¿¡æģ¯åıijå¸ĥ
    -0.18
     gì
    -0.17
    å§ĭ建
    -0.17
    åľ©
    -0.17
     mediator
    -0.17
    дв
    -0.17
    æĭĶ
    -0.17
    ish
    -0.16
    POSITIVE LOGITS
    èĢħ
    0.21
    ucz
    0.19
    çļĦæĥħåĨµ
    0.19
    å±ŀå®ŀ
    0.18
    0.18
    leck
    0.18
    edin
    0.18
    çļĦä¸Ģ
    0.17
    à¹ĥà¸Ļà¸ģาร
    0.17
    urations
    0.17
    Activations Density 0.188%

    No Known Activations