© 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. 13282
    Prev
    Next
    INDEX
    Explanations

    déjÃł vu, fraction, way, screen

    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
     lou
    -0.22
    isters
    -0.19
    thon
    -0.19
     unm
    -0.19
    âĹ»
    -0.19
    éħį
    -0.18
    asca
    -0.18
     Aleppo
    -0.18
    黣
    -0.18
    é¢ĦåijĬ
    -0.17
    POSITIVE LOGITS
    è¿Ļ个åIJįåŃĹ
    0.31
    è¿Ļ个è¯į
    0.26
    åIJį称
    0.25
     name
    0.25
    ä»Ģä¹ĪæĦıæĢĿ
    0.25
    æĦıæĢĿæĺ¯
    0.23
    åIJįåŃĹ
    0.22
    name
    0.21
     nome
    0.20
    åĸĨ
    0.20
    Activations Density 0.165%

    No Known Activations