© 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. 15430
    Prev
    Next
    INDEX
    Explanations

    linking words with prior information

    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
    è¯Ħå§Ķ
    -0.21
    ä¸ŃåĽ½çī¹èī²
    -0.20
    å°ļ书
    -0.20
    å¿ĥæĦı
    -0.19
    admins
    -0.19
    楼梯
    -0.19
    éĺ¶æ¢¯
    -0.19
    æģ¯æģ¯
    -0.19
    ìĤ´
    -0.18
    éªĸ
    -0.18
    POSITIVE LOGITS
    æīįçŁ¥éģĵ
    0.25
     prior
    0.23
     previous
    0.23
    ä¸Ĭä¸ĭ游
    0.21
     previously
    0.21
    æµģæ°´
    0.21
    èµĦæºIJä¼ĺåĬ¿
    0.20
    渺
    0.20
     data
    0.20
    缸åħ³ä¿¡æģ¯
    0.19
    Activations Density 0.044%

    No Known Activations