© 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. 10828
    Prev
    Next
    INDEX
    Explanations

    be separated into

    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
    ãģĽãģĦ
    -0.21
     maybe
    -0.20
     finally
    -0.20
     somehow
    -0.19
     fault
    -0.19
     again
    -0.19
    åıĪæĺ¯
    -0.18
    .ax
    -0.18
    èĥĮ
    -0.18
     probably
    -0.18
    POSITIVE LOGITS
    æľīå¾Īå¤ļç§į
    0.32
    å¾Īå¤ļäºĭæĥħ
    0.31
    æľī许å¤ļ
    0.31
    éĿŀ常å¤ļçļĦ
    0.27
    許å¤ļ
    0.25
    ä¹Łæľīå¾Īå¤ļ
    0.25
    å¾Īéĩįè¦ģçļĦ
    0.25
    ãģĦãģıãģ¤ãģĭ
    0.24
    å¾Īæĸ¹ä¾¿
    0.24
    ãģķãģ¾ãģĸãģ¾ãģª
    0.24
    Activations Density 0.939%

    No Known Activations