© 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. 11820
    Prev
    Next
    INDEX
    Explanations

    continue to increase

    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
    ç³Ĭ
    -0.21
    第ä¸Ģ天
    -0.20
    åĴĮ个人
    -0.20
    né
    -0.20
    love
    -0.19
    .hashCode
    -0.19
    -contrib
    -0.18
    ç´¯
    -0.18
    æ¶Ĥ
    -0.18
    ä»ħ为
    -0.18
    POSITIVE LOGITS
    继ç»Ńä¿ĿæĮģ
    0.31
    æľªæĿ¥çļĦ
    0.30
    继ç»Ń
    0.30
     future
    0.29
    åľ¨æľªæĿ¥
    0.27
    æľªæĿ¥
    0.27
    future
    0.26
     further
    0.25
    è¶ĭåĬ¿
    0.24
     далÑĮÑĪе
    0.24
    Activations Density 0.050%

    No Known Activations