© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 15248
    Prev
    Next
    INDEX
    Explanations

    then followed by action

    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
    inges
    -0.21
     certain
    -0.20
    带ç»Ļ
    -0.20
    èŀŃ
    -0.20
    常çĶ¨çļĦ
    -0.19
    æ¯Ķè¾ĥå¤ļ
    -0.19
    lis
    -0.18
    .rpm
    -0.18
    äºĨåIJ§
    -0.18
    ç»ıéªĮåĴĮ
    -0.18
    POSITIVE LOGITS
    others
    0.23
    两项
    0.23
     thereafter
    0.21
    åIJĮäºĭ们
    0.19
    åľ°ä¸Ń
    0.19
     заÑĤем
    0.19
    ante
    0.19
    两次
    0.19
     té
    0.19
    ANTE
    0.19
    Activations Density 0.013%

    No Known Activations