© 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. 16834
    Prev
    Next
    INDEX
    Explanations

    zone or terminal

    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
    åħ¬åı¸å°Ĩ
    -0.20
    åºIJ
    -0.19
    jas
    -0.19
    èĩ³å°Ĭ
    -0.18
    PROP
    -0.18
    å¿ĥè¡Ģ
    -0.18
    åģļçĶŁæĦı
    -0.18
    éĿ¢çĽ®
    -0.17
    æĨ¾
    -0.17
    广
    -0.17
    POSITIVE LOGITS
    ä¸ĵå±ŀ
    0.34
     riêng
    0.32
    ä¸ĵçĶ¨
    0.28
    ä¸ĵåľº
    0.27
    åįķçĭ¬
    0.26
    ä¸ĵåĮº
    0.24
    å°Ī
    0.24
    çĭ¬èĩª
    0.23
    åĪĨå¼Ģ
    0.23
     separate
    0.22
    Activations Density 0.262%

    No Known Activations