© 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. 14431
    Prev
    Next
    INDEX
    Explanations

    tapering, diminishing, cone shape

    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·³è·ĥ
    -0.20
    èµ·æĿ¥äºĨ
    -0.20
    蹦
    -0.20
    å¾Ĺèµ·
    -0.19
    漪
    -0.19
    çħ¨
    -0.19
    åİĨå¹´
    -0.18
    èµ·ä¹ī
    -0.18
    åİĨ
    -0.17
    fulness
    -0.17
    POSITIVE LOGITS
    ToPoint
    0.19
    åIJij社ä¼ļ
    0.19
    çĦ¶
    0.19
     endregion
    0.19
    缩å°ı
    0.18
     Ont
    0.18
     Apt
    0.18
    oes
    0.18
    uj
    0.18
    ä¿©
    0.18
    Activations Density 0.018%

    No Known Activations