© 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. 10942
    Prev
    Next
    INDEX
    Explanations

    thousand, ten

    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
    ros
    -0.21
    çİ°æľī
    -0.19
    ç§ij
    -0.19
    (main
    -0.19
    .main
    -0.19
    èĩªæķij
    -0.18
    主
    -0.18
    èĩªå·±
    -0.18
    èĩªå®¶
    -0.18
     supers
    -0.17
    POSITIVE LOGITS
    VES
    0.20
    bens
    0.20
    ught
    0.19
    讼
    0.19
     tán
    0.18
    é¡·
    0.18
    æī«ä¸Ģ
    0.17
    -ves
    0.17
     Harbor
    0.17
    æģ¼
    0.17
    Activations Density 0.005%

    No Known Activations