© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 17243
    Prev
    Next
    INDEX
    Explanations

    making things worse

    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
    Negative Logits
    çłĶåıijä¸Ńå¿ĥ
    -0.21
    åĨ³å®ļäºĨ
    -0.19
    èģĶåIJĪ
    -0.18
    ois
    -0.18
    oise
    -0.18
    ä¸įæĸŃå¢ŀ强
    -0.18
     undermin
    -0.18
     Pis
    -0.18
    ression
    -0.18
    pis
    -0.18
    POSITIVE LOGITS
    åİŁæľīçļĦ
    0.30
    åİŁæľī
    0.29
    åıįèĢĮ
    0.26
    åİŁæĿ¥çļĦ
    0.25
    æĸ°çļĦ
    0.25
    è¿ĺæĺ¯ä¼ļ
    0.24
     ìĺ¤íŀĪ볤
    0.23
    缸åıį
    0.23
     worse
    0.21
    åĬłåī§
    0.20
    Activations Density 0.071%

    No Known Activations