© 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. 15349
    Prev
    Next
    INDEX
    Explanations

    describing shape or state

    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
    æķ¬
    -0.21
    âīĤ
    -0.21
    çĤ·
    -0.20
     program
    -0.20
    éĨª
    -0.20
    ä¸įåĨį
    -0.19
    á´¼
    -0.19
    交
    -0.19
     resign
    -0.19
    åıłåĬł
    -0.19
    POSITIVE LOGITS
    辨
    0.24
    æ¨¡æł·
    0.23
    æŀĦéĢł
    0.21
    å»ĵ
    0.21
    append
    0.21
     distinctive
    0.21
    æł·åŃIJ
    0.20
    extends
    0.20
    eus
    0.19
    读åIJİ
    0.19
    Activations Density 0.066%

    No Known Activations