© 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. 16365
    Prev
    Next
    INDEX
    Explanations

    Self-poisoning

    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
    éĿ¶
    -0.21
    æĸĻ
    -0.20
    ANNEL
    -0.19
    plat
    -0.19
    ilestone
    -0.19
    ģ
    -0.18
    å¿Ļ
    -0.18
    åģıå·®
    -0.18
    arget
    -0.18
    CU
    -0.18
    POSITIVE LOGITS
    ãģ¤ãģij
    0.22
    ozy
    0.22
    è¿ĺ羣
    0.21
    è¿ĺ羣æĺ¯
    0.21
    åħ¬å¯ĵ
    0.19
    æľīèĩªå·±çļĦ
    0.19
    享çĶ¨
    0.18
    计è¾ĥ
    0.18
    è¯ĹæĦı
    0.18
     Apartment
    0.18
    Activations Density 0.003%

    No Known Activations