© 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. 13464
    Prev
    Next
    INDEX
    Explanations

    politeness and social graces

    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
    çĿĽ
    -0.21
    æµij身
    -0.19
    osto
    -0.18
    éĶ»çĤ¼
    -0.18
    æĪ®
    -0.18
    æ¿Ģæĥħ
    -0.18
    æŀľæłij
    -0.17
    幸
    -0.17
    ecs
    -0.17
    osis
    -0.17
    POSITIVE LOGITS
    大æĸ¹
    0.26
    礼è²Į
    0.25
    ä¼ĺéĽħ
    0.25
    ç»ħ
    0.23
     manner
    0.23
    åħ¸éĽħ
    0.22
    彬
    0.22
    礼仪
    0.21
     elegant
    0.20
     dign
    0.20
    Activations Density 0.031%

    No Known Activations