© 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. 14310
    Prev
    Next
    INDEX
    Explanations

    describing qualities

    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
    亲çĪ±
    -0.22
    æĹłç¼ĺ
    -0.21
    主线
    -0.19
    ç£IJ
    -0.19
    Synopsis
    -0.19
    主æµģ
    -0.19
    anine
    -0.19
    jos
    -0.19
    彩èĻ¹
    -0.19
    ç»´å°Ķ
    -0.18
    POSITIVE LOGITS
     (?)
    0.24
    ly
    0.20
     but
    0.19
    lux
    0.19
    ALLY
    0.18
    à¸Ļาม
    0.18
    çĤ¸
    0.18
     cứ
    0.18
    ably
    0.18
    ally
    0.18
    Activations Density 0.163%

    No Known Activations