© 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. 14408
    Prev
    Next
    INDEX
    Explanations

    'hidden', calories, considered, caffeine, less

    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
    á´¼
    -0.21
    ilingual
    -0.21
    alesce
    -0.21
    å·¡åĽŀ
    -0.21
    ledon
    -0.21
    æĻ·
    -0.20
    erp
    -0.20
    urname
    -0.19
    oppable
    -0.19
     Rad
    -0.18
    POSITIVE LOGITS
    æijĦåħ¥
    0.26
     calories
    0.24
    çľĭä½ł
    0.23
    ä¸ĭ车
    0.23
    æ·»åĬł
    0.22
     grams
    0.21
    added
    0.21
    вед
    0.21
    åIJ«æľī
    0.21
    èIJ¥åħ»ä»·åĢ¼
    0.20
    Activations Density 0.029%

    No Known Activations