© 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. 14379
    Prev
    Next
    INDEX
    Explanations

    superlatives and qualifiers

    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´æķ´
    -0.21
    ål
    -0.20
    UAGE
    -0.20
    uable
    -0.20
    uien
    -0.20
    ÑģÑĤвие
    -0.20
    ECH
    -0.18
    alaria
    -0.18
    -authored
    -0.18
    apel
    -0.18
    POSITIVE LOGITS
    ä¹Łä¸įæĺ¯
    0.35
    èĢĮæĺ¯
    0.30
     Äijâu
    0.29
     lắm
    0.28
     nor
    0.28
    ä¸įé«ĺ
    0.27
     necessarily
    0.23
    ä¸įæĺ¯å¾Ī
    0.22
    rocket
    0.21
    ä¹Łä¸įæĥ³
    0.20
    Activations Density 0.047%

    No Known Activations