© 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. 14723
    Prev
    Next
    INDEX
    Explanations

    isotope analysis

    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
    æĶĺ
    -0.24
    OLOR
    -0.21
    classname
    -0.20
    зол
    -0.20
    æĶ¶çº³
    -0.20
    'class
    -0.19
    oucher
    -0.19
    ä¿ĿèįIJ
    -0.19
    ä¸ŀ
    -0.19
    è¿«ä¸įåıĬ
    -0.19
    POSITIVE LOGITS
     ratios
    0.25
    trace
    0.24
    æłĩè®°
    0.23
    æ¯Ķçİĩ
    0.23
     trace
    0.22
     ratio
    0.21
     tracing
    0.20
    追溯
    0.20
     Hut
    0.20
     isot
    0.20
    Activations Density 0.121%

    No Known Activations