© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 15222
    Prev
    Next
    INDEX
    Explanations

    accuracy or inaccuracy

    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
    åįłæį®äºĨ
    -0.20
    ä½»
    -0.19
    åįłæį®
    -0.19
     Seat
    -0.18
    大人
    -0.18
    æĪIJ人
    -0.18
    åIJĮåŁİ
    -0.18
    çŀ»
    -0.17
     Philipp
    -0.17
    ä½Ķ
    -0.17
    POSITIVE LOGITS
    æݪ
    0.25
    è°¬
    0.23
    èĪĽ
    0.21
    ä¹Łæľīä¸ĢäºĽ
    0.21
    accuracy
    0.21
     flaw
    0.19
    表述
    0.19
    éĶĻ误
    0.19
    åĩĨç¡®æĢ§
    0.19
    ä¿®æŃ£
    0.19
    Activations Density 0.080%

    No Known Activations