© 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. 12180
    Prev
    Next
    INDEX
    Explanations

    water stains and damage

    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
    ysical
    -0.20
    çijĽ
    -0.20
    .generate
    -0.20
     Chop
    -0.19
    èĢĮåĩº
    -0.19
    çļĦä¸ĢåĪĩ
    -0.19
     Closed
    -0.18
    .flat
    -0.18
    éĸ
    -0.18
    osl
    -0.18
    POSITIVE LOGITS
     spots
    0.27
    spots
    0.27
    æ¸į
    0.27
    éħĴç²¾
    0.27
    æĸij
    0.25
    æ¶Ĥ
    0.25
    浸
    0.25
    æ¸ĹéĢı
    0.25
    涴
    0.23
     spot
    0.23
    Activations Density 0.117%

    No Known Activations