© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 34060
    Prev
    Next
    INDEX
    Explanations

    genuine or genuinely

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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
    Trail
    -0.09
    éĹ´æİ¥
    -0.09
    Liv
    -0.09
     öl
    -0.09
    è¡İ
    -0.08
     Ginger
    -0.08
     Tic
    -0.08
    bundle
    -0.08
    å¦Ļ
    -0.08
    оÑĢа
    -0.08
    POSITIVE LOGITS
     genuinely
    0.18
     genuine
    0.18
    羣è¯ļ
    0.17
    羣å¿ĥ
    0.14
    羣å®ŀçļĦ
    0.14
    羣æĥħ
    0.13
     Genuine
    0.13
     legitimately
    0.13
    èªł
    0.12
    ç¡®å®ŀæĺ¯
    0.12
    Activations Density 0.032%

    No Known Activations