© 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. 49053
    Prev
    Next
    INDEX
    Explanations

    more and then comparative adjectives

    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
    使ä¹ĭ
    -0.10
    ity
    -0.10
    çĭ¸
    -0.10
    é¢Į
    -0.09
    æĽ´å¤ļçļĦ
    -0.09
    æĭĤ
    -0.09
    精度
    -0.09
    .Box
    -0.08
    èĪĮå°ĸ
    -0.08
    ä½ļ
    -0.08
    POSITIVE LOGITS
    itz
    0.10
    uario
    0.10
    è¡°
    0.09
    èĩªçĦ¶èĢĮ
    0.09
    iversary
    0.09
    èij©
    0.09
     Zac
    0.09
     RU
    0.08
     diam
    0.08
    åĪĨæĪIJ
    0.08
    Activations Density 0.216%

    No Known Activations