© 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. 49603
    Prev
    Next
    INDEX
    Explanations

    continents and oceans

    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
    åĥ®
    -0.09
    æĬ¬å¤´
    -0.09
    æĸ¥
    -0.09
    çĸ¾
    -0.08
    å°º
    -0.08
    å¤ĸç±į
    -0.08
     ÙħÙĨÙĩا
    -0.08
    ØŃاد
    -0.08
    ÑĦоÑĢ
    -0.08
    ãĤģ
    -0.08
    POSITIVE LOGITS
    太平æ´ĭ
    0.13
    æ´²
    0.13
     continents
    0.12
     continent
    0.12
    è¯
    0.10
    大æ´ĭ
    0.10
    大éĻĨ
    0.09
     planet
    0.09
    æĺŁçIJĥ
    0.09
    大éĻ¸
    0.09
    Activations Density 0.073%

    No Known Activations