© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51956
    Prev
    Next
    INDEX
    Explanations

    Mediterranean

    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
    fra
    -0.10
    esthetic
    -0.09
    ầm
    -0.09
    迤
    -0.08
    ãģĩ
    -0.08
    atura
    -0.08
    ä¸ļ
    -0.08
    ìľ¨
    -0.08
    çľ¼æ³ª
    -0.08
    ANCH
    -0.08
    POSITIVE LOGITS
    çĪ·
    0.11
    ä¸įå®ļ
    0.10
    headed
    0.10
    xec
    0.10
    æĥľ
    0.09
    éĢįéģ¥
    0.09
    azy
    0.09
    ä¸įåłª
    0.08
    iffs
    0.08
    æĥħ人
    0.08
    Activations Density 0.005%

    No Known Activations