© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19765
    Prev
    Next
    INDEX
    Explanations

    contrast words

    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
    ä»»
    -0.11
    欢
    -0.10
    èĥĮåıĽ
    -0.09
    诬
    -0.09
    æ¼ıæ´ŀ
    -0.08
    çĪ±åIJĥ
    -0.08
    æĺ±
    -0.08
    尤为
    -0.08
    è¯ĺ
    -0.08
     Country
    -0.08
    POSITIVE LOGITS
    ä½Ĩæĺ¯
    0.20
    ä½Ĩæĺ¯åľ¨
    0.19
    ä½Ĩ
    0.19
    ä½Ĩåį´
    0.19
     but
    0.18
    ä½Ĩæĺ¯å¦Ĥæŀľ
    0.18
    ä½Ĩå¦Ĥæŀľ
    0.17
    ä½Ĩæĺ¯å¯¹äºİ
    0.17
    ä½Ĩåľ¨
    0.17
    åı¯æĺ¯
    0.17
    Activations Density 0.317%

    No Known Activations