© 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. 30887
    Prev
    Next
    INDEX
    Explanations

    followed by punctuation comparison

    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
    AGER
    -0.10
    ager
    -0.09
    stor
    -0.09
    sert
    -0.08
    ç£ħ
    -0.08
    åıĭæĥħ
    -0.08
    çĶ¨åľ°
    -0.08
    رÙĩ
    -0.08
    æĽ´å¤ļ精彩
    -0.08
    åı¦è¡Į
    -0.08
    POSITIVE LOGITS
     compared
    0.14
    ä¼ĺäºİ
    0.13
    ä¸įåIJĮäºİ
    0.13
    ä»ħ次äºİ
    0.12
    æ¯Ķèµ·
    0.12
     Compared
    0.11
    è¦ģæ¯Ķ
    0.11
    æ¯Ķ
    0.11
    æ¯ĶæĪij
    0.11
    åīįæīĢæľª
    0.10
    Activations Density 0.102%

    No Known Activations