© 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. 36970
    Prev
    Next
    INDEX
    Explanations

    penalties and consequences

    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
    æ¯ĴæĢ§
    -0.10
    æľīæ¯Ĵ
    -0.08
    enth
    -0.08
     оÑĢг
    -0.08
    ué
    -0.08
     frustr
    -0.08
    slt
    -0.08
    è¾Ĺ
    -0.08
    mods
    -0.08
    缼
    -0.08
    POSITIVE LOGITS
    ç½ļ
    0.13
    æĬ¹
    0.12
    æĥ©ç½ļ
    0.11
    ç½°
    0.10
    åĩºå±Ģ
    0.10
    大å¦Ī
    0.10
    ä¸Ģé¡¿
    0.10
    éĻ¢éķ¿
    0.10
    读åIJİ
    0.10
    ä¸Ģ个æĺŁæľŁ
    0.09
    Activations Density 0.074%

    No Known Activations