© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 3061
    Prev
    Next
    INDEX
    Explanations

    starts with a symbol or specific word

    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
    åĬ¿
    -0.11
    ofil
    -0.09
    裸
    -0.08
    åĭ¢
    -0.08
    suffix
    -0.08
    uffix
    -0.08
    Suffix
    -0.08
    ç¨į
    -0.08
    UFFIX
    -0.08
    èIJ½æĪ·
    -0.08
    POSITIVE LOGITS
    å¼Ģ头
    0.10
    çļĦ年轻人
    0.09
    .Collectors
    0.09
    ä¸Ģæłĭ
    0.09
    åĨį说
    0.09
    è¿Ļçīĩ
    0.09
    åı¯è§ģ
    0.08
    ç»Ħå§Ķä¼ļ
    0.08
    æĢľ
    0.08
    头
    0.08
    Activations Density 0.131%

    No Known Activations