© 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. 5009
    Prev
    Next
    INDEX
    Explanations

    punctuation followed by specific 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
    管å±Ģ
    -0.10
    GNU
    -0.09
    æĽ´å¥½çļĦ
    -0.08
    çĦī
    -0.08
    æĺ¯éĿŀ
    -0.08
    ç¡®
    -0.08
    大è¡Ĺ
    -0.08
    æĺ¯ä¸Ģ个
    -0.08
    ÙĨØ©
    -0.08
    æĸ°ä¸Ģè½®
    -0.08
    POSITIVE LOGITS
    æĸ°éĹ»åıijå¸ĥ
    0.10
    ness
    0.09
    -looking
    0.09
    ãģĮãĤī
    0.09
    èĬľ
    0.08
    æ¡Ī
    0.08
    å¨ĥå¨ĥ
    0.08
    -than
    0.08
     evidence
    0.08
    è¦ģç´§
    0.08
    Activations Density 0.217%

    No Known Activations