© 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. 63215
    Prev
    Next
    INDEX
    Explanations

    we have or are

    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
     himself
    -0.14
    è·ŁæĪij说
    -0.13
    æĪijåİ»
    -0.13
    åĴ±ä»¬
    -0.12
     itself
    -0.12
    ä»ĸèĩªå·±
    -0.12
     us
    -0.12
    åĴ±
    -0.12
    åĴĮä»ĸçļĦ
    -0.11
    æľ¬äºº
    -0.11
    POSITIVE LOGITS
     ourselves
    0.24
    eping
    0.13
     selves
    0.12
    можем
    0.12
    ëģ¼
    0.12
     Ñģами
    0.11
    éĥ½çŁ¥éģĵ
    0.11
    åĽłä¸ºæĪij们
    0.11
    å½¼æѤ
    0.11
     avons
    0.11
    Activations Density 1.381%

    No Known Activations