© 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. 39134
    Prev
    Next
    INDEX
    Explanations

    prefix of technical terms

    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
    ymb
    -0.10
    ippy
    -0.10
    ายà¸Ļ
    -0.09
    æĶ¿åĬ¡
    -0.09
    ORM
    -0.09
    istine
    -0.09
    iram
    -0.09
    çłļ
    -0.09
    iple
    -0.09
    _rt
    -0.09
    POSITIVE LOGITS
    æ¯ıæĹ¥
    0.10
     ST
    0.08
     cogn
    0.08
     daily
    0.08
    Ñįк
    0.08
    èģĮ
    0.08
    ä¸Ģè½®
    0.08
    ÑģÑĤв
    0.08
    ä½Ĩæĺ¯ä»ĸ
    0.08
    crew
    0.07
    Activations Density 0.004%

    No Known Activations