© 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. 62555
    Prev
    Next
    INDEX
    Explanations

    TV network names

    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
    åĸĦæĦı
    -0.09
    头æĿ¡
    -0.09
    strap
    -0.09
    ä¸ĵ线
    -0.09
    å°ij女
    -0.09
    AST
    -0.09
    subpackage
    -0.08
    åħ¬åĬ¡
    -0.08
    æ³ī
    -0.08
    atz
    -0.08
    POSITIVE LOGITS
    çĺ¦
    0.10
    ä¸įæĿ¥
    0.09
    è£Ļ
    0.09
    æĪIJ绩
    0.08
    宿
    0.08
    æĭ±
    0.08
     needle
    0.08
    çĸ¸
    0.08
    isms
    0.08
    åIJIJ
    0.08
    Activations Density 0.011%

    No Known Activations