© 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. 27825
    Prev
    Next
    INDEX
    Explanations

    J.P. and JP abbreviations

    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
     Watkins
    -0.10
    ledged
    -0.10
    ptune
    -0.09
    æ·®åįĹ
    -0.08
    OUNT
    -0.08
    лем
    -0.08
    баÑĢ
    -0.08
     opener
    -0.08
    usiness
    -0.08
    æ½´
    -0.08
    POSITIVE LOGITS
    зо
    0.10
    ä»Ģä¹Īæł·
    0.09
    çĩĬ
    0.09
    ฯ
    0.09
    éĤ¸
    0.09
    terior
    0.09
    æĪı
    0.08
    territ
    0.08
    ZR
    0.08
    æıı
    0.08
    Activations Density 0.004%

    No Known Activations