© 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. 3501
    Prev
    Next
    INDEX
    Explanations

    home

    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
    çľ¼åīį
    -0.09
    iert
    -0.09
     lame
    -0.09
    æĽ·
    -0.09
    _sensitive
    -0.08
    VAL
    -0.08
     Cann
    -0.08
     lif
    -0.08
    ihan
    -0.08
    Looper
    -0.08
    POSITIVE LOGITS
    zel
    0.10
    åľ¨çĶŁæ´»ä¸Ń
    0.10
    inson
    0.09
    éĽį
    0.09
    å¤ļå¹´çļĦ
    0.08
    æĥħåĨµè¿Ľè¡Į
    0.08
    comed
    0.08
    ÑĢаг
    0.08
    éļª
    0.08
    ç®ĹäºĨ
    0.08
    Activations Density 0.032%

    No Known Activations