© 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. 33750
    Prev
    Next
    INDEX
    Explanations

    wise and wisdom

    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
    ãģĩ
    -0.10
    èµĸ以
    -0.09
    oder
    -0.08
    tered
    -0.08
    à¹Į
    -0.08
    iers
    -0.08
    ter
    -0.08
    weet
    -0.08
    ils
    -0.08
    amar
    -0.08
    POSITIVE LOGITS
    fully
    0.14
    acre
    0.13
    0.11
    -looking
    0.11
    NESS
    0.10
    ä¹ĭ举
    0.10
    ë¡Ń
    0.10
    fulness
    0.10
     Ñĥзн
    0.10
     جدا
    0.09
    Activations Density 0.009%

    No Known Activations