© 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. 27654
    Prev
    Next
    INDEX
    Explanations

    McKinney, McKinley, MacKenzie, McKenna

    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
    ities
    -0.10
    stead
    -0.09
     predecessor
    -0.09
    stick
    -0.09
    ision
    -0.08
    å²ĺ
    -0.08
    chestra
    -0.08
    ways
    -0.08
    oral
    -0.08
    èĺ¸
    -0.08
    POSITIVE LOGITS
    ney
    0.12
    upt
    0.11
    ä¸Ģçľĭ
    0.10
    otre
    0.10
    inker
    0.09
    جا
    0.09
    clusive
    0.09
    floating
    0.09
    ¤
    0.09
    ugar
    0.09
    Activations Density 0.029%

    No Known Activations