© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20988
    Prev
    Next
    INDEX
    Explanations

    comm + suffixes

    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
    -0.11
    grim
    -0.11
    ½
    -0.11
    hood
    -0.11
    teen
    -0.10
    аÑĤелÑĮ
    -0.10
    lop
    -0.10
    裳
    -0.09
    ettes
    -0.09
    ternal
    -0.09
    POSITIVE LOGITS
    roach
    0.10
    unal
    0.10
    忤
    0.10
     Sund
    0.10
    èĩ¾
    0.09
    ihilation
    0.09
    ãģ¹ãģį
    0.09
    UNITY
    0.09
    preh
    0.09
    â»
    0.09
    Activations Density 0.035%

    No Known Activations