© 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. 30196
    Prev
    Next
    INDEX
    Explanations

    premise

    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
    inn
    -0.11
    yst
    -0.10
    meal
    -0.10
    åIJĮæľŁ
    -0.10
    ered
    -0.09
    thon
    -0.09
    oku
    -0.09
    qing
    -0.09
    icians
    -0.09
    itian
    -0.09
    POSITIVE LOGITS
    å·²ä¹ħçļĦ
    0.11
    и
    0.10
     Ñģобой
    0.10
    esa
    0.10
    ÙĬدÙĬ
    0.10
    CISION
    0.09
    blem
    0.09
    ned
    0.08
    é«ĺ涨
    0.08
    etheus
    0.08
    Activations Density 0.021%

    No Known Activations