© 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. 48685
    Prev
    Next
    INDEX
    Explanations

    Thomas followed by names and places

    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
    ERAL
    -0.11
    ture
    -0.10
    ÑĤÑı
    -0.10
    deen
    -0.09
    鸮
    -0.09
    asan
    -0.09
    ICODE
    -0.09
    opa
    -0.09
    DEFINED
    -0.09
    ķ
    -0.09
    POSITIVE LOGITS
    ãģĹãģ¾ãģĨ
    0.11
    icken
    0.10
    ains
    0.10
    ickey
    0.10
    .TryParse
    0.09
    ses
    0.09
    alia
    0.09
    ilde
    0.09
    ively
    0.09
    iveness
    0.09
    Activations Density 0.034%

    No Known Activations