© 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. 53462
    Prev
    Next
    INDEX
    Explanations

    poor people and neighborhoods

    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
    ç½Ķ
    -0.10
    çĦ¦
    -0.09
    çīŁ
    -0.08
    >Lorem
    -0.08
    ãĤ£
    -0.08
    onian
    -0.08
     Tanner
    -0.08
    <![
    -0.08
    illes
    -0.08
    (tol
    -0.08
    POSITIVE LOGITS
    窮
    0.12
    ç©·
    0.12
    ÑĥÑĢÑģ
    0.10
     poorer
    0.10
    (er
    0.09
    дол
    0.09
    ategy
    0.09
    ãģ»ãģĨ
    0.09
     Poor
    0.09
    erg
    0.08
    Activations Density 0.020%

    No Known Activations