© 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. 2794
    Prev
    Next
    INDEX
    Explanations

    Abu Dhabi or Abuse

    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
    unn
    -0.11
    opi
    -0.10
    istic
    -0.09
    女ç¥ŀ
    -0.09
    führer
    -0.09
    大æĸ¹
    -0.09
    åģĥ
    -0.09
    èĩªè´Ł
    -0.08
    馥
    -0.08
    hop
    -0.08
    POSITIVE LOGITS
    raries
    0.10
    ../../../
    0.10
    Ìģ
    0.10
    å¾Ĺèµ·
    0.10
    teenth
    0.09
    centage
    0.09
    ucker
    0.09
    ANTS
    0.09
    antry
    0.09
    riad
    0.09
    Activations Density 0.014%

    No Known Activations