© 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. 32202
    Prev
    Next
    INDEX
    Explanations

    companion, company, accompanying words

    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
    .osgi
    -0.11
    ä¿£
    -0.10
    cloth
    -0.09
    太快
    -0.09
    á¼´
    -0.09
    ạt
    -0.09
    coat
    -0.09
    cher
    -0.09
    èĨĽ
    -0.09
    裱
    -0.09
    POSITIVE LOGITS
    hip
    0.15
    hood
    0.10
    å¾Ĺ天
    0.09
    hips
    0.09
    éļıæĦı
    0.09
    enate
    0.08
    odal
    0.08
    好åıĭ
    0.08
    ä¼´éļı
    0.08
    éĢ£ãĤĮ
    0.08
    Activations Density 0.018%

    No Known Activations