© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51888
    Prev
    Next
    INDEX
    Explanations

    deck

    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
    atti
    -0.12
    ç¼ķ
    -0.10
    ãĥ¼ãĥī
    -0.09
     Khánh
    -0.09
    keit
    -0.09
    affle
    -0.08
    гла
    -0.08
    .TO
    -0.08
     Kurd
    -0.08
    ãĤĤãģ£ãģ¨
    -0.08
    POSITIVE LOGITS
    artment
    0.11
    ster
    0.11
    поÑĤ
    0.10
    acon
    0.10
    大æĪĺ
    0.10
    velt
    0.10
    nuts
    0.09
    amp
    0.09
    rais
    0.09
    tones
    0.09
    Activations Density 0.010%

    No Known Activations