© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 40236
    Prev
    Next
    INDEX
    Explanations

    moments

    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
    Negative Logits
    yard
    -0.11
    om
    -0.10
    elta
    -0.09
    head
    -0.08
    arest
    -0.08
    iland
    -0.08
    irma
    -0.08
    -0.08
    меÑĢ
    -0.08
    isser
    -0.08
    POSITIVE LOGITS
    aneously
    0.10
    ãģĬãģĭ
    0.10
    (stypy
    0.10
    举æĬ¥çĶµè¯Ŀ
    0.09
    åħĴ
    0.09
    侯
    0.09
    aneous
    0.09
    ARY
    0.09
    å¦Ĥæľīä¾µæĿĥ
    0.09
    çļĦãģ«
    0.09
    Activations Density 0.015%

    No Known Activations