© 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. 24164
    Prev
    Next
    INDEX
    Explanations

    Sar, Mar, Lar, ar prefixes

    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
    ãģĹãģ¾ãģĨ
    -0.12
    ivities
    -0.11
    ãģĹãģ¾ãģĦ
    -0.11
    rich
    -0.10
    ally
    -0.10
    etur
    -0.10
    ãģĹãģ¾
    -0.10
    аÑĢÑĮ
    -0.10
    armed
    -0.10
    rong
    -0.10
    POSITIVE LOGITS
    ithmetic
    0.12
    ın
    0.12
    becue
    0.12
    iances
    0.11
    çĹª
    0.11
    imony
    0.11
    á¹ĩ
    0.11
    celona
    0.11
    bitrary
    0.11
    itecture
    0.10
    Activations Density 0.649%

    No Known Activations