© 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. 57017
    Prev
    Next
    INDEX
    Explanations

    pick in drafts and selections

    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
    gers
    -0.10
    urga
    -0.09
    æģĤ
    -0.08
    æĢ§çļĦ
    -0.08
    ulle
    -0.08
     incident
    -0.08
    cape
    -0.08
    cro
    -0.08
     lust
    -0.08
    .runner
    -0.08
    POSITIVE LOGITS
    æªĢ
    0.11
    çĽĺçĤ¹
    0.10
    åijĵ
    0.10
    æħ¨
    0.09
    à¸ķà¸Ļ
    0.09
     draft
    0.09
    æį§
    0.08
    å®Ħ
    0.08
    baum
    0.08
    draft
    0.08
    Activations Density 0.005%

    No Known Activations