© 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. 1406
    Prev
    Next
    INDEX
    Explanations

    Daniel Stenberg copyright

    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
    çĦ¶æĺ¯
    -0.11
    اذ
    -0.10
    ness
    -0.10
    ÑĪий
    -0.09
    eing
    -0.09
    วล
    -0.09
    Ģ
    -0.09
     coh
    -0.09
    shire
    -0.08
    еÑĩ
    -0.08
    POSITIVE LOGITS
    /or
    0.12
    ities
    0.11
    erve
    0.10
    ÑģÑı
    0.09
    imu
    0.09
    累累
    0.09
    elian
    0.09
    nest
    0.08
    Ķ
    0.08
    AndUpdate
    0.08
    Activations Density 0.638%

    No Known Activations