© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50595
    Prev
    Next
    INDEX
    Explanations

    J + word (query, dialog, unit, boss, etc.)

    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
    ½
    -0.10
    ental
    -0.09
    æĵį
    -0.09
    леÑĩ
    -0.09
    å°ı康
    -0.09
    ilha
    -0.09
    isi
    -0.09
    ç¬ŀ
    -0.08
    egin
    -0.08
    à¥įर
    -0.08
    POSITIVE LOGITS
    oning
    0.10
    åŃij
    0.10
    æµ®
    0.10
    igsaw
    0.09
    à¹ģà¸Ķà¸ĩ
    0.09
    .Cancel
    0.08
    ÙĪار
    0.08
    wahl
    0.08
    upp
    0.08
    çĪ·
    0.08
    Activations Density 0.065%

    No Known Activations