© 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. 61665
    Prev
    Next
    INDEX
    Explanations

    Ro followed by common suffixes

    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
    ities
    -0.13
    æĬ¥éħ¬
    -0.10
    iasi
    -0.09
    à¹Į
    -0.09
    ÑĢеги
    -0.09
    orce
    -0.09
    sa
    -0.09
    urations
    -0.09
    çĦ¶æĺ¯
    -0.09
    ertia
    -0.08
    POSITIVE LOGITS
    èĩ¾
    0.11
    mary
    0.10
    åīįæīĢ
    0.10
    pping
    0.10
    ../
    0.10
    INGS
    0.10
    pty
    0.09
    strup
    0.09
    æł·åŃIJ
    0.09
    ning
    0.09
    Activations Density 0.028%

    No Known Activations