© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51983
    Prev
    Next
    INDEX
    Explanations

    aligning shapes, text, or components

    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
    illion
    -0.10
    ially
    -0.10
    ¿
    -0.09
    æł·åŃIJ
    -0.09
    ively
    -0.09
    ials
    -0.09
    aceous
    -0.09
    åIJĮè¡Į
    -0.09
    ional
    -0.09
    leet
    -0.09
    POSITIVE LOGITS
    .TRAILING
    0.13
    .LEADING
    0.13
    raph
    0.10
    MENT
    0.10
    견
    0.09
    ظ
    0.09
    OrCreate
    0.09
    ments
    0.09
    icut
    0.09
    ément
    0.09
    Activations Density 0.052%

    No Known Activations