© 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. 9106
    Prev
    Next
    INDEX
    Explanations

    Fraser University

    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
     Bradford
    -0.11
    aje
    -0.10
    ered
    -0.09
    ees
    -0.09
     ÑģÑĤоÑĢ
    -0.09
    ÑĢок
    -0.08
    ibr
    -0.08
    ãĤ§
    -0.08
    ived
    -0.08
    LEX
    -0.08
    POSITIVE LOGITS
    ij
    0.09
    ä¸Ģæŀļ
    0.09
    gend
    0.09
    ģ
    0.08
    ĭ
    0.08
    quelle
    0.08
     commuter
    0.08
    é»Ħçĵľ
    0.08
    otland
    0.08
    orthand
    0.08
    Activations Density 0.003%

    No Known Activations