© 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. 20901
    Prev
    Next
    INDEX
    Explanations

    years followed by nouns

    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
    她们
    -0.10
    社ä¼ļ责任
    -0.09
    壶
    -0.09
    è´ŀ
    -0.09
    两年åīį
    -0.09
    çIJĨä¼ļ
    -0.09
    apel
    -0.08
    çĸĹ
    -0.08
    è´´è¿ij
    -0.08
    ä¹IJæĦı
    -0.08
    POSITIVE LOGITS
    ä¹ħ
    0.09
    abic
    0.08
    çīĪæĿĥå½Ĵ
    0.08
    çīĪ
    0.08
    wil
    0.08
    ather
    0.08
    ably
    0.07
    å¹´
    0.07
     equivalents
    0.07
    adel
    0.07
    Activations Density 0.233%

    No Known Activations