© 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. 21153
    Prev
    Next
    INDEX
    Explanations

    album

    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
    inine
    -0.09
    IFO
    -0.09
    ýt
    -0.08
     inne
    -0.08
    acad
    -0.08
    ivor
    -0.08
     rug
    -0.08
    èľ»
    -0.08
    Cop
    -0.08
     Evil
    -0.08
    POSITIVE LOGITS
    ey
    0.11
    ä¸Ģ页
    0.11
    ä¸Ģæľ¬
    0.11
     Shore
    0.09
    æĽĿ
    0.09
    æĩĭ
    0.09
    à¸ķะ
    0.09
    å¼ıçļĦ
    0.09
    ä¸Ģæľ¬ä¹¦
    0.09
    å¹³åĩ¡
    0.09
    Activations Density 0.013%

    No Known Activations