© 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. 20700
    Prev
    Next
    INDEX
    Explanations

    sword in the stone

    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
    åħħå®ŀ
    -0.10
    åľ¨é¦Ļ港
    -0.09
    伪
    -0.08
    è¯ł
    -0.08
    éĵ¨
    -0.08
    orsch
    -0.08
    æ°¸ä¹ħ
    -0.08
    æĪIJçĨŁ
    -0.08
    åĢĻ
    -0.08
    æĢ§å¼º
    -0.08
    POSITIVE LOGITS
    ijd
    0.10
    é¢ĵ
    0.10
     обÑĭ
    0.10
    管å±Ģ
    0.09
    æѤæŶ
    0.09
    icts
    0.09
    èĮ¨
    0.09
     اÙĦبÙĬ
    0.09
    zig
    0.09
    ipl
    0.08
    Activations Density 0.008%

    No Known Activations