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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 58952
    Prev
    Next
    INDEX
    Explanations

    my options

    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
    íĢ´
    -0.10
    몰
    -0.10
    quisitions
    -0.10
    é¢ī
    -0.10
    é£ŀè·ĥ
    -0.09
    åľ¨è¿ĩ渡
    -0.09
     xuyên
    -0.09
    angu
    -0.09
    ören
    -0.09
     ìŀIJ리
    -0.09
    POSITIVE LOGITS
     options
    0.35
    éĢī项
    0.29
    options
    0.27
     Options
    0.26
     option
    0.23
    	options
    0.23
     choices
    0.22
    Options
    0.22
     opciones
    0.22
    .options
    0.22
    Activations Density 0.137%

    No Known Activations