© 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. 54616
    Prev
    Next
    INDEX
    Explanations

    full copyright

    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
    éĴī
    -0.10
     Hanson
    -0.09
    åĬ¡
    -0.08
    ocus
    -0.08
    è¿Ļ件äºĭæĥħ
    -0.08
     veter
    -0.08
     Crack
    -0.08
    APPER
    -0.08
    åĪ°äºĨ
    -0.08
    对æѤ
    -0.08
    POSITIVE LOGITS
    å®Įæķ´çļĦ
    0.14
    å®Įæķ´
    0.14
    åIJįåįķ
    0.13
     full
    0.12
     list
    0.11
     complete
    0.10
    åħ¨æĸĩ
    0.10
    åıĤä¼ļ
    0.09
     completa
    0.09
    æĺİç»Ĩ
    0.09
    Activations Density 0.129%

    No Known Activations