© 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. 58221
    Prev
    Next
    INDEX
    Explanations

    SS followed by abbreviations or specific words

    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
    ä¸Ģ个éĩįè¦ģ
    -0.09
    绢
    -0.09
    箫
    -0.09
    ÑıÑģ
    -0.09
    ishment
    -0.09
    SAT
    -0.08
    èIJ±
    -0.08
    媳
    -0.08
    äºĴéĢļ
    -0.08
    isman
    -0.08
    POSITIVE LOGITS
    canf
    0.12
    ively
    0.10
    rip
    0.10
    rü
    0.09
    moth
    0.09
    utherland
    0.09
    ening
    0.09
    ignment
    0.09
    ould
    0.09
    mpeg
    0.09
    Activations Density 0.050%

    No Known Activations