© 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. 11483
    Prev
    Next
    INDEX
    Explanations

    DataGridView, grid, rows, columns

    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
    缴è¾ĸå¸Ĥ
    -0.11
    çĤĢ
    -0.09
    åħ¬åħĥ
    -0.09
    æ¶īå«Į
    -0.09
    ä¼ĹçŃ¹
    -0.09
    .writeFile
    -0.09
    æīĵåIJ¬
    -0.08
    åħ±éĿĴåĽ¢
    -0.08
     UIAlert
    -0.08
     Caul
    -0.08
    POSITIVE LOGITS
    jet
    0.09
    Jet
    0.09
    ADATA
    0.09
    åĵĢ
    0.08
     Jet
    0.08
    èĬĤ
    0.08
     Sorting
    0.08
     row
    0.08
    ält
    0.08
    åĩĭ
    0.08
    Activations Density 0.784%

    No Known Activations