© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10437
    Prev
    Next
    INDEX
    Explanations

    con (followed by certain suffixes)

    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
     constexpr
    -0.14
    -0.13
    itecture
    -0.10
    ÙĥÙĪÙĨ
    -0.10
    ================================================================
    -0.10
    ----------------------------------------------------------------
    -0.10
    ernes
    -0.10
    conference
    -0.10
    fully
    -0.09
    ugin
    -0.09
    POSITIVE LOGITS
    ality
    0.16
    ément
    0.11
    CEPT
    0.11
    icut
    0.11
    ider
    0.10
    ible
    0.10
    verting
    0.10
     gái
    0.10
    idor
    0.10
    er
    0.10
    Activations Density 0.198%

    No Known Activations