© 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. 31086
    Prev
    Next
    INDEX
    Explanations

    text or textual content

    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
     lẽ
    -0.10
    /dis
    -0.09
    ment
    -0.09
    ately
    -0.09
    ments
    -0.09
    stration
    -0.09
    /her
    -0.09
    athy
    -0.09
    thest
    -0.09
    ishment
    -0.08
    POSITIVE LOGITS
    MeshPro
    0.16
    ual
    0.13
    ÙĪص
    0.12
    ually
    0.12
    .RegularExpressions
    0.12
    uality
    0.12
    ural
    0.12
    imonial
    0.11
    mate
    0.10
    овÑĭй
    0.10
    Activations Density 0.077%

    No Known Activations