© 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. 10834
    Prev
    Next
    INDEX
    Explanations

    string formatting and manipulation

    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
    asString
    -0.11
    éc
    -0.10
    odie
    -0.10
    ÑģÑĤÑĢо
    -0.10
    /she
    -0.09
    ÑģÑĤав
    -0.09
    -minus
    -0.09
    oldemort
    -0.08
    inition
    -0.08
    /mit
    -0.08
    POSITIVE LOGITS
    .fromCharCode
    0.17
    ified
    0.16
    []{"
    0.14
    IFY
    0.13
    (describing
    0.12
    ify
    0.12
    ency
    0.11
    ifier
    0.11
    ifying
    0.11
    .IsNullOrWhiteSpace
    0.10
    Activations Density 0.147%

    No Known Activations