© 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. 5479
    Prev
    Next
    INDEX
    Explanations

    string 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
    =Integer
    -0.09
    =$((
    -0.09
    eeper
    -0.08
    ä¸Ģåľº
    -0.08
    .HasValue
    -0.08
    lectic
    -0.08
    /high
    -0.08
    è·
    -0.08
     Whitney
    -0.08
    EEK
    -0.08
    POSITIVE LOGITS
    åŃĹ符串
    0.14
    -string
    0.13
     string
    0.13
     strings
    0.12
    strings
    0.11
     str
    0.11
    (strings
    0.10
    string
    0.10
    å¾Īéķ¿
    0.10
    _string
    0.10
    Activations Density 0.926%

    No Known Activations