© 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. 17474
    Prev
    Next
    INDEX
    Explanations

    chunking data

    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
    竺
    -0.10
    .surface
    -0.09
    boro
    -0.09
    xing
    -0.09
    交ä»ĺ
    -0.09
    奢
    -0.08
    hood
    -0.08
    _coin
    -0.08
    inoa
    -0.08
    folk
    -0.08
    POSITIVE LOGITS
    åĦ¿
    0.13
    åħĴ
    0.11
    busters
    0.10
    icles
    0.10
    åŁ
    0.10
    0.10
    åĿĹ
    0.10
    èĦ¸
    0.09
    agus
    0.09
    lets
    0.09
    Activations Density 0.009%

    No Known Activations