© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20558
    Prev
    Next
    INDEX
    Explanations

    make/allow/be specific

    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
    ê±´ìĦ¤
    -0.09
     гоÑĢ
    -0.09
     McGregor
    -0.09
    ((&
    -0.09
    é£İåIJ¹
    -0.09
    .statusText
    -0.09
    изм
    -0.09
    aison
    -0.08
    ElementException
    -0.08
    ataka
    -0.08
    POSITIVE LOGITS
    Specific
    0.12
    specific
    0.11
    å°½åı¯èĥ½
    0.11
    çī¹å®ļ
    0.10
     Specific
    0.10
    _specific
    0.10
     specifically
    0.10
    à¹Ĥà¸Ķยà¹Ģà¸īà¸ŀาะ
    0.09
     specific
    0.09
    åĴı
    0.09
    Activations Density 0.134%

    No Known Activations