© 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. 2582
    Prev
    Next
    INDEX
    Explanations

    counting up or down

    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
    åĴı
    -0.10
    rag
    -0.09
    StateManager
    -0.09
    ä¸Ģå¹´ä¸Ģ度
    -0.08
     Zwe
    -0.08
     ÙĪاÙĦع
    -0.08
    ycle
    -0.08
    ligt
    -0.08
    .Static
    -0.08
    ot
    -0.08
    POSITIVE LOGITS
    åĩłä¸ª
    0.14
    å¹¾åĢĭ
    0.12
     two
    0.12
    åĩłåIJį
    0.11
    两个
    0.11
    åĩłçĤ¹
    0.11
     few
    0.11
    åĩłä½į
    0.11
    ä¸ī个
    0.11
    ä¿©
    0.11
    Activations Density 0.305%

    No Known Activations