© 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. 2997
    Prev
    Next
    INDEX
    Explanations

    gerunds acting as verbs

    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
    иÑĤели
    -0.09
    .Lib
    -0.08
    ingu
    -0.08
    æ¹¾
    -0.08
    ĮĢ
    -0.08
     loạn
    -0.08
    _Equals
    -0.08
    .Networking
    -0.08
    æĪĭ
    -0.08
    èĦ©
    -0.08
    POSITIVE LOGITS
    æĿ¥çļĦ
    0.10
     oneself
    0.10
    çľĭ好
    0.08
    æĤį
    0.08
    好çľĭ
    0.08
    ç³Ł
    0.08
     future
    0.08
    ç»ĵå®ŀ
    0.08
    æ´¾
    0.08
    rs
    0.07
    Activations Density 0.218%

    No Known Activations