© 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. 63402
    Prev
    Next
    INDEX
    Explanations

    describing qualities

    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
    editary
    -0.09
    ÙħÙĬر
    -0.09
    maf
    -0.08
    æݨçIJĨ
    -0.08
    een
    -0.08
    ToArray
    -0.08
    usz
    -0.08
    овиÑĩ
    -0.08
    Fmt
    -0.08
    ä»Ģä¹Īäºĭ
    -0.08
    POSITIVE LOGITS
    åıĪ好
    0.15
     yet
    0.14
    èĢĮåıĪ
    0.13
    -looking
    0.12
    ä¸Ķ
    0.12
    yet
    0.11
    ness
    0.11
     enough
    0.11
    æĹłæ¯Ķ
    0.11
     جدا
    0.10
    Activations Density 0.372%

    No Known Activations