© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22287
    Prev
    Next
    INDEX
    Explanations

    change and alteration

    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
    æĮ«æĬĺ
    -0.10
    æĬĢæľ¯çłĶåıij
    -0.10
    aget
    -0.09
    erus
    -0.09
    acia
    -0.08
    Projection
    -0.08
    æĸĩåŃĹ
    -0.08
    Interop
    -0.08
    åĤ¢
    -0.08
     Strange
    -0.08
    POSITIVE LOGITS
    ä¸Ģ座
    0.09
    ful
    0.09
    æıIJåīį
    0.09
    å°ıèħ¿
    0.09
    åĽŀ头
    0.09
    è¡Įãģį
    0.09
    æĹ¥åĨħ
    0.09
     perf
    0.09
    ä¸Ńåħ±
    0.09
    人们çļĦ
    0.09
    Activations Density 0.049%

    No Known Activations