© 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. 51231
    Prev
    Next
    INDEX
    Explanations

    before doing something

    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
    æĽ´å¥½çļĦ
    -0.09
    æĿ¥åĪ°äºĨ
    -0.09
    移
    -0.09
    cade
    -0.09
    çĽijçĿ£æ£ĢæŁ¥
    -0.09
    arde
    -0.09
    abet
    -0.09
    .Matcher
    -0.09
    åĮĸè¿Ľç¨ĭ
    -0.09
    éªĹ
    -0.08
    POSITIVE LOGITS
     before
    0.29
    before
    0.25
     пÑĢежде
    0.24
    _before
    0.24
    Before
    0.23
     Before
    0.23
     antes
    0.22
    	before
    0.22
     trÆ°á»Ľc
    0.22
     ÙĤبÙĦ
    0.21
    Activations Density 0.187%

    No Known Activations