© 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. 36384
    Prev
    Next
    INDEX
    Explanations

    a followed by code or technical terms

    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
    åIJĮä¸Ģ个
    -0.10
    åħ®
    -0.09
     same
    -0.09
    wor
    -0.09
    ä¸Ģ个个
    -0.09
    ir
    -0.09
    æľīå¤ļ大
    -0.08
    на
    -0.08
    a
    -0.08
    vol
    -0.08
    POSITIVE LOGITS
    ustralian
    0.14
    /the
    0.12
    ustin
    0.12
    å¾Īéĩįè¦ģçļĦ
    0.12
     handful
    0.11
    egis
    0.11
    ä¸įå°ıçļĦ
    0.11
    plied
    0.11
    å¾Īåĸľæ¬¢
    0.11
    ther
    0.11
    Activations Density 1.962%

    No Known Activations