© 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. 23032
    Prev
    Next
    INDEX
    Explanations

    of the

    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
    å°½
    -0.09
    éĤ£äºĽ
    -0.09
    ä¸Ģ个éĩįè¦ģ
    -0.08
    errs
    -0.08
     Nug
    -0.08
    ær
    -0.08
    errupt
    -0.08
    ãģĤãĤĮ
    -0.08
    ä¸ĭä¸Ģ个
    -0.08
    éĤ£ç§į
    -0.08
    POSITIVE LOGITS
    该
    0.21
    該
    0.19
    该游æĪı
    0.15
    è¯¥é¡¹çĽ®
    0.15
    该项
    0.15
    该åī§
    0.14
    è¿Ļ项
    0.13
    è¿Ļåľº
    0.13
     íķ´ëĭ¹
    0.12
     данного
    0.12
    Activations Density 0.684%

    No Known Activations