© 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. 34357
    Prev
    Next
    INDEX
    Explanations

    throw away or down

    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
    çļĦä¸ĢåĪĩ
    -0.09
    èı½
    -0.09
    eer
    -0.08
    wand
    -0.08
    æ¸ĹéĢı
    -0.08
    _extensions
    -0.08
     Imperial
    -0.08
    ucc
    -0.08
    èı²
    -0.08
    社ä¼ļ责任
    -0.08
    POSITIVE LOGITS
    æİ·
    0.11
    aker
    0.11
    为ä¹ĭ
    0.10
    ables
    0.10
    è¿Ļåıªæĺ¯
    0.10
    æ¯Ľæ³½ä¸ľ
    0.09
    pad
    0.09
    OrUpdate
    0.09
    lish
    0.09
    é¦ģ
    0.09
    Activations Density 0.017%

    No Known Activations