© 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. 39861
    Prev
    Next
    INDEX
    Explanations

    more, then added nouns

    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
    èİĺ
    -0.11
    .reporting
    -0.10
    羸
    -0.09
     Escorts
    -0.09
    è¡Į车
    -0.09
    两级
    -0.09
    çµĤãĤıãģ£ãģŁ
    -0.08
    æ¡£
    -0.08
    _ENTER
    -0.08
    STALL
    -0.08
    POSITIVE LOGITS
    åĨįæĿ¥
    0.11
    åĨį次
    0.10
    éĵº
    0.10
     again
    0.10
     nữa
    0.10
    another
    0.10
    åıĪä¸Ģ次
    0.09
     оÑĩеÑĢед
    0.09
    åıĪ被
    0.09
     another
    0.09
    Activations Density 0.061%

    No Known Activations