© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 12348
    Prev
    Next
    INDEX
    Explanations

    to 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
    Negative Logits
    æľīçļĦ
    -0.09
    5
    -0.09
    0
    -0.09
    å¸ĪçĶŁ
    -0.09
    4
    -0.09
    7
    -0.08
    2
    -0.08
    1
    -0.08
    cloud
    -0.08
    iman
    -0.08
    POSITIVE LOGITS
    ä»¥ä¹łè¿ijå¹³
    0.10
    çݳ
    0.10
    éĤ£ä»½
    0.10
    åij¨åĽ´çļĦ
    0.09
    ä¸ĩæĪ·
    0.09
    ä¸Ĭåı¸
    0.09
    æĭĪ
    0.09
    åłĩ
    0.09
    portion
    0.09
    æīĢæıIJä¾ĽçļĦ
    0.09
    Activations Density 0.032%

    No Known Activations