© 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. 21616
    Prev
    Next
    INDEX
    Explanations

    loss or upset

    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
    çĨ
    -0.10
    èĶ¼
    -0.10
    çĵł
    -0.09
    :disable
    -0.09
    æĹłæķ°
    -0.09
    讪
    -0.09
    åŃº
    -0.09
    ä¸įåIJĪ
    -0.09
    _RING
    -0.09
    bane
    -0.09
    POSITIVE LOGITS
    èĩªèº«çļĦ
    0.08
    å·Ŀ
    0.08
     themselves
    0.07
    åĨ³
    0.07
    labs
    0.07
    å°Ĩèĩªå·±çļĦ
    0.07
    ä¸Ģ线
    0.07
    åı¯ä¿¡
    0.07
    CONS
    0.07
    /tiny
    0.07
    Activations Density 0.044%

    No Known Activations