© 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. 55931
    Prev
    Next
    INDEX
    Explanations

    contrasting conjunction "but"

    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
    åĹŁ
    -0.11
    绿è±Ĩ
    -0.11
    å®Łãģ¯
    -0.09
    compressed
    -0.09
    好æ¶Īæģ¯
    -0.09
    ÑĮев
    -0.09
    æ£ĺ
    -0.08
    åį´æĺ¯
    -0.08
    éķ¿å¯¿
    -0.08
    èĵĦ
    -0.08
    POSITIVE LOGITS
    è¿Ļæĺ¯ä¸Ģç§į
    0.11
    ãi
    0.09
    è¿Ļæĺ¯
    0.09
    ãĥ¼ãĤ
    0.09
    以æѤ
    0.08
    è¿Ļæĺ¯æĪij们
    0.08
    è¿Ļæĺ¯ä¸Ģ个
    0.08
    ĺ
    0.08
    è¿Ļæĺ¯æĪij
    0.08
    liner
    0.08
    Activations Density 0.082%

    No Known Activations