Pinned Loading

1. MLA_Output_Projection_Extension MLA_Output_Projection_Extension Public

   Applying Multi-head Latent Attention (MLA) style latent compression to the attention output projection (write side), with small-scale BERT/GPT2 experiments.

   Jupyter Notebook

2. huggingface/transformers huggingface/transformers Public

   🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

   Python 167k 34.7k