We simply applied the methodology from the paper titled "TransMLA: MLA Is All You Need" to convert the attention mechanism within the Llama-3 model from Grouped Query Attention (GQA) to Multi-Head Latent Attention (MLA), Kinda.
- Downloads last month
- 629
Model tree for ForSureTesterSim/Llama-3.1-8B-TransMLA-PreRecovery
Base model
meta-llama/Llama-3.1-8B Finetuned
meta-llama/Llama-3.1-8B-Instruct