baihaochen bc3d137248 [feat] add MLA with decoupled RoPE (DeepSeek-V2 style) ... - Add model_minimind_mla.py with Q/KV low-rank compression and decoupled RoPE - Add kv_norm on kv_latent, q_compress + q_up + q_rope_proj for Q path - Cache format: (kv_latent, k_rope) tuple, 176 floats/tok/layer (4.4x compression) - Add --use_mla flag to all training scripts and utils - Add benchmark_gqa_vs_mla.py script		2026-05-12 18:38:47 +08:00
..
__init__.py	250426	2025-04-26 10:05:47 +08:00
model_lora.py	[update] minimind-3	2026-03-25 23:57:45 +08:00
model_minimind.py	add Chinese annotations pretrain	2026-05-08 20:44:02 +08:00
model_minimind_mla.py	[feat] add MLA with decoupled RoPE (DeepSeek-V2 style)	2026-05-12 18:38:47 +08:00
tokenizer.json	[update] minimind-3	2026-03-25 23:57:45 +08:00
tokenizer_config.json	[update] minimind-3	2026-03-25 23:57:45 +08:00