docs: update README and remove ulysses_size from ep_fsdp_qwen3_moe.py

- Split combined "ep/sp FSDP MoE finetuning" entry in README into separate "ep FSDP MoE finetuning" and "sp FSDP finetuning" rows for clarity
- Remove ulysses_size parameter from device_mesh initialization in ep_fsdp_qwen3_moe.py as it is no longer needed for the example configuration

Author: meichangsu1

README.md

@@ -69,7 +69,8 @@ pip install -e .
 | --------------------------------- | --------------- | ------------------------------------------------- |
 | FSDP finetuning                   | transformers    | [Script](cookbook/transformers/fsdp2.py)             |
 | FSDP MoE finetuning               | transformers    | [Script](cookbook/transformers/fsdp2_moe.py)         |
-| ep/sp FSDP MoE finetuning              | transformers    | [Script](cookbook/transformers/ep_fsdp_qwen3_moe.py)         |
+| ep FSDP MoE finetuning            | transformers    | [Script](cookbook/transformers/ep_fsdp_qwen3_moe.py)|
+| sp FSDP finetuning              | transformers    | [Script](cookbook/transformers/sp_fsdp_dense.py) |
 | EP MoE finetuning                 | transformers    | [Script](cookbook/transformers/ep_fsdp_qwen3_moe.py) |
 | pp/tp/cp finetuning               | megatron        | [Script](cookbook/megatron/tp.py)                    |
 | pp/tp/cp MoE finetuning           | megatron        | [Script](cookbook/megatron/tp_moe.py)                |

cookbook/transformers/ep_fsdp_qwen3_moe.py

@@ -21,13 +21,11 @@
 # 4 gpus, dp=2, ep=2
 dp_size = 2
 ep_size = 2
-ulysses_size = 2
 
 device_mesh = DeviceMesh(
     device_type=Platform.get_platform().device_prefix(),
     mesh=np.arange(dp_size * ep_size).reshape(dp_size, ep_size),
     mesh_dim_names=('dp', 'ep'),
-    ulysses_size=ulysses_size, # enable sp
 )
 
 twinkle.initialize(