千橙
浏览器或者油猴脚本，chrome extension 中如何应用此分词工具？ 如果有个示例就好了，后端开发不是很熟悉，见谅啊
[System Setting] -> [General] -> [Device Management] ->[install a profile]
When I try to set packed=True, it's rejected by the parent class PreferenceDataset which raises a ValueError with "Packed is currently not supported for preference datasets." Is there a way...
Qwen3--vl-30b-a3b verl:vllm011:latest ```sh verl/workers/rollout/vllm_rollout/vllm_async_server.py", line 347, in run_server engine_client = AsyncLLM.from_vllm_config( ^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/vllm/utils/__init__.py", line 1572, in inner return fn(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/async_llm.py", line 207, in from_vllm_config return cls(...
when training qwen3-vl-30b-a3b with env ```sh deep_ep 1.0.0+a84a248 /home/dpsk_a2a/DeepEP deep_gemm 2.0.0+ea9c5d9 mbridge 0.15.1 megatron-core 0.14.0rc7 pplx-kernels 0.0.1 /vllm-workspace/ep_kernels_workspace/pplx-kernels torch 2.8.0+cu128 torch-tb-profiler 0.4.3 torchaudio 2.8.0+cu128 torchdata 0.11.0 torchvision 0.23.0+cu128 transformers 4.57.3...
```yaml model: _component_: torchtune.models.qwen3.lora_qwen3_moe_30b_a3b_instruct lora_attn_modules: ['q_proj', 'k_proj', 'v_proj', 'output_proj', 'w1', 'w2', 'w3', 'output'] apply_lora_to_mlp: True apply_lora_to_output: True ``` The LoRA training layer extends from the attention layer `['q_proj', 'v_proj', 'output_proj']`...