Merge branch 'PaddlePaddle:develop' into dev_20250106_update_auto_tokenizer

Author: DrownFish19

.github/PULL_REQUEST_TEMPLATE.md

@@ -1,4 +1,18 @@
 <!-- Demo: https://github.com/PaddlePaddle/PaddleNLP/pull/26 -->
+#### Before submitting
+
+- [ ] Lint code. If there are lint issues, please format the code first.
+
+```shell
+# Install and register `pre-commit` in the project folder
+pip install pre-commit && pre-commit install
+
+# Process previous code files separately
+pre-commit run --file XXXX.py
+```
+
+- [ ] Add test cases into `tests` folder. If there are codecov issues, please add tests cases first.
+
 ### PR types
 <!-- One of [ New features | Bug fixes | Function optimization | Performance optimization | Breaking changes | Others ] -->
 

README.md

@@ -204,7 +204,8 @@ mkdir -p llm/data && cd llm/data
 wget https://bj.bcebos.com/paddlenlp/models/transformers/llama/data/llama_openwebtext_100k.bin
 wget https://bj.bcebos.com/paddlenlp/models/transformers/llama/data/llama_openwebtext_100k.idx
 cd .. # change folder to PaddleNLP/llm
-python -u -m paddle.distributed.launch --gpus "0,1,2,3,4,5,6,7" run_pretrain.py ./config/llama/pretrain_argument.json
+# 如需使用use_fused_rms_norm=true，需要前往slm/model_zoo/gpt-3/external_ops安装fused_ln
+python -u -m paddle.distributed.launch --gpus "0,1,2,3,4,5,6,7" run_pretrain.py ./config/llama/pretrain_argument.json --use_fused_rms_norm false
 ```
 
 ### 大模型 SFT 精调

docs/llm/devices/dcu/llama/README.md

@@ -0,0 +1 @@
+../../../../../llm/devices/dcu/llama/README.md

docs/llm/devices/gcu/llama/README.md

@@ -0,0 +1 @@
+../../../../../llm/devices/gcu/llama/README.md

docs/llm/devices/iluvatar/llama/README.md

@@ -0,0 +1 @@
+../../../../../llm/devices/iluvatar/llama/README.md

docs/llm/devices/intel_hpu/llama/README.md

@@ -0,0 +1 @@
+../../../../../llm/devices/intel_hpu/llama/README.md

docs/llm/devices/metax/llama/README.md

@@ -0,0 +1 @@
+../../../../../llm/devices/metax/llama/README.md

docs/llm/devices/mlu/llama/README.md

@@ -0,0 +1 @@
+../../../../../llm/devices/mlu/llama/README.md

docs/llm/devices/mthreads/llama/README.md

@@ -0,0 +1 @@
+../../../../../llm/devices/mthreads/llama/README.md

docs/llm/devices/npu/llama/README.md

@@ -0,0 +1 @@
+../../../../../llm/devices/npu/llama/README.md