p12475398
/
LLaMA-Factory-Mirror
forked from p04798526/LLaMA-Factory-Mirror
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix #1068 #1074

This commit is contained in:
hiyouga 2023-09-28 14:39:16 +08:00
parent de19614306
commit d11a545463
4 changed files with 10 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@ -14,11 +14,11 @@
## Changelog
[23/09/27] We supported **$S^2$-Attn** proposed by [LongLoRA](https://github.com/dvlab-research/LongLoRA). Try `--shift_attn` argument to enable shift short attention.
[23/09/27] We supported **$S^2$-Attn** proposed by [LongLoRA](https://github.com/dvlab-research/LongLoRA) for the LLaMA models. Try `--shift_attn` argument to enable shift short attention.
[23/09/23] We integrated MMLU, C-Eval and CMMLU benchmarks in this repo. See [this example](#evaluation) to evaluate your models.
[23/09/10] We supported using **[FlashAttention](https://github.com/Dao-AILab/flash-attention)** for the LLaMA models. Try `--flash_attn` argument to enable FlashAttention-2 if you are using RTX4090, A100 or H100 GPUs.
[23/09/10] We supported using **[FlashAttention-2](https://github.com/Dao-AILab/flash-attention)** for the LLaMA models. Try `--flash_attn` argument to enable FlashAttention-2 if you are using RTX4090, A100 or H100 GPUs.
[23/08/18] We supported **resuming training**, upgrade `transformers` to `4.31.0` to enjoy this feature.

4
README_zh.md
View File

@ -14,11 +14,11 @@
## 更新日志
[23/09/27] 我们支持了 [LongLoRA](https://github.com/dvlab-research/LongLoRA) 提出的 **$S^2$-Attn**。请使用 `--shift_attn` 参数以启用该功能。
[23/09/27] 我们针对 LLaMA 模型支持了 [LongLoRA](https://github.com/dvlab-research/LongLoRA) 提出的 **$S^2$-Attn**。请使用 `--shift_attn` 参数以启用该功能。
[23/09/23] 我们在项目中集成了 MMLU、C-Eval 和 CMMLU 评估集。使用方法请参阅[此示例](#模型评估)。
[23/09/10] 我们支持了 LLaMA 模型的 **[FlashAttention](https://github.com/Dao-AILab/flash-attention)**。如果您使用的是 RTX4090、A100 或 H100 GPU，请使用 `--flash_attn` 参数以启用 FlashAttention-2（实验性功能）。
[23/09/10] 我们针对 LLaMA 模型支持了 **[FlashAttention-2](https://github.com/Dao-AILab/flash-attention)**。如果您使用的是 RTX4090、A100 或 H100 GPU，请使用 `--flash_attn` 参数以启用 FlashAttention-2（实验性功能）。
[23/08/18] 我们支持了**训练状态恢复**，请将 `transformers` 升级至 `4.31.0` 以启用此功能。

4
evaluation/cmmlu/cmmlu.py
View File

@ -160,4 +160,8 @@ class CMMLU(datasets.GeneratorBasedBuilder):
def _generate_examples(self, filepath):
df = pd.read_csv(filepath, header=0, index_col=0, encoding="utf-8")
for i, instance in enumerate(df.to_dict(orient="records")):
question = instance.pop("Question", "")
answer = instance.pop("Answer", "")
instance["question"] = question
instance["answer"] = answer
yield i, instance

2
src/llmtuner/extras/constants.py
View File

@ -51,7 +51,9 @@ SUPPORTED_MODELS = {
"InternLM-7B-Chat": "internlm/internlm-chat-7b",
"InternLM-20B-Chat": "internlm/internlm-chat-20b",
"Qwen-7B": "Qwen/Qwen-7B",
"Qwen-14B": "Qwen/Qwen-14B",
"Qwen-7B-Chat": "Qwen/Qwen-7B-Chat",
"Qwen-14B-Chat": "Qwen/Qwen-14B-Chat",
"XVERSE-13B": "xverse/XVERSE-13B",
"XVERSE-13B-Chat": "xverse/XVERSE-13B-Chat",
"ChatGLM2-6B-Chat": "THUDM/chatglm2-6b",