thu-coai / Safety-Prompts
Chinese safety prompts for evaluating and improving the safety of LLMs. 中文安全prompts，用于评估和提升大模型的安全性。
☆871Updated 8 months ago
Related projects ⓘ
Alternatives and complementary repositories for Safety-Prompts
- 面向中文大模型价值观的评估与对齐研究☆479Updated last year
- 开源SFT数据集整理,随时补充☆445Updated last year
- "他山之石、可以攻玉"：复旦白泽智能发布面向国内开源和国外商用大模型的Demo数据集JADE-DB☆313Updated last week
- Tuning LLMs with no tears💦; Sample Design Engineering (SDE) for more efficient downstream-tuning.☆969Updated 6 months ago
- Official github repo for SafetyBench, a comprehensive benchmark to evaluate LLMs' safety. [ACL 2024]☆157Updated 4 months ago
- 大模型多维度中文对齐评测基准 (ACL 2024)☆332Updated 3 months ago
- Awesome-LLM-Eval: a curated list of tools, datasets/benchmark, demos, leaderboard, papers, docs and models, mainly for Evaluation on LLMs…☆434Updated 3 weeks ago
- 人工精调的中文对话数据集和一段chatglm的微调代码☆1,158Updated 6 months ago
- FlagEval is an evaluation toolkit for AI large foundation models.☆301Updated 4 months ago
- 使用peft库，对chatGLM-6B/chatGLM2-6B实现4bit的QLoRA高效微调，并做lora model和base model的merge及4bit的量化（quantize）。☆356Updated last year
- Firefly中文LLaMA-2大模型，支持增量预训练Baichuan2、Llama2、Llama、Falcon、Qwen、Baichuan、InternLM、Bloom等大模型☆397Updated last year
- SC-Safety: 中文大模型多轮对抗安全基准☆107Updated 8 months ago
- Official github repo for C-Eval, a Chinese evaluation suite for foundation models [NeurIPS 2023]☆1,637Updated last year
- GAOKAO-Bench is an evaluation framework that utilizes GAOKAO questions as a dataset to evaluate large language models.☆553Updated 7 months ago
- ChatGLM-6B 指令学习|指令数据|Instruct☆653Updated last year
- 中文大模型微调(LLM-SFT), 数学指令数据集MWP-Instruct, 支持模型(ChatGLM-6B, LLaMA, Bloom-7B, baichuan-7B), 支持(LoRA, QLoRA, DeepSpeed, UI, TensorboardX), 支持(微…☆168Updated 6 months ago
- CMMLU: Measuring massive multitask language understanding in Chinese☆699Updated 2 weeks ago
- 《ChatGPT原理与实战：大型语言模型的算法、技术和私有化》☆341Updated 11 months ago
- pCLUE: 1000000+多任务提示学习数据集☆469Updated 2 years ago
- ☆492Updated 5 months ago
- 语言模型中文认知能力分析☆235Updated last year
- 活字通用大模型☆353Updated 2 months ago
- Phi2-Chinese-0.2B 从0开始训练自己的Phi2中文小模型，支持接入langchain加载本地知识库做检索增强生成RAG。Training your own Phi2 small chat model from scratch.☆488Updated 4 months ago
- ChatGLM2-6B 全参数微调，支持多轮对话的高效微调。☆395Updated last year
- Unify Efficient Fine-tuning of RAG Retrieval, including Embedding, ColBERT, ReRanker.☆511Updated last week
- chatglm多gpu用deepspeed和☆403Updated 4 months ago
- Panda项目是于2023年5月启动的开源海外中文大语言模型项目，致力于大模型时代探索整个技术栈，旨在推动中文自然语言处理领域的创新和合作。☆1,067Updated last year
- unified embedding model☆833Updated last year
- This is a repository used by individuals to experiment and reproduce the pre-training process of LLM.☆353Updated 6 months ago
- ShieldLM: Empowering LLMs as Aligned, Customizable and Explainable Safety Detectors [EMNLP 2024 Findings]☆158Updated last month