OlaGPT is a newly developed framework that enhances large language models by simulating human-like problem-solving abilities. It incorporates six cognitive modules, including attention, memory, reasoning, learning, decision-making, and action selection. The model was evaluated on algebraic word problems and analogical reasoning questions, showing superior performance against existing benchmarks. OlaGPT is integrated with pre-existing models such as GPT-3 as base models, with different cognitive modules added. Although the framework has limitations that prevent it from providing a creative solution, it is a promising tool that could approximate the human brain model.
The Open LLM Leaderboard tracks, ranks, and evaluates language models and chatbots based on various benchmarks. Anyone from the community can submit a model for automated evaluation on the GPU cluster, as long as it is a Transformers model with weights on the Hub. The leaderboard evaluates models on four benchmarks, including AI2 Reasoning Challenge, HellaSwag, MMLU, and TruthfulQA, to test reasoning and general knowledge in both zero-shot and few-shot settings.
QLoRA allows fine-tuning of large language models on a single GPU. Using this method, they trained Guanaco, a family of chatbots based on Meta's LLaMA models, achieving over 99% of ChatGPT's performance. QLoRA reduces the memory requirement by quantizing models to 4 bits and adding low-rank adaptive weights. The team found that data quality is more important than quantity for fine-tuning, with models trained on OpenAssistant data performing better. Even the smallest Guanaco model outperformed other models, and the team believes that QLoRA will make fine-tuning more accessible, bridging the resource gap between large corporations and small teams. They also see potential for private models on mobile devices, enabling privacy-preserving fine-tuning on smartphones.
A collection of LLM (large language models) resources that can be used to build products or perform reproducible research. The table of contents includes local LLMs, LLM-based tools, training and quantization resources, non-English models, and autonomous agents.
Alpaca Electron is a user-friendly chatbot application built to interact with Alpaca AI models. The application runs locally on a user's computer, and an internet connection is only needed to download models. Alpaca Electron is compact and efficient, using alpaca.cpp as its backend and running on a CPU, making it accessible to anyone without an expensive graphics card. The application does not require external dependencies, and everything is included in the installer. The UI is borrowed from a popular chat AI, and the application supports Windows, MacOS, and Linux (untested). Alpaca Electron is dockerized and includes context memory.
Koala is a new model fine-tuned on freely available interaction data scraped from the web, with a specific focus on data that includes interaction with highly capable closed-source models such as ChatGPT. The LLaMA base model is fine-tuned on dialogue data scraped from the web and public datasets, including high-quality responses to user queries from other large language models, question answering datasets, and human feedback datasets. The resulting model, Koala-13B, shows competitive performance compared to existing models as demonstrated by human evaluation on real-world user prompts. The post suggests that learning from high-quality datasets can mitigate some of the shortcomings of smaller models and may even match the capabilities of large closed-source models in the future. This implies that the community should put more effort into curating high-quality datasets to enable safer, more factual, and more capable models instead of simply increasing the size of existing systems.
Vicuna-13B is an open-source chatbot based on a fine-tuned LLaMA base model trained on user-shared conversations collected from ShareGPT.com. The enhanced dataset and easy-to-use, scalable infrastructure of Vicuna-13B provide a competitive performance compared to other open-source models like Stanford Alpaca, as demonstrated in a preliminary evaluation. The training scripts were enhanced to handle multi-round conversations and long sequences, and the training was done with PyTorch FSDP on 8 A100 GPUs in one day. A lightweight distributed serving system was implemented for the demo. The model quality was evaluated using a set of 80 diverse questions and GPT-4 to judge the model outputs by combining the outputs from each model into a single prompt for each question. The post invites the community to interact with the online demo to test the chatbot's capabilities.
ChatGPT is GPT-3.5 finetuned with RLHF (Reinforcement Learning with Human Feedback) for human instruction and chat. Alternatives are projects featuring different instruct finetuned language models for chat. Projects are not counted if they are:
Alternative frontend projects which simply call OpenAI's APIs.
Using language models which are not finetuned for human instruction or chat.
FoldFold allExpandExpand allAre you sure you want to delete this link?Are you sure you want to delete this tag?
The personal, minimalist, super-fast, database free, bookmarking service by the Shaarli community