LLaMA-LoRA-Tuner
stanford_alpaca
LLaMA-LoRA-Tuner | stanford_alpaca | |
---|---|---|
6 | 108 | |
430 | 29,026 | |
- | 0.7% | |
7.9 | 2.0 | |
about 1 year ago | 3 months ago | |
Python | Python | |
- | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
LLaMA-LoRA-Tuner
- [P] Uptraining a pretrained model using company data?
- (HELP) Token Issue on Generation
- Help with Random Characters and Words on Output
-
Fine-tuning LLaMA for research without Meta license
I would like to fine-tune LLaMA using this tuner for a research paper, but I am wondering if it is legal to do so. If it isn't, does anyone have suggestions for alternatives which are similarly user-friendly as the one above, since I am not a good programmer? Any advice would be greatly appreciated, thank you!
-
Why run LLMs locally?
The bad news is that, as far as I know, it does require a GPU. The good news is that I've gotten training done with a 7b model on both google colab and kaggle with free accounts. Both have 'just' enough vram to make it work as long as you use load the model in 8bit. Like --load-in-8bit on the command line with oobabooga. The Lora Tuner frontend even has a colab notebook set up to simplify things even more. Though the frontend keeps the LoRA Rank and LoRA Alpha values capped pretty low. Thankfully that's just set in the GUI though. I think it was one of the files in its UI directory. Pretty easy to just hand edit it to allow for higher values if desired.
- How can I train my custom dataset on top of Vicuna?
stanford_alpaca
-
How Open is Generative AI? Part 2
Alpaca is an instruction-oriented LLM derived from LLaMA, enhanced by Stanford researchers with a dataset of 52,000 examples of following instructions, sourced from OpenAI’s InstructGPT through the self-instruct method. The extensive self-instruct dataset, details of data generation, and the model refinement code were publicly disclosed. This model complies with the licensing requirements of its base model. Due to the utilization of InstructGPT for data generation, it also adheres to OpenAI’s usage terms, which prohibit the creation of models competing with OpenAI. This illustrates how dataset restrictions can indirectly affect the resulting fine-tuned model.
- Ask HN: AI/ML papers to catch up with current state of AI?
- OpenAI board in discussions with Sam Altman to return as CEO
- Are there any AI like ChatGPT without content restrictions?
-
Fine-tuning LLMs with LoRA: A Gentle Introduction
In this article, we're going to experiment with LoRA and fine-tune Llama Alpaca using commercial hardware.
-
Creating a new Finetuned model
Most papers I did read showed at least a thousand, even 10000 at several cases, so I assumed that to be the trend in the case of Low rank adapter(PEFT) training.(source: [2305.14314] QLoRA: Efficient Finetuning of Quantized LLMs (arxiv.org) , Stanford CRFM (Alpaca) and the minimum being openchat/openchat · Hugging Face ; There are a lot more examples)
-
Shock tick up for wage growth to 7.3% in blow for Bank of England
I'm not talking about OpenAI ChatGPT I'm talking about things ALPACA, and where did they train these models? Off the existing models for a fraction of a fraction of a fraction of the cost: https://crfm.stanford.edu/2023/03/13/alpaca.html
- Bye bye Bing
-
The idea maze for AI startups (2015)
I think there's a new approach for “How do you get the data?” that wasn't available when this article was written in 2015. The new text and image generative models can now be used to synthesize training datasets.
I was working on an typing autocorrect project and needed a corpus of "text messages". Most of the traditional NLP corpuses like those available through NLTK [0] aren't suitable. But it was easy to script ChatGPT to generate thousands of believable text messages by throwing random topics at it.
Similarly, you can synthesize a training dataset by giving GPT the outputs/labels and asking it to generate a variety of inputs. For sentiment analysis... "Give me 1000 negative movie reviews" and "Now give me 1000 positive movie reviews".
The Alpaca folks used GPT-3 to generate high-quality instruction-following datasets [1] based on a small set of human samples.
Etc.
[0] https://www.nltk.org/nltk_data/
[1] https://crfm.stanford.edu/2023/03/13/alpaca.html
-
Repos and tutorials for a full finetune (not LoRA)
AFAIK, the original alpaca repo was a full finetune. https://github.com/tatsu-lab/stanford_alpaca
What are some alternatives?
CodeCapybara - Open-source Self-Instruction Tuning Code LLM
alpaca-lora - Instruct-tune LLaMA on consumer hardware
AlpacaDataCleaned - Alpaca dataset from Stanford, cleaned and curated
ChatGLM-6B - ChatGLM-6B: An Open Bilingual Dialogue Language Model | 开源双语对话语言模型
CodeCapypara - [Moved to: https://github.com/FSoft-AI4Code/CodeCapybara]
Open-Assistant - OpenAssistant is a chat-based assistant that understands tasks, can interact with third-party systems, and retrieve information dynamically to do so.
BELLE - BELLE: Be Everyone's Large Language model Engine(开源中文对话大模型)
llama.cpp - LLM inference in C/C++
lora - Train Large Language Models (LLM) using LoRA
GPTQ-for-LLaMa - 4 bits quantization of LLaMA using GPTQ
koboldcpp - A simple one-file way to run various GGML and GGUF models with KoboldAI's UI
text-generation-webui - A Gradio web UI for Large Language Models. Supports transformers, GPTQ, AWQ, EXL2, llama.cpp (GGUF), Llama models.