huggingface / trl
Train transformer language models with reinforcement learning.
β9,288Updated this week
Related projects: β
- π€ PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.β15,839Updated this week
- QLoRA: Efficient Finetuning of Quantized LLMsβ9,906Updated 3 months ago
- Fast and memory-efficient exact attentionβ13,401Updated this week
- A framework for few-shot evaluation of language models.β6,426Updated this week
- Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"β10,327Updated last month
- Accessible large language models via k-bit quantization for PyTorch.β6,029Updated this week
- Implementation of the LLaMA language model based on nanoGPT. Supports flash attention, Int8 and GPTQ 4bit quantization, LoRA and LLaMA-Adβ¦β5,958Updated last week
- A repo for distributed training of language models with Reinforcement Learning via Human Feedback (RLHF)β4,442Updated 8 months ago
- Large Language Model Text Generation Inferenceβ8,762Updated this week
- π A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (iβ¦β7,687Updated this week
- [ICLR 2024] Fine-tuning LLaMA to follow Instructions within 1 Hour and 1.2M Parametersβ5,691Updated 6 months ago
- [ICLR 2024] Efficient Streaming Language Models with Attention Sinksβ6,537Updated 2 months ago
- A high-throughput and memory-efficient inference and serving engine for LLMsβ26,822Updated this week
- Aligning pretrained language models with instruction data generated by themselves.β4,062Updated last year
- General technology for enabling AI capabilities w/ LLMs and MLLMsβ3,561Updated this week
- Ongoing research training transformer models at scaleβ9,949Updated this week
- Scripts for fine-tuning Meta Llama3 with composable FSDP & PEFT methods to cover single/multi-node GPUs. Supports default & custom dataseβ¦β11,582Updated last week
- 20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.β9,780Updated this week
- OpenLLaMA, a permissively licensed open source reproduction of Meta AIβs LLaMA 7B trained on the RedPajama datasetβ7,339Updated last year
- The TinyLlama project is an open endeavor to pretrain a 1.1B Llama model on 3 trillion tokens.β7,620Updated 4 months ago
- Robust recipes to align language models with human and AI preferencesβ4,481Updated 3 weeks ago
- Go ahead and axolotl questionsβ7,554Updated this week
- The RedPajama-Data repository contains code for preparing large datasets for training large language models.β4,522Updated last month
- An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.β4,326Updated last month
- LLM training code for Databricks foundation modelsβ3,964Updated this week
- LLMs build upon Evol Insturct: WizardLM, WizardCoder, WizardMathβ9,194Updated last month
- Instruction Tuning with GPT-4β4,165Updated last year
- β4,006Updated 3 months ago
- RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the bestβ¦β12,397Updated 2 weeks ago
- A curated list of practical guide resources of LLMs (LLMs Tree, Examples, Papers)β9,282Updated 3 months ago