AblateIt / finetune-studyLinks
Comprehensive analysis of difference in performance of QLora, Lora, and Full Finetunes.
☆83Updated 2 years ago
Alternatives and similar repositories for finetune-study
Users that are interested in finetune-study are comparing it to the libraries listed below
Sorting:
- Just a bunch of benchmark logs for different LLMs☆119Updated last year
- ☆94Updated 2 years ago
- Fully fine-tune large models like Mistral, Llama-2-13B, or Qwen-14B completely for free☆232Updated last year
- Full finetuning of large language models without large memory requirements☆94Updated 3 months ago
- ☆95Updated 2 years ago
- ☆45Updated 2 years ago
- ☆22Updated 2 years ago
- QLoRA with Enhanced Multi GPU Support☆37Updated 2 years ago
- ☆86Updated last year
- MiniHF is an inference, human preference data collection, and fine-tuning tool for local language models. It is intended to help the user…☆183Updated last month
- Multi-Domain Expert Learning☆67Updated last year
- QLoRA: Efficient Finetuning of Quantized LLMs☆79Updated last year
- ☆23Updated 2 years ago
- an implementation of Self-Extend, to expand the context window via grouped attention☆119Updated last year
- Zeus LLM Trainer is a rewrite of Stanford Alpaca aiming to be the trainer for all Large Language Models☆70Updated 2 years ago
- inference code for mixtral-8x7b-32kseqlen☆104Updated 2 years ago
- Low-Rank adapter extraction for fine-tuned transformers models☆180Updated last year
- ☆198Updated last year
- This is our own implementation of 'Layer Selective Rank Reduction'☆240Updated last year
- An introduction to LLM Sampling☆79Updated last year
- a pipeline for using api calls to agnostically convert unstructured data into structured training data☆32Updated last year
- ☆21Updated 2 years ago
- Multipack distributed sampler for fast padding-free training of LLMs☆202Updated last year
- Reimplementation of the task generation part from the Alpaca paper☆119Updated 2 years ago
- Manage scalable open LLM inference endpoints in Slurm clusters☆278Updated last year
- ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward exp…☆226Updated 3 months ago
- Generate Synthetic Data Using OpenAI, MistralAI or AnthropicAI☆222Updated last year
- Fast & more realistic evaluation of chat language models. Includes leaderboard.☆189Updated 2 years ago
- ☆74Updated 2 years ago
- Minimal example scripts of the Hugging Face Trainer, focused on staying under 150 lines☆196Updated last year