abacaj / train-with-fsdpLinks
☆94Updated 2 years ago
Alternatives and similar repositories for train-with-fsdp
Users that are interested in train-with-fsdp are comparing it to the libraries listed below
Sorting:
- Comprehensive analysis of difference in performance of QLora, Lora, and Full Finetunes.☆82Updated 2 years ago
- Fully fine-tune large models like Mistral, Llama-2-13B, or Qwen-14B completely for free☆231Updated last year
- Full finetuning of large language models without large memory requirements☆93Updated last month
- Minimal example scripts of the Hugging Face Trainer, focused on staying under 150 lines☆195Updated last year
- ☆45Updated 2 years ago
- Just a bunch of benchmark logs for different LLMs☆118Updated last year
- ☆95Updated 2 years ago
- ☆88Updated last year
- ☆22Updated 2 years ago
- QLoRA with Enhanced Multi GPU Support☆37Updated 2 years ago
- Manage scalable open LLM inference endpoints in Slurm clusters☆274Updated last year
- batched loras☆347Updated 2 years ago
- A set of scripts and notebooks on LLM finetunning and dataset creation☆111Updated last year
- ☆415Updated 2 years ago
- experiments with inference on llama☆103Updated last year
- Large scale 4D parallelism pre-training for 🤗 transformers in Mixture of Experts *(still work in progress)*☆87Updated last year
- Fast & more realistic evaluation of chat language models. Includes leaderboard.☆189Updated last year
- git extension for {collaborative, communal, continual} model development☆215Updated 11 months ago
- Multipack distributed sampler for fast padding-free training of LLMs☆201Updated last year
- ☆197Updated last year
- Helpers and such for working with Lambda Cloud☆51Updated 2 years ago
- Minimalistic, extremely fast, and hackable researcher's toolbench for GPT models in 307 lines of code. Reaches <3.8 validation loss on wi…☆351Updated last year
- Doing simple retrieval from LLM models at various context lengths to measure accuracy☆106Updated last month
- ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward exp…☆224Updated last month
- MiniHF is an inference, human preference data collection, and fine-tuning tool for local language models. It is intended to help the user…☆181Updated last week
- ☆159Updated 11 months ago
- A bagel, with everything.☆324Updated last year
- Multi-Domain Expert Learning☆66Updated last year
- QLoRA: Efficient Finetuning of Quantized LLMs☆78Updated last year
- an implementation of Self-Extend, to expand the context window via grouped attention☆118Updated last year