rmihaylov / mpttune
Tune MPTs
☆84Updated last year
Related projects ⓘ
Alternatives and complementary repositories for mpttune
- QLoRA: Efficient Finetuning of Quantized LLMs☆77Updated 6 months ago
- Full finetuning of large language models without large memory requirements☆93Updated 10 months ago
- Comprehensive analysis of difference in performance of QLora, Lora, and Full Finetunes.☆81Updated last year
- ☆49Updated 7 months ago
- Multipack distributed sampler for fast padding-free training of LLMs☆175Updated 3 months ago
- ☆91Updated last year
- ☆91Updated last month
- ☆116Updated 2 months ago
- GPTQLoRA: Efficient Finetuning of Quantized LLMs with GPTQ☆96Updated last year
- Finetune Falcon, LLaMA, MPT, and RedPajama on consumer hardware using PEFT LoRA☆101Updated 3 months ago
- Manage scalable open LLM inference endpoints in Slurm clusters☆237Updated 3 months ago
- an implementation of Self-Extend, to expand the context window via grouped attention☆118Updated 10 months ago
- experiments with inference on llama☆105Updated 5 months ago
- Fully fine-tune large models like Mistral, Llama-2-13B, or Qwen-14B completely for free☆219Updated last week
- Low-Rank adapter extraction for fine-tuned transformers model☆162Updated 6 months ago
- ☆72Updated last year
- This is our own implementation of 'Layer Selective Rank Reduction'☆231Updated 5 months ago
- Q-GaLore: Quantized GaLore with INT4 Projection and Layer-Adaptive Low-Rank Gradients.☆171Updated 3 months ago
- An easy-to-understand framework for LLM samplers that rewind and revise generated tokens☆105Updated last week
- Load multiple LoRA modules simultaneously and automatically switch the appropriate combination of LoRA modules to generate the best answe…☆141Updated 9 months ago
- Zeus LLM Trainer is a rewrite of Stanford Alpaca aiming to be the trainer for all Large Language Models☆69Updated last year
- Landmark Attention: Random-Access Infinite Context Length for Transformers QLoRA☆124Updated last year
- Notus is a collection of fine-tuned LLMs using SFT, DPO, SFT+DPO, and/or any other RLHF techniques, while always keeping a data-first app…☆161Updated 9 months ago
- Exploring finetuning public checkpoints on filter 8K sequences on Pile☆115Updated last year
- Merge Transformers language models by use of gradient parameters.☆202Updated 3 months ago
- Spherical Merge Pytorch/HF format Language Models with minimal feature loss.☆111Updated last year
- QLoRA with Enhanced Multi GPU Support☆36Updated last year
- ☆200Updated 9 months ago
- inference code for mixtral-8x7b-32kseqlen☆98Updated 10 months ago
- The GeoV model is a large langauge model designed by Georges Harik and uses Rotary Positional Embeddings with Relative distances (RoPER).…☆122Updated last year