geov-ai / geov
The GeoV model is a large langauge model designed by Georges Harik and uses Rotary Positional Embeddings with Relative distances (RoPER). We have shared a pre-trained 9B parameter model.
☆122Updated last year
Related projects: ⓘ
- Finetune Falcon, LLaMA, MPT, and RedPajama on consumer hardware using PEFT LoRA☆99Updated last month
- Reimplementation of the task generation part from the Alpaca paper☆118Updated last year
- ☆89Updated 11 months ago
- Code repository for the c-BTM paper☆105Updated 11 months ago
- Full finetuning of large language models without large memory requirements☆94Updated 8 months ago
- Comprehensive analysis of difference in performance of QLora, Lora, and Full Finetunes.☆81Updated 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…☆143Updated this week
- Command-line script for inferencing from models such as MPT-7B-Chat☆102Updated last year
- Exploring finetuning public checkpoints on filter 8K sequences on Pile☆115Updated last year
- An all-new Language Model That Processes Ultra-Long Sequences of 100,000+ Ultra-Fast☆131Updated 2 weeks ago
- an implementation of Self-Extend, to expand the context window via grouped attention☆117Updated 8 months ago
- Tune MPTs☆84Updated last year
- Multi-Domain Expert Learning☆67Updated 7 months ago
- ☆62Updated 5 months ago
- Mixing Language Models with Self-Verification and Meta-Verification☆96Updated 10 months ago
- TART: A plug-and-play Transformer module for task-agnostic reasoning☆188Updated last year
- A Multilingual Dataset for Parsing Realistic Task-Oriented Dialogs☆112Updated last year
- Used for adaptive human in the loop evaluation of language and embedding models.☆300Updated last year
- Instruct-tuning LLaMA on consumer hardware☆66Updated last year
- ☆91Updated 5 months ago
- ☆22Updated last year
- The Next Generation Multi-Modality Superintelligence☆69Updated 2 weeks ago
- Evaluating LLMs with CommonGen-Lite☆83Updated 5 months ago
- Zeus LLM Trainer is a rewrite of Stanford Alpaca aiming to be the trainer for all Large Language Models☆67Updated last year
- ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward exp…☆218Updated 5 months ago
- Modified Stanford-Alpaca Trainer for Training Replit's Code Model☆40Updated last year
- ☆48Updated 11 months ago
- ☆92Updated last year
- Multipack distributed sampler for fast padding-free training of LLMs☆170Updated last month