LLM360 / amber-train
Pre-training code for Amber 7B LLM
☆154Updated 6 months ago
Related projects ⓘ
Alternatives and complementary repositories for amber-train
- Data preparation code for Amber 7B LLM☆82Updated 6 months ago
- Positional Skip-wise Training for Efficient Context Window Extension of LLMs to Extremely Length (ICLR 2024)☆199Updated 6 months ago
- Multipack distributed sampler for fast padding-free training of LLMs☆178Updated 3 months ago
- Scaling Data-Constrained Language Models☆321Updated last month
- Pre-training code for CrystalCoder 7B LLM☆53Updated 6 months ago
- Experiments on speculative sampling with Llama models☆118Updated last year
- Parameter-Efficient Sparsity Crafting From Dense to Mixture-of-Experts for Instruction Tuning on General Tasks☆129Updated 2 months ago
- ☆93Updated last year
- A pipeline for LLM knowledge distillation☆78Updated 3 months ago
- ☆247Updated last year
- Repo for Rho-1: Token-level Data Selection & Selective Pretraining of LLMs.☆307Updated 7 months ago
- Implementation of the LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens Paper☆124Updated 4 months ago
- batched loras☆336Updated last year
- Code repository for the c-BTM paper☆105Updated last year
- Implementation of paper Data Engineering for Scaling Language Models to 128K Context☆438Updated 8 months ago
- [ICML'24] Data and code for our paper "Training-Free Long-Context Scaling of Large Language Models"☆358Updated last month
- Official code for "MAmmoTH2: Scaling Instructions from the Web" [NeurIPS 2024]☆124Updated 3 weeks ago
- Spherical Merge Pytorch/HF format Language Models with minimal feature loss.☆112Updated last year
- Manage scalable open LLM inference endpoints in Slurm clusters☆236Updated 4 months ago
- Expert Specialized Fine-Tuning☆145Updated last month
- A pipeline to improve skills of large language models☆191Updated this week
- Codes for the paper "∞Bench: Extending Long Context Evaluation Beyond 100K Tokens": https://arxiv.org/abs/2402.13718☆285Updated last month
- Benchmarking LLMs with Challenging Tasks from Real Users☆195Updated 2 weeks ago
- This is the official repository for Inheritune.☆105Updated last month
- Load multiple LoRA modules simultaneously and automatically switch the appropriate combination of LoRA modules to generate the best answe…☆142Updated 9 months ago
- ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward exp…☆216Updated 7 months ago
- DSIR large-scale data selection framework for language model training☆230Updated 7 months ago
- ☆451Updated 3 weeks ago
- Implementation of CALM from the paper "LLM Augmented LLMs: Expanding Capabilities through Composition", out of Google Deepmind☆169Updated 2 months ago
- A bagel, with everything.☆312Updated 7 months ago