---
license: apache-2.0
datasets:
- bigcode/starcoderdata
- tiiuae/falcon-refinedweb
language:
- en
---

A Reproduction of [OpenLLaMA](https://github.com/openlm-research/open_llama) using 128 H100 GPUs in Bfloat16.


The pretrain data consists of Falcon, Starcoder, and the wikipedia, arxiv, books, stackexchange from RedPajama. In total, this encompassed nearly 1 trillion tokens.


The model was trained over a single epoch, incorporating 2000 warm-up steps and a cosine learning rate schedule, starting at 3e-5 with 4M batch size.



![image/png](https://cdn-uploads.huggingface.co/production/uploads/643fb889b9ba82afb66d6b36/7nSjTJNB7qjwIa74Rr6kd.png)