budecosystem/boomer-1b
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Jeevan Joy
2023-10-09 14:53:29 +00:00
committed by huggingface-web
parent 06ef4bdd47
commit bb25b84b7b
1 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
README.md
View File

@@ -4,6 +4,7 @@ language:
- en
library_name: transformers
---
<div align="center"><img src="https://github.com/BudEcosystem/boomer/blob/main/assets/boomer-logo.png" width=200></div>
@@ -15,7 +16,7 @@ library_name: transformers
We are open-sourcing one of our early experiments of pretraining with custom architecture and datasets. This 1.1B parameter model is pre-trained from scratch using a custom-curated dataset of 41B tokens. The model's architecture experiments contain the addition of flash attention and a higher intermediate dimension of the MLP layer. The dataset is a combination of wiki, stories, arxiv, math and code. The model is available on huggingface [Boomer1B](https://huggingface.co/budecosystem/boomer-1b)
<div align="center"><img src="https://github.com/BudEcosystem/boomer/blob/main/assets/boomer-arch.jpg" width=500></div>
<div align="center"><img src="https://accubits-assests.s3.ap-south-1.amazonaws.com/boomer/boomer-arch.jpg" width=500></div>
## Getting Started on GitHub 💻