Improve metadata, add link to paper
Browse filesThis PR improves the discoverability of your models
README.md
CHANGED
|
@@ -1,9 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
Artifacts for the GoldFinch paper
|
| 2 |
|
| 3 |
Code available at https://github.com/recursal/GoldFinch-paper
|
| 4 |
|
| 5 |
Contains checkpoints for various small scale ablations and 1.5B scale models, all trained on minipile
|
| 6 |
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pipeline_tag: text-generation
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
Artifacts for the GoldFinch paper
|
| 7 |
|
| 8 |
Code available at https://github.com/recursal/GoldFinch-paper
|
| 9 |
|
| 10 |
Contains checkpoints for various small scale ablations and 1.5B scale models, all trained on minipile
|
| 11 |
|
| 12 |
+
## Bibtex citation
|
| 13 |
+
|
| 14 |
+
```bibtex
|
| 15 |
+
@misc{goldstein2024goldfinchhighperformancerwkvtransformer,
|
| 16 |
+
title={GoldFinch: High Performance RWKV/Transformer Hybrid with Linear Pre-Fill and Extreme KV-Cache Compression},
|
| 17 |
+
author={Daniel Goldstein and Fares Obeid and Eric Alcaide and Guangyu Song and Eugene Cheah},
|
| 18 |
+
year={2024},
|
| 19 |
+
eprint={2407.12077},
|
| 20 |
+
archivePrefix={arXiv},
|
| 21 |
+
primaryClass={cs.CL},
|
| 22 |
+
url={https://arxiv.org/abs/2407.12077},
|
| 23 |
+
}
|
| 24 |
+
```
|