Commit
•
f1c4afe
1
Parent(s):
3c5bf87
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,8 +1,15 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
| 4 |
|
| 5 |
-
This repo contains a low-rank adapter for LLaMA-7B trained on generated (not translated!) 55125 instructions in Polish.
|
| 6 |
|
| 7 |
The training took almost 16 hours on a single RTX 4090 with the following hyperparameters:
|
| 8 |
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- pl
|
| 5 |
+
tags:
|
| 6 |
+
- llama
|
| 7 |
+
- alpaca
|
| 8 |
+
- lora
|
| 9 |
+
- self-instruct
|
| 10 |
---
|
| 11 |
|
| 12 |
+
This repo contains a low-rank adapter for LLaMA-7B trained on generated (not translated!) 55125 [instructions](https://huggingface.co/datasets/chrisociepa/self-generated-instructions-pl) in Polish.
|
| 13 |
|
| 14 |
The training took almost 16 hours on a single RTX 4090 with the following hyperparameters:
|
| 15 |
|