Update README.md
Browse files
README.md
CHANGED
|
@@ -1,6 +1,12 @@
|
|
| 1 |
---
|
|
|
|
| 2 |
library_name: transformers
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
---
|
| 5 |
|
| 6 |
# Llama-3.2-3B-Malaysian-Reasoning LoRA
|
|
@@ -11,6 +17,6 @@ Full README at [mesolitica/Llama-3.2-3B-Malaysian-Reasoning](https://huggingface
|
|
| 11 |
|
| 12 |
## Merging
|
| 13 |
|
| 14 |
-
Because Llama 3.2 3B is using tied weight embedding, so merging required to clone embedding into lm head, script at https://github.com/mesolitica/malaya/blob/master/session/small-malaysian-reasoning/merge-3b.ipynb
|
| 15 |
|
| 16 |
If the model is not a tied weight, you can use default `merge_and_unload` function from PEFT or unsloth.
|
|
|
|
| 1 |
---
|
| 2 |
+
base_model: unsloth/Llama-3.2-3B-Instruct
|
| 3 |
library_name: transformers
|
| 4 |
+
language:
|
| 5 |
+
- en
|
| 6 |
+
- ms
|
| 7 |
+
- id
|
| 8 |
+
- ta
|
| 9 |
+
- zh
|
| 10 |
---
|
| 11 |
|
| 12 |
# Llama-3.2-3B-Malaysian-Reasoning LoRA
|
|
|
|
| 17 |
|
| 18 |
## Merging
|
| 19 |
|
| 20 |
+
Because Llama 3.2 3B is using tied weight embedding, so merging required to clone embedding into lm head after that `addmm` as usual, script at https://github.com/mesolitica/malaya/blob/master/session/small-malaysian-reasoning/merge-3b.ipynb
|
| 21 |
|
| 22 |
If the model is not a tied weight, you can use default `merge_and_unload` function from PEFT or unsloth.
|