Commit
•
1a9f336
1
Parent(s):
b284269
up
Browse files
README.md
CHANGED
@@ -14,12 +14,12 @@ inference: false
|
|
14 |
---
|
15 |
|
16 |
<p align="center" width="100%">
|
17 |
-
<img src="https://huggingface.co/bofenghuang/vigogne-
|
18 |
</p>
|
19 |
|
20 |
-
# Vigogne-
|
21 |
|
22 |
-
Vigogne-
|
23 |
|
24 |
For more information, please visit the Github repo: https://github.com/bofenghuang/vigogne
|
25 |
|
@@ -33,23 +33,23 @@ This repo only contains the low-rank adapter. In order to access the complete mo
|
|
33 |
from peft import PeftModel
|
34 |
from transformers import LlamaForCausalLM, LlamaTokenizer
|
35 |
|
36 |
-
|
|
|
|
|
|
|
37 |
model = LlamaForCausalLM.from_pretrained(
|
38 |
-
|
39 |
load_in_8bit=True,
|
|
|
40 |
device_map="auto",
|
41 |
)
|
42 |
-
model = PeftModel.from_pretrained(model,
|
43 |
```
|
44 |
|
45 |
You can infer this model by using the following Google Colab Notebook.
|
46 |
|
47 |
-
<a href="https://colab.research.google.com/github/bofenghuang/vigogne/blob/main/
|
48 |
|
49 |
## Limitations
|
50 |
|
51 |
Vigogne is still under development, and there are many limitations that have to be addressed. Please note that it is possible that the model generates harmful or biased content, incorrect information or generally unhelpful answers.
|
52 |
-
|
53 |
-
## Next Steps
|
54 |
-
|
55 |
-
- Add output examples
|
|
|
14 |
---
|
15 |
|
16 |
<p align="center" width="100%">
|
17 |
+
<img src="https://huggingface.co/bofenghuang/vigogne-instruct-13b/resolve/main/vigogne_logo.png" alt="Vigogne" style="width: 40%; min-width: 300px; display: block; margin: auto;">
|
18 |
</p>
|
19 |
|
20 |
+
# Vigogne-instruct-13b: A French Instruction-following LLaMA Model
|
21 |
|
22 |
+
Vigogne-instruct-13b is a LLaMA-13B model fine-tuned to follow the 🇫🇷 French instructions.
|
23 |
|
24 |
For more information, please visit the Github repo: https://github.com/bofenghuang/vigogne
|
25 |
|
|
|
33 |
from peft import PeftModel
|
34 |
from transformers import LlamaForCausalLM, LlamaTokenizer
|
35 |
|
36 |
+
base_model_name_or_path = "name/or/path/to/hf/llama/13b/model"
|
37 |
+
lora_model_name_or_path = "bofenghuang/vigogne-instruct-13b"
|
38 |
+
|
39 |
+
tokenizer = LlamaTokenizer.from_pretrained(base_model_name_or_path, padding_side="right", use_fast=False))
|
40 |
model = LlamaForCausalLM.from_pretrained(
|
41 |
+
base_model_name_or_path,
|
42 |
load_in_8bit=True,
|
43 |
+
torch_dtype=torch.float16,
|
44 |
device_map="auto",
|
45 |
)
|
46 |
+
model = PeftModel.from_pretrained(model, lora_model_name_or_path)
|
47 |
```
|
48 |
|
49 |
You can infer this model by using the following Google Colab Notebook.
|
50 |
|
51 |
+
<a href="https://colab.research.google.com/github/bofenghuang/vigogne/blob/main/notebooks/infer_instruct.ipynb" target="_blank"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/></a>
|
52 |
|
53 |
## Limitations
|
54 |
|
55 |
Vigogne is still under development, and there are many limitations that have to be addressed. Please note that it is possible that the model generates harmful or biased content, incorrect information or generally unhelpful answers.
|
|
|
|
|
|
|
|
adapter_config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"base_model_name_or_path": "
|
3 |
"bias": "none",
|
4 |
"enable_lora": null,
|
5 |
"fan_in_fan_out": false,
|
|
|
1 |
{
|
2 |
+
"base_model_name_or_path": "hf_models/llama-13b-hf",
|
3 |
"bias": "none",
|
4 |
"enable_lora": null,
|
5 |
"fan_in_fan_out": false,
|
runs/Mar26_00-26-15_koios.zaion.ai/1679786775.1921213/events.out.tfevents.1679786775.koios.zaion.ai
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:6ca642fe1808e886cf3384445799fd148aa28011a18c04f0799cba919e5e4785
|
3 |
-
size 5584
|
|
|
|
|
|
|
|
runs/Mar26_00-26-15_koios.zaion.ai/events.out.tfevents.1679786775.koios.zaion.ai
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:c6c19f4204fa57f6071b9d238843b6330ebadf3524dd95c3e463d503e34d98a2
|
3 |
-
size 12806
|
|
|
|
|
|
|
|