Commit
•
0a303a8
1
Parent(s):
0b74fff
"Update README.md"
Browse files
README.md
CHANGED
@@ -10,23 +10,26 @@ inference: false
|
|
10 |
---
|
11 |
[![banner](https://maddes8cht.github.io/assets/buttons/Huggingface-banner.jpg)]()
|
12 |
|
13 |
-
I
|
14 |
|
15 |
# vigogne-falcon-7b-instruct - GGUF
|
16 |
- Model creator: [bofenghuang](https://huggingface.co/bofenghuang)
|
17 |
- Original model: [vigogne-falcon-7b-instruct](https://huggingface.co/bofenghuang/vigogne-falcon-7b-instruct)
|
18 |
|
19 |
-
#
|
20 |
|
21 |
-
As noted on the [Llama.cpp
|
22 |
|
23 |
-
|
24 |
|
25 |
-
**
|
26 |
-
|
27 |
-
**
|
28 |
-
|
29 |
-
|
|
|
|
|
|
|
30 |
|
31 |
|
32 |
|
@@ -35,6 +38,8 @@ As a solo operator of this page, I'm doing my best to expedite the process, but
|
|
35 |
Vigogne-Falcon-7B-Instruct is a Falcon-7B model fine-tuned to follow the French instructions.
|
36 |
|
37 |
|
|
|
|
|
38 |
|
39 |
|
40 |
# About GGUF format
|
|
|
10 |
---
|
11 |
[![banner](https://maddes8cht.github.io/assets/buttons/Huggingface-banner.jpg)]()
|
12 |
|
13 |
+
I'm constantly enhancing these model descriptions to provide you with the most relevant and comprehensive information
|
14 |
|
15 |
# vigogne-falcon-7b-instruct - GGUF
|
16 |
- Model creator: [bofenghuang](https://huggingface.co/bofenghuang)
|
17 |
- Original model: [vigogne-falcon-7b-instruct](https://huggingface.co/bofenghuang/vigogne-falcon-7b-instruct)
|
18 |
|
19 |
+
# Important Update for Falcon Models in llama.cpp Versions After October 18, 2023
|
20 |
|
21 |
+
As noted on the [Llama.cpp GitHub repository](https://github.com/ggerganov/llama.cpp#hot-topics), all new Llama.cpp releases after October 18, 2023, will require a re-quantization due to the new BPE tokenizer.
|
22 |
|
23 |
+
**Good news!** I am glad that my re-quantization process for Falcon Models is nearly complete. Download the latest quantized models to ensure compatibility with recent llama.cpp software.
|
24 |
|
25 |
+
**Key Points:**
|
26 |
+
|
27 |
+
- **Stay Informed:** Keep an eye on software application release schedules using llama.cpp libraries.
|
28 |
+
- **Monitor Upload Times:** Re-quantization is *almost* done. Watch for updates on my Hugging Face Model pages.
|
29 |
+
|
30 |
+
**Important Compatibility Note:** Old software will work with old Falcon models, but expect updated software to exclusively support the new models.
|
31 |
+
|
32 |
+
This change primarily affects **Falcon** and **Starcoder** models, with other models remaining unaffected.
|
33 |
|
34 |
|
35 |
|
|
|
38 |
Vigogne-Falcon-7B-Instruct is a Falcon-7B model fine-tuned to follow the French instructions.
|
39 |
|
40 |
|
41 |
+
---
|
42 |
+
|
43 |
|
44 |
|
45 |
# About GGUF format
|