Improved quantization using Openvino 2024.5.0rc1
Browse files- README.md +6 -7
- config.json +1 -1
- generation_config.json +1 -1
- openvino_model.bin +2 -2
- openvino_model.xml +0 -0
README.md
CHANGED
@@ -7,21 +7,20 @@ tags:
|
|
7 |
|
8 |
This is an INT4 quantized version of the `meta-llama/Llama-3.1-8B-Instruct` model. The Python packages used in creating this model are as follows:
|
9 |
```
|
10 |
-
openvino==2024.
|
11 |
optimum==1.23.3
|
12 |
optimum-intel==1.20.1
|
13 |
nncf==2.13.0
|
14 |
torch==2.5.1
|
15 |
-
transformers==4.46.
|
16 |
```
|
17 |
This quantized model is created using the following command:
|
18 |
```
|
19 |
-
optimum-cli export openvino
|
20 |
```
|
21 |
For more details, run the following command from your Python environment: `optimum-cli export openvino --help`
|
22 |
|
23 |
INFO:nncf:Statistics of the bitwidth distribution:
|
24 |
-
| Num bits (N)
|
25 |
-
|
26 |
-
|
|
27 |
-
| 4 | 87% (224 / 226) | 100% (224 / 224) |
|
|
|
7 |
|
8 |
This is an INT4 quantized version of the `meta-llama/Llama-3.1-8B-Instruct` model. The Python packages used in creating this model are as follows:
|
9 |
```
|
10 |
+
openvino==2024.5.0rc1
|
11 |
optimum==1.23.3
|
12 |
optimum-intel==1.20.1
|
13 |
nncf==2.13.0
|
14 |
torch==2.5.1
|
15 |
+
transformers==4.46.2
|
16 |
```
|
17 |
This quantized model is created using the following command:
|
18 |
```
|
19 |
+
optimum-cli export openvino --model "meta-llama/Llama-3.1-8B-Instruct" --weight-format int4 --group-size 128 --sym --ratio 1 --all-layers ./llama-3_1-8b-instruct-ov-int4
|
20 |
```
|
21 |
For more details, run the following command from your Python environment: `optimum-cli export openvino --help`
|
22 |
|
23 |
INFO:nncf:Statistics of the bitwidth distribution:
|
24 |
+
| Num bits (N) | % all parameters (layers) | % ratio-defining parameters (layers) |
|
25 |
+
|----------------|-----------------------------|----------------------------------------|
|
26 |
+
| 4 | 100% (226 / 226) | 100% (226 / 226) |
|
|
config.json
CHANGED
@@ -35,7 +35,7 @@
|
|
35 |
"rope_theta": 500000.0,
|
36 |
"tie_word_embeddings": false,
|
37 |
"torch_dtype": "bfloat16",
|
38 |
-
"transformers_version": "4.46.
|
39 |
"use_cache": true,
|
40 |
"vocab_size": 128256
|
41 |
}
|
|
|
35 |
"rope_theta": 500000.0,
|
36 |
"tie_word_embeddings": false,
|
37 |
"torch_dtype": "bfloat16",
|
38 |
+
"transformers_version": "4.46.2",
|
39 |
"use_cache": true,
|
40 |
"vocab_size": 128256
|
41 |
}
|
generation_config.json
CHANGED
@@ -8,5 +8,5 @@
|
|
8 |
],
|
9 |
"temperature": 0.6,
|
10 |
"top_p": 0.9,
|
11 |
-
"transformers_version": "4.46.
|
12 |
}
|
|
|
8 |
],
|
9 |
"temperature": 0.6,
|
10 |
"top_p": 0.9,
|
11 |
+
"transformers_version": "4.46.2"
|
12 |
}
|
openvino_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3e412669f665674fd8cab4953934a2de259a0d25efb6a6b4bd114f781efab67e
|
3 |
+
size 4141531712
|
openvino_model.xml
CHANGED
The diff for this file is too large to render.
See raw diff
|
|