satreysa commited on
Commit
2846660
·
verified ·
1 Parent(s): aa9d978

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +32 -24
README.md CHANGED
@@ -1,40 +1,48 @@
1
  ---
 
2
  language:
3
- - zh
4
  - en
 
 
 
 
5
  tags:
6
- - glm
7
- - chatglm
8
- - thudm
9
- - ryzenai-npu
10
- base_model: THUDM/chatglm3-6b
11
  ---
12
-
13
- # chatglm3-6b
14
  - ## Introduction
15
- This model was created using Quark Quantization, followed by OGA Model Builder, and finalized with post-processing for NPU deployment.
 
16
  - ## Quantization Strategy
17
- - AWQ / Group 128 / Asymmetric / BF16 activations / UINT4 weights
18
-
 
19
  - ## Quick Start
20
- For quickstart, refer to [Ryzen AI doucmentation](https://ryzenai.docs.amd.com/en/latest/npu_oga.html)
21
 
22
  #### Evaluation scores
23
- The perplexity measurement is run on the wikitext-2-raw-v1 (raw data) dataset provided by Hugging Face. Perplexity score measured for prompt length 2k is 29.81679.
24
-
25
-
26
 
27
  #### License
 
28
  Modifications copyright(c) 2024 Advanced Micro Devices,Inc. All rights reserved.
29
 
30
- Licensed under the Apache License, Version 2.0 (the "License");
31
- you may not use this file except in compliance with the License.
32
- You may obtain a copy of the License at
 
 
 
 
 
33
 
34
- http://www.apache.org/licenses/LICENSE-2.0
 
 
 
35
 
36
- Unless required by applicable law or agreed to in writing, software
37
- distributed under the License is distributed on an "AS IS" BASIS,
38
- WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
39
- See the License for the specific language governing permissions and
40
- limitations under the License.
 
1
  ---
2
+ license: mit
3
  language:
 
4
  - en
5
+ - fr
6
+ base_model:
7
+ - microsoft/Phi-4-mini-instruct
8
+ pipeline_tag: text-generation
9
  tags:
10
+ - nlp
11
+ - code
12
+ - onnx
13
+ - amd
14
+ - ryzenai-hybrid
15
  ---
16
+ # microsoft/Phi-4-mini-instruct
 
17
  - ## Introduction
18
+ This model was prepared using the AMD Quark Quantization tool, followed by necessary post-processing.
19
+
20
  - ## Quantization Strategy
21
+ - GPTQ / Group size per layer lm_head 32 / Asymmetric / UINT4 Weights / BFP16 activations
22
+ - Excluded Layers: None
23
+
24
  - ## Quick Start
25
+ For quickstart, refer to [Ryzen AI doucmentation](https://ryzenai.docs.amd.com/en/latest/hybrid_oga.html)
26
 
27
  #### Evaluation scores
28
+ The MMLU scores are, astronomy: 67.76, philosophy: 69.45, and management: 72.82
 
 
29
 
30
  #### License
31
+
32
  Modifications copyright(c) 2024 Advanced Micro Devices,Inc. All rights reserved.
33
 
34
+ MIT License
35
+
36
+ Copyright (c) 2024 Advanced Micro Devices, Inc
37
+
38
+ Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal
39
+ in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
40
+ copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
41
+ The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
42
 
43
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
44
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
45
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
46
+ SOFTWARE.
47
 
48
+ license: MIT license