lvkaokao bconsolvo commited on
Commit
558b0ef
1 Parent(s): 1b2b346

Minor updates to README (#1)

Browse files

- Minor updates to README (a2453d5ad543a5383a9b2bc80ef1bd0306f20e5d)


Co-authored-by: Benjamin Consolvo <bconsolvo@users.noreply.huggingface.co>

Files changed (1) hide show
  1. README.md +37 -26
README.md CHANGED
@@ -2,21 +2,34 @@
2
  license: apache-2.0
3
  ---
4
 
5
- ## Fine-tuning on Intel Gaudi2
6
 
7
- This model is a fine-tuned model based on [Intel/neural-chat-7b-v3-1](https://huggingface.co/Intel/neural-chat-7b-v3-1) on the [meta-math/MetaMathQA](https://huggingface.co/datasets/meta-math/MetaMathQA) dataset. And make a alignment using DPO method with [Intel/orca_dpo_pairs](https://huggingface.co/datasets/Intel/orca_dpo_pairs). For more details about [Intel/neural-chat-7b-v3-1](https://huggingface.co/Intel/neural-chat-7b-v3-1), you can refer the [model card](https://huggingface.co/Intel/neural-chat-7b-v3-1) and our blog [The Practice of Supervised Fine-tuning and Direct Preference Optimization on Intel Gaudi2](https://medium.com/@NeuralCompressor/the-practice-of-supervised-finetuning-and-direct-preference-optimization-on-habana-gaudi2-a1197d8a3cd3)
8
 
9
  **Note:** Adjust lora modules to trade off truthfulqa and gsm8k performance on DPO stage.
10
 
11
- ## Model date
12
- Neural-chat-7b-v3-3 was trained at December, 2023.
13
 
14
- ### Training sample code
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
15
  Here is the sample code to reproduce the model: [Sample Code](https://github.com/intel/intel-extension-for-transformers/blob/main/intel_extension_for_transformers/neural_chat/examples/finetuning/finetune_neuralchat_v3/README.md).
16
 
17
  ## Prompt Template
18
-
19
- ```
20
  ### System:
21
  {system}
22
  ### User:
@@ -25,25 +38,8 @@ Here is the sample code to reproduce the model: [Sample Code](https://github.com
25
 
26
  ```
27
 
28
- ## Ethical Considerations and Limitations
29
- neural-chat-7b-v3-3 can produce factually incorrect output, and should not be relied on to produce factually accurate information. Because of the limitations of the pretrained model and the finetuning datasets, it is possible that this model could generate lewd, biased or otherwise offensive outputs.
30
-
31
- Therefore, before deploying any applications of neural-chat-7b-v3-3, developers should perform safety testing.
32
-
33
- ## Disclaimer
34
-
35
- The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please cosult an attorney before using this model for commercial purposes.
36
-
37
- ## Organizations developing the model
38
-
39
- The NeuralChat team with members from Intel/DCAI/AISE/AIPT. Core team members: Kaokao Lv, Liang Lv, Chang Wang, Wenxin Zhang, Xuhui Ren, and Haihao Shen.
40
-
41
- ## Useful links
42
- * Intel Neural Compressor [link](https://github.com/intel/neural-compressor)
43
- * Intel Extension for Transformers [link](https://github.com/intel/intel-extension-for-transformers)
44
-
45
- # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
46
- Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Intel__neural-chat-7b-v3-1) (**note:** the leaderboard removes drop task)
47
 
48
  | Metric | Value |
49
  |-----------------------|---------------------------|
@@ -54,3 +50,18 @@ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-le
54
  | TruthfulQA (0-shot) | 63.01 |
55
  | Winogrande (5-shot) | 79.64 |
56
  | GSM8K (5-shot) | 61.11 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  license: apache-2.0
3
  ---
4
 
5
+ ## Model Details: Neural-Chat-v3-3
6
 
7
+ This model is a fine-tuned 7B parameter LLM on the Intel Gaudi 2 processor from the [Intel/neural-chat-7b-v3-1](https://huggingface.co/Intel/neural-chat-7b-v3-1) on the [meta-math/MetaMathQA](https://huggingface.co/datasets/meta-math/MetaMathQA) dataset. The model was aligned using the Direct Performance Optimization (DPO) method with [Intel/orca_dpo_pairs](https://huggingface.co/datasets/Intel/orca_dpo_pairs). The [Intel/neural-chat-7b-v3-1](https://huggingface.co/Intel/neural-chat-7b-v3-1) was originally fine-tuned from [mistralai/Mistral-7B-v-0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1). For more information, refer to our blog [The Practice of Supervised Fine-tuning and Direct Preference Optimization on Intel Gaudi2](https://medium.com/@NeuralCompressor/the-practice-of-supervised-finetuning-and-direct-preference-optimization-on-habana-gaudi2-a1197d8a3cd3).
8
 
9
  **Note:** Adjust lora modules to trade off truthfulqa and gsm8k performance on DPO stage.
10
 
 
 
11
 
12
+ | Model Detail | Description |
13
+ | ----------- | ----------- |
14
+ | Model Authors - Company | Intel. The NeuralChat team with members from Intel/DCAI/AISE/AIPT. Core team members: Kaokao Lv, Liang Lv, Chang Wang, Wenxin Zhang, Xuhui Ren, and Haihao Shen.|
15
+ | Date | December, 2023 |
16
+ | Version | v3-3 |
17
+ | Type | 7B Large Language Model |
18
+ | Paper or Other Resources | [Medium Blog](https://medium.com/@NeuralCompressor/the-practice-of-supervised-finetuning-and-direct-preference-optimization-on-habana-gaudi2-a1197d8a3cd3) |
19
+ | License | Apache 2.0 |
20
+ | Questions or Comments | [Community Tab](https://huggingface.co/Intel/neural-chat-7b-v3-3/discussions) and [Intel Developers Discord](https://discord.gg/rv2Gp55UJQ)|
21
+
22
+ | Intended Use | Description |
23
+ | ----------- | ----------- |
24
+ | Primary intended uses | You can use the fine-tuned model for several language-related tasks. Checkout the [LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) to see how this model and others from Intel are doing. |
25
+ | Primary intended users | Anyone doing inference on language-related tasks. |
26
+ | Out-of-scope uses | This model in most cases will need to be fine-tuned for your particular task. The model should not be used to intentionally create hostile or alienating environments for people.|
27
+
28
+ ## How to use and Sample Code
29
  Here is the sample code to reproduce the model: [Sample Code](https://github.com/intel/intel-extension-for-transformers/blob/main/intel_extension_for_transformers/neural_chat/examples/finetuning/finetune_neuralchat_v3/README.md).
30
 
31
  ## Prompt Template
32
+ ```plaintext
 
33
  ### System:
34
  {system}
35
  ### User:
 
38
 
39
  ```
40
 
41
+ ## [Quantitative Analyses: Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
42
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Intel__neural-chat-7b-v3-3) (**note:** the leaderboard removed drop task)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
43
 
44
  | Metric | Value |
45
  |-----------------------|---------------------------|
 
50
  | TruthfulQA (0-shot) | 63.01 |
51
  | Winogrande (5-shot) | 79.64 |
52
  | GSM8K (5-shot) | 61.11 |
53
+
54
+ ## Useful links
55
+ * Intel Neural Compressor [link](https://github.com/intel/neural-compressor)
56
+ * Intel Extension for Transformers [link](https://github.com/intel/intel-extension-for-transformers)
57
+
58
+ ## Ethical Considerations and Limitations
59
+ neural-chat-7b-v3-3 can produce factually incorrect output, and should not be relied on to produce factually accurate information. Because of the limitations of the pretrained model and the finetuning datasets, it is possible that this model could generate lewd, biased or otherwise offensive outputs.
60
+
61
+ Therefore, before deploying any applications of neural-chat-7b-v3-3, developers should perform safety testing.
62
+
63
+ ## Disclaimer
64
+
65
+ The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please cosult an attorney before using this model for commercial purposes.
66
+
67
+