bconsolvo commited on
Commit
c83335a
1 Parent(s): 6dbd30b

update README.md

Browse files

Revamped the README.md to include a lot more information, including more code snippets. I also included all of the previous PR suggestions in here as well.

Files changed (1) hide show
  1. README.md +149 -44
README.md CHANGED
@@ -1,25 +1,81 @@
1
  ---
2
  license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
4
 
5
- ## Fine-tuning on Intel Gaudi2
6
 
7
- This model is a fine-tuned model based on [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the open source dataset [Open-Orca/SlimOrca](https://huggingface.co/datasets/Open-Orca/SlimOrca). Then we align it with DPO algorithm. For more details, you can refer 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
- ## Model date
10
- Neural-chat-7b-v3-1 was trained between September and October, 2023.
 
 
11
 
12
- ## Evaluation
 
 
 
 
 
 
 
 
13
 
14
- We submit our model to [open_llm_leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard), and the model performance has been **improved significantly** as we see from the average metric of 7 tasks from the leaderboard.
 
 
 
 
15
 
16
- | Model | Average ⬆️| ARC (25-s) ⬆️ | HellaSwag (10-s) ⬆️ | MMLU (5-s) ⬆️| TruthfulQA (MC) (0-s) ⬆️ | Winogrande (5-s) | GSM8K (5-s) | DROP (3-s) |
17
- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
18
- |[mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) | 50.32 | 59.58 | 83.31 | 64.16 | 42.15 | 78.37 | 18.12 | 6.14 |
19
- | [Intel/neural-chat-7b-v3](https://huggingface.co/Intel/neural-chat-7b-v3) | **57.31** | 67.15 | 83.29 | 62.26 | 58.77 | 78.06 | 1.21 | 50.43 |
20
- | [Intel/neural-chat-7b-v3-1](https://huggingface.co/Intel/neural-chat-7b-v3-1) | **59.06** | 66.21 | 83.64 | 62.37 | 59.65 | 78.14 | 19.56 | 43.84 |
21
-
22
- ## Training procedure
23
 
24
  ### Training hyperparameters
25
 
@@ -38,27 +94,46 @@ The following hyperparameters were used during training:
38
  - lr_scheduler_warmup_ratio: 0.03
39
  - num_epochs: 2.0
40
 
41
- ### Training sample code
42
- 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).
 
 
 
 
 
 
 
 
43
 
44
- ## Prompt Template
 
45
 
46
  ```
47
- ### System:
48
- {system}
49
- ### User:
50
- {usr}
51
- ### Assistant:
52
 
 
 
 
 
 
 
53
  ```
54
 
 
 
 
 
 
 
 
 
55
 
56
- ## Inference with transformers
 
57
 
58
  ```python
59
  import transformers
60
 
61
-
62
  model_name = 'Intel/neural-chat-7b-v3-1'
63
  model = transformers.AutoModelForCausalLM.from_pretrained(model_name)
64
  tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
@@ -100,36 +175,66 @@ Step 2: Add the result from step 1 to the third number (60)
100
 
101
  So, the sum of 100, 520, and 60 is 680.
102
  """
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
103
 
104
  ```
105
 
106
- ## Ethical Considerations and Limitations
107
- neural-chat-7b-v3-1 can produce factually incorrect output, and should not be relied on to produce factually accurate information. neural-chat-7b-v3-1 was trained on [Open-Orca/SlimOrca](https://huggingface.co/datasets/Open-Orca/SlimOrca) based on [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1). 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.
 
 
 
 
108
 
109
- Therefore, before deploying any applications of neural-chat-7b-v3-1, developers should perform safety testing.
 
 
 
 
110
 
111
- ## Disclaimer
 
 
 
 
 
 
 
 
 
 
 
 
 
112
 
113
- 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.
 
114
 
115
- ## Organizations developing the model
 
 
116
 
117
- 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.
118
 
119
- ## Useful links
120
  * Intel Neural Compressor [link](https://github.com/intel/neural-compressor)
121
  * Intel Extension for Transformers [link](https://github.com/intel/intel-extension-for-transformers)
122
 
123
- # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
124
- Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Intel__neural-chat-7b-v3-1)
125
-
126
- | Metric | Value |
127
- |-----------------------|---------------------------|
128
- | Avg. | 59.06 |
129
- | ARC (25-shot) | 66.21 |
130
- | HellaSwag (10-shot) | 83.64 |
131
- | MMLU (5-shot) | 62.37 |
132
- | TruthfulQA (0-shot) | 59.65 |
133
- | Winogrande (5-shot) | 78.14 |
134
- | GSM8K (5-shot) | 19.56 |
135
- | DROP (3-shot) | 43.84 |
 
1
  ---
2
  license: apache-2.0
3
+ tags:
4
+ - LLMs
5
+ - mistral
6
+ - Intel
7
+ pipeline_tag: text-generation
8
+ base_model: mistralai/Mistral-7B-v0.1
9
+ model-index:
10
+ - name: neural-chat-7b-v3-1
11
+ results:
12
+ - task:
13
+ type: Large Language Model
14
+ name: Large Language Model
15
+ dataset:
16
+ type: Open-Orca/SlimOrca
17
+ name: Open-Orca/SlimOrca
18
+ metrics:
19
+ - type: ARC (25-shot)
20
+ value: 66.21
21
+ name: ARC (25-shot)
22
+ verified: true
23
+ - type: HellaSwag (10-shot)
24
+ value: 83.64
25
+ name: HellaSwag (10-shot)
26
+ verified: true
27
+ - type: MMLU (5-shot)
28
+ value: 62.37
29
+ name: MMLU (5-shot)
30
+ verified: true
31
+ - type: TruthfulQA (0-shot)
32
+ value: 59.65
33
+ name: TruthfulQA (0-shot)
34
+ verified: true
35
+ - type: Winogrande (5-shot)
36
+ value: 78.14
37
+ name: Winogrande (5-shot)
38
+ verified: true
39
+ - type: GSM8K (5-shot)
40
+ value: 19.56
41
+ name: GSM8K (5-shot)
42
+ verified: true
43
+ - type: DROP (3-shot)
44
+ value: 43.84
45
+ name: DROP (3-shot)
46
+ verified: true
47
+ datasets:
48
+ - Open-Orca/SlimOrca
49
+ language:
50
+ - en
51
  ---
52
 
53
+ ## Model Details: Neural-Chat-v3-1
54
 
55
+ This model is a fine-tuned 7B parameter LLM on the Intel Gaudi 2 processor from the [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the open source dataset [Open-Orca/SlimOrca](https://huggingface.co/datasets/Open-Orca/SlimOrca). The model was aligned using the Direct Performance Optimization (DPO) method with [Intel/orca_dpo_pairs](https://huggingface.co/datasets/Intel/orca_dpo_pairs). For more information, refer to the Medium article [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).
56
 
57
+ <p align="center">
58
+ <img src="https://cdn-uploads.huggingface.co/production/uploads/6297f0e30bd2f58c647abb1d/ctASHUT5QYIxMsOFa-sHC.webp" width="500"/>
59
+ Photo by [Google DeepMind on Unsplash](https://unsplash.com/@googledeepmind?utm_source=medium&utm_medium=referral)
60
+ </p>
61
 
62
+ | Model Detail | Description |
63
+ | ----------- | ----------- |
64
+ | Model Authors - Company | Intel. The NeuralChat team with members from DCAI/AISE/AIPT. Core team members: Kaokao Lv, Liang Lv, Chang Wang, Wenxin Zhang, Xuhui Ren, and Haihao Shen.|
65
+ | Date | October, 2023 |
66
+ | Version | v3-1 |
67
+ | Type | 7B Large Language Model |
68
+ | Paper or Other Resources | [Medium Blog](https://medium.com/@NeuralCompressor/the-practice-of-supervised-finetuning-and-direct-preference-optimization-on-habana-gaudi2-a1197d8a3cd3) |
69
+ | License | Apache 2.0 |
70
+ | Questions or Comments | [Community Tab](https://huggingface.co/Intel/neural-chat-7b-v3-1/discussions) and [Intel DevHub Discord](https://discord.gg/rv2Gp55UJQ)|
71
 
72
+ | Intended Use | Description |
73
+ | ----------- | ----------- |
74
+ | 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 is doing. |
75
+ | Primary intended users | Anyone doing inference on language-related tasks. |
76
+ | 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.|
77
 
78
+ ## How To Use
 
 
 
 
 
 
79
 
80
  ### Training hyperparameters
81
 
 
94
  - lr_scheduler_warmup_ratio: 0.03
95
  - num_epochs: 2.0
96
 
97
+ ### Reproduce the model
98
+ Here is the sample code to reproduce the model: [GitHub sample code](https://github.com/intel/intel-extension-for-transformers/blob/main/intel_extension_for_transformers/neural_chat/examples/finetuning/finetune_neuralchat_v3). Here is the documentation to reproduce building the model:
99
+
100
+ ```bash
101
+ git clone https://github.com/intel/intel-extension-for-transformers.git
102
+ cd intel-extension-for-transformers
103
+
104
+ docker build --no-cache ./ --target hpu --build-arg REPO=https://github.com/intel/intel-extension-for-transformers.git --build-arg ITREX_VER=main -f ./intel_extension_for_transformers/neural_chat/docker/Dockerfile -t chatbot_finetuning:latest
105
+
106
+ docker run -it --runtime=habana -e HABANA_VISIBLE_DEVICES=all -e OMPI_MCA_btl_vader_single_copy_mechanism=none --cap-add=sys_nice --net=host --ipc=host chatbot_finetuning:latest
107
 
108
+ # after entering docker container
109
+ cd examples/finetuning/finetune_neuralchat_v3
110
 
111
  ```
112
+ We select the latest pretrained mistralai/Mistral-7B-v0.1 and the open source dataset Open-Orca/SlimOrca to conduct the experiment.
 
 
 
 
113
 
114
+ The below script use deepspeed zero2 to lanuch the training with 8 cards Gaudi2. In the `finetune_neuralchat_v3.py`, the default `use_habana=True, use_lazy_mode=True, device="hpu"` for Gaudi2. And if you want to run it on NVIDIA GPU, you can set them `use_habana=False, use_lazy_mode=False, device="auto"`.
115
+
116
+ ```python
117
+ deepspeed --include localhost:0,1,2,3,4,5,6,7 \
118
+ --master_port 29501 \
119
+ finetune_neuralchat_v3.py
120
  ```
121
 
122
+ Merge the LoRA weights:
123
+
124
+ ```python
125
+ python apply_lora.py \
126
+ --base-model-path mistralai/Mistral-7B-v0.1 \
127
+ --lora-model-path finetuned_model/ \
128
+ --output-path finetuned_model_lora
129
+ ```
130
 
131
+
132
+ ### FP32 Inference with Transformers
133
 
134
  ```python
135
  import transformers
136
 
 
137
  model_name = 'Intel/neural-chat-7b-v3-1'
138
  model = transformers.AutoModelForCausalLM.from_pretrained(model_name)
139
  tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
 
175
 
176
  So, the sum of 100, 520, and 60 is 680.
177
  """
178
+ ```
179
+
180
+ ### INT4 Inference with Transformers and Intel Extension for Transformers
181
+ ```python
182
+ from transformers import AutoTokenizer, TextStreamer
183
+ from intel_extension_for_transformers.transformers import AutoModelForCausalLM, WeightOnlyQuantConfig
184
+ model_name = "Intel/neural-chat-7b-v3-1"
185
+ config = WeightOnlyQuantConfig(compute_dtype="int8", weight_dtype="int4")
186
+ prompt = "Once upon a time, there existed a little girl,"
187
+
188
+ tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
189
+ inputs = tokenizer(prompt, return_tensors="pt").input_ids
190
+ streamer = TextStreamer(tokenizer)
191
+
192
+ model = AutoModelForCausalLM.from_pretrained(model_name, quantization_config=config)
193
+ outputs = model.generate(inputs, streamer=streamer, max_new_tokens=300)
194
 
195
  ```
196
 
197
+ | Factors | Description |
198
+ | ----------- | ----------- |
199
+ | Groups | More details about the dataset and annotations can be found at [Open-Orca/SlimOrca](https://huggingface.co/datasets/Open-Orca/SlimOrca) and the associated paper at https://arxiv.org/abs/2306.02707. |
200
+ | Instrumentation | The performance of the model can vary depending on the inputs to the model. In this case, the prompts provided can drastically change the prediction of the language model. |
201
+ | Environment | The model was trained on the Intel Gaudi 2 processor (8 cards). |
202
+ | Card Prompts | Model deployment on alternate hardware and software will change model performance. The model evaluation factors are from the Hugging Face LLM leaderboard: ARC, HellaSwag, MMLU, TruthfulQA, Winogrande, GSM8K, and DROP (see Quantitative Analyses below). |
203
 
204
+ | Metrics | Description |
205
+ | ----------- | ----------- |
206
+ | Model performance measures | The model performance was evaluated against other LLMs according to the measures on the LLM leaderboard. These were selected as this has become the standard for LLM performance. |
207
+ | Decision thresholds | No decision thresholds were used. |
208
+ | Approaches to uncertainty and variability | - |
209
 
210
+ | Training and Evaluation Data | Description |
211
+ | ----------- | ----------- |
212
+ | Datasets | The training data are from [Open-Orca/SlimOrca](https://huggingface.co/datasets/Open-Orca/SlimOrca). There is no contamination from the GSM8k test set, as this is not a part of the Open-Orca/SlimOrca dataset.|
213
+ | Motivation | - |
214
+ | Preprocessing | - |
215
+
216
+ ## Quantitative Analyses
217
+ The model was submitted to the [LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard). The detailed submission can be found here: [https://huggingface.co/datasets/open-llm-leaderboard/details_Intel__neural-chat-7b-v3-1](https://huggingface.co/datasets/open-llm-leaderboard/details_Intel__neural-chat-7b-v3-1). The metrics can be found below and show that the model has significantly improved performance from Mistral-7B-v0.1 and neural-chat-7b-v3.
218
+
219
+ | Model | Average ⬆️| ARC (25-s) ⬆️ | HellaSwag (10-s) ⬆️ | MMLU (5-s) ⬆️| TruthfulQA (MC) (0-s) ⬆️ | Winogrande (5-s) | GSM8K (5-s) | DROP (3-s) |
220
+ | --- | --- | --- | --- | --- | --- | --- | --- | --- |
221
+ |[mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) | 50.32 | 59.58 | 83.31 | 64.16 | 42.15 | 78.37 | 18.12 | 6.14 |
222
+ | [Intel/neural-chat-7b-v3](https://huggingface.co/Intel/neural-chat-7b-v3) | **57.31** | 67.15 | 83.29 | 62.26 | 58.77 | 78.06 | 1.21 | 50.43 |
223
+ | [Intel/neural-chat-7b-v3-1](https://huggingface.co/Intel/neural-chat-7b-v3-1) | **59.06** | 66.21 | 83.64 | 62.37 | 59.65 | 78.14 | 19.56 | 43.84 |
224
 
225
+ ## Ethical Considerations and Limitations
226
+ Neural-chat-7b-v3-1 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.
227
 
228
+ Therefore, before deploying any applications of neural-chat-7b-v3-1, developers should perform safety testing.
229
+
230
+ ## Caveats and Recommendations
231
 
232
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model.
233
 
234
+ Here are a couple of useful links to learn more about Intel's AI software:
235
  * Intel Neural Compressor [link](https://github.com/intel/neural-compressor)
236
  * Intel Extension for Transformers [link](https://github.com/intel/intel-extension-for-transformers)
237
 
238
+ ## Disclaimer
239
+
240
+ 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.