LoneStriker commited on
Commit
c35561d
1 Parent(s): 4a4610b

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ codefuse-deepseek-33b-nlp.png filter=lfs diff=lfs merge=lfs -text
LEGAL.md ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # CodeFuse COMMUNITY LICENSE AGREEMENT
2
+ CodeFuse Release Date: September 8, 2023
3
+
4
+ By clicking to agree or by using or distributing any portion or element of the Materials, you will be deemed to have recognized and accepted the content of this Agreement, which is effective immediately.
5
+
6
+ 1. Definitions.
7
+ a. This CodeFuse COMMUNITY LICENSE AGREEMENT (this "Agreement") shall mean the terms and conditions for use, reproduction, distribution and modification of the Materials as defined by this Agreement.
8
+ b. "Ant" or "We" (or "Us") shall mean Ant Group.
9
+ c. "CodeFuse" shall mean the large language models (including CodeFuse-13B and CodeFuse-CodeLlaMa-34B), and software and algorithms, consisting of trained model weights, parameters (including optimizer states), machine-learning model code, and other elements of the foregoing distributed by Us.
10
+ d. "Documentation" shall mean the specifications, manuals and documentation accompanying CodeFuse distributed by Us.
11
+ e. "Materials" shall mean, collectively, Ant's proprietary CodeFuse and Documentation (and any portion thereof) made available under this Agreement.
12
+ f. "Object" form shall mean any form resulting from mechanical transformation or translation of a Source form, including but not limited to compiled object code, generated documentation, and conversions to other media types.
13
+ g. "Source" form shall mean the preferred form for making modifications, including but not limited to model source code, documentation source, and configuration files.
14
+ h. "Third Parties" (or "Third Party") shall mean individuals or legal entities that are not controlling, controlled by Us or You, or under common control with Us or You.
15
+ i. "You" (or "Your") shall mean a natural person or legal entity exercising the rights granted by this Agreement and/or using the Materials for any purpose and in any field of use.
16
+
17
+ 2. Grant of Rights.
18
+ You are granted a non-exclusive, worldwide, non-transferable and royalty-free limited license under Ant's intellectual property or other rights owned by Ant embodied in the Materials to use, reproduce, distribute, copy, create derivative works of, and make modifications to the Materials.
19
+
20
+ 3. Redistribution.
21
+ You may distribute or make the Materials or derivative works thereof available to a Third Party in any medium, with or without modifications, and in Source or Object form, provided that You meet the following conditions:
22
+ a. You shall provide a copy of this Agreement to such Third Party;
23
+ b. if You modify the CodeFuse model, You shall provide a prominent notice, stating how You have modified the CodeFuse model, to such Third Party; and
24
+ c. You shall retain in all copies of the Materials that You distribute the following attribution notices within a "Notice" text file distributed as a part of such copies: "CodeFuse is licensed under the CodeFuse COMMUNITY LICENSE AGREEMENT, Copyright (c) Ant Group. All Rights Reserved."
25
+ You may add Your own copyright statement to Your modifications and may provide additional or different license terms and conditions for use, reproduction, or distribution of Your modifications, or for any such derivative works as a whole, provided Your use, reproduction, and distribution of the work otherwise complies with the terms and conditions of this Agreement.
26
+
27
+ 4. Rules of Use.
28
+ You shall comply with applicable laws and regulations (including without limitation export controls or restrictions) in Your use of the Materials.
29
+
30
+ 5. Intellectual Property.
31
+ a. Ant retains ownership of all intellectual property rights in and to the Materials and derivatives made by or for Ant. Conditioned upon compliance with the terms and conditions of this Agreement, with respect to any derivative works and modifications of the Materials that are made by You, You are and will be the owner of such derivative works and modifications.
32
+ b. No trademark license is granted to use the trade names, trademarks, service marks, or product names of Ant, except as required to fulfill notice requirements under this Agreement or as required for reasonable and customary use in describing and redistributing the Materials.
33
+ c. If You commence a lawsuit or other proceedings (including a cross-claim or counterclaim in a lawsuit) against Ant or any entity alleging that the Materials or any output therefrom, or any part of the foregoing, infringe any intellectual property or other right owned or licensable by You, then all licences granted to You under this Agreement shall terminate as of the date such lawsuit or other proceeding is commenced or brought.
34
+
35
+ 6. Disclaimer of Warranty and Limitation of Liability.
36
+ a. Ant is not obligated to support, update, provide training for, or develop any further version of the Materials or to grant any license thereto.
37
+ b. THE MATERIALS ARE PROVIDED "AS IS" WITHOUT ANY EXPRESS OR IMPLIED WARRANTY OF ANY KIND INCLUDING WARRANTIES OF TITLE, MERCHANTABILITY, NONINFRINGEMENT, OR FITNESS FOR A PARTICULAR PURPOSE. WE MAKE NO WARRANTY AND ASSUME NO RESPONSIBILITY FOR THE SAFETY OR STABILITY OF THE MATERIALS AND ANY OUTPUT THEREFROM.
38
+ c. YOU ARE SOLELY RESPONSIBLE FOR DETERMINING THE APPROPRIATENESS OF USING OR REDISTRIBUTING THE MATERIALS AND ASSUME ANY RISKS ASSOCIATED WITH YOUR USE OF THE MATERIALS AND ANY OUTPUT AND RESULTS. IN NO EVENT SHALL WE BE LIABLE TO YOU FOR ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING OUT OF THIS AGREEMENT OR ARISING FROM YOUR USE OR INABILITY TO USE THE MATERIALS OR ANY OUTPUT OF IT, FOR ANY DIRECT, OR INDIRECT, SPECIAL, CONSEQUENTIAL, INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, NO MATTER HOW IT'S CAUSED OR EVEN IF ANT OR ITS AFFILIATES HAVE BEEN ADVISED OF THE POSSIBILITY OF ANY OF THE FOREGOING.
39
+ d. You will defend, indemnify and hold harmless Ant from and against any claim by any Third Party arising out of or related to Your use or distribution of the Materials.
40
+
41
+ 7. Survival and Termination.
42
+ a. The term of this Agreement shall commence upon Your acceptance of this Agreement or access to the Materials and will continue in full force and effect until terminated in accordance with the terms and conditions herein.
43
+ b. We may terminate this Agreement if You breach any of the terms or conditions of this Agreement. Upon termination of this Agreement, You must delete and cease use of the Materials. Sections 6 and 8 shall survive the termination of this Agreement.
44
+
45
+ 8. Governing Law and Jurisdiction.
46
+ a. This Agreement and any dispute arising out of or relating to it, whether in contract, tort, negligence, products liability, or otherwise, will be governed by the laws of China, without regard to conflict of law principles, and the UN Convention on Contracts for the International Sale of Goods does not apply to this Agreement.
47
+ b. The People's Courts in Hangzhou City shall have exclusive jurisdiction over any dispute arising out of this Agreement.
LOGO.jpg ADDED
MODEL_LICENSE.md ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # CodeFuse COMMUNITY LICENSE AGREEMENT
2
+ CodeFuse Release Date: September 8, 2023
3
+
4
+ By clicking to agree or by using or distributing any portion or element of the Materials, you will be deemed to have recognized and accepted the content of this Agreement, which is effective immediately.
5
+
6
+ 1. Definitions.
7
+ a. This CodeFuse COMMUNITY LICENSE AGREEMENT (this "Agreement") shall mean the terms and conditions for use, reproduction, distribution and modification of the Materials as defined by this Agreement.
8
+ b. "Ant" or "We" (or "Us") shall mean Ant Group.
9
+ c. "CodeFuse" shall mean the large language models (including CodeFuse-13B and CodeFuse-CodeLlaMa-34B), and software and algorithms, consisting of trained model weights, parameters (including optimizer states), machine-learning model code, and other elements of the foregoing distributed by Us.
10
+ d. "Documentation" shall mean the specifications, manuals and documentation accompanying CodeFuse distributed by Us.
11
+ e. "Materials" shall mean, collectively, Ant's proprietary CodeFuse and Documentation (and any portion thereof) made available under this Agreement.
12
+ f. "Object" form shall mean any form resulting from mechanical transformation or translation of a Source form, including but not limited to compiled object code, generated documentation, and conversions to other media types.
13
+ g. "Source" form shall mean the preferred form for making modifications, including but not limited to model source code, documentation source, and configuration files.
14
+ h. "Third Parties" (or "Third Party") shall mean individuals or legal entities that are not controlling, controlled by Us or You, or under common control with Us or You.
15
+ i. "You" (or "Your") shall mean a natural person or legal entity exercising the rights granted by this Agreement and/or using the Materials for any purpose and in any field of use.
16
+
17
+ 2. Grant of Rights.
18
+ You are granted a non-exclusive, worldwide, non-transferable and royalty-free limited license under Ant's intellectual property or other rights owned by Ant embodied in the Materials to use, reproduce, distribute, copy, create derivative works of, and make modifications to the Materials.
19
+
20
+ 3. Redistribution.
21
+ You may distribute or make the Materials or derivative works thereof available to a Third Party in any medium, with or without modifications, and in Source or Object form, provided that You meet the following conditions:
22
+ a. You shall provide a copy of this Agreement to such Third Party;
23
+ b. if You modify the CodeFuse model, You shall provide a prominent notice, stating how You have modified the CodeFuse model, to such Third Party; and
24
+ c. You shall retain in all copies of the Materials that You distribute the following attribution notices within a "Notice" text file distributed as a part of such copies: "CodeFuse is licensed under the CodeFuse COMMUNITY LICENSE AGREEMENT, Copyright (c) Ant Group. All Rights Reserved."
25
+ You may add Your own copyright statement to Your modifications and may provide additional or different license terms and conditions for use, reproduction, or distribution of Your modifications, or for any such derivative works as a whole, provided Your use, reproduction, and distribution of the work otherwise complies with the terms and conditions of this Agreement.
26
+
27
+ 4. Rules of Use.
28
+ You shall comply with applicable laws and regulations (including without limitation export controls or restrictions) in Your use of the Materials.
29
+
30
+ 5. Intellectual Property.
31
+ a. Ant retains ownership of all intellectual property rights in and to the Materials and derivatives made by or for Ant. Conditioned upon compliance with the terms and conditions of this Agreement, with respect to any derivative works and modifications of the Materials that are made by You, You are and will be the owner of such derivative works and modifications.
32
+ b. No trademark license is granted to use the trade names, trademarks, service marks, or product names of Ant, except as required to fulfill notice requirements under this Agreement or as required for reasonable and customary use in describing and redistributing the Materials.
33
+ c. If You commence a lawsuit or other proceedings (including a cross-claim or counterclaim in a lawsuit) against Ant or any entity alleging that the Materials or any output therefrom, or any part of the foregoing, infringe any intellectual property or other right owned or licensable by You, then all licences granted to You under this Agreement shall terminate as of the date such lawsuit or other proceeding is commenced or brought.
34
+
35
+ 6. Disclaimer of Warranty and Limitation of Liability.
36
+ a. Ant is not obligated to support, update, provide training for, or develop any further version of the Materials or to grant any license thereto.
37
+ b. THE MATERIALS ARE PROVIDED "AS IS" WITHOUT ANY EXPRESS OR IMPLIED WARRANTY OF ANY KIND INCLUDING WARRANTIES OF TITLE, MERCHANTABILITY, NONINFRINGEMENT, OR FITNESS FOR A PARTICULAR PURPOSE. WE MAKE NO WARRANTY AND ASSUME NO RESPONSIBILITY FOR THE SAFETY OR STABILITY OF THE MATERIALS AND ANY OUTPUT THEREFROM.
38
+ c. YOU ARE SOLELY RESPONSIBLE FOR DETERMINING THE APPROPRIATENESS OF USING OR REDISTRIBUTING THE MATERIALS AND ASSUME ANY RISKS ASSOCIATED WITH YOUR USE OF THE MATERIALS AND ANY OUTPUT AND RESULTS. IN NO EVENT SHALL WE BE LIABLE TO YOU FOR ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING OUT OF THIS AGREEMENT OR ARISING FROM YOUR USE OR INABILITY TO USE THE MATERIALS OR ANY OUTPUT OF IT, FOR ANY DIRECT, OR INDIRECT, SPECIAL, CONSEQUENTIAL, INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, NO MATTER HOW IT'S CAUSED OR EVEN IF ANT OR ITS AFFILIATES HAVE BEEN ADVISED OF THE POSSIBILITY OF ANY OF THE FOREGOING.
39
+ d. You will defend, indemnify and hold harmless Ant from and against any claim by any Third Party arising out of or related to Your use or distribution of the Materials.
40
+
41
+ 7. Survival and Termination.
42
+ a. The term of this Agreement shall commence upon Your acceptance of this Agreement or access to the Materials and will continue in full force and effect until terminated in accordance with the terms and conditions herein.
43
+ b. We may terminate this Agreement if You breach any of the terms or conditions of this Agreement. Upon termination of this Agreement, You must delete and cease use of the Materials. Sections 6 and 8 shall survive the termination of this Agreement.
44
+
45
+ 8. Governing Law and Jurisdiction.
46
+ a. This Agreement and any dispute arising out of or relating to it, whether in contract, tort, negligence, products liability, or otherwise, will be governed by the laws of China, without regard to conflict of law principles, and the UN Convention on Contracts for the International Sale of Goods does not apply to this Agreement.
47
+ b. The People's Courts in Hangzhou City shall have exclusive jurisdiction over any dispute arising out of this Agreement.
README.md ADDED
@@ -0,0 +1,384 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ tasks:
4
+ - code-generation
5
+ ---
6
+ # Model Card for CodeFuse-DeepSeek-33B
7
+ ![logo](LOGO.jpg)
8
+
9
+ [[中文]](#chinese) [[English]](#english)
10
+
11
+
12
+
13
+ <a id="english"></a>
14
+
15
+ ## Model Description
16
+
17
+ CodeFuse-DeepSeek-33B is a 33B Code-LLM finetuned by QLoRA on multiple code-related tasks on the base model DeepSeek-Coder-33B.
18
+
19
+ <br>
20
+
21
+ ## News and Updates
22
+
23
+ 🔥🔥🔥 2024-01-12 CodeFuse-DeepSeek-33B has been released, achieving a pass@1 (greedy decoding) score of 78.65% on HumanEval.
24
+
25
+ 🔥🔥🔥 2024-01-12 CodeFuse-Mixtral-8x7B has been released, achieving a pass@1 (greedy decoding) score of 56.1% on HumanEval, which is a 15% increase compared to Mixtral-8x7b's 40%.
26
+
27
+ 🔥🔥 2023-11-10 CodeFuse-CodeGeeX2-6B has been released, achieving a pass@1 (greedy decoding) score of 45.12% on HumanEval, which is a 9.22% increase compared to CodeGeeX2 35.9%.
28
+
29
+ 🔥🔥 2023-10-20 CodeFuse-QWen-14B technical documentation has been released. For those interested, please refer to the CodeFuse article on our WeChat official account via the provided link.(https://mp.weixin.qq.com/s/PCQPkvbvfxSPzsqjOILCDw)
30
+
31
+ 🔥🔥 2023-10-16 CodeFuse-QWen-14B has been released, achieving a pass@1 (greedy decoding) score of 48.78% on HumanEval, which is a 16% increase compared to Qwen-14b's 32.3%.
32
+
33
+ 🔥🔥 2023-09-27 CodeFuse-StarCoder-15B has been released, achieving a pass@1 (greedy decoding) score of 54.9% on HumanEval, which is a 21% increase compared to StarCoder's 33.6%.
34
+
35
+ 🔥🔥 2023-09-26 We are pleased to announce the release of the 4-bit quantized version of CodeFuse-CodeLlama-34B. Despite the quantization process, the model still achieves a remarkable 73.8% accuracy (greedy decoding) on the HumanEval pass@1 metric.
36
+
37
+ 🔥🔥 2023-09-11 CodeFuse-CodeLlama-34B has achieved 74.4% of pass@1 (greedy decoding) on HumanEval, which is SOTA results for openspurced LLMs at present.
38
+
39
+ <br>
40
+
41
+ ## Code Community
42
+
43
+ **Homepage**: 🏡 https://github.com/codefuse-ai (**Please give us your support with a Star🌟 + Fork🚀 + Watch👀**)
44
+
45
+ + If you wish to fine-tune the model yourself, you can visit ✨[MFTCoder](https://github.com/codefuse-ai/MFTCoder)✨✨
46
+
47
+
48
+ + If you wish to see a demo of the model, you can visit ✨[CodeFuse Demo](https://github.com/codefuse-ai/codefuse)✨✨
49
+
50
+ <br>
51
+
52
+ ## Performance
53
+
54
+ ### Code
55
+
56
+ | Model | HumanEval(pass@1) | Date |
57
+ |:----------------------------|:-----------------:|:-------:|
58
+ | **CodeFuse-DeepSeek-33B** | **78.65%** | 2024.01 |
59
+ | **CodeFuse-Mixtral-8x7B** | **56.10%** | 2024.01 |
60
+ | **CodeFuse-CodeLlama-34B** | 74.4% | 2023.9 |
61
+ |**CodeFuse-CodeLlama-34B-4bits** | 73.8% | 2023.9 |
62
+ | **CodeFuse-StarCoder-15B** | 54.9% | 2023.9 |
63
+ | **CodeFuse-QWen-14B** | 48.78% | 2023.10 |
64
+ | **CodeFuse-CodeGeeX2-6B** | 45.12% | 2023.11 |
65
+ | WizardCoder-Python-34B-V1.0 | 73.2% | 2023.8 |
66
+ | GPT-4(zero-shot) | 67.0% | 2023.3 |
67
+ | PanGu-Coder2 15B | 61.6% | 2023.8 |
68
+ | CodeLlama-34b-Python | 53.7% | 2023.8 |
69
+ | CodeLlama-34b | 48.8% | 2023.8 |
70
+ | GPT-3.5(zero-shot) | 48.1% | 2022.11 |
71
+ | OctoCoder | 46.2% | 2023.8 |
72
+ | StarCoder-15B | 33.6% | 2023.5 |
73
+ | Qwen-14b | 32.3% | 2023.10 |
74
+
75
+
76
+
77
+
78
+ ### NLP
79
+
80
+ ![NLP Performance Radar](codefuse-deepseek-33b-nlp.png)
81
+
82
+ <br>
83
+
84
+ ## Requirements
85
+
86
+ * python>=3.8
87
+ * pytorch>=2.0.0
88
+ * transformers>=4.33.2
89
+ * Sentencepiece
90
+ * CUDA 11.4
91
+ <br>
92
+
93
+ ## Inference String Format
94
+
95
+ The inference string is a concatenated string formed by combining conversation data(system, human and bot contents) in the training data format. It is used as input during the inference process.
96
+ Here are examples of prompts used to request the model:
97
+
98
+ **Multi-Round with System Prompt:**
99
+ ```python
100
+ """
101
+ <s>system
102
+ System instruction
103
+ <s>human
104
+ Human 1st round input
105
+ <s>bot
106
+ Bot 1st round output<|end▁of▁sentence|>
107
+ <s>human
108
+ Human 2nd round input
109
+ <s>bot
110
+ Bot 2nd round output<|end▁of▁sentence|>
111
+ ...
112
+ ...
113
+ ...
114
+ <s>human
115
+ Human nth round input
116
+ <s>bot
117
+ """
118
+ ```
119
+
120
+ **Single-Round without System Prompt:**
121
+ ```python
122
+ """
123
+ <s>human
124
+ User prompt...
125
+ <s>bot
126
+
127
+ """
128
+ ```
129
+
130
+ In this format, the system section is optional and the conversation can be either single-turn or multi-turn. When applying inference, you always make your input string end with "\<s\>bot" to ask the model generating answers.
131
+
132
+ For example, the format used to infer HumanEval is like the following:
133
+
134
+ ```
135
+ <s>human
136
+ # language: Python
137
+ from typing import List
138
+ def separate_paren_groups(paren_string: str) -> List[str]:
139
+ """ Input to this function is a string containing multiple groups of nested parentheses. Your goal is to
140
+ separate those group into separate strings and return the list of those.
141
+ Separate groups are balanced (each open brace is properly closed) and not nested within each other
142
+ Ignore any spaces in the input string.
143
+ >>> separate_paren_groups('( ) (( )) (( )( ))')
144
+ ['()', '(())', '(()())']
145
+ """
146
+ <s>bot
147
+
148
+ ```
149
+
150
+ Specifically, we also add the Programming Language Tag (e.g. "```# language: Python```" for Python) used by CodeGeex models.
151
+
152
+ ## Quickstart
153
+
154
+
155
+ ```python
156
+ import torch
157
+ from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig
158
+
159
+ model_dir = "codefuse-ai/CodeFuse-DeepSeek-33B"
160
+
161
+ def load_model_tokenizer(model_path):
162
+ tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
163
+ tokenizer.eos_token = "<|end▁of▁sentence|>"
164
+ tokenizer.pad_token = "<|end▁of▁sentence|>"
165
+ tokenizer.eos_token_id = tokenizer.convert_tokens_to_ids(tokenizer.eos_token)
166
+ tokenizer.pad_token_id = tokenizer.convert_tokens_to_ids(tokenizer.pad_token)
167
+ tokenizer.padding_side = "left"
168
+
169
+ model = AutoModelForCausalLM.from_pretrained(model_path, device_map='auto',torch_dtype=torch.bfloat16, trust_remote_code=True)
170
+ return model, tokenizer
171
+
172
+
173
+ HUMAN_ROLE_START_TAG = "<s>human\n"
174
+ BOT_ROLE_START_TAG = "<s>bot\n"
175
+
176
+ text_list = [f'{HUMAN_ROLE_START_TAG}Write a QuickSort program\n#Python\n{BOT_ROLE_START_TAG}']
177
+
178
+ model, tokenizer = load_model_tokenizer(model_dir)
179
+ inputs = tokenizer(text_list, return_tensors='pt', padding=True, add_special_tokens=False).to('cuda')
180
+ input_ids = inputs["input_ids"]
181
+ attention_mask = inputs["attention_mask"]
182
+ generation_config = GenerationConfig(
183
+ eos_token_id=tokenizer.eos_token_id,
184
+ pad_token_id=tokenizer.pad_token_id,
185
+ temperature=0.1,
186
+ max_new_tokens=512,
187
+ num_return_sequences=1,
188
+ num_beams=1,
189
+ top_p=0.95,
190
+ do_sample=False
191
+ )
192
+ outputs = model.generate(
193
+ inputs= input_ids,
194
+ attention_mask=attention_mask,
195
+ **generation_config.to_dict()
196
+ )
197
+ gen_text = tokenizer.batch_decode(outputs[:, input_ids.shape[1]:], skip_special_tokens=True)
198
+ print(gen_text[0])
199
+ ```
200
+
201
+
202
+
203
+
204
+
205
+
206
+
207
+
208
+ <a id="chinese"></a>
209
+
210
+ ## 模型简介
211
+
212
+ CodeFuse-DeepSeek-33B 是一个通过QLoRA对基座模型DeepSeek-Coder-33B进行多代码任务微调而得到的代码大模型。
213
+ <br>
214
+
215
+ ## 新闻
216
+
217
+ 🔥🔥🔥 2024-01-12 CodeFuse-DeepSeek-33B模型发布,模型在HumanEval pass@1指标为78.65% (贪婪解码)。
218
+
219
+ 🔥🔥🔥 2023-11-10 开源了CodeFuse-CodeGeeX2-6B模型,在HumanEval pass@1(greedy decoding)上可以达到48.12%, 比CodeGeeX2提高了9.22%的代码能力(HumanEval)
220
+
221
+ 🔥🔥🔥 2023-10-20 公布了CodeFuse-QWen-14B技术文档,感兴趣详见微信公众号CodeFuse文章:https://mp.weixin.qq.com/s/PCQPkvbvfxSPzsqjOILCDw
222
+
223
+ 🔥🔥🔥 2023-10-16开源了CodeFuse-QWen-14B模型,在HumanEval pass@1(greedy decoding)上可以达到48.78%, 比Qwen-14b提高了16%的代码能力(HumanEval)
224
+
225
+ 🔥🔥🔥 2023-09-27开源了CodeFuse-StarCoder-15B模型,在HumanEval pass@1(greedy decoding)上可以达到54.9%, 比StarCoder提高了21%的代码能力(HumanEval)
226
+
227
+ 🔥🔥🔥 2023-09-26 [CodeFuse-CodeLlama-34B 4bits](https://modelscope.cn/models/codefuse-ai/CodeFuse-CodeLlama-34B-4bits/summary)量化版本发布,量化后模型在HumanEval pass@1指标为73.8% (贪婪解码)。
228
+
229
+ 🔥🔥🔥 2023-09-11 [CodeFuse-CodeLlama-34B](https://modelscope.cn/models/codefuse-ai/CodeFuse-CodeLlama-34B/summary)发布,HumanEval pass@1指标达到74.4% (贪婪解码), 为当前开源SOTA。
230
+
231
+ <br>
232
+
233
+ ## 代码社区
234
+ **大本营**: 🏡 https://github.com/codefuse-ai (**请支持我们的项目Star🌟 + Fork🚀 + Watch👀**)
235
+
236
+ + 如果您想自己微调该模型,可以访问 ✨[MFTCoder](https://github.com/codefuse-ai/MFTCoder)✨✨
237
+
238
+ + 如果您想观看该模型示例,可以访问 ✨[CodeFuse Demo](https://github.com/codefuse-ai/codefuse)✨✨
239
+
240
+ <br>
241
+
242
+
243
+ ## 评测表现
244
+
245
+ ### 代码
246
+
247
+
248
+ | 模型 | HumanEval(pass@1) | 日期 |
249
+ |:----------------------------|:-----------------:|:-------:|
250
+ | **CodeFuse-CodeLlama-34B** | 74.4% | 2023.9 |
251
+ |**CodeFuse-CodeLlama-34B-4bits** | 73.8% | 2023.9 |
252
+ | WizardCoder-Python-34B-V1.0 | 73.2% | 2023.8 |
253
+ | GPT-4(zero-shot) | 67.0% | 2023.3 |
254
+ | PanGu-Coder2 15B | 61.6% | 2023.8 |
255
+ | CodeLlama-34b-Python | 53.7% | 2023.8 |
256
+ | CodeLlama-34b | 48.8% | 2023.8 |
257
+ | GPT-3.5(zero-shot) | 48.1% | 2022.11 |
258
+ | OctoCoder | 46.2% | 2023.8 |
259
+ | StarCoder-15B | 33.6% | 2023.5 |
260
+ | Qwen-14b | 32.3% | 2023.10 |
261
+ | **CodeFuse-StarCoder-15B** | 54.9% | 2023.9 |
262
+ | **CodeFuse-QWen-14B** | 48.78% | 2023.8 |
263
+ | **CodeFuse-CodeGeeX2-6B** | 45.12% | 2023.11 |
264
+ | **CodeFuse-DeepSeek-33B**. | **78.65%** | 2024.01 |
265
+
266
+
267
+ ### NLP
268
+
269
+ ![NLP Performance Radar](codefuse-deepseek-33b-nlp.png)
270
+
271
+ ## Requirements
272
+
273
+ * python>=3.8
274
+ * pytorch>=2.0.0
275
+ * transformers>=4.33.2
276
+ * Sentencepiece
277
+ * CUDA 11.4
278
+ <br>
279
+
280
+ ## 推理数据格式
281
+
282
+ 推理数据为模型在训练数据格式下拼接的字符串形式,它也是推理时输入prompt拼接的方式. 下面分别是带系统提示的多轮会话格式和不带系统提示的单轮会话格式:
283
+
284
+ **带System提示的多轮会话格式:**
285
+ ```python
286
+ """
287
+ <s>system
288
+ System instruction
289
+ <s>human
290
+ Human 1st round input
291
+ <s>bot
292
+ Bot 1st round output<|end▁of▁sentence|>
293
+ <s>human
294
+ Human 2nd round input
295
+ <s>bot
296
+ Bot 2nd round output<|end▁of▁sentence|>
297
+ ...
298
+ ...
299
+ ...
300
+ <s>human
301
+ Human nth round input
302
+ <s>bot
303
+ """
304
+ ```
305
+
306
+ **不带System提示的单轮会话格式:**
307
+ ```python
308
+ """
309
+ <s>human
310
+ User prompt...
311
+ <s>bot
312
+
313
+ """
314
+ ```
315
+
316
+ 在这个格式中,System提示是可选的(按需设定),支持单轮会话也支持多轮会话。推理时,请确保拼接的prompt字符串以"\<s\>bot\n"结尾,引导模型生成回答。
317
+
318
+ 例如,推理HumanEval数据时使用的格式如下所示:
319
+
320
+ ```python
321
+ <s>human
322
+ # language: Python
323
+ from typing import List
324
+ def separate_paren_groups(paren_string: str) -> List[str]:
325
+ """ Input to this function is a string containing multiple groups of nested parentheses. Your goal is to
326
+ separate those group into separate strings and return the list of those.
327
+ Separate groups are balanced (each open brace is properly closed) and not nested within each other
328
+ Ignore any spaces in the input string.
329
+ >>> separate_paren_groups('( ) (( )) (( )( ))')
330
+ ['()', '(())', '(()())']
331
+ """
332
+ <s>bot
333
+
334
+ ```
335
+
336
+ 特别地,我们也使用了CodeGeeX系列模型采用的编程语言区分标签(例如,对于Python语言,我们会使用"```# language: Python```")。
337
+
338
+ ## 快速使用
339
+
340
+ ```python
341
+ import torch
342
+ from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig
343
+
344
+ model_dir = "codefuse-ai/CodeFuse-DeepSeek-33B"
345
+
346
+ def load_model_tokenizer(model_path):
347
+ tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
348
+ tokenizer.eos_token = "<|end▁of▁sentence|>"
349
+ tokenizer.pad_token = "<|end▁of▁sentence|>"
350
+ tokenizer.eos_token_id = tokenizer.convert_tokens_to_ids(tokenizer.eos_token)
351
+ tokenizer.pad_token_id = tokenizer.convert_tokens_to_ids(tokenizer.pad_token)
352
+ tokenizer.padding_side = "left"
353
+
354
+ model = AutoModelForCausalLM.from_pretrained(model_path, device_map='auto',torch_dtype=torch.bfloat16, trust_remote_code=True)
355
+ return model, tokenizer
356
+
357
+ HUMAN_ROLE_START_TAG = "<s>human\n"
358
+ BOT_ROLE_START_TAG = "<s>bot\n"
359
+
360
+ text_list = [f'{HUMAN_ROLE_START_TAG}请写一个快排程序\n#Python\n{BOT_ROLE_START_TAG}']
361
+
362
+ model, tokenizer = load_model_tokenizer(model_dir)
363
+ inputs = tokenizer(text_list, return_tensors='pt', padding=True, add_special_tokens=False).to('cuda')
364
+ input_ids = inputs["input_ids"]
365
+ attention_mask = inputs["attention_mask"]
366
+ generation_config = GenerationConfig(
367
+ eos_token_id=tokenizer.eos_token_id,
368
+ pad_token_id=tokenizer.pad_token_id,
369
+ temperature=0.2,
370
+ max_new_tokens=512,
371
+ num_return_sequences=1,
372
+ num_beams=1,
373
+ top_p=0.95,
374
+ do_sample=False
375
+ )
376
+ outputs = model.generate(
377
+ inputs= input_ids,
378
+ attention_mask=attention_mask,
379
+ **generation_config.to_dict()
380
+ )
381
+ gen_text = tokenizer.batch_decode(outputs[:, input_ids.shape[1]:], skip_special_tokens=True)
382
+ print(gen_text[0])
383
+ ```
384
+
codefuse-deepseek-33b-nlp.png ADDED

Git LFS Details

  • SHA256: e78294652b904ef725866bade3e720fb9a15423841051033fc6e1d5718b4c66c
  • Pointer size: 132 Bytes
  • Size of remote file: 1.23 MB
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "bos_token_id": 32013,
6
+ "eos_token_id": 32014,
7
+ "hidden_act": "silu",
8
+ "hidden_size": 7168,
9
+ "initializer_range": 0.02,
10
+ "intermediate_size": 19200,
11
+ "max_position_embeddings": 16384,
12
+ "model_type": "llama",
13
+ "num_attention_heads": 56,
14
+ "num_hidden_layers": 62,
15
+ "num_key_value_heads": 8,
16
+ "pretraining_tp": 1,
17
+ "rms_norm_eps": 1e-06,
18
+ "rope_scaling": {
19
+ "factor": 4.0,
20
+ "type": "linear"
21
+ },
22
+ "rope_theta": 100000,
23
+ "tie_word_embeddings": false,
24
+ "torch_dtype": "bfloat16",
25
+ "transformers_version": "4.34.1",
26
+ "use_cache": true,
27
+ "vocab_size": 32256
28
+ }
configuration.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"framework": "pytorch", "task": "text-generation", "allow_remote": true}
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 32013,
4
+ "eos_token_id": 32014,
5
+ "transformers_version": "4.34.0"
6
+ }
output-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4af378ba7b33eee8edf8ca86b2fa1af56c90e94579e5a143ee882f1e21a39891
3
+ size 8572320616
output-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c08cc46b5cff114bd35e9f540059fb1f1b0ed08ef4a7f34503d482172c2721c1
3
+ size 8522776048
output-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4851455b46612f60a7e5187454bc2700cf15cee7f52121a4b78786b1a4c7120d
3
+ size 8529677648
output-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33864e09f4631aaf7a18b9da6fc1d44c090866f1201185e4a4f4135fd3655d32
3
+ size 7938693704
pytorch_model.bin.index.json ADDED
@@ -0,0 +1,568 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 66685982720
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "pytorch_model-00007-of-00007.bin",
7
+ "model.embed_tokens.weight": "pytorch_model-00001-of-00007.bin",
8
+ "model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
9
+ "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00007.bin",
10
+ "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00007.bin",
11
+ "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00007.bin",
12
+ "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
13
+ "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00007.bin",
14
+ "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00007.bin",
15
+ "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00007.bin",
16
+ "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00007.bin",
17
+ "model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
18
+ "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00007.bin",
19
+ "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00007.bin",
20
+ "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00007.bin",
21
+ "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
22
+ "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00007.bin",
23
+ "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00007.bin",
24
+ "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00007.bin",
25
+ "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00007.bin",
26
+ "model.layers.10.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
27
+ "model.layers.10.mlp.down_proj.weight": "pytorch_model-00002-of-00007.bin",
28
+ "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00002-of-00007.bin",
29
+ "model.layers.10.mlp.up_proj.weight": "pytorch_model-00002-of-00007.bin",
30
+ "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
31
+ "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00002-of-00007.bin",
32
+ "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00002-of-00007.bin",
33
+ "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00002-of-00007.bin",
34
+ "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00002-of-00007.bin",
35
+ "model.layers.11.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
36
+ "model.layers.11.mlp.down_proj.weight": "pytorch_model-00002-of-00007.bin",
37
+ "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00002-of-00007.bin",
38
+ "model.layers.11.mlp.up_proj.weight": "pytorch_model-00002-of-00007.bin",
39
+ "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
40
+ "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00002-of-00007.bin",
41
+ "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00002-of-00007.bin",
42
+ "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00002-of-00007.bin",
43
+ "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00002-of-00007.bin",
44
+ "model.layers.12.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
45
+ "model.layers.12.mlp.down_proj.weight": "pytorch_model-00002-of-00007.bin",
46
+ "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00002-of-00007.bin",
47
+ "model.layers.12.mlp.up_proj.weight": "pytorch_model-00002-of-00007.bin",
48
+ "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
49
+ "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00002-of-00007.bin",
50
+ "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00002-of-00007.bin",
51
+ "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00002-of-00007.bin",
52
+ "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00002-of-00007.bin",
53
+ "model.layers.13.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
54
+ "model.layers.13.mlp.down_proj.weight": "pytorch_model-00002-of-00007.bin",
55
+ "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00002-of-00007.bin",
56
+ "model.layers.13.mlp.up_proj.weight": "pytorch_model-00002-of-00007.bin",
57
+ "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
58
+ "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00002-of-00007.bin",
59
+ "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00002-of-00007.bin",
60
+ "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00002-of-00007.bin",
61
+ "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00002-of-00007.bin",
62
+ "model.layers.14.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
63
+ "model.layers.14.mlp.down_proj.weight": "pytorch_model-00002-of-00007.bin",
64
+ "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00002-of-00007.bin",
65
+ "model.layers.14.mlp.up_proj.weight": "pytorch_model-00002-of-00007.bin",
66
+ "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
67
+ "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00002-of-00007.bin",
68
+ "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00002-of-00007.bin",
69
+ "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00002-of-00007.bin",
70
+ "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00002-of-00007.bin",
71
+ "model.layers.15.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
72
+ "model.layers.15.mlp.down_proj.weight": "pytorch_model-00002-of-00007.bin",
73
+ "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00002-of-00007.bin",
74
+ "model.layers.15.mlp.up_proj.weight": "pytorch_model-00002-of-00007.bin",
75
+ "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
76
+ "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00002-of-00007.bin",
77
+ "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00002-of-00007.bin",
78
+ "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00002-of-00007.bin",
79
+ "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00002-of-00007.bin",
80
+ "model.layers.16.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
81
+ "model.layers.16.mlp.down_proj.weight": "pytorch_model-00002-of-00007.bin",
82
+ "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00002-of-00007.bin",
83
+ "model.layers.16.mlp.up_proj.weight": "pytorch_model-00002-of-00007.bin",
84
+ "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
85
+ "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00002-of-00007.bin",
86
+ "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00002-of-00007.bin",
87
+ "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00002-of-00007.bin",
88
+ "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00002-of-00007.bin",
89
+ "model.layers.17.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
90
+ "model.layers.17.mlp.down_proj.weight": "pytorch_model-00002-of-00007.bin",
91
+ "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00002-of-00007.bin",
92
+ "model.layers.17.mlp.up_proj.weight": "pytorch_model-00002-of-00007.bin",
93
+ "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
94
+ "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00002-of-00007.bin",
95
+ "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00002-of-00007.bin",
96
+ "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00002-of-00007.bin",
97
+ "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00002-of-00007.bin",
98
+ "model.layers.18.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
99
+ "model.layers.18.mlp.down_proj.weight": "pytorch_model-00003-of-00007.bin",
100
+ "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00003-of-00007.bin",
101
+ "model.layers.18.mlp.up_proj.weight": "pytorch_model-00003-of-00007.bin",
102
+ "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
103
+ "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00002-of-00007.bin",
104
+ "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00003-of-00007.bin",
105
+ "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00002-of-00007.bin",
106
+ "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00002-of-00007.bin",
107
+ "model.layers.19.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
108
+ "model.layers.19.mlp.down_proj.weight": "pytorch_model-00003-of-00007.bin",
109
+ "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00003-of-00007.bin",
110
+ "model.layers.19.mlp.up_proj.weight": "pytorch_model-00003-of-00007.bin",
111
+ "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
112
+ "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00003-of-00007.bin",
113
+ "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00003-of-00007.bin",
114
+ "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00003-of-00007.bin",
115
+ "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00003-of-00007.bin",
116
+ "model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
117
+ "model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00007.bin",
118
+ "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00007.bin",
119
+ "model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00007.bin",
120
+ "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
121
+ "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00007.bin",
122
+ "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00007.bin",
123
+ "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00007.bin",
124
+ "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00007.bin",
125
+ "model.layers.20.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
126
+ "model.layers.20.mlp.down_proj.weight": "pytorch_model-00003-of-00007.bin",
127
+ "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00003-of-00007.bin",
128
+ "model.layers.20.mlp.up_proj.weight": "pytorch_model-00003-of-00007.bin",
129
+ "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
130
+ "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00003-of-00007.bin",
131
+ "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00003-of-00007.bin",
132
+ "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00003-of-00007.bin",
133
+ "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00003-of-00007.bin",
134
+ "model.layers.21.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
135
+ "model.layers.21.mlp.down_proj.weight": "pytorch_model-00003-of-00007.bin",
136
+ "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00003-of-00007.bin",
137
+ "model.layers.21.mlp.up_proj.weight": "pytorch_model-00003-of-00007.bin",
138
+ "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
139
+ "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00003-of-00007.bin",
140
+ "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00003-of-00007.bin",
141
+ "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00003-of-00007.bin",
142
+ "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00003-of-00007.bin",
143
+ "model.layers.22.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
144
+ "model.layers.22.mlp.down_proj.weight": "pytorch_model-00003-of-00007.bin",
145
+ "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00003-of-00007.bin",
146
+ "model.layers.22.mlp.up_proj.weight": "pytorch_model-00003-of-00007.bin",
147
+ "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
148
+ "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00003-of-00007.bin",
149
+ "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00003-of-00007.bin",
150
+ "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00003-of-00007.bin",
151
+ "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00003-of-00007.bin",
152
+ "model.layers.23.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
153
+ "model.layers.23.mlp.down_proj.weight": "pytorch_model-00003-of-00007.bin",
154
+ "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00003-of-00007.bin",
155
+ "model.layers.23.mlp.up_proj.weight": "pytorch_model-00003-of-00007.bin",
156
+ "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
157
+ "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00003-of-00007.bin",
158
+ "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00003-of-00007.bin",
159
+ "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00003-of-00007.bin",
160
+ "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00003-of-00007.bin",
161
+ "model.layers.24.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
162
+ "model.layers.24.mlp.down_proj.weight": "pytorch_model-00003-of-00007.bin",
163
+ "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00003-of-00007.bin",
164
+ "model.layers.24.mlp.up_proj.weight": "pytorch_model-00003-of-00007.bin",
165
+ "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
166
+ "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00003-of-00007.bin",
167
+ "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00003-of-00007.bin",
168
+ "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00003-of-00007.bin",
169
+ "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00003-of-00007.bin",
170
+ "model.layers.25.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
171
+ "model.layers.25.mlp.down_proj.weight": "pytorch_model-00003-of-00007.bin",
172
+ "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00003-of-00007.bin",
173
+ "model.layers.25.mlp.up_proj.weight": "pytorch_model-00003-of-00007.bin",
174
+ "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
175
+ "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00003-of-00007.bin",
176
+ "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00003-of-00007.bin",
177
+ "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00003-of-00007.bin",
178
+ "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00003-of-00007.bin",
179
+ "model.layers.26.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
180
+ "model.layers.26.mlp.down_proj.weight": "pytorch_model-00003-of-00007.bin",
181
+ "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00003-of-00007.bin",
182
+ "model.layers.26.mlp.up_proj.weight": "pytorch_model-00003-of-00007.bin",
183
+ "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
184
+ "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00003-of-00007.bin",
185
+ "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00003-of-00007.bin",
186
+ "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00003-of-00007.bin",
187
+ "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00003-of-00007.bin",
188
+ "model.layers.27.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
189
+ "model.layers.27.mlp.down_proj.weight": "pytorch_model-00004-of-00007.bin",
190
+ "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00003-of-00007.bin",
191
+ "model.layers.27.mlp.up_proj.weight": "pytorch_model-00004-of-00007.bin",
192
+ "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
193
+ "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00003-of-00007.bin",
194
+ "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00003-of-00007.bin",
195
+ "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00003-of-00007.bin",
196
+ "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00003-of-00007.bin",
197
+ "model.layers.28.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
198
+ "model.layers.28.mlp.down_proj.weight": "pytorch_model-00004-of-00007.bin",
199
+ "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00004-of-00007.bin",
200
+ "model.layers.28.mlp.up_proj.weight": "pytorch_model-00004-of-00007.bin",
201
+ "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
202
+ "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00004-of-00007.bin",
203
+ "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00004-of-00007.bin",
204
+ "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00004-of-00007.bin",
205
+ "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00004-of-00007.bin",
206
+ "model.layers.29.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
207
+ "model.layers.29.mlp.down_proj.weight": "pytorch_model-00004-of-00007.bin",
208
+ "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00004-of-00007.bin",
209
+ "model.layers.29.mlp.up_proj.weight": "pytorch_model-00004-of-00007.bin",
210
+ "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
211
+ "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00004-of-00007.bin",
212
+ "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00004-of-00007.bin",
213
+ "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00004-of-00007.bin",
214
+ "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00004-of-00007.bin",
215
+ "model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
216
+ "model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00007.bin",
217
+ "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00007.bin",
218
+ "model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00007.bin",
219
+ "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
220
+ "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00007.bin",
221
+ "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00007.bin",
222
+ "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00007.bin",
223
+ "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00007.bin",
224
+ "model.layers.30.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
225
+ "model.layers.30.mlp.down_proj.weight": "pytorch_model-00004-of-00007.bin",
226
+ "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00004-of-00007.bin",
227
+ "model.layers.30.mlp.up_proj.weight": "pytorch_model-00004-of-00007.bin",
228
+ "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
229
+ "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00004-of-00007.bin",
230
+ "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00004-of-00007.bin",
231
+ "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00004-of-00007.bin",
232
+ "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00004-of-00007.bin",
233
+ "model.layers.31.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
234
+ "model.layers.31.mlp.down_proj.weight": "pytorch_model-00004-of-00007.bin",
235
+ "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00004-of-00007.bin",
236
+ "model.layers.31.mlp.up_proj.weight": "pytorch_model-00004-of-00007.bin",
237
+ "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
238
+ "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00004-of-00007.bin",
239
+ "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00004-of-00007.bin",
240
+ "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00004-of-00007.bin",
241
+ "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00004-of-00007.bin",
242
+ "model.layers.32.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
243
+ "model.layers.32.mlp.down_proj.weight": "pytorch_model-00004-of-00007.bin",
244
+ "model.layers.32.mlp.gate_proj.weight": "pytorch_model-00004-of-00007.bin",
245
+ "model.layers.32.mlp.up_proj.weight": "pytorch_model-00004-of-00007.bin",
246
+ "model.layers.32.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
247
+ "model.layers.32.self_attn.k_proj.weight": "pytorch_model-00004-of-00007.bin",
248
+ "model.layers.32.self_attn.o_proj.weight": "pytorch_model-00004-of-00007.bin",
249
+ "model.layers.32.self_attn.q_proj.weight": "pytorch_model-00004-of-00007.bin",
250
+ "model.layers.32.self_attn.v_proj.weight": "pytorch_model-00004-of-00007.bin",
251
+ "model.layers.33.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
252
+ "model.layers.33.mlp.down_proj.weight": "pytorch_model-00004-of-00007.bin",
253
+ "model.layers.33.mlp.gate_proj.weight": "pytorch_model-00004-of-00007.bin",
254
+ "model.layers.33.mlp.up_proj.weight": "pytorch_model-00004-of-00007.bin",
255
+ "model.layers.33.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
256
+ "model.layers.33.self_attn.k_proj.weight": "pytorch_model-00004-of-00007.bin",
257
+ "model.layers.33.self_attn.o_proj.weight": "pytorch_model-00004-of-00007.bin",
258
+ "model.layers.33.self_attn.q_proj.weight": "pytorch_model-00004-of-00007.bin",
259
+ "model.layers.33.self_attn.v_proj.weight": "pytorch_model-00004-of-00007.bin",
260
+ "model.layers.34.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
261
+ "model.layers.34.mlp.down_proj.weight": "pytorch_model-00004-of-00007.bin",
262
+ "model.layers.34.mlp.gate_proj.weight": "pytorch_model-00004-of-00007.bin",
263
+ "model.layers.34.mlp.up_proj.weight": "pytorch_model-00004-of-00007.bin",
264
+ "model.layers.34.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
265
+ "model.layers.34.self_attn.k_proj.weight": "pytorch_model-00004-of-00007.bin",
266
+ "model.layers.34.self_attn.o_proj.weight": "pytorch_model-00004-of-00007.bin",
267
+ "model.layers.34.self_attn.q_proj.weight": "pytorch_model-00004-of-00007.bin",
268
+ "model.layers.34.self_attn.v_proj.weight": "pytorch_model-00004-of-00007.bin",
269
+ "model.layers.35.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
270
+ "model.layers.35.mlp.down_proj.weight": "pytorch_model-00004-of-00007.bin",
271
+ "model.layers.35.mlp.gate_proj.weight": "pytorch_model-00004-of-00007.bin",
272
+ "model.layers.35.mlp.up_proj.weight": "pytorch_model-00004-of-00007.bin",
273
+ "model.layers.35.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
274
+ "model.layers.35.self_attn.k_proj.weight": "pytorch_model-00004-of-00007.bin",
275
+ "model.layers.35.self_attn.o_proj.weight": "pytorch_model-00004-of-00007.bin",
276
+ "model.layers.35.self_attn.q_proj.weight": "pytorch_model-00004-of-00007.bin",
277
+ "model.layers.35.self_attn.v_proj.weight": "pytorch_model-00004-of-00007.bin",
278
+ "model.layers.36.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
279
+ "model.layers.36.mlp.down_proj.weight": "pytorch_model-00005-of-00007.bin",
280
+ "model.layers.36.mlp.gate_proj.weight": "pytorch_model-00004-of-00007.bin",
281
+ "model.layers.36.mlp.up_proj.weight": "pytorch_model-00004-of-00007.bin",
282
+ "model.layers.36.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
283
+ "model.layers.36.self_attn.k_proj.weight": "pytorch_model-00004-of-00007.bin",
284
+ "model.layers.36.self_attn.o_proj.weight": "pytorch_model-00004-of-00007.bin",
285
+ "model.layers.36.self_attn.q_proj.weight": "pytorch_model-00004-of-00007.bin",
286
+ "model.layers.36.self_attn.v_proj.weight": "pytorch_model-00004-of-00007.bin",
287
+ "model.layers.37.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
288
+ "model.layers.37.mlp.down_proj.weight": "pytorch_model-00005-of-00007.bin",
289
+ "model.layers.37.mlp.gate_proj.weight": "pytorch_model-00005-of-00007.bin",
290
+ "model.layers.37.mlp.up_proj.weight": "pytorch_model-00005-of-00007.bin",
291
+ "model.layers.37.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
292
+ "model.layers.37.self_attn.k_proj.weight": "pytorch_model-00005-of-00007.bin",
293
+ "model.layers.37.self_attn.o_proj.weight": "pytorch_model-00005-of-00007.bin",
294
+ "model.layers.37.self_attn.q_proj.weight": "pytorch_model-00005-of-00007.bin",
295
+ "model.layers.37.self_attn.v_proj.weight": "pytorch_model-00005-of-00007.bin",
296
+ "model.layers.38.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
297
+ "model.layers.38.mlp.down_proj.weight": "pytorch_model-00005-of-00007.bin",
298
+ "model.layers.38.mlp.gate_proj.weight": "pytorch_model-00005-of-00007.bin",
299
+ "model.layers.38.mlp.up_proj.weight": "pytorch_model-00005-of-00007.bin",
300
+ "model.layers.38.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
301
+ "model.layers.38.self_attn.k_proj.weight": "pytorch_model-00005-of-00007.bin",
302
+ "model.layers.38.self_attn.o_proj.weight": "pytorch_model-00005-of-00007.bin",
303
+ "model.layers.38.self_attn.q_proj.weight": "pytorch_model-00005-of-00007.bin",
304
+ "model.layers.38.self_attn.v_proj.weight": "pytorch_model-00005-of-00007.bin",
305
+ "model.layers.39.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
306
+ "model.layers.39.mlp.down_proj.weight": "pytorch_model-00005-of-00007.bin",
307
+ "model.layers.39.mlp.gate_proj.weight": "pytorch_model-00005-of-00007.bin",
308
+ "model.layers.39.mlp.up_proj.weight": "pytorch_model-00005-of-00007.bin",
309
+ "model.layers.39.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
310
+ "model.layers.39.self_attn.k_proj.weight": "pytorch_model-00005-of-00007.bin",
311
+ "model.layers.39.self_attn.o_proj.weight": "pytorch_model-00005-of-00007.bin",
312
+ "model.layers.39.self_attn.q_proj.weight": "pytorch_model-00005-of-00007.bin",
313
+ "model.layers.39.self_attn.v_proj.weight": "pytorch_model-00005-of-00007.bin",
314
+ "model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
315
+ "model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00007.bin",
316
+ "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00007.bin",
317
+ "model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00007.bin",
318
+ "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
319
+ "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00007.bin",
320
+ "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00007.bin",
321
+ "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00007.bin",
322
+ "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00007.bin",
323
+ "model.layers.40.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
324
+ "model.layers.40.mlp.down_proj.weight": "pytorch_model-00005-of-00007.bin",
325
+ "model.layers.40.mlp.gate_proj.weight": "pytorch_model-00005-of-00007.bin",
326
+ "model.layers.40.mlp.up_proj.weight": "pytorch_model-00005-of-00007.bin",
327
+ "model.layers.40.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
328
+ "model.layers.40.self_attn.k_proj.weight": "pytorch_model-00005-of-00007.bin",
329
+ "model.layers.40.self_attn.o_proj.weight": "pytorch_model-00005-of-00007.bin",
330
+ "model.layers.40.self_attn.q_proj.weight": "pytorch_model-00005-of-00007.bin",
331
+ "model.layers.40.self_attn.v_proj.weight": "pytorch_model-00005-of-00007.bin",
332
+ "model.layers.41.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
333
+ "model.layers.41.mlp.down_proj.weight": "pytorch_model-00005-of-00007.bin",
334
+ "model.layers.41.mlp.gate_proj.weight": "pytorch_model-00005-of-00007.bin",
335
+ "model.layers.41.mlp.up_proj.weight": "pytorch_model-00005-of-00007.bin",
336
+ "model.layers.41.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
337
+ "model.layers.41.self_attn.k_proj.weight": "pytorch_model-00005-of-00007.bin",
338
+ "model.layers.41.self_attn.o_proj.weight": "pytorch_model-00005-of-00007.bin",
339
+ "model.layers.41.self_attn.q_proj.weight": "pytorch_model-00005-of-00007.bin",
340
+ "model.layers.41.self_attn.v_proj.weight": "pytorch_model-00005-of-00007.bin",
341
+ "model.layers.42.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
342
+ "model.layers.42.mlp.down_proj.weight": "pytorch_model-00005-of-00007.bin",
343
+ "model.layers.42.mlp.gate_proj.weight": "pytorch_model-00005-of-00007.bin",
344
+ "model.layers.42.mlp.up_proj.weight": "pytorch_model-00005-of-00007.bin",
345
+ "model.layers.42.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
346
+ "model.layers.42.self_attn.k_proj.weight": "pytorch_model-00005-of-00007.bin",
347
+ "model.layers.42.self_attn.o_proj.weight": "pytorch_model-00005-of-00007.bin",
348
+ "model.layers.42.self_attn.q_proj.weight": "pytorch_model-00005-of-00007.bin",
349
+ "model.layers.42.self_attn.v_proj.weight": "pytorch_model-00005-of-00007.bin",
350
+ "model.layers.43.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
351
+ "model.layers.43.mlp.down_proj.weight": "pytorch_model-00005-of-00007.bin",
352
+ "model.layers.43.mlp.gate_proj.weight": "pytorch_model-00005-of-00007.bin",
353
+ "model.layers.43.mlp.up_proj.weight": "pytorch_model-00005-of-00007.bin",
354
+ "model.layers.43.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
355
+ "model.layers.43.self_attn.k_proj.weight": "pytorch_model-00005-of-00007.bin",
356
+ "model.layers.43.self_attn.o_proj.weight": "pytorch_model-00005-of-00007.bin",
357
+ "model.layers.43.self_attn.q_proj.weight": "pytorch_model-00005-of-00007.bin",
358
+ "model.layers.43.self_attn.v_proj.weight": "pytorch_model-00005-of-00007.bin",
359
+ "model.layers.44.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
360
+ "model.layers.44.mlp.down_proj.weight": "pytorch_model-00005-of-00007.bin",
361
+ "model.layers.44.mlp.gate_proj.weight": "pytorch_model-00005-of-00007.bin",
362
+ "model.layers.44.mlp.up_proj.weight": "pytorch_model-00005-of-00007.bin",
363
+ "model.layers.44.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
364
+ "model.layers.44.self_attn.k_proj.weight": "pytorch_model-00005-of-00007.bin",
365
+ "model.layers.44.self_attn.o_proj.weight": "pytorch_model-00005-of-00007.bin",
366
+ "model.layers.44.self_attn.q_proj.weight": "pytorch_model-00005-of-00007.bin",
367
+ "model.layers.44.self_attn.v_proj.weight": "pytorch_model-00005-of-00007.bin",
368
+ "model.layers.45.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
369
+ "model.layers.45.mlp.down_proj.weight": "pytorch_model-00005-of-00007.bin",
370
+ "model.layers.45.mlp.gate_proj.weight": "pytorch_model-00005-of-00007.bin",
371
+ "model.layers.45.mlp.up_proj.weight": "pytorch_model-00005-of-00007.bin",
372
+ "model.layers.45.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
373
+ "model.layers.45.self_attn.k_proj.weight": "pytorch_model-00005-of-00007.bin",
374
+ "model.layers.45.self_attn.o_proj.weight": "pytorch_model-00005-of-00007.bin",
375
+ "model.layers.45.self_attn.q_proj.weight": "pytorch_model-00005-of-00007.bin",
376
+ "model.layers.45.self_attn.v_proj.weight": "pytorch_model-00005-of-00007.bin",
377
+ "model.layers.46.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
378
+ "model.layers.46.mlp.down_proj.weight": "pytorch_model-00006-of-00007.bin",
379
+ "model.layers.46.mlp.gate_proj.weight": "pytorch_model-00006-of-00007.bin",
380
+ "model.layers.46.mlp.up_proj.weight": "pytorch_model-00006-of-00007.bin",
381
+ "model.layers.46.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
382
+ "model.layers.46.self_attn.k_proj.weight": "pytorch_model-00005-of-00007.bin",
383
+ "model.layers.46.self_attn.o_proj.weight": "pytorch_model-00006-of-00007.bin",
384
+ "model.layers.46.self_attn.q_proj.weight": "pytorch_model-00005-of-00007.bin",
385
+ "model.layers.46.self_attn.v_proj.weight": "pytorch_model-00005-of-00007.bin",
386
+ "model.layers.47.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
387
+ "model.layers.47.mlp.down_proj.weight": "pytorch_model-00006-of-00007.bin",
388
+ "model.layers.47.mlp.gate_proj.weight": "pytorch_model-00006-of-00007.bin",
389
+ "model.layers.47.mlp.up_proj.weight": "pytorch_model-00006-of-00007.bin",
390
+ "model.layers.47.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
391
+ "model.layers.47.self_attn.k_proj.weight": "pytorch_model-00006-of-00007.bin",
392
+ "model.layers.47.self_attn.o_proj.weight": "pytorch_model-00006-of-00007.bin",
393
+ "model.layers.47.self_attn.q_proj.weight": "pytorch_model-00006-of-00007.bin",
394
+ "model.layers.47.self_attn.v_proj.weight": "pytorch_model-00006-of-00007.bin",
395
+ "model.layers.48.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
396
+ "model.layers.48.mlp.down_proj.weight": "pytorch_model-00006-of-00007.bin",
397
+ "model.layers.48.mlp.gate_proj.weight": "pytorch_model-00006-of-00007.bin",
398
+ "model.layers.48.mlp.up_proj.weight": "pytorch_model-00006-of-00007.bin",
399
+ "model.layers.48.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
400
+ "model.layers.48.self_attn.k_proj.weight": "pytorch_model-00006-of-00007.bin",
401
+ "model.layers.48.self_attn.o_proj.weight": "pytorch_model-00006-of-00007.bin",
402
+ "model.layers.48.self_attn.q_proj.weight": "pytorch_model-00006-of-00007.bin",
403
+ "model.layers.48.self_attn.v_proj.weight": "pytorch_model-00006-of-00007.bin",
404
+ "model.layers.49.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
405
+ "model.layers.49.mlp.down_proj.weight": "pytorch_model-00006-of-00007.bin",
406
+ "model.layers.49.mlp.gate_proj.weight": "pytorch_model-00006-of-00007.bin",
407
+ "model.layers.49.mlp.up_proj.weight": "pytorch_model-00006-of-00007.bin",
408
+ "model.layers.49.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
409
+ "model.layers.49.self_attn.k_proj.weight": "pytorch_model-00006-of-00007.bin",
410
+ "model.layers.49.self_attn.o_proj.weight": "pytorch_model-00006-of-00007.bin",
411
+ "model.layers.49.self_attn.q_proj.weight": "pytorch_model-00006-of-00007.bin",
412
+ "model.layers.49.self_attn.v_proj.weight": "pytorch_model-00006-of-00007.bin",
413
+ "model.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
414
+ "model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00007.bin",
415
+ "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00007.bin",
416
+ "model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00007.bin",
417
+ "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
418
+ "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00007.bin",
419
+ "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00007.bin",
420
+ "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00007.bin",
421
+ "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00007.bin",
422
+ "model.layers.50.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
423
+ "model.layers.50.mlp.down_proj.weight": "pytorch_model-00006-of-00007.bin",
424
+ "model.layers.50.mlp.gate_proj.weight": "pytorch_model-00006-of-00007.bin",
425
+ "model.layers.50.mlp.up_proj.weight": "pytorch_model-00006-of-00007.bin",
426
+ "model.layers.50.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
427
+ "model.layers.50.self_attn.k_proj.weight": "pytorch_model-00006-of-00007.bin",
428
+ "model.layers.50.self_attn.o_proj.weight": "pytorch_model-00006-of-00007.bin",
429
+ "model.layers.50.self_attn.q_proj.weight": "pytorch_model-00006-of-00007.bin",
430
+ "model.layers.50.self_attn.v_proj.weight": "pytorch_model-00006-of-00007.bin",
431
+ "model.layers.51.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
432
+ "model.layers.51.mlp.down_proj.weight": "pytorch_model-00006-of-00007.bin",
433
+ "model.layers.51.mlp.gate_proj.weight": "pytorch_model-00006-of-00007.bin",
434
+ "model.layers.51.mlp.up_proj.weight": "pytorch_model-00006-of-00007.bin",
435
+ "model.layers.51.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
436
+ "model.layers.51.self_attn.k_proj.weight": "pytorch_model-00006-of-00007.bin",
437
+ "model.layers.51.self_attn.o_proj.weight": "pytorch_model-00006-of-00007.bin",
438
+ "model.layers.51.self_attn.q_proj.weight": "pytorch_model-00006-of-00007.bin",
439
+ "model.layers.51.self_attn.v_proj.weight": "pytorch_model-00006-of-00007.bin",
440
+ "model.layers.52.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
441
+ "model.layers.52.mlp.down_proj.weight": "pytorch_model-00006-of-00007.bin",
442
+ "model.layers.52.mlp.gate_proj.weight": "pytorch_model-00006-of-00007.bin",
443
+ "model.layers.52.mlp.up_proj.weight": "pytorch_model-00006-of-00007.bin",
444
+ "model.layers.52.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
445
+ "model.layers.52.self_attn.k_proj.weight": "pytorch_model-00006-of-00007.bin",
446
+ "model.layers.52.self_attn.o_proj.weight": "pytorch_model-00006-of-00007.bin",
447
+ "model.layers.52.self_attn.q_proj.weight": "pytorch_model-00006-of-00007.bin",
448
+ "model.layers.52.self_attn.v_proj.weight": "pytorch_model-00006-of-00007.bin",
449
+ "model.layers.53.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
450
+ "model.layers.53.mlp.down_proj.weight": "pytorch_model-00006-of-00007.bin",
451
+ "model.layers.53.mlp.gate_proj.weight": "pytorch_model-00006-of-00007.bin",
452
+ "model.layers.53.mlp.up_proj.weight": "pytorch_model-00006-of-00007.bin",
453
+ "model.layers.53.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
454
+ "model.layers.53.self_attn.k_proj.weight": "pytorch_model-00006-of-00007.bin",
455
+ "model.layers.53.self_attn.o_proj.weight": "pytorch_model-00006-of-00007.bin",
456
+ "model.layers.53.self_attn.q_proj.weight": "pytorch_model-00006-of-00007.bin",
457
+ "model.layers.53.self_attn.v_proj.weight": "pytorch_model-00006-of-00007.bin",
458
+ "model.layers.54.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
459
+ "model.layers.54.mlp.down_proj.weight": "pytorch_model-00006-of-00007.bin",
460
+ "model.layers.54.mlp.gate_proj.weight": "pytorch_model-00006-of-00007.bin",
461
+ "model.layers.54.mlp.up_proj.weight": "pytorch_model-00006-of-00007.bin",
462
+ "model.layers.54.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
463
+ "model.layers.54.self_attn.k_proj.weight": "pytorch_model-00006-of-00007.bin",
464
+ "model.layers.54.self_attn.o_proj.weight": "pytorch_model-00006-of-00007.bin",
465
+ "model.layers.54.self_attn.q_proj.weight": "pytorch_model-00006-of-00007.bin",
466
+ "model.layers.54.self_attn.v_proj.weight": "pytorch_model-00006-of-00007.bin",
467
+ "model.layers.55.input_layernorm.weight": "pytorch_model-00007-of-00007.bin",
468
+ "model.layers.55.mlp.down_proj.weight": "pytorch_model-00007-of-00007.bin",
469
+ "model.layers.55.mlp.gate_proj.weight": "pytorch_model-00006-of-00007.bin",
470
+ "model.layers.55.mlp.up_proj.weight": "pytorch_model-00007-of-00007.bin",
471
+ "model.layers.55.post_attention_layernorm.weight": "pytorch_model-00007-of-00007.bin",
472
+ "model.layers.55.self_attn.k_proj.weight": "pytorch_model-00006-of-00007.bin",
473
+ "model.layers.55.self_attn.o_proj.weight": "pytorch_model-00006-of-00007.bin",
474
+ "model.layers.55.self_attn.q_proj.weight": "pytorch_model-00006-of-00007.bin",
475
+ "model.layers.55.self_attn.v_proj.weight": "pytorch_model-00006-of-00007.bin",
476
+ "model.layers.56.input_layernorm.weight": "pytorch_model-00007-of-00007.bin",
477
+ "model.layers.56.mlp.down_proj.weight": "pytorch_model-00007-of-00007.bin",
478
+ "model.layers.56.mlp.gate_proj.weight": "pytorch_model-00007-of-00007.bin",
479
+ "model.layers.56.mlp.up_proj.weight": "pytorch_model-00007-of-00007.bin",
480
+ "model.layers.56.post_attention_layernorm.weight": "pytorch_model-00007-of-00007.bin",
481
+ "model.layers.56.self_attn.k_proj.weight": "pytorch_model-00007-of-00007.bin",
482
+ "model.layers.56.self_attn.o_proj.weight": "pytorch_model-00007-of-00007.bin",
483
+ "model.layers.56.self_attn.q_proj.weight": "pytorch_model-00007-of-00007.bin",
484
+ "model.layers.56.self_attn.v_proj.weight": "pytorch_model-00007-of-00007.bin",
485
+ "model.layers.57.input_layernorm.weight": "pytorch_model-00007-of-00007.bin",
486
+ "model.layers.57.mlp.down_proj.weight": "pytorch_model-00007-of-00007.bin",
487
+ "model.layers.57.mlp.gate_proj.weight": "pytorch_model-00007-of-00007.bin",
488
+ "model.layers.57.mlp.up_proj.weight": "pytorch_model-00007-of-00007.bin",
489
+ "model.layers.57.post_attention_layernorm.weight": "pytorch_model-00007-of-00007.bin",
490
+ "model.layers.57.self_attn.k_proj.weight": "pytorch_model-00007-of-00007.bin",
491
+ "model.layers.57.self_attn.o_proj.weight": "pytorch_model-00007-of-00007.bin",
492
+ "model.layers.57.self_attn.q_proj.weight": "pytorch_model-00007-of-00007.bin",
493
+ "model.layers.57.self_attn.v_proj.weight": "pytorch_model-00007-of-00007.bin",
494
+ "model.layers.58.input_layernorm.weight": "pytorch_model-00007-of-00007.bin",
495
+ "model.layers.58.mlp.down_proj.weight": "pytorch_model-00007-of-00007.bin",
496
+ "model.layers.58.mlp.gate_proj.weight": "pytorch_model-00007-of-00007.bin",
497
+ "model.layers.58.mlp.up_proj.weight": "pytorch_model-00007-of-00007.bin",
498
+ "model.layers.58.post_attention_layernorm.weight": "pytorch_model-00007-of-00007.bin",
499
+ "model.layers.58.self_attn.k_proj.weight": "pytorch_model-00007-of-00007.bin",
500
+ "model.layers.58.self_attn.o_proj.weight": "pytorch_model-00007-of-00007.bin",
501
+ "model.layers.58.self_attn.q_proj.weight": "pytorch_model-00007-of-00007.bin",
502
+ "model.layers.58.self_attn.v_proj.weight": "pytorch_model-00007-of-00007.bin",
503
+ "model.layers.59.input_layernorm.weight": "pytorch_model-00007-of-00007.bin",
504
+ "model.layers.59.mlp.down_proj.weight": "pytorch_model-00007-of-00007.bin",
505
+ "model.layers.59.mlp.gate_proj.weight": "pytorch_model-00007-of-00007.bin",
506
+ "model.layers.59.mlp.up_proj.weight": "pytorch_model-00007-of-00007.bin",
507
+ "model.layers.59.post_attention_layernorm.weight": "pytorch_model-00007-of-00007.bin",
508
+ "model.layers.59.self_attn.k_proj.weight": "pytorch_model-00007-of-00007.bin",
509
+ "model.layers.59.self_attn.o_proj.weight": "pytorch_model-00007-of-00007.bin",
510
+ "model.layers.59.self_attn.q_proj.weight": "pytorch_model-00007-of-00007.bin",
511
+ "model.layers.59.self_attn.v_proj.weight": "pytorch_model-00007-of-00007.bin",
512
+ "model.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
513
+ "model.layers.6.mlp.down_proj.weight": "pytorch_model-00001-of-00007.bin",
514
+ "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00007.bin",
515
+ "model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00007.bin",
516
+ "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
517
+ "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00007.bin",
518
+ "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00007.bin",
519
+ "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00007.bin",
520
+ "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00007.bin",
521
+ "model.layers.60.input_layernorm.weight": "pytorch_model-00007-of-00007.bin",
522
+ "model.layers.60.mlp.down_proj.weight": "pytorch_model-00007-of-00007.bin",
523
+ "model.layers.60.mlp.gate_proj.weight": "pytorch_model-00007-of-00007.bin",
524
+ "model.layers.60.mlp.up_proj.weight": "pytorch_model-00007-of-00007.bin",
525
+ "model.layers.60.post_attention_layernorm.weight": "pytorch_model-00007-of-00007.bin",
526
+ "model.layers.60.self_attn.k_proj.weight": "pytorch_model-00007-of-00007.bin",
527
+ "model.layers.60.self_attn.o_proj.weight": "pytorch_model-00007-of-00007.bin",
528
+ "model.layers.60.self_attn.q_proj.weight": "pytorch_model-00007-of-00007.bin",
529
+ "model.layers.60.self_attn.v_proj.weight": "pytorch_model-00007-of-00007.bin",
530
+ "model.layers.61.input_layernorm.weight": "pytorch_model-00007-of-00007.bin",
531
+ "model.layers.61.mlp.down_proj.weight": "pytorch_model-00007-of-00007.bin",
532
+ "model.layers.61.mlp.gate_proj.weight": "pytorch_model-00007-of-00007.bin",
533
+ "model.layers.61.mlp.up_proj.weight": "pytorch_model-00007-of-00007.bin",
534
+ "model.layers.61.post_attention_layernorm.weight": "pytorch_model-00007-of-00007.bin",
535
+ "model.layers.61.self_attn.k_proj.weight": "pytorch_model-00007-of-00007.bin",
536
+ "model.layers.61.self_attn.o_proj.weight": "pytorch_model-00007-of-00007.bin",
537
+ "model.layers.61.self_attn.q_proj.weight": "pytorch_model-00007-of-00007.bin",
538
+ "model.layers.61.self_attn.v_proj.weight": "pytorch_model-00007-of-00007.bin",
539
+ "model.layers.7.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
540
+ "model.layers.7.mlp.down_proj.weight": "pytorch_model-00001-of-00007.bin",
541
+ "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00001-of-00007.bin",
542
+ "model.layers.7.mlp.up_proj.weight": "pytorch_model-00001-of-00007.bin",
543
+ "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
544
+ "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00001-of-00007.bin",
545
+ "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00001-of-00007.bin",
546
+ "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00001-of-00007.bin",
547
+ "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00001-of-00007.bin",
548
+ "model.layers.8.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
549
+ "model.layers.8.mlp.down_proj.weight": "pytorch_model-00002-of-00007.bin",
550
+ "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00001-of-00007.bin",
551
+ "model.layers.8.mlp.up_proj.weight": "pytorch_model-00001-of-00007.bin",
552
+ "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
553
+ "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00001-of-00007.bin",
554
+ "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00001-of-00007.bin",
555
+ "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00001-of-00007.bin",
556
+ "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00001-of-00007.bin",
557
+ "model.layers.9.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
558
+ "model.layers.9.mlp.down_proj.weight": "pytorch_model-00002-of-00007.bin",
559
+ "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00002-of-00007.bin",
560
+ "model.layers.9.mlp.up_proj.weight": "pytorch_model-00002-of-00007.bin",
561
+ "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
562
+ "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00002-of-00007.bin",
563
+ "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00002-of-00007.bin",
564
+ "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00002-of-00007.bin",
565
+ "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00002-of-00007.bin",
566
+ "model.norm.weight": "pytorch_model-00007-of-00007.bin"
567
+ }
568
+ }
requirements.txt ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ numpy
2
+ pandas
3
+ einops
4
+ sentencepiece
5
+ deepspeed==0.9.3
6
+ transformers==4.33.2
7
+ accelerate==0.21.0
8
+ peft==0.4.0
9
+ BitsAndBytes==0.40.2
10
+ xformers==0.0.21
11
+ ujson
12
+ jsonlines
13
+ tiktoken
14
+ transformers_stream_generator
special_tokens_map.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|begin▁of▁sentence|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|end▁of▁sentence|>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|end▁of▁sentence|>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ }
23
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "bos_token": {
5
+ "__type": "AddedToken",
6
+ "content": "<|begin▁of▁sentence|>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "clean_up_tokenization_spaces": false,
13
+ "eos_token": {
14
+ "__type": "AddedToken",
15
+ "content": "<|end▁of▁sentence|>",
16
+ "lstrip": false,
17
+ "normalized": true,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "legacy": true,
22
+ "model_max_length": 16384,
23
+ "pad_token": {
24
+ "__type": "AddedToken",
25
+ "content": "<|end▁of▁sentence|>",
26
+ "lstrip": false,
27
+ "normalized": true,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ },
31
+ "sp_model_kwargs": {},
32
+ "unk_token": null,
33
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user' or message['role'] == 'human') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.index0 == 0 and system_message != false %}{% set content = '<s>system\n' + system_message + '\n' %}{% else %}{% set content = '' %}{% endif %}{% if message['role'] == 'user' or message['role'] == 'human' %}{{ content + '<s>human\n' + message['content'] + '\n' }}{% elif message['role'] == 'assistant' or message['role'] == 'bot' %}{{ '<s>bot\n' + message['content'] + '\n' + eos_token + '\n'}}{% else %}{{ raise_exception('Only user/human and assistant/bot roles are supported!') }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<s>bot\n' }}{% endif %}",
34
+ "tokenizer_class": "LlamaTokenizerFast"
35
+ }