LoneStriker commited on
Commit
731f728
1 Parent(s): c68be37

Upload folder using huggingface_hub

Browse files
LICENSE ADDED
@@ -0,0 +1,324 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Yi Series Models Community License Agreement
2
+ Version: 2.1
3
+ Date of Release: November 23, 2023
4
+
5
+ 1. Definition
6
+
7
+ “Agreement” refers to the terms and conditions defined in this Yi Series Models
8
+ Community License Agreement for the use, reproduction and distribution of Yi
9
+ Series Models.
10
+
11
+ “Model” refers to associated components (including checkpoints) developed based
12
+ on machine learning, including learned weights and parameters (including the
13
+ status of optimizer).
14
+
15
+ “Yi Series Models” refers to opensource models with different specifications and
16
+ capabilities named “Yi” provided by the Licensor, including Yi-6B, Yi-34B etc.
17
+
18
+ “Derivatives” refers to all modifications to Yi Series Models, work based on Yi
19
+ Series Models, or any other models created or initialized by transferring the
20
+ weights, parameters, activations, or output patterns of Yi Series Models to
21
+ other models to achieve similar performance, including but not limited to
22
+ methods that require using intermediate data representations or generating
23
+ synthetic data based on Yi Series Models to train other models.
24
+
25
+ “Licensor” refers to Beijing Lingyiwanwu Information Technology Co., Ltd.
26
+
27
+ “you” refers to an individual or legal entity that exercises the license granted
28
+ by this Agreement and/or uses the Yi Series Models for any purpose and in any
29
+ field of use.
30
+
31
+ “Third Party” refers to any individuals, legal entities or non-legal
32
+ organizations other than you.
33
+
34
+ “Distribute” refers to transmitting, copying, publishing, or otherwise sharing
35
+ the Yi Series Models with third parties, including providing the Yi Series
36
+ Models through electronic or other remote means (such as any SaaS software or
37
+ PaaS software accessed via API or web access).
38
+
39
+ “Commercial Purposes” refers to the use of the Yi Series Models, directly or
40
+ indirectly, for the operation, promotion, revenue generation, or any other
41
+ profit-making purposes for entities or individuals.
42
+
43
+ “Laws and Regulations” refers to the laws and administrative regulations of the
44
+ mainland of the People's Republic of China (for the purposes of this Agreement
45
+ only, excluding Hong Kong, Macau, and Taiwan).
46
+
47
+ “Personal Information” refers to various information related to identified or
48
+ identifiable natural persons recorded electronically or by other means,
49
+ excluding information that has been anonymized.
50
+
51
+ “Logo” refers to any trademark, service mark, trade name, domain name, website
52
+ name, or other distinctive branding marks.
53
+
54
+ 2. License and License Restrictions
55
+ The Licensor hereby grants you a non-exclusive, global, non-transferable,
56
+ non-sub-licensable, revocable, and royalty-free copyright license. You must
57
+ adhere to the following license restrictions:
58
+
59
+ 1) Your use of the Yi Series Models must comply with the Laws and Regulations as
60
+ well as applicable legal requirements of other countries/regions, and respect
61
+ social ethics and moral standards, including but not limited to, not using the
62
+ Yi Series Models for purposes prohibited by Laws and Regulations as well as
63
+ applicable legal requirements of other countries/regions, such as harming
64
+ national security, promoting terrorism, extremism, inciting ethnic or racial
65
+ hatred, discrimination, violence, or pornography, and spreading false harmful
66
+ information.
67
+
68
+ 2) You shall not, for military or unlawful purposes or in ways not allowed by
69
+ Laws and Regulations as well as applicable legal requirements of other
70
+ countries/regions, a) use, copy or Distribute the Yi Series Models, or b) create
71
+ complete or partial Derivatives of the Yi Series Models.
72
+
73
+ 3) Your use of the Yi Series Models (including using the output of the Yi Series
74
+ Models) and the creation of Derivatives must not infringe upon the legitimate
75
+ rights of any Third Party, including but not limited to the rights of personal
76
+ rights such as the right to likeness, reputation, and privacy, as well as
77
+ intellectual property rights such as copyrights, patents, trade secrets, and
78
+ other property rights.
79
+
80
+ 4) You must clearly attribute the source of the Yi Series Models to the Licensor
81
+ and provide a copy of this Agreement to any Third-Party users of the Yi Series
82
+ Models and Derivatives.
83
+
84
+ 5) If you modify the Yi Series Models to create Derivatives, you must clearly
85
+ indicate the substantial modifications made, and these modifications shall not
86
+ violate the license restrictions of this Agreement. You shall not enable,
87
+ assist, or in any way facilitate Third Parties to violate the license
88
+ restrictions of this Agreement.
89
+
90
+ If you plan to use the Yi Series Models and Derivatives for Commercial Purposes,
91
+ please refer to the Registration Form of Yi Series Models for Commercial Purposes
92
+ (“Registration Form”), as provided in Attachment 1 of the Yi Series Models
93
+ Commercial License Agreement (available at https://www.lingyiwanwu.com/yi-license)
94
+ and send completed Registration Form to the email: yi@01.ai to complete the
95
+ registration and obtain the license for Commercial Purposes. If you obtained the
96
+ license for Commercial Purposes and use the Yi Series Models and Derivatives for
97
+ Commercial Purposes, you must comply with the afore-mentioned license restrictions
98
+ and restrictions specified under the Yi Series Models Commercial License Agreement.
99
+
100
+
101
+ 3. Intellectual Property
102
+ The ownership of the Yi Series Models and their related intellectual property
103
+ rights is solely held by the Licensor.
104
+
105
+ In any circumstance, without the prior written consent of the Licensor, you are
106
+ not allowed to use any Logo associated with the Licensor. If your use of
107
+ Licensor's Logo in violation of this Agreement causes any losses to the Licensor
108
+ or others, you will bear full legal responsibility.
109
+
110
+
111
+ 4. Disclaimer and Limitation of Liability
112
+
113
+ The Yi Series Models are provided "AS IS." The Licensor does not provide any
114
+ express or implied warranties for the Yi Series Models, including but not
115
+ limited to stability, ownership, merchantability, non-infringement, or fitness
116
+ for a specific purpose of the Yi Series Models and their output results. You
117
+ assume all responsibilities for the risks and consequences arising from the use,
118
+ reproduction, distribution of the Yi Series Models, and the creation of
119
+ Derivatives.
120
+
121
+ The Licensor complies with Laws and Regulations at all stages of model training,
122
+ maintaining the legality, authenticity, accuracy, objectivity, and diversity of
123
+ data and algorithms. The Licensor is not liable for any direct, indirect,
124
+ incidental consequences, and other losses or damages related to your use,
125
+ reproduction, and distribution of the Yi Series Models, and the creation of
126
+ Derivatives under this Agreement. This includes but is not limited to:
127
+
128
+ 1) The Licensor is not responsible for data security risks resulting from your
129
+ use of the Yi Series Models.
130
+
131
+ 2) The Yi Series Models may contain Personal Information. When you use Yi Series
132
+ Models, you acknowledge that you are the data processing entity as defined under
133
+ the Laws and Regulations responsible for determining the processing methods and
134
+ purposes of Personal Information. You must comply with legal requirements for
135
+ processing any Personal Information that may be contained in the Yi Series
136
+ Models and assume the associated legal responsibilities, as well as the risks
137
+ and consequences of processing Personal Information.
138
+
139
+ 3) The Licensor is not liable for reputation risks arising from your use of the
140
+ Yi Series Models or the output results of the Yi Series Models.
141
+
142
+ 4) The Licensor is not liable for intellectual property risks associated with
143
+ your use of the Yi Series Models’ output results.
144
+
145
+ If your use, reproduction, distribution of the Yi Series Models, or the creation
146
+ of Derivatives result in losses to the Licensor, the Licensor has the right to
147
+ seek compensation from you. For any claims made by Third Parties against the
148
+ Licensor related to your use, reproduction, and distribution of the Yi Series
149
+ Models, or the creation of Derivatives, the Licensor has the right to demand
150
+ that you defend, compensate, and indemnify the Licensor and protect the Licensor
151
+ from harm.
152
+
153
+
154
+ 5. Dispute Resolution
155
+
156
+ The stipulation, effectiveness, interpretation, performance, modification, and
157
+ termination of the Agreement, the use, copy and Distribute of the Yi Series
158
+ Models, and dispute resolution associated with your use, copy and distribution
159
+ shall be governed by the laws of the mainland of the People's Republic of China
160
+ (for the purposes of this agreement only, excluding Hong Kong, Macau, and
161
+ Taiwan), and the application of conflict of laws is excluded.
162
+
163
+ Any disputes arising from the use, copy or distribution of the Yi Series Models
164
+ should first be resolved through amicable negotiations. If negotiations fail,
165
+ legal proceedings should be initiated in the People's Court at the location of
166
+ the Licensor.
167
+
168
+
169
+ 6. Effectiveness and Termination of the Agreement
170
+
171
+ Your use of the Yi Series Models signifies that you have read and agreed to be
172
+ bound by the terms of the Agreement. The Agreement becomes effective from the
173
+ date of your use of the Yi Series Models and will terminate from the date you
174
+ cease using the Yi Series Models. If you violate any terms or restrictions in
175
+ the Agreement, the Licensor reserves the right to terminate the Agreement.
176
+
177
+ Upon termination of the Agreement, you must immediately cease using the Yi
178
+ Series Models. Section 4, "Disclaimer and Limitation of Liability," and Section
179
+ 5, "Dispute Resolution," of this Agreement remain in effect after the
180
+ termination of this Agreement.
181
+
182
+
183
+ 7. Updates to the Agreement and Contact Information
184
+
185
+ The Licensor reserves the right to update the Agreement from time to time. The
186
+ latest version of the Agreement will be posted by the Licensor through
187
+ https://01.ai.
188
+
189
+ For any questions related to licensing and copyright, please contact the
190
+ Licensor at yi@01.ai.
191
+
192
+
193
+
194
+ Yi系列模型社区许可协议
195
+ 版本: 2.1
196
+ 发布日期: 2023年11月23日
197
+
198
+ 1. 定义
199
+
200
+ “协议”是指本协议中定义Yi系列模型使用、复制和分发的条款和条件。
201
+
202
+ “模型”是指任何附带的基于机器学习的组件(包括检查点),包括学习的权重、参数(包括优
203
+ 化器状态)。
204
+
205
+ “Yi系列模型”是指许可方开源的以Yi命名的不同规格、不同能力的模型,包括
206
+ Yi-6B、Yi-34B等。
207
+
208
+ “模型衍生品”是指对Yi系列模型的所有修改、基于Yi系列模型的工作,或通过将Yi系列模型
209
+ 的权重、参数、激活或输出模式转移到其他模型而创建或初始化的任何其他模型,以使其他
210
+ 模型的性能与Yi系列模型类似,包括但不限于需要使用中间数据表示的提取方法或基于Yi系
211
+ 列模型生成合成数据来训练其他模型的方法。
212
+
213
+ “许可方”是指北京零一万物信息技术有限公司。
214
+
215
+ “您”是指行使本协议授予的权限和/或出于任何目的和在任何使用领域使用Yi系列模型的个
216
+ 人或法人实体。
217
+
218
+ “第三方”是指您之外的任何个人、法人实体或非法人组织。
219
+
220
+ “分发”是指向第三方传输、复制、发布或以其他方式共享Yi系列模型,包括将Yi系列模型作
221
+ 为通过电子或其他远程方式(例如基于 API 或 Web 访问的任何 SaaS 软件或 PaaS 软
222
+ 件)提供。
223
+
224
+ “商业用途”是指使用Yi系列模型,直接或间接为实体或个人进行运营、推广或产生收入,或
225
+ 用于任何其他盈利目的。
226
+
227
+ “法律法规”是指中华人民共和国大陆地区(仅为本协议之目的,不包括香港、澳门和台湾)
228
+ 的法律及行政法规。
229
+
230
+ “个人信息”是指以电子或者其他方式记录的与已识别或者可识别的自然人有关的各种信息,
231
+ 不包括匿名化处理后的信息。
232
+
233
+ “标识” 是指任何商标、服务标记、商号、域名、网站名称或其他带有显著品牌特征的标记。
234
+
235
+
236
+ 2. 许可及许可限制
237
+
238
+ 许可方特此授予您非排他性、全球性、不可转让、不可再许可、可撤销、免版税的版权许可。
239
+ 您必须满足如下许可限制条件:
240
+
241
+ 1) 您对Yi系列模型的使用应遵守法律法规以及其他国家/地区适用的法律要求、尊重社会公
242
+ 德和伦理道德。包括但不限于您不得将Yi系列模型用作危害国家安全、宣扬恐怖主义、极端
243
+ 主义,宣扬民族及种族仇恨、歧视,暴力、色情,以及虚假有害信息等法律法规以及其他国
244
+ 家/地区适用的法律要求禁止的目的。
245
+
246
+ 2) 您不得出于军事或非法目的,或以法律法规以及其他国家/地区适用的法律要求所不允许
247
+ 的方式 a) 使用、复制、或分发Yi系列模型; 或 b) 创建Yi系列模型的全部或部分衍生品。
248
+
249
+ 3) 您对Yi系列模型的使用(包括使用Yi系列模型的输出)以及模型衍生品的创建不得侵犯
250
+ 任何第三方的合法权益,包括但不限于他人肖像权、名誉权、隐私权等人格权,著作权、专
251
+ 利权、商业秘密等知识产权,或其他财产权益。
252
+
253
+ 4) 您必须向Yi系列模型及Yi系列模型衍生品的任何第三方使用者明确Yi系列模型的来源为
254
+ 许可方并向其提供本协议的副本。
255
+
256
+ 5) 若您修改Yi系列模型得到模型衍生品,您必须以显著的方式说明修改的内容,且上述修
257
+ 改不得违反本协议的许可限制条件,也不能允许、协助或以其他方式使得第三方违反本协议
258
+ 中的许可限制条件。
259
+
260
+ 如果您计划将Yi系列模型及模型衍生品用作商业用途,请参见《Yi系列模型商用许可协议》
261
+ (参见:https://www.lingyiwanwu.com/yi-license)附件一《Yi系列模型商用登
262
+ 记表》(“登记表”)并将填写完毕的登记表发送至 yi@01.ai 邮箱完成登记即可获得商用
263
+ 许可。若您获得商用许可并将Yi系列模型及模型衍生品用作商业用途,您应满足许可方上述
264
+ 许可限制条件及《Yi系列模型商用许可协议》中的商业许可限制。
265
+
266
+ 3. 知识产权
267
+
268
+ Yi系列模型的所有权及其相关知识产权,由许可方单独所有。
269
+
270
+ 在任何情况下,未经许可方事先书面同意,您不得以任何方式使用许可方的任何标识。由于
271
+ 您违反本协议使用许可方的标识给许可方或他人造成损失的,由您承担全部法律责任。
272
+
273
+
274
+ 4. 免责声明及责任限制
275
+
276
+ Yi系列模型按“原样”提供。许可方不对Yi系列模型提供任何明示或暗示的保证,包括但不限
277
+ 于:模型及输出结果的稳定性、所有权、适销性、非侵权性、或特定用途适用性。您将对适
278
+ 用、复制及分发Yi系列模型以及创建模型衍生品所产生的风险与后果承担所有责任。
279
+
280
+ 许可方在模型训练的所有阶段都遵守法律法规,坚持维护数据和算法的合法、真实、准确、
281
+ 客观和多样性。许可方不对您根据本协议使用、复制及分发Yi系列模型,以及创建模��衍生
282
+ 品而产生或与之相关的任何直接、间接、附带的后果、以及其他损失或损害承担责任。包括
283
+ 但不限于:
284
+
285
+ 1) 许可方不承担您因使用Yi系列模型而导致的数据安全风险。
286
+
287
+ 2) Yi系列模型中可能包含个人信息。在您使用Yi系列模型的过程中,您承认您为法律法规
288
+ 定义下决定个人信息处理方式和目的的个人信息处理者。您应遵守法律法规要求处理Yi系列
289
+ 模型中可能包含的个人信息,并承担相应的法律责任,以及处理个人信息的风险和后果。
290
+
291
+ 3) 许可方不承担您使用Yi系列模型或模型输出结果而产生的声誉风险。
292
+
293
+ 4) 许可方不承担您使用Yi系列模型的输出结果涉及的知识产权风险。
294
+
295
+ 若由于您对Yi系列模型的使用、复制或分发,或者创建模型衍生品而导致许可方遭受损失,
296
+ 许可方有权要求您对许可方的损失进行赔偿。对于任何第三方向许可方提出的因您使用、复
297
+ 制或分发Yi系列模型或创建模型衍生品行为的相关索赔,许可方有权要求您为许可方进行辩
298
+ 护、赔偿并使许可方免受损害。
299
+
300
+
301
+ 5. 争议解决
302
+
303
+ 协议的订立、效力、解释、履行、修改和终止,使用、复制和分发Yi系列模型以及争议解决
304
+ 均适用中华人民共和国大陆地区(仅为本协议之目的,不包括香港、澳门和台湾)法律,并
305
+ 排除冲突法的适用。
306
+
307
+ 因使用、复制和分发Yi系列模型而发生的任何争议,各方应首先通过友好协商的方式加以解
308
+ 决。协商不成时,应向许可方所在地人民法院提起诉讼。
309
+
310
+
311
+ 6. 协议的生效及终止
312
+
313
+ 您使用Yi系列模型即表示您已阅读并同意接受协议的约束。协议自您使用Yi系列模型之日起
314
+ 生效并将在您停止使用Yi系列模型之日起终止。若您违反协议中的任何条款或限制,许可方
315
+ 有权终止协议。
316
+
317
+ 若协议终止,您需立即停止使用Yi系列模型。本协议第4条“免责声明及责任限制”及第5条
318
+ “争议解决”在协议终止后仍有效。
319
+
320
+
321
+ 7. 协议更新及联系方式
322
+
323
+ 许可方有权对协议进行不时更新。许可方将通过 https://01.ai 公布协议最新版本。有关
324
+ 许可和版权的任何问题,请通过 yi@01.ai 与许可方联系。
README.md CHANGED
@@ -2,93 +2,1260 @@
2
  license: other
3
  license_name: yi-license
4
  license_link: LICENSE
 
 
 
 
 
 
 
 
 
 
5
  ---
 
6
  <div align="center">
7
 
8
- <img src="./Yi.svg" width="200px">
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9
 
10
  </div>
11
 
12
- ## Introduction
13
-
14
- The **Yi** series models are large language models trained from scratch by
15
- developers at [01.AI](https://01.ai/). The first public release contains two
16
- bilingual(English/Chinese) base models with the parameter sizes of 6B([`Yi-6B`](https://huggingface.co/01-ai/Yi-6B))
17
- and 34B([`Yi-34B`](https://huggingface.co/01-ai/Yi-34B)). Both of them are trained
18
- with 4K sequence length and can be extended to 32K during inference time.
19
- The [`Yi-6B-200K`](https://huggingface.co/01-ai/Yi-6B-200K)
20
- and [`Yi-34B-200K`](https://huggingface.co/01-ai/Yi-34B-200K) are base model with
21
- 200K context length.
22
-
23
- ## News
24
-
25
- - 🎯 **2023/11/06**: The base model of [`Yi-6B-200K`](https://huggingface.co/01-ai/Yi-6B-200K)
26
- and [`Yi-34B-200K`](https://huggingface.co/01-ai/Yi-34B-200K) with 200K context length.
27
- - 🎯 **2023/11/02**: The base model of [`Yi-6B`](https://huggingface.co/01-ai/Yi-6B) and
28
- [`Yi-34B`](https://huggingface.co/01-ai/Yi-34B).
29
-
30
-
31
- ## Model Performance
32
-
33
- | Model | MMLU | CMMLU | C-Eval | GAOKAO | BBH | Common-sense Reasoning | Reading Comprehension | Math & Code |
34
- | :------------ | :------: | :------: | :------: | :------: | :------: | :--------------------: | :-------------------: | :---------: |
35
- | | 5-shot | 5-shot | 5-shot | 0-shot | 3-shot@1 | - | - | - |
36
- | LLaMA2-34B | 62.6 | - | - | - | 44.1 | 69.9 | 68.0 | 26.0 |
37
- | LLaMA2-70B | 68.9 | 53.3 | - | 49.8 | 51.2 | 71.9 | 69.4 | 36.8 |
38
- | Baichuan2-13B | 59.2 | 62.0 | 58.1 | 54.3 | 48.8 | 64.3 | 62.4 | 23.0 |
39
- | Qwen-14B | 66.3 | 71.0 | 72.1 | 62.5 | 53.4 | 73.3 | 72.5 | **39.8** |
40
- | Skywork-13B | 62.1 | 61.8 | 60.6 | 68.1 | 41.7 | 72.4 | 61.4 | 24.9 |
41
- | InternLM-20B | 62.1 | 59.0 | 58.8 | 45.5 | 52.5 | 78.3 | - | 30.4 |
42
- | Aquila-34B | 67.8 | 71.4 | 63.1 | - | - | - | - | - |
43
- | Falcon-180B | 70.4 | 58.0 | 57.8 | 59.0 | 54.0 | 77.3 | 68.8 | 34.0 |
44
- | Yi-6B | 63.2 | 75.5 | 72.0 | 72.2 | 42.8 | 72.3 | 68.7 | 19.8 |
45
- | Yi-6B-200K | 64.0 | 75.3 | 73.5 | 73.9 | 42.0 | 72.0 | 69.1 | 19.0 |
46
- | **Yi-34B** | **76.3** | **83.7** | 81.4 | 82.8 | **54.3** | **80.1** | 76.4 | 37.1 |
47
- | Yi-34B-200K | 76.1 | 83.6 | **81.9** | **83.4** | 52.7 | 79.7 | **76.6** | 36.3 |
48
-
49
- While benchmarking open-source models, we have observed a disparity between the
50
- results generated by our pipeline and those reported in public sources (e.g.
51
- OpenCompass). Upon conducting a more in-depth investigation of this difference,
52
- we have discovered that various models may employ different prompts,
53
- post-processing strategies, and sampling techniques, potentially resulting in
54
- significant variations in the outcomes. Our prompt and post-processing strategy
55
- remains consistent with the original benchmark, and greedy decoding is employed
56
- during evaluation without any post-processing for the generated content. For
57
- scores that were not reported by the original authors (including scores reported
58
- with different settings), we try to get results with our pipeline.
59
-
60
- To evaluate the model's capability extensively, we adopted the methodology
61
- outlined in Llama2. Specifically, we included PIQA, SIQA, HellaSwag, WinoGrande,
62
- ARC, OBQA, and CSQA to assess common sense reasoning. SquAD, QuAC, and BoolQ
63
- were incorporated to evaluate reading comprehension. CSQA was exclusively tested
64
- using a 7-shot setup, while all other tests were conducted with a 0-shot
65
- configuration. Additionally, we introduced GSM8K (8-shot@1), MATH (4-shot@1),
66
- HumanEval (0-shot@1), and MBPP (3-shot@1) under the category "Math & Code". Due
67
- to technical constraints, we did not test Falcon-180 on QuAC and OBQA; the score
68
- is derived by averaging the scores on the remaining tasks. Since the scores for
69
- these two tasks are generally lower than the average, we believe that
70
- Falcon-180B's performance was not underestimated.
71
-
72
- ## Usage
73
-
74
- Please visit our [github repository](https://github.com/01-ai/Yi) for general
75
- guidance on how to use this model.
76
-
77
- ## Disclaimer
78
-
79
- Although we use data compliance checking algorithms during the training process
80
- to ensure the compliance of the trained model to the best of our ability, due to
81
- the complexity of the data and the diversity of language model usage scenarios,
82
- we cannot guarantee that the model will generate correct and reasonable output
83
- in all scenarios. Please be aware that there is still a risk of the model
84
- producing problematic outputs. We will not be responsible for any risks and
85
- issues resulting from misuse, misguidance, illegal usage, and related
86
- misinformation, as well as any associated data security concerns.
87
-
88
- ## License
89
-
90
- The Yi series models are fully open for academic research and free commercial
91
- usage with permission via applications. All usage must adhere to the [Model
92
- License Agreement 2.0](https://huggingface.co/01-ai/Yi-6B-200K/blob/main/LICENSE). To
93
- apply for the official commercial license, please contact us
94
- ([yi@01.ai](mailto:yi@01.ai)).
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  license: other
3
  license_name: yi-license
4
  license_link: LICENSE
5
+ widget:
6
+ - example_title: "Yi-34B-Chat"
7
+ text: "hi"
8
+ output:
9
+ text: " Hello! How can I assist you today?"
10
+ - example_title: "Yi-34B"
11
+ text: "There's a place where time stands still. A place of breath taking wonder, but also"
12
+ output:
13
+ text: " an eerie sense that something is just not right…\nBetween the two worlds lies The Forgotten Kingdom - home to creatures long since thought extinct and ancient magic so strong it defies belief! Only here can you find what has been lost for centuries: An Elixir Of Life which will restore youth and vitality if only those who seek its power are brave enough to face up against all manner of dangers lurking in this mysterious land! But beware; some say there may even exist powerful entities beyond our comprehension whose intentions towards humanity remain unclear at best ---- they might want nothing more than destruction itself rather then anything else from their quest after immortality (and maybe someone should tell them about modern medicine)? In any event though – one thing remains true regardless : whether or not success comes easy depends entirely upon how much effort we put into conquering whatever challenges lie ahead along with having faith deep down inside ourselves too ;) So let’s get started now shall We?"
14
+ pipeline_tag: text-generation
15
  ---
16
+
17
  <div align="center">
18
 
19
+ <picture>
20
+ <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/01-ai/Yi/main/assets/img/Yi_logo_icon_dark.svg" width="200px">
21
+ <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/01-ai/Yi/main/assets/img/Yi_logo_icon_light.svg" width="200px">
22
+ <img alt="specify theme context for images" src="https://raw.githubusercontent.com/01-ai/Yi/main/assets/img/Yi_logo_icon_light.svg">
23
+ </picture>
24
+
25
+ </br>
26
+ </br>
27
+
28
+ <div style="display: inline-block;">
29
+ <a href="https://github.com/01-ai/Yi/actions/workflows/build_docker_image.yml">
30
+ <img src="https://github.com/01-ai/Yi/actions/workflows/build_docker_image.yml/badge.svg">
31
+ </a>
32
+ </div>
33
+
34
+ <div style="display: inline-block;">
35
+ <a href="https://github.com/01-ai/Yi/blob/main/LICENSE">
36
+ <img src="https://img.shields.io/badge/Code_License-Apache_2.0-lightblue">
37
+ </a>
38
+ </div>
39
+
40
+ <div style="display: inline-block;">
41
+ <a href="https://github.com/01-ai/Yi/blob/main/MODEL_LICENSE_AGREEMENT.txt">
42
+ <img src="https://img.shields.io/badge/Model_License-Yi_License-lightblue">
43
+ </a>
44
+ </div>
45
+
46
+ <div style="display: inline-block;">
47
+ <a href="mailto:oss@01.ai">
48
+ <img src="https://img.shields.io/badge/✉️-yi@01.ai-FFE01B">
49
+ </a>
50
+ </div>
51
 
52
  </div>
53
 
54
+ <div align="center">
55
+ <h3 align="center">Building the Next Generation of Open-Source and Bilingual LLMs</h3>
56
+ </div>
57
+
58
+ <p align="center">
59
+ 🤗 <a href="https://huggingface.co/01-ai" target="_blank">Hugging Face</a> • 🤖 <a href="https://www.modelscope.cn/organization/01ai/" target="_blank">ModelScope</a> ✡️ <a href="https://wisemodel.cn/organization/01.AI" target="_blank">WiseModel</a>
60
+ </p>
61
+
62
+ <p align="center">
63
+ 👋 Join us 💬 <a href="https://github.com/01-ai/Yi/issues/43#issuecomment-1827285245" target="_blank"> WeChat (Chinese) </a>!
64
+ </p>
65
+
66
+
67
+ <!-- DO NOT REMOVE ME -->
68
+
69
+ <hr>
70
+
71
+ <details open>
72
+ <summary></b>📕 Table of Contents</b></summary>
73
+
74
+ - [What is Yi?](#what-is-yi)
75
+ - [Introduction](#introduction)
76
+ - [Models](#models)
77
+ - [Chat models](#chat-models)
78
+ - [Base models](#base-models)
79
+ - [Other info](#other-info)
80
+ - [News](#news)
81
+ - [How to use Yi?](#how-to-use-yi)
82
+ - [Quick start](#quick-start)
83
+ - [Choose your path](#choose-your-path)
84
+ - [pip](#quick-start---pip)
85
+ - [docker](#quick-start---docker)
86
+ - [llama.cpp](#quick-start---llamacpp)
87
+ - [conda-lock](#quick-start---conda-lock)
88
+ - [Web demo](#web-demo)
89
+ - [Fine-tuning](#fine-tuning)
90
+ - [Quantization](#quantization)
91
+ - [Deployment](#deployment)
92
+ - [Learning hub](#learning-hub)
93
+ - [Why Yi?](#why-yi)
94
+ - [Ecosystem](#ecosystem)
95
+ - [Upstream](#upstream)
96
+ - [Downstream](#downstream)
97
+ - [Serving](#serving)
98
+ - [Quantization](#quantization-1)
99
+ - [Fine-tuning](#fine-tuning-1)
100
+ - [API](#api)
101
+ - [Benchmarks](#benchmarks)
102
+ - [Base model performance](#base-model-performance)
103
+ - [Chat model performance](#chat-model-performance)
104
+ - [Tech report](#tech-report)
105
+ - [Citation](#citation)
106
+ - [Who can use Yi?](#who-can-use-yi)
107
+ - [Misc.](#misc)
108
+ - [Acknowledgements](#acknowledgments)
109
+ - [Disclaimer](#disclaimer)
110
+ - [License](#license)
111
+
112
+ </details>
113
+
114
+ <hr>
115
+
116
+ # What is Yi?
117
+
118
+ ## Introduction
119
+
120
+ - 🤖 The Yi series models are the next generation of open-source large language models trained from scratch by [01.AI](https://01.ai/).
121
+
122
+ - 🙌 Targeted as a bilingual language model and trained on 3T multilingual corpus, the Yi series models become one of the strongest LLM worldwide, showing promise in language understanding, commonsense reasoning, reading comprehension, and more. For example,
123
+
124
+ - Yi-34B-Chat model **landed in second place (following GPT-4 Turbo)**, outperforming other LLMs (such as GPT-4, Mixtral, Claude) on the AlpacaEval Leaderboard (based on data available up to January 2024).
125
+
126
+ - Yi-34B model **ranked first among all existing open-source models** (such as Falcon-180B, Llama-70B, Claude) in **both English and Chinese** on various benchmarks, including Hugging Face Open LLM Leaderboard (pre-trained) and C-Eval (based on data available up to November 2023).
127
+
128
+ - 🙏 (Credits to Llama) Thanks to the Transformer and Llama open-source communities, as they reduce the efforts required to build from scratch and enable the utilization of the same tools within the AI ecosystem.
129
+
130
+ <details style="display: inline;"><summary> If you're interested in Yi's adoption of Llama architecture and license usage policy, see <span style="color: green;">Yi's relation with Llama.</span> ⬇️</summary> <ul> <br>
131
+
132
+ > 💡 TL;DR
133
+ >
134
+ > The Yi series models adopt the same model architecture as Llama but are **NOT** derivatives of Llama.
135
+
136
+ - Both Yi and Llama are based on the Transformer structure, which has been the standard architecture for large language models since 2018.
137
+
138
+ - Grounded in the Transformer architecture, Llama has become a new cornerstone for the majority of state-of-the-art open-source models due to its excellent stability, reliable convergence, and robust compatibility. This positions Llama as the recognized foundational framework for models including Yi.
139
+
140
+ - Thanks to the Transformer and Llama architectures, other models can leverage their power, reducing the effort required to build from scratch and enabling the utilization of the same tools within their ecosystems.
141
+
142
+ - However, the Yi series models are NOT derivatives of Llama, as they do not use Llama's weights.
143
+
144
+ - As Llama's structure is employed by the majority of open-source models, the key factors of determining model performance are training datasets, training pipelines, and training infrastructure.
145
+
146
+ - Developing in a unique and proprietary way, Yi has independently created its own high-quality training datasets, efficient training pipelines, and robust training infrastructure entirely from the ground up. This effort has led to excellent performance with Yi series models ranking just behind GPT4 and surpassing Llama on the [Alpaca Leaderboard in Dec 2023](https://tatsu-lab.github.io/alpaca_eval/).
147
+ </ul>
148
+ </details>
149
+
150
+ <p align="right"> [
151
+ <a href="#top">Back to top ⬆️ </a> ]
152
+ </p>
153
+
154
+ ## News
155
+
156
+ <details open>
157
+ <summary>🎯 <b>2024-03-08</b>: <a href="https://arxiv.org/abs/2403.04652">Yi Tech Report</a> is published! </summary>
158
+ </details>
159
+
160
+
161
+ <details open>
162
+ <summary>🔔 <b>2024-03-07</b>: The long text capability of the Yi-34B-200K has been enhanced. </summary>
163
+ <br>
164
+ In the "Needle-in-a-Haystack" test, the Yi-34B-200K's performance is improved by 10.5%, rising from 89.3% to an impressive 99.8%. We continue to pre-train the model on 5B tokens long-context data mixture and demonstrate a near-all-green performance.
165
+ </details>
166
+
167
+ <details open>
168
+ <summary>🎯 <b>2024-03-06</b>: The <code>Yi-9B</code> is open-sourced and available to the public.</summary>
169
+ <br>
170
+ <code>Yi-9B</code> stands out as the top performer among a range of similar-sized open-source models (including Mistral-7B, SOLAR-10.7B, Gemma-7B, DeepSeek-Coder-7B-Base-v1.5 and more), particularly excelling in code, math, common-sense reasoning, and reading comprehension.
171
+ </details>
172
+
173
+ <details open>
174
+ <summary>🎯 <b>2024-01-23</b>: The Yi-VL models, <code><a href="https://huggingface.co/01-ai/Yi-VL-34B">Yi-VL-34B</a></code> and <code><a href="https://huggingface.co/01-ai/Yi-VL-6B">Yi-VL-6B</a></code>, are open-sourced and available to the public.</summary>
175
+ <br>
176
+ <code><a href="https://huggingface.co/01-ai/Yi-VL-34B">Yi-VL-34B</a></code> has ranked <strong>first</strong> among all existing open-source models in the latest benchmarks, including <a href="https://arxiv.org/abs/2311.16502">MMMU</a> and <a href="https://arxiv.org/abs/2401.11944">CMMMU</a> (based on data available up to January 2024).</li>
177
+ </details>
178
+
179
+
180
+ <details>
181
+ <summary>🎯 <b>2023-11-23</b>: <a href="#chat-models">Chat models</a> are open-sourced and available to the public.</summary>
182
+ <br>This release contains two chat models based on previously released base models, two 8-bit models quantized by GPTQ, and two 4-bit models quantized by AWQ.
183
+
184
+ - `Yi-34B-Chat`
185
+ - `Yi-34B-Chat-4bits`
186
+ - `Yi-34B-Chat-8bits`
187
+ - `Yi-6B-Chat`
188
+ - `Yi-6B-Chat-4bits`
189
+ - `Yi-6B-Chat-8bits`
190
+
191
+ You can try some of them interactively at:
192
+
193
+ - [Hugging Face](https://huggingface.co/spaces/01-ai/Yi-34B-Chat)
194
+ - [Replicate](https://replicate.com/01-ai)
195
+ </details>
196
+
197
+ <details>
198
+ <summary>🔔 <b>2023-11-23</b>: The Yi Series Models Community License Agreement is updated to <a href="https://github.com/01-ai/Yi/blob/main/MODEL_LICENSE_AGREEMENT.txt">v2.1</a>.</summary>
199
+ </details>
200
+
201
+ <details>
202
+ <summary>🔥 <b>2023-11-08</b>: Invited test of Yi-34B chat model.</summary>
203
+ <br>Application form:
204
+
205
+ - [English](https://cn.mikecrm.com/l91ODJf)
206
+ - [Chinese](https://cn.mikecrm.com/gnEZjiQ)
207
+ </details>
208
+
209
+ <details>
210
+ <summary>🎯 <b>2023-11-05</b>: <a href="#base-models">The base models, </a><code>Yi-6B-200K</code> and <code>Yi-34B-200K</code>, are open-sourced and available to the public.</summary>
211
+ <br>This release contains two base models with the same parameter sizes as the previous
212
+ release, except that the context window is extended to 200K.
213
+ </details>
214
+
215
+ <details>
216
+ <summary>🎯 <b>2023-11-02</b>: <a href="#base-models">The base models, </a><code>Yi-6B</code> and <code>Yi-34B</code>, are open-sourced and available to the public.</summary>
217
+ <br>The first public release contains two bilingual (English/Chinese) base models
218
+ with the parameter sizes of 6B and 34B. Both of them are trained with 4K
219
+ sequence length and can be extended to 32K during inference time.
220
+
221
+ </details>
222
+
223
+ <p align="right"> [
224
+ <a href="#top">Back to top ⬆️ </a> ]
225
+ </p>
226
+
227
+ ## Models
228
+
229
+ Yi models come in multiple sizes and cater to different use cases. You can also fine-tune Yi models to meet your specific requirements.
230
+
231
+ If you want to deploy Yi models, make sure you meet the [software and hardware requirements](#deployment).
232
+
233
+ ### Chat models
234
+
235
+ | Model | Download
236
+ |---|---
237
+ Yi-34B-Chat | • [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-34B-Chat) • [🤖 ModelScope](https://www.modelscope.cn/models/01ai/Yi-34B-Chat/summary)
238
+ Yi-34B-Chat-4bits | • [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-34B-Chat-4bits) • [🤖 ModelScope](https://www.modelscope.cn/models/01ai/Yi-34B-Chat-4bits/summary)
239
+ Yi-34B-Chat-8bits | • [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-34B-Chat-8bits) • [🤖 ModelScope](https://www.modelscope.cn/models/01ai/Yi-34B-Chat-8bits/summary)
240
+ Yi-6B-Chat| • [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-6B-Chat) • [🤖 ModelScope](https://www.modelscope.cn/models/01ai/Yi-6B-Chat/summary)
241
+ Yi-6B-Chat-4bits | • [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-6B-Chat-4bits) • [🤖 ModelScope](https://www.modelscope.cn/models/01ai/Yi-6B-Chat-4bits/summary)
242
+ Yi-6B-Chat-8bits | • [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-6B-Chat-8bits) • [🤖 ModelScope](https://www.modelscope.cn/models/01ai/Yi-6B-Chat-8bits/summary)
243
+
244
+
245
+ <sub><sup> - 4-bit series models are quantized by AWQ. <br> - 8-bit series models are quantized by GPTQ <br> - All quantized models have a low barrier to use since they can be deployed on consumer-grade GPUs (e.g., 3090, 4090). </sup></sub>
246
+
247
+ ### Base models
248
+
249
+ | Model | Download |
250
+ |---|---|
251
+ Yi-34B| • [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-34B) • [🤖 ModelScope](https://www.modelscope.cn/models/01ai/Yi-34B/summary)
252
+ Yi-34B-200K|• [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-34B-200K) • [🤖 ModelScope](https://www.modelscope.cn/models/01ai/Yi-34B-200K/summary)
253
+ Yi-9B|• [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-9B)
254
+ Yi-9B-200K | • [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-9B-200K)
255
+ Yi-6B| • [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-6B) • [🤖 ModelScope](https://www.modelscope.cn/models/01ai/Yi-6B/summary)
256
+ Yi-6B-200K | • [🤗 Hugging Face](https://huggingface.co/01-ai/Yi-6B-200K) • [🤖 ModelScope](https://www.modelscope.cn/models/01ai/Yi-6B-200K/summary)
257
+
258
+ <sub><sup> - 200k is roughly equivalent to 400,000 Chinese characters. <br> - If you want to use the previous version of the Yi-34B-200K (released on Nov 5, 2023), run `git checkout 069cd341d60f4ce4b07ec394e82b79e94f656cf` to download the weight. </sup></sub>
259
+
260
+ ### Model info
261
+
262
+ - For chat and base models
263
+
264
+ Model | Intro | Default context window | Pretrained tokens | Training Data Date
265
+ |---|---|---|---|---
266
+ 6B series models |They are suitable for personal and academic use. | 4K | 3T | Up to June 2023
267
+ 9B model| It is the best at coding and math in the Yi series models.|4K | Yi-9B is continuously trained based on Yi-6B, using 0.8T tokens. | Up to June 2023
268
+ 34B series models | They are suitable for personal, academic, and commercial (particularly for small and medium-sized enterprises) purposes. It's a cost-effective solution that's affordable and equipped with emergent ability.|4K | 3T | Up to June 2023
269
+
270
+ - For chat models
271
+
272
+ <details style="display: inline;"><summary>For chat model limitations, see the explanations below. ⬇️</summary>
273
+ <ul>
274
+ <br>The released chat model has undergone exclusive training using Supervised Fine-Tuning (SFT). Compared to other standard chat models, our model produces more diverse responses, making it suitable for various downstream tasks, such as creative scenarios. Furthermore, this diversity is expected to enhance the likelihood of generating higher quality responses, which will be advantageous for subsequent Reinforcement Learning (RL) training.
275
+
276
+ <br>However, this higher diversity might amplify certain existing issues, including:
277
+ <li>Hallucination: This refers to the model generating factually incorrect or nonsensical information. With the model's responses being more varied, there's a higher chance of hallucination that are not based on accurate data or logical reasoning.</li>
278
+ <li>Non-determinism in re-generation: When attempting to regenerate or sample responses, inconsistencies in the outcomes may occur. The increased diversity can lead to varying results even under similar input conditions.</li>
279
+ <li>Cumulative Error: This occurs when errors in the model's responses compound over time. As the model generates more diverse responses, the likelihood of small inaccuracies building up into larger errors increases, especially in complex tasks like extended reasoning, mathematical problem-solving, etc.</li>
280
+ <li>To achieve more coherent and consistent responses, it is advisable to adjust generation configuration parameters such as temperature, top_p, or top_k. These adjustments can help in the balance between creativity and coherence in the model's outputs.</li>
281
+ </ul>
282
+ </details>
283
+
284
+ <p align="right"> [
285
+ <a href="#top">Back to top ⬆️ </a> ]
286
+ </p>
287
+
288
+
289
+ # How to use Yi?
290
+
291
+ - [Quick start](#quick-start)
292
+ - [Choose your path](#choose-your-path)
293
+ - [pip](#quick-start---pip)
294
+ - [docker](#quick-start---docker)
295
+ - [conda-lock](#quick-start---conda-lock)
296
+ - [llama.cpp](#quick-start---llamacpp)
297
+ - [Web demo](#web-demo)
298
+ - [Fine-tuning](#fine-tuning)
299
+ - [Quantization](#quantization)
300
+ - [Deployment](#deployment)
301
+ - [Learning hub](#learning-hub)
302
+
303
+ ## Quick start
304
+
305
+ Getting up and running with Yi models is simple with multiple choices available.
306
+
307
+ ### Choose your path
308
+
309
+ Select one of the following paths to begin your journey with Yi!
310
+
311
+ ![Quick start - Choose your path](https://github.com/01-ai/Yi/blob/main/assets/img/quick_start_path.png?raw=true)
312
+
313
+ #### 🎯 Deploy Yi locally
314
+
315
+ If you prefer to deploy Yi models locally,
316
+
317
+ - 🙋‍♀️ and you have **sufficient** resources (for example, NVIDIA A800 80GB), you can choose one of the following methods:
318
+ - [pip](#quick-start---pip)
319
+ - [Docker](#quick-start---docker)
320
+ - [conda-lock](#quick-start---conda-lock)
321
+
322
+ - 🙋‍♀️ and you have **limited** resources (for example, a MacBook Pro), you can use [llama.cpp](#quick-start---llamacpp).
323
+
324
+ #### 🎯 Not to deploy Yi locally
325
+
326
+ If you prefer not to deploy Yi models locally, you can explore Yi's capabilities using any of the following options.
327
+
328
+ ##### 🙋‍♀️ Run Yi with APIs
329
+
330
+ If you want to explore more features of Yi, you can adopt one of these methods:
331
+
332
+ - Yi APIs (Yi official)
333
+ - [Early access has been granted](https://x.com/01AI_Yi/status/1735728934560600536?s=20) to some applicants. Stay tuned for the next round of access!
334
+
335
+ - [Yi APIs](https://replicate.com/01-ai/yi-34b-chat/api?tab=nodejs) (Replicate)
336
+
337
+ ##### 🙋‍♀️ Run Yi in playground
338
+
339
+ If you want to chat with Yi with more customizable options (e.g., system prompt, temperature, repetition penalty, etc.), you can try one of the following options:
340
+
341
+ - [Yi-34B-Chat-Playground](https://platform.lingyiwanwu.com/prompt/playground) (Yi official)
342
+ - Access is available through a whitelist. Welcome to apply (fill out a form in [English](https://cn.mikecrm.com/l91ODJf) or [Chinese](https://cn.mikecrm.com/gnEZjiQ)).
343
+
344
+ - [Yi-34B-Chat-Playground](https://replicate.com/01-ai/yi-34b-chat) (Replicate)
345
+
346
+ ##### 🙋‍♀️ Chat with Yi
347
+
348
+ If you want to chat with Yi, you can use one of these online services, which offer a similar user experience:
349
+
350
+ - [Yi-34B-Chat](https://huggingface.co/spaces/01-ai/Yi-34B-Chat) (Yi official on Hugging Face)
351
+ - No registration is required.
352
+
353
+ - [Yi-34B-Chat](https://platform.lingyiwanwu.com/) (Yi official beta)
354
+ - Access is available through a whitelist. Welcome to apply (fill out a form in [English](https://cn.mikecrm.com/l91ODJf) or [Chinese](https://cn.mikecrm.com/gnEZjiQ)).
355
+
356
+ <p align="right"> [
357
+ <a href="#top">Back to top ⬆️ </a> ]
358
+ </p>
359
+
360
+ ### Quick start - pip
361
+
362
+ This tutorial guides you through every step of running **Yi-34B-Chat locally on an A800 (80G)** and then performing inference.
363
+
364
+ #### Step 0: Prerequisites
365
+
366
+ - Make sure Python 3.10 or a later version is installed.
367
+
368
+ - If you want to run other Yi models, see [software and hardware requirements](#deployment).
369
+
370
+ #### Step 1: Prepare your environment
371
+
372
+ To set up the environment and install the required packages, execute the following command.
373
+
374
+ ```bash
375
+ git clone https://github.com/01-ai/Yi.git
376
+ cd yi
377
+ pip install -r requirements.txt
378
+ ```
379
+
380
+ #### Step 2: Download the Yi model
381
+
382
+ You can download the weights and tokenizer of Yi models from the following sources:
383
+
384
+ - [Hugging Face](https://huggingface.co/01-ai)
385
+ - [ModelScope](https://www.modelscope.cn/organization/01ai/)
386
+ - [WiseModel](https://wisemodel.cn/organization/01.AI)
387
+
388
+ #### Step 3: Perform inference
389
+
390
+ You can perform inference with Yi chat or base models as below.
391
+
392
+ ##### Perform inference with Yi chat model
393
+
394
+ 1. Create a file named `quick_start.py` and copy the following content to it.
395
+
396
+ ```python
397
+ from transformers import AutoModelForCausalLM, AutoTokenizer
398
+
399
+ model_path = '<your-model-path>'
400
+
401
+ tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=False)
402
+
403
+ # Since transformers 4.35.0, the GPT-Q/AWQ model can be loaded using AutoModelForCausalLM.
404
+ model = AutoModelForCausalLM.from_pretrained(
405
+ model_path,
406
+ device_map="auto",
407
+ torch_dtype='auto'
408
+ ).eval()
409
+
410
+ # Prompt content: "hi"
411
+ messages = [
412
+ {"role": "user", "content": "hi"}
413
+ ]
414
+
415
+ input_ids = tokenizer.apply_chat_template(conversation=messages, tokenize=True, add_generation_prompt=True, return_tensors='pt')
416
+ output_ids = model.generate(input_ids.to('cuda'))
417
+ response = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)
418
+
419
+ # Model response: "Hello! How can I assist you today?"
420
+ print(response)
421
+ ```
422
+
423
+ 2. Run `quick_start.py`.
424
+
425
+ ```bash
426
+ python quick_start.py
427
+ ```
428
+
429
+ Then you can see an output similar to the one below. 🥳
430
+
431
+ ```bash
432
+ Hello! How can I assist you today?
433
+ ```
434
+
435
+ ##### Perform inference with Yi base model
436
+
437
+ - Yi-34B
438
+
439
+ The steps are similar to [pip - Perform inference with Yi chat model](#perform-inference-with-yi-chat-model).
440
+
441
+ You can use the existing file [`text_generation.py`](https://github.com/01-ai/Yi/tree/main/demo).
442
+
443
+ ```bash
444
+ python demo/text_generation.py --model <your-model-path>
445
+ ```
446
+
447
+ Then you can see an output similar to the one below. 🥳
448
+
449
+ <details>
450
+
451
+ <summary>Output. ⬇️ </summary>
452
+
453
+ <br>
454
+
455
+ **Prompt**: Let me tell you an interesting story about cat Tom and mouse Jerry,
456
+
457
+ **Generation**: Let me tell you an interesting story about cat Tom and mouse Jerry, which happened in my childhood. My father had a big house with two cats living inside it to kill mice. One day when I was playing at home alone, I found one of the tomcats lying on his back near our kitchen door, looking very much like he wanted something from us but couldn’t get up because there were too many people around him! He kept trying for several minutes before finally giving up...
458
+
459
+ </details>
460
+
461
+ - Yi-9B
462
+
463
+ Input
464
+
465
+ ```bash
466
+ from transformers import AutoModelForCausalLM, AutoTokenizer
467
+
468
+ MODEL_DIR = "01-ai/Yi-9B"
469
+ model = AutoModelForCausalLM.from_pretrained(MODEL_DIR, torch_dtype="auto")
470
+ tokenizer = AutoTokenizer.from_pretrained(MODEL_DIR, use_fast=False)
471
+
472
+ input_text = "# write the quick sort algorithm"
473
+ inputs = tokenizer(input_text, return_tensors="pt").to(model.device)
474
+ outputs = model.generate(**inputs, max_length=256)
475
+ print(tokenizer.decode(outputs[0], skip_special_tokens=True))
476
+ ```
477
+
478
+ Output
479
+
480
+ ```bash
481
+ # write the quick sort algorithm
482
+ def quick_sort(arr):
483
+ if len(arr) <= 1:
484
+ return arr
485
+ pivot = arr[len(arr) // 2]
486
+ left = [x for x in arr if x < pivot]
487
+ middle = [x for x in arr if x == pivot]
488
+ right = [x for x in arr if x > pivot]
489
+ return quick_sort(left) + middle + quick_sort(right)
490
+
491
+ # test the quick sort algorithm
492
+ print(quick_sort([3, 6, 8, 10, 1, 2, 1]))
493
+ ```
494
+
495
+ <p align="right"> [
496
+ <a href="#top">Back to top ⬆️ </a> ]
497
+ </p>
498
+
499
+ ### Quick start - Docker
500
+ <details>
501
+ <summary> Run Yi-34B-chat locally with Docker: a step-by-step guide. ⬇️</summary>
502
+ <br>This tutorial guides you through every step of running <strong>Yi-34B-Chat on an A800 GPU</strong> or <strong>4*4090</strong> locally and then performing inference.
503
+ <h4>Step 0: Prerequisites</h4>
504
+ <p>Make sure you've installed <a href="https://docs.docker.com/engine/install/?open_in_browser=true">Docker</a> and <a href="https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html">nvidia-container-toolkit</a>.</p>
505
+
506
+ <h4> Step 1: Start Docker </h4>
507
+ <pre><code>docker run -it --gpus all \
508
+ -v &lt;your-model-path&gt;: /models
509
+ ghcr.io/01-ai/yi:latest
510
+ </code></pre>
511
+ <p>Alternatively, you can pull the Yi Docker image from <code>registry.lingyiwanwu.com/ci/01-ai/yi:latest</code>.</p>
512
+
513
+ <h4>Step 2: Perform inference</h4>
514
+ <p>You can perform inference with Yi chat or base models as below.</p>
515
+
516
+ <h5>Perform inference with Yi chat model</h5>
517
+ <p>The steps are similar to <a href="#perform-inference-with-yi-chat-model">pip - Perform inference with Yi chat model</a>.</p>
518
+ <p><strong>Note</strong> that the only difference is to set <code>model_path = '&lt;your-model-mount-path&gt;'</code> instead of <code>model_path = '&lt;your-model-path&gt;'</code>.</p>
519
+ <h5>Perform inference with Yi base model</h5>
520
+ <p>The steps are similar to <a href="#perform-inference-with-yi-base-model">pip - Perform inference with Yi base model</a>.</p>
521
+ <p><strong>Note</strong> that the only difference is to set <code>--model &lt;your-model-mount-path&gt;'</code> instead of <code>model &lt;your-model-path&gt;</code>.</p>
522
+ </details>
523
+
524
+ ### Quick start - conda-lock
525
+
526
+ <details>
527
+ <summary>You can use <code><a href="https://github.com/conda/conda-lock">conda-lock</a></code> to generate fully reproducible lock files for conda environments. ⬇️</summary>
528
+ <br>
529
+ You can refer to <a href="https://github.com/01-ai/Yi/blob/ebba23451d780f35e74a780987ad377553134f68/conda-lock.yml">conda-lock.yml</a> for the exact versions of the dependencies. Additionally, you can utilize <code><a href="https://mamba.readthedocs.io/en/latest/user_guide/micromamba.html">micromamba</a></code> for installing these dependencies.
530
+ <br>
531
+ To install the dependencies, follow these steps:
532
+
533
+ 1. Install micromamba by following the instructions available <a href="https://mamba.readthedocs.io/en/latest/installation/micromamba-installation.html">here</a>.
534
+
535
+ 2. Execute <code>micromamba install -y -n yi -f conda-lock.yml</code> to create a conda environment named <code>yi</code> and install the necessary dependencies.
536
+ </details>
537
+
538
+
539
+ ### Quick start - llama.cpp
540
+ <details>
541
+ <summary> Run Yi-chat-6B-2bits locally with llama.cpp: a step-by-step guide. ⬇️</summary>
542
+ <br>This tutorial guides you through every step of running a quantized model (<a href="https://huggingface.co/XeIaso/yi-chat-6B-GGUF/tree/main">Yi-chat-6B-2bits</a>) locally and then performing inference.</p>
543
+
544
+ - [Step 0: Prerequisites](#step-0-prerequisites)
545
+ - [Step 1: Download llama.cpp](#step-1-download-llamacpp)
546
+ - [Step 2: Download Yi model](#step-2-download-yi-model)
547
+ - [Step 3: Perform inference](#step-3-perform-inference)
548
+
549
+ #### Step 0: Prerequisites
550
+
551
+ - This tutorial assumes you use a MacBook Pro with 16GB of memory and an Apple M2 Pro chip.
552
+
553
+ - Make sure [`git-lfs`](https://git-lfs.com/) is installed on your machine.
554
+
555
+ #### Step 1: Download `llama.cpp`
556
+
557
+ To clone the [`llama.cpp`](https://github.com/ggerganov/llama.cpp) repository, run the following command.
558
+
559
+ ```bash
560
+ git clone git@github.com:ggerganov/llama.cpp.git
561
+ ```
562
+
563
+ #### Step 2: Download Yi model
564
+
565
+ 2.1 To clone [XeIaso/yi-chat-6B-GGUF](https://huggingface.co/XeIaso/yi-chat-6B-GGUF/tree/main) with just pointers, run the following command.
566
+
567
+ ```bash
568
+ GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/XeIaso/yi-chat-6B-GGUF
569
+ ```
570
+
571
+ 2.2 To download a quantized Yi model ([yi-chat-6b.Q2_K.gguf](https://huggingface.co/XeIaso/yi-chat-6B-GGUF/blob/main/yi-chat-6b.Q2_K.gguf)), run the following command.
572
+
573
+ ```bash
574
+ git-lfs pull --include yi-chat-6b.Q2_K.gguf
575
+ ```
576
+
577
+ #### Step 3: Perform inference
578
+
579
+ To perform inference with the Yi model, you can use one of the following methods.
580
+
581
+ - [Method 1: Perform inference in terminal](#method-1-perform-inference-in-terminal)
582
+
583
+ - [Method 2: Perform inference in web](#method-2-perform-inference-in-web)
584
+
585
+ ##### Method 1: Perform inference in terminal
586
+
587
+ To compile `llama.cpp` using 4 threads and then conduct inference, navigate to the `llama.cpp` directory, and run the following command.
588
+
589
+ > ##### Tips
590
+ >
591
+ > - Replace `/Users/yu/yi-chat-6B-GGUF/yi-chat-6b.Q2_K.gguf` with the actual path of your model.
592
+ >
593
+ > - By default, the model operates in completion mode.
594
+ >
595
+ > - For additional output customization options (for example, system prompt, temperature, repetition penalty, etc.), run `./main -h` to check detailed descriptions and usage.
596
+
597
+ ```bash
598
+ make -j4 && ./main -m /Users/yu/yi-chat-6B-GGUF/yi-chat-6b.Q2_K.gguf -p "How do you feed your pet fox? Please answer this question in 6 simple steps:\nStep 1:" -n 384 -e
599
+
600
+ ...
601
+
602
+ How do you feed your pet fox? Please answer this question in 6 simple steps:
603
+
604
+ Step 1: Select the appropriate food for your pet fox. You should choose high-quality, balanced prey items that are suitable for their unique dietary needs. These could include live or frozen mice, rats, pigeons, or other small mammals, as well as fresh fruits and vegetables.
605
+
606
+ Step 2: Feed your pet fox once or twice a day, depending on the species and its individual preferences. Always ensure that they have access to fresh water throughout the day.
607
+
608
+ Step 3: Provide an appropriate environment for your pet fox. Ensure it has a comfortable place to rest, plenty of space to move around, and opportunities to play and exercise.
609
+
610
+ Step 4: Socialize your pet with other animals if possible. Interactions with other creatures can help them develop social skills and prevent boredom or stress.
611
+
612
+ Step 5: Regularly check for signs of illness or discomfort in your fox. Be prepared to provide veterinary care as needed, especially for common issues such as parasites, dental health problems, or infections.
613
+
614
+ Step 6: Educate yourself about the needs of your pet fox and be aware of any potential risks or concerns that could affect their well-being. Regularly consult with a veterinarian to ensure you are providing the best care.
615
+
616
+ ...
617
+
618
+ ```
619
+
620
+ Now you have successfully asked a question to the Yi model and got an answer! 🥳
621
+
622
+ ##### Method 2: Perform inference in web
623
+
624
+ 1. To initialize a lightweight and swift chatbot, run the following command.
625
+
626
+ ```bash
627
+ cd llama.cpp
628
+ ./server --ctx-size 2048 --host 0.0.0.0 --n-gpu-layers 64 --model /Users/yu/yi-chat-6B-GGUF/yi-chat-6b.Q2_K.gguf
629
+ ```
630
+
631
+ Then you can get an output like this:
632
+
633
+
634
+ ```bash
635
+ ...
636
+
637
+ llama_new_context_with_model: n_ctx = 2048
638
+ llama_new_context_with_model: freq_base = 5000000.0
639
+ llama_new_context_with_model: freq_scale = 1
640
+ ggml_metal_init: allocating
641
+ ggml_metal_init: found device: Apple M2 Pro
642
+ ggml_metal_init: picking default device: Apple M2 Pro
643
+ ggml_metal_init: ggml.metallib not found, loading from source
644
+ ggml_metal_init: GGML_METAL_PATH_RESOURCES = nil
645
+ ggml_metal_init: loading '/Users/yu/llama.cpp/ggml-metal.metal'
646
+ ggml_metal_init: GPU name: Apple M2 Pro
647
+ ggml_metal_init: GPU family: MTLGPUFamilyApple8 (1008)
648
+ ggml_metal_init: hasUnifiedMemory = true
649
+ ggml_metal_init: recommendedMaxWorkingSetSize = 11453.25 MB
650
+ ggml_metal_init: maxTransferRate = built-in GPU
651
+ ggml_backend_metal_buffer_type_alloc_buffer: allocated buffer, size = 128.00 MiB, ( 2629.44 / 10922.67)
652
+ llama_new_context_with_model: KV self size = 128.00 MiB, K (f16): 64.00 MiB, V (f16): 64.00 MiB
653
+ ggml_backend_metal_buffer_type_alloc_buffer: allocated buffer, size = 0.02 MiB, ( 2629.45 / 10922.67)
654
+ llama_build_graph: non-view tensors processed: 676/676
655
+ llama_new_context_with_model: compute buffer total size = 159.19 MiB
656
+ ggml_backend_metal_buffer_type_alloc_buffer: allocated buffer, size = 156.02 MiB, ( 2785.45 / 10922.67)
657
+ Available slots:
658
+ -> Slot 0 - max context: 2048
659
+
660
+ llama server listening at http://0.0.0.0:8080
661
+ ```
662
+
663
+ 2. To access the chatbot interface, open your web browser and enter `http://0.0.0.0:8080` into the address bar.
664
+
665
+ ![Yi model chatbot interface - llama.cpp](https://github.com/01-ai/Yi/blob/main/assets/img/yi_llama_cpp1.png?raw=true)
666
+
667
+
668
+ 3. Enter a question, such as "How do you feed your pet fox? Please answer this question in 6 simple steps" into the prompt window, and you will receive a corresponding answer.
669
+
670
+ ![Ask a question to Yi model - llama.cpp](https://github.com/01-ai/Yi/blob/main/assets/img/yi_llama_cpp2.png?raw=true)
671
+
672
+ </ul>
673
+ </details>
674
+
675
+ <p align="right"> [
676
+ <a href="#top">Back to top ⬆️ </a> ]
677
+ </p>
678
+
679
+ ### Web demo
680
+
681
+ You can build a web UI demo for Yi **chat** models (note that Yi base models are not supported in this senario).
682
+
683
+ [Step 1: Prepare your environment](#step-1-prepare-your-environment).
684
+
685
+ [Step 2: Download the Yi model](#step-2-download-the-yi-model).
686
+
687
+ Step 3. To start a web service locally, run the following command.
688
+
689
+ ```bash
690
+ python demo/web_demo.py -c <your-model-path>
691
+ ```
692
+
693
+ You can access the web UI by entering the address provided in the console into your browser.
694
+
695
+ ![Quick start - web demo](https://github.com/01-ai/Yi/blob/main/assets/img/yi_34b_chat_web_demo.gif?raw=true)
696
+
697
+ <p align="right"> [
698
+ <a href="#top">Back to top ⬆️ </a> ]
699
+ </p>
700
+
701
+ ### Fine-tuning
702
+
703
+ ```bash
704
+ bash finetune/scripts/run_sft_Yi_6b.sh
705
+ ```
706
+
707
+ Once finished, you can compare the finetuned model and the base model with the following command:
708
+
709
+ ```bash
710
+ bash finetune/scripts/run_eval.sh
711
+ ```
712
+ <details style="display: inline;"><summary>For advanced usage (like fine-tuning based on your custom data), see the explanations below. ⬇️ </summary> <ul>
713
+
714
+ ### Finetune code for Yi 6B and 34B
715
+
716
+ #### Preparation
717
+
718
+ ##### From Image
719
+
720
+ By default, we use a small dataset from [BAAI/COIG](https://huggingface.co/datasets/BAAI/COIG) to finetune the base model.
721
+ You can also prepare your customized dataset in the following `jsonl` format:
722
+
723
+ ```json
724
+ { "prompt": "Human: Who are you? Assistant:", "chosen": "I'm Yi." }
725
+ ```
726
+
727
+ And then mount them in the container to replace the default ones:
728
+
729
+ ```bash
730
+ docker run -it \
731
+ -v /path/to/save/finetuned/model/:/finetuned-model \
732
+ -v /path/to/train.jsonl:/yi/finetune/data/train.json \
733
+ -v /path/to/eval.jsonl:/yi/finetune/data/eval.json \
734
+ ghcr.io/01-ai/yi:latest \
735
+ bash finetune/scripts/run_sft_Yi_6b.sh
736
+ ```
737
+
738
+ ##### From Local Server
739
+
740
+ Make sure you have conda. If not, use
741
+
742
+ ```bash
743
+ mkdir -p ~/miniconda3
744
+ wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda3/miniconda.sh
745
+ bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3
746
+ rm -rf ~/miniconda3/miniconda.sh
747
+ ~/miniconda3/bin/conda init bash
748
+ source ~/.bashrc
749
+ ```
750
+
751
+ Then, create a conda env:
752
+
753
+ ```bash
754
+ conda create -n dev_env python=3.10 -y
755
+ conda activate dev_env
756
+ pip install torch==2.0.1 deepspeed==0.10 tensorboard transformers datasets sentencepiece accelerate ray==2.7
757
+ ```
758
+
759
+ #### Hardware Setup
760
+
761
+ For the Yi-6B model, a node with 4 GPUs, each has GPU mem larger than 60GB is recommended.
762
+
763
+ For the Yi-34B model, because the usage of zero-offload technique takes a lot CPU memory, please be careful to limit the GPU numbers in 34B finetune training. Please use CUDA_VISIBLE_DEVICES to limit the GPU number (as shown in scripts/run_sft_Yi_34b.sh).
764
+
765
+ A typical hardware setup for finetuning 34B model is a node with 8GPUS (limit to 4 in running by CUDA_VISIBLE_DEVICES=0,1,2,3), each has GPU mem larger than 80GB, with total CPU mem larger than 900GB.
766
+
767
+ #### Quick Start
768
+
769
+ Download a LLM-base model to MODEL_PATH (6B and 34B). A typical folder of models is like:
770
+
771
+ ```bash
772
+ |-- $MODEL_PATH
773
+ | |-- config.json
774
+ | |-- pytorch_model-00001-of-00002.bin
775
+ | |-- pytorch_model-00002-of-00002.bin
776
+ | |-- pytorch_model.bin.index.json
777
+ | |-- tokenizer_config.json
778
+ | |-- tokenizer.model
779
+ | |-- ...
780
+ ```
781
+
782
+ Download a dataset from huggingface to local storage DATA_PATH, e.g. Dahoas/rm-static.
783
+
784
+ ```bash
785
+ |-- $DATA_PATH
786
+ | |-- data
787
+ | | |-- train-00000-of-00001-2a1df75c6bce91ab.parquet
788
+ | | |-- test-00000-of-00001-8c7c51afc6d45980.parquet
789
+ | |-- dataset_infos.json
790
+ | |-- README.md
791
+ ```
792
+
793
+ `finetune/yi_example_dataset` has example datasets, which are modified from [BAAI/COIG](https://huggingface.co/datasets/BAAI/COIG)
794
+
795
+ ```bash
796
+ |-- $DATA_PATH
797
+ |--data
798
+ |-- train.jsonl
799
+ |-- eval.jsonl
800
+ ```
801
+
802
+ `cd` into the scripts folder, copy and paste the script, and run. For example:
803
+
804
+ ```bash
805
+ cd finetune/scripts
806
+
807
+ bash run_sft_Yi_6b.sh
808
+ ```
809
+
810
+ For the Yi-6B base model, setting training_debug_steps=20 and num_train_epochs=4 can output a chat model, which takes about 20 minutes.
811
+
812
+ For the Yi-34B base model, it takes a relatively long time for initialization. Please be patient.
813
+
814
+ #### Evaluation
815
+
816
+ ```bash
817
+ cd finetune/scripts
818
+
819
+ bash run_eval.sh
820
+ ```
821
+
822
+ Then you'll see the answer from both the base model and the finetuned model.
823
+ </ul>
824
+ </details>
825
+
826
+ <p align="right"> [
827
+ <a href="#top">Back to top ⬆️ </a> ]
828
+ </p>
829
+
830
+ ### Quantization
831
+
832
+ #### GPT-Q
833
+ ```bash
834
+ python quantization/gptq/quant_autogptq.py \
835
+ --model /base_model \
836
+ --output_dir /quantized_model \
837
+ --trust_remote_code
838
+ ```
839
+
840
+ Once finished, you can then evaluate the resulting model as follows:
841
+
842
+ ```bash
843
+ python quantization/gptq/eval_quantized_model.py \
844
+ --model /quantized_model \
845
+ --trust_remote_code
846
+ ```
847
+
848
+ <details style="display: inline;"><summary>For a more detailed explanation, see the explanations below. ⬇️</summary> <ul>
849
+
850
+ #### GPT-Q quantization
851
+
852
+ [GPT-Q](https://github.com/IST-DASLab/gptq) is a PTQ(Post-Training Quantization)
853
+ method. It's memory saving and provides potential speedups while retaining the accuracy
854
+ of the model.
855
+
856
+ Yi models can be GPT-Q quantized without a lot of efforts.
857
+ We provide a step-by-step tutorial below.
858
+
859
+ To run GPT-Q, we will use [AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ) and
860
+ [exllama](https://github.com/turboderp/exllama).
861
+ And the huggingface transformers has integrated optimum and auto-gptq to perform
862
+ GPTQ quantization on language models.
863
+
864
+ ##### Do Quantization
865
+
866
+ The `quant_autogptq.py` script is provided for you to perform GPT-Q quantization:
867
+
868
+ ```bash
869
+ python quant_autogptq.py --model /base_model \
870
+ --output_dir /quantized_model --bits 4 --group_size 128 --trust_remote_code
871
+ ```
872
+
873
+
874
+ ##### Run Quantized Model
875
+
876
+ You can run a quantized model using the `eval_quantized_model.py`:
877
+
878
+ ```bash
879
+ python eval_quantized_model.py --model /quantized_model --trust_remote_code
880
+ ```
881
+ </ul>
882
+ </details>
883
+
884
+ #### AWQ
885
+ ```bash
886
+ python quantization/awq/quant_autoawq.py \
887
+ --model /base_model \
888
+ --output_dir /quantized_model \
889
+ --trust_remote_code
890
+ ```
891
+
892
+ Once finished, you can then evaluate the resulting model as follows:
893
+
894
+ ```bash
895
+ python quantization/awq/eval_quantized_model.py \
896
+ --model /quantized_model \
897
+ --trust_remote_code
898
+ ```
899
+ <details style="display: inline;"><summary>For detailed explanations, see the explanations below. ⬇️</summary> <ul>
900
+
901
+ #### AWQ quantization
902
+
903
+ [AWQ](https://github.com/mit-han-lab/llm-awq) is a PTQ(Post-Training Quantization)
904
+ method. It's an efficient and accurate low-bit weight quantization (INT3/4) for LLMs.
905
+
906
+ Yi models can be AWQ quantized without a lot of efforts.
907
+ We provide a step-by-step tutorial below.
908
+
909
+ To run AWQ, we will use [AutoAWQ](https://github.com/casper-hansen/AutoAWQ).
910
+
911
+ ##### Do Quantization
912
+
913
+ The `quant_autoawq.py` script is provided for you to perform AWQ quantization:
914
+
915
+ ```bash
916
+ python quant_autoawq.py --model /base_model \
917
+ --output_dir /quantized_model --bits 4 --group_size 128 --trust_remote_code
918
+ ```
919
+
920
+ ##### Run Quantized Model
921
+
922
+ You can run a quantized model using the `eval_quantized_model.py`:
923
+
924
+ ```bash
925
+ python eval_quantized_model.py --model /quantized_model --trust_remote_code
926
+ ```
927
+
928
+
929
+ </ul>
930
+ </details>
931
+ <p align="right"> [
932
+ <a href="#top">Back to top ⬆️ </a> ]
933
+ </p>
934
+
935
+ ### Deployment
936
+
937
+ If you want to deploy Yi models, make sure you meet the software and hardware requirements.
938
+
939
+ #### Software requirements
940
+
941
+ Before using Yi quantized models, make sure you've installed the correct software listed below.
942
+
943
+ | Model | Software
944
+ |---|---
945
+ Yi 4-bit quantized models | [AWQ and CUDA](https://github.com/casper-hansen/AutoAWQ?tab=readme-ov-file#install-from-pypi)
946
+ Yi 8-bit quantized models | [GPTQ and CUDA](https://github.com/PanQiWei/AutoGPTQ?tab=readme-ov-file#quick-installation)
947
+
948
+ #### Hardware requirements
949
+
950
+ Before deploying Yi in your environment, make sure your hardware meets the following requirements.
951
+
952
+ ##### Chat models
953
+
954
+ | Model | Minimum VRAM | Recommended GPU Example |
955
+ |:----------------------|:--------------|:-------------------------------------:|
956
+ | Yi-6B-Chat | 15 GB | 1 x RTX 3090 (24 GB) <br> 1 x RTX 4090 (24 GB) <br> 1 x A10 (24 GB) <br> 1 x A30 (24 GB) |
957
+ | Yi-6B-Chat-4bits | 4 GB | 1 x RTX 3060 (12 GB)<br> 1 x RTX 4060 (8 GB) |
958
+ | Yi-6B-Chat-8bits | 8 GB | 1 x RTX 3070 (8 GB) <br> 1 x RTX 4060 (8 GB) |
959
+ | Yi-34B-Chat | 72 GB | 4 x RTX 4090 (24 GB)<br> 1 x A800 (80GB) |
960
+ | Yi-34B-Chat-4bits | 20 GB | 1 x RTX 3090 (24 GB) <br> 1 x RTX 4090 (24 GB) <br> 1 x A10 (24 GB) <br> 1 x A30 (24 GB) <br> 1 x A100 (40 GB) |
961
+ | Yi-34B-Chat-8bits | 38 GB | 2 x RTX 3090 (24 GB) <br> 2 x RTX 4090 (24 GB)<br> 1 x A800 (40 GB) |
962
+
963
+ Below are detailed minimum VRAM requirements under different batch use cases.
964
+
965
+ | Model | batch=1 | batch=4 | batch=16 | batch=32 |
966
+ | ----------------------- | ------- | ------- | -------- | -------- |
967
+ | Yi-6B-Chat | 12 GB | 13 GB | 15 GB | 18 GB |
968
+ | Yi-6B-Chat-4bits | 4 GB | 5 GB | 7 GB | 10 GB |
969
+ | Yi-6B-Chat-8bits | 7 GB | 8 GB | 10 GB | 14 GB |
970
+ | Yi-34B-Chat | 65 GB | 68 GB | 76 GB | > 80 GB |
971
+ | Yi-34B-Chat-4bits | 19 GB | 20 GB | 30 GB | 40 GB |
972
+ | Yi-34B-Chat-8bits | 35 GB | 37 GB | 46 GB | 58 GB |
973
+
974
+ ##### Base models
975
+
976
+ | Model | Minimum VRAM | Recommended GPU Example |
977
+ |----------------------|--------------|:-------------------------------------:|
978
+ | Yi-6B | 15 GB | 1 x RTX 3090 (24 GB) <br> 1 x RTX 4090 (24 GB) <br> 1 x A10 (24 GB) <br> 1 x A30 (24 GB) |
979
+ | Yi-6B-200K | 50 GB | 1 x A800 (80 GB) |
980
+ | Yi-9B | 20 GB | 1 x RTX 4090 (24 GB) |
981
+ | Yi-34B | 72 GB | 4 x RTX 4090 (24 GB) <br> 1 x A800 (80 GB) |
982
+ | Yi-34B-200K | 200 GB | 4 x A800 (80 GB) |
983
+
984
+ <p align="right"> [
985
+ <a href="#top">Back to top ⬆️ </a> ]
986
+ </p>
987
+
988
+ ### Learning hub
989
+
990
+ <details>
991
+ <summary> If you want to learn Yi, you can find a wealth of helpful educational resources here. ⬇️</summary>
992
+ <br>
993
+
994
+ Welcome to the Yi learning hub!
995
+
996
+ Whether you're a seasoned developer or a newcomer, you can find a wealth of helpful educational resources to enhance your understanding and skills with Yi models, including insightful blog posts, comprehensive video tutorials, hands-on guides, and more.
997
+
998
+ The content you find here has been generously contributed by knowledgeable Yi experts and passionate enthusiasts. We extend our heartfelt gratitude for your invaluable contributions!
999
+
1000
+ At the same time, we also warmly invite you to join our collaborative effort by contributing to Yi. If you have already made contributions to Yi, please don't hesitate to showcase your remarkable work in the table below.
1001
+
1002
+ With all these resources at your fingertips, you're ready to start your exciting journey with Yi. Happy learning! 🥳
1003
+
1004
+ #### Tutorials
1005
+ ##### English tutorials
1006
+ | Type | Deliverable | Date | Author |
1007
+ |-------------|--------------------------------------------------------|----------------|----------------|
1008
+ | Video | [Run dolphin-2.2-yi-34b on IoT Devices](https://www.youtube.com/watch?v=NJ89T5mO25Y) | 2023-11-30 | [Second State](https://github.com/second-state) |
1009
+ | Blog | [Running Yi-34B-Chat locally using LlamaEdge](https://www.secondstate.io/articles/yi-34b/) | 2023-11-30 | [Second State](https://github.com/second-state) |
1010
+ | Video | [Install Yi 34B Locally - Chinese English Bilingual LLM](https://www.youtube.com/watch?v=CVQvj4Wrh4w&t=476s) | 2023-11-05 | [Fahd Mirza](https://www.youtube.com/@fahdmirza) |
1011
+ | Video | [Dolphin Yi 34b - Brand New Foundational Model TESTED](https://www.youtube.com/watch?v=On3Zuv27V3k&t=85s) | 2023-11-27 | [Matthew Berman](https://www.youtube.com/@matthew_berman) |
1012
+
1013
+
1014
+ ##### Chinese tutorials
1015
+ | Type | Deliverable | Date | Author |
1016
+ |-------------|--------------------------------------------------------|----------------|----------------|
1017
+ | Blog | [实测零一万物Yi-VL多模态语言模型:能准确“识图吃瓜”](https://mp.weixin.qq.com/s/fu4O9XvJ03JhimsEyI-SsQ) | 2024-02-02 | [苏洋](https://github.com/soulteary) |
1018
+ | Blog | [本地运行零一万物 34B 大模型,使用 Llama.cpp & 21G 显存](https://zhuanlan.zhihu.com/p/668921042) | 2023-11-26 | [苏洋](https://github.com/soulteary) |
1019
+ | Blog | [零一万物模型折腾笔记:官方 Yi-34B 模型基础使用](https://zhuanlan.zhihu.com/p/671387298) | 2023-12-10 | [苏洋](https://github.com/soulteary) |
1020
+ | Blog | [CPU 混合推理,非常见大模型量化方案:“二三五六” 位量化方案](https://zhuanlan.zhihu.com/p/671698216) | 2023-12-12 | [苏洋](https://github.com/soulteary) |
1021
+ | Blog | [单卡 3 小时训练 Yi-6B 大模型 Agent:基于 Llama Factory 实战](https://zhuanlan.zhihu.com/p/678989191) | 2024-01-22 | [郑耀威](https://github.com/hiyouga) |
1022
+ | Blog | [零一万物开源Yi-VL多模态大模型,魔搭社区推理&微调最佳实践来啦!](https://zhuanlan.zhihu.com/p/680098411) | 2024-01-26 | [ModelScope](https://github.com/modelscope) |
1023
+ | Video | [只需 24G 显存,用 vllm 跑起来 Yi-34B 中英双语大模型](https://www.bilibili.com/video/BV17t4y1f7Ee/) | 2023-12-28 | [漆妮妮](https://space.bilibili.com/1262370256) |
1024
+ | Video | [Yi-VL-34B 多模态大模型 - 用两张 A40 显卡跑起来](https://www.bilibili.com/video/BV1Q5411y7AG/) | 2023-01-28 | [漆妮妮](https://space.bilibili.com/1262370256) |
1025
+
1026
+ </details>
1027
+
1028
+
1029
+ # Why Yi?
1030
+
1031
+ - [Ecosystem](#ecosystem)
1032
+ - [Upstream](#upstream)
1033
+ - [Downstream](#downstream)
1034
+ - [Serving](#serving)
1035
+ - [Quantization](#quantization-1)
1036
+ - [Fine-tuning](#fine-tuning-1)
1037
+ - [API](#api)
1038
+ - [Benchmarks](#benchmarks)
1039
+ - [Chat model performance](#chat-model-performance)
1040
+ - [Base model performance](#base-model-performance)
1041
+ - [Yi-34B and Yi-34B-200K](#yi-34b-and-yi-34b-200k)
1042
+ - [Yi-9B](#yi-9b)
1043
+
1044
+ ## Ecosystem
1045
+
1046
+ Yi has a comprehensive ecosystem, offering a range of tools, services, and models to enrich your experiences and maximize productivity.
1047
+
1048
+ - [Upstream](#upstream)
1049
+ - [Downstream](#downstream)
1050
+ - [Serving](#serving)
1051
+ - [Quantization](#quantization-1)
1052
+ - [Fine-tuning](#fine-tuning-1)
1053
+ - [API](#api)
1054
+
1055
+ ### Upstream
1056
+
1057
+ The Yi series models follow the same model architecture as Llama. By choosing Yi, you can leverage existing tools, libraries, and resources within the Llama ecosystem, eliminating the need to create new tools and enhancing development efficiency.
1058
+
1059
+ For example, the Yi series models are saved in the format of the Llama model. You can directly use `LlamaForCausalLM` and `LlamaTokenizer` to load the model. For more information, see [Use the chat model](#31-use-the-chat-model).
1060
+
1061
+ ```python
1062
+ from transformers import AutoModelForCausalLM, AutoTokenizer
1063
+
1064
+ tokenizer = AutoTokenizer.from_pretrained("01-ai/Yi-34b", use_fast=False)
1065
+
1066
+ model = AutoModelForCausalLM.from_pretrained("01-ai/Yi-34b", device_map="auto")
1067
+ ```
1068
+
1069
+ <p align="right"> [
1070
+ <a href="#top">Back to top ⬆️ </a> ]
1071
+ </p>
1072
+
1073
+ ### Downstream
1074
+
1075
+ > 💡 Tip
1076
+ >
1077
+ > - Feel free to create a PR and share the fantastic work you've built using the Yi series models.
1078
+ >
1079
+ > - To help others quickly understand your work, it is recommended to use the format of `<model-name>: <model-intro> + <model-highlights>`.
1080
+
1081
+ #### Serving
1082
+
1083
+ If you want to get up with Yi in a few minutes, you can use the following services built upon Yi.
1084
+
1085
+ - Yi-34B-Chat: you can chat with Yi using one of the following platforms:
1086
+ - [Yi-34B-Chat | Hugging Face](https://huggingface.co/spaces/01-ai/Yi-34B-Chat)
1087
+ - [Yi-34B-Chat | Yi Platform](https://platform.lingyiwanwu.com/): **Note** that currently it's available through a whitelist. Welcome to apply (fill out a form in [English](https://cn.mikecrm.com/l91ODJf) or [Chinese](https://cn.mikecrm.com/gnEZjiQ)) and experience it firsthand!
1088
+
1089
+ - [Yi-6B-Chat (Replicate)](https://replicate.com/01-ai): you can use this model with more options by setting additional parameters and calling APIs.
1090
+
1091
+ - [ScaleLLM](https://github.com/vectorch-ai/ScaleLLM#supported-models): you can use this service to run Yi models locally with added flexibility and customization.
1092
+
1093
+ #### Quantization
1094
+
1095
+ If you have limited computational capabilities, you can use Yi's quantized models as follows.
1096
+
1097
+ These quantized models have reduced precision but offer increased efficiency, such as faster inference speed and smaller RAM usage.
1098
+
1099
+ - [TheBloke/Yi-34B-GPTQ](https://huggingface.co/TheBloke/Yi-34B-GPTQ)
1100
+ - [TheBloke/Yi-34B-GGUF](https://huggingface.co/TheBloke/Yi-34B-GGUF)
1101
+ - [TheBloke/Yi-34B-AWQ](https://huggingface.co/TheBloke/Yi-34B-AWQ)
1102
+
1103
+ #### Fine-tuning
1104
+
1105
+ If you're seeking to explore the diverse capabilities within Yi's thriving family, you can delve into Yi's fine-tuned models as below.
1106
+
1107
+ - [TheBloke Models](https://huggingface.co/TheBloke): this site hosts numerous fine-tuned models derived from various LLMs including Yi.
1108
+
1109
+ This is not an exhaustive list for Yi, but to name a few sorted on downloads:
1110
+ - [TheBloke/dolphin-2_2-yi-34b-AWQ](https://huggingface.co/TheBloke/dolphin-2_2-yi-34b-AWQ)
1111
+ - [TheBloke/Yi-34B-Chat-AWQ](https://huggingface.co/TheBloke/Yi-34B-Chat-AWQ)
1112
+ - [TheBloke/Yi-34B-Chat-GPTQ](https://huggingface.co/TheBloke/Yi-34B-Chat-GPTQ)
1113
+
1114
+ - [SUSTech/SUS-Chat-34B](https://huggingface.co/SUSTech/SUS-Chat-34B): this model ranked first among all models below 70B and outperformed the twice larger deepseek-llm-67b-chat. You can check the result on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).
1115
+
1116
+ - [OrionStarAI/OrionStar-Yi-34B-Chat-Llama](https://huggingface.co/OrionStarAI/OrionStar-Yi-34B-Chat-Llama): this model excelled beyond other models (such as GPT-4, Qwen-14B-Chat, Baichuan2-13B-Chat) in C-Eval and CMMLU evaluations on the [OpenCompass LLM Leaderboard](https://opencompass.org.cn/leaderboard-llm).
1117
+
1118
+ - [NousResearch/Nous-Capybara-34B](https://huggingface.co/NousResearch/Nous-Capybara-34B): this model is trained with 200K context length and 3 epochs on the Capybara dataset.
1119
+
1120
+ #### API
1121
+
1122
+ - [amazing-openai-api](https://github.com/soulteary/amazing-openai-api): this tool converts Yi model APIs into the OpenAI API format out of the box.
1123
+ - [LlamaEdge](https://www.secondstate.io/articles/yi-34b/#create-an-openai-compatible-api-service-for-the-yi-34b-chat-model): this tool builds an OpenAI-compatible API server for Yi-34B-Chat using a portable Wasm (WebAssembly) file, powered by Rust.
1124
+
1125
+ <p align="right"> [
1126
+ <a href="#top">Back to top ⬆️ </a> ]
1127
+ </p>
1128
+
1129
+ ## Tech report
1130
+
1131
+ For detailed capabilities of the Yi series model, see [Yi: Open Foundation Models by 01.AI](https://arxiv.org/abs/2403.04652).
1132
+
1133
+ ### Citation
1134
+
1135
+ ```
1136
+ @misc{ai2024yi,
1137
+ title={Yi: Open Foundation Models by 01.AI},
1138
+ author={01. AI and : and Alex Young and Bei Chen and Chao Li and Chengen Huang and Ge Zhang and Guanwei Zhang and Heng Li and Jiangcheng Zhu and Jianqun Chen and Jing Chang and Kaidong Yu and Peng Liu and Qiang Liu and Shawn Yue and Senbin Yang and Shiming Yang and Tao Yu and Wen Xie and Wenhao Huang and Xiaohui Hu and Xiaoyi Ren and Xinyao Niu and Pengcheng Nie and Yuchi Xu and Yudong Liu and Yue Wang and Yuxuan Cai and Zhenyu Gu and Zhiyuan Liu and Zonghong Dai},
1139
+ year={2024},
1140
+ eprint={2403.04652},
1141
+ archivePrefix={arXiv},
1142
+ primaryClass={cs.CL}
1143
+ }
1144
+ ```
1145
+
1146
+ ## Benchmarks
1147
+
1148
+ - [Chat model performance](#-chat-model-performance)
1149
+ - [Base model performance](#-base-model-performance)
1150
+
1151
+ ### Chat model performance
1152
+
1153
+ Yi-34B-Chat model demonstrates exceptional performance, ranking first among all existing open-source models in the benchmarks including MMLU, CMMLU, BBH, GSM8k, and more.
1154
+
1155
+ ![Chat model performance](https://github.com/01-ai/Yi/blob/main/assets/img/benchmark_chat.png?raw=true)
1156
+
1157
+ <details>
1158
+ <summary> Evaluation methods and challenges. ⬇️ </summary>
1159
+
1160
+ - **Evaluation methods**: we evaluated various benchmarks using both zero-shot and few-shot methods, except for TruthfulQA.
1161
+ - **Zero-shot vs. few-shot**: in chat models, the zero-shot approach is more commonly employed.
1162
+ - **Evaluation strategy**: our evaluation strategy involves generating responses while following instructions explicitly or implicitly (such as using few-shot examples). We then isolate relevant answers from the generated text.
1163
+ - **Challenges faced**: some models are not well-suited to produce output in the specific format required by instructions in few datasets, which leads to suboptimal results.
1164
+
1165
+ <strong>*</strong>: C-Eval results are evaluated on the validation datasets
1166
+ </details>
1167
+
1168
+ ### Base model performance
1169
+
1170
+ #### Yi-34B and Yi-34B-200K
1171
+
1172
+ The Yi-34B and Yi-34B-200K models stand out as the top performers among open-source models, especially excelling in MMLU, CMMLU, common-sense reasoning, reading comprehension, and more.
1173
+
1174
+ ![Base model performance](https://github.com/01-ai/Yi/blob/main/assets/img/benchmark_base.png?raw=true)
1175
+
1176
+ <details>
1177
+ <summary> Evaluation methods. ⬇️</summary>
1178
+
1179
+ - **Disparity in results**: while benchmarking open-source models, a disparity has been noted between results from our pipeline and those reported by public sources like OpenCompass.
1180
+ - **Investigation findings**: a deeper investigation reveals that variations in prompts, post-processing strategies, and sampling techniques across models may lead to significant outcome differences.
1181
+ - **Uniform benchmarking process**: our methodology aligns with the original benchmarks—consistent prompts and post-processing strategies are used, and greedy decoding is applied during evaluations without any post-processing for the generated content.
1182
+ - **Efforts to retrieve unreported scores**: for scores that were not reported by the original authors (including scores reported with different settings), we try to get results with our pipeline.
1183
+ - **Extensive model evaluation**: to evaluate the model’s capability extensively, we adopted the methodology outlined in Llama2. Specifically, we included PIQA, SIQA, HellaSwag, WinoGrande, ARC, OBQA, and CSQA to assess common sense reasoning. SquAD, QuAC, and BoolQ were incorporated to evaluate reading comprehension.
1184
+ - **Special configurations**: CSQA was exclusively tested using a 7-shot setup, while all other tests were conducted with a 0-shot configuration. Additionally, we introduced GSM8K (8-shot@1), MATH (4-shot@1), HumanEval (0-shot@1), and MBPP (3-shot@1) under the category "Math & Code".
1185
+ - **Falcon-180B caveat**: Falcon-180B was not tested on QuAC and OBQA due to technical constraints. Its performance score is an average from other tasks, and considering the generally lower scores of these two tasks, Falcon-180B's capabilities are likely not underestimated.
1186
+ </details>
1187
+
1188
+ #### Yi-9B
1189
+
1190
+ Yi-9B is almost the best among a range of similar-sized open-source models (including Mistral-7B, SOLAR-10.7B, Gemma-7B, DeepSeek-Coder-7B-Base-v1.5 and more), particularly excelling in code, math, common-sense reasoning, and reading comprehension.
1191
+
1192
+ ![Yi-9B benchmark - details](https://github.com/01-ai/Yi/blob/main/assets/img/Yi-9B_benchmark_details.png?raw=true)
1193
+
1194
+ - In terms of **overall** ability (Mean-All), Yi-9B performs the best among similarly sized open-source models, surpassing DeepSeek-Coder, DeepSeek-Math, Mistral-7B, SOLAR-10.7B, and Gemma-7B.
1195
+
1196
+ ![Yi-9B benchmark - overall](https://github.com/01-ai/Yi/blob/main/assets/img/Yi-9B_benchmark_overall.png?raw=true)
1197
+
1198
+ - In terms of **coding** ability (Mean-Code), Yi-9B's performance is second only to DeepSeek-Coder-7B, surpassing Yi-34B, SOLAR-10.7B, Mistral-7B, and Gemma-7B.
1199
+
1200
+ ![Yi-9B benchmark - code](https://github.com/01-ai/Yi/blob/main/assets/img/Yi-9B_benchmark_code.png?raw=true)
1201
+
1202
+ - In terms of **math** ability (Mean-Math), Yi-9B's performance is second only to DeepSeek-Math-7B, surpassing SOLAR-10.7B, Mistral-7B, and Gemma-7B.
1203
+
1204
+ ![Yi-9B benchmark - math](https://github.com/01-ai/Yi/blob/main/assets/img/Yi-9B_benchmark_math.png?raw=true)
1205
+
1206
+ - In terms of **common sense and reasoning** ability (Mean-Text), Yi-9B's performance is on par with Mistral-7B, SOLAR-10.7B, and Gemma-7B.
1207
+
1208
+ ![Yi-9B benchmark - text](https://github.com/01-ai/Yi/blob/main/assets/img/Yi-9B_benchmark_text.png?raw=true)
1209
+
1210
+ <p align="right"> [
1211
+ <a href="#top">Back to top ⬆️ </a> ]
1212
+ </p>
1213
+
1214
+ # Who can use Yi?
1215
+
1216
+ Everyone! 🙌 ✅
1217
+
1218
+ - The Yi series models are free for personal usage, academic purposes, and commercial use. All usage must adhere to the [Yi Series Models Community License Agreement 2.1](https://github.com/01-ai/Yi/blob/main/MODEL_LICENSE_AGREEMENT.txt)
1219
+
1220
+ - For free commercial use, you only need to [complete this form](https://www.lingyiwanwu.com/yi-license) to get a Yi Model Commercial License.
1221
+
1222
+ <p align="right"> [
1223
+ <a href="#top">Back to top ⬆️ </a> ]
1224
+ </p>
1225
+
1226
+ # Misc.
1227
+
1228
+ ### Acknowledgments
1229
+
1230
+ A heartfelt thank you to each of you who have made contributions to the Yi community! You have helped Yi not just a project, but a vibrant, growing home for innovation.
1231
+
1232
+ [![yi contributors](https://contrib.rocks/image?repo=01-ai/yi&max=2000&columns=15)](https://github.com/01-ai/yi/graphs/contributors)
1233
+
1234
+ <p align="right"> [
1235
+ <a href="#top">Back to top ⬆️ </a> ]
1236
+ </p>
1237
+
1238
+ ### Disclaimer
1239
+
1240
+ We use data compliance checking algorithms during the training process, to
1241
+ ensure the compliance of the trained model to the best of our ability. Due to
1242
+ complex data and the diversity of language model usage scenarios, we cannot
1243
+ guarantee that the model will generate correct, and reasonable output in all
1244
+ scenarios. Please be aware that there is still a risk of the model producing
1245
+ problematic outputs. We will not be responsible for any risks and issues
1246
+ resulting from misuse, misguidance, illegal usage, and related misinformation,
1247
+ as well as any associated data security concerns.
1248
+
1249
+ <p align="right"> [
1250
+ <a href="#top">Back to top ⬆️ </a> ]
1251
+ </p>
1252
+
1253
+ ### License
1254
+
1255
+ The source code in this repo is licensed under the [Apache 2.0
1256
+ license](https://github.com/01-ai/Yi/blob/main/LICENSE). The Yi series models are fully open for academic research and free for commercial use, with automatic permission granted upon application. All usage must adhere to the [Yi Series Models Community License Agreement 2.1](https://github.com/01-ai/Yi/blob/main/MODEL_LICENSE_AGREEMENT.txt).
1257
+ For free commercial use, you only need to send an email to [get official commercial permission](https://www.lingyiwanwu.com/yi-license).
1258
+
1259
+ <p align="right"> [
1260
+ <a href="#top">Back to top ⬆️ </a> ]
1261
+ </p>
Yi.svg ADDED
config.json CHANGED
@@ -1,29 +1,37 @@
1
  {
2
- "architectures": [
3
- "YiForCausalLM"
4
- ],
5
- "auto_map": {
6
- "AutoConfig": "configuration_yi.YiConfig",
7
- "AutoModel": "modeling_yi.YiModel",
8
- "AutoModelForCausalLM":"modeling_yi.YiForCausalLM"
9
- },
10
- "bos_token_id": 1,
11
- "eos_token_id": 2,
12
- "hidden_act": "silu",
13
- "hidden_size": 4096,
14
- "initializer_range": 0.02,
15
- "intermediate_size": 11008,
16
- "max_position_embeddings": 200000,
17
- "model_type": "Yi",
18
- "num_attention_heads": 32,
19
- "num_hidden_layers": 32,
20
- "num_key_value_heads": 4,
21
- "pad_token_id": 0,
22
- "rms_norm_eps": 1e-05,
23
- "rope_theta": 5000000.0,
24
- "tie_word_embeddings": false,
25
- "torch_dtype": "bfloat16",
26
- "transformers_version": "4.34.0",
27
- "use_cache": true,
28
- "vocab_size": 64000
29
- }
 
 
 
 
 
 
 
 
 
1
  {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "bos_token_id": 1,
6
+ "eos_token_id": 2,
7
+ "hidden_act": "silu",
8
+ "hidden_size": 4096,
9
+ "initializer_range": 0.02,
10
+ "intermediate_size": 11008,
11
+ "max_position_embeddings": 200000,
12
+ "model_type": "llama",
13
+ "num_attention_heads": 32,
14
+ "num_hidden_layers": 32,
15
+ "num_key_value_heads": 4,
16
+ "pad_token_id": 0,
17
+ "pretraining_tp": 1,
18
+ "rms_norm_eps": 1e-05,
19
+ "rope_scaling": null,
20
+ "rope_theta": 10000000.0,
21
+ "tie_word_embeddings": false,
22
+ "torch_dtype": "bfloat16",
23
+ "transformers_version": "4.34.0",
24
+ "use_cache": true,
25
+ "vocab_size": 64000,
26
+ "quantization_config": {
27
+ "quant_method": "exl2",
28
+ "version": "0.0.15",
29
+ "bits": 8.0,
30
+ "head_bits": 8,
31
+ "calibration": {
32
+ "rows": 100,
33
+ "length": 2048,
34
+ "dataset": "(default)"
35
+ }
36
+ }
37
+ }
generation_config.json CHANGED
@@ -3,5 +3,5 @@
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
  "pad_token_id": 0,
6
- "transformers_version": "4.31.0"
7
  }
 
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
  "pad_token_id": 0,
6
+ "transformers_version": "4.34.0"
7
  }
md5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ 79e86bca154db97ec5a44409913f2e6b pytorch_model-00001-of-00002.bin
2
+ 189489a23e33259290db34c69682f72e pytorch_model-00002-of-00002.bin
3
+ 291724ef50f729e45d68f474a7755bbc tokenizer.model
model.safetensors.index.json CHANGED
@@ -1,298 +1,330 @@
1
  {
2
- "metadata": {
3
- "total_size": 12122071040
4
- },
5
- "weight_map": {
6
- "lm_head.weight": "model-00002-of-00002.safetensors",
7
- "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
8
- "model.layers.0.ln1.weight": "model-00001-of-00002.safetensors",
9
- "model.layers.0.ln2.weight": "model-00001-of-00002.safetensors",
10
- "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
11
- "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
12
- "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
13
- "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
14
- "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
15
- "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
16
- "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
17
- "model.layers.1.ln1.weight": "model-00001-of-00002.safetensors",
18
- "model.layers.1.ln2.weight": "model-00001-of-00002.safetensors",
19
- "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
20
- "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
21
- "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
22
- "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
23
- "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
24
- "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
25
- "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
26
- "model.layers.10.ln1.weight": "model-00001-of-00002.safetensors",
27
- "model.layers.10.ln2.weight": "model-00001-of-00002.safetensors",
28
- "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
29
- "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
30
- "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
31
- "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
32
- "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
33
- "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
34
- "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
35
- "model.layers.11.ln1.weight": "model-00001-of-00002.safetensors",
36
- "model.layers.11.ln2.weight": "model-00001-of-00002.safetensors",
37
- "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
38
- "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
39
- "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
40
- "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
41
- "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
42
- "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
43
- "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
44
- "model.layers.12.ln1.weight": "model-00001-of-00002.safetensors",
45
- "model.layers.12.ln2.weight": "model-00001-of-00002.safetensors",
46
- "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
47
- "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
48
- "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
49
- "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
50
- "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
51
- "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
52
- "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
53
- "model.layers.13.ln1.weight": "model-00001-of-00002.safetensors",
54
- "model.layers.13.ln2.weight": "model-00001-of-00002.safetensors",
55
- "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
56
- "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
57
- "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
58
- "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
59
- "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
60
- "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
61
- "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
62
- "model.layers.14.ln1.weight": "model-00001-of-00002.safetensors",
63
- "model.layers.14.ln2.weight": "model-00001-of-00002.safetensors",
64
- "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
65
- "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
66
- "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
67
- "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
68
- "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
69
- "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
70
- "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
71
- "model.layers.15.ln1.weight": "model-00001-of-00002.safetensors",
72
- "model.layers.15.ln2.weight": "model-00001-of-00002.safetensors",
73
- "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
74
- "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
75
- "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
76
- "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
77
- "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
78
- "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
79
- "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
80
- "model.layers.16.ln1.weight": "model-00001-of-00002.safetensors",
81
- "model.layers.16.ln2.weight": "model-00001-of-00002.safetensors",
82
- "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
83
- "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
84
- "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
85
- "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
86
- "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
87
- "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
88
- "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
89
- "model.layers.17.ln1.weight": "model-00001-of-00002.safetensors",
90
- "model.layers.17.ln2.weight": "model-00001-of-00002.safetensors",
91
- "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
92
- "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
93
- "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
94
- "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
95
- "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
96
- "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
97
- "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
98
- "model.layers.18.ln1.weight": "model-00001-of-00002.safetensors",
99
- "model.layers.18.ln2.weight": "model-00001-of-00002.safetensors",
100
- "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
101
- "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
102
- "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
103
- "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
104
- "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
105
- "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
106
- "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
107
- "model.layers.19.ln1.weight": "model-00001-of-00002.safetensors",
108
- "model.layers.19.ln2.weight": "model-00001-of-00002.safetensors",
109
- "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
110
- "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
111
- "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
112
- "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
113
- "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
114
- "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
115
- "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
116
- "model.layers.2.ln1.weight": "model-00001-of-00002.safetensors",
117
- "model.layers.2.ln2.weight": "model-00001-of-00002.safetensors",
118
- "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
119
- "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
120
- "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
121
- "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
122
- "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
123
- "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
124
- "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
125
- "model.layers.20.ln1.weight": "model-00001-of-00002.safetensors",
126
- "model.layers.20.ln2.weight": "model-00001-of-00002.safetensors",
127
- "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
128
- "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
129
- "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
130
- "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
131
- "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
132
- "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
133
- "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
134
- "model.layers.21.ln1.weight": "model-00001-of-00002.safetensors",
135
- "model.layers.21.ln2.weight": "model-00001-of-00002.safetensors",
136
- "model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
137
- "model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
138
- "model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
139
- "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
140
- "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
141
- "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
142
- "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
143
- "model.layers.22.ln1.weight": "model-00001-of-00002.safetensors",
144
- "model.layers.22.ln2.weight": "model-00001-of-00002.safetensors",
145
- "model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
146
- "model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
147
- "model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
148
- "model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
149
- "model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
150
- "model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
151
- "model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
152
- "model.layers.23.ln1.weight": "model-00001-of-00002.safetensors",
153
- "model.layers.23.ln2.weight": "model-00001-of-00002.safetensors",
154
- "model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
155
- "model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
156
- "model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
157
- "model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
158
- "model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
159
- "model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
160
- "model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
161
- "model.layers.24.ln1.weight": "model-00001-of-00002.safetensors",
162
- "model.layers.24.ln2.weight": "model-00001-of-00002.safetensors",
163
- "model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
164
- "model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
165
- "model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
166
- "model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
167
- "model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
168
- "model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
169
- "model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
170
- "model.layers.25.ln1.weight": "model-00001-of-00002.safetensors",
171
- "model.layers.25.ln2.weight": "model-00001-of-00002.safetensors",
172
- "model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
173
- "model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
174
- "model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
175
- "model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
176
- "model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
177
- "model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
178
- "model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
179
- "model.layers.26.ln1.weight": "model-00001-of-00002.safetensors",
180
- "model.layers.26.ln2.weight": "model-00001-of-00002.safetensors",
181
- "model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
182
- "model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
183
- "model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
184
- "model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
185
- "model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
186
- "model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
187
- "model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
188
- "model.layers.27.ln1.weight": "model-00002-of-00002.safetensors",
189
- "model.layers.27.ln2.weight": "model-00002-of-00002.safetensors",
190
- "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
191
- "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
192
- "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
193
- "model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
194
- "model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
195
- "model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
196
- "model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
197
- "model.layers.28.ln1.weight": "model-00002-of-00002.safetensors",
198
- "model.layers.28.ln2.weight": "model-00002-of-00002.safetensors",
199
- "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
200
- "model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
201
- "model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
202
- "model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
203
- "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
204
- "model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
205
- "model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
206
- "model.layers.29.ln1.weight": "model-00002-of-00002.safetensors",
207
- "model.layers.29.ln2.weight": "model-00002-of-00002.safetensors",
208
- "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
209
- "model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
210
- "model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
211
- "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
212
- "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
213
- "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
214
- "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
215
- "model.layers.3.ln1.weight": "model-00001-of-00002.safetensors",
216
- "model.layers.3.ln2.weight": "model-00001-of-00002.safetensors",
217
- "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
218
- "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
219
- "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
220
- "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
221
- "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
222
- "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
223
- "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
224
- "model.layers.30.ln1.weight": "model-00002-of-00002.safetensors",
225
- "model.layers.30.ln2.weight": "model-00002-of-00002.safetensors",
226
- "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
227
- "model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
228
- "model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
229
- "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
230
- "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
231
- "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
232
- "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
233
- "model.layers.31.ln1.weight": "model-00002-of-00002.safetensors",
234
- "model.layers.31.ln2.weight": "model-00002-of-00002.safetensors",
235
- "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
236
- "model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
237
- "model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
238
- "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
239
- "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
240
- "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
241
- "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
242
- "model.layers.4.ln1.weight": "model-00001-of-00002.safetensors",
243
- "model.layers.4.ln2.weight": "model-00001-of-00002.safetensors",
244
- "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
245
- "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
246
- "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
247
- "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
248
- "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
249
- "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
250
- "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
251
- "model.layers.5.ln1.weight": "model-00001-of-00002.safetensors",
252
- "model.layers.5.ln2.weight": "model-00001-of-00002.safetensors",
253
- "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
254
- "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
255
- "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
256
- "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
257
- "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
258
- "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
259
- "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
260
- "model.layers.6.ln1.weight": "model-00001-of-00002.safetensors",
261
- "model.layers.6.ln2.weight": "model-00001-of-00002.safetensors",
262
- "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
263
- "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
264
- "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
265
- "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
266
- "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
267
- "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
268
- "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
269
- "model.layers.7.ln1.weight": "model-00001-of-00002.safetensors",
270
- "model.layers.7.ln2.weight": "model-00001-of-00002.safetensors",
271
- "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
272
- "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
273
- "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
274
- "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
275
- "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
276
- "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
277
- "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
278
- "model.layers.8.ln1.weight": "model-00001-of-00002.safetensors",
279
- "model.layers.8.ln2.weight": "model-00001-of-00002.safetensors",
280
- "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
281
- "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
282
- "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
283
- "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
284
- "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
285
- "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
286
- "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
287
- "model.layers.9.ln1.weight": "model-00001-of-00002.safetensors",
288
- "model.layers.9.ln2.weight": "model-00001-of-00002.safetensors",
289
- "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
290
- "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
291
- "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
292
- "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
293
- "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
294
- "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
295
- "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
296
- "model.norm.weight": "model-00002-of-00002.safetensors"
297
- }
298
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  {
2
+ "metadata": {
3
+ "total_size": 12122075136
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model-00002-of-00002.safetensors",
7
+ "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
8
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
9
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
10
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
11
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
12
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
13
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
14
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
15
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
16
+ "model.layers.0.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
17
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
18
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
19
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
20
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
21
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
22
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
23
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
24
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
25
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
26
+ "model.layers.1.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
27
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
28
+ "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
29
+ "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
30
+ "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
31
+ "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
32
+ "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
33
+ "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
34
+ "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
35
+ "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
36
+ "model.layers.10.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
37
+ "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
38
+ "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
39
+ "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
40
+ "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
41
+ "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
42
+ "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
43
+ "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
44
+ "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
45
+ "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
46
+ "model.layers.11.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
47
+ "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
48
+ "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
49
+ "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
50
+ "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
51
+ "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
52
+ "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
53
+ "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
54
+ "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
55
+ "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
56
+ "model.layers.12.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
57
+ "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
58
+ "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
59
+ "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
60
+ "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
61
+ "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
62
+ "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
63
+ "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
64
+ "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
65
+ "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
66
+ "model.layers.13.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
67
+ "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
68
+ "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
69
+ "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
70
+ "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
71
+ "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
72
+ "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
73
+ "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
74
+ "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
75
+ "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
76
+ "model.layers.14.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
77
+ "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
78
+ "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
79
+ "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
80
+ "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
81
+ "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
82
+ "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
83
+ "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
84
+ "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
85
+ "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
86
+ "model.layers.15.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
87
+ "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
88
+ "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
89
+ "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
90
+ "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
91
+ "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
92
+ "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
93
+ "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
94
+ "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
95
+ "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
96
+ "model.layers.16.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
97
+ "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
98
+ "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
99
+ "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
100
+ "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
101
+ "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
102
+ "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
103
+ "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
104
+ "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
105
+ "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
106
+ "model.layers.17.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
107
+ "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
108
+ "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
109
+ "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
110
+ "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
111
+ "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
112
+ "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
113
+ "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
114
+ "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
115
+ "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
116
+ "model.layers.18.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
117
+ "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
118
+ "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
119
+ "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
120
+ "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
121
+ "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
122
+ "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
123
+ "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
124
+ "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
125
+ "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
126
+ "model.layers.19.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
127
+ "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
128
+ "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
129
+ "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
130
+ "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
131
+ "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
132
+ "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
133
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
134
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
135
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
136
+ "model.layers.2.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
137
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
138
+ "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
139
+ "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
140
+ "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
141
+ "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
142
+ "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
143
+ "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
144
+ "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
145
+ "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
146
+ "model.layers.20.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
147
+ "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
148
+ "model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
149
+ "model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
150
+ "model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
151
+ "model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
152
+ "model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
153
+ "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
154
+ "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
155
+ "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
156
+ "model.layers.21.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
157
+ "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
158
+ "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
159
+ "model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
160
+ "model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
161
+ "model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
162
+ "model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
163
+ "model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
164
+ "model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
165
+ "model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
166
+ "model.layers.22.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
167
+ "model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
168
+ "model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
169
+ "model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
170
+ "model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
171
+ "model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
172
+ "model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
173
+ "model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
174
+ "model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
175
+ "model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
176
+ "model.layers.23.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
177
+ "model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
178
+ "model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
179
+ "model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
180
+ "model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
181
+ "model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
182
+ "model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
183
+ "model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
184
+ "model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
185
+ "model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
186
+ "model.layers.24.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
187
+ "model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
188
+ "model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
189
+ "model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
190
+ "model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
191
+ "model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
192
+ "model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
193
+ "model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
194
+ "model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
195
+ "model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
196
+ "model.layers.25.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
197
+ "model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
198
+ "model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
199
+ "model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
200
+ "model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
201
+ "model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
202
+ "model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
203
+ "model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
204
+ "model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
205
+ "model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
206
+ "model.layers.26.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
207
+ "model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
208
+ "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
209
+ "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
210
+ "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
211
+ "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
212
+ "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
213
+ "model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
214
+ "model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
215
+ "model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
216
+ "model.layers.27.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
217
+ "model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
218
+ "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
219
+ "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
220
+ "model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
221
+ "model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
222
+ "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
223
+ "model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
224
+ "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
225
+ "model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
226
+ "model.layers.28.self_attn.rotary_emb.inv_freq": "model-00002-of-00002.safetensors",
227
+ "model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
228
+ "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
229
+ "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
230
+ "model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
231
+ "model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
232
+ "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
233
+ "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
234
+ "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
235
+ "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
236
+ "model.layers.29.self_attn.rotary_emb.inv_freq": "model-00002-of-00002.safetensors",
237
+ "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
238
+ "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
239
+ "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
240
+ "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
241
+ "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
242
+ "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
243
+ "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
244
+ "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
245
+ "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
246
+ "model.layers.3.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
247
+ "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
248
+ "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
249
+ "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
250
+ "model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
251
+ "model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
252
+ "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
253
+ "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
254
+ "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
255
+ "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
256
+ "model.layers.30.self_attn.rotary_emb.inv_freq": "model-00002-of-00002.safetensors",
257
+ "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
258
+ "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
259
+ "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
260
+ "model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
261
+ "model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
262
+ "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
263
+ "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
264
+ "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
265
+ "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
266
+ "model.layers.31.self_attn.rotary_emb.inv_freq": "model-00002-of-00002.safetensors",
267
+ "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
268
+ "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
269
+ "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
270
+ "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
271
+ "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
272
+ "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
273
+ "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
274
+ "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
275
+ "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
276
+ "model.layers.4.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
277
+ "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
278
+ "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
279
+ "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
280
+ "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
281
+ "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
282
+ "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
283
+ "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
284
+ "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
285
+ "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
286
+ "model.layers.5.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
287
+ "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
288
+ "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
289
+ "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
290
+ "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
291
+ "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
292
+ "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
293
+ "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
294
+ "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
295
+ "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
296
+ "model.layers.6.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
297
+ "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
298
+ "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
299
+ "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
300
+ "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
301
+ "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
302
+ "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
303
+ "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
304
+ "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
305
+ "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
306
+ "model.layers.7.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
307
+ "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
308
+ "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
309
+ "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
310
+ "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
311
+ "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
312
+ "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
313
+ "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
314
+ "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
315
+ "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
316
+ "model.layers.8.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
317
+ "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
318
+ "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
319
+ "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
320
+ "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
321
+ "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
322
+ "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
323
+ "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
324
+ "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
325
+ "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
326
+ "model.layers.9.self_attn.rotary_emb.inv_freq": "model-00001-of-00002.safetensors",
327
+ "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
328
+ "model.norm.weight": "model-00002-of-00002.safetensors"
329
+ }
330
+ }
output.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5250f05927e3a3179c01d2d8c19e3adf54f932683cef7146522c81b32e1fd789
3
- size 6337663352
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d34dc76d91813c464eaa1360925b8b20d8dd1a7e0869a13e01e6fe6dc75c0dd6
3
+ size 6322661592
pytorch_model.bin.index.json CHANGED
@@ -1,297 +1,329 @@
1
  {
2
  "metadata": {
3
- "total_size": 12122071040
4
  },
5
  "weight_map": {
6
  "lm_head.weight": "pytorch_model-00002-of-00002.bin",
7
  "model.embed_tokens.weight": "pytorch_model-00001-of-00002.bin",
8
- "model.layers.0.ln1.weight": "pytorch_model-00001-of-00002.bin",
9
- "model.layers.0.ln2.weight": "pytorch_model-00001-of-00002.bin",
10
  "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
11
  "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
12
  "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
13
  "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
14
  "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
15
  "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
16
  "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
17
- "model.layers.1.ln1.weight": "pytorch_model-00001-of-00002.bin",
18
- "model.layers.1.ln2.weight": "pytorch_model-00001-of-00002.bin",
19
  "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
20
  "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
21
  "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
22
  "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
23
  "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
24
  "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
25
  "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
26
- "model.layers.10.ln1.weight": "pytorch_model-00001-of-00002.bin",
27
- "model.layers.10.ln2.weight": "pytorch_model-00001-of-00002.bin",
28
  "model.layers.10.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
29
  "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
30
  "model.layers.10.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
31
  "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
32
  "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
33
  "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
34
  "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
35
- "model.layers.11.ln1.weight": "pytorch_model-00001-of-00002.bin",
36
- "model.layers.11.ln2.weight": "pytorch_model-00001-of-00002.bin",
37
  "model.layers.11.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
38
  "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
39
  "model.layers.11.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
40
  "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
41
  "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
42
  "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
43
  "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
44
- "model.layers.12.ln1.weight": "pytorch_model-00001-of-00002.bin",
45
- "model.layers.12.ln2.weight": "pytorch_model-00001-of-00002.bin",
46
  "model.layers.12.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
47
  "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
48
  "model.layers.12.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
49
  "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
50
  "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
51
  "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
52
  "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
53
- "model.layers.13.ln1.weight": "pytorch_model-00001-of-00002.bin",
54
- "model.layers.13.ln2.weight": "pytorch_model-00001-of-00002.bin",
55
  "model.layers.13.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
56
  "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
57
  "model.layers.13.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
58
  "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
59
  "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
60
  "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
61
  "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
62
- "model.layers.14.ln1.weight": "pytorch_model-00001-of-00002.bin",
63
- "model.layers.14.ln2.weight": "pytorch_model-00001-of-00002.bin",
64
  "model.layers.14.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
65
  "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
66
  "model.layers.14.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
67
  "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
68
  "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
69
  "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
70
  "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
71
- "model.layers.15.ln1.weight": "pytorch_model-00001-of-00002.bin",
72
- "model.layers.15.ln2.weight": "pytorch_model-00001-of-00002.bin",
73
  "model.layers.15.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
74
  "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
75
  "model.layers.15.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
76
  "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
77
  "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
78
  "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
79
  "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
80
- "model.layers.16.ln1.weight": "pytorch_model-00001-of-00002.bin",
81
- "model.layers.16.ln2.weight": "pytorch_model-00001-of-00002.bin",
82
  "model.layers.16.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
83
  "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
84
  "model.layers.16.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
85
  "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
86
  "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
87
  "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
88
  "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
89
- "model.layers.17.ln1.weight": "pytorch_model-00001-of-00002.bin",
90
- "model.layers.17.ln2.weight": "pytorch_model-00001-of-00002.bin",
91
  "model.layers.17.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
92
  "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
93
  "model.layers.17.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
94
  "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
95
  "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
96
  "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
97
  "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
98
- "model.layers.18.ln1.weight": "pytorch_model-00001-of-00002.bin",
99
- "model.layers.18.ln2.weight": "pytorch_model-00001-of-00002.bin",
100
  "model.layers.18.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
101
  "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
102
  "model.layers.18.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
103
  "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
104
  "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
105
  "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
106
  "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
107
- "model.layers.19.ln1.weight": "pytorch_model-00001-of-00002.bin",
108
- "model.layers.19.ln2.weight": "pytorch_model-00001-of-00002.bin",
109
  "model.layers.19.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
110
  "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
111
  "model.layers.19.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
112
  "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
113
  "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
114
  "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
115
  "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
116
- "model.layers.2.ln1.weight": "pytorch_model-00001-of-00002.bin",
117
- "model.layers.2.ln2.weight": "pytorch_model-00001-of-00002.bin",
118
  "model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
119
  "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
120
  "model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
121
  "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
122
  "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
123
  "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
124
  "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
125
- "model.layers.20.ln1.weight": "pytorch_model-00001-of-00002.bin",
126
- "model.layers.20.ln2.weight": "pytorch_model-00001-of-00002.bin",
127
  "model.layers.20.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
128
  "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
129
  "model.layers.20.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
130
  "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
131
  "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
132
  "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
133
  "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
134
- "model.layers.21.ln1.weight": "pytorch_model-00001-of-00002.bin",
135
- "model.layers.21.ln2.weight": "pytorch_model-00001-of-00002.bin",
136
  "model.layers.21.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
137
  "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
138
  "model.layers.21.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
139
  "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
140
  "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
141
  "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
142
  "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
143
- "model.layers.22.ln1.weight": "pytorch_model-00001-of-00002.bin",
144
- "model.layers.22.ln2.weight": "pytorch_model-00001-of-00002.bin",
145
  "model.layers.22.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
146
  "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
147
  "model.layers.22.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
148
  "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
149
  "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
150
  "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
151
  "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
152
- "model.layers.23.ln1.weight": "pytorch_model-00001-of-00002.bin",
153
- "model.layers.23.ln2.weight": "pytorch_model-00001-of-00002.bin",
154
  "model.layers.23.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
155
  "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
156
  "model.layers.23.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
157
  "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
158
  "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
159
  "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
160
  "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
161
- "model.layers.24.ln1.weight": "pytorch_model-00001-of-00002.bin",
162
- "model.layers.24.ln2.weight": "pytorch_model-00001-of-00002.bin",
163
  "model.layers.24.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
164
  "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
165
  "model.layers.24.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
166
  "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
167
  "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
168
  "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
169
  "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
170
- "model.layers.25.ln1.weight": "pytorch_model-00001-of-00002.bin",
171
- "model.layers.25.ln2.weight": "pytorch_model-00001-of-00002.bin",
172
  "model.layers.25.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
173
  "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
174
  "model.layers.25.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
175
  "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
176
  "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
177
  "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
178
  "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
179
- "model.layers.26.ln1.weight": "pytorch_model-00001-of-00002.bin",
180
- "model.layers.26.ln2.weight": "pytorch_model-00001-of-00002.bin",
181
  "model.layers.26.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
182
  "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
183
  "model.layers.26.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
184
  "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
185
  "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
186
  "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
187
  "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
188
- "model.layers.27.ln1.weight": "pytorch_model-00002-of-00002.bin",
189
- "model.layers.27.ln2.weight": "pytorch_model-00002-of-00002.bin",
190
  "model.layers.27.mlp.down_proj.weight": "pytorch_model-00002-of-00002.bin",
191
  "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00002-of-00002.bin",
192
  "model.layers.27.mlp.up_proj.weight": "pytorch_model-00002-of-00002.bin",
 
193
  "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
194
  "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
195
  "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
196
  "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
197
- "model.layers.28.ln1.weight": "pytorch_model-00002-of-00002.bin",
198
- "model.layers.28.ln2.weight": "pytorch_model-00002-of-00002.bin",
199
  "model.layers.28.mlp.down_proj.weight": "pytorch_model-00002-of-00002.bin",
200
  "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00002-of-00002.bin",
201
  "model.layers.28.mlp.up_proj.weight": "pytorch_model-00002-of-00002.bin",
 
202
  "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
203
  "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00002-of-00002.bin",
204
  "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
 
205
  "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
206
- "model.layers.29.ln1.weight": "pytorch_model-00002-of-00002.bin",
207
- "model.layers.29.ln2.weight": "pytorch_model-00002-of-00002.bin",
208
  "model.layers.29.mlp.down_proj.weight": "pytorch_model-00002-of-00002.bin",
209
  "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00002-of-00002.bin",
210
  "model.layers.29.mlp.up_proj.weight": "pytorch_model-00002-of-00002.bin",
 
211
  "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
212
  "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00002-of-00002.bin",
213
  "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
 
214
  "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
215
- "model.layers.3.ln1.weight": "pytorch_model-00001-of-00002.bin",
216
- "model.layers.3.ln2.weight": "pytorch_model-00001-of-00002.bin",
217
  "model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
218
  "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
219
  "model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
220
  "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
221
  "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
222
  "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
223
  "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
224
- "model.layers.30.ln1.weight": "pytorch_model-00002-of-00002.bin",
225
- "model.layers.30.ln2.weight": "pytorch_model-00002-of-00002.bin",
226
  "model.layers.30.mlp.down_proj.weight": "pytorch_model-00002-of-00002.bin",
227
  "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00002-of-00002.bin",
228
  "model.layers.30.mlp.up_proj.weight": "pytorch_model-00002-of-00002.bin",
 
229
  "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
230
  "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00002-of-00002.bin",
231
  "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
 
232
  "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
233
- "model.layers.31.ln1.weight": "pytorch_model-00002-of-00002.bin",
234
- "model.layers.31.ln2.weight": "pytorch_model-00002-of-00002.bin",
235
  "model.layers.31.mlp.down_proj.weight": "pytorch_model-00002-of-00002.bin",
236
  "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00002-of-00002.bin",
237
  "model.layers.31.mlp.up_proj.weight": "pytorch_model-00002-of-00002.bin",
 
238
  "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
239
  "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00002-of-00002.bin",
240
  "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
 
241
  "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
242
- "model.layers.4.ln1.weight": "pytorch_model-00001-of-00002.bin",
243
- "model.layers.4.ln2.weight": "pytorch_model-00001-of-00002.bin",
244
  "model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
245
  "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
246
  "model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
247
  "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
248
  "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
249
  "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
250
  "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
251
- "model.layers.5.ln1.weight": "pytorch_model-00001-of-00002.bin",
252
- "model.layers.5.ln2.weight": "pytorch_model-00001-of-00002.bin",
253
  "model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
254
  "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
255
  "model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
256
  "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
257
  "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
258
  "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
259
  "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
260
- "model.layers.6.ln1.weight": "pytorch_model-00001-of-00002.bin",
261
- "model.layers.6.ln2.weight": "pytorch_model-00001-of-00002.bin",
262
  "model.layers.6.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
263
  "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
264
  "model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
265
  "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
266
  "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
267
  "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
268
  "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
269
- "model.layers.7.ln1.weight": "pytorch_model-00001-of-00002.bin",
270
- "model.layers.7.ln2.weight": "pytorch_model-00001-of-00002.bin",
271
  "model.layers.7.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
272
  "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
273
  "model.layers.7.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
274
  "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
275
  "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
276
  "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
277
  "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
278
- "model.layers.8.ln1.weight": "pytorch_model-00001-of-00002.bin",
279
- "model.layers.8.ln2.weight": "pytorch_model-00001-of-00002.bin",
280
  "model.layers.8.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
281
  "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
282
  "model.layers.8.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
283
  "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
284
  "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
285
  "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
286
  "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
287
- "model.layers.9.ln1.weight": "pytorch_model-00001-of-00002.bin",
288
- "model.layers.9.ln2.weight": "pytorch_model-00001-of-00002.bin",
289
  "model.layers.9.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
290
  "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
291
  "model.layers.9.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
 
292
  "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
293
  "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
294
  "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
 
295
  "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
296
  "model.norm.weight": "pytorch_model-00002-of-00002.bin"
297
  }
 
1
  {
2
  "metadata": {
3
+ "total_size": 12122075136
4
  },
5
  "weight_map": {
6
  "lm_head.weight": "pytorch_model-00002-of-00002.bin",
7
  "model.embed_tokens.weight": "pytorch_model-00001-of-00002.bin",
8
+ "model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
9
  "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
10
  "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
11
  "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
12
+ "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
13
  "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
14
  "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
15
  "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
16
+ "model.layers.0.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
17
  "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
18
+ "model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
19
  "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
20
  "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
21
  "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
22
+ "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
23
  "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
24
  "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
25
  "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
26
+ "model.layers.1.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
27
  "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
28
+ "model.layers.10.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
29
  "model.layers.10.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
30
  "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
31
  "model.layers.10.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
32
+ "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
33
  "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
34
  "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
35
  "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
36
+ "model.layers.10.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
37
  "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
38
+ "model.layers.11.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
39
  "model.layers.11.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
40
  "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
41
  "model.layers.11.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
42
+ "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
43
  "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
44
  "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
45
  "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
46
+ "model.layers.11.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
47
  "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
48
+ "model.layers.12.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
49
  "model.layers.12.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
50
  "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
51
  "model.layers.12.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
52
+ "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
53
  "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
54
  "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
55
  "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
56
+ "model.layers.12.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
57
  "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
58
+ "model.layers.13.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
59
  "model.layers.13.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
60
  "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
61
  "model.layers.13.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
62
+ "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
63
  "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
64
  "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
65
  "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
66
+ "model.layers.13.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
67
  "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
68
+ "model.layers.14.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
69
  "model.layers.14.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
70
  "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
71
  "model.layers.14.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
72
+ "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
73
  "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
74
  "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
75
  "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
76
+ "model.layers.14.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
77
  "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
78
+ "model.layers.15.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
79
  "model.layers.15.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
80
  "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
81
  "model.layers.15.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
82
+ "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
83
  "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
84
  "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
85
  "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
86
+ "model.layers.15.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
87
  "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
88
+ "model.layers.16.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
89
  "model.layers.16.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
90
  "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
91
  "model.layers.16.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
92
+ "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
93
  "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
94
  "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
95
  "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
96
+ "model.layers.16.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
97
  "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
98
+ "model.layers.17.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
99
  "model.layers.17.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
100
  "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
101
  "model.layers.17.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
102
+ "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
103
  "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
104
  "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
105
  "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
106
+ "model.layers.17.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
107
  "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
108
+ "model.layers.18.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
109
  "model.layers.18.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
110
  "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
111
  "model.layers.18.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
112
+ "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
113
  "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
114
  "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
115
  "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
116
+ "model.layers.18.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
117
  "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
118
+ "model.layers.19.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
119
  "model.layers.19.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
120
  "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
121
  "model.layers.19.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
122
+ "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
123
  "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
124
  "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
125
  "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
126
+ "model.layers.19.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
127
  "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
128
+ "model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
129
  "model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
130
  "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
131
  "model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
132
+ "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
133
  "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
134
  "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
135
  "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
136
+ "model.layers.2.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
137
  "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
138
+ "model.layers.20.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
139
  "model.layers.20.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
140
  "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
141
  "model.layers.20.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
142
+ "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
143
  "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
144
  "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
145
  "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
146
+ "model.layers.20.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
147
  "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
148
+ "model.layers.21.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
149
  "model.layers.21.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
150
  "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
151
  "model.layers.21.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
152
+ "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
153
  "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
154
  "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
155
  "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
156
+ "model.layers.21.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
157
  "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
158
+ "model.layers.22.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
159
  "model.layers.22.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
160
  "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
161
  "model.layers.22.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
162
+ "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
163
  "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
164
  "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
165
  "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
166
+ "model.layers.22.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
167
  "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
168
+ "model.layers.23.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
169
  "model.layers.23.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
170
  "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
171
  "model.layers.23.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
172
+ "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
173
  "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
174
  "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
175
  "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
176
+ "model.layers.23.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
177
  "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
178
+ "model.layers.24.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
179
  "model.layers.24.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
180
  "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
181
  "model.layers.24.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
182
+ "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
183
  "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
184
  "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
185
  "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
186
+ "model.layers.24.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
187
  "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
188
+ "model.layers.25.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
189
  "model.layers.25.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
190
  "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
191
  "model.layers.25.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
192
+ "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
193
  "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
194
  "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
195
  "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
196
+ "model.layers.25.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
197
  "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
198
+ "model.layers.26.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
199
  "model.layers.26.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
200
  "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
201
  "model.layers.26.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
202
+ "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
203
  "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
204
  "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
205
  "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
206
+ "model.layers.26.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
207
  "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
208
+ "model.layers.27.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
 
209
  "model.layers.27.mlp.down_proj.weight": "pytorch_model-00002-of-00002.bin",
210
  "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00002-of-00002.bin",
211
  "model.layers.27.mlp.up_proj.weight": "pytorch_model-00002-of-00002.bin",
212
+ "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
213
  "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
214
  "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
215
  "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
216
+ "model.layers.27.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
217
  "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
218
+ "model.layers.28.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
 
219
  "model.layers.28.mlp.down_proj.weight": "pytorch_model-00002-of-00002.bin",
220
  "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00002-of-00002.bin",
221
  "model.layers.28.mlp.up_proj.weight": "pytorch_model-00002-of-00002.bin",
222
+ "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
223
  "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
224
  "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00002-of-00002.bin",
225
  "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
226
+ "model.layers.28.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
227
  "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
228
+ "model.layers.29.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
 
229
  "model.layers.29.mlp.down_proj.weight": "pytorch_model-00002-of-00002.bin",
230
  "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00002-of-00002.bin",
231
  "model.layers.29.mlp.up_proj.weight": "pytorch_model-00002-of-00002.bin",
232
+ "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
233
  "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
234
  "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00002-of-00002.bin",
235
  "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
236
+ "model.layers.29.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
237
  "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
238
+ "model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
239
  "model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
240
  "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
241
  "model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
242
+ "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
243
  "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
244
  "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
245
  "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
246
+ "model.layers.3.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
247
  "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
248
+ "model.layers.30.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
 
249
  "model.layers.30.mlp.down_proj.weight": "pytorch_model-00002-of-00002.bin",
250
  "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00002-of-00002.bin",
251
  "model.layers.30.mlp.up_proj.weight": "pytorch_model-00002-of-00002.bin",
252
+ "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
253
  "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
254
  "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00002-of-00002.bin",
255
  "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
256
+ "model.layers.30.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
257
  "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
258
+ "model.layers.31.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
 
259
  "model.layers.31.mlp.down_proj.weight": "pytorch_model-00002-of-00002.bin",
260
  "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00002-of-00002.bin",
261
  "model.layers.31.mlp.up_proj.weight": "pytorch_model-00002-of-00002.bin",
262
+ "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
263
  "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
264
  "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00002-of-00002.bin",
265
  "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
266
+ "model.layers.31.self_attn.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
267
  "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
268
+ "model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
269
  "model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
270
  "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
271
  "model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
272
+ "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
273
  "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
274
  "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
275
  "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
276
+ "model.layers.4.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
277
  "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
278
+ "model.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
279
  "model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
280
  "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
281
  "model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
282
+ "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
283
  "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
284
  "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
285
  "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
286
+ "model.layers.5.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
287
  "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
288
+ "model.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
289
  "model.layers.6.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
290
  "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
291
  "model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
292
+ "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
293
  "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
294
  "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
295
  "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
296
+ "model.layers.6.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
297
  "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
298
+ "model.layers.7.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
299
  "model.layers.7.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
300
  "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
301
  "model.layers.7.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
302
+ "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
303
  "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
304
  "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
305
  "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
306
+ "model.layers.7.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
307
  "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
308
+ "model.layers.8.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
309
  "model.layers.8.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
310
  "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
311
  "model.layers.8.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
312
+ "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
313
  "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
314
  "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
315
  "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
316
+ "model.layers.8.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
317
  "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
318
+ "model.layers.9.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
 
319
  "model.layers.9.mlp.down_proj.weight": "pytorch_model-00001-of-00002.bin",
320
  "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00001-of-00002.bin",
321
  "model.layers.9.mlp.up_proj.weight": "pytorch_model-00001-of-00002.bin",
322
+ "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
323
  "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
324
  "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00001-of-00002.bin",
325
  "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
326
+ "model.layers.9.self_attn.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
327
  "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
328
  "model.norm.weight": "pytorch_model-00002-of-00002.bin"
329
  }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -1,9 +1,13 @@
1
  {
2
- "auto_map": {
3
- "AutoTokenizer": ["tokenization_yi.YiTokenizer", null]
4
- },
5
  "add_bos_token": false,
6
  "add_eos_token": false,
7
  "model_max_length": 200000,
8
- "tokenizer_class": "YiTokenizer"
 
 
 
 
 
 
 
9
  }
 
1
  {
 
 
 
2
  "add_bos_token": false,
3
  "add_eos_token": false,
4
  "model_max_length": 200000,
5
+ "unk_token": "<unk>",
6
+ "bos_token": "<|startoftext|>",
7
+ "eos_token": "<|endoftext|>",
8
+ "pad_token": "<unk>",
9
+ "sp_model_kwargs": {},
10
+ "clean_up_tokenization_spaces": false,
11
+ "legacy": true,
12
+ "tokenizer_class": "LlamaTokenizer"
13
  }