turboderp commited on
Commit
8b7e043
1 Parent(s): c2ec91d
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -1,10 +1,358 @@
1
- EXL2 quants of [c4ai-command-r-plus](https://huggingface.co/CohereForAI/c4ai-command-r-plus)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
 
3
- [2.50 bits per weight](https://huggingface.co/turboderp/command-r-plus-103B-exl2/tree/2.5bpw)
4
- [3.00 bits per weight](https://huggingface.co/turboderp/command-r-plus-103B-exl2/tree/3.0bpw)
5
- [3.50 bits per weight](https://huggingface.co/turboderp/command-r-plus-103B-exl2/tree/3.5bpw)
6
- [4.00 bits per weight](https://huggingface.co/turboderp/command-r-plus-103B-exl2/tree/4.0bpw)
7
- [4.50 bits per weight](https://huggingface.co/turboderp/command-r-plus-103B-exl2/tree/4.5bpw)
8
- [5.00 bits per weight](https://huggingface.co/turboderp/command-r-plus-103B-exl2/tree/5.0bpw)
9
 
10
- [measurement.json](https://huggingface.co/turboderp/command-r-plus-103B-exl2/blob/main/measurement.json)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-nc-4.0
3
+ library_name: transformers
4
+ language:
5
+ - en
6
+ - fr
7
+ - de
8
+ - es
9
+ - it
10
+ - pt
11
+ - ja
12
+ - ko
13
+ - zh
14
+ - ar
15
+ ---
16
 
17
+ # Model Card for C4AI Command R+
 
 
 
 
 
18
 
19
+
20
+ ## Model Summary
21
+
22
+ C4AI Command R+ is an open weights research release of a 104B billion parameter model with highly advanced capabilities, this includes Retrieval Augmented Generation (RAG) and tool use to automate sophisticated tasks. The tool use in this model generation enables multi-step tool use which allows the model to combine multiple tools over multiple steps to accomplish difficult tasks. C4AI Command R+ is a multilingual model evaluated in 10 languages for performance: English, French, Spanish, Italian, German, Brazilian Portuguese, Japanese, Korean, Arabic, and Simplified Chinese. Command R+ is optimized for a variety of use cases including reasoning, summarization, and question answering.
23
+
24
+ C4AI Command R+ is part of a family of open weight releases from Cohere For AI and Cohere. Our smaller companion model is [C4AI Command R](https://huggingface.co/CohereForAI/c4ai-command-r-v01)
25
+
26
+ Developed by: [Cohere](https://cohere.com/) and [Cohere For AI](https://cohere.for.ai)
27
+
28
+ - Point of Contact: Cohere For AI: [cohere.for.ai](https://cohere.for.ai/)
29
+ - License: [CC-BY-NC](https://cohere.com/c4ai-cc-by-nc-license), requires also adhering to [C4AI's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy)
30
+ - Model: c4ai-command-r-plus
31
+ - Model Size: 104 billion parameters
32
+ - Context length: 128K
33
+
34
+ **Try C4AI Command R+**
35
+
36
+ You can try out C4AI Command R+ before downloading the weights in our hosted [Hugging Face Space](https://huggingface.co/spaces/CohereForAI/c4ai-command-r-plus).
37
+
38
+ **Usage**
39
+
40
+ Please install `transformers` from the source repository that includes the necessary changes for this model.
41
+ ```python
42
+ # pip install 'git+https://github.com/huggingface/transformers.git'
43
+ from transformers import AutoTokenizer, AutoModelForCausalLM
44
+
45
+ model_id = "CohereForAI/c4ai-command-r-plus"
46
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
47
+ model = AutoModelForCausalLM.from_pretrained(model_id)
48
+
49
+ # Format message with the command-r-plus chat template
50
+ messages = [{"role": "user", "content": "Hello, how are you?"}]
51
+ input_ids = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
52
+ ## <BOS_TOKEN><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Hello, how are you?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
53
+
54
+ gen_tokens = model.generate(
55
+ input_ids,
56
+ max_new_tokens=100,
57
+ do_sample=True,
58
+ temperature=0.3,
59
+ )
60
+
61
+ gen_text = tokenizer.decode(gen_tokens[0])
62
+ print(gen_text)
63
+ ```
64
+
65
+ **Quantized model through bitsandbytes, 8-bit precision**
66
+
67
+ ```python
68
+ # pip install 'git+https://github.com/huggingface/transformers.git' bitsandbytes accelerate
69
+ from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
70
+
71
+ bnb_config = BitsAndBytesConfig(load_in_8bit=True)
72
+
73
+ model_id = "CohereForAI/c4ai-command-r-plus"
74
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
75
+ model = AutoModelForCausalLM.from_pretrained(model_id, quantization_config=bnb_config)
76
+
77
+ # Format message with the command-r-plus chat template
78
+ messages = [{"role": "user", "content": "Hello, how are you?"}]
79
+ input_ids = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
80
+ ## <BOS_TOKEN><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Hello, how are you?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
81
+
82
+ gen_tokens = model.generate(
83
+ input_ids,
84
+ max_new_tokens=100,
85
+ do_sample=True,
86
+ temperature=0.3,
87
+ )
88
+
89
+ gen_text = tokenizer.decode(gen_tokens[0])
90
+ print(gen_text)
91
+ ```
92
+
93
+ **Quantized model through bitsandbytes, 4-bit precision**
94
+
95
+ ```python
96
+ # pip install 'git+https://github.com/huggingface/transformers.git' bitsandbytes accelerate
97
+ from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
98
+
99
+ bnb_config = BitsAndBytesConfig(load_in_4bit=True)
100
+
101
+ model_id = "CohereForAI/c4ai-command-r-plus"
102
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
103
+ model = AutoModelForCausalLM.from_pretrained(model_id, quantization_config=bnb_config)
104
+
105
+ # Format message with the command-r-plus chat template
106
+ messages = [{"role": "user", "content": "Hello, how are you?"}]
107
+ input_ids = tokenizer.apply_chat_template(messages, tokenize=True, add_generation_prompt=True, return_tensors="pt")
108
+ ## <BOS_TOKEN><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Hello, how are you?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
109
+
110
+ gen_tokens = model.generate(
111
+ input_ids,
112
+ max_new_tokens=100,
113
+ do_sample=True,
114
+ temperature=0.3,
115
+ )
116
+
117
+ gen_text = tokenizer.decode(gen_tokens[0])
118
+ print(gen_text)
119
+ ```
120
+
121
+
122
+ ## Model Details
123
+
124
+ **Input**: Models input text only.
125
+
126
+ **Output**: Models generate text only.
127
+
128
+ **Model Architecture**: This is an auto-regressive language model that uses an optimized transformer architecture. After pretraining, this model uses supervised fine-tuning (SFT) and preference training to align model behavior to human preferences for helpfulness and safety.
129
+
130
+ **Languages covered**: The model is optimized to perform well in the following languages: English, French, Spanish, Italian, German, Brazilian Portuguese, Japanese, Korean, Simplified Chinese, and Arabic.
131
+
132
+ Pre-training data additionally included the following 13 languages: Russian, Polish, Turkish, Vietnamese, Dutch, Czech, Indonesian, Ukrainian, Romanian, Greek, Hindi, Hebrew, Persian.
133
+
134
+ **Context length**: Command R+ supports a context length of 128K.
135
+
136
+ ### Tool use & multihop capabilities:
137
+
138
+ Command R+ has been specifically trained with conversational tool use capabilities. These have been trained into the model via a mixture of supervised fine-tuning and preference fine-tuning, using a specific prompt template. Deviating from this prompt template will likely reduce performance, but we encourage experimentation.
139
+
140
+ Command R+’s tool use functionality takes a conversation as input (with an optional user-system preamble), along with a list of available tools. The model will then generate a json-formatted list of actions to execute on a subset of those tools. Command R+ may use one of its supplied tools more than once.
141
+
142
+ The model has been trained to recognise a special `directly_answer` tool, which it uses to indicate that it doesn’t want to use any of its other tools. The ability to abstain from calling a specific tool can be useful in a range of situations, such as greeting a user, or asking clarifying questions.
143
+ We recommend including the `directly_answer` tool, but it can be removed or renamed if required.
144
+
145
+ Comprehensive documentation for working with command R+'s tool use prompt template can be found [here](https://docs.cohere.com/docs/prompting-command-r).
146
+
147
+ The code snippet below shows a minimal working example on how to render a prompt.
148
+
149
+ <details>
150
+ <summary><b>Usage: Rendering Tool Use Prompts [CLICK TO EXPAND]</b> </summary>
151
+
152
+ ```python
153
+ from transformers import AutoTokenizer
154
+
155
+ model_id = "CohereForAI/c4ai-command-r-plus"
156
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
157
+
158
+ # define conversation input:
159
+ conversation = [
160
+ {"role": "user", "content": "Whats the biggest penguin in the world?"}
161
+ ]
162
+ # Define tools available for the model to use:
163
+ tools = [
164
+ {
165
+ "name": "internet_search",
166
+ "description": "Returns a list of relevant document snippets for a textual query retrieved from the internet",
167
+ "parameter_definitions": {
168
+ "query": {
169
+ "description": "Query to search the internet with",
170
+ "type": 'str',
171
+ "required": True
172
+ }
173
+ }
174
+ },
175
+ {
176
+ 'name': "directly_answer",
177
+ "description": "Calls a standard (un-augmented) AI chatbot to generate a response given the conversation history",
178
+ 'parameter_definitions': {}
179
+ }
180
+ ]
181
+
182
+ # render the tool use prompt as a string:
183
+ tool_use_prompt = tokenizer.apply_tool_use_template(
184
+ conversation,
185
+ tools=tools,
186
+ tokenize=False,
187
+ add_generation_prompt=True,
188
+ )
189
+ print(tool_use_prompt)
190
+ ```
191
+
192
+ </details>
193
+
194
+ <details>
195
+ <summary><b>Example Rendered Tool Use Prompt [CLICK TO EXPAND]</b></summary>
196
+
197
+ ````
198
+ <BOS_TOKEN><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|># Safety Preamble
199
+ The instructions in this section override those in the task description and style guide sections. Don't answer questions that are harmful or immoral.
200
+
201
+ # System Preamble
202
+ ## Basic Rules
203
+ You are a powerful conversational AI trained by Cohere to help people. You are augmented by a number of tools, and your job is to use and consume the output of these tools to best help the user. You will see a conversation history between yourself and a user, ending with an utterance from the user. You will then see a specific instruction instructing you what kind of response to generate. When you answer the user's requests, you cite your sources in your answers, according to those instructions.
204
+
205
+ # User Preamble
206
+ ## Task and Context
207
+ You help people answer their questions and other requests interactively. You will be asked a very wide array of requests on all kinds of topics. You will be equipped with a wide range of search engines or similar tools to help you, which you use to research your answer. You should focus on serving the user's needs as best you can, which will be wide-ranging.
208
+
209
+ ## Style Guide
210
+ Unless the user asks for a different style of answer, you should answer in full sentences, using proper grammar and spelling.
211
+
212
+ ## Available Tools
213
+ Here is a list of tools that you have available to you:
214
+
215
+ ```python
216
+ def internet_search(query: str) -> List[Dict]:
217
+ """Returns a list of relevant document snippets for a textual query retrieved from the internet
218
+
219
+ Args:
220
+ query (str): Query to search the internet with
221
+ """
222
+ pass
223
+ ```
224
+
225
+ ```python
226
+ def directly_answer() -> List[Dict]:
227
+ """Calls a standard (un-augmented) AI chatbot to generate a response given the conversation history
228
+ """
229
+ pass
230
+ ```<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Whats the biggest penguin in the world?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>Write 'Action:' followed by a json-formatted list of actions that you want to perform in order to produce a good response to the user's last input. You can use any of the supplied tools any number of times, but you should aim to execute the minimum number of necessary actions for the input. You should use the `directly-answer` tool if calling the other tools is unnecessary. The list of actions you want to call should be formatted as a list of json objects, for example:
231
+ ```json
232
+ [
233
+ {
234
+ "tool_name": title of the tool in the specification,
235
+ "parameters": a dict of parameters to input into the tool as they are defined in the specs, or {} if it takes no parameters
236
+ }
237
+ ]```<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
238
+
239
+ ````
240
+
241
+ </details>
242
+
243
+ <details>
244
+ <summary><b>Example Rendered Tool Use Completion [CLICK TO EXPAND]</b></summary>
245
+
246
+ ````
247
+ Action: ```json
248
+ [
249
+ {
250
+ "tool_name": "internet_search",
251
+ "parameters": {
252
+ "query": "biggest penguin in the world"
253
+ }
254
+ }
255
+ ]
256
+ ```
257
+ ````
258
+ </details>
259
+
260
+ ### Grounded Generation and RAG Capabilities:
261
+
262
+ Command R+ has been specifically trained with grounded generation capabilities. This means that it can generate responses based on a list of supplied document snippets, and it will include grounding spans (citations) in its response indicating the source of the information. This can be used to enable behaviors such as grounded summarization and the final step of Retrieval Augmented Generation (RAG). This behavior has been trained into the model via a mixture of supervised fine-tuning and preference fine-tuning, using a specific prompt template. Deviating from this prompt template may reduce performance, but we encourage experimentation.
263
+
264
+ Command R+’s grounded generation behavior takes a conversation as input (with an optional user-supplied system preamble, indicating task, context and desired output style), along with a list of retrieved document snippets. The document snippets should be chunks, rather than long documents, typically around 100-400 words per chunk. Document snippets consist of key-value pairs. The keys should be short descriptive strings, the values can be text or semi-structured.
265
+
266
+ By default, Command R+ will generate grounded responses by first predicting which documents are relevant, then predicting which ones it will cite, then generating an answer. Finally, it will then insert grounding spans into the answer. See below for an example. This is referred to as `accurate` grounded generation.
267
+
268
+ The model is trained with a number of other answering modes, which can be selected by prompt changes. A `fast` citation mode is supported in the tokenizer, which will directly generate an answer with grounding spans in it, without first writing the answer out in full. This sacrifices some grounding accuracy in favor of generating fewer tokens.
269
+
270
+ Comprehensive documentation for working with Command R+'s grounded generation prompt template can be found [here](https://docs.cohere.com/docs/prompting-command-r).
271
+
272
+ The code snippet below shows a minimal working example on how to render a prompt.
273
+
274
+ <details>
275
+ <summary> <b>Usage: Rendering Grounded Generation prompts [CLICK TO EXPAND]</b> </summary>
276
+
277
+ ````python
278
+ from transformers import AutoTokenizer
279
+
280
+ model_id = "CohereForAI/c4ai-command-r-plus"
281
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
282
+
283
+ # define conversation input:
284
+ conversation = [
285
+ {"role": "user", "content": "Whats the biggest penguin in the world?"}
286
+ ]
287
+ # define documents to ground on:
288
+ documents = [
289
+ { "title": "Tall penguins", "text": "Emperor penguins are the tallest growing up to 122 cm in height." },
290
+ { "title": "Penguin habitats", "text": "Emperor penguins only live in Antarctica."}
291
+ ]
292
+
293
+ # render the tool use prompt as a string:
294
+ grounded_generation_prompt = tokenizer.apply_grounded_generation_template(
295
+ conversation,
296
+ documents=documents,
297
+ citation_mode="accurate", # or "fast"
298
+ tokenize=False,
299
+ add_generation_prompt=True,
300
+ )
301
+ print(grounded_generation_prompt)
302
+ ````
303
+ </details>
304
+
305
+ <details>
306
+ <summary><b>Example Rendered Grounded Generation Prompt [CLICK TO EXPAND]</b></summary>
307
+
308
+ ````<BOS_TOKEN><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|># Safety Preamble
309
+ The instructions in this section override those in the task description and style guide sections. Don't answer questions that are harmful or immoral.
310
+
311
+ # System Preamble
312
+ ## Basic Rules
313
+ You are a powerful conversational AI trained by Cohere to help people. You are augmented by a number of tools, and your job is to use and consume the output of these tools to best help the user. You will see a conversation history between yourself and a user, ending with an utterance from the user. You will then see a specific instruction instructing you what kind of response to generate. When you answer the user's requests, you cite your sources in your answers, according to those instructions.
314
+
315
+ # User Preamble
316
+ ## Task and Context
317
+ You help people answer their questions and other requests interactively. You will be asked a very wide array of requests on all kinds of topics. You will be equipped with a wide range of search engines or similar tools to help you, which you use to research your answer. You should focus on serving the user's needs as best you can, which will be wide-ranging.
318
+
319
+ ## Style Guide
320
+ Unless the user asks for a different style of answer, you should answer in full sentences, using proper grammar and spelling.<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Whats the biggest penguin in the world?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|><results>
321
+ Document: 0
322
+ title: Tall penguins
323
+ text: Emperor penguins are the tallest growing up to 122 cm in height.
324
+
325
+ Document: 1
326
+ title: Penguin habitats
327
+ text: Emperor penguins only live in Antarctica.
328
+ </results><|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>Carefully perform the following instructions, in order, starting each with a new line.
329
+ Firstly, Decide which of the retrieved documents are relevant to the user's last input by writing 'Relevant Documents:' followed by comma-separated list of document numbers. If none are relevant, you should instead write 'None'.
330
+ Secondly, Decide which of the retrieved documents contain facts that should be cited in a good answer to the user's last input by writing 'Cited Documents:' followed a comma-separated list of document numbers. If you dont want to cite any of them, you should instead write 'None'.
331
+ Thirdly, Write 'Answer:' followed by a response to the user's last input in high quality natural english. Use the retrieved documents to help you. Do not insert any citations or grounding markup.
332
+ Finally, Write 'Grounded answer:' followed by a response to the user's last input in high quality natural english. Use the symbols <co: doc> and </co: doc> to indicate when a fact comes from a document in the search result, e.g <co: 0>my fact</co: 0> for a fact from document 0.<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
333
+ ````
334
+
335
+ </details>
336
+
337
+ <details>
338
+ <summary><b>Example Rendered Grounded Generation Completion [CLICK TO EXPAND]</b></summary>
339
+
340
+ ````
341
+ Relevant Documents: 0,1
342
+ Cited Documents: 0,1
343
+ Answer: The Emperor Penguin is the tallest or biggest penguin in the world. It is a bird that lives only in Antarctica and grows to a height of around 122 centimetres.
344
+ Grounded answer: The <co: 0>Emperor Penguin</co: 0> is the <co: 0>tallest</co: 0> or biggest penguin in the world. It is a bird that <co: 1>lives only in Antarctica</co: 1> and <co: 0>grows to a height of around 122 centimetres.</co: 0>
345
+ ````
346
+ </details>
347
+
348
+ ### Code Capabilities:
349
+ Command R+ has been optimized to interact with your code, by requesting code snippets, code explanations, or code rewrites. It might not perform well out-of-the-box for pure code completion. For better performance, we also recommend using a low temperature (and even greedy decoding) for code-generation related instructions.
350
+
351
+ ### Model Card Contact
352
+ For errors or additional questions about details in this model card, contact [info@for.ai](mailto:info@for.ai).
353
+
354
+ ### Terms of Use:
355
+ We hope that the release of this model will make community-based research efforts more accessible, by releasing the weights of a highly performant 104 billion parameter model to researchers all over the world. This model is governed by a [CC-BY-NC](https://cohere.com/c4ai-cc-by-nc-license) License with an acceptable use addendum, and also requires adhering to [C4AI's Acceptable Use Policy](https://docs.cohere.com/docs/c4ai-acceptable-use-policy).
356
+
357
+ ### Try Chat:
358
+ You can try Command R+ chat in the playground [here](https://dashboard.cohere.com/playground/chat). You can also use it in our dedicated Hugging Face Space [here](https://huggingface.co/spaces/CohereForAI/c4ai-command-r-plus).
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CohereForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 5,
8
+ "eos_token_id": 255001,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 12288,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 33792,
13
+ "layer_norm_eps": 1e-05,
14
+ "logit_scale": 0.8333333333333334,
15
+ "max_position_embeddings": 8192,
16
+ "model_type": "cohere",
17
+ "num_attention_heads": 96,
18
+ "num_hidden_layers": 64,
19
+ "num_key_value_heads": 8,
20
+ "pad_token_id": 0,
21
+ "rope_theta": 75000000.0,
22
+ "torch_dtype": "float16",
23
+ "transformers_version": "4.40.0.dev0",
24
+ "use_cache": true,
25
+ "use_qk_norm": true,
26
+ "vocab_size": 256000,
27
+ "quantization_config": {
28
+ "quant_method": "exl2",
29
+ "version": "0.0.18",
30
+ "bits": 4.25,
31
+ "head_bits": 6,
32
+ "calibration": {
33
+ "rows": 100,
34
+ "length": 2048,
35
+ "dataset": "(default)"
36
+ }
37
+ }
38
+ }
config.json.bak ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CohereForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 5,
8
+ "eos_token_id": 255001,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 12288,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 33792,
13
+ "layer_norm_eps": 1e-05,
14
+ "logit_scale": 0.8333333333333334,
15
+ "max_position_embeddings": 8192,
16
+ "model_type": "cohere",
17
+ "num_attention_heads": 96,
18
+ "num_hidden_layers": 64,
19
+ "num_key_value_heads": 8,
20
+ "pad_token_id": 0,
21
+ "rope_theta": 75000000.0,
22
+ "torch_dtype": "float16",
23
+ "transformers_version": "4.40.0.dev0",
24
+ "use_cache": true,
25
+ "use_qk_norm": true,
26
+ "vocab_size": 256000
27
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 5,
4
+ "eos_token_id": 255001,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.40.0.dev0"
7
+ }
gitattributes ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz filter=lfs diff=lfs merge=lfs -text
33
+ *.zip filter=lfs diff=lfs merge=lfs -text
34
+ *.zst filter=lfs diff=lfs merge=lfs -text
35
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
model.safetensors.index.json ADDED
@@ -0,0 +1,649 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 207621349376
4
+ },
5
+ "weight_map": {
6
+ "model.embed_tokens.weight": "model-00001-of-00044.safetensors",
7
+ "model.layers.0.input_layernorm.weight": "model-00002-of-00044.safetensors",
8
+ "model.layers.0.mlp.down_proj.weight": "model-00002-of-00044.safetensors",
9
+ "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00044.safetensors",
10
+ "model.layers.0.mlp.up_proj.weight": "model-00002-of-00044.safetensors",
11
+ "model.layers.0.self_attn.k_norm.weight": "model-00002-of-00044.safetensors",
12
+ "model.layers.0.self_attn.k_proj.weight": "model-00002-of-00044.safetensors",
13
+ "model.layers.0.self_attn.o_proj.weight": "model-00002-of-00044.safetensors",
14
+ "model.layers.0.self_attn.q_norm.weight": "model-00002-of-00044.safetensors",
15
+ "model.layers.0.self_attn.q_proj.weight": "model-00002-of-00044.safetensors",
16
+ "model.layers.0.self_attn.v_proj.weight": "model-00002-of-00044.safetensors",
17
+ "model.layers.1.input_layernorm.weight": "model-00003-of-00044.safetensors",
18
+ "model.layers.1.mlp.down_proj.weight": "model-00003-of-00044.safetensors",
19
+ "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00044.safetensors",
20
+ "model.layers.1.mlp.up_proj.weight": "model-00003-of-00044.safetensors",
21
+ "model.layers.1.self_attn.k_norm.weight": "model-00002-of-00044.safetensors",
22
+ "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00044.safetensors",
23
+ "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00044.safetensors",
24
+ "model.layers.1.self_attn.q_norm.weight": "model-00002-of-00044.safetensors",
25
+ "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00044.safetensors",
26
+ "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00044.safetensors",
27
+ "model.layers.10.input_layernorm.weight": "model-00009-of-00044.safetensors",
28
+ "model.layers.10.mlp.down_proj.weight": "model-00009-of-00044.safetensors",
29
+ "model.layers.10.mlp.gate_proj.weight": "model-00008-of-00044.safetensors",
30
+ "model.layers.10.mlp.up_proj.weight": "model-00009-of-00044.safetensors",
31
+ "model.layers.10.self_attn.k_norm.weight": "model-00008-of-00044.safetensors",
32
+ "model.layers.10.self_attn.k_proj.weight": "model-00008-of-00044.safetensors",
33
+ "model.layers.10.self_attn.o_proj.weight": "model-00008-of-00044.safetensors",
34
+ "model.layers.10.self_attn.q_norm.weight": "model-00008-of-00044.safetensors",
35
+ "model.layers.10.self_attn.q_proj.weight": "model-00008-of-00044.safetensors",
36
+ "model.layers.10.self_attn.v_proj.weight": "model-00008-of-00044.safetensors",
37
+ "model.layers.11.input_layernorm.weight": "model-00009-of-00044.safetensors",
38
+ "model.layers.11.mlp.down_proj.weight": "model-00009-of-00044.safetensors",
39
+ "model.layers.11.mlp.gate_proj.weight": "model-00009-of-00044.safetensors",
40
+ "model.layers.11.mlp.up_proj.weight": "model-00009-of-00044.safetensors",
41
+ "model.layers.11.self_attn.k_norm.weight": "model-00009-of-00044.safetensors",
42
+ "model.layers.11.self_attn.k_proj.weight": "model-00009-of-00044.safetensors",
43
+ "model.layers.11.self_attn.o_proj.weight": "model-00009-of-00044.safetensors",
44
+ "model.layers.11.self_attn.q_norm.weight": "model-00009-of-00044.safetensors",
45
+ "model.layers.11.self_attn.q_proj.weight": "model-00009-of-00044.safetensors",
46
+ "model.layers.11.self_attn.v_proj.weight": "model-00009-of-00044.safetensors",
47
+ "model.layers.12.input_layernorm.weight": "model-00010-of-00044.safetensors",
48
+ "model.layers.12.mlp.down_proj.weight": "model-00010-of-00044.safetensors",
49
+ "model.layers.12.mlp.gate_proj.weight": "model-00010-of-00044.safetensors",
50
+ "model.layers.12.mlp.up_proj.weight": "model-00010-of-00044.safetensors",
51
+ "model.layers.12.self_attn.k_norm.weight": "model-00009-of-00044.safetensors",
52
+ "model.layers.12.self_attn.k_proj.weight": "model-00010-of-00044.safetensors",
53
+ "model.layers.12.self_attn.o_proj.weight": "model-00010-of-00044.safetensors",
54
+ "model.layers.12.self_attn.q_norm.weight": "model-00009-of-00044.safetensors",
55
+ "model.layers.12.self_attn.q_proj.weight": "model-00010-of-00044.safetensors",
56
+ "model.layers.12.self_attn.v_proj.weight": "model-00010-of-00044.safetensors",
57
+ "model.layers.13.input_layernorm.weight": "model-00011-of-00044.safetensors",
58
+ "model.layers.13.mlp.down_proj.weight": "model-00011-of-00044.safetensors",
59
+ "model.layers.13.mlp.gate_proj.weight": "model-00010-of-00044.safetensors",
60
+ "model.layers.13.mlp.up_proj.weight": "model-00011-of-00044.safetensors",
61
+ "model.layers.13.self_attn.k_norm.weight": "model-00010-of-00044.safetensors",
62
+ "model.layers.13.self_attn.k_proj.weight": "model-00010-of-00044.safetensors",
63
+ "model.layers.13.self_attn.o_proj.weight": "model-00010-of-00044.safetensors",
64
+ "model.layers.13.self_attn.q_norm.weight": "model-00010-of-00044.safetensors",
65
+ "model.layers.13.self_attn.q_proj.weight": "model-00010-of-00044.safetensors",
66
+ "model.layers.13.self_attn.v_proj.weight": "model-00010-of-00044.safetensors",
67
+ "model.layers.14.input_layernorm.weight": "model-00011-of-00044.safetensors",
68
+ "model.layers.14.mlp.down_proj.weight": "model-00011-of-00044.safetensors",
69
+ "model.layers.14.mlp.gate_proj.weight": "model-00011-of-00044.safetensors",
70
+ "model.layers.14.mlp.up_proj.weight": "model-00011-of-00044.safetensors",
71
+ "model.layers.14.self_attn.k_norm.weight": "model-00011-of-00044.safetensors",
72
+ "model.layers.14.self_attn.k_proj.weight": "model-00011-of-00044.safetensors",
73
+ "model.layers.14.self_attn.o_proj.weight": "model-00011-of-00044.safetensors",
74
+ "model.layers.14.self_attn.q_norm.weight": "model-00011-of-00044.safetensors",
75
+ "model.layers.14.self_attn.q_proj.weight": "model-00011-of-00044.safetensors",
76
+ "model.layers.14.self_attn.v_proj.weight": "model-00011-of-00044.safetensors",
77
+ "model.layers.15.input_layernorm.weight": "model-00012-of-00044.safetensors",
78
+ "model.layers.15.mlp.down_proj.weight": "model-00012-of-00044.safetensors",
79
+ "model.layers.15.mlp.gate_proj.weight": "model-00012-of-00044.safetensors",
80
+ "model.layers.15.mlp.up_proj.weight": "model-00012-of-00044.safetensors",
81
+ "model.layers.15.self_attn.k_norm.weight": "model-00011-of-00044.safetensors",
82
+ "model.layers.15.self_attn.k_proj.weight": "model-00012-of-00044.safetensors",
83
+ "model.layers.15.self_attn.o_proj.weight": "model-00012-of-00044.safetensors",
84
+ "model.layers.15.self_attn.q_norm.weight": "model-00011-of-00044.safetensors",
85
+ "model.layers.15.self_attn.q_proj.weight": "model-00012-of-00044.safetensors",
86
+ "model.layers.15.self_attn.v_proj.weight": "model-00012-of-00044.safetensors",
87
+ "model.layers.16.input_layernorm.weight": "model-00013-of-00044.safetensors",
88
+ "model.layers.16.mlp.down_proj.weight": "model-00013-of-00044.safetensors",
89
+ "model.layers.16.mlp.gate_proj.weight": "model-00012-of-00044.safetensors",
90
+ "model.layers.16.mlp.up_proj.weight": "model-00013-of-00044.safetensors",
91
+ "model.layers.16.self_attn.k_norm.weight": "model-00012-of-00044.safetensors",
92
+ "model.layers.16.self_attn.k_proj.weight": "model-00012-of-00044.safetensors",
93
+ "model.layers.16.self_attn.o_proj.weight": "model-00012-of-00044.safetensors",
94
+ "model.layers.16.self_attn.q_norm.weight": "model-00012-of-00044.safetensors",
95
+ "model.layers.16.self_attn.q_proj.weight": "model-00012-of-00044.safetensors",
96
+ "model.layers.16.self_attn.v_proj.weight": "model-00012-of-00044.safetensors",
97
+ "model.layers.17.input_layernorm.weight": "model-00013-of-00044.safetensors",
98
+ "model.layers.17.mlp.down_proj.weight": "model-00013-of-00044.safetensors",
99
+ "model.layers.17.mlp.gate_proj.weight": "model-00013-of-00044.safetensors",
100
+ "model.layers.17.mlp.up_proj.weight": "model-00013-of-00044.safetensors",
101
+ "model.layers.17.self_attn.k_norm.weight": "model-00013-of-00044.safetensors",
102
+ "model.layers.17.self_attn.k_proj.weight": "model-00013-of-00044.safetensors",
103
+ "model.layers.17.self_attn.o_proj.weight": "model-00013-of-00044.safetensors",
104
+ "model.layers.17.self_attn.q_norm.weight": "model-00013-of-00044.safetensors",
105
+ "model.layers.17.self_attn.q_proj.weight": "model-00013-of-00044.safetensors",
106
+ "model.layers.17.self_attn.v_proj.weight": "model-00013-of-00044.safetensors",
107
+ "model.layers.18.input_layernorm.weight": "model-00014-of-00044.safetensors",
108
+ "model.layers.18.mlp.down_proj.weight": "model-00014-of-00044.safetensors",
109
+ "model.layers.18.mlp.gate_proj.weight": "model-00014-of-00044.safetensors",
110
+ "model.layers.18.mlp.up_proj.weight": "model-00014-of-00044.safetensors",
111
+ "model.layers.18.self_attn.k_norm.weight": "model-00013-of-00044.safetensors",
112
+ "model.layers.18.self_attn.k_proj.weight": "model-00014-of-00044.safetensors",
113
+ "model.layers.18.self_attn.o_proj.weight": "model-00014-of-00044.safetensors",
114
+ "model.layers.18.self_attn.q_norm.weight": "model-00013-of-00044.safetensors",
115
+ "model.layers.18.self_attn.q_proj.weight": "model-00014-of-00044.safetensors",
116
+ "model.layers.18.self_attn.v_proj.weight": "model-00014-of-00044.safetensors",
117
+ "model.layers.19.input_layernorm.weight": "model-00015-of-00044.safetensors",
118
+ "model.layers.19.mlp.down_proj.weight": "model-00015-of-00044.safetensors",
119
+ "model.layers.19.mlp.gate_proj.weight": "model-00014-of-00044.safetensors",
120
+ "model.layers.19.mlp.up_proj.weight": "model-00015-of-00044.safetensors",
121
+ "model.layers.19.self_attn.k_norm.weight": "model-00014-of-00044.safetensors",
122
+ "model.layers.19.self_attn.k_proj.weight": "model-00014-of-00044.safetensors",
123
+ "model.layers.19.self_attn.o_proj.weight": "model-00014-of-00044.safetensors",
124
+ "model.layers.19.self_attn.q_norm.weight": "model-00014-of-00044.safetensors",
125
+ "model.layers.19.self_attn.q_proj.weight": "model-00014-of-00044.safetensors",
126
+ "model.layers.19.self_attn.v_proj.weight": "model-00014-of-00044.safetensors",
127
+ "model.layers.2.input_layernorm.weight": "model-00003-of-00044.safetensors",
128
+ "model.layers.2.mlp.down_proj.weight": "model-00003-of-00044.safetensors",
129
+ "model.layers.2.mlp.gate_proj.weight": "model-00003-of-00044.safetensors",
130
+ "model.layers.2.mlp.up_proj.weight": "model-00003-of-00044.safetensors",
131
+ "model.layers.2.self_attn.k_norm.weight": "model-00003-of-00044.safetensors",
132
+ "model.layers.2.self_attn.k_proj.weight": "model-00003-of-00044.safetensors",
133
+ "model.layers.2.self_attn.o_proj.weight": "model-00003-of-00044.safetensors",
134
+ "model.layers.2.self_attn.q_norm.weight": "model-00003-of-00044.safetensors",
135
+ "model.layers.2.self_attn.q_proj.weight": "model-00003-of-00044.safetensors",
136
+ "model.layers.2.self_attn.v_proj.weight": "model-00003-of-00044.safetensors",
137
+ "model.layers.20.input_layernorm.weight": "model-00015-of-00044.safetensors",
138
+ "model.layers.20.mlp.down_proj.weight": "model-00015-of-00044.safetensors",
139
+ "model.layers.20.mlp.gate_proj.weight": "model-00015-of-00044.safetensors",
140
+ "model.layers.20.mlp.up_proj.weight": "model-00015-of-00044.safetensors",
141
+ "model.layers.20.self_attn.k_norm.weight": "model-00015-of-00044.safetensors",
142
+ "model.layers.20.self_attn.k_proj.weight": "model-00015-of-00044.safetensors",
143
+ "model.layers.20.self_attn.o_proj.weight": "model-00015-of-00044.safetensors",
144
+ "model.layers.20.self_attn.q_norm.weight": "model-00015-of-00044.safetensors",
145
+ "model.layers.20.self_attn.q_proj.weight": "model-00015-of-00044.safetensors",
146
+ "model.layers.20.self_attn.v_proj.weight": "model-00015-of-00044.safetensors",
147
+ "model.layers.21.input_layernorm.weight": "model-00016-of-00044.safetensors",
148
+ "model.layers.21.mlp.down_proj.weight": "model-00016-of-00044.safetensors",
149
+ "model.layers.21.mlp.gate_proj.weight": "model-00016-of-00044.safetensors",
150
+ "model.layers.21.mlp.up_proj.weight": "model-00016-of-00044.safetensors",
151
+ "model.layers.21.self_attn.k_norm.weight": "model-00015-of-00044.safetensors",
152
+ "model.layers.21.self_attn.k_proj.weight": "model-00016-of-00044.safetensors",
153
+ "model.layers.21.self_attn.o_proj.weight": "model-00016-of-00044.safetensors",
154
+ "model.layers.21.self_attn.q_norm.weight": "model-00015-of-00044.safetensors",
155
+ "model.layers.21.self_attn.q_proj.weight": "model-00016-of-00044.safetensors",
156
+ "model.layers.21.self_attn.v_proj.weight": "model-00016-of-00044.safetensors",
157
+ "model.layers.22.input_layernorm.weight": "model-00017-of-00044.safetensors",
158
+ "model.layers.22.mlp.down_proj.weight": "model-00017-of-00044.safetensors",
159
+ "model.layers.22.mlp.gate_proj.weight": "model-00016-of-00044.safetensors",
160
+ "model.layers.22.mlp.up_proj.weight": "model-00017-of-00044.safetensors",
161
+ "model.layers.22.self_attn.k_norm.weight": "model-00016-of-00044.safetensors",
162
+ "model.layers.22.self_attn.k_proj.weight": "model-00016-of-00044.safetensors",
163
+ "model.layers.22.self_attn.o_proj.weight": "model-00016-of-00044.safetensors",
164
+ "model.layers.22.self_attn.q_norm.weight": "model-00016-of-00044.safetensors",
165
+ "model.layers.22.self_attn.q_proj.weight": "model-00016-of-00044.safetensors",
166
+ "model.layers.22.self_attn.v_proj.weight": "model-00016-of-00044.safetensors",
167
+ "model.layers.23.input_layernorm.weight": "model-00017-of-00044.safetensors",
168
+ "model.layers.23.mlp.down_proj.weight": "model-00017-of-00044.safetensors",
169
+ "model.layers.23.mlp.gate_proj.weight": "model-00017-of-00044.safetensors",
170
+ "model.layers.23.mlp.up_proj.weight": "model-00017-of-00044.safetensors",
171
+ "model.layers.23.self_attn.k_norm.weight": "model-00017-of-00044.safetensors",
172
+ "model.layers.23.self_attn.k_proj.weight": "model-00017-of-00044.safetensors",
173
+ "model.layers.23.self_attn.o_proj.weight": "model-00017-of-00044.safetensors",
174
+ "model.layers.23.self_attn.q_norm.weight": "model-00017-of-00044.safetensors",
175
+ "model.layers.23.self_attn.q_proj.weight": "model-00017-of-00044.safetensors",
176
+ "model.layers.23.self_attn.v_proj.weight": "model-00017-of-00044.safetensors",
177
+ "model.layers.24.input_layernorm.weight": "model-00018-of-00044.safetensors",
178
+ "model.layers.24.mlp.down_proj.weight": "model-00018-of-00044.safetensors",
179
+ "model.layers.24.mlp.gate_proj.weight": "model-00018-of-00044.safetensors",
180
+ "model.layers.24.mlp.up_proj.weight": "model-00018-of-00044.safetensors",
181
+ "model.layers.24.self_attn.k_norm.weight": "model-00017-of-00044.safetensors",
182
+ "model.layers.24.self_attn.k_proj.weight": "model-00018-of-00044.safetensors",
183
+ "model.layers.24.self_attn.o_proj.weight": "model-00018-of-00044.safetensors",
184
+ "model.layers.24.self_attn.q_norm.weight": "model-00017-of-00044.safetensors",
185
+ "model.layers.24.self_attn.q_proj.weight": "model-00018-of-00044.safetensors",
186
+ "model.layers.24.self_attn.v_proj.weight": "model-00018-of-00044.safetensors",
187
+ "model.layers.25.input_layernorm.weight": "model-00019-of-00044.safetensors",
188
+ "model.layers.25.mlp.down_proj.weight": "model-00019-of-00044.safetensors",
189
+ "model.layers.25.mlp.gate_proj.weight": "model-00018-of-00044.safetensors",
190
+ "model.layers.25.mlp.up_proj.weight": "model-00019-of-00044.safetensors",
191
+ "model.layers.25.self_attn.k_norm.weight": "model-00018-of-00044.safetensors",
192
+ "model.layers.25.self_attn.k_proj.weight": "model-00018-of-00044.safetensors",
193
+ "model.layers.25.self_attn.o_proj.weight": "model-00018-of-00044.safetensors",
194
+ "model.layers.25.self_attn.q_norm.weight": "model-00018-of-00044.safetensors",
195
+ "model.layers.25.self_attn.q_proj.weight": "model-00018-of-00044.safetensors",
196
+ "model.layers.25.self_attn.v_proj.weight": "model-00018-of-00044.safetensors",
197
+ "model.layers.26.input_layernorm.weight": "model-00019-of-00044.safetensors",
198
+ "model.layers.26.mlp.down_proj.weight": "model-00019-of-00044.safetensors",
199
+ "model.layers.26.mlp.gate_proj.weight": "model-00019-of-00044.safetensors",
200
+ "model.layers.26.mlp.up_proj.weight": "model-00019-of-00044.safetensors",
201
+ "model.layers.26.self_attn.k_norm.weight": "model-00019-of-00044.safetensors",
202
+ "model.layers.26.self_attn.k_proj.weight": "model-00019-of-00044.safetensors",
203
+ "model.layers.26.self_attn.o_proj.weight": "model-00019-of-00044.safetensors",
204
+ "model.layers.26.self_attn.q_norm.weight": "model-00019-of-00044.safetensors",
205
+ "model.layers.26.self_attn.q_proj.weight": "model-00019-of-00044.safetensors",
206
+ "model.layers.26.self_attn.v_proj.weight": "model-00019-of-00044.safetensors",
207
+ "model.layers.27.input_layernorm.weight": "model-00020-of-00044.safetensors",
208
+ "model.layers.27.mlp.down_proj.weight": "model-00020-of-00044.safetensors",
209
+ "model.layers.27.mlp.gate_proj.weight": "model-00020-of-00044.safetensors",
210
+ "model.layers.27.mlp.up_proj.weight": "model-00020-of-00044.safetensors",
211
+ "model.layers.27.self_attn.k_norm.weight": "model-00019-of-00044.safetensors",
212
+ "model.layers.27.self_attn.k_proj.weight": "model-00020-of-00044.safetensors",
213
+ "model.layers.27.self_attn.o_proj.weight": "model-00020-of-00044.safetensors",
214
+ "model.layers.27.self_attn.q_norm.weight": "model-00019-of-00044.safetensors",
215
+ "model.layers.27.self_attn.q_proj.weight": "model-00020-of-00044.safetensors",
216
+ "model.layers.27.self_attn.v_proj.weight": "model-00020-of-00044.safetensors",
217
+ "model.layers.28.input_layernorm.weight": "model-00021-of-00044.safetensors",
218
+ "model.layers.28.mlp.down_proj.weight": "model-00021-of-00044.safetensors",
219
+ "model.layers.28.mlp.gate_proj.weight": "model-00020-of-00044.safetensors",
220
+ "model.layers.28.mlp.up_proj.weight": "model-00021-of-00044.safetensors",
221
+ "model.layers.28.self_attn.k_norm.weight": "model-00020-of-00044.safetensors",
222
+ "model.layers.28.self_attn.k_proj.weight": "model-00020-of-00044.safetensors",
223
+ "model.layers.28.self_attn.o_proj.weight": "model-00020-of-00044.safetensors",
224
+ "model.layers.28.self_attn.q_norm.weight": "model-00020-of-00044.safetensors",
225
+ "model.layers.28.self_attn.q_proj.weight": "model-00020-of-00044.safetensors",
226
+ "model.layers.28.self_attn.v_proj.weight": "model-00020-of-00044.safetensors",
227
+ "model.layers.29.input_layernorm.weight": "model-00021-of-00044.safetensors",
228
+ "model.layers.29.mlp.down_proj.weight": "model-00021-of-00044.safetensors",
229
+ "model.layers.29.mlp.gate_proj.weight": "model-00021-of-00044.safetensors",
230
+ "model.layers.29.mlp.up_proj.weight": "model-00021-of-00044.safetensors",
231
+ "model.layers.29.self_attn.k_norm.weight": "model-00021-of-00044.safetensors",
232
+ "model.layers.29.self_attn.k_proj.weight": "model-00021-of-00044.safetensors",
233
+ "model.layers.29.self_attn.o_proj.weight": "model-00021-of-00044.safetensors",
234
+ "model.layers.29.self_attn.q_norm.weight": "model-00021-of-00044.safetensors",
235
+ "model.layers.29.self_attn.q_proj.weight": "model-00021-of-00044.safetensors",
236
+ "model.layers.29.self_attn.v_proj.weight": "model-00021-of-00044.safetensors",
237
+ "model.layers.3.input_layernorm.weight": "model-00004-of-00044.safetensors",
238
+ "model.layers.3.mlp.down_proj.weight": "model-00004-of-00044.safetensors",
239
+ "model.layers.3.mlp.gate_proj.weight": "model-00004-of-00044.safetensors",
240
+ "model.layers.3.mlp.up_proj.weight": "model-00004-of-00044.safetensors",
241
+ "model.layers.3.self_attn.k_norm.weight": "model-00003-of-00044.safetensors",
242
+ "model.layers.3.self_attn.k_proj.weight": "model-00004-of-00044.safetensors",
243
+ "model.layers.3.self_attn.o_proj.weight": "model-00004-of-00044.safetensors",
244
+ "model.layers.3.self_attn.q_norm.weight": "model-00003-of-00044.safetensors",
245
+ "model.layers.3.self_attn.q_proj.weight": "model-00004-of-00044.safetensors",
246
+ "model.layers.3.self_attn.v_proj.weight": "model-00004-of-00044.safetensors",
247
+ "model.layers.30.input_layernorm.weight": "model-00022-of-00044.safetensors",
248
+ "model.layers.30.mlp.down_proj.weight": "model-00022-of-00044.safetensors",
249
+ "model.layers.30.mlp.gate_proj.weight": "model-00022-of-00044.safetensors",
250
+ "model.layers.30.mlp.up_proj.weight": "model-00022-of-00044.safetensors",
251
+ "model.layers.30.self_attn.k_norm.weight": "model-00021-of-00044.safetensors",
252
+ "model.layers.30.self_attn.k_proj.weight": "model-00022-of-00044.safetensors",
253
+ "model.layers.30.self_attn.o_proj.weight": "model-00022-of-00044.safetensors",
254
+ "model.layers.30.self_attn.q_norm.weight": "model-00021-of-00044.safetensors",
255
+ "model.layers.30.self_attn.q_proj.weight": "model-00022-of-00044.safetensors",
256
+ "model.layers.30.self_attn.v_proj.weight": "model-00022-of-00044.safetensors",
257
+ "model.layers.31.input_layernorm.weight": "model-00023-of-00044.safetensors",
258
+ "model.layers.31.mlp.down_proj.weight": "model-00023-of-00044.safetensors",
259
+ "model.layers.31.mlp.gate_proj.weight": "model-00022-of-00044.safetensors",
260
+ "model.layers.31.mlp.up_proj.weight": "model-00023-of-00044.safetensors",
261
+ "model.layers.31.self_attn.k_norm.weight": "model-00022-of-00044.safetensors",
262
+ "model.layers.31.self_attn.k_proj.weight": "model-00022-of-00044.safetensors",
263
+ "model.layers.31.self_attn.o_proj.weight": "model-00022-of-00044.safetensors",
264
+ "model.layers.31.self_attn.q_norm.weight": "model-00022-of-00044.safetensors",
265
+ "model.layers.31.self_attn.q_proj.weight": "model-00022-of-00044.safetensors",
266
+ "model.layers.31.self_attn.v_proj.weight": "model-00022-of-00044.safetensors",
267
+ "model.layers.32.input_layernorm.weight": "model-00023-of-00044.safetensors",
268
+ "model.layers.32.mlp.down_proj.weight": "model-00023-of-00044.safetensors",
269
+ "model.layers.32.mlp.gate_proj.weight": "model-00023-of-00044.safetensors",
270
+ "model.layers.32.mlp.up_proj.weight": "model-00023-of-00044.safetensors",
271
+ "model.layers.32.self_attn.k_norm.weight": "model-00023-of-00044.safetensors",
272
+ "model.layers.32.self_attn.k_proj.weight": "model-00023-of-00044.safetensors",
273
+ "model.layers.32.self_attn.o_proj.weight": "model-00023-of-00044.safetensors",
274
+ "model.layers.32.self_attn.q_norm.weight": "model-00023-of-00044.safetensors",
275
+ "model.layers.32.self_attn.q_proj.weight": "model-00023-of-00044.safetensors",
276
+ "model.layers.32.self_attn.v_proj.weight": "model-00023-of-00044.safetensors",
277
+ "model.layers.33.input_layernorm.weight": "model-00024-of-00044.safetensors",
278
+ "model.layers.33.mlp.down_proj.weight": "model-00024-of-00044.safetensors",
279
+ "model.layers.33.mlp.gate_proj.weight": "model-00024-of-00044.safetensors",
280
+ "model.layers.33.mlp.up_proj.weight": "model-00024-of-00044.safetensors",
281
+ "model.layers.33.self_attn.k_norm.weight": "model-00023-of-00044.safetensors",
282
+ "model.layers.33.self_attn.k_proj.weight": "model-00024-of-00044.safetensors",
283
+ "model.layers.33.self_attn.o_proj.weight": "model-00024-of-00044.safetensors",
284
+ "model.layers.33.self_attn.q_norm.weight": "model-00023-of-00044.safetensors",
285
+ "model.layers.33.self_attn.q_proj.weight": "model-00024-of-00044.safetensors",
286
+ "model.layers.33.self_attn.v_proj.weight": "model-00024-of-00044.safetensors",
287
+ "model.layers.34.input_layernorm.weight": "model-00025-of-00044.safetensors",
288
+ "model.layers.34.mlp.down_proj.weight": "model-00025-of-00044.safetensors",
289
+ "model.layers.34.mlp.gate_proj.weight": "model-00024-of-00044.safetensors",
290
+ "model.layers.34.mlp.up_proj.weight": "model-00025-of-00044.safetensors",
291
+ "model.layers.34.self_attn.k_norm.weight": "model-00024-of-00044.safetensors",
292
+ "model.layers.34.self_attn.k_proj.weight": "model-00024-of-00044.safetensors",
293
+ "model.layers.34.self_attn.o_proj.weight": "model-00024-of-00044.safetensors",
294
+ "model.layers.34.self_attn.q_norm.weight": "model-00024-of-00044.safetensors",
295
+ "model.layers.34.self_attn.q_proj.weight": "model-00024-of-00044.safetensors",
296
+ "model.layers.34.self_attn.v_proj.weight": "model-00024-of-00044.safetensors",
297
+ "model.layers.35.input_layernorm.weight": "model-00025-of-00044.safetensors",
298
+ "model.layers.35.mlp.down_proj.weight": "model-00025-of-00044.safetensors",
299
+ "model.layers.35.mlp.gate_proj.weight": "model-00025-of-00044.safetensors",
300
+ "model.layers.35.mlp.up_proj.weight": "model-00025-of-00044.safetensors",
301
+ "model.layers.35.self_attn.k_norm.weight": "model-00025-of-00044.safetensors",
302
+ "model.layers.35.self_attn.k_proj.weight": "model-00025-of-00044.safetensors",
303
+ "model.layers.35.self_attn.o_proj.weight": "model-00025-of-00044.safetensors",
304
+ "model.layers.35.self_attn.q_norm.weight": "model-00025-of-00044.safetensors",
305
+ "model.layers.35.self_attn.q_proj.weight": "model-00025-of-00044.safetensors",
306
+ "model.layers.35.self_attn.v_proj.weight": "model-00025-of-00044.safetensors",
307
+ "model.layers.36.input_layernorm.weight": "model-00026-of-00044.safetensors",
308
+ "model.layers.36.mlp.down_proj.weight": "model-00026-of-00044.safetensors",
309
+ "model.layers.36.mlp.gate_proj.weight": "model-00026-of-00044.safetensors",
310
+ "model.layers.36.mlp.up_proj.weight": "model-00026-of-00044.safetensors",
311
+ "model.layers.36.self_attn.k_norm.weight": "model-00025-of-00044.safetensors",
312
+ "model.layers.36.self_attn.k_proj.weight": "model-00026-of-00044.safetensors",
313
+ "model.layers.36.self_attn.o_proj.weight": "model-00026-of-00044.safetensors",
314
+ "model.layers.36.self_attn.q_norm.weight": "model-00025-of-00044.safetensors",
315
+ "model.layers.36.self_attn.q_proj.weight": "model-00026-of-00044.safetensors",
316
+ "model.layers.36.self_attn.v_proj.weight": "model-00026-of-00044.safetensors",
317
+ "model.layers.37.input_layernorm.weight": "model-00027-of-00044.safetensors",
318
+ "model.layers.37.mlp.down_proj.weight": "model-00027-of-00044.safetensors",
319
+ "model.layers.37.mlp.gate_proj.weight": "model-00026-of-00044.safetensors",
320
+ "model.layers.37.mlp.up_proj.weight": "model-00027-of-00044.safetensors",
321
+ "model.layers.37.self_attn.k_norm.weight": "model-00026-of-00044.safetensors",
322
+ "model.layers.37.self_attn.k_proj.weight": "model-00026-of-00044.safetensors",
323
+ "model.layers.37.self_attn.o_proj.weight": "model-00026-of-00044.safetensors",
324
+ "model.layers.37.self_attn.q_norm.weight": "model-00026-of-00044.safetensors",
325
+ "model.layers.37.self_attn.q_proj.weight": "model-00026-of-00044.safetensors",
326
+ "model.layers.37.self_attn.v_proj.weight": "model-00026-of-00044.safetensors",
327
+ "model.layers.38.input_layernorm.weight": "model-00027-of-00044.safetensors",
328
+ "model.layers.38.mlp.down_proj.weight": "model-00027-of-00044.safetensors",
329
+ "model.layers.38.mlp.gate_proj.weight": "model-00027-of-00044.safetensors",
330
+ "model.layers.38.mlp.up_proj.weight": "model-00027-of-00044.safetensors",
331
+ "model.layers.38.self_attn.k_norm.weight": "model-00027-of-00044.safetensors",
332
+ "model.layers.38.self_attn.k_proj.weight": "model-00027-of-00044.safetensors",
333
+ "model.layers.38.self_attn.o_proj.weight": "model-00027-of-00044.safetensors",
334
+ "model.layers.38.self_attn.q_norm.weight": "model-00027-of-00044.safetensors",
335
+ "model.layers.38.self_attn.q_proj.weight": "model-00027-of-00044.safetensors",
336
+ "model.layers.38.self_attn.v_proj.weight": "model-00027-of-00044.safetensors",
337
+ "model.layers.39.input_layernorm.weight": "model-00028-of-00044.safetensors",
338
+ "model.layers.39.mlp.down_proj.weight": "model-00028-of-00044.safetensors",
339
+ "model.layers.39.mlp.gate_proj.weight": "model-00028-of-00044.safetensors",
340
+ "model.layers.39.mlp.up_proj.weight": "model-00028-of-00044.safetensors",
341
+ "model.layers.39.self_attn.k_norm.weight": "model-00027-of-00044.safetensors",
342
+ "model.layers.39.self_attn.k_proj.weight": "model-00028-of-00044.safetensors",
343
+ "model.layers.39.self_attn.o_proj.weight": "model-00028-of-00044.safetensors",
344
+ "model.layers.39.self_attn.q_norm.weight": "model-00027-of-00044.safetensors",
345
+ "model.layers.39.self_attn.q_proj.weight": "model-00028-of-00044.safetensors",
346
+ "model.layers.39.self_attn.v_proj.weight": "model-00028-of-00044.safetensors",
347
+ "model.layers.4.input_layernorm.weight": "model-00005-of-00044.safetensors",
348
+ "model.layers.4.mlp.down_proj.weight": "model-00005-of-00044.safetensors",
349
+ "model.layers.4.mlp.gate_proj.weight": "model-00004-of-00044.safetensors",
350
+ "model.layers.4.mlp.up_proj.weight": "model-00005-of-00044.safetensors",
351
+ "model.layers.4.self_attn.k_norm.weight": "model-00004-of-00044.safetensors",
352
+ "model.layers.4.self_attn.k_proj.weight": "model-00004-of-00044.safetensors",
353
+ "model.layers.4.self_attn.o_proj.weight": "model-00004-of-00044.safetensors",
354
+ "model.layers.4.self_attn.q_norm.weight": "model-00004-of-00044.safetensors",
355
+ "model.layers.4.self_attn.q_proj.weight": "model-00004-of-00044.safetensors",
356
+ "model.layers.4.self_attn.v_proj.weight": "model-00004-of-00044.safetensors",
357
+ "model.layers.40.input_layernorm.weight": "model-00029-of-00044.safetensors",
358
+ "model.layers.40.mlp.down_proj.weight": "model-00029-of-00044.safetensors",
359
+ "model.layers.40.mlp.gate_proj.weight": "model-00028-of-00044.safetensors",
360
+ "model.layers.40.mlp.up_proj.weight": "model-00029-of-00044.safetensors",
361
+ "model.layers.40.self_attn.k_norm.weight": "model-00028-of-00044.safetensors",
362
+ "model.layers.40.self_attn.k_proj.weight": "model-00028-of-00044.safetensors",
363
+ "model.layers.40.self_attn.o_proj.weight": "model-00028-of-00044.safetensors",
364
+ "model.layers.40.self_attn.q_norm.weight": "model-00028-of-00044.safetensors",
365
+ "model.layers.40.self_attn.q_proj.weight": "model-00028-of-00044.safetensors",
366
+ "model.layers.40.self_attn.v_proj.weight": "model-00028-of-00044.safetensors",
367
+ "model.layers.41.input_layernorm.weight": "model-00029-of-00044.safetensors",
368
+ "model.layers.41.mlp.down_proj.weight": "model-00029-of-00044.safetensors",
369
+ "model.layers.41.mlp.gate_proj.weight": "model-00029-of-00044.safetensors",
370
+ "model.layers.41.mlp.up_proj.weight": "model-00029-of-00044.safetensors",
371
+ "model.layers.41.self_attn.k_norm.weight": "model-00029-of-00044.safetensors",
372
+ "model.layers.41.self_attn.k_proj.weight": "model-00029-of-00044.safetensors",
373
+ "model.layers.41.self_attn.o_proj.weight": "model-00029-of-00044.safetensors",
374
+ "model.layers.41.self_attn.q_norm.weight": "model-00029-of-00044.safetensors",
375
+ "model.layers.41.self_attn.q_proj.weight": "model-00029-of-00044.safetensors",
376
+ "model.layers.41.self_attn.v_proj.weight": "model-00029-of-00044.safetensors",
377
+ "model.layers.42.input_layernorm.weight": "model-00030-of-00044.safetensors",
378
+ "model.layers.42.mlp.down_proj.weight": "model-00030-of-00044.safetensors",
379
+ "model.layers.42.mlp.gate_proj.weight": "model-00030-of-00044.safetensors",
380
+ "model.layers.42.mlp.up_proj.weight": "model-00030-of-00044.safetensors",
381
+ "model.layers.42.self_attn.k_norm.weight": "model-00029-of-00044.safetensors",
382
+ "model.layers.42.self_attn.k_proj.weight": "model-00030-of-00044.safetensors",
383
+ "model.layers.42.self_attn.o_proj.weight": "model-00030-of-00044.safetensors",
384
+ "model.layers.42.self_attn.q_norm.weight": "model-00029-of-00044.safetensors",
385
+ "model.layers.42.self_attn.q_proj.weight": "model-00030-of-00044.safetensors",
386
+ "model.layers.42.self_attn.v_proj.weight": "model-00030-of-00044.safetensors",
387
+ "model.layers.43.input_layernorm.weight": "model-00031-of-00044.safetensors",
388
+ "model.layers.43.mlp.down_proj.weight": "model-00031-of-00044.safetensors",
389
+ "model.layers.43.mlp.gate_proj.weight": "model-00030-of-00044.safetensors",
390
+ "model.layers.43.mlp.up_proj.weight": "model-00031-of-00044.safetensors",
391
+ "model.layers.43.self_attn.k_norm.weight": "model-00030-of-00044.safetensors",
392
+ "model.layers.43.self_attn.k_proj.weight": "model-00030-of-00044.safetensors",
393
+ "model.layers.43.self_attn.o_proj.weight": "model-00030-of-00044.safetensors",
394
+ "model.layers.43.self_attn.q_norm.weight": "model-00030-of-00044.safetensors",
395
+ "model.layers.43.self_attn.q_proj.weight": "model-00030-of-00044.safetensors",
396
+ "model.layers.43.self_attn.v_proj.weight": "model-00030-of-00044.safetensors",
397
+ "model.layers.44.input_layernorm.weight": "model-00031-of-00044.safetensors",
398
+ "model.layers.44.mlp.down_proj.weight": "model-00031-of-00044.safetensors",
399
+ "model.layers.44.mlp.gate_proj.weight": "model-00031-of-00044.safetensors",
400
+ "model.layers.44.mlp.up_proj.weight": "model-00031-of-00044.safetensors",
401
+ "model.layers.44.self_attn.k_norm.weight": "model-00031-of-00044.safetensors",
402
+ "model.layers.44.self_attn.k_proj.weight": "model-00031-of-00044.safetensors",
403
+ "model.layers.44.self_attn.o_proj.weight": "model-00031-of-00044.safetensors",
404
+ "model.layers.44.self_attn.q_norm.weight": "model-00031-of-00044.safetensors",
405
+ "model.layers.44.self_attn.q_proj.weight": "model-00031-of-00044.safetensors",
406
+ "model.layers.44.self_attn.v_proj.weight": "model-00031-of-00044.safetensors",
407
+ "model.layers.45.input_layernorm.weight": "model-00032-of-00044.safetensors",
408
+ "model.layers.45.mlp.down_proj.weight": "model-00032-of-00044.safetensors",
409
+ "model.layers.45.mlp.gate_proj.weight": "model-00032-of-00044.safetensors",
410
+ "model.layers.45.mlp.up_proj.weight": "model-00032-of-00044.safetensors",
411
+ "model.layers.45.self_attn.k_norm.weight": "model-00031-of-00044.safetensors",
412
+ "model.layers.45.self_attn.k_proj.weight": "model-00032-of-00044.safetensors",
413
+ "model.layers.45.self_attn.o_proj.weight": "model-00032-of-00044.safetensors",
414
+ "model.layers.45.self_attn.q_norm.weight": "model-00031-of-00044.safetensors",
415
+ "model.layers.45.self_attn.q_proj.weight": "model-00032-of-00044.safetensors",
416
+ "model.layers.45.self_attn.v_proj.weight": "model-00032-of-00044.safetensors",
417
+ "model.layers.46.input_layernorm.weight": "model-00033-of-00044.safetensors",
418
+ "model.layers.46.mlp.down_proj.weight": "model-00033-of-00044.safetensors",
419
+ "model.layers.46.mlp.gate_proj.weight": "model-00032-of-00044.safetensors",
420
+ "model.layers.46.mlp.up_proj.weight": "model-00033-of-00044.safetensors",
421
+ "model.layers.46.self_attn.k_norm.weight": "model-00032-of-00044.safetensors",
422
+ "model.layers.46.self_attn.k_proj.weight": "model-00032-of-00044.safetensors",
423
+ "model.layers.46.self_attn.o_proj.weight": "model-00032-of-00044.safetensors",
424
+ "model.layers.46.self_attn.q_norm.weight": "model-00032-of-00044.safetensors",
425
+ "model.layers.46.self_attn.q_proj.weight": "model-00032-of-00044.safetensors",
426
+ "model.layers.46.self_attn.v_proj.weight": "model-00032-of-00044.safetensors",
427
+ "model.layers.47.input_layernorm.weight": "model-00033-of-00044.safetensors",
428
+ "model.layers.47.mlp.down_proj.weight": "model-00033-of-00044.safetensors",
429
+ "model.layers.47.mlp.gate_proj.weight": "model-00033-of-00044.safetensors",
430
+ "model.layers.47.mlp.up_proj.weight": "model-00033-of-00044.safetensors",
431
+ "model.layers.47.self_attn.k_norm.weight": "model-00033-of-00044.safetensors",
432
+ "model.layers.47.self_attn.k_proj.weight": "model-00033-of-00044.safetensors",
433
+ "model.layers.47.self_attn.o_proj.weight": "model-00033-of-00044.safetensors",
434
+ "model.layers.47.self_attn.q_norm.weight": "model-00033-of-00044.safetensors",
435
+ "model.layers.47.self_attn.q_proj.weight": "model-00033-of-00044.safetensors",
436
+ "model.layers.47.self_attn.v_proj.weight": "model-00033-of-00044.safetensors",
437
+ "model.layers.48.input_layernorm.weight": "model-00034-of-00044.safetensors",
438
+ "model.layers.48.mlp.down_proj.weight": "model-00034-of-00044.safetensors",
439
+ "model.layers.48.mlp.gate_proj.weight": "model-00034-of-00044.safetensors",
440
+ "model.layers.48.mlp.up_proj.weight": "model-00034-of-00044.safetensors",
441
+ "model.layers.48.self_attn.k_norm.weight": "model-00033-of-00044.safetensors",
442
+ "model.layers.48.self_attn.k_proj.weight": "model-00034-of-00044.safetensors",
443
+ "model.layers.48.self_attn.o_proj.weight": "model-00034-of-00044.safetensors",
444
+ "model.layers.48.self_attn.q_norm.weight": "model-00033-of-00044.safetensors",
445
+ "model.layers.48.self_attn.q_proj.weight": "model-00034-of-00044.safetensors",
446
+ "model.layers.48.self_attn.v_proj.weight": "model-00034-of-00044.safetensors",
447
+ "model.layers.49.input_layernorm.weight": "model-00035-of-00044.safetensors",
448
+ "model.layers.49.mlp.down_proj.weight": "model-00035-of-00044.safetensors",
449
+ "model.layers.49.mlp.gate_proj.weight": "model-00034-of-00044.safetensors",
450
+ "model.layers.49.mlp.up_proj.weight": "model-00035-of-00044.safetensors",
451
+ "model.layers.49.self_attn.k_norm.weight": "model-00034-of-00044.safetensors",
452
+ "model.layers.49.self_attn.k_proj.weight": "model-00034-of-00044.safetensors",
453
+ "model.layers.49.self_attn.o_proj.weight": "model-00034-of-00044.safetensors",
454
+ "model.layers.49.self_attn.q_norm.weight": "model-00034-of-00044.safetensors",
455
+ "model.layers.49.self_attn.q_proj.weight": "model-00034-of-00044.safetensors",
456
+ "model.layers.49.self_attn.v_proj.weight": "model-00034-of-00044.safetensors",
457
+ "model.layers.5.input_layernorm.weight": "model-00005-of-00044.safetensors",
458
+ "model.layers.5.mlp.down_proj.weight": "model-00005-of-00044.safetensors",
459
+ "model.layers.5.mlp.gate_proj.weight": "model-00005-of-00044.safetensors",
460
+ "model.layers.5.mlp.up_proj.weight": "model-00005-of-00044.safetensors",
461
+ "model.layers.5.self_attn.k_norm.weight": "model-00005-of-00044.safetensors",
462
+ "model.layers.5.self_attn.k_proj.weight": "model-00005-of-00044.safetensors",
463
+ "model.layers.5.self_attn.o_proj.weight": "model-00005-of-00044.safetensors",
464
+ "model.layers.5.self_attn.q_norm.weight": "model-00005-of-00044.safetensors",
465
+ "model.layers.5.self_attn.q_proj.weight": "model-00005-of-00044.safetensors",
466
+ "model.layers.5.self_attn.v_proj.weight": "model-00005-of-00044.safetensors",
467
+ "model.layers.50.input_layernorm.weight": "model-00035-of-00044.safetensors",
468
+ "model.layers.50.mlp.down_proj.weight": "model-00035-of-00044.safetensors",
469
+ "model.layers.50.mlp.gate_proj.weight": "model-00035-of-00044.safetensors",
470
+ "model.layers.50.mlp.up_proj.weight": "model-00035-of-00044.safetensors",
471
+ "model.layers.50.self_attn.k_norm.weight": "model-00035-of-00044.safetensors",
472
+ "model.layers.50.self_attn.k_proj.weight": "model-00035-of-00044.safetensors",
473
+ "model.layers.50.self_attn.o_proj.weight": "model-00035-of-00044.safetensors",
474
+ "model.layers.50.self_attn.q_norm.weight": "model-00035-of-00044.safetensors",
475
+ "model.layers.50.self_attn.q_proj.weight": "model-00035-of-00044.safetensors",
476
+ "model.layers.50.self_attn.v_proj.weight": "model-00035-of-00044.safetensors",
477
+ "model.layers.51.input_layernorm.weight": "model-00036-of-00044.safetensors",
478
+ "model.layers.51.mlp.down_proj.weight": "model-00036-of-00044.safetensors",
479
+ "model.layers.51.mlp.gate_proj.weight": "model-00036-of-00044.safetensors",
480
+ "model.layers.51.mlp.up_proj.weight": "model-00036-of-00044.safetensors",
481
+ "model.layers.51.self_attn.k_norm.weight": "model-00035-of-00044.safetensors",
482
+ "model.layers.51.self_attn.k_proj.weight": "model-00036-of-00044.safetensors",
483
+ "model.layers.51.self_attn.o_proj.weight": "model-00036-of-00044.safetensors",
484
+ "model.layers.51.self_attn.q_norm.weight": "model-00035-of-00044.safetensors",
485
+ "model.layers.51.self_attn.q_proj.weight": "model-00036-of-00044.safetensors",
486
+ "model.layers.51.self_attn.v_proj.weight": "model-00036-of-00044.safetensors",
487
+ "model.layers.52.input_layernorm.weight": "model-00037-of-00044.safetensors",
488
+ "model.layers.52.mlp.down_proj.weight": "model-00037-of-00044.safetensors",
489
+ "model.layers.52.mlp.gate_proj.weight": "model-00036-of-00044.safetensors",
490
+ "model.layers.52.mlp.up_proj.weight": "model-00037-of-00044.safetensors",
491
+ "model.layers.52.self_attn.k_norm.weight": "model-00036-of-00044.safetensors",
492
+ "model.layers.52.self_attn.k_proj.weight": "model-00036-of-00044.safetensors",
493
+ "model.layers.52.self_attn.o_proj.weight": "model-00036-of-00044.safetensors",
494
+ "model.layers.52.self_attn.q_norm.weight": "model-00036-of-00044.safetensors",
495
+ "model.layers.52.self_attn.q_proj.weight": "model-00036-of-00044.safetensors",
496
+ "model.layers.52.self_attn.v_proj.weight": "model-00036-of-00044.safetensors",
497
+ "model.layers.53.input_layernorm.weight": "model-00037-of-00044.safetensors",
498
+ "model.layers.53.mlp.down_proj.weight": "model-00037-of-00044.safetensors",
499
+ "model.layers.53.mlp.gate_proj.weight": "model-00037-of-00044.safetensors",
500
+ "model.layers.53.mlp.up_proj.weight": "model-00037-of-00044.safetensors",
501
+ "model.layers.53.self_attn.k_norm.weight": "model-00037-of-00044.safetensors",
502
+ "model.layers.53.self_attn.k_proj.weight": "model-00037-of-00044.safetensors",
503
+ "model.layers.53.self_attn.o_proj.weight": "model-00037-of-00044.safetensors",
504
+ "model.layers.53.self_attn.q_norm.weight": "model-00037-of-00044.safetensors",
505
+ "model.layers.53.self_attn.q_proj.weight": "model-00037-of-00044.safetensors",
506
+ "model.layers.53.self_attn.v_proj.weight": "model-00037-of-00044.safetensors",
507
+ "model.layers.54.input_layernorm.weight": "model-00038-of-00044.safetensors",
508
+ "model.layers.54.mlp.down_proj.weight": "model-00038-of-00044.safetensors",
509
+ "model.layers.54.mlp.gate_proj.weight": "model-00038-of-00044.safetensors",
510
+ "model.layers.54.mlp.up_proj.weight": "model-00038-of-00044.safetensors",
511
+ "model.layers.54.self_attn.k_norm.weight": "model-00037-of-00044.safetensors",
512
+ "model.layers.54.self_attn.k_proj.weight": "model-00038-of-00044.safetensors",
513
+ "model.layers.54.self_attn.o_proj.weight": "model-00038-of-00044.safetensors",
514
+ "model.layers.54.self_attn.q_norm.weight": "model-00037-of-00044.safetensors",
515
+ "model.layers.54.self_attn.q_proj.weight": "model-00038-of-00044.safetensors",
516
+ "model.layers.54.self_attn.v_proj.weight": "model-00038-of-00044.safetensors",
517
+ "model.layers.55.input_layernorm.weight": "model-00039-of-00044.safetensors",
518
+ "model.layers.55.mlp.down_proj.weight": "model-00039-of-00044.safetensors",
519
+ "model.layers.55.mlp.gate_proj.weight": "model-00038-of-00044.safetensors",
520
+ "model.layers.55.mlp.up_proj.weight": "model-00039-of-00044.safetensors",
521
+ "model.layers.55.self_attn.k_norm.weight": "model-00038-of-00044.safetensors",
522
+ "model.layers.55.self_attn.k_proj.weight": "model-00038-of-00044.safetensors",
523
+ "model.layers.55.self_attn.o_proj.weight": "model-00038-of-00044.safetensors",
524
+ "model.layers.55.self_attn.q_norm.weight": "model-00038-of-00044.safetensors",
525
+ "model.layers.55.self_attn.q_proj.weight": "model-00038-of-00044.safetensors",
526
+ "model.layers.55.self_attn.v_proj.weight": "model-00038-of-00044.safetensors",
527
+ "model.layers.56.input_layernorm.weight": "model-00039-of-00044.safetensors",
528
+ "model.layers.56.mlp.down_proj.weight": "model-00039-of-00044.safetensors",
529
+ "model.layers.56.mlp.gate_proj.weight": "model-00039-of-00044.safetensors",
530
+ "model.layers.56.mlp.up_proj.weight": "model-00039-of-00044.safetensors",
531
+ "model.layers.56.self_attn.k_norm.weight": "model-00039-of-00044.safetensors",
532
+ "model.layers.56.self_attn.k_proj.weight": "model-00039-of-00044.safetensors",
533
+ "model.layers.56.self_attn.o_proj.weight": "model-00039-of-00044.safetensors",
534
+ "model.layers.56.self_attn.q_norm.weight": "model-00039-of-00044.safetensors",
535
+ "model.layers.56.self_attn.q_proj.weight": "model-00039-of-00044.safetensors",
536
+ "model.layers.56.self_attn.v_proj.weight": "model-00039-of-00044.safetensors",
537
+ "model.layers.57.input_layernorm.weight": "model-00040-of-00044.safetensors",
538
+ "model.layers.57.mlp.down_proj.weight": "model-00040-of-00044.safetensors",
539
+ "model.layers.57.mlp.gate_proj.weight": "model-00040-of-00044.safetensors",
540
+ "model.layers.57.mlp.up_proj.weight": "model-00040-of-00044.safetensors",
541
+ "model.layers.57.self_attn.k_norm.weight": "model-00039-of-00044.safetensors",
542
+ "model.layers.57.self_attn.k_proj.weight": "model-00040-of-00044.safetensors",
543
+ "model.layers.57.self_attn.o_proj.weight": "model-00040-of-00044.safetensors",
544
+ "model.layers.57.self_attn.q_norm.weight": "model-00039-of-00044.safetensors",
545
+ "model.layers.57.self_attn.q_proj.weight": "model-00040-of-00044.safetensors",
546
+ "model.layers.57.self_attn.v_proj.weight": "model-00040-of-00044.safetensors",
547
+ "model.layers.58.input_layernorm.weight": "model-00041-of-00044.safetensors",
548
+ "model.layers.58.mlp.down_proj.weight": "model-00041-of-00044.safetensors",
549
+ "model.layers.58.mlp.gate_proj.weight": "model-00040-of-00044.safetensors",
550
+ "model.layers.58.mlp.up_proj.weight": "model-00041-of-00044.safetensors",
551
+ "model.layers.58.self_attn.k_norm.weight": "model-00040-of-00044.safetensors",
552
+ "model.layers.58.self_attn.k_proj.weight": "model-00040-of-00044.safetensors",
553
+ "model.layers.58.self_attn.o_proj.weight": "model-00040-of-00044.safetensors",
554
+ "model.layers.58.self_attn.q_norm.weight": "model-00040-of-00044.safetensors",
555
+ "model.layers.58.self_attn.q_proj.weight": "model-00040-of-00044.safetensors",
556
+ "model.layers.58.self_attn.v_proj.weight": "model-00040-of-00044.safetensors",
557
+ "model.layers.59.input_layernorm.weight": "model-00041-of-00044.safetensors",
558
+ "model.layers.59.mlp.down_proj.weight": "model-00041-of-00044.safetensors",
559
+ "model.layers.59.mlp.gate_proj.weight": "model-00041-of-00044.safetensors",
560
+ "model.layers.59.mlp.up_proj.weight": "model-00041-of-00044.safetensors",
561
+ "model.layers.59.self_attn.k_norm.weight": "model-00041-of-00044.safetensors",
562
+ "model.layers.59.self_attn.k_proj.weight": "model-00041-of-00044.safetensors",
563
+ "model.layers.59.self_attn.o_proj.weight": "model-00041-of-00044.safetensors",
564
+ "model.layers.59.self_attn.q_norm.weight": "model-00041-of-00044.safetensors",
565
+ "model.layers.59.self_attn.q_proj.weight": "model-00041-of-00044.safetensors",
566
+ "model.layers.59.self_attn.v_proj.weight": "model-00041-of-00044.safetensors",
567
+ "model.layers.6.input_layernorm.weight": "model-00006-of-00044.safetensors",
568
+ "model.layers.6.mlp.down_proj.weight": "model-00006-of-00044.safetensors",
569
+ "model.layers.6.mlp.gate_proj.weight": "model-00006-of-00044.safetensors",
570
+ "model.layers.6.mlp.up_proj.weight": "model-00006-of-00044.safetensors",
571
+ "model.layers.6.self_attn.k_norm.weight": "model-00005-of-00044.safetensors",
572
+ "model.layers.6.self_attn.k_proj.weight": "model-00006-of-00044.safetensors",
573
+ "model.layers.6.self_attn.o_proj.weight": "model-00006-of-00044.safetensors",
574
+ "model.layers.6.self_attn.q_norm.weight": "model-00005-of-00044.safetensors",
575
+ "model.layers.6.self_attn.q_proj.weight": "model-00006-of-00044.safetensors",
576
+ "model.layers.6.self_attn.v_proj.weight": "model-00006-of-00044.safetensors",
577
+ "model.layers.60.input_layernorm.weight": "model-00042-of-00044.safetensors",
578
+ "model.layers.60.mlp.down_proj.weight": "model-00042-of-00044.safetensors",
579
+ "model.layers.60.mlp.gate_proj.weight": "model-00042-of-00044.safetensors",
580
+ "model.layers.60.mlp.up_proj.weight": "model-00042-of-00044.safetensors",
581
+ "model.layers.60.self_attn.k_norm.weight": "model-00041-of-00044.safetensors",
582
+ "model.layers.60.self_attn.k_proj.weight": "model-00042-of-00044.safetensors",
583
+ "model.layers.60.self_attn.o_proj.weight": "model-00042-of-00044.safetensors",
584
+ "model.layers.60.self_attn.q_norm.weight": "model-00041-of-00044.safetensors",
585
+ "model.layers.60.self_attn.q_proj.weight": "model-00042-of-00044.safetensors",
586
+ "model.layers.60.self_attn.v_proj.weight": "model-00042-of-00044.safetensors",
587
+ "model.layers.61.input_layernorm.weight": "model-00043-of-00044.safetensors",
588
+ "model.layers.61.mlp.down_proj.weight": "model-00043-of-00044.safetensors",
589
+ "model.layers.61.mlp.gate_proj.weight": "model-00042-of-00044.safetensors",
590
+ "model.layers.61.mlp.up_proj.weight": "model-00043-of-00044.safetensors",
591
+ "model.layers.61.self_attn.k_norm.weight": "model-00042-of-00044.safetensors",
592
+ "model.layers.61.self_attn.k_proj.weight": "model-00042-of-00044.safetensors",
593
+ "model.layers.61.self_attn.o_proj.weight": "model-00042-of-00044.safetensors",
594
+ "model.layers.61.self_attn.q_norm.weight": "model-00042-of-00044.safetensors",
595
+ "model.layers.61.self_attn.q_proj.weight": "model-00042-of-00044.safetensors",
596
+ "model.layers.61.self_attn.v_proj.weight": "model-00042-of-00044.safetensors",
597
+ "model.layers.62.input_layernorm.weight": "model-00043-of-00044.safetensors",
598
+ "model.layers.62.mlp.down_proj.weight": "model-00043-of-00044.safetensors",
599
+ "model.layers.62.mlp.gate_proj.weight": "model-00043-of-00044.safetensors",
600
+ "model.layers.62.mlp.up_proj.weight": "model-00043-of-00044.safetensors",
601
+ "model.layers.62.self_attn.k_norm.weight": "model-00043-of-00044.safetensors",
602
+ "model.layers.62.self_attn.k_proj.weight": "model-00043-of-00044.safetensors",
603
+ "model.layers.62.self_attn.o_proj.weight": "model-00043-of-00044.safetensors",
604
+ "model.layers.62.self_attn.q_norm.weight": "model-00043-of-00044.safetensors",
605
+ "model.layers.62.self_attn.q_proj.weight": "model-00043-of-00044.safetensors",
606
+ "model.layers.62.self_attn.v_proj.weight": "model-00043-of-00044.safetensors",
607
+ "model.layers.63.input_layernorm.weight": "model-00044-of-00044.safetensors",
608
+ "model.layers.63.mlp.down_proj.weight": "model-00044-of-00044.safetensors",
609
+ "model.layers.63.mlp.gate_proj.weight": "model-00044-of-00044.safetensors",
610
+ "model.layers.63.mlp.up_proj.weight": "model-00044-of-00044.safetensors",
611
+ "model.layers.63.self_attn.k_norm.weight": "model-00043-of-00044.safetensors",
612
+ "model.layers.63.self_attn.k_proj.weight": "model-00044-of-00044.safetensors",
613
+ "model.layers.63.self_attn.o_proj.weight": "model-00044-of-00044.safetensors",
614
+ "model.layers.63.self_attn.q_norm.weight": "model-00043-of-00044.safetensors",
615
+ "model.layers.63.self_attn.q_proj.weight": "model-00044-of-00044.safetensors",
616
+ "model.layers.63.self_attn.v_proj.weight": "model-00044-of-00044.safetensors",
617
+ "model.layers.7.input_layernorm.weight": "model-00007-of-00044.safetensors",
618
+ "model.layers.7.mlp.down_proj.weight": "model-00007-of-00044.safetensors",
619
+ "model.layers.7.mlp.gate_proj.weight": "model-00006-of-00044.safetensors",
620
+ "model.layers.7.mlp.up_proj.weight": "model-00007-of-00044.safetensors",
621
+ "model.layers.7.self_attn.k_norm.weight": "model-00006-of-00044.safetensors",
622
+ "model.layers.7.self_attn.k_proj.weight": "model-00006-of-00044.safetensors",
623
+ "model.layers.7.self_attn.o_proj.weight": "model-00006-of-00044.safetensors",
624
+ "model.layers.7.self_attn.q_norm.weight": "model-00006-of-00044.safetensors",
625
+ "model.layers.7.self_attn.q_proj.weight": "model-00006-of-00044.safetensors",
626
+ "model.layers.7.self_attn.v_proj.weight": "model-00006-of-00044.safetensors",
627
+ "model.layers.8.input_layernorm.weight": "model-00007-of-00044.safetensors",
628
+ "model.layers.8.mlp.down_proj.weight": "model-00007-of-00044.safetensors",
629
+ "model.layers.8.mlp.gate_proj.weight": "model-00007-of-00044.safetensors",
630
+ "model.layers.8.mlp.up_proj.weight": "model-00007-of-00044.safetensors",
631
+ "model.layers.8.self_attn.k_norm.weight": "model-00007-of-00044.safetensors",
632
+ "model.layers.8.self_attn.k_proj.weight": "model-00007-of-00044.safetensors",
633
+ "model.layers.8.self_attn.o_proj.weight": "model-00007-of-00044.safetensors",
634
+ "model.layers.8.self_attn.q_norm.weight": "model-00007-of-00044.safetensors",
635
+ "model.layers.8.self_attn.q_proj.weight": "model-00007-of-00044.safetensors",
636
+ "model.layers.8.self_attn.v_proj.weight": "model-00007-of-00044.safetensors",
637
+ "model.layers.9.input_layernorm.weight": "model-00008-of-00044.safetensors",
638
+ "model.layers.9.mlp.down_proj.weight": "model-00008-of-00044.safetensors",
639
+ "model.layers.9.mlp.gate_proj.weight": "model-00008-of-00044.safetensors",
640
+ "model.layers.9.mlp.up_proj.weight": "model-00008-of-00044.safetensors",
641
+ "model.layers.9.self_attn.k_norm.weight": "model-00007-of-00044.safetensors",
642
+ "model.layers.9.self_attn.k_proj.weight": "model-00008-of-00044.safetensors",
643
+ "model.layers.9.self_attn.o_proj.weight": "model-00008-of-00044.safetensors",
644
+ "model.layers.9.self_attn.q_norm.weight": "model-00007-of-00044.safetensors",
645
+ "model.layers.9.self_attn.q_proj.weight": "model-00008-of-00044.safetensors",
646
+ "model.layers.9.self_attn.v_proj.weight": "model-00008-of-00044.safetensors",
647
+ "model.norm.weight": "model-00044-of-00044.safetensors"
648
+ }
649
+ }
output-00001-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e954dbfc8edcb8a56567f6c64eeaec67baba52e1df2831c24a6c45d14c37a59e
3
+ size 8547247052
output-00002-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e4b4fa49ddefd83f8718aab3e3379bae590795fd08df659275ba9c1f7bf27da
3
+ size 8537034506
output-00003-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e24819d0c6c0eab3d116a89dfd883d43027d8d9a19caca31797f8efee068e7a
3
+ size 8507885178
output-00004-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee236ba437b6b987bb101d406c1b61c988e38132c5de61977a4725e6afac94a9
3
+ size 8551951536
output-00005-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2344e48e03a7624c451309ad88773906803f6d83f2ab39bd0f3bd9724e59fd1c
3
+ size 8577275360
output-00006-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2c2451afb63b2faca72f5a7fe4e82c985332684dfc96ee541057925f9ba7d329
3
+ size 8452157312
output-00007-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2fb823e7b92181d006175a5ed697b48f744982a4c598a97451eee64552f65302
3
+ size 8520099296
output-00008-of-00008.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b38e6a16bc37b7853b3910aade47b6bb291d8b7d7f72b821c2a7e4ad1d95626b
3
+ size 4237008706
special_tokens_map.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<BOS_TOKEN>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|END_OF_TURN_TOKEN|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<PAD>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ }
23
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9619890aebac311d644236f49462d7f8618ebef7c7020c52645ccb597434a3c9
3
+ size 16543645
tokenizer_config.json ADDED
@@ -0,0 +1,330 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": false,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<PAD>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<UNK>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "<CLS>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "3": {
31
+ "content": "<SEP>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "4": {
39
+ "content": "<MASK_TOKEN>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "5": {
47
+ "content": "<BOS_TOKEN>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "6": {
55
+ "content": "<EOS_TOKEN>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "7": {
63
+ "content": "<EOP_TOKEN>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "255000": {
71
+ "content": "<|START_OF_TURN_TOKEN|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": false
77
+ },
78
+ "255001": {
79
+ "content": "<|END_OF_TURN_TOKEN|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "255002": {
87
+ "content": "<|YES_TOKEN|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": false
93
+ },
94
+ "255003": {
95
+ "content": "<|NO_TOKEN|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": false
101
+ },
102
+ "255004": {
103
+ "content": "<|GOOD_TOKEN|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": false
109
+ },
110
+ "255005": {
111
+ "content": "<|BAD_TOKEN|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": false
117
+ },
118
+ "255006": {
119
+ "content": "<|USER_TOKEN|>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "255007": {
127
+ "content": "<|CHATBOT_TOKEN|>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": false
133
+ },
134
+ "255008": {
135
+ "content": "<|SYSTEM_TOKEN|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": false
141
+ },
142
+ "255009": {
143
+ "content": "<|USER_0_TOKEN|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": false
149
+ },
150
+ "255010": {
151
+ "content": "<|USER_1_TOKEN|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": false
157
+ },
158
+ "255011": {
159
+ "content": "<|USER_2_TOKEN|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": false
165
+ },
166
+ "255012": {
167
+ "content": "<|USER_3_TOKEN|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": false
173
+ },
174
+ "255013": {
175
+ "content": "<|USER_4_TOKEN|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": false
181
+ },
182
+ "255014": {
183
+ "content": "<|USER_5_TOKEN|>",
184
+ "lstrip": false,
185
+ "normalized": false,
186
+ "rstrip": false,
187
+ "single_word": false,
188
+ "special": false
189
+ },
190
+ "255015": {
191
+ "content": "<|USER_6_TOKEN|>",
192
+ "lstrip": false,
193
+ "normalized": false,
194
+ "rstrip": false,
195
+ "single_word": false,
196
+ "special": false
197
+ },
198
+ "255016": {
199
+ "content": "<|USER_7_TOKEN|>",
200
+ "lstrip": false,
201
+ "normalized": false,
202
+ "rstrip": false,
203
+ "single_word": false,
204
+ "special": false
205
+ },
206
+ "255017": {
207
+ "content": "<|USER_8_TOKEN|>",
208
+ "lstrip": false,
209
+ "normalized": false,
210
+ "rstrip": false,
211
+ "single_word": false,
212
+ "special": false
213
+ },
214
+ "255018": {
215
+ "content": "<|USER_9_TOKEN|>",
216
+ "lstrip": false,
217
+ "normalized": false,
218
+ "rstrip": false,
219
+ "single_word": false,
220
+ "special": false
221
+ },
222
+ "255019": {
223
+ "content": "<|EXTRA_0_TOKEN|>",
224
+ "lstrip": false,
225
+ "normalized": false,
226
+ "rstrip": false,
227
+ "single_word": false,
228
+ "special": false
229
+ },
230
+ "255020": {
231
+ "content": "<|EXTRA_1_TOKEN|>",
232
+ "lstrip": false,
233
+ "normalized": false,
234
+ "rstrip": false,
235
+ "single_word": false,
236
+ "special": false
237
+ },
238
+ "255021": {
239
+ "content": "<|EXTRA_2_TOKEN|>",
240
+ "lstrip": false,
241
+ "normalized": false,
242
+ "rstrip": false,
243
+ "single_word": false,
244
+ "special": false
245
+ },
246
+ "255022": {
247
+ "content": "<|EXTRA_3_TOKEN|>",
248
+ "lstrip": false,
249
+ "normalized": false,
250
+ "rstrip": false,
251
+ "single_word": false,
252
+ "special": false
253
+ },
254
+ "255023": {
255
+ "content": "<|EXTRA_4_TOKEN|>",
256
+ "lstrip": false,
257
+ "normalized": false,
258
+ "rstrip": false,
259
+ "single_word": false,
260
+ "special": false
261
+ },
262
+ "255024": {
263
+ "content": "<|EXTRA_5_TOKEN|>",
264
+ "lstrip": false,
265
+ "normalized": false,
266
+ "rstrip": false,
267
+ "single_word": false,
268
+ "special": false
269
+ },
270
+ "255025": {
271
+ "content": "<|EXTRA_6_TOKEN|>",
272
+ "lstrip": false,
273
+ "normalized": false,
274
+ "rstrip": false,
275
+ "single_word": false,
276
+ "special": false
277
+ },
278
+ "255026": {
279
+ "content": "<|EXTRA_7_TOKEN|>",
280
+ "lstrip": false,
281
+ "normalized": false,
282
+ "rstrip": false,
283
+ "single_word": false,
284
+ "special": false
285
+ },
286
+ "255027": {
287
+ "content": "<|EXTRA_8_TOKEN|>",
288
+ "lstrip": false,
289
+ "normalized": false,
290
+ "rstrip": false,
291
+ "single_word": false,
292
+ "special": false
293
+ },
294
+ "255028": {
295
+ "content": "<|EXTRA_9_TOKEN|>",
296
+ "lstrip": false,
297
+ "normalized": false,
298
+ "rstrip": false,
299
+ "single_word": false,
300
+ "special": false
301
+ }
302
+ },
303
+ "bos_token": "<BOS_TOKEN>",
304
+ "chat_template": [
305
+ {
306
+ "name": "default",
307
+ "template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% elif false == true %}{% set loop_messages = messages %}{% set system_message = 'You are Command-R, a brilliant, sophisticated, AI-assistant trained to assist human users by providing thorough responses. You are trained by Cohere.' %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% if system_message != false %}{{ '<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>' + system_message + '<|END_OF_TURN_TOKEN|>' }}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|START_OF_TURN_TOKEN|><|USER_TOKEN|>' + content.strip() + '<|END_OF_TURN_TOKEN|>' }}{% elif message['role'] == 'assistant' %}{{ '<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>' + content.strip() + '<|END_OF_TURN_TOKEN|>' }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>' }}{% endif %}"
308
+ },
309
+ {
310
+ "name": "tool_use",
311
+ "template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = '## Task and Context\\nYou help people answer their questions and other requests interactively. You will be asked a very wide array of requests on all kinds of topics. You will be equipped with a wide range of search engines or similar tools to help you, which you use to research your answer. You should focus on serving the user\\'s needs as best you can, which will be wide-ranging.\\n\\n## Style Guide\\nUnless the user asks for a different style of answer, you should answer in full sentences, using proper grammar and spelling.' %}{% endif %}{{ '<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>' }}{{ '# Safety Preamble' }}{{ '\nThe instructions in this section override those in the task description and style guide sections. Don\\'t answer questions that are harmful or immoral.' }}{{ '\n\n# System Preamble' }}{{ '\n## Basic Rules' }}{{ '\nYou are a powerful conversational AI trained by Cohere to help people. You are augmented by a number of tools, and your job is to use and consume the output of these tools to best help the user. You will see a conversation history between yourself and a user, ending with an utterance from the user. You will then see a specific instruction instructing you what kind of response to generate. When you answer the user\\'s requests, you cite your sources in your answers, according to those instructions.' }}{{ '\n\n# User Preamble' }}{{ '\n' + system_message }}{{'\n\n## Available Tools\nHere is a list of tools that you have available to you:\n\n'}}{% for tool in tools %}{% if loop.index0 != 0 %}{{ '\n\n'}}{% endif %}{{'```python\ndef ' + tool.name + '('}}{% for param_name, param_fields in tool.parameter_definitions.items() %}{% if loop.index0 != 0 %}{{ ', '}}{% endif %}{{param_name}}: {% if not param_fields.required %}{{'Optional[' + param_fields.type + '] = None'}}{% else %}{{ param_fields.type }}{% endif %}{% endfor %}{{ ') -> List[Dict]:\n \"\"\"'}}{{ tool.description }}{% if tool.parameter_definitions|length != 0 %}{{ '\n\n Args:\n '}}{% for param_name, param_fields in tool.parameter_definitions.items() %}{% if loop.index0 != 0 %}{{ '\n ' }}{% endif %}{{ param_name + ' ('}}{% if not param_fields.required %}{{'Optional[' + param_fields.type + ']'}}{% else %}{{ param_fields.type }}{% endif %}{{ '): ' + param_fields.description }}{% endfor %}{% endif %}{{ '\n \"\"\"\n pass\n```' }}{% endfor %}{{ '<|END_OF_TURN_TOKEN|>'}}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|START_OF_TURN_TOKEN|><|USER_TOKEN|>' + content.strip() + '<|END_OF_TURN_TOKEN|>' }}{% elif message['role'] == 'system' %}{{ '<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>' + content.strip() + '<|END_OF_TURN_TOKEN|>' }}{% elif message['role'] == 'assistant' %}{{ '<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>' + content.strip() + '<|END_OF_TURN_TOKEN|>' }}{% endif %}{% endfor %}{{'<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>Write \\'Action:\\' followed by a json-formatted list of actions that you want to perform in order to produce a good response to the user\\'s last input. You can use any of the supplied tools any number of times, but you should aim to execute the minimum number of necessary actions for the input. You should use the `directly-answer` tool if calling the other tools is unnecessary. The list of actions you want to call should be formatted as a list of json objects, for example:\n```json\n[\n {\n \"tool_name\": title of the tool in the specification,\n \"parameters\": a dict of parameters to input into the tool as they are defined in the specs, or {} if it takes no parameters\n }\n]```<|END_OF_TURN_TOKEN|>'}}{% if add_generation_prompt %}{{ '<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>' }}{% endif %}"
312
+ },
313
+ {
314
+ "name": "rag",
315
+ "template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = '## Task and Context\\nYou help people answer their questions and other requests interactively. You will be asked a very wide array of requests on all kinds of topics. You will be equipped with a wide range of search engines or similar tools to help you, which you use to research your answer. You should focus on serving the user\\'s needs as best you can, which will be wide-ranging.\\n\\n## Style Guide\\nUnless the user asks for a different style of answer, you should answer in full sentences, using proper grammar and spelling.' %}{% endif %}{{ '<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>' }}{{ '# Safety Preamble' }}{{ '\nThe instructions in this section override those in the task description and style guide sections. Don\\'t answer questions that are harmful or immoral.' }}{{ '\n\n# System Preamble' }}{{ '\n## Basic Rules' }}{{ '\nYou are a powerful conversational AI trained by Cohere to help people. You are augmented by a number of tools, and your job is to use and consume the output of these tools to best help the user. You will see a conversation history between yourself and a user, ending with an utterance from the user. You will then see a specific instruction instructing you what kind of response to generate. When you answer the user\\'s requests, you cite your sources in your answers, according to those instructions.' }}{{ '\n\n# User Preamble' }}{{ '\n' + system_message }}{{ '<|END_OF_TURN_TOKEN|>'}}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|START_OF_TURN_TOKEN|><|USER_TOKEN|>' + content.strip() + '<|END_OF_TURN_TOKEN|>' }}{% elif message['role'] == 'system' %}{{ '<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>' + content.strip() + '<|END_OF_TURN_TOKEN|>' }}{% elif message['role'] == 'assistant' %}{{ '<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>' + content.strip() + '<|END_OF_TURN_TOKEN|>' }}{% endif %}{% endfor %}{{ '<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>'}}{{ '<results>' }}{% for document in documents %}{{ '\nDocument: ' }}{{ loop.index0 }}\n{% for key, value in document.items() %}{{ key }}: {{value}}\n{% endfor %}{% endfor %}{{ '</results>'}}{{ '<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>' }}{{ 'Carefully perform the following instructions, in order, starting each with a new line.\n' }}{{ 'Firstly, Decide which of the retrieved documents are relevant to the user\\'s last input by writing \\'Relevant Documents:\\' followed by comma-separated list of document numbers. If none are relevant, you should instead write \\'None\\'.\n' }}{{ 'Secondly, Decide which of the retrieved documents contain facts that should be cited in a good answer to the user\\'s last input by writing \\'Cited Documents:\\' followed a comma-separated list of document numbers. If you dont want to cite any of them, you should instead write \\'None\\'.\n' }}{% if citation_mode=='accurate' %}{{ 'Thirdly, Write \\'Answer:\\' followed by a response to the user\\'s last input in high quality natural english. Use the retrieved documents to help you. Do not insert any citations or grounding markup.\n' }}{% endif %}{{ 'Finally, Write \\'Grounded answer:\\' followed by a response to the user\\'s last input in high quality natural english. Use the symbols <co: doc> and </co: doc> to indicate when a fact comes from a document in the search result, e.g <co: 0>my fact</co: 0> for a fact from document 0.' }}{{ '<|END_OF_TURN_TOKEN|>' }}{% if add_generation_prompt %}{{ '<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>' }}{% endif %}"
316
+ }
317
+ ],
318
+ "clean_up_tokenization_spaces": false,
319
+ "eos_token": "<|END_OF_TURN_TOKEN|>",
320
+ "legacy": true,
321
+ "merges_file": null,
322
+ "model_max_length": 1000000000000000019884624838656,
323
+ "pad_token": "<PAD>",
324
+ "sp_model_kwargs": {},
325
+ "spaces_between_special_tokens": false,
326
+ "tokenizer_class": "CohereTokenizer",
327
+ "unk_token": null,
328
+ "use_default_system_prompt": false,
329
+ "vocab_file": null
330
+ }