jartine commited on
Commit
80cb95a
1 Parent(s): 4e99edc

Add README.md to repo

Browse files
Files changed (1) hide show
  1. README.md +404 -0
README.md ADDED
@@ -0,0 +1,404 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: mistralai/Mistral-7B-Instruct-v0.2
3
+ inference: false
4
+ license: apache-2.0
5
+ model_creator: Mistral AI_
6
+ model_name: Mistral 7B Instruct v0.2
7
+ model_type: mistral
8
+ pipeline_tag: text-generation
9
+ prompt_template: '<s>[INST] {prompt} [/INST]
10
+
11
+ '
12
+ quantized_by: TheBloke
13
+ tags:
14
+ - finetuned
15
+ ---
16
+ <!-- markdownlint-disable MD041 -->
17
+
18
+ <!-- header start -->
19
+ <!-- 200823 -->
20
+ <div style="width: auto; margin-left: auto; margin-right: auto">
21
+ <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
22
+ </div>
23
+ <div style="display: flex; justify-content: space-between; width: 100%;">
24
+ <div style="display: flex; flex-direction: column; align-items: flex-start;">
25
+ <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
26
+ </div>
27
+ <div style="display: flex; flex-direction: column; align-items: flex-end;">
28
+ <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
29
+ </div>
30
+ </div>
31
+ <div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
32
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
33
+ <!-- header end -->
34
+
35
+ # Mistral 7B Instruct v0.2 - GGUF
36
+ - Model creator: [Mistral AI_](https://huggingface.co/mistralai)
37
+ - Original model: [Mistral 7B Instruct v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2)
38
+
39
+ <!-- description start -->
40
+ ## Description
41
+
42
+ This repo contains GGUF format model files for [Mistral AI_'s Mistral 7B Instruct v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2).
43
+
44
+ These files were quantised using hardware kindly provided by [Massed Compute](https://massedcompute.com/).
45
+
46
+ <!-- description end -->
47
+ <!-- README_GGUF.md-about-gguf start -->
48
+ ### About GGUF
49
+
50
+ GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp.
51
+
52
+ Here is an incomplete list of clients and libraries that are known to support GGUF:
53
+
54
+ * [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
55
+ * [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
56
+ * [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
57
+ * [GPT4All](https://gpt4all.io/index.html), a free and open source local running GUI, supporting Windows, Linux and macOS with full GPU accel.
58
+ * [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration. Linux available, in beta as of 27/11/2023.
59
+ * [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
60
+ * [Faraday.dev](https://faraday.dev/), an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
61
+ * [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
62
+ * [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
63
+ * [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server. Note, as of time of writing (November 27th 2023), ctransformers has not been updated in a long time and does not support many recent models.
64
+
65
+ <!-- README_GGUF.md-about-gguf end -->
66
+ <!-- repositories-available start -->
67
+ ## Repositories available
68
+
69
+ * [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-AWQ)
70
+ * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GPTQ)
71
+ * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF)
72
+ * [Mistral AI_'s original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2)
73
+ <!-- repositories-available end -->
74
+
75
+ <!-- prompt-template start -->
76
+ ## Prompt template: Mistral
77
+
78
+ ```
79
+ <s>[INST] {prompt} [/INST]
80
+
81
+ ```
82
+
83
+ <!-- prompt-template end -->
84
+
85
+
86
+ <!-- compatibility_gguf start -->
87
+ ## Compatibility
88
+
89
+ These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
90
+
91
+ They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
92
+
93
+ ## Explanation of quantisation methods
94
+
95
+ <details>
96
+ <summary>Click to see details</summary>
97
+
98
+ The new methods available are:
99
+
100
+ * GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
101
+ * GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
102
+ * GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
103
+ * GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
104
+ * GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw
105
+
106
+ Refer to the Provided Files table below to see what files use which methods, and how.
107
+ </details>
108
+ <!-- compatibility_gguf end -->
109
+
110
+ <!-- README_GGUF.md-provided-files start -->
111
+ ## Provided files
112
+
113
+ | Name | Quant method | Bits | Size | Max RAM required | Use case |
114
+ | ---- | ---- | ---- | ---- | ---- | ----- |
115
+ | [mistral-7b-instruct-v0.2.Q2_K.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q2_K.gguf) | Q2_K | 2 | 3.08 GB| 5.58 GB | smallest, significant quality loss - not recommended for most purposes |
116
+ | [mistral-7b-instruct-v0.2.Q3_K_S.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q3_K_S.gguf) | Q3_K_S | 3 | 3.16 GB| 5.66 GB | very small, high quality loss |
117
+ | [mistral-7b-instruct-v0.2.Q3_K_M.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q3_K_M.gguf) | Q3_K_M | 3 | 3.52 GB| 6.02 GB | very small, high quality loss |
118
+ | [mistral-7b-instruct-v0.2.Q3_K_L.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q3_K_L.gguf) | Q3_K_L | 3 | 3.82 GB| 6.32 GB | small, substantial quality loss |
119
+ | [mistral-7b-instruct-v0.2.Q4_0.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q4_0.gguf) | Q4_0 | 4 | 4.11 GB| 6.61 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
120
+ | [mistral-7b-instruct-v0.2.Q4_K_S.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q4_K_S.gguf) | Q4_K_S | 4 | 4.14 GB| 6.64 GB | small, greater quality loss |
121
+ | [mistral-7b-instruct-v0.2.Q4_K_M.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q4_K_M.gguf) | Q4_K_M | 4 | 4.37 GB| 6.87 GB | medium, balanced quality - recommended |
122
+ | [mistral-7b-instruct-v0.2.Q5_0.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q5_0.gguf) | Q5_0 | 5 | 5.00 GB| 7.50 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
123
+ | [mistral-7b-instruct-v0.2.Q5_K_S.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q5_K_S.gguf) | Q5_K_S | 5 | 5.00 GB| 7.50 GB | large, low quality loss - recommended |
124
+ | [mistral-7b-instruct-v0.2.Q5_K_M.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q5_K_M.gguf) | Q5_K_M | 5 | 5.13 GB| 7.63 GB | large, very low quality loss - recommended |
125
+ | [mistral-7b-instruct-v0.2.Q6_K.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q6_K.gguf) | Q6_K | 6 | 5.94 GB| 8.44 GB | very large, extremely low quality loss |
126
+ | [mistral-7b-instruct-v0.2.Q8_0.gguf](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/blob/main/mistral-7b-instruct-v0.2.Q8_0.gguf) | Q8_0 | 8 | 7.70 GB| 10.20 GB | very large, extremely low quality loss - not recommended |
127
+
128
+ **Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
129
+
130
+
131
+
132
+ <!-- README_GGUF.md-provided-files end -->
133
+
134
+ <!-- README_GGUF.md-how-to-download start -->
135
+ ## How to download GGUF files
136
+
137
+ **Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
138
+
139
+ The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
140
+
141
+ * LM Studio
142
+ * LoLLMS Web UI
143
+ * Faraday.dev
144
+
145
+ ### In `text-generation-webui`
146
+
147
+ Under Download Model, you can enter the model repo: TheBloke/Mistral-7B-Instruct-v0.2-GGUF and below it, a specific filename to download, such as: mistral-7b-instruct-v0.2.Q4_K_M.gguf.
148
+
149
+ Then click Download.
150
+
151
+ ### On the command line, including multiple files at once
152
+
153
+ I recommend using the `huggingface-hub` Python library:
154
+
155
+ ```shell
156
+ pip3 install huggingface-hub
157
+ ```
158
+
159
+ Then you can download any individual model file to the current directory, at high speed, with a command like this:
160
+
161
+ ```shell
162
+ huggingface-cli download TheBloke/Mistral-7B-Instruct-v0.2-GGUF mistral-7b-instruct-v0.2.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
163
+ ```
164
+
165
+ <details>
166
+ <summary>More advanced huggingface-cli download usage (click to read)</summary>
167
+
168
+ You can also download multiple files at once with a pattern:
169
+
170
+ ```shell
171
+ huggingface-cli download TheBloke/Mistral-7B-Instruct-v0.2-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
172
+ ```
173
+
174
+ For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
175
+
176
+ To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
177
+
178
+ ```shell
179
+ pip3 install hf_transfer
180
+ ```
181
+
182
+ And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
183
+
184
+ ```shell
185
+ HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Mistral-7B-Instruct-v0.2-GGUF mistral-7b-instruct-v0.2.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
186
+ ```
187
+
188
+ Windows Command Line users: You can set the environment variable by running `set HF_HUB_ENABLE_HF_TRANSFER=1` before the download command.
189
+ </details>
190
+ <!-- README_GGUF.md-how-to-download end -->
191
+
192
+ <!-- README_GGUF.md-how-to-run start -->
193
+ ## Example `llama.cpp` command
194
+
195
+ Make sure you are using `llama.cpp` from commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
196
+
197
+ ```shell
198
+ ./main -ngl 35 -m mistral-7b-instruct-v0.2.Q4_K_M.gguf --color -c 32768 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "<s>[INST] {prompt} [/INST]"
199
+ ```
200
+
201
+ Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
202
+
203
+ Change `-c 32768` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically. Note that longer sequence lengths require much more resources, so you may need to reduce this value.
204
+
205
+ If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
206
+
207
+ For other parameters and how to use them, please refer to [the llama.cpp documentation](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)
208
+
209
+ ## How to run in `text-generation-webui`
210
+
211
+ Further instructions can be found in the text-generation-webui documentation, here: [text-generation-webui/docs/04 ‐ Model Tab.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/04%20%E2%80%90%20Model%20Tab.md#llamacpp).
212
+
213
+ ## How to run from Python code
214
+
215
+ You can use GGUF models from Python using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) or [ctransformers](https://github.com/marella/ctransformers) libraries. Note that at the time of writing (Nov 27th 2023), ctransformers has not been updated for some time and is not compatible with some recent models. Therefore I recommend you use llama-cpp-python.
216
+
217
+ ### How to load this model in Python code, using llama-cpp-python
218
+
219
+ For full documentation, please see: [llama-cpp-python docs](https://abetlen.github.io/llama-cpp-python/).
220
+
221
+ #### First install the package
222
+
223
+ Run one of the following commands, according to your system:
224
+
225
+ ```shell
226
+ # Base ctransformers with no GPU acceleration
227
+ pip install llama-cpp-python
228
+ # With NVidia CUDA acceleration
229
+ CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
230
+ # Or with OpenBLAS acceleration
231
+ CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
232
+ # Or with CLBLast acceleration
233
+ CMAKE_ARGS="-DLLAMA_CLBLAST=on" pip install llama-cpp-python
234
+ # Or with AMD ROCm GPU acceleration (Linux only)
235
+ CMAKE_ARGS="-DLLAMA_HIPBLAS=on" pip install llama-cpp-python
236
+ # Or with Metal GPU acceleration for macOS systems only
237
+ CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
238
+
239
+ # In windows, to set the variables CMAKE_ARGS in PowerShell, follow this format; eg for NVidia CUDA:
240
+ $env:CMAKE_ARGS = "-DLLAMA_OPENBLAS=on"
241
+ pip install llama-cpp-python
242
+ ```
243
+
244
+ #### Simple llama-cpp-python example code
245
+
246
+ ```python
247
+ from llama_cpp import Llama
248
+
249
+ # Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
250
+ llm = Llama(
251
+ model_path="./mistral-7b-instruct-v0.2.Q4_K_M.gguf", # Download the model file first
252
+ n_ctx=32768, # The max sequence length to use - note that longer sequence lengths require much more resources
253
+ n_threads=8, # The number of CPU threads to use, tailor to your system and the resulting performance
254
+ n_gpu_layers=35 # The number of layers to offload to GPU, if you have GPU acceleration available
255
+ )
256
+
257
+ # Simple inference example
258
+ output = llm(
259
+ "<s>[INST] {prompt} [/INST]", # Prompt
260
+ max_tokens=512, # Generate up to 512 tokens
261
+ stop=["</s>"], # Example stop token - not necessarily correct for this specific model! Please check before using.
262
+ echo=True # Whether to echo the prompt
263
+ )
264
+
265
+ # Chat Completion API
266
+
267
+ llm = Llama(model_path="./mistral-7b-instruct-v0.2.Q4_K_M.gguf", chat_format="llama-2") # Set chat_format according to the model you are using
268
+ llm.create_chat_completion(
269
+ messages = [
270
+ {"role": "system", "content": "You are a story writing assistant."},
271
+ {
272
+ "role": "user",
273
+ "content": "Write a story about llamas."
274
+ }
275
+ ]
276
+ )
277
+ ```
278
+
279
+ ## How to use with LangChain
280
+
281
+ Here are guides on using llama-cpp-python and ctransformers with LangChain:
282
+
283
+ * [LangChain + llama-cpp-python](https://python.langchain.com/docs/integrations/llms/llamacpp)
284
+ * [LangChain + ctransformers](https://python.langchain.com/docs/integrations/providers/ctransformers)
285
+
286
+ <!-- README_GGUF.md-how-to-run end -->
287
+
288
+ <!-- footer start -->
289
+ <!-- 200823 -->
290
+ ## Discord
291
+
292
+ For further support, and discussions on these models and AI in general, join us at:
293
+
294
+ [TheBloke AI's Discord server](https://discord.gg/theblokeai)
295
+
296
+ ## Thanks, and how to contribute
297
+
298
+ Thanks to the [chirper.ai](https://chirper.ai) team!
299
+
300
+ Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
301
+
302
+ I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
303
+
304
+ If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
305
+
306
+ Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
307
+
308
+ * Patreon: https://patreon.com/TheBlokeAI
309
+ * Ko-Fi: https://ko-fi.com/TheBlokeAI
310
+
311
+ **Special thanks to**: Aemon Algiz.
312
+
313
+ **Patreon special mentions**: Michael Levine, 阿明, Trailburnt, Nikolai Manek, John Detwiler, Randy H, Will Dee, Sebastain Graf, NimbleBox.ai, Eugene Pentland, Emad Mostaque, Ai Maven, Jim Angel, Jeff Scroggin, Michael Davis, Manuel Alberto Morcote, Stephen Murray, Robert, Justin Joy, Luke @flexchar, Brandon Frisco, Elijah Stavena, S_X, Dan Guido, Undi ., Komninos Chatzipapas, Shadi, theTransient, Lone Striker, Raven Klaugh, jjj, Cap'n Zoog, Michel-Marie MAUDET (LINAGORA), Matthew Berman, David, Fen Risland, Omer Bin Jawed, Luke Pendergrass, Kalila, OG, Erik Bjäreholt, Rooh Singh, Joseph William Delisle, Dan Lewis, TL, John Villwock, AzureBlack, Brad, Pedro Madruga, Caitlyn Gatomon, K, jinyuan sun, Mano Prime, Alex, Jeffrey Morgan, Alicia Loh, Illia Dulskyi, Chadd, transmissions 11, fincy, Rainer Wilmers, ReadyPlayerEmma, knownsqashed, Mandus, biorpg, Deo Leter, Brandon Phillips, SuperWojo, Sean Connelly, Iucharbius, Jack West, Harry Royden McLaughlin, Nicholas, terasurfer, Vitor Caleffi, Duane Dunston, Johann-Peter Hartmann, David Ziegler, Olakabola, Ken Nordquist, Trenton Dambrowitz, Tom X Nguyen, Vadim, Ajan Kanaga, Leonard Tan, Clay Pascal, Alexandros Triantafyllidis, JM33133, Xule, vamX, ya boyyy, subjectnull, Talal Aujan, Alps Aficionado, wassieverse, Ari Malik, James Bentley, Woland, Spencer Kim, Michael Dempsey, Fred von Graf, Elle, zynix, William Richards, Stanislav Ovsiannikov, Edmond Seymore, Jonathan Leane, Martin Kemka, usrbinkat, Enrico Ros
314
+
315
+
316
+ Thank you to all my generous patrons and donaters!
317
+
318
+ And thank you again to a16z for their generous grant.
319
+
320
+ <!-- footer end -->
321
+
322
+ <!-- original-model-card start -->
323
+ # Original model card: Mistral AI_'s Mistral 7B Instruct v0.2
324
+
325
+
326
+ # Model Card for Mistral-7B-Instruct-v0.2
327
+
328
+ The Mistral-7B-Instruct-v0.2 Large Language Model (LLM) is an improved instruct fine-tuned version of [Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1).
329
+
330
+ For full details of this model please read our [paper](https://arxiv.org/abs/2310.06825) and [release blog post](https://mistral.ai/news/la-plateforme/).
331
+
332
+ ## Instruction format
333
+
334
+ In order to leverage instruction fine-tuning, your prompt should be surrounded by `[INST]` and `[/INST]` tokens. The very first instruction should begin with a begin of sentence id. The next instructions should not. The assistant generation will be ended by the end-of-sentence token id.
335
+
336
+ E.g.
337
+ ```
338
+ text = "<s>[INST] What is your favourite condiment? [/INST]"
339
+ "Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!</s> "
340
+ "[INST] Do you have mayonnaise recipes? [/INST]"
341
+ ```
342
+
343
+ This format is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating) via the `apply_chat_template()` method:
344
+
345
+ ```python
346
+ from transformers import AutoModelForCausalLM, AutoTokenizer
347
+
348
+ device = "cuda" # the device to load the model onto
349
+
350
+ model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.1")
351
+ tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.1")
352
+
353
+ messages = [
354
+ {"role": "user", "content": "What is your favourite condiment?"},
355
+ {"role": "assistant", "content": "Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!"},
356
+ {"role": "user", "content": "Do you have mayonnaise recipes?"}
357
+ ]
358
+
359
+ encodeds = tokenizer.apply_chat_template(messages, return_tensors="pt")
360
+
361
+ model_inputs = encodeds.to(device)
362
+ model.to(device)
363
+
364
+ generated_ids = model.generate(model_inputs, max_new_tokens=1000, do_sample=True)
365
+ decoded = tokenizer.batch_decode(generated_ids)
366
+ print(decoded[0])
367
+ ```
368
+
369
+ ## Model Architecture
370
+ This instruction model is based on Mistral-7B-v0.1, a transformer model with the following architecture choices:
371
+ - Grouped-Query Attention
372
+ - Sliding-Window Attention
373
+ - Byte-fallback BPE tokenizer
374
+
375
+ ## Troubleshooting
376
+ - If you see the following error:
377
+ ```
378
+ Traceback (most recent call last):
379
+ File "", line 1, in
380
+ File "/transformers/models/auto/auto_factory.py", line 482, in from_pretrained
381
+ config, kwargs = AutoConfig.from_pretrained(
382
+ File "/transformers/models/auto/configuration_auto.py", line 1022, in from_pretrained
383
+ config_class = CONFIG_MAPPING[config_dict["model_type"]]
384
+ File "/transformers/models/auto/configuration_auto.py", line 723, in getitem
385
+ raise KeyError(key)
386
+ KeyError: 'mistral'
387
+ ```
388
+
389
+ Installing transformers from source should solve the issue
390
+ pip install git+https://github.com/huggingface/transformers
391
+
392
+ This should not be required after transformers-v4.33.4.
393
+
394
+ ## Limitations
395
+
396
+ The Mistral 7B Instruct model is a quick demonstration that the base model can be easily fine-tuned to achieve compelling performance.
397
+ It does not have any moderation mechanisms. We're looking forward to engaging with the community on ways to
398
+ make the model finely respect guardrails, allowing for deployment in environments requiring moderated outputs.
399
+
400
+ ## The Mistral AI Team
401
+
402
+ Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Blanche Savary, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Emma Bou Hanna, Florian Bressand, Gianna Lengyel, Guillaume Bour, Guillaume Lample, Lélio Renard Lavaud, Louis Ternon, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Théophile Gervet, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.
403
+
404
+ <!-- original-model-card end -->