TheBloke commited on
Commit
2d7f929
1 Parent(s): 776ed81

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +394 -0
README.md ADDED
@@ -0,0 +1,394 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Doctor-Shotgun/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES
3
+ inference: false
4
+ model_creator: Doctor Shotgun
5
+ model_name: Mixtral 8X7B Instruct V0.1 LimaRP ZLoss DARE TIES
6
+ model_type: mixtral
7
+ prompt_template: '{prompt}
8
+
9
+ '
10
+ quantized_by: TheBloke
11
+ tags:
12
+ - mergekit
13
+ - merge
14
+ ---
15
+ <!-- markdownlint-disable MD041 -->
16
+
17
+ <!-- header start -->
18
+ <!-- 200823 -->
19
+ <div style="width: auto; margin-left: auto; margin-right: auto">
20
+ <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
21
+ </div>
22
+ <div style="display: flex; justify-content: space-between; width: 100%;">
23
+ <div style="display: flex; flex-direction: column; align-items: flex-start;">
24
+ <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
25
+ </div>
26
+ <div style="display: flex; flex-direction: column; align-items: flex-end;">
27
+ <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>
28
+ </div>
29
+ </div>
30
+ <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>
31
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
32
+ <!-- header end -->
33
+
34
+ # Mixtral 8X7B Instruct V0.1 LimaRP ZLoss DARE TIES - AWQ
35
+ - Model creator: [Doctor Shotgun](https://huggingface.co/Doctor-Shotgun)
36
+ - Original model: [Mixtral 8X7B Instruct V0.1 LimaRP ZLoss DARE TIES](https://huggingface.co/Doctor-Shotgun/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES)
37
+
38
+ <!-- description start -->
39
+ ## Description
40
+
41
+ This repo contains AWQ model files for [Doctor Shotgun's Mixtral 8X7B Instruct V0.1 LimaRP ZLoss DARE TIES](https://huggingface.co/Doctor-Shotgun/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES).
42
+
43
+ These files were quantised using hardware kindly provided by [Massed Compute](https://massedcompute.com/).
44
+
45
+
46
+ **MIXTRAL AWQ**
47
+
48
+ This is a Mixtral AWQ model.
49
+
50
+ For AutoAWQ inference, please install AutoAWQ 0.1.8 or later.
51
+
52
+ Support via Transformers is also available, but currently requires installing Transformers from Github: `pip3 install git+https://github.com/huggingface/transformers.git`
53
+
54
+ vLLM: version 0.2.6 is confirmed to support Mixtral AWQs.
55
+
56
+ TGI: I tested version 1.3.3 and it loaded the model fine, but I was not able to get any output back. Further testing/debug is required. (Let me know if you get it working!)
57
+
58
+ ### About AWQ
59
+
60
+ AWQ is an efficient, accurate and blazing-fast low-bit weight quantization method, currently supporting 4-bit quantization. Compared to GPTQ, it offers faster Transformers-based inference with equivalent or better quality compared to the most commonly used GPTQ settings.
61
+
62
+ AWQ models are currently supported on Linux and Windows, with NVidia GPUs only. macOS users: please use GGUF models instead.
63
+
64
+ AWQ models are supported by (note that not all of these may support Mixtral models yet - see above):
65
+
66
+ - [Text Generation Webui](https://github.com/oobabooga/text-generation-webui) - using Loader: AutoAWQ
67
+ - [vLLM](https://github.com/vllm-project/vllm) - version 0.2.2 or later for support for all model types.
68
+ - [Hugging Face Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference)
69
+ - [Transformers](https://huggingface.co/docs/transformers) version 4.35.0 and later, from any code or client that supports Transformers
70
+ - [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) - for use from Python code
71
+
72
+ <!-- description end -->
73
+ <!-- repositories-available start -->
74
+ ## Repositories available
75
+
76
+ * [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES-AWQ)
77
+ * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES-GPTQ)
78
+ * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES-GGUF)
79
+ * [Doctor Shotgun's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/Doctor-Shotgun/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES)
80
+ <!-- repositories-available end -->
81
+
82
+ <!-- prompt-template start -->
83
+ ## Prompt template: Unknown
84
+
85
+ ```
86
+ {prompt}
87
+
88
+ ```
89
+
90
+ <!-- prompt-template end -->
91
+
92
+
93
+ <!-- README_AWQ.md-provided-files start -->
94
+ ## Provided files, and AWQ parameters
95
+
96
+ I currently release 128g GEMM models only. The addition of group_size 32 models, and GEMV kernel models, is being actively considered.
97
+
98
+ Models are released as sharded safetensors files.
99
+
100
+ | Branch | Bits | GS | AWQ Dataset | Seq Len | Size |
101
+ | ------ | ---- | -- | ----------- | ------- | ---- |
102
+ | [main](https://huggingface.co/TheBloke/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES-AWQ/tree/main) | 4 | 128 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 8192 | 24.65 GB
103
+
104
+ <!-- README_AWQ.md-provided-files end -->
105
+
106
+ <!-- README_AWQ.md-text-generation-webui start -->
107
+ ## How to easily download and use this model in [text-generation-webui](https://github.com/oobabooga/text-generation-webui)
108
+
109
+ Please make sure you're using the latest version of [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
110
+
111
+ It is strongly recommended to use the text-generation-webui one-click-installers unless you're sure you know how to make a manual install.
112
+
113
+ 1. Click the **Model tab**.
114
+ 2. Under **Download custom model or LoRA**, enter `TheBloke/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES-AWQ`.
115
+ 3. Click **Download**.
116
+ 4. The model will start downloading. Once it's finished it will say "Done".
117
+ 5. In the top left, click the refresh icon next to **Model**.
118
+ 6. In the **Model** dropdown, choose the model you just downloaded: `Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES-AWQ`
119
+ 7. Select **Loader: AutoAWQ**.
120
+ 8. Click Load, and the model will load and is now ready for use.
121
+ 9. If you want any custom settings, set them and then click **Save settings for this model** followed by **Reload the Model** in the top right.
122
+ 10. Once you're ready, click the **Text Generation** tab and enter a prompt to get started!
123
+ <!-- README_AWQ.md-text-generation-webui end -->
124
+
125
+ <!-- README_AWQ.md-use-from-vllm start -->
126
+ ## Multi-user inference server: vLLM
127
+
128
+ Documentation on installing and using vLLM [can be found here](https://vllm.readthedocs.io/en/latest/).
129
+
130
+ - Please ensure you are using vLLM version 0.2 or later.
131
+ - When using vLLM as a server, pass the `--quantization awq` parameter.
132
+
133
+ For example:
134
+
135
+ ```shell
136
+ python3 -m vllm.entrypoints.api_server --model TheBloke/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES-AWQ --quantization awq --dtype auto
137
+ ```
138
+
139
+ - When using vLLM from Python code, again set `quantization=awq`.
140
+
141
+ For example:
142
+
143
+ ```python
144
+ from vllm import LLM, SamplingParams
145
+
146
+ prompts = [
147
+ "Tell me about AI",
148
+ "Write a story about llamas",
149
+ "What is 291 - 150?",
150
+ "How much wood would a woodchuck chuck if a woodchuck could chuck wood?",
151
+ ]
152
+ prompt_template=f'''{prompt}
153
+ '''
154
+
155
+ prompts = [prompt_template.format(prompt=prompt) for prompt in prompts]
156
+
157
+ sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
158
+
159
+ llm = LLM(model="TheBloke/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES-AWQ", quantization="awq", dtype="auto")
160
+
161
+ outputs = llm.generate(prompts, sampling_params)
162
+
163
+ # Print the outputs.
164
+ for output in outputs:
165
+ prompt = output.prompt
166
+ generated_text = output.outputs[0].text
167
+ print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
168
+ ```
169
+ <!-- README_AWQ.md-use-from-vllm start -->
170
+
171
+ <!-- README_AWQ.md-use-from-tgi start -->
172
+ ## Multi-user inference server: Hugging Face Text Generation Inference (TGI)
173
+
174
+ Use TGI version 1.1.0 or later. The official Docker container is: `ghcr.io/huggingface/text-generation-inference:1.1.0`
175
+
176
+ Example Docker parameters:
177
+
178
+ ```shell
179
+ --model-id TheBloke/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES-AWQ --port 3000 --quantize awq --max-input-length 3696 --max-total-tokens 4096 --max-batch-prefill-tokens 4096
180
+ ```
181
+
182
+ Example Python code for interfacing with TGI (requires [huggingface-hub](https://github.com/huggingface/huggingface_hub) 0.17.0 or later):
183
+
184
+ ```shell
185
+ pip3 install huggingface-hub
186
+ ```
187
+
188
+ ```python
189
+ from huggingface_hub import InferenceClient
190
+
191
+ endpoint_url = "https://your-endpoint-url-here"
192
+
193
+ prompt = "Tell me about AI"
194
+ prompt_template=f'''{prompt}
195
+ '''
196
+
197
+ client = InferenceClient(endpoint_url)
198
+ response = client.text_generation(prompt,
199
+ max_new_tokens=128,
200
+ do_sample=True,
201
+ temperature=0.7,
202
+ top_p=0.95,
203
+ top_k=40,
204
+ repetition_penalty=1.1)
205
+
206
+ print(f"Model output: ", response)
207
+ ```
208
+ <!-- README_AWQ.md-use-from-tgi end -->
209
+
210
+ <!-- README_AWQ.md-use-from-python start -->
211
+ ## Inference from Python code using Transformers
212
+
213
+ ### Install the necessary packages
214
+
215
+ - Requires: [Transformers](https://huggingface.co/docs/transformers) 4.35.0 or later.
216
+ - Requires: [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) 0.1.6 or later.
217
+
218
+ ```shell
219
+ pip3 install --upgrade "autoawq>=0.1.6" "transformers>=4.35.0"
220
+ ```
221
+
222
+ Note that if you are using PyTorch 2.0.1, the above AutoAWQ command will automatically upgrade you to PyTorch 2.1.0.
223
+
224
+ If you are using CUDA 11.8 and wish to continue using PyTorch 2.0.1, instead run this command:
225
+
226
+ ```shell
227
+ pip3 install https://github.com/casper-hansen/AutoAWQ/releases/download/v0.1.6/autoawq-0.1.6+cu118-cp310-cp310-linux_x86_64.whl
228
+ ```
229
+
230
+ If you have problems installing [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) using the pre-built wheels, install it from source instead:
231
+
232
+ ```shell
233
+ pip3 uninstall -y autoawq
234
+ git clone https://github.com/casper-hansen/AutoAWQ
235
+ cd AutoAWQ
236
+ pip3 install .
237
+ ```
238
+
239
+ ### Transformers example code (requires Transformers 4.35.0 and later)
240
+
241
+ ```python
242
+ from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
243
+
244
+ model_name_or_path = "TheBloke/Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES-AWQ"
245
+
246
+ tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
247
+ model = AutoModelForCausalLM.from_pretrained(
248
+ model_name_or_path,
249
+ low_cpu_mem_usage=True,
250
+ device_map="cuda:0"
251
+ )
252
+
253
+ # Using the text streamer to stream output one token at a time
254
+ streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
255
+
256
+ prompt = "Tell me about AI"
257
+ prompt_template=f'''{prompt}
258
+ '''
259
+
260
+ # Convert prompt to tokens
261
+ tokens = tokenizer(
262
+ prompt_template,
263
+ return_tensors='pt'
264
+ ).input_ids.cuda()
265
+
266
+ generation_params = {
267
+ "do_sample": True,
268
+ "temperature": 0.7,
269
+ "top_p": 0.95,
270
+ "top_k": 40,
271
+ "max_new_tokens": 512,
272
+ "repetition_penalty": 1.1
273
+ }
274
+
275
+ # Generate streamed output, visible one token at a time
276
+ generation_output = model.generate(
277
+ tokens,
278
+ streamer=streamer,
279
+ **generation_params
280
+ )
281
+
282
+ # Generation without a streamer, which will include the prompt in the output
283
+ generation_output = model.generate(
284
+ tokens,
285
+ **generation_params
286
+ )
287
+
288
+ # Get the tokens from the output, decode them, print them
289
+ token_output = generation_output[0]
290
+ text_output = tokenizer.decode(token_output)
291
+ print("model.generate output: ", text_output)
292
+
293
+ # Inference is also possible via Transformers' pipeline
294
+ from transformers import pipeline
295
+
296
+ pipe = pipeline(
297
+ "text-generation",
298
+ model=model,
299
+ tokenizer=tokenizer,
300
+ **generation_params
301
+ )
302
+
303
+ pipe_output = pipe(prompt_template)[0]['generated_text']
304
+ print("pipeline output: ", pipe_output)
305
+
306
+ ```
307
+ <!-- README_AWQ.md-use-from-python end -->
308
+
309
+ <!-- README_AWQ.md-compatibility start -->
310
+ ## Compatibility
311
+
312
+ The files provided are tested to work with:
313
+
314
+ - [text-generation-webui](https://github.com/oobabooga/text-generation-webui) using `Loader: AutoAWQ`.
315
+ - [vLLM](https://github.com/vllm-project/vllm) version 0.2.0 and later.
316
+ - [Hugging Face Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference) version 1.1.0 and later.
317
+ - [Transformers](https://huggingface.co/docs/transformers) version 4.35.0 and later.
318
+ - [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) version 0.1.1 and later.
319
+
320
+ <!-- README_AWQ.md-compatibility end -->
321
+
322
+ <!-- footer start -->
323
+ <!-- 200823 -->
324
+ ## Discord
325
+
326
+ For further support, and discussions on these models and AI in general, join us at:
327
+
328
+ [TheBloke AI's Discord server](https://discord.gg/theblokeai)
329
+
330
+ ## Thanks, and how to contribute
331
+
332
+ Thanks to the [chirper.ai](https://chirper.ai) team!
333
+
334
+ Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
335
+
336
+ 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.
337
+
338
+ 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.
339
+
340
+ Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
341
+
342
+ * Patreon: https://patreon.com/TheBlokeAI
343
+ * Ko-Fi: https://ko-fi.com/TheBlokeAI
344
+
345
+ **Special thanks to**: Aemon Algiz.
346
+
347
+ **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
348
+
349
+
350
+ Thank you to all my generous patrons and donaters!
351
+
352
+ And thank you again to a16z for their generous grant.
353
+
354
+ <!-- footer end -->
355
+
356
+ # Original model card: Doctor Shotgun's Mixtral 8X7B Instruct V0.1 LimaRP ZLoss DARE TIES
357
+
358
+ # Mixtral-8x7B-Instruct-v0.1-LimaRP-ZLoss-DARE-TIES
359
+
360
+ This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
361
+
362
+ ## Merge Details
363
+ ### Merge Method
364
+
365
+ This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using ./extra_hdd/Mixtral-8x7B-v0.1 as a base.
366
+
367
+ ### Models Merged
368
+
369
+ The following models were included in the merge:
370
+ * ./extra_hdd2/Mixtral-8x7B-Instruct-v0.1
371
+ * ./extra_hdd/Mixtral-8x7B-v0.1-LimaRP-ZLoss
372
+
373
+ ### Configuration
374
+
375
+ The following YAML configuration was used to produce this model:
376
+
377
+ ```yaml
378
+ models:
379
+ - model: ./extra_hdd2/Mixtral-8x7B-Instruct-v0.1
380
+ parameters:
381
+ density: 0.5
382
+ weight: 1.0
383
+ - model: ./extra_hdd/Mixtral-8x7B-v0.1-LimaRP-ZLoss
384
+ parameters:
385
+ density: 0.5
386
+ weight: 0.5
387
+ merge_method: dare_ties
388
+ base_model: ./extra_hdd/Mixtral-8x7B-v0.1
389
+ parameters:
390
+ #normalize: false
391
+ #int8_mask: true
392
+ dtype: bfloat16
393
+
394
+ ```