TheBloke commited on
Commit
4611d1f
1 Parent(s): 3cba302

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +375 -0
README.md ADDED
@@ -0,0 +1,375 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: TencentARC/LLaMA-Pro-8B-Instruct
3
+ inference: false
4
+ license: llama2
5
+ model_creator: ARC Lab, Tencent PCG
6
+ model_name: Llama Pro 8B Instruct
7
+ model_type: llama
8
+ prompt_template: '<|user|>
9
+
10
+ {prompt}
11
+
12
+ <|assistant|>
13
+
14
+ '
15
+ quantized_by: TheBloke
16
+ ---
17
+ <!-- markdownlint-disable MD041 -->
18
+
19
+ <!-- header start -->
20
+ <!-- 200823 -->
21
+ <div style="width: auto; margin-left: auto; margin-right: auto">
22
+ <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
23
+ </div>
24
+ <div style="display: flex; justify-content: space-between; width: 100%;">
25
+ <div style="display: flex; flex-direction: column; align-items: flex-start;">
26
+ <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
27
+ </div>
28
+ <div style="display: flex; flex-direction: column; align-items: flex-end;">
29
+ <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>
30
+ </div>
31
+ </div>
32
+ <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>
33
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
34
+ <!-- header end -->
35
+
36
+ # Llama Pro 8B Instruct - AWQ
37
+ - Model creator: [ARC Lab, Tencent PCG](https://huggingface.co/TencentARC)
38
+ - Original model: [Llama Pro 8B Instruct](https://huggingface.co/TencentARC/LLaMA-Pro-8B-Instruct)
39
+
40
+ <!-- description start -->
41
+ ## Description
42
+
43
+ This repo contains AWQ model files for [ARC Lab, Tencent PCG's Llama Pro 8B Instruct](https://huggingface.co/TencentARC/LLaMA-Pro-8B-Instruct).
44
+
45
+ These files were quantised using hardware kindly provided by [Massed Compute](https://massedcompute.com/).
46
+
47
+
48
+ ### About AWQ
49
+
50
+ 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.
51
+
52
+ AWQ models are currently supported on Linux and Windows, with NVidia GPUs only. macOS users: please use GGUF models instead.
53
+
54
+ It is supported by:
55
+
56
+ - [Text Generation Webui](https://github.com/oobabooga/text-generation-webui) - using Loader: AutoAWQ
57
+ - [vLLM](https://github.com/vllm-project/vllm) - version 0.2.2 or later for support for all model types.
58
+ - [Hugging Face Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference)
59
+ - [Transformers](https://huggingface.co/docs/transformers) version 4.35.0 and later, from any code or client that supports Transformers
60
+ - [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) - for use from Python code
61
+
62
+ <!-- description end -->
63
+ <!-- repositories-available start -->
64
+ ## Repositories available
65
+
66
+ * [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/LLaMA-Pro-8B-Instruct-AWQ)
67
+ * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/LLaMA-Pro-8B-Instruct-GPTQ)
68
+ * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/LLaMA-Pro-8B-Instruct-GGUF)
69
+ * [ARC Lab, Tencent PCG's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/TencentARC/LLaMA-Pro-8B-Instruct)
70
+ <!-- repositories-available end -->
71
+
72
+ <!-- prompt-template start -->
73
+ ## Prompt template: ToRA
74
+
75
+ ```
76
+ <|user|>
77
+ {prompt}
78
+ <|assistant|>
79
+
80
+ ```
81
+
82
+ <!-- prompt-template end -->
83
+
84
+
85
+ <!-- README_AWQ.md-provided-files start -->
86
+ ## Provided files, and AWQ parameters
87
+
88
+ I currently release 128g GEMM models only. The addition of group_size 32 models, and GEMV kernel models, is being actively considered.
89
+
90
+ Models are released as sharded safetensors files.
91
+
92
+ | Branch | Bits | GS | AWQ Dataset | Seq Len | Size |
93
+ | ------ | ---- | -- | ----------- | ------- | ---- |
94
+ | [main](https://huggingface.co/TheBloke/LLaMA-Pro-8B-Instruct-AWQ/tree/main) | 4 | 128 | [VMware Open Instruct](https://huggingface.co/datasets/VMware/open-instruct/viewer/) | 4096 | 4.73 GB
95
+
96
+ <!-- README_AWQ.md-provided-files end -->
97
+
98
+ <!-- README_AWQ.md-text-generation-webui start -->
99
+ ## How to easily download and use this model in [text-generation-webui](https://github.com/oobabooga/text-generation-webui)
100
+
101
+ Please make sure you're using the latest version of [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
102
+
103
+ 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.
104
+
105
+ 1. Click the **Model tab**.
106
+ 2. Under **Download custom model or LoRA**, enter `TheBloke/LLaMA-Pro-8B-Instruct-AWQ`.
107
+ 3. Click **Download**.
108
+ 4. The model will start downloading. Once it's finished it will say "Done".
109
+ 5. In the top left, click the refresh icon next to **Model**.
110
+ 6. In the **Model** dropdown, choose the model you just downloaded: `LLaMA-Pro-8B-Instruct-AWQ`
111
+ 7. Select **Loader: AutoAWQ**.
112
+ 8. Click Load, and the model will load and is now ready for use.
113
+ 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.
114
+ 10. Once you're ready, click the **Text Generation** tab and enter a prompt to get started!
115
+ <!-- README_AWQ.md-text-generation-webui end -->
116
+
117
+ <!-- README_AWQ.md-use-from-vllm start -->
118
+ ## Multi-user inference server: vLLM
119
+
120
+ Documentation on installing and using vLLM [can be found here](https://vllm.readthedocs.io/en/latest/).
121
+
122
+ - Please ensure you are using vLLM version 0.2 or later.
123
+ - When using vLLM as a server, pass the `--quantization awq` parameter.
124
+
125
+ For example:
126
+
127
+ ```shell
128
+ python3 -m vllm.entrypoints.api_server --model TheBloke/LLaMA-Pro-8B-Instruct-AWQ --quantization awq --dtype auto
129
+ ```
130
+
131
+ - When using vLLM from Python code, again set `quantization=awq`.
132
+
133
+ For example:
134
+
135
+ ```python
136
+ from vllm import LLM, SamplingParams
137
+
138
+ prompts = [
139
+ "Tell me about AI",
140
+ "Write a story about llamas",
141
+ "What is 291 - 150?",
142
+ "How much wood would a woodchuck chuck if a woodchuck could chuck wood?",
143
+ ]
144
+ prompt_template=f'''<|user|>
145
+ {prompt}
146
+ <|assistant|>
147
+ '''
148
+
149
+ prompts = [prompt_template.format(prompt=prompt) for prompt in prompts]
150
+
151
+ sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
152
+
153
+ llm = LLM(model="TheBloke/LLaMA-Pro-8B-Instruct-AWQ", quantization="awq", dtype="auto")
154
+
155
+ outputs = llm.generate(prompts, sampling_params)
156
+
157
+ # Print the outputs.
158
+ for output in outputs:
159
+ prompt = output.prompt
160
+ generated_text = output.outputs[0].text
161
+ print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
162
+ ```
163
+ <!-- README_AWQ.md-use-from-vllm start -->
164
+
165
+ <!-- README_AWQ.md-use-from-tgi start -->
166
+ ## Multi-user inference server: Hugging Face Text Generation Inference (TGI)
167
+
168
+ Use TGI version 1.1.0 or later. The official Docker container is: `ghcr.io/huggingface/text-generation-inference:1.1.0`
169
+
170
+ Example Docker parameters:
171
+
172
+ ```shell
173
+ --model-id TheBloke/LLaMA-Pro-8B-Instruct-AWQ --port 3000 --quantize awq --max-input-length 3696 --max-total-tokens 4096 --max-batch-prefill-tokens 4096
174
+ ```
175
+
176
+ Example Python code for interfacing with TGI (requires [huggingface-hub](https://github.com/huggingface/huggingface_hub) 0.17.0 or later):
177
+
178
+ ```shell
179
+ pip3 install huggingface-hub
180
+ ```
181
+
182
+ ```python
183
+ from huggingface_hub import InferenceClient
184
+
185
+ endpoint_url = "https://your-endpoint-url-here"
186
+
187
+ prompt = "Tell me about AI"
188
+ prompt_template=f'''<|user|>
189
+ {prompt}
190
+ <|assistant|>
191
+ '''
192
+
193
+ client = InferenceClient(endpoint_url)
194
+ response = client.text_generation(prompt,
195
+ max_new_tokens=128,
196
+ do_sample=True,
197
+ temperature=0.7,
198
+ top_p=0.95,
199
+ top_k=40,
200
+ repetition_penalty=1.1)
201
+
202
+ print(f"Model output: ", response)
203
+ ```
204
+ <!-- README_AWQ.md-use-from-tgi end -->
205
+
206
+ <!-- README_AWQ.md-use-from-python start -->
207
+ ## Inference from Python code using Transformers
208
+
209
+ ### Install the necessary packages
210
+
211
+ - Requires: [Transformers](https://huggingface.co/docs/transformers) 4.35.0 or later.
212
+ - Requires: [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) 0.1.6 or later.
213
+
214
+ ```shell
215
+ pip3 install --upgrade "autoawq>=0.1.6" "transformers>=4.35.0"
216
+ ```
217
+
218
+ Note that if you are using PyTorch 2.0.1, the above AutoAWQ command will automatically upgrade you to PyTorch 2.1.0.
219
+
220
+ If you are using CUDA 11.8 and wish to continue using PyTorch 2.0.1, instead run this command:
221
+
222
+ ```shell
223
+ pip3 install https://github.com/casper-hansen/AutoAWQ/releases/download/v0.1.6/autoawq-0.1.6+cu118-cp310-cp310-linux_x86_64.whl
224
+ ```
225
+
226
+ If you have problems installing [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) using the pre-built wheels, install it from source instead:
227
+
228
+ ```shell
229
+ pip3 uninstall -y autoawq
230
+ git clone https://github.com/casper-hansen/AutoAWQ
231
+ cd AutoAWQ
232
+ pip3 install .
233
+ ```
234
+
235
+ ### Transformers example code (requires Transformers 4.35.0 and later)
236
+
237
+ ```python
238
+ from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
239
+
240
+ model_name_or_path = "TheBloke/LLaMA-Pro-8B-Instruct-AWQ"
241
+
242
+ tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
243
+ model = AutoModelForCausalLM.from_pretrained(
244
+ model_name_or_path,
245
+ low_cpu_mem_usage=True,
246
+ device_map="cuda:0"
247
+ )
248
+
249
+ # Using the text streamer to stream output one token at a time
250
+ streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
251
+
252
+ prompt = "Tell me about AI"
253
+ prompt_template=f'''<|user|>
254
+ {prompt}
255
+ <|assistant|>
256
+ '''
257
+
258
+ # Convert prompt to tokens
259
+ tokens = tokenizer(
260
+ prompt_template,
261
+ return_tensors='pt'
262
+ ).input_ids.cuda()
263
+
264
+ generation_params = {
265
+ "do_sample": True,
266
+ "temperature": 0.7,
267
+ "top_p": 0.95,
268
+ "top_k": 40,
269
+ "max_new_tokens": 512,
270
+ "repetition_penalty": 1.1
271
+ }
272
+
273
+ # Generate streamed output, visible one token at a time
274
+ generation_output = model.generate(
275
+ tokens,
276
+ streamer=streamer,
277
+ **generation_params
278
+ )
279
+
280
+ # Generation without a streamer, which will include the prompt in the output
281
+ generation_output = model.generate(
282
+ tokens,
283
+ **generation_params
284
+ )
285
+
286
+ # Get the tokens from the output, decode them, print them
287
+ token_output = generation_output[0]
288
+ text_output = tokenizer.decode(token_output)
289
+ print("model.generate output: ", text_output)
290
+
291
+ # Inference is also possible via Transformers' pipeline
292
+ from transformers import pipeline
293
+
294
+ pipe = pipeline(
295
+ "text-generation",
296
+ model=model,
297
+ tokenizer=tokenizer,
298
+ **generation_params
299
+ )
300
+
301
+ pipe_output = pipe(prompt_template)[0]['generated_text']
302
+ print("pipeline output: ", pipe_output)
303
+
304
+ ```
305
+ <!-- README_AWQ.md-use-from-python end -->
306
+
307
+ <!-- README_AWQ.md-compatibility start -->
308
+ ## Compatibility
309
+
310
+ The files provided are tested to work with:
311
+
312
+ - [text-generation-webui](https://github.com/oobabooga/text-generation-webui) using `Loader: AutoAWQ`.
313
+ - [vLLM](https://github.com/vllm-project/vllm) version 0.2.0 and later.
314
+ - [Hugging Face Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference) version 1.1.0 and later.
315
+ - [Transformers](https://huggingface.co/docs/transformers) version 4.35.0 and later.
316
+ - [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) version 0.1.1 and later.
317
+
318
+ <!-- README_AWQ.md-compatibility end -->
319
+
320
+ <!-- footer start -->
321
+ <!-- 200823 -->
322
+ ## Discord
323
+
324
+ For further support, and discussions on these models and AI in general, join us at:
325
+
326
+ [TheBloke AI's Discord server](https://discord.gg/theblokeai)
327
+
328
+ ## Thanks, and how to contribute
329
+
330
+ Thanks to the [chirper.ai](https://chirper.ai) team!
331
+
332
+ Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
333
+
334
+ 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.
335
+
336
+ 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.
337
+
338
+ Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
339
+
340
+ * Patreon: https://patreon.com/TheBlokeAI
341
+ * Ko-Fi: https://ko-fi.com/TheBlokeAI
342
+
343
+ **Special thanks to**: Aemon Algiz.
344
+
345
+ **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
346
+
347
+
348
+ Thank you to all my generous patrons and donaters!
349
+
350
+ And thank you again to a16z for their generous grant.
351
+
352
+ <!-- footer end -->
353
+
354
+ # Original model card: ARC Lab, Tencent PCG's Llama Pro 8B Instruct
355
+
356
+
357
+ # LLaMA-PRO-Instruct Model Card
358
+
359
+ ## Model Description
360
+ LLaMA-PRO-Instruct is a transformative expansion of the LLaMA2-7B model, now boasting 8.3 billion parameters. It uniquely specializes in programming, coding, and mathematical reasoning, maintaining versatility in general language tasks.
361
+
362
+ ## Development and Training
363
+ This model, developed by Tencent ARC team, extends LLaMA2-7B using innovative block expansion techniques. It's meticulously trained on a diverse blend of coding and mathematical data, encompassing over 80 billion tokens.
364
+
365
+ ## Intended Use
366
+ LLaMA-PRO-Instruct is ideal for complex NLP challenges, excelling in programming, mathematical reasoning, and general language processing, suitable for both specialized and broad applications.
367
+
368
+ ## Performance
369
+ It surpasses its predecessors in the LLaMA series, especially in code domains, demonstrating exceptional competence as a comprehensive language model.
370
+
371
+ ## Limitations
372
+ Despite advancements, it may encounter difficulties in highly niche or nuanced tasks.
373
+
374
+ ## Ethical Considerations
375
+ Users are advised to consider inherent biases and responsibly manage its application across various fields.