TheBloke commited on
Commit
46414ad
1 Parent(s): 0f32112

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +426 -0
README.md ADDED
@@ -0,0 +1,426 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Nondzu/zephyr-7b-beta-pl
3
+ inference: false
4
+ language:
5
+ - pl
6
+ license: mit
7
+ model_creator: Kamil
8
+ model_name: Zephyr 7B Beta PL
9
+ model_type: mistral
10
+ prompt_template: 'Below is an instruction that describes a task. Write a response
11
+ that appropriately completes the request.
12
+
13
+
14
+ ### Instruction:
15
+
16
+ {prompt}
17
+
18
+
19
+ ### Response:
20
+
21
+ '
22
+ quantized_by: TheBloke
23
+ ---
24
+ <!-- markdownlint-disable MD041 -->
25
+
26
+ <!-- header start -->
27
+ <!-- 200823 -->
28
+ <div style="width: auto; margin-left: auto; margin-right: auto">
29
+ <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
30
+ </div>
31
+ <div style="display: flex; justify-content: space-between; width: 100%;">
32
+ <div style="display: flex; flex-direction: column; align-items: flex-start;">
33
+ <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
34
+ </div>
35
+ <div style="display: flex; flex-direction: column; align-items: flex-end;">
36
+ <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>
37
+ </div>
38
+ </div>
39
+ <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>
40
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
41
+ <!-- header end -->
42
+
43
+ # Zephyr 7B Beta PL - AWQ
44
+ - Model creator: [Kamil](https://huggingface.co/Nondzu)
45
+ - Original model: [Zephyr 7B Beta PL](https://huggingface.co/Nondzu/zephyr-7b-beta-pl)
46
+
47
+ <!-- description start -->
48
+ ## Description
49
+
50
+ This repo contains AWQ model files for [Kamil's Zephyr 7B Beta PL](https://huggingface.co/Nondzu/zephyr-7b-beta-pl).
51
+
52
+ These files were quantised using hardware kindly provided by [Massed Compute](https://massedcompute.com/).
53
+
54
+
55
+ ### About AWQ
56
+
57
+ 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.
58
+
59
+ It is supported by:
60
+
61
+ - [Text Generation Webui](https://github.com/oobabooga/text-generation-webui) - using Loader: AutoAWQ
62
+ - [vLLM](https://github.com/vllm-project/vllm) - Llama and Mistral models only
63
+ - [Hugging Face Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference)
64
+ - [Transformers](https://huggingface.co/docs/transformers) version 4.35.0 and later, from any code or client that supports Transformers
65
+ - [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) - for use from Python code
66
+
67
+ <!-- description end -->
68
+ <!-- repositories-available start -->
69
+ ## Repositories available
70
+
71
+ * [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/zephyr-7B-beta-pl-AWQ)
72
+ * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/zephyr-7B-beta-pl-GPTQ)
73
+ * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/zephyr-7B-beta-pl-GGUF)
74
+ * [Kamil's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/Nondzu/zephyr-7b-beta-pl)
75
+ <!-- repositories-available end -->
76
+
77
+ <!-- prompt-template start -->
78
+ ## Prompt template: Alpaca
79
+
80
+ ```
81
+ Below is an instruction that describes a task. Write a response that appropriately completes the request.
82
+
83
+ ### Instruction:
84
+ {prompt}
85
+
86
+ ### Response:
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/zephyr-7B-beta-pl-AWQ/tree/main) | 4 | 128 | [polish](https://huggingface.co/datasets/WiktorS/polish-news/viewer/) | 4096 | 4.15 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/zephyr-7B-beta-pl-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: `zephyr-7B-beta-pl-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/zephyr-7B-beta-pl-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'''Below is an instruction that describes a task. Write a response that appropriately completes the request.
153
+
154
+ ### Instruction:
155
+ {prompt}
156
+
157
+ ### Response:
158
+ '''
159
+
160
+ prompts = [prompt_template.format(prompt=prompt) for prompt in prompts]
161
+
162
+ sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
163
+
164
+ llm = LLM(model="TheBloke/zephyr-7B-beta-pl-AWQ", quantization="awq", dtype="auto")
165
+
166
+ outputs = llm.generate(prompts, sampling_params)
167
+
168
+ # Print the outputs.
169
+ for output in outputs:
170
+ prompt = output.prompt
171
+ generated_text = output.outputs[0].text
172
+ print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
173
+ ```
174
+ <!-- README_AWQ.md-use-from-vllm start -->
175
+
176
+ <!-- README_AWQ.md-use-from-tgi start -->
177
+ ## Multi-user inference server: Hugging Face Text Generation Inference (TGI)
178
+
179
+ Use TGI version 1.1.0 or later. The official Docker container is: `ghcr.io/huggingface/text-generation-inference:1.1.0`
180
+
181
+ Example Docker parameters:
182
+
183
+ ```shell
184
+ --model-id TheBloke/zephyr-7B-beta-pl-AWQ --port 3000 --quantize awq --max-input-length 3696 --max-total-tokens 4096 --max-batch-prefill-tokens 4096
185
+ ```
186
+
187
+ Example Python code for interfacing with TGI (requires [huggingface-hub](https://github.com/huggingface/huggingface_hub) 0.17.0 or later):
188
+
189
+ ```shell
190
+ pip3 install huggingface-hub
191
+ ```
192
+
193
+ ```python
194
+ from huggingface_hub import InferenceClient
195
+
196
+ endpoint_url = "https://your-endpoint-url-here"
197
+
198
+ prompt = "Tell me about AI"
199
+ prompt_template=f'''Below is an instruction that describes a task. Write a response that appropriately completes the request.
200
+
201
+ ### Instruction:
202
+ {prompt}
203
+
204
+ ### Response:
205
+ '''
206
+
207
+ client = InferenceClient(endpoint_url)
208
+ response = client.text_generation(prompt,
209
+ max_new_tokens=128,
210
+ do_sample=True,
211
+ temperature=0.7,
212
+ top_p=0.95,
213
+ top_k=40,
214
+ repetition_penalty=1.1)
215
+
216
+ print(f"Model output: ", response)
217
+ ```
218
+ <!-- README_AWQ.md-use-from-tgi end -->
219
+
220
+ <!-- README_AWQ.md-use-from-python start -->
221
+ ## Inference from Python code using Transformers
222
+
223
+ ### Install the necessary packages
224
+
225
+ - Requires: [Transformers](https://huggingface.co/docs/transformers) 4.35.0 or later.
226
+ - Requires: [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) 0.1.6 or later.
227
+
228
+ ```shell
229
+ pip3 install --upgrade "autoawq>=0.1.6" "transformers>=4.35.0"
230
+ ```
231
+
232
+ Note that if you are using PyTorch 2.0.1, the above AutoAWQ command will automatically upgrade you to PyTorch 2.1.0.
233
+
234
+ If you are using CUDA 11.8 and wish to continue using PyTorch 2.0.1, instead run this command:
235
+
236
+ ```shell
237
+ pip3 install https://github.com/casper-hansen/AutoAWQ/releases/download/v0.1.6/autoawq-0.1.6+cu118-cp310-cp310-linux_x86_64.whl
238
+ ```
239
+
240
+ If you have problems installing [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) using the pre-built wheels, install it from source instead:
241
+
242
+ ```shell
243
+ pip3 uninstall -y autoawq
244
+ git clone https://github.com/casper-hansen/AutoAWQ
245
+ cd AutoAWQ
246
+ pip3 install .
247
+ ```
248
+
249
+ ### Transformers example code (requires Transformers 4.35.0 and later)
250
+
251
+ ```python
252
+ from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
253
+
254
+ model_name_or_path = "TheBloke/zephyr-7B-beta-pl-AWQ"
255
+
256
+ tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
257
+ model = AutoModelForCausalLM.from_pretrained(
258
+ model_name_or_path,
259
+ low_cpu_mem_usage=True,
260
+ device_map="cuda:0"
261
+ )
262
+
263
+ # Using the text streamer to stream output one token at a time
264
+ streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
265
+
266
+ prompt = "Tell me about AI"
267
+ prompt_template=f'''Below is an instruction that describes a task. Write a response that appropriately completes the request.
268
+
269
+ ### Instruction:
270
+ {prompt}
271
+
272
+ ### Response:
273
+ '''
274
+
275
+ # Convert prompt to tokens
276
+ tokens = tokenizer(
277
+ prompt_template,
278
+ return_tensors='pt'
279
+ ).input_ids.cuda()
280
+
281
+ generation_params = {
282
+ "do_sample": True,
283
+ "temperature": 0.7,
284
+ "top_p": 0.95,
285
+ "top_k": 40,
286
+ "max_new_tokens": 512,
287
+ "repetition_penalty": 1.1
288
+ }
289
+
290
+ # Generate streamed output, visible one token at a time
291
+ generation_output = model.generate(
292
+ tokens,
293
+ streamer=streamer,
294
+ **generation_params
295
+ )
296
+
297
+ # Generation without a streamer, which will include the prompt in the output
298
+ generation_output = model.generate(
299
+ tokens,
300
+ **generation_params
301
+ )
302
+
303
+ # Get the tokens from the output, decode them, print them
304
+ token_output = generation_output[0]
305
+ text_output = tokenizer.decode(token_output)
306
+ print("model.generate output: ", text_output)
307
+
308
+ # Inference is also possible via Transformers' pipeline
309
+ from transformers import pipeline
310
+
311
+ pipe = pipeline(
312
+ "text-generation",
313
+ model=model,
314
+ tokenizer=tokenizer,
315
+ **generation_params
316
+ )
317
+
318
+ pipe_output = pipe(prompt_template)[0]['generated_text']
319
+ print("pipeline output: ", pipe_output)
320
+
321
+ ```
322
+ <!-- README_AWQ.md-use-from-python end -->
323
+
324
+ <!-- README_AWQ.md-compatibility start -->
325
+ ## Compatibility
326
+
327
+ The files provided are tested to work with:
328
+
329
+ - [text-generation-webui](https://github.com/oobabooga/text-generation-webui) using `Loader: AutoAWQ`.
330
+ - [vLLM](https://github.com/vllm-project/vllm) version 0.2.0 and later.
331
+ - [Hugging Face Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference) version 1.1.0 and later.
332
+ - [Transformers](https://huggingface.co/docs/transformers) version 4.35.0 and later.
333
+ - [AutoAWQ](https://github.com/casper-hansen/AutoAWQ) version 0.1.1 and later.
334
+
335
+ <!-- README_AWQ.md-compatibility end -->
336
+
337
+ <!-- footer start -->
338
+ <!-- 200823 -->
339
+ ## Discord
340
+
341
+ For further support, and discussions on these models and AI in general, join us at:
342
+
343
+ [TheBloke AI's Discord server](https://discord.gg/theblokeai)
344
+
345
+ ## Thanks, and how to contribute
346
+
347
+ Thanks to the [chirper.ai](https://chirper.ai) team!
348
+
349
+ Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
350
+
351
+ 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.
352
+
353
+ 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.
354
+
355
+ Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
356
+
357
+ * Patreon: https://patreon.com/TheBlokeAI
358
+ * Ko-Fi: https://ko-fi.com/TheBlokeAI
359
+
360
+ **Special thanks to**: Aemon Algiz.
361
+
362
+ **Patreon special mentions**: Brandon Frisco, LangChain4j, Spiking Neurons AB, transmissions 11, Joseph William Delisle, Nitin Borwankar, Willem Michiel, Michael Dempsey, vamX, Jeffrey Morgan, zynix, jjj, Omer Bin Jawed, Sean Connelly, jinyuan sun, Jeromy Smith, Shadi, Pawan Osman, Chadd, Elijah Stavena, Illia Dulskyi, Sebastain Graf, Stephen Murray, terasurfer, Edmond Seymore, Celu Ramasamy, Mandus, Alex, biorpg, Ajan Kanaga, Clay Pascal, Raven Klaugh, 阿明, K, ya boyyy, usrbinkat, Alicia Loh, John Villwock, ReadyPlayerEmma, Chris Smitley, Cap'n Zoog, fincy, GodLy, S_X, sidney chen, Cory Kujawski, OG, Mano Prime, AzureBlack, Pieter, Kalila, Spencer Kim, Tom X Nguyen, Stanislav Ovsiannikov, Michael Levine, Andrey, Trailburnt, Vadim, Enrico Ros, Talal Aujan, Brandon Phillips, Jack West, Eugene Pentland, Michael Davis, Will Dee, webtim, Jonathan Leane, Alps Aficionado, Rooh Singh, Tiffany J. Kim, theTransient, Luke @flexchar, Elle, Caitlyn Gatomon, Ari Malik, subjectnull, Johann-Peter Hartmann, Trenton Dambrowitz, Imad Khwaja, Asp the Wyvern, Emad Mostaque, Rainer Wilmers, Alexandros Triantafyllidis, Nicholas, Pedro Madruga, SuperWojo, Harry Royden McLaughlin, James Bentley, Olakabola, David Ziegler, Ai Maven, Jeff Scroggin, Nikolai Manek, Deo Leter, Matthew Berman, Fen Risland, Ken Nordquist, Manuel Alberto Morcote, Luke Pendergrass, TL, Fred von Graf, Randy H, Dan Guido, NimbleBox.ai, Vitor Caleffi, Gabriel Tamborski, knownsqashed, Lone Striker, Erik Bjäreholt, John Detwiler, Leonard Tan, Iucharbius
363
+
364
+
365
+ Thank you to all my generous patrons and donaters!
366
+
367
+ And thank you again to a16z for their generous grant.
368
+
369
+ <!-- footer end -->
370
+
371
+ # Original model card: Kamil's Zephyr 7B Beta PL
372
+
373
+
374
+ ## Model Overview
375
+ The model is a result of advanced fine-tuning methods applied to a base model, focusing on enhancing its capabilities for specific Polish language datasets. It incorporates cutting-edge techniques and is built upon the Zephyr Beta model framework.
376
+
377
+ ## Current Status: Alpha
378
+ - **Stage**: Alpha-Alpaca
379
+
380
+ ## Training Details
381
+
382
+ I trained the model using 3xRTX 3090 for 163 hours.
383
+ [![Built with Axolotl](https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png)](https://github.com/OpenAccess-AI-Collective/axolotl)
384
+
385
+ ## Model Specifics
386
+ - **Base Model**: HuggingFaceH4/zephyr-7b-beta
387
+ - **Fine-Tuning Method**: QLORA
388
+ - **Primary Focus**: Polish language datasets
389
+
390
+ ## Datasets:
391
+ - Dataset 1 Name: Lajonbot/alpaca-dolly-chrisociepa-instruction-only-polish
392
+ - Dataset 1 Link: [Lajonbot/alpaca-dolly-chrisociepa-instruction-only-polish](https://huggingface.co/datasets/Lajonbot/alpaca-dolly-chrisociepa-instruction-only-polish?row=16)
393
+ - Dataset 2 Name: klima7/polish-prose
394
+ - Dataset 2 Link: [klima7/polish-prose](https://huggingface.co/datasets/klima7/polish-prose)
395
+
396
+ ## Usage Warning
397
+ As this is an experimental model, users should be aware of the following:
398
+ - **Reliability**: The model has not been fully tested and may exhibit unexpected behaviors or performance issues.
399
+ - **Updates**: The model is subject to change based on ongoing testing and feedback.
400
+ - **Data Sensitivity**: Users should exercise caution when using sensitive or private data, as the model's output and behavior are not fully predictable at this stage.
401
+
402
+ ## Prompt template: Alpaca
403
+
404
+ ```
405
+ Below is an instruction that describes a task. Write a response that appropriately completes the request.
406
+
407
+ ### Instruction:
408
+ {prompt}
409
+
410
+ ### Response:
411
+
412
+ ```
413
+
414
+ ## Example
415
+
416
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/63729f35acef705233c87909/1WYp9Su1NYvYCIU-2J7TG.png)
417
+
418
+ ## Feedback and Contribution
419
+ User feedback is crucial during this testing phase. We encourage users to provide feedback on model performance, issues encountered, and any suggestions for improvements. Contributions in terms of shared test results, datasets, or code improvements are also welcome.
420
+
421
+ ---
422
+
423
+ **Disclaimer**: This experimental model is provided 'as is', without warranty of any kind. Users should use the model at their own risk. The creators or maintainers of the model are not responsible for any consequences arising from its use.
424
+
425
+
426
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/63729f35acef705233c87909/CPClYNIMp3Qswt2F0Y9B3.png)