TheBloke commited on
Commit
833652b
1 Parent(s): f8c1bee

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +55 -42
README.md CHANGED
@@ -44,39 +44,47 @@ tags:
44
  - Model creator: [Phind](https://huggingface.co/Phind)
45
  - Original model: [Phind CodeLlama 34B Python v1](https://huggingface.co/Phind/Phind-CodeLlama-34B-Python-v1)
46
 
 
47
  ## Description
48
 
49
  This repo contains GPTQ model files for [Phind's Phind CodeLlama 34B Python v1](https://huggingface.co/Phind/Phind-CodeLlama-34B-Python-v1).
50
 
51
  Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.
52
 
 
 
53
  ## Repositories available
54
 
55
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ)
56
  * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GGUF)
57
- * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference (deprecated)](https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GGML)
58
  * [Phind's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/Phind/Phind-CodeLlama-34B-Python-v1)
 
59
 
 
60
  ## Prompt template: Plain-with-newline
61
 
62
  ```
63
  {prompt} \n
 
64
  ```
65
 
 
 
 
66
  ## Provided files and GPTQ parameters
67
 
68
  Multiple quantisation parameters are provided, to allow you to choose the best one for your hardware and requirements.
69
 
70
  Each separate quant is in a different branch. See below for instructions on fetching from different branches.
71
 
72
- All GPTQ files are made with AutoGPTQ.
73
 
74
  <details>
75
  <summary>Explanation of GPTQ parameters</summary>
76
 
77
  - Bits: The bit size of the quantised model.
78
  - GS: GPTQ group size. Higher numbers use less VRAM, but have lower quantisation accuracy. "None" is the lowest possible value.
79
- - Act Order: True or False. Also known as `desc_act`. True results in better quantisation accuracy. Some GPTQ clients have issues with models that use Act Order plus Group Size.
80
  - Damp %: A GPTQ parameter that affects how samples are processed for quantisation. 0.01 is default, but 0.1 results in slightly better accuracy.
81
  - GPTQ dataset: The dataset used for quantisation. Using a dataset more appropriate to the model's training can improve quantisation accuracy. Note that the GPTQ dataset is not the same as the dataset used to train the model - please refer to the original model repo for details of the training dataset(s).
82
  - Sequence Length: The length of the dataset sequences used for quantisation. Ideally this is the same as the model sequence length. For some very long sequence models (16+K), a lower sequence length may have to be used. Note that a lower sequence length does not limit the sequence length of the quantised model. It only impacts the quantisation accuracy on longer inference sequences.
@@ -93,6 +101,9 @@ All GPTQ files are made with AutoGPTQ.
93
  | [gptq-3bit--1g-actorder_True](https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ/tree/gptq-3bit--1g-actorder_True) | 3 | None | Yes | 0.1 | [Evol Instruct Code](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1) | 8192 | 13.54 GB | No | 3-bit, with Act Order and no group size. Lowest possible VRAM requirements. May be lower quality than 3-bit 128g. |
94
  | [gptq-3bit-128g-actorder_True](https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ/tree/gptq-3bit-128g-actorder_True) | 3 | 128 | Yes | 0.1 | [Evol Instruct Code](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1) | 8192 | 14.14 GB | No | 3-bit, with group size 128g and act-order. Higher quality than 128g-False but poor AutoGPTQ CUDA speed. |
95
 
 
 
 
96
  ## How to download from branches
97
 
98
  - In text-generation-webui, you can add `:branch` to the end of the download name, eg `TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ:gptq-4bit-32g-actorder_True`
@@ -101,75 +112,75 @@ All GPTQ files are made with AutoGPTQ.
101
  git clone --single-branch --branch gptq-4bit-32g-actorder_True https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ
102
  ```
103
  - In Python Transformers code, the branch is the `revision` parameter; see below.
104
-
 
105
  ## How to easily download and use this model in [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
106
 
107
  Please make sure you're using the latest version of [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
108
 
109
- It is strongly recommended to use the text-generation-webui one-click-installers unless you know how to make a manual install.
110
 
111
  1. Click the **Model tab**.
112
  2. Under **Download custom model or LoRA**, enter `TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ`.
113
  - To download from a specific branch, enter for example `TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ:gptq-4bit-32g-actorder_True`
114
  - see Provided Files above for the list of branches for each option.
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: `Phind-CodeLlama-34B-Python-v1-GPTQ`
119
  7. The model will automatically load, and is now ready for use!
120
  8. 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.
121
- * Note that you do not need to set GPTQ parameters any more. These are set automatically from the file `quantize_config.json`.
122
  9. Once you're ready, click the **Text Generation tab** and enter a prompt to get started!
 
123
 
 
124
  ## How to use this GPTQ model from Python code
125
 
126
- First make sure you have [AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ) 0.3.1 or later installed:
127
 
128
- ```
129
- pip3 install auto-gptq
130
- ```
131
 
132
- If you have problems installing AutoGPTQ, please build from source instead:
 
 
133
  ```
 
 
 
 
134
  pip3 uninstall -y auto-gptq
135
  git clone https://github.com/PanQiWei/AutoGPTQ
136
  cd AutoGPTQ
137
  pip3 install .
138
  ```
139
 
140
- Then try the following example code:
 
 
 
 
 
 
 
 
141
 
142
  ```python
143
- from transformers import AutoTokenizer, pipeline, logging
144
- from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
145
 
146
  model_name_or_path = "TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ"
147
-
148
- use_triton = False
 
 
 
 
149
 
150
  tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
151
 
152
- model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
153
- use_safetensors=True,
154
- trust_remote_code=False,
155
- device="cuda:0",
156
- use_triton=use_triton,
157
- quantize_config=None)
158
-
159
- """
160
- # To download from a specific branch, use the revision parameter, as in this example:
161
- # Note that `revision` requires AutoGPTQ 0.3.1 or later!
162
-
163
- model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
164
- revision="gptq-4bit-32g-actorder_True",
165
- use_safetensors=True,
166
- trust_remote_code=False,
167
- device="cuda:0",
168
- quantize_config=None)
169
- """
170
-
171
  prompt = "Tell me about AI"
172
  prompt_template=f'''{prompt} \n
 
173
  '''
174
 
175
  print("\n\n*** Generate:")
@@ -180,9 +191,6 @@ print(tokenizer.decode(output[0]))
180
 
181
  # Inference can also be done using transformers' pipeline
182
 
183
- # Prevent printing spurious transformers error when using pipeline with AutoGPTQ
184
- logging.set_verbosity(logging.CRITICAL)
185
-
186
  print("*** Pipeline:")
187
  pipe = pipeline(
188
  "text-generation",
@@ -196,12 +204,17 @@ pipe = pipeline(
196
 
197
  print(pipe(prompt_template)[0]['generated_text'])
198
  ```
 
199
 
 
200
  ## Compatibility
201
 
202
- The files provided will work with AutoGPTQ (CUDA and Triton modes), GPTQ-for-LLaMa (only CUDA has been tested), and Occ4m's GPTQ-for-LLaMa fork.
 
 
203
 
204
- ExLlama works with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
 
205
 
206
  <!-- footer start -->
207
  <!-- 200823 -->
@@ -226,7 +239,7 @@ Donaters will get priority support on any and all AI/LLM/model questions and req
226
 
227
  **Special thanks to**: Aemon Algiz.
228
 
229
- **Patreon special mentions**: Kacper Wikieł, knownsqashed, Leonard Tan, Asp the Wyvern, Daniel P. Andersen, Luke Pendergrass, Stanislav Ovsiannikov, RoA, Dave, Ai Maven, Kalila, Will Dee, Imad Khwaja, Nitin Borwankar, Joseph William Delisle, Tony Hughes, Cory Kujawski, Rishabh Srivastava, Russ Johnson, Stephen Murray, Lone Striker, Johann-Peter Hartmann, Elle, J, Deep Realms, SuperWojo, Raven Klaugh, Sebastain Graf, ReadyPlayerEmma, Alps Aficionado, Mano Prime, Derek Yates, Gabriel Puliatti, Mesiah Bishop, Magnesian, Sean Connelly, biorpg, Iucharbius, Olakabola, Fen Risland, Space Cruiser, theTransient, Illia Dulskyi, Thomas Belote, Spencer Kim, Pieter, John Detwiler, Fred von Graf, Michael Davis, Swaroop Kallakuri, subjectnull, Clay Pascal, Subspace Studios, Chris Smitley, Enrico Ros, usrbinkat, Steven Wood, alfie_i, David Ziegler, Willem Michiel, Matthew Berman, Andrey, Pyrater, Jeffrey Morgan, vamX, LangChain4j, Luke @flexchar, Trenton Dambrowitz, Pierre Kircher, Alex, Sam, James Bentley, Edmond Seymore, Eugene Pentland, Pedro Madruga, Rainer Wilmers, Dan Guido, Nathan LeClaire, Spiking Neurons AB, Talal Aujan, zynix, Artur Olbinski, Michael Levine, 阿明, K, John Villwock, Nikolai Manek, Femi Adebogun, senxiiz, Deo Leter, NimbleBox.ai, Viktor Bowallius, Geoffrey Montalvo, Mandus, Ajan Kanaga, ya boyyy, Jonathan Leane, webtim, Brandon Frisco, danny, Alexandros Triantafyllidis, Gabriel Tamborski, Randy H, terasurfer, Vadim, Junyu Yang, Vitor Caleffi, Chadd, transmissions 11
230
 
231
 
232
  Thank you to all my generous patrons and donaters!
 
44
  - Model creator: [Phind](https://huggingface.co/Phind)
45
  - Original model: [Phind CodeLlama 34B Python v1](https://huggingface.co/Phind/Phind-CodeLlama-34B-Python-v1)
46
 
47
+ <!-- description start -->
48
  ## Description
49
 
50
  This repo contains GPTQ model files for [Phind's Phind CodeLlama 34B Python v1](https://huggingface.co/Phind/Phind-CodeLlama-34B-Python-v1).
51
 
52
  Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.
53
 
54
+ <!-- description end -->
55
+ <!-- repositories-available start -->
56
  ## Repositories available
57
 
58
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ)
59
  * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GGUF)
 
60
  * [Phind's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/Phind/Phind-CodeLlama-34B-Python-v1)
61
+ <!-- repositories-available end -->
62
 
63
+ <!-- prompt-template start -->
64
  ## Prompt template: Plain-with-newline
65
 
66
  ```
67
  {prompt} \n
68
+
69
  ```
70
 
71
+ <!-- prompt-template end -->
72
+
73
+ <!-- README_GPTQ.md-provided-files start -->
74
  ## Provided files and GPTQ parameters
75
 
76
  Multiple quantisation parameters are provided, to allow you to choose the best one for your hardware and requirements.
77
 
78
  Each separate quant is in a different branch. See below for instructions on fetching from different branches.
79
 
80
+ All recent GPTQ files are made with AutoGPTQ, and all files in non-main branches are made with AutoGPTQ. Files in the `main` branch which were uploaded before August 2023 were made with GPTQ-for-LLaMa.
81
 
82
  <details>
83
  <summary>Explanation of GPTQ parameters</summary>
84
 
85
  - Bits: The bit size of the quantised model.
86
  - GS: GPTQ group size. Higher numbers use less VRAM, but have lower quantisation accuracy. "None" is the lowest possible value.
87
+ - Act Order: True or False. Also known as `desc_act`. True results in better quantisation accuracy. Some GPTQ clients have had issues with models that use Act Order plus Group Size, but this is generally resolved now.
88
  - Damp %: A GPTQ parameter that affects how samples are processed for quantisation. 0.01 is default, but 0.1 results in slightly better accuracy.
89
  - GPTQ dataset: The dataset used for quantisation. Using a dataset more appropriate to the model's training can improve quantisation accuracy. Note that the GPTQ dataset is not the same as the dataset used to train the model - please refer to the original model repo for details of the training dataset(s).
90
  - Sequence Length: The length of the dataset sequences used for quantisation. Ideally this is the same as the model sequence length. For some very long sequence models (16+K), a lower sequence length may have to be used. Note that a lower sequence length does not limit the sequence length of the quantised model. It only impacts the quantisation accuracy on longer inference sequences.
 
101
  | [gptq-3bit--1g-actorder_True](https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ/tree/gptq-3bit--1g-actorder_True) | 3 | None | Yes | 0.1 | [Evol Instruct Code](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1) | 8192 | 13.54 GB | No | 3-bit, with Act Order and no group size. Lowest possible VRAM requirements. May be lower quality than 3-bit 128g. |
102
  | [gptq-3bit-128g-actorder_True](https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ/tree/gptq-3bit-128g-actorder_True) | 3 | 128 | Yes | 0.1 | [Evol Instruct Code](https://huggingface.co/datasets/nickrosh/Evol-Instruct-Code-80k-v1) | 8192 | 14.14 GB | No | 3-bit, with group size 128g and act-order. Higher quality than 128g-False but poor AutoGPTQ CUDA speed. |
103
 
104
+ <!-- README_GPTQ.md-provided-files end -->
105
+
106
+ <!-- README_GPTQ.md-download-from-branches start -->
107
  ## How to download from branches
108
 
109
  - In text-generation-webui, you can add `:branch` to the end of the download name, eg `TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ:gptq-4bit-32g-actorder_True`
 
112
  git clone --single-branch --branch gptq-4bit-32g-actorder_True https://huggingface.co/TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ
113
  ```
114
  - In Python Transformers code, the branch is the `revision` parameter; see below.
115
+ <!-- README_GPTQ.md-download-from-branches end -->
116
+ <!-- README_GPTQ.md-text-generation-webui start -->
117
  ## How to easily download and use this model in [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
118
 
119
  Please make sure you're using the latest version of [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
120
 
121
+ 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.
122
 
123
  1. Click the **Model tab**.
124
  2. Under **Download custom model or LoRA**, enter `TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ`.
125
  - To download from a specific branch, enter for example `TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ:gptq-4bit-32g-actorder_True`
126
  - see Provided Files above for the list of branches for each option.
127
  3. Click **Download**.
128
+ 4. The model will start downloading. Once it's finished it will say "Done".
129
  5. In the top left, click the refresh icon next to **Model**.
130
  6. In the **Model** dropdown, choose the model you just downloaded: `Phind-CodeLlama-34B-Python-v1-GPTQ`
131
  7. The model will automatically load, and is now ready for use!
132
  8. 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.
133
+ * Note that you do not need to and should not set manual GPTQ parameters any more. These are set automatically from the file `quantize_config.json`.
134
  9. Once you're ready, click the **Text Generation tab** and enter a prompt to get started!
135
+ <!-- README_GPTQ.md-text-generation-webui end -->
136
 
137
+ <!-- README_GPTQ.md-use-from-python start -->
138
  ## How to use this GPTQ model from Python code
139
 
140
+ ### Install the necessary packages
141
 
142
+ Requires: Transformers 4.32.0 or later, Optimum 1.12.0 or later, and AutoGPTQ 0.4.2 or later.
 
 
143
 
144
+ ```shell
145
+ pip3 install transformers>=4.32.0 optimum>=1.12.0
146
+ pip3 install auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/ # Use cu117 if on CUDA 11.7
147
  ```
148
+
149
+ If you have problems installing AutoGPTQ using the pre-built wheels, install it from source instead:
150
+
151
+ ```shell
152
  pip3 uninstall -y auto-gptq
153
  git clone https://github.com/PanQiWei/AutoGPTQ
154
  cd AutoGPTQ
155
  pip3 install .
156
  ```
157
 
158
+ ### For CodeLlama models only: you must use Transformers 4.33.0 or later.
159
+
160
+ If 4.33.0 is not yet released when you read this, you will need to install Transformers from source:
161
+ ```shell
162
+ pip3 uninstall -y transformers
163
+ pip3 install git+https://github.com/huggingface/transformers.git
164
+ ```
165
+
166
+ ### You can then use the following code
167
 
168
  ```python
169
+ from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
 
170
 
171
  model_name_or_path = "TheBloke/Phind-CodeLlama-34B-Python-v1-GPTQ"
172
+ # To use a different branch, change revision
173
+ # For example: revision="gptq-4bit-32g-actorder_True"
174
+ model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
175
+ torch_dtype=torch.float16,
176
+ device_map="auto",
177
+ revision="main")
178
 
179
  tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
180
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
181
  prompt = "Tell me about AI"
182
  prompt_template=f'''{prompt} \n
183
+
184
  '''
185
 
186
  print("\n\n*** Generate:")
 
191
 
192
  # Inference can also be done using transformers' pipeline
193
 
 
 
 
194
  print("*** Pipeline:")
195
  pipe = pipeline(
196
  "text-generation",
 
204
 
205
  print(pipe(prompt_template)[0]['generated_text'])
206
  ```
207
+ <!-- README_GPTQ.md-use-from-python end -->
208
 
209
+ <!-- README_GPTQ.md-compatibility start -->
210
  ## Compatibility
211
 
212
+ The files provided are tested to work with AutoGPTQ, both via Transformers and using AutoGPTQ directly. They should also work with [Occ4m's GPTQ-for-LLaMa fork](https://github.com/0cc4m/KoboldAI).
213
+
214
+ [ExLlama](https://github.com/turboderp/exllama) is compatible with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
215
 
216
+ [Huggingface Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference) is compatible with all GPTQ models.
217
+ <!-- README_GPTQ.md-compatibility end -->
218
 
219
  <!-- footer start -->
220
  <!-- 200823 -->
 
239
 
240
  **Special thanks to**: Aemon Algiz.
241
 
242
+ **Patreon special mentions**: Russ Johnson, J, alfie_i, Alex, NimbleBox.ai, Chadd, Mandus, Nikolai Manek, Ken Nordquist, ya boyyy, Illia Dulskyi, Viktor Bowallius, vamX, Iucharbius, zynix, Magnesian, Clay Pascal, Pierre Kircher, Enrico Ros, Tony Hughes, Elle, Andrey, knownsqashed, Deep Realms, Jerry Meng, Lone Striker, Derek Yates, Pyrater, Mesiah Bishop, James Bentley, Femi Adebogun, Brandon Frisco, SuperWojo, Alps Aficionado, Michael Dempsey, Vitor Caleffi, Will Dee, Edmond Seymore, usrbinkat, LangChain4j, Kacper Wikieł, Luke Pendergrass, John Detwiler, theTransient, Nathan LeClaire, Tiffany J. Kim, biorpg, Eugene Pentland, Stanislav Ovsiannikov, Fred von Graf, terasurfer, Kalila, Dan Guido, Nitin Borwankar, 阿明, Ai Maven, John Villwock, Gabriel Puliatti, Stephen Murray, Asp the Wyvern, danny, Chris Smitley, ReadyPlayerEmma, S_X, Daniel P. Andersen, Olakabola, Jeffrey Morgan, Imad Khwaja, Caitlyn Gatomon, webtim, Alicia Loh, Trenton Dambrowitz, Swaroop Kallakuri, Erik Bjäreholt, Leonard Tan, Spiking Neurons AB, Luke @flexchar, Ajan Kanaga, Thomas Belote, Deo Leter, RoA, Willem Michiel, transmissions 11, subjectnull, Matthew Berman, Joseph William Delisle, David Ziegler, Michael Davis, Johann-Peter Hartmann, Talal Aujan, senxiiz, Artur Olbinski, Rainer Wilmers, Spencer Kim, Fen Risland, Cap'n Zoog, Rishabh Srivastava, Michael Levine, Geoffrey Montalvo, Sean Connelly, Alexandros Triantafyllidis, Pieter, Gabriel Tamborski, Sam, Subspace Studios, Junyu Yang, Pedro Madruga, Vadim, Cory Kujawski, K, Raven Klaugh, Randy H, Mano Prime, Sebastain Graf, Space Cruiser
243
 
244
 
245
  Thank you to all my generous patrons and donaters!