TheBloke commited on
Commit
a0b2080
1 Parent(s): 74e55d7

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +28 -18
README.md CHANGED
@@ -1,10 +1,15 @@
1
  ---
 
2
  inference: false
3
  license: llama2
4
  model_creator: lmsys
5
- model_link: https://huggingface.co/lmsys/vicuna-13b-v1.5-16k
6
  model_name: Vicuna 13B v1.5 16K
7
  model_type: llama
 
 
 
 
 
8
  quantized_by: TheBloke
9
  ---
10
 
@@ -40,9 +45,9 @@ Multiple GPTQ parameter permutations are provided; see Provided Files below for
40
  <!-- repositories-available start -->
41
  ## Repositories available
42
 
 
43
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ)
44
  * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GGUF)
45
- * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference (deprecated)](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GGML)
46
  * [lmsys's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/lmsys/vicuna-13b-v1.5-16k)
47
  <!-- repositories-available end -->
48
 
@@ -56,6 +61,7 @@ A chat between a curious user and an artificial intelligence assistant. The assi
56
 
57
  <!-- prompt-template end -->
58
 
 
59
  <!-- README_GPTQ.md-provided-files start -->
60
  ## Provided files and GPTQ parameters
61
 
@@ -80,24 +86,24 @@ All recent GPTQ files are made with AutoGPTQ, and all files in non-main branches
80
 
81
  | Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
82
  | ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
83
- | [main](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/main) | 4 | 128 | No | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 7.26 GB | Yes | Most compatible option. Good inference speed in AutoGPTQ and GPTQ-for-LLaMa. Lower inference quality than other options. |
84
- | [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 8.00 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. Poor AutoGPTQ CUDA speed. |
85
- | [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 7.51 GB | Yes | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
86
- | [gptq-4bit-128g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-4bit-128g-actorder_True) | 4 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 7.26 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
87
- | [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 13.36 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements and to improve AutoGPTQ speed. |
88
  | [gptq-8bit-64g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-8bit-64g-actorder_True) | 8 | 64 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 13.95 GB | No | 8-bit, with group size 64g and Act Order for even higher inference quality. Poor AutoGPTQ CUDA speed. |
89
- | [gptq-8bit-32g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-8bit-32g-actorder_True) | 8 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 14.54 GB | No | 8-bit, with group size 32g and Act Order for maximum inference quality. Poor AutoGPTQ CUDA speed. |
90
- | [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 13.65 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. Poor AutoGPTQ CUDA speed. |
91
 
92
  <!-- README_GPTQ.md-provided-files end -->
93
 
94
  <!-- README_GPTQ.md-download-from-branches start -->
95
  ## How to download from branches
96
 
97
- - In text-generation-webui, you can add `:branch` to the end of the download name, eg `TheBloke/vicuna-13B-v1.5-16K-GPTQ:gptq-4bit-32g-actorder_True`
98
  - With Git, you can clone a branch with:
99
  ```
100
- git clone --single-branch --branch gptq-4bit-32g-actorder_True https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ
101
  ```
102
  - In Python Transformers code, the branch is the `revision` parameter; see below.
103
  <!-- README_GPTQ.md-download-from-branches end -->
@@ -110,7 +116,7 @@ It is strongly recommended to use the text-generation-webui one-click-installers
110
 
111
  1. Click the **Model tab**.
112
  2. Under **Download custom model or LoRA**, enter `TheBloke/vicuna-13B-v1.5-16K-GPTQ`.
113
- - To download from a specific branch, enter for example `TheBloke/vicuna-13B-v1.5-16K-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".
@@ -158,10 +164,10 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
158
 
159
  model_name_or_path = "TheBloke/vicuna-13B-v1.5-16K-GPTQ"
160
  # To use a different branch, change revision
161
- # For example: revision="gptq-4bit-32g-actorder_True"
162
  model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
163
- torch_dtype=torch.float16,
164
  device_map="auto",
 
165
  revision="main")
166
 
167
  tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
@@ -174,7 +180,7 @@ prompt_template=f'''A chat between a curious user and an artificial intelligence
174
  print("\n\n*** Generate:")
175
 
176
  input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
177
- output = model.generate(inputs=input_ids, temperature=0.7, max_new_tokens=512)
178
  print(tokenizer.decode(output[0]))
179
 
180
  # Inference can also be done using transformers' pipeline
@@ -185,9 +191,11 @@ pipe = pipeline(
185
  model=model,
186
  tokenizer=tokenizer,
187
  max_new_tokens=512,
 
188
  temperature=0.7,
189
  top_p=0.95,
190
- repetition_penalty=1.15
 
191
  )
192
 
193
  print(pipe(prompt_template)[0]['generated_text'])
@@ -212,10 +220,12 @@ For further support, and discussions on these models and AI in general, join us
212
 
213
  [TheBloke AI's Discord server](https://discord.gg/theblokeai)
214
 
215
- ## Thanks, and how to contribute.
216
 
217
  Thanks to the [chirper.ai](https://chirper.ai) team!
218
 
 
 
219
  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.
220
 
221
  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.
@@ -227,7 +237,7 @@ Donaters will get priority support on any and all AI/LLM/model questions and req
227
 
228
  **Special thanks to**: Aemon Algiz.
229
 
230
- **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
231
 
232
 
233
  Thank you to all my generous patrons and donaters!
 
1
  ---
2
+ base_model: https://huggingface.co/lmsys/vicuna-13b-v1.5-16k
3
  inference: false
4
  license: llama2
5
  model_creator: lmsys
 
6
  model_name: Vicuna 13B v1.5 16K
7
  model_type: llama
8
+ prompt_template: 'A chat between a curious user and an artificial intelligence assistant.
9
+ The assistant gives helpful, detailed, and polite answers to the user''s questions.
10
+ USER: {prompt} ASSISTANT:
11
+
12
+ '
13
  quantized_by: TheBloke
14
  ---
15
 
 
45
  <!-- repositories-available start -->
46
  ## Repositories available
47
 
48
+ * [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-AWQ)
49
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ)
50
  * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GGUF)
 
51
  * [lmsys's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/lmsys/vicuna-13b-v1.5-16k)
52
  <!-- repositories-available end -->
53
 
 
61
 
62
  <!-- prompt-template end -->
63
 
64
+
65
  <!-- README_GPTQ.md-provided-files start -->
66
  ## Provided files and GPTQ parameters
67
 
 
86
 
87
  | Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
88
  | ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
89
+ | [main](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/main) | 4 | 128 | No | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 7.26 GB | Yes | 4-bit, without Act Order and group size 128g. |
90
+ | [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 8.00 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. |
91
+ | [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 7.51 GB | Yes | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. |
92
+ | [gptq-4bit-128g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-4bit-128g-actorder_True) | 4 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 7.26 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. |
93
+ | [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 13.36 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements. |
94
  | [gptq-8bit-64g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-8bit-64g-actorder_True) | 8 | 64 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 13.95 GB | No | 8-bit, with group size 64g and Act Order for even higher inference quality. Poor AutoGPTQ CUDA speed. |
95
+ | [gptq-8bit-32g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-8bit-32g-actorder_True) | 8 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 14.54 GB | No | 8-bit, with group size 32g and Act Order for maximum inference quality. |
96
+ | [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 8192 | 13.65 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. |
97
 
98
  <!-- README_GPTQ.md-provided-files end -->
99
 
100
  <!-- README_GPTQ.md-download-from-branches start -->
101
  ## How to download from branches
102
 
103
+ - In text-generation-webui, you can add `:branch` to the end of the download name, eg `TheBloke/vicuna-13B-v1.5-16K-GPTQ:main`
104
  - With Git, you can clone a branch with:
105
  ```
106
+ git clone --single-branch --branch main https://huggingface.co/TheBloke/vicuna-13B-v1.5-16K-GPTQ
107
  ```
108
  - In Python Transformers code, the branch is the `revision` parameter; see below.
109
  <!-- README_GPTQ.md-download-from-branches end -->
 
116
 
117
  1. Click the **Model tab**.
118
  2. Under **Download custom model or LoRA**, enter `TheBloke/vicuna-13B-v1.5-16K-GPTQ`.
119
+ - To download from a specific branch, enter for example `TheBloke/vicuna-13B-v1.5-16K-GPTQ:main`
120
  - see Provided Files above for the list of branches for each option.
121
  3. Click **Download**.
122
  4. The model will start downloading. Once it's finished it will say "Done".
 
164
 
165
  model_name_or_path = "TheBloke/vicuna-13B-v1.5-16K-GPTQ"
166
  # To use a different branch, change revision
167
+ # For example: revision="main"
168
  model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
 
169
  device_map="auto",
170
+ trust_remote_code=False,
171
  revision="main")
172
 
173
  tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
 
180
  print("\n\n*** Generate:")
181
 
182
  input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
183
+ output = model.generate(inputs=input_ids, temperature=0.7, do_sample=True, top_p=0.95, top_k=40, max_new_tokens=512)
184
  print(tokenizer.decode(output[0]))
185
 
186
  # Inference can also be done using transformers' pipeline
 
191
  model=model,
192
  tokenizer=tokenizer,
193
  max_new_tokens=512,
194
+ do_sample=True,
195
  temperature=0.7,
196
  top_p=0.95,
197
+ top_k=40,
198
+ repetition_penalty=1.1
199
  )
200
 
201
  print(pipe(prompt_template)[0]['generated_text'])
 
220
 
221
  [TheBloke AI's Discord server](https://discord.gg/theblokeai)
222
 
223
+ ## Thanks, and how to contribute
224
 
225
  Thanks to the [chirper.ai](https://chirper.ai) team!
226
 
227
+ Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
228
+
229
  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.
230
 
231
  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.
 
237
 
238
  **Special thanks to**: Aemon Algiz.
239
 
240
+ **Patreon special mentions**: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov
241
 
242
 
243
  Thank you to all my generous patrons and donaters!