TheBloke commited on
Commit
14553fb
1 Parent(s): b25010c

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +85 -19
README.md CHANGED
@@ -1,4 +1,5 @@
1
  ---
 
2
  datasets:
3
  - flozi00/conversations
4
  inference: false
@@ -7,9 +8,13 @@ language:
7
  - de
8
  license: llama2
9
  model_creator: Florian Zimmermeister
10
- model_link: https://huggingface.co/flozi00/Llama-2-13b-german-assistant-v4
11
  model_name: Llama 2 13B German Assistant v4
12
  model_type: llama
 
 
 
 
 
13
  quantized_by: TheBloke
14
  ---
15
 
@@ -34,23 +39,25 @@ quantized_by: TheBloke
34
  - Model creator: [Florian Zimmermeister](https://huggingface.co/flozi00)
35
  - Original model: [Llama 2 13B German Assistant v4](https://huggingface.co/flozi00/Llama-2-13b-german-assistant-v4)
36
 
 
37
  ## Description
38
 
39
  This repo contains GGUF format model files for [Florian Zimmermeister's Llama 2 13B German Assistant v4](https://huggingface.co/flozi00/Llama-2-13b-german-assistant-v4).
40
 
 
41
  <!-- README_GGUF.md-about-gguf start -->
42
  ### About GGUF
43
 
44
- GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp.
45
 
46
- The key benefit of GGUF is that it is a extensible, future-proof format which stores more information about the model as metadata. It also includes significantly improved tokenization code, including for the first time full support for special tokens. This should improve performance, especially with models that use new special tokens and implement custom prompt templates.
47
 
48
- Here are a list of clients and libraries that are known to support GGUF:
49
- * [llama.cpp](https://github.com/ggerganov/llama.cpp).
50
- * [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions.
51
- * [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with full GPU accel across multiple platforms and GPU architectures. Especially good for story telling.
52
- * [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI with GPU acceleration on both Windows (NVidia and AMD), and macOS.
53
  * [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
 
54
  * [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
55
  * [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
56
  * [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
@@ -59,6 +66,7 @@ Here are a list of clients and libraries that are known to support GGUF:
59
  <!-- repositories-available start -->
60
  ## Repositories available
61
 
 
62
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Llama-2-13B-German-Assistant-v4-GPTQ)
63
  * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Llama-2-13B-German-Assistant-v4-GGUF)
64
  * [Florian Zimmermeister's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/flozi00/Llama-2-13b-german-assistant-v4)
@@ -74,12 +82,14 @@ Here are a list of clients and libraries that are known to support GGUF:
74
  ```
75
 
76
  <!-- prompt-template end -->
 
 
77
  <!-- compatibility_gguf start -->
78
  ## Compatibility
79
 
80
- These quantised GGUF files are compatible with llama.cpp from August 21st 2023 onwards, as of commit [6381d4e110bd0ec02843a60bbeb8b6fc37a9ace9](https://github.com/ggerganov/llama.cpp/commit/6381d4e110bd0ec02843a60bbeb8b6fc37a9ace9)
81
 
82
- They are now also compatible with many third party UIs and libraries - please see the list at the top of the README.
83
 
84
  ## Explanation of quantisation methods
85
  <details>
@@ -120,21 +130,75 @@ Refer to the Provided Files table below to see what files use which methods, and
120
 
121
  <!-- README_GGUF.md-provided-files end -->
122
 
123
- <!-- README_GGUF.md-how-to-run start -->
124
- ## Example `llama.cpp` command
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
125
 
126
- Make sure you are using `llama.cpp` from commit [6381d4e110bd0ec02843a60bbeb8b6fc37a9ace9](https://github.com/ggerganov/llama.cpp/commit/6381d4e110bd0ec02843a60bbeb8b6fc37a9ace9) or later.
 
127
 
128
- For compatibility with older versions of llama.cpp, or for any third-party libraries or clients that haven't yet updated for GGUF, please use GGML files instead.
129
 
 
 
130
  ```
131
- ./main -t 10 -ngl 32 -m llama-2-13b-german-assistant-v4.q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "### User: {prompt}\n### Assistant:"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
132
  ```
133
- Change `-t 10` to the number of physical CPU cores you have. For example if your system has 8 cores/16 threads, use `-t 8`. If offloading all layers to GPU, set `-t 1`.
134
 
135
  Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
136
 
137
- Change `-c 4096` to the desired sequence length for this model. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
138
 
139
  If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
140
 
@@ -191,10 +255,12 @@ For further support, and discussions on these models and AI in general, join us
191
 
192
  [TheBloke AI's Discord server](https://discord.gg/theblokeai)
193
 
194
- ## Thanks, and how to contribute.
195
 
196
  Thanks to the [chirper.ai](https://chirper.ai) team!
197
 
 
 
198
  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.
199
 
200
  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.
@@ -206,7 +272,7 @@ Donaters will get priority support on any and all AI/LLM/model questions and req
206
 
207
  **Special thanks to**: Aemon Algiz.
208
 
209
- **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
210
 
211
 
212
  Thank you to all my generous patrons and donaters!
 
1
  ---
2
+ base_model: https://huggingface.co/flozi00/Llama-2-13b-german-assistant-v4
3
  datasets:
4
  - flozi00/conversations
5
  inference: false
 
8
  - de
9
  license: llama2
10
  model_creator: Florian Zimmermeister
 
11
  model_name: Llama 2 13B German Assistant v4
12
  model_type: llama
13
+ prompt_template: '### User: {prompt}
14
+
15
+ ### Assistant:
16
+
17
+ '
18
  quantized_by: TheBloke
19
  ---
20
 
 
39
  - Model creator: [Florian Zimmermeister](https://huggingface.co/flozi00)
40
  - Original model: [Llama 2 13B German Assistant v4](https://huggingface.co/flozi00/Llama-2-13b-german-assistant-v4)
41
 
42
+ <!-- description start -->
43
  ## Description
44
 
45
  This repo contains GGUF format model files for [Florian Zimmermeister's Llama 2 13B German Assistant v4](https://huggingface.co/flozi00/Llama-2-13b-german-assistant-v4).
46
 
47
+ <!-- description end -->
48
  <!-- README_GGUF.md-about-gguf start -->
49
  ### About GGUF
50
 
51
+ GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp. GGUF offers numerous advantages over GGML, such as better tokenisation, and support for special tokens. It is also supports metadata, and is designed to be extensible.
52
 
53
+ Here is an incomplate list of clients and libraries that are known to support GGUF:
54
 
55
+ * [llama.cpp](https://github.com/ggerganov/llama.cpp). The source project for GGUF. Offers a CLI and a server option.
56
+ * [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
57
+ * [KoboldCpp](https://github.com/LostRuins/koboldcpp), a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
58
+ * [LM Studio](https://lmstudio.ai/), an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
 
59
  * [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with many interesting and unique features, including a full model library for easy model selection.
60
+ * [Faraday.dev](https://faraday.dev/), an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
61
  * [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
62
  * [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
63
  * [candle](https://github.com/huggingface/candle), a Rust ML framework with a focus on performance, including GPU support, and ease of use.
 
66
  <!-- repositories-available start -->
67
  ## Repositories available
68
 
69
+ * [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/Llama-2-13B-German-Assistant-v4-AWQ)
70
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Llama-2-13B-German-Assistant-v4-GPTQ)
71
  * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Llama-2-13B-German-Assistant-v4-GGUF)
72
  * [Florian Zimmermeister's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/flozi00/Llama-2-13b-german-assistant-v4)
 
82
  ```
83
 
84
  <!-- prompt-template end -->
85
+
86
+
87
  <!-- compatibility_gguf start -->
88
  ## Compatibility
89
 
90
+ These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit [d0cee0d36d5be95a0d9088b674dbb27354107221](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221)
91
 
92
+ They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
93
 
94
  ## Explanation of quantisation methods
95
  <details>
 
130
 
131
  <!-- README_GGUF.md-provided-files end -->
132
 
133
+ <!-- README_GGUF.md-how-to-download start -->
134
+ ## How to download GGUF files
135
+
136
+ **Note for manual downloaders:** You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
137
+
138
+ The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
139
+ - LM Studio
140
+ - LoLLMS Web UI
141
+ - Faraday.dev
142
+
143
+ ### In `text-generation-webui`
144
+
145
+ Under Download Model, you can enter the model repo: TheBloke/Llama-2-13B-German-Assistant-v4-GGUF and below it, a specific filename to download, such as: llama-2-13b-german-assistant-v4.q4_K_M.gguf.
146
+
147
+ Then click Download.
148
+
149
+ ### On the command line, including multiple files at once
150
+
151
+ I recommend using the `huggingface-hub` Python library:
152
+
153
+ ```shell
154
+ pip3 install huggingface-hub>=0.17.1
155
+ ```
156
+
157
+ Then you can download any individual model file to the current directory, at high speed, with a command like this:
158
+
159
+ ```shell
160
+ huggingface-cli download TheBloke/Llama-2-13B-German-Assistant-v4-GGUF llama-2-13b-german-assistant-v4.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
161
+ ```
162
 
163
+ <details>
164
+ <summary>More advanced huggingface-cli download usage</summary>
165
 
166
+ You can also download multiple files at once with a pattern:
167
 
168
+ ```shell
169
+ huggingface-cli download TheBloke/Llama-2-13B-German-Assistant-v4-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
170
  ```
171
+
172
+ For more documentation on downloading with `huggingface-cli`, please see: [HF -> Hub Python Library -> Download files -> Download from the CLI](https://huggingface.co/docs/huggingface_hub/guides/download#download-from-the-cli).
173
+
174
+ To accelerate downloads on fast connections (1Gbit/s or higher), install `hf_transfer`:
175
+
176
+ ```shell
177
+ pip3 install hf_transfer
178
+ ```
179
+
180
+ And set environment variable `HF_HUB_ENABLE_HF_TRANSFER` to `1`:
181
+
182
+ ```shell
183
+ HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Llama-2-13B-German-Assistant-v4-GGUF llama-2-13b-german-assistant-v4.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
184
+ ```
185
+
186
+ Windows CLI users: Use `set HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1` before running the download command.
187
+ </details>
188
+ <!-- README_GGUF.md-how-to-download end -->
189
+
190
+ <!-- README_GGUF.md-how-to-run start -->
191
+ ## Example `llama.cpp` command
192
+
193
+ Make sure you are using `llama.cpp` from commit [d0cee0d36d5be95a0d9088b674dbb27354107221](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
194
+
195
+ ```shell
196
+ ./main -ngl 32 -m llama-2-13b-german-assistant-v4.q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "### User: {prompt}\n### Assistant:"
197
  ```
 
198
 
199
  Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
200
 
201
+ Change `-c 4096` to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
202
 
203
  If you want to have a chat-style conversation, replace the `-p <PROMPT>` argument with `-i -ins`
204
 
 
255
 
256
  [TheBloke AI's Discord server](https://discord.gg/theblokeai)
257
 
258
+ ## Thanks, and how to contribute
259
 
260
  Thanks to the [chirper.ai](https://chirper.ai) team!
261
 
262
+ Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
263
+
264
  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.
265
 
266
  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.
 
272
 
273
  **Special thanks to**: Aemon Algiz.
274
 
275
+ **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
276
 
277
 
278
  Thank you to all my generous patrons and donaters!