TheBloke commited on
Commit
3fb78f4
1 Parent(s): 4105c76

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +27 -12
README.md CHANGED
@@ -43,6 +43,13 @@ tags:
43
 
44
  This repo contains GGML format model files for [Voicelab's Trurl 2 13B](https://huggingface.co/Voicelab/trurl-2-13b).
45
 
 
 
 
 
 
 
 
46
  GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/ggerganov/llama.cpp) and libraries and UIs which support this format, such as:
47
  * [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most popular web UI. Supports NVidia CUDA GPU acceleration.
48
  * [KoboldCpp](https://github.com/LostRuins/koboldcpp), a powerful GGML web UI with GPU acceleration on all platforms (CUDA and OpenCL). Especially good for story telling.
@@ -54,7 +61,8 @@ GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/gger
54
  ## Repositories available
55
 
56
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Trurl-2-13B-GPTQ)
57
- * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/Trurl-2-13B-GGML)
 
58
  * [Voicelab's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/Voicelab/trurl-2-13b)
59
 
60
  ## Prompt template: Llama-2-Chat
@@ -64,14 +72,19 @@ GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/gger
64
  You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.
65
  <</SYS>>
66
  {prompt}[/INST]
 
67
  ```
68
 
69
  <!-- compatibility_ggml start -->
70
  ## Compatibility
71
 
72
- These quantised GGML files are compatible with llama.cpp as of June 6th, commit `2d43387`.
 
 
73
 
74
- They should also be compatible with all UIs, libraries and utilities which use GGML.
 
 
75
 
76
  ## Explanation of the new k-quant methods
77
  <details>
@@ -94,17 +107,17 @@ Refer to the Provided Files table below to see what files use which methods, and
94
  | Name | Quant method | Bits | Size | Max RAM required | Use case |
95
  | ---- | ---- | ---- | ---- | ---- | ----- |
96
  | [trurl-2-13b.ggmlv3.q2_K.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q2_K.bin) | q2_K | 2 | 5.74 GB| 8.24 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.vw and feed_forward.w2 tensors, GGML_TYPE_Q2_K for the other tensors. |
97
- | [trurl-2-13b.ggmlv3.q3_K_L.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q3_K_L.bin) | q3_K_L | 3 | 7.14 GB| 9.64 GB | New k-quant method. Uses GGML_TYPE_Q5_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
98
- | [trurl-2-13b.ggmlv3.q3_K_M.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q3_K_M.bin) | q3_K_M | 3 | 6.53 GB| 9.03 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
99
  | [trurl-2-13b.ggmlv3.q3_K_S.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q3_K_S.bin) | q3_K_S | 3 | 5.87 GB| 8.37 GB | New k-quant method. Uses GGML_TYPE_Q3_K for all tensors |
 
 
100
  | [trurl-2-13b.ggmlv3.q4_0.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q4_0.bin) | q4_0 | 4 | 7.32 GB| 9.82 GB | Original quant method, 4-bit. |
101
- | [trurl-2-13b.ggmlv3.q4_1.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q4_1.bin) | q4_1 | 4 | 8.14 GB| 10.64 GB | Original quant method, 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
102
- | [trurl-2-13b.ggmlv3.q4_K_M.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q4_K_M.bin) | q4_K_M | 4 | 8.06 GB| 10.56 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q4_K |
103
  | [trurl-2-13b.ggmlv3.q4_K_S.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q4_K_S.bin) | q4_K_S | 4 | 7.56 GB| 10.06 GB | New k-quant method. Uses GGML_TYPE_Q4_K for all tensors |
 
 
104
  | [trurl-2-13b.ggmlv3.q5_0.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q5_0.bin) | q5_0 | 5 | 8.95 GB| 11.45 GB | Original quant method, 5-bit. Higher accuracy, higher resource usage and slower inference. |
105
- | [trurl-2-13b.ggmlv3.q5_1.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q5_1.bin) | q5_1 | 5 | 9.76 GB| 12.26 GB | Original quant method, 5-bit. Even higher accuracy, resource usage and slower inference. |
106
- | [trurl-2-13b.ggmlv3.q5_K_M.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q5_K_M.bin) | q5_K_M | 5 | 9.40 GB| 11.90 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q5_K |
107
  | [trurl-2-13b.ggmlv3.q5_K_S.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q5_K_S.bin) | q5_K_S | 5 | 9.14 GB| 11.64 GB | New k-quant method. Uses GGML_TYPE_Q5_K for all tensors |
 
 
108
  | [trurl-2-13b.ggmlv3.q6_K.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q6_K.bin) | q6_K | 6 | 10.83 GB| 13.33 GB | New k-quant method. Uses GGML_TYPE_Q8_K for all tensors - 6-bit quantization |
109
  | [trurl-2-13b.ggmlv3.q8_0.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q8_0.bin) | q8_0 | 8 | 13.83 GB| 16.33 GB | Original quant method, 8-bit. Almost indistinguishable from float16. High resource use and slow. Not recommended for most users. |
110
 
@@ -112,10 +125,12 @@ Refer to the Provided Files table below to see what files use which methods, and
112
 
113
  ## How to run in `llama.cpp`
114
 
115
- I use the following command line; adjust for your tastes and needs:
 
 
116
 
117
  ```
118
- ./main -t 10 -ngl 32 -m trurl-2-13b.ggmlv3.q4_K_M.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "### Instruction: Write a story about llamas\n### Response:"
119
  ```
120
  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`.
121
 
@@ -154,7 +169,7 @@ Donaters will get priority support on any and all AI/LLM/model questions and req
154
 
155
  **Special thanks to**: Aemon Algiz.
156
 
157
- **Patreon special mentions**: Sam, theTransient, Jonathan Leane, Steven Wood, webtim, Johann-Peter Hartmann, Geoffrey Montalvo, Gabriel Tamborski, Willem Michiel, John Villwock, Derek Yates, Mesiah Bishop, Eugene Pentland, Pieter, Chadd, Stephen Murray, Daniel P. Andersen, terasurfer, Brandon Frisco, Thomas Belote, Sid, Nathan LeClaire, Magnesian, Alps Aficionado, Stanislav Ovsiannikov, Alex, Joseph William Delisle, Nikolai Manek, Michael Davis, Junyu Yang, K, J, Spencer Kim, Stefan Sabev, Olusegun Samson, transmissions 11, Michael Levine, Cory Kujawski, Rainer Wilmers, zynix, Kalila, Luke @flexchar, Ajan Kanaga, Mandus, vamX, Ai Maven, Mano Prime, Matthew Berman, subjectnull, Vitor Caleffi, Clay Pascal, biorpg, alfie_i, 阿明, Jeffrey Morgan, ya boyyy, Raymond Fosdick, knownsqashed, Olakabola, Leonard Tan, ReadyPlayerEmma, Enrico Ros, Dave, Talal Aujan, Illia Dulskyi, Sean Connelly, senxiiz, Artur Olbinski, Elle, Raven Klaugh, Fen Risland, Deep Realms, Imad Khwaja, Fred von Graf, Will Dee, usrbinkat, SuperWojo, Alexandros Triantafyllidis, Swaroop Kallakuri, Dan Guido, John Detwiler, Pedro Madruga, Iucharbius, Viktor Bowallius, Asp the Wyvern, Edmond Seymore, Trenton Dambrowitz, Space Cruiser, Spiking Neurons AB, Pyrater, LangChain4j, Tony Hughes, Kacper Wikieł, Rishabh Srivastava, David Ziegler, Luke Pendergrass, Andrey, Gabriel Puliatti, Lone Striker, Sebastain Graf, Pierre Kircher, Randy H, NimbleBox.ai, Vadim, danny, Deo Leter
158
 
159
 
160
  Thank you to all my generous patrons and donaters!
 
43
 
44
  This repo contains GGML format model files for [Voicelab's Trurl 2 13B](https://huggingface.co/Voicelab/trurl-2-13b).
45
 
46
+ ### Important note regarding GGML files.
47
+
48
+ The GGML format has now been superseded by GGUF. As of August 21st 2023, [llama.cpp](https://github.com/ggerganov/llama.cpp) no longer supports GGML models. Third party clients and libraries are expected to still support it for a time, but many may also drop support.
49
+
50
+ Please use the GGUF models instead.
51
+ ### About GGML
52
+
53
  GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/ggerganov/llama.cpp) and libraries and UIs which support this format, such as:
54
  * [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most popular web UI. Supports NVidia CUDA GPU acceleration.
55
  * [KoboldCpp](https://github.com/LostRuins/koboldcpp), a powerful GGML web UI with GPU acceleration on all platforms (CUDA and OpenCL). Especially good for story telling.
 
61
  ## Repositories available
62
 
63
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Trurl-2-13B-GPTQ)
64
+ * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Trurl-2-13B-GGUF)
65
+ * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference (deprecated)](https://huggingface.co/TheBloke/Trurl-2-13B-GGML)
66
  * [Voicelab's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/Voicelab/trurl-2-13b)
67
 
68
  ## Prompt template: Llama-2-Chat
 
72
  You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.
73
  <</SYS>>
74
  {prompt}[/INST]
75
+
76
  ```
77
 
78
  <!-- compatibility_ggml start -->
79
  ## Compatibility
80
 
81
+ These quantised GGML files are compatible with llama.cpp between June 6th (commit `2d43387`) and August 21st 2023.
82
+
83
+ For support with latest llama.cpp, please use GGUF files instead.
84
 
85
+ The final llama.cpp commit with support for GGML was: [dadbed99e65252d79f81101a392d0d6497b86caa](https://github.com/ggerganov/llama.cpp/commit/dadbed99e65252d79f81101a392d0d6497b86caa)
86
+
87
+ As of August 23rd 2023 they are still compatible with all UIs, libraries and utilities which use GGML. This may change in the future.
88
 
89
  ## Explanation of the new k-quant methods
90
  <details>
 
107
  | Name | Quant method | Bits | Size | Max RAM required | Use case |
108
  | ---- | ---- | ---- | ---- | ---- | ----- |
109
  | [trurl-2-13b.ggmlv3.q2_K.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q2_K.bin) | q2_K | 2 | 5.74 GB| 8.24 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.vw and feed_forward.w2 tensors, GGML_TYPE_Q2_K for the other tensors. |
 
 
110
  | [trurl-2-13b.ggmlv3.q3_K_S.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q3_K_S.bin) | q3_K_S | 3 | 5.87 GB| 8.37 GB | New k-quant method. Uses GGML_TYPE_Q3_K for all tensors |
111
+ | [trurl-2-13b.ggmlv3.q3_K_M.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q3_K_M.bin) | q3_K_M | 3 | 6.53 GB| 9.03 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
112
+ | [trurl-2-13b.ggmlv3.q3_K_L.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q3_K_L.bin) | q3_K_L | 3 | 7.14 GB| 9.64 GB | New k-quant method. Uses GGML_TYPE_Q5_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
113
  | [trurl-2-13b.ggmlv3.q4_0.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q4_0.bin) | q4_0 | 4 | 7.32 GB| 9.82 GB | Original quant method, 4-bit. |
 
 
114
  | [trurl-2-13b.ggmlv3.q4_K_S.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q4_K_S.bin) | q4_K_S | 4 | 7.56 GB| 10.06 GB | New k-quant method. Uses GGML_TYPE_Q4_K for all tensors |
115
+ | [trurl-2-13b.ggmlv3.q4_K_M.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q4_K_M.bin) | q4_K_M | 4 | 8.06 GB| 10.56 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q4_K |
116
+ | [trurl-2-13b.ggmlv3.q4_1.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q4_1.bin) | q4_1 | 4 | 8.14 GB| 10.64 GB | Original quant method, 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
117
  | [trurl-2-13b.ggmlv3.q5_0.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q5_0.bin) | q5_0 | 5 | 8.95 GB| 11.45 GB | Original quant method, 5-bit. Higher accuracy, higher resource usage and slower inference. |
 
 
118
  | [trurl-2-13b.ggmlv3.q5_K_S.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q5_K_S.bin) | q5_K_S | 5 | 9.14 GB| 11.64 GB | New k-quant method. Uses GGML_TYPE_Q5_K for all tensors |
119
+ | [trurl-2-13b.ggmlv3.q5_K_M.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q5_K_M.bin) | q5_K_M | 5 | 9.40 GB| 11.90 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q5_K |
120
+ | [trurl-2-13b.ggmlv3.q5_1.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q5_1.bin) | q5_1 | 5 | 9.76 GB| 12.26 GB | Original quant method, 5-bit. Even higher accuracy, resource usage and slower inference. |
121
  | [trurl-2-13b.ggmlv3.q6_K.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q6_K.bin) | q6_K | 6 | 10.83 GB| 13.33 GB | New k-quant method. Uses GGML_TYPE_Q8_K for all tensors - 6-bit quantization |
122
  | [trurl-2-13b.ggmlv3.q8_0.bin](https://huggingface.co/TheBloke/Trurl-2-13B-GGML/blob/main/trurl-2-13b.ggmlv3.q8_0.bin) | q8_0 | 8 | 13.83 GB| 16.33 GB | Original quant method, 8-bit. Almost indistinguishable from float16. High resource use and slow. Not recommended for most users. |
123
 
 
125
 
126
  ## How to run in `llama.cpp`
127
 
128
+ Make sure you are using `llama.cpp` from commit [dadbed99e65252d79f81101a392d0d6497b86caa](https://github.com/ggerganov/llama.cpp/commit/dadbed99e65252d79f81101a392d0d6497b86caa) or earlier.
129
+
130
+ For compatibility with latest llama.cpp, please use GGUF files instead.
131
 
132
  ```
133
+ ./main -t 10 -ngl 32 -m trurl-2-13b.ggmlv3.q4_K_M.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "[INST] <<SYS>>\nYou are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.\n<</SYS>>\n{prompt}[/INST]"
134
  ```
135
  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`.
136
 
 
169
 
170
  **Special thanks to**: Aemon Algiz.
171
 
172
+ **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
173
 
174
 
175
  Thank you to all my generous patrons and donaters!