TheBloke commited on
Commit
ad21829
1 Parent(s): a907ab0

Update for Transformers GPTQ support

Browse files
README.md CHANGED
@@ -17,17 +17,20 @@ tags:
17
  ---
18
 
19
  <!-- header start -->
20
- <div style="width: 100%;">
21
- <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
 
22
  </div>
23
  <div style="display: flex; justify-content: space-between; width: 100%;">
24
  <div style="display: flex; flex-direction: column; align-items: flex-start;">
25
- <p><a href="https://discord.gg/theblokeai">Chat & support: my new Discord server</a></p>
26
  </div>
27
  <div style="display: flex; flex-direction: column; align-items: flex-end;">
28
- <p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
29
  </div>
30
  </div>
 
 
31
  <!-- header end -->
32
 
33
  # Vigogne 2 7B Instruct - GPTQ
@@ -65,13 +68,13 @@ Each separate quant is in a different branch. See below for instructions on fet
65
 
66
  | Branch | Bits | Group Size | Act Order (desc_act) | GPTQ Dataset | Size | ExLlama Compat? | Made With | Desc |
67
  | ------ | ---- | ---------- | -------------------- | ------------ | ---- | --------------- | --------- | ---- |
68
- | [main](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/main) | 4 | 128 | No | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 3.90 GB | Yes | AutoGPTQ | Most compatible option. Good inference speed in AutoGPTQ and GPTQ-for-LLaMa. Lower inference quality than other options. |
69
- | [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 4.28 GB | Yes | AutoGPTQ | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. Poor AutoGPTQ CUDA speed. |
70
- | [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 4.02 GB | Yes | AutoGPTQ | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
71
- | [gptq-4bit-128g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-4bit-128g-actorder_True) | 4 | 128 | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 3.90 GB | Yes | AutoGPTQ | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
72
- | [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 7.01 GB | No | AutoGPTQ | 8-bit, with Act Order. No group size, to lower VRAM requirements and to improve AutoGPTQ speed. |
73
- | [gptq-8bit-128g-actorder_False](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-8bit-128g-actorder_False) | 8 | 128 | No | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 7.16 GB | No | AutoGPTQ | 8-bit, with group size 128g for higher inference quality and without Act Order to improve AutoGPTQ speed. |
74
- | [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 7.16 GB | No | AutoGPTQ | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. Poor AutoGPTQ CUDA speed. |
75
  | [gptq-8bit-64g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-8bit-64g-actorder_True) | 8 | 64 | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 7.31 GB | No | AutoGPTQ | 8-bit, with group size 64g and Act Order for maximum inference quality. Poor AutoGPTQ CUDA speed. |
76
 
77
  ## How to download from branches
@@ -115,7 +118,7 @@ from transformers import AutoTokenizer, pipeline, logging
115
  from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
116
 
117
  model_name_or_path = "TheBloke/Vigogne-2-7B-Instruct-GPTQ"
118
- model_basename = "gptq_model-4bit-128g"
119
 
120
  use_triton = False
121
 
@@ -182,6 +185,7 @@ The files provided will work with AutoGPTQ (CUDA and Triton modes), GPTQ-for-LLa
182
  ExLlama works with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
183
 
184
  <!-- footer start -->
 
185
  ## Discord
186
 
187
  For further support, and discussions on these models and AI in general, join us at:
@@ -201,13 +205,15 @@ Donaters will get priority support on any and all AI/LLM/model questions and req
201
  * Patreon: https://patreon.com/TheBlokeAI
202
  * Ko-Fi: https://ko-fi.com/TheBlokeAI
203
 
204
- **Special thanks to**: Luke from CarbonQuill, Aemon Algiz.
205
 
206
- **Patreon special mentions**: Slarti, Chadd, John Detwiler, Pieter, zynix, K, Mano Prime, ReadyPlayerEmma, Ai Maven, Leonard Tan, Edmond Seymore, Joseph William Delisle, Luke @flexchar, Fred von Graf, Viktor Bowallius, Rishabh Srivastava, Nikolai Manek, Matthew Berman, Johann-Peter Hartmann, ya boyyy, Greatston Gnanesh, Femi Adebogun, Talal Aujan, Jonathan Leane, terasurfer, David Flickinger, William Sang, Ajan Kanaga, Vadim, Artur Olbinski, Raven Klaugh, Michael Levine, Oscar Rangel, Randy H, Cory Kujawski, RoA, Dave, Alex, Alexandros Triantafyllidis, Fen Risland, Eugene Pentland, vamX, Elle, Nathan LeClaire, Khalefa Al-Ahmad, Rainer Wilmers, subjectnull, Junyu Yang, Daniel P. Andersen, SuperWojo, LangChain4j, Mandus, Kalila, Illia Dulskyi, Trenton Dambrowitz, Asp the Wyvern, Derek Yates, Jeffrey Morgan, Deep Realms, Imad Khwaja, Pyrater, Preetika Verma, biorpg, Gabriel Tamborski, Stephen Murray, Spiking Neurons AB, Iucharbius, Chris Smitley, Willem Michiel, Luke Pendergrass, Sebastain Graf, senxiiz, Will Dee, Space Cruiser, Karl Bernard, Clay Pascal, Lone Striker, transmissions 11, webtim, WelcomeToTheClub, Sam, theTransient, Pierre Kircher, chris gileta, John Villwock, Sean Connelly, Willian Hasse
207
 
208
 
209
  Thank you to all my generous patrons and donaters!
210
 
 
 
211
  <!-- footer end -->
212
 
213
  # Original model card: bofenghuang's Vigogne 2 7B Instruct
@@ -267,7 +273,7 @@ You can also infer this model by using the following Google Colab Notebook.
267
  Explique le théorème central limite à un enfant de six ans.
268
 
269
  ***Vigogne-2-7B-Instruct:***
270
-
271
  Le théorème central limite est un concept mathématique qui dit que si vous prenez beaucoup de mesures de la même chose, la moyenne de ces mesures sera proche de la moyenne de toutes les mesures. C'est comme si vous prenait des photos d'un objet et que la moyenne de toutes les photos était proche de la vraie forme de l'objet.
272
 
273
  <details><summary>Other models</summary>
@@ -294,7 +300,7 @@ Le théorème central limite est un concept mathématique qui explique comment l
294
  Je suis actuellement orienté vers l'est. Ensuite je vais tourner à droite. De quel côté serai-je orienté ?
295
 
296
  ***Vigogne-2-7B-Instruct:***
297
-
298
  Vous serez orienté vers le sud.
299
 
300
  <details><summary>Other models</summary>
@@ -318,7 +324,7 @@ Vous serez orienté vers le nord.
318
  Parmi les aliments suivants, lequel est le plus susceptible de provoquer une intoxication alimentaire ? A, poisson cru, B, poulet cuit, C, légumes cuits, D, pain grillé.
319
 
320
  ***Vigogne-2-7B-Instruct:***
321
-
322
  A, poisson cru.
323
 
324
  <details><summary>Other models</summary>
@@ -341,7 +347,7 @@ Le plus susceptible de provoquer une intoxication alimentaire est le poisson cru
341
  Écrivez une fonction qui prend une liste de chaînes de caractères et renvoie une liste sans doublons.
342
 
343
  ***Vigogne-2-7B-Instruct:***
344
-
345
  ```
346
  def remove_duplicates(lst):
347
  return list(set(lst))
 
17
  ---
18
 
19
  <!-- header start -->
20
+ <!-- 200823 -->
21
+ <div style="width: auto; margin-left: auto; margin-right: auto">
22
+ <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
23
  </div>
24
  <div style="display: flex; justify-content: space-between; width: 100%;">
25
  <div style="display: flex; flex-direction: column; align-items: flex-start;">
26
+ <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
27
  </div>
28
  <div style="display: flex; flex-direction: column; align-items: flex-end;">
29
+ <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>
30
  </div>
31
  </div>
32
+ <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>
33
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
34
  <!-- header end -->
35
 
36
  # Vigogne 2 7B Instruct - GPTQ
 
68
 
69
  | Branch | Bits | Group Size | Act Order (desc_act) | GPTQ Dataset | Size | ExLlama Compat? | Made With | Desc |
70
  | ------ | ---- | ---------- | -------------------- | ------------ | ---- | --------------- | --------- | ---- |
71
+ | [main](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/main) | 4 | 128 | No | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 3.90 GB | Yes | AutoGPTQ | Most compatible option. Good inference speed in AutoGPTQ and GPTQ-for-LLaMa. Lower inference quality than other options. |
72
+ | [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 4.28 GB | Yes | AutoGPTQ | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. Poor AutoGPTQ CUDA speed. |
73
+ | [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 4.02 GB | Yes | AutoGPTQ | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
74
+ | [gptq-4bit-128g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-4bit-128g-actorder_True) | 4 | 128 | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 3.90 GB | Yes | AutoGPTQ | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
75
+ | [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 7.01 GB | No | AutoGPTQ | 8-bit, with Act Order. No group size, to lower VRAM requirements and to improve AutoGPTQ speed. |
76
+ | [gptq-8bit-128g-actorder_False](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-8bit-128g-actorder_False) | 8 | 128 | No | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 7.16 GB | No | AutoGPTQ | 8-bit, with group size 128g for higher inference quality and without Act Order to improve AutoGPTQ speed. |
77
+ | [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 7.16 GB | No | AutoGPTQ | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. Poor AutoGPTQ CUDA speed. |
78
  | [gptq-8bit-64g-actorder_True](https://huggingface.co/TheBloke/Vigogne-2-7B-Instruct-GPTQ/tree/gptq-8bit-64g-actorder_True) | 8 | 64 | Yes | [French news](https://huggingface.co/datasets/gustavecortal/diverse_french_news) | 7.31 GB | No | AutoGPTQ | 8-bit, with group size 64g and Act Order for maximum inference quality. Poor AutoGPTQ CUDA speed. |
79
 
80
  ## How to download from branches
 
118
  from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
119
 
120
  model_name_or_path = "TheBloke/Vigogne-2-7B-Instruct-GPTQ"
121
+ model_basename = "model"
122
 
123
  use_triton = False
124
 
 
185
  ExLlama works with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
186
 
187
  <!-- footer start -->
188
+ <!-- 200823 -->
189
  ## Discord
190
 
191
  For further support, and discussions on these models and AI in general, join us at:
 
205
  * Patreon: https://patreon.com/TheBlokeAI
206
  * Ko-Fi: https://ko-fi.com/TheBlokeAI
207
 
208
+ **Special thanks to**: Aemon Algiz.
209
 
210
+ **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
211
 
212
 
213
  Thank you to all my generous patrons and donaters!
214
 
215
+ And thank you again to a16z for their generous grant.
216
+
217
  <!-- footer end -->
218
 
219
  # Original model card: bofenghuang's Vigogne 2 7B Instruct
 
273
  Explique le théorème central limite à un enfant de six ans.
274
 
275
  ***Vigogne-2-7B-Instruct:***
276
+
277
  Le théorème central limite est un concept mathématique qui dit que si vous prenez beaucoup de mesures de la même chose, la moyenne de ces mesures sera proche de la moyenne de toutes les mesures. C'est comme si vous prenait des photos d'un objet et que la moyenne de toutes les photos était proche de la vraie forme de l'objet.
278
 
279
  <details><summary>Other models</summary>
 
300
  Je suis actuellement orienté vers l'est. Ensuite je vais tourner à droite. De quel côté serai-je orienté ?
301
 
302
  ***Vigogne-2-7B-Instruct:***
303
+
304
  Vous serez orienté vers le sud.
305
 
306
  <details><summary>Other models</summary>
 
324
  Parmi les aliments suivants, lequel est le plus susceptible de provoquer une intoxication alimentaire ? A, poisson cru, B, poulet cuit, C, légumes cuits, D, pain grillé.
325
 
326
  ***Vigogne-2-7B-Instruct:***
327
+
328
  A, poisson cru.
329
 
330
  <details><summary>Other models</summary>
 
347
  Écrivez une fonction qui prend une liste de chaînes de caractères et renvoie une liste sans doublons.
348
 
349
  ***Vigogne-2-7B-Instruct:***
350
+
351
  ```
352
  def remove_duplicates(lst):
353
  return list(set(lst))
config.json CHANGED
@@ -1,27 +1,38 @@
1
  {
2
- "_name_or_path": "meta-llama/Llama-2-7b-hf",
3
- "architectures": [
4
- "LlamaForCausalLM"
5
- ],
6
- "bos_token_id": 1,
7
- "eos_token_id": 2,
8
- "hidden_act": "silu",
9
- "hidden_size": 4096,
10
- "initializer_range": 0.02,
11
- "intermediate_size": 11008,
12
- "max_length": 4096,
13
- "max_position_embeddings": 2048,
14
- "model_type": "llama",
15
- "num_attention_heads": 32,
16
- "num_hidden_layers": 32,
17
- "num_key_value_heads": 32,
18
- "pad_token_id": 0,
19
- "pretraining_tp": 1,
20
- "rms_norm_eps": 1e-05,
21
- "rope_scaling": null,
22
- "tie_word_embeddings": false,
23
- "torch_dtype": "float16",
24
- "transformers_version": "4.32.0.dev0",
25
- "use_cache": true,
26
- "vocab_size": 32000
 
 
 
 
 
 
 
 
 
 
 
27
  }
 
1
  {
2
+ "_name_or_path": "meta-llama/Llama-2-7b-hf",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 4096,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 11008,
12
+ "max_length": 4096,
13
+ "max_position_embeddings": 2048,
14
+ "model_type": "llama",
15
+ "num_attention_heads": 32,
16
+ "num_hidden_layers": 32,
17
+ "num_key_value_heads": 32,
18
+ "pad_token_id": 0,
19
+ "pretraining_tp": 1,
20
+ "rms_norm_eps": 1e-05,
21
+ "rope_scaling": null,
22
+ "tie_word_embeddings": false,
23
+ "torch_dtype": "float16",
24
+ "transformers_version": "4.32.0.dev0",
25
+ "use_cache": true,
26
+ "vocab_size": 32000,
27
+ "quantization_config": {
28
+ "bits": 8,
29
+ "group_size": 128,
30
+ "damp_percent": 0.1,
31
+ "desc_act": true,
32
+ "sym": true,
33
+ "true_sequential": true,
34
+ "model_name_or_path": null,
35
+ "model_file_base_name": "model",
36
+ "quant_method": "gptq"
37
+ }
38
  }
gptq_model-8bit-128g.safetensors → model.safetensors RENAMED
File without changes
quantize_config.json CHANGED
@@ -6,5 +6,5 @@
6
  "sym": true,
7
  "true_sequential": true,
8
  "model_name_or_path": null,
9
- "model_file_base_name": null
10
  }
 
6
  "sym": true,
7
  "true_sequential": true,
8
  "model_name_or_path": null,
9
+ "model_file_base_name": "model"
10
  }