TheBloke commited on
Commit
d89af6d
1 Parent(s): 6764435

Update for Transformers GPTQ support

Browse files
README.md CHANGED
@@ -13,17 +13,20 @@ tags:
13
  ---
14
 
15
  <!-- header start -->
16
- <div style="width: 100%;">
17
- <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
 
18
  </div>
19
  <div style="display: flex; justify-content: space-between; width: 100%;">
20
  <div style="display: flex; flex-direction: column; align-items: flex-start;">
21
- <p><a href="https://discord.gg/theblokeai">Chat & support: my new Discord server</a></p>
22
  </div>
23
  <div style="display: flex; flex-direction: column; align-items: flex-end;">
24
- <p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
25
  </div>
26
  </div>
 
 
27
  <!-- header end -->
28
 
29
  # CalderaAI's 13B BlueMethod GPTQ
@@ -57,13 +60,13 @@ Each separate quant is in a different branch. See below for instructions on fet
57
 
58
  | Branch | Bits | Group Size | Act Order (desc_act) | File Size | ExLlama Compatible? | Made With | Description |
59
  | ------ | ---- | ---------- | -------------------- | --------- | ------------------- | --------- | ----------- |
60
- | main | 4 | 128 | False | 7.26 GB | True | AutoGPTQ | Most compatible option. Good inference speed in AutoGPTQ and GPTQ-for-LLaMa. Lower inference quality than other options. |
61
- | gptq-4bit-32g-actorder_True | 4 | 32 | True | 8.00 GB | True | AutoGPTQ | 4-bit, with Act Order and group size. 32g gives highest possible inference quality, with maximum VRAM usage. Poor AutoGPTQ CUDA speed. |
62
- | gptq-4bit-64g-actorder_True | 4 | 64 | True | 7.51 GB | True | AutoGPTQ | 4-bit, with Act Order and group size. 64g uses less VRAM than 32g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
63
- | gptq-4bit-128g-actorder_True | 4 | 128 | True | 7.26 GB | True | AutoGPTQ | 4-bit, with Act Order and group size. 128g uses even less VRAM, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
64
- | gptq-8bit-128g-actorder_True | 8 | 128 | True | 13.65 GB | False | AutoGPTQ | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. Poor AutoGPTQ CUDA speed. |
65
- | gptq-8bit-64g-actorder_True | 8 | 64 | True | 13.95 GB | False | AutoGPTQ | 8-bit, with group size 64g and Act Order for maximum inference quality. Poor AutoGPTQ CUDA speed. |
66
- | gptq-8bit-128g-actorder_False | 8 | 128 | False | 13.65 GB | False | AutoGPTQ | 8-bit, with group size 128g for higher inference quality and without Act Order to improve AutoGPTQ speed. |
67
  | gptq-8bit--1g-actorder_True | 8 | None | True | 13.36 GB | False | AutoGPTQ | 8-bit, with Act Order. No group size, to lower VRAM requirements and to improve AutoGPTQ speed. |
68
 
69
  ## How to download from branches
@@ -107,7 +110,7 @@ from transformers import AutoTokenizer, pipeline, logging
107
  from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
108
 
109
  model_name_or_path = "TheBloke/13B-BlueMethod-GPTQ"
110
- model_basename = "gptq_model-4bit-128g"
111
 
112
  use_triton = False
113
 
@@ -173,6 +176,7 @@ The files provided will work with AutoGPTQ (CUDA and Triton modes), GPTQ-for-LLa
173
  ExLlama works with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
174
 
175
  <!-- footer start -->
 
176
  ## Discord
177
 
178
  For further support, and discussions on these models and AI in general, join us at:
@@ -192,13 +196,15 @@ Donaters will get priority support on any and all AI/LLM/model questions and req
192
  * Patreon: https://patreon.com/TheBlokeAI
193
  * Ko-Fi: https://ko-fi.com/TheBlokeAI
194
 
195
- **Special thanks to**: Luke from CarbonQuill, Aemon Algiz.
196
 
197
- **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
198
 
199
 
200
  Thank you to all my generous patrons and donaters!
201
 
 
 
202
  <!-- footer end -->
203
 
204
  # Original model card: CalderaAI's 13B BlueMethod
 
13
  ---
14
 
15
  <!-- header start -->
16
+ <!-- 200823 -->
17
+ <div style="width: auto; margin-left: auto; margin-right: auto">
18
+ <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
19
  </div>
20
  <div style="display: flex; justify-content: space-between; width: 100%;">
21
  <div style="display: flex; flex-direction: column; align-items: flex-start;">
22
+ <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
23
  </div>
24
  <div style="display: flex; flex-direction: column; align-items: flex-end;">
25
+ <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>
26
  </div>
27
  </div>
28
+ <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>
29
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
30
  <!-- header end -->
31
 
32
  # CalderaAI's 13B BlueMethod GPTQ
 
60
 
61
  | Branch | Bits | Group Size | Act Order (desc_act) | File Size | ExLlama Compatible? | Made With | Description |
62
  | ------ | ---- | ---------- | -------------------- | --------- | ------------------- | --------- | ----------- |
63
+ | main | 4 | 128 | False | 7.26 GB | True | AutoGPTQ | Most compatible option. Good inference speed in AutoGPTQ and GPTQ-for-LLaMa. Lower inference quality than other options. |
64
+ | gptq-4bit-32g-actorder_True | 4 | 32 | True | 8.00 GB | True | AutoGPTQ | 4-bit, with Act Order and group size. 32g gives highest possible inference quality, with maximum VRAM usage. Poor AutoGPTQ CUDA speed. |
65
+ | gptq-4bit-64g-actorder_True | 4 | 64 | True | 7.51 GB | True | AutoGPTQ | 4-bit, with Act Order and group size. 64g uses less VRAM than 32g, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
66
+ | gptq-4bit-128g-actorder_True | 4 | 128 | True | 7.26 GB | True | AutoGPTQ | 4-bit, with Act Order and group size. 128g uses even less VRAM, but with slightly lower accuracy. Poor AutoGPTQ CUDA speed. |
67
+ | gptq-8bit-128g-actorder_True | 8 | 128 | True | 13.65 GB | False | AutoGPTQ | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. Poor AutoGPTQ CUDA speed. |
68
+ | gptq-8bit-64g-actorder_True | 8 | 64 | True | 13.95 GB | False | AutoGPTQ | 8-bit, with group size 64g and Act Order for maximum inference quality. Poor AutoGPTQ CUDA speed. |
69
+ | gptq-8bit-128g-actorder_False | 8 | 128 | False | 13.65 GB | False | AutoGPTQ | 8-bit, with group size 128g for higher inference quality and without Act Order to improve AutoGPTQ speed. |
70
  | gptq-8bit--1g-actorder_True | 8 | None | True | 13.36 GB | False | AutoGPTQ | 8-bit, with Act Order. No group size, to lower VRAM requirements and to improve AutoGPTQ speed. |
71
 
72
  ## How to download from branches
 
110
  from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
111
 
112
  model_name_or_path = "TheBloke/13B-BlueMethod-GPTQ"
113
+ model_basename = "model"
114
 
115
  use_triton = False
116
 
 
176
  ExLlama works with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
177
 
178
  <!-- footer start -->
179
+ <!-- 200823 -->
180
  ## Discord
181
 
182
  For further support, and discussions on these models and AI in general, join us at:
 
196
  * Patreon: https://patreon.com/TheBlokeAI
197
  * Ko-Fi: https://ko-fi.com/TheBlokeAI
198
 
199
+ **Special thanks to**: Aemon Algiz.
200
 
201
+ **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
202
 
203
 
204
  Thank you to all my generous patrons and donaters!
205
 
206
+ And thank you again to a16z for their generous grant.
207
+
208
  <!-- footer end -->
209
 
210
  # Original model card: CalderaAI's 13B BlueMethod
config.json CHANGED
@@ -1,24 +1,35 @@
1
  {
2
- "_name_or_path": "13B-BlueMethod",
3
- "architectures": [
4
- "LlamaForCausalLM"
5
- ],
6
- "bos_token_id": 1,
7
- "eos_token_id": 2,
8
- "hidden_act": "silu",
9
- "hidden_size": 5120,
10
- "initializer_range": 0.02,
11
- "intermediate_size": 13824,
12
- "max_position_embeddings": 2048,
13
- "max_seq_len": 4096,
14
- "model_type": "llama",
15
- "num_attention_heads": 40,
16
- "num_hidden_layers": 40,
17
- "pad_token_id": 0,
18
- "rms_norm_eps": 1e-06,
19
- "tie_word_embeddings": false,
20
- "torch_dtype": "float16",
21
- "transformers_version": "4.30.2",
22
- "use_cache": true,
23
- "vocab_size": 32000
 
 
 
 
 
 
 
 
 
 
 
24
  }
 
1
  {
2
+ "_name_or_path": "13B-BlueMethod",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 5120,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 13824,
12
+ "max_position_embeddings": 2048,
13
+ "max_seq_len": 4096,
14
+ "model_type": "llama",
15
+ "num_attention_heads": 40,
16
+ "num_hidden_layers": 40,
17
+ "pad_token_id": 0,
18
+ "rms_norm_eps": 1e-06,
19
+ "tie_word_embeddings": false,
20
+ "torch_dtype": "float16",
21
+ "transformers_version": "4.30.2",
22
+ "use_cache": true,
23
+ "vocab_size": 32000,
24
+ "quantization_config": {
25
+ "bits": 8,
26
+ "group_size": 128,
27
+ "damp_percent": 0.01,
28
+ "desc_act": false,
29
+ "sym": true,
30
+ "true_sequential": true,
31
+ "model_name_or_path": null,
32
+ "model_file_base_name": "model",
33
+ "quant_method": "gptq"
34
+ }
35
  }
gptq_model-8bit-128g.safetensors → model.safetensors RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:eff9b783d9c3630bc10f4e59ae99b0be9b6094d3d177076a0e7f8b855a0806e2
3
- size 13652898056
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17c60291ea740edc5a938cb6f110d53b94ffea38b495808369588ce684d917f4
3
+ size 13652898120
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
  }