TheBloke commited on
Commit
741af1e
1 Parent(s): 7191759

Update for Transformers GPTQ support

Browse files
README.md CHANGED
@@ -10,17 +10,20 @@ tags:
10
  ---
11
 
12
  <!-- header start -->
13
- <div style="width: 100%;">
14
- <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
 
15
  </div>
16
  <div style="display: flex; justify-content: space-between; width: 100%;">
17
  <div style="display: flex; flex-direction: column; align-items: flex-start;">
18
- <p><a href="https://discord.gg/theblokeai">Chat & support: my new Discord server</a></p>
19
  </div>
20
  <div style="display: flex; flex-direction: column; align-items: flex-end;">
21
- <p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
22
  </div>
23
  </div>
 
 
24
  <!-- header end -->
25
 
26
  # Mikael110's Llama2 13B Guanaco QLoRA GPTQ
@@ -52,13 +55,13 @@ Each separate quant is in a different branch. See below for instructions on fet
52
 
53
  | Branch | Bits | Group Size | Act Order (desc_act) | File Size | ExLlama Compatible? | Made With | Description |
54
  | ------ | ---- | ---------- | -------------------- | --------- | ------------------- | --------- | ----------- |
55
- | 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. |
56
- | 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. |
57
- | 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. |
58
- | 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. |
59
- | 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. |
60
- | 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. |
61
- | 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. |
62
  | 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. |
63
 
64
  ## How to download from branches
@@ -102,7 +105,7 @@ from transformers import AutoTokenizer, pipeline, logging
102
  from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
103
 
104
  model_name_or_path = "TheBloke/llama-2-13B-Guanaco-QLoRA-GPTQ"
105
- model_basename = "gptq_model-4bit-128g"
106
 
107
  use_triton = False
108
 
@@ -165,6 +168,7 @@ The files provided will work with AutoGPTQ (CUDA and Triton modes), GPTQ-for-LLa
165
  ExLlama works with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
166
 
167
  <!-- footer start -->
 
168
  ## Discord
169
 
170
  For further support, and discussions on these models and AI in general, join us at:
@@ -184,13 +188,15 @@ Donaters will get priority support on any and all AI/LLM/model questions and req
184
  * Patreon: https://patreon.com/TheBlokeAI
185
  * Ko-Fi: https://ko-fi.com/TheBlokeAI
186
 
187
- **Special thanks to**: Luke from CarbonQuill, Aemon Algiz.
188
 
189
- **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
190
 
191
 
192
  Thank you to all my generous patrons and donaters!
193
 
 
 
194
  <!-- footer end -->
195
 
196
  # Original model card: Mikael110's Llama2 13B Guanaco QLoRA
 
10
  ---
11
 
12
  <!-- header start -->
13
+ <!-- 200823 -->
14
+ <div style="width: auto; margin-left: auto; margin-right: auto">
15
+ <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
16
  </div>
17
  <div style="display: flex; justify-content: space-between; width: 100%;">
18
  <div style="display: flex; flex-direction: column; align-items: flex-start;">
19
+ <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
20
  </div>
21
  <div style="display: flex; flex-direction: column; align-items: flex-end;">
22
+ <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>
23
  </div>
24
  </div>
25
+ <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>
26
+ <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
27
  <!-- header end -->
28
 
29
  # Mikael110's Llama2 13B Guanaco QLoRA GPTQ
 
55
 
56
  | Branch | Bits | Group Size | Act Order (desc_act) | File Size | ExLlama Compatible? | Made With | Description |
57
  | ------ | ---- | ---------- | -------------------- | --------- | ------------------- | --------- | ----------- |
58
+ | 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. |
59
+ | 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. |
60
+ | 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. |
61
+ | 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. |
62
+ | 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. |
63
+ | 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. |
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
 
67
  ## How to download from branches
 
105
  from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
106
 
107
  model_name_or_path = "TheBloke/llama-2-13B-Guanaco-QLoRA-GPTQ"
108
+ model_basename = "model"
109
 
110
  use_triton = False
111
 
 
168
  ExLlama works with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
169
 
170
  <!-- footer start -->
171
+ <!-- 200823 -->
172
  ## Discord
173
 
174
  For further support, and discussions on these models and AI in general, join us at:
 
188
  * Patreon: https://patreon.com/TheBlokeAI
189
  * Ko-Fi: https://ko-fi.com/TheBlokeAI
190
 
191
+ **Special thanks to**: Aemon Algiz.
192
 
193
+ **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
194
 
195
 
196
  Thank you to all my generous patrons and donaters!
197
 
198
+ And thank you again to a16z for their generous grant.
199
+
200
  <!-- footer end -->
201
 
202
  # Original model card: Mikael110's Llama2 13B Guanaco QLoRA
config.json CHANGED
@@ -1,26 +1,37 @@
1
  {
2
- "_name_or_path": "/root/llama2-13b-hf/",
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
- "model_type": "llama",
14
- "num_attention_heads": 40,
15
- "num_hidden_layers": 40,
16
- "num_key_value_heads": 40,
17
- "pad_token_id": 0,
18
- "pretraining_tp": 1,
19
- "rms_norm_eps": 1e-05,
20
- "rope_scaling": null,
21
- "tie_word_embeddings": false,
22
- "torch_dtype": "float16",
23
- "transformers_version": "4.32.0.dev0",
24
- "use_cache": true,
25
- "vocab_size": 32000
 
 
 
 
 
 
 
 
 
 
 
26
  }
 
1
  {
2
+ "_name_or_path": "/root/llama2-13b-hf/",
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
+ "model_type": "llama",
14
+ "num_attention_heads": 40,
15
+ "num_hidden_layers": 40,
16
+ "num_key_value_heads": 40,
17
+ "pad_token_id": 0,
18
+ "pretraining_tp": 1,
19
+ "rms_norm_eps": 1e-05,
20
+ "rope_scaling": null,
21
+ "tie_word_embeddings": false,
22
+ "torch_dtype": "float16",
23
+ "transformers_version": "4.32.0.dev0",
24
+ "use_cache": true,
25
+ "vocab_size": 32000,
26
+ "quantization_config": {
27
+ "bits": 4,
28
+ "group_size": 128,
29
+ "damp_percent": 0.01,
30
+ "desc_act": false,
31
+ "sym": true,
32
+ "true_sequential": true,
33
+ "model_name_or_path": null,
34
+ "model_file_base_name": "model",
35
+ "quant_method": "gptq"
36
+ }
37
  }
gptq_model-4bit-128g.safetensors → model.safetensors RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:49670c08473faccd5c9be3aad8025b042c1040a3757c9b905c27e76f7ee6a092
3
- size 7259449424
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc5253f4cb6de1ee6d3e59be857d269ce4096b1f156dd1b4bb4d8645390a5de5
3
+ size 7259449480
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
  }