Update for Transformers GPTQ support
Browse files- README.md +19 -12
- config.json +48 -37
- gptq_model-4bit-128g.safetensors → model.safetensors +2 -2
- quantize_config.json +1 -1
README.md
CHANGED
@@ -40,23 +40,26 @@ extra_gated_prompt: >-
|
|
40 |
Please read the BigCode [OpenRAIL-M
|
41 |
license](https://huggingface.co/spaces/bigcode/bigcode-model-license-agreement)
|
42 |
agreement before accepting it.
|
43 |
-
|
44 |
extra_gated_fields:
|
45 |
I accept the above license agreement, and will use the Model complying with the set of use restrictions and sharing requirements: checkbox
|
46 |
---
|
47 |
|
48 |
<!-- header start -->
|
49 |
-
|
50 |
-
|
|
|
51 |
</div>
|
52 |
<div style="display: flex; justify-content: space-between; width: 100%;">
|
53 |
<div style="display: flex; flex-direction: column; align-items: flex-start;">
|
54 |
-
<p><a href="https://discord.gg/
|
55 |
</div>
|
56 |
<div style="display: flex; flex-direction: column; align-items: flex-end;">
|
57 |
-
<p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
|
58 |
</div>
|
59 |
</div>
|
|
|
|
|
60 |
<!-- header end -->
|
61 |
|
62 |
# OpenAccess AI Collective's Minotaur 15B GPTQ
|
@@ -111,7 +114,7 @@ from transformers import AutoTokenizer, pipeline, logging
|
|
111 |
from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
|
112 |
|
113 |
model_name_or_path = "TheBloke/minotaur-15B-GPTQ"
|
114 |
-
model_basename = "
|
115 |
|
116 |
use_triton = False
|
117 |
|
@@ -171,11 +174,12 @@ It was created with group_size 128 to increase inference accuracy, but without -
|
|
171 |
* Parameters: Groupsize = 128. Act Order / desc_act = False.
|
172 |
|
173 |
<!-- footer start -->
|
|
|
174 |
## Discord
|
175 |
|
176 |
For further support, and discussions on these models and AI in general, join us at:
|
177 |
|
178 |
-
[TheBloke AI's Discord server](https://discord.gg/
|
179 |
|
180 |
## Thanks, and how to contribute.
|
181 |
|
@@ -190,12 +194,15 @@ Donaters will get priority support on any and all AI/LLM/model questions and req
|
|
190 |
* Patreon: https://patreon.com/TheBlokeAI
|
191 |
* Ko-Fi: https://ko-fi.com/TheBlokeAI
|
192 |
|
193 |
-
**Special thanks to**:
|
|
|
|
|
194 |
|
195 |
-
**Patreon special mentions**: vamX, K, Jonathan Leane, Lone Striker, Sean Connelly, Chris McCloskey, WelcomeToTheClub, Nikolai Manek, John Detwiler, Kalila, David Flickinger, Fen Risland, subjectnull, Johann-Peter Hartmann, Talal Aujan, John Villwock, senxiiz, Khalefa Al-Ahmad, Kevin Schuppel, Alps Aficionado, Derek Yates, Mano Prime, Nathan LeClaire, biorpg, trip7s trip, Asp the Wyvern, chris gileta, Iucharbius , Artur Olbinski, Ai Maven, Joseph William Delisle, Luke Pendergrass, Illia Dulskyi, Eugene Pentland, Ajan Kanaga, Willem Michiel, Space Cruiser, Pyrater, Preetika Verma, Junyu Yang, Oscar Rangel, Spiking Neurons AB, Pierre Kircher, webtim, Cory Kujawski, terasurfer , Trenton Dambrowitz, Gabriel Puliatti, Imad Khwaja, Luke.
|
196 |
|
197 |
Thank you to all my generous patrons and donaters!
|
198 |
|
|
|
|
|
199 |
<!-- footer end -->
|
200 |
|
201 |
# Original model card: OpenAccess AI Collective's Minotaur 15B
|
@@ -282,10 +289,10 @@ Play with the instruction-tuned StarCoderPlus at [StarChat-Beta](https://hugging
|
|
282 |
|
283 |
## Model Summary
|
284 |
|
285 |
-
StarCoderPlus is a fine-tuned version of [StarCoderBase](https://huggingface.co/bigcode/starcoderbase) on 600B tokens from the English web dataset [RedefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb)
|
286 |
combined with [StarCoderData](https://huggingface.co/datasets/bigcode/starcoderdata) from [The Stack (v1.2)](https://huggingface.co/datasets/bigcode/the-stack) and a Wikipedia dataset.
|
287 |
It's a 15.5B parameter Language Model trained on English and 80+ programming languages. The model uses [Multi Query Attention](https://arxiv.org/abs/1911.02150),
|
288 |
-
[a context window of 8192 tokens](https://arxiv.org/abs/2205.14135), and was trained using the [Fill-in-the-Middle objective](https://arxiv.org/abs/2207.14255) on 1.6 trillion tokens.
|
289 |
|
290 |
- **Repository:** [bigcode/Megatron-LM](https://github.com/bigcode-project/Megatron-LM)
|
291 |
- **Project Website:** [bigcode-project.org](https://www.bigcode-project.org)
|
@@ -334,7 +341,7 @@ The training code dataset of the model was filtered for permissive licenses only
|
|
334 |
# Limitations
|
335 |
|
336 |
The model has been trained on a mixture of English text from the web and GitHub code. Therefore it might encounter limitations when working with non-English text, and can carry the stereotypes and biases commonly encountered online.
|
337 |
-
Additionally, the generated code should be used with caution as it may contain errors, inefficiencies, or potential vulnerabilities. For a more comprehensive understanding of the base model's code limitations, please refer to See [StarCoder paper](hhttps://arxiv.org/abs/2305.06161).
|
338 |
|
339 |
# Training
|
340 |
StarCoderPlus is a fine-tuned version on 600B English and code tokens of StarCoderBase, which was pre-trained on 1T code tokens. Below are the fine-tuning details:
|
|
|
40 |
Please read the BigCode [OpenRAIL-M
|
41 |
license](https://huggingface.co/spaces/bigcode/bigcode-model-license-agreement)
|
42 |
agreement before accepting it.
|
43 |
+
|
44 |
extra_gated_fields:
|
45 |
I accept the above license agreement, and will use the Model complying with the set of use restrictions and sharing requirements: checkbox
|
46 |
---
|
47 |
|
48 |
<!-- header start -->
|
49 |
+
<!-- 200823 -->
|
50 |
+
<div style="width: auto; margin-left: auto; margin-right: auto">
|
51 |
+
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
|
52 |
</div>
|
53 |
<div style="display: flex; justify-content: space-between; width: 100%;">
|
54 |
<div style="display: flex; flex-direction: column; align-items: flex-start;">
|
55 |
+
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
|
56 |
</div>
|
57 |
<div style="display: flex; flex-direction: column; align-items: flex-end;">
|
58 |
+
<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>
|
59 |
</div>
|
60 |
</div>
|
61 |
+
<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>
|
62 |
+
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
|
63 |
<!-- header end -->
|
64 |
|
65 |
# OpenAccess AI Collective's Minotaur 15B GPTQ
|
|
|
114 |
from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
|
115 |
|
116 |
model_name_or_path = "TheBloke/minotaur-15B-GPTQ"
|
117 |
+
model_basename = "model"
|
118 |
|
119 |
use_triton = False
|
120 |
|
|
|
174 |
* Parameters: Groupsize = 128. Act Order / desc_act = False.
|
175 |
|
176 |
<!-- footer start -->
|
177 |
+
<!-- 200823 -->
|
178 |
## Discord
|
179 |
|
180 |
For further support, and discussions on these models and AI in general, join us at:
|
181 |
|
182 |
+
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
|
183 |
|
184 |
## Thanks, and how to contribute.
|
185 |
|
|
|
194 |
* Patreon: https://patreon.com/TheBlokeAI
|
195 |
* Ko-Fi: https://ko-fi.com/TheBlokeAI
|
196 |
|
197 |
+
**Special thanks to**: Aemon Algiz.
|
198 |
+
|
199 |
+
**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
|
200 |
|
|
|
201 |
|
202 |
Thank you to all my generous patrons and donaters!
|
203 |
|
204 |
+
And thank you again to a16z for their generous grant.
|
205 |
+
|
206 |
<!-- footer end -->
|
207 |
|
208 |
# Original model card: OpenAccess AI Collective's Minotaur 15B
|
|
|
289 |
|
290 |
## Model Summary
|
291 |
|
292 |
+
StarCoderPlus is a fine-tuned version of [StarCoderBase](https://huggingface.co/bigcode/starcoderbase) on 600B tokens from the English web dataset [RedefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb)
|
293 |
combined with [StarCoderData](https://huggingface.co/datasets/bigcode/starcoderdata) from [The Stack (v1.2)](https://huggingface.co/datasets/bigcode/the-stack) and a Wikipedia dataset.
|
294 |
It's a 15.5B parameter Language Model trained on English and 80+ programming languages. The model uses [Multi Query Attention](https://arxiv.org/abs/1911.02150),
|
295 |
+
[a context window of 8192 tokens](https://arxiv.org/abs/2205.14135), and was trained using the [Fill-in-the-Middle objective](https://arxiv.org/abs/2207.14255) on 1.6 trillion tokens.
|
296 |
|
297 |
- **Repository:** [bigcode/Megatron-LM](https://github.com/bigcode-project/Megatron-LM)
|
298 |
- **Project Website:** [bigcode-project.org](https://www.bigcode-project.org)
|
|
|
341 |
# Limitations
|
342 |
|
343 |
The model has been trained on a mixture of English text from the web and GitHub code. Therefore it might encounter limitations when working with non-English text, and can carry the stereotypes and biases commonly encountered online.
|
344 |
+
Additionally, the generated code should be used with caution as it may contain errors, inefficiencies, or potential vulnerabilities. For a more comprehensive understanding of the base model's code limitations, please refer to See [StarCoder paper](hhttps://arxiv.org/abs/2305.06161).
|
345 |
|
346 |
# Training
|
347 |
StarCoderPlus is a fine-tuned version on 600B English and code tokens of StarCoderBase, which was pre-trained on 1T code tokens. Below are the fine-tuning details:
|
config.json
CHANGED
@@ -1,39 +1,50 @@
|
|
1 |
{
|
2 |
-
|
3 |
-
|
4 |
-
|
5 |
-
|
6 |
-
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
38 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
39 |
}
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "/fsx/bigcode/experiments/pretraining/conversions/starcoderplus/large-model",
|
3 |
+
"activation_function": "gelu",
|
4 |
+
"architectures": [
|
5 |
+
"GPTBigCodeForCausalLM"
|
6 |
+
],
|
7 |
+
"attention_softmax_in_fp32": true,
|
8 |
+
"multi_query": true,
|
9 |
+
"attn_pdrop": 0.1,
|
10 |
+
"bos_token_id": 0,
|
11 |
+
"embd_pdrop": 0.1,
|
12 |
+
"eos_token_id": 0,
|
13 |
+
"inference_runner": 0,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"layer_norm_epsilon": 1e-05,
|
16 |
+
"max_batch_size": null,
|
17 |
+
"max_sequence_length": null,
|
18 |
+
"model_type": "gpt_bigcode",
|
19 |
+
"n_embd": 6144,
|
20 |
+
"n_head": 48,
|
21 |
+
"n_inner": 24576,
|
22 |
+
"n_layer": 40,
|
23 |
+
"n_positions": 8192,
|
24 |
+
"pad_key_length": true,
|
25 |
+
"pre_allocate_kv_cache": false,
|
26 |
+
"resid_pdrop": 0.1,
|
27 |
+
"scale_attention_softmax_in_fp32": true,
|
28 |
+
"scale_attn_weights": true,
|
29 |
+
"summary_activation": null,
|
30 |
+
"summary_first_dropout": 0.1,
|
31 |
+
"summary_proj_to_labels": true,
|
32 |
+
"summary_type": "cls_index",
|
33 |
+
"summary_use_proj": true,
|
34 |
+
"torch_dtype": "float32",
|
35 |
+
"transformers_version": "4.28.1",
|
36 |
+
"use_cache": true,
|
37 |
+
"validate_runner_input": true,
|
38 |
+
"vocab_size": 49152,
|
39 |
+
"quantization_config": {
|
40 |
+
"bits": 4,
|
41 |
+
"group_size": 128,
|
42 |
+
"damp_percent": 0.01,
|
43 |
+
"desc_act": false,
|
44 |
+
"sym": true,
|
45 |
+
"true_sequential": true,
|
46 |
+
"model_name_or_path": null,
|
47 |
+
"model_file_base_name": "model",
|
48 |
+
"quant_method": "gptq"
|
49 |
+
}
|
50 |
}
|
gptq_model-4bit-128g.safetensors → model.safetensors
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5244916171978b2a27c3d20ab73e86bfc2bb0f149877adf20d7f9748463b2ec5
|
3 |
+
size 9198404376
|
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":
|
10 |
}
|
|
|
6 |
"sym": true,
|
7 |
"true_sequential": true,
|
8 |
"model_name_or_path": null,
|
9 |
+
"model_file_base_name": "model"
|
10 |
}
|