TheBloke commited on
Commit
efe4117
1 Parent(s): 34586d0

Initial GPTQ model commit

Browse files
Files changed (1) hide show
  1. README.md +73 -82
README.md CHANGED
@@ -1,14 +1,16 @@
1
  ---
2
  inference: false
 
 
3
  license: llama2
4
  model_creator: Meta
5
- model_link: https://ai.meta.com/resources/models-and-libraries/llama-downloads
6
  model_name: CodeLlama 13B
7
  model_type: llama
 
8
  quantized_by: TheBloke
9
  tags:
10
  - llama-2
11
- - codellama
12
  ---
13
 
14
  <!-- header start -->
@@ -30,11 +32,11 @@ tags:
30
 
31
  # CodeLlama 13B - GPTQ
32
  - Model creator: [Meta](https://huggingface.co/meta-llama)
33
- - Original model: [CodeLlama 13B](https://ai.meta.com/resources/models-and-libraries/llama-downloads)
34
 
35
  ## Description
36
 
37
- This repo contains GPTQ model files for [Meta's CodeLlama 13B](https://ai.meta.com/resources/models-and-libraries/llama-downloads).
38
 
39
  Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.
40
 
@@ -43,7 +45,7 @@ Multiple GPTQ parameter permutations are provided; see Provided Files below for
43
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/CodeLlama-13B-GPTQ)
44
  * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/CodeLlama-13B-GGUF)
45
  * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference (deprecated)](https://huggingface.co/TheBloke/CodeLlama-13B-GGML)
46
- * [Meta's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/TheBloke/CodeLlama-13B-fp16)
47
 
48
  ## Prompt template: TBC
49
 
@@ -214,7 +216,7 @@ Donaters will get priority support on any and all AI/LLM/model questions and req
214
 
215
  **Special thanks to**: Aemon Algiz.
216
 
217
- **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
218
 
219
 
220
  Thank you to all my generous patrons and donaters!
@@ -225,123 +227,112 @@ And thank you again to a16z for their generous grant.
225
 
226
  # Original model card: Meta's CodeLlama 13B
227
 
 
 
228
 
229
- <!-- header start -->
230
- <!-- 200823 -->
231
- <div style="width: auto; margin-left: auto; margin-right: auto">
232
- <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
233
- </div>
234
- <div style="display: flex; justify-content: space-between; width: 100%;">
235
- <div style="display: flex; flex-direction: column; align-items: flex-start;">
236
- <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
237
- </div>
238
- <div style="display: flex; flex-direction: column; align-items: flex-end;">
239
- <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>
240
- </div>
241
- </div>
242
- <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>
243
- <hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
244
- <!-- header end -->
245
-
246
- # CodeLlama 13B fp16
247
- - Model creator: [Meta](https://ai.meta.com/llama/)
248
-
249
- ## Description
250
-
251
- This is Transformers/HF format fp16 weights for CodeLlama 13B. It is the result of downloading CodeLlama 13B from [Meta](https://ai.meta.com/blog/code-llama-large-language-model-coding/) and converting to HF using `convert_llama_weights_to_hf.py`.
252
-
253
- Quantisations will be coming shortly.
254
-
255
- Please note that due to a change in the RoPE Theta value, for correct results you must load these FP16 models with `trust_remote_code=True`
256
-
257
- Credit to @emozilla for creating the necessary modelling code to achieve this!
258
-
259
- ## Prompt template: TBC
260
-
261
-
262
- <!-- footer start -->
263
- <!-- 200823 -->
264
- ## Discord
265
 
266
- For further support, and discussions on these models and AI in general, join us at:
267
-
268
- [TheBloke AI's Discord server](https://discord.gg/theblokeai)
269
-
270
- ## Thanks, and how to contribute.
271
 
272
- Thanks to the [chirper.ai](https://chirper.ai) team!
273
 
274
- I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
275
 
276
- If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
 
 
277
 
278
- Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
279
 
280
- * Patreon: https://patreon.com/TheBlokeAI
281
- * Ko-Fi: https://ko-fi.com/TheBlokeAI
 
 
282
 
283
- **Special thanks to**: Aemon Algiz.
284
 
285
- **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
 
 
 
286
 
 
287
 
288
- Thank you to all my generous patrons and donaters!
 
 
 
 
 
 
289
 
290
- And thank you again to a16z for their generous grant.
 
 
 
 
 
 
 
 
 
 
 
 
291
 
292
- <!-- footer end -->
293
 
294
- # Original model card
 
295
 
296
- # Code Llama
297
 
298
- ## **Model Details**
299
 
300
- **Model Developers** Meta AI
 
 
301
 
302
- **Variations** Code Llama comes in three model sizes, and three variants:
303
- 1) Code Llama: our base models designed for general code synthesis and understanding
304
- 2) Code Llama - Python: designed specifically for Python
305
- 3) Code Llama - Instruct: for instruction following and safer deployment
306
-
307
  All variants are available in sizes of 7B, 13B and 34B parameters.
308
 
 
 
309
  **Input** Models input text only.
310
 
311
- **Output** Models output text only.
312
 
313
- **Model Architecture** Code Llama and its variants are autoregressive language models using optimized transformer architectures. Code Llama 7B and 13B additionally support infilling text generation. All models were fine-tuned with up to 16K tokens, and support up to 100K tokens at inference time.
314
 
315
  **Model Dates** Code Llama and its variants have been trained between January 2023 and July 2023.
316
 
317
- **Status** This is a static model trained on an offline dataset. Future versions of Code Llama - Instruct will be released as we improve model safety with community feedback.
318
 
319
- **Licence** A custom commercial license is available at: [https://ai.meta.com/resources/models-and-libraries/llama-downloads/](https://ai.meta.com/resources/models-and-libraries/llama-downloads/).
320
 
321
  **Research Paper** More information can be found in the paper "[Code Llama: Open Foundation Models for Code](https://ai.meta.com/research/publications/code-llama-open-foundation-models-for-code/)".
322
 
323
- **Where to send comments** Instructions on how to provide feedback or comments on the model can be found in the model [README](README.md), or by opening an issue in the GitHub repository ([https://github.com/facebookresearch/codellama/](https://github.com/facebookresearch/codellama/)).
324
-
325
- ## **Intended Use**
326
  **Intended Use Cases** Code Llama and its variants is intended for commercial and research use in English and relevant programming languages. The base model Code Llama can be adapted for a variety of code synthesis and understanding tasks, Code Llama - Python is designed specifically to handle the Python programming language, and Code Llama - Instruct is intended to be safer to use for code assistant and generation applications.
327
 
328
  **Out-of-Scope Uses** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in languages other than English. Use in any other way that is prohibited by the Acceptable Use Policy and Licensing Agreement for Code Llama and its variants.
329
 
330
- ## **Hardware and Software**
331
- **Training Factors**
332
- We used custom training libraries. The training and fine-tuning of the released models have been performed Meta’s Research Super Cluster.
333
 
334
  **Carbon Footprint** In aggregate, training all 9 Code Llama models required 400K GPU hours of computation on hardware of type A100-80GB (TDP of 350-400W). Estimated total emissions were 65.3 tCO2eq, 100% of which were offset by Meta’s sustainability program.
335
 
336
- **Training data**
 
337
  All experiments reported here and the released models have been trained and fine-tuned using the same data as Llama 2 with different weights (see Section 2 and Table 1 in the [research paper](https://ai.meta.com/research/publications/code-llama-open-foundation-models-for-code/) for details).
338
- Code Llama - Instruct uses additional instruction fine-tuning data.
339
 
340
- **Evaluation Results**
 
341
  See evaluations for the main models and detailed ablations in Section 3 and safety evaluations in Section 4 of the research paper.
342
 
343
- ## **Ethical Considerations and Limitations**
 
 
344
  Code Llama and its variants are a new technology that carries risks with use. Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, Code Llama’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate or objectionable responses to user prompts. Therefore, before deploying any applications of Code Llama, developers should perform safety testing and tuning tailored to their specific applications of the model.
345
 
346
  Please see the Responsible Use Guide available available at [https://ai.meta.com/llama/responsible-user-guide](https://ai.meta.com/llama/responsible-user-guide).
347
-
 
1
  ---
2
  inference: false
3
+ language:
4
+ - code
5
  license: llama2
6
  model_creator: Meta
7
+ model_link: https://huggingface.co/codellama/CodeLlama-13b-hf
8
  model_name: CodeLlama 13B
9
  model_type: llama
10
+ pipeline_tag: text-generation
11
  quantized_by: TheBloke
12
  tags:
13
  - llama-2
 
14
  ---
15
 
16
  <!-- header start -->
 
32
 
33
  # CodeLlama 13B - GPTQ
34
  - Model creator: [Meta](https://huggingface.co/meta-llama)
35
+ - Original model: [CodeLlama 13B](https://huggingface.co/codellama/CodeLlama-13b-hf)
36
 
37
  ## Description
38
 
39
+ This repo contains GPTQ model files for [Meta's CodeLlama 13B](https://huggingface.co/codellama/CodeLlama-13b-hf).
40
 
41
  Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.
42
 
 
45
  * [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/CodeLlama-13B-GPTQ)
46
  * [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/CodeLlama-13B-GGUF)
47
  * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference (deprecated)](https://huggingface.co/TheBloke/CodeLlama-13B-GGML)
48
+ * [Meta's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/codellama/CodeLlama-13b-hf)
49
 
50
  ## Prompt template: TBC
51
 
 
216
 
217
  **Special thanks to**: Aemon Algiz.
218
 
219
+ **Patreon special mentions**: Kacper Wikieł, knownsqashed, Leonard Tan, Asp the Wyvern, Daniel P. Andersen, Luke Pendergrass, Stanislav Ovsiannikov, RoA, Dave, Ai Maven, Kalila, Will Dee, Imad Khwaja, Nitin Borwankar, Joseph William Delisle, Tony Hughes, Cory Kujawski, Rishabh Srivastava, Russ Johnson, Stephen Murray, Lone Striker, Johann-Peter Hartmann, Elle, J, Deep Realms, SuperWojo, Raven Klaugh, Sebastain Graf, ReadyPlayerEmma, Alps Aficionado, Mano Prime, Derek Yates, Gabriel Puliatti, Mesiah Bishop, Magnesian, Sean Connelly, biorpg, Iucharbius, Olakabola, Fen Risland, Space Cruiser, theTransient, Illia Dulskyi, Thomas Belote, Spencer Kim, Pieter, John Detwiler, Fred von Graf, Michael Davis, Swaroop Kallakuri, subjectnull, Clay Pascal, Subspace Studios, Chris Smitley, Enrico Ros, usrbinkat, Steven Wood, alfie_i, David Ziegler, Willem Michiel, Matthew Berman, Andrey, Pyrater, Jeffrey Morgan, vamX, LangChain4j, Luke @flexchar, Trenton Dambrowitz, Pierre Kircher, Alex, Sam, James Bentley, Edmond Seymore, Eugene Pentland, Pedro Madruga, Rainer Wilmers, Dan Guido, Nathan LeClaire, Spiking Neurons AB, Talal Aujan, zynix, Artur Olbinski, Michael Levine, 阿明, K, John Villwock, Nikolai Manek, Femi Adebogun, senxiiz, Deo Leter, NimbleBox.ai, Viktor Bowallius, Geoffrey Montalvo, Mandus, Ajan Kanaga, ya boyyy, Jonathan Leane, webtim, Brandon Frisco, danny, Alexandros Triantafyllidis, Gabriel Tamborski, Randy H, terasurfer, Vadim, Junyu Yang, Vitor Caleffi, Chadd, transmissions 11
220
 
221
 
222
  Thank you to all my generous patrons and donaters!
 
227
 
228
  # Original model card: Meta's CodeLlama 13B
229
 
230
+ # **Code Llama**
231
+ Code Llama is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 34 billion parameters. This is the repository for the base 13B version in the Hugging Face Transformers format. This model is designed for general code synthesis and understanding. Links to other models can be found in the index at the bottom.
232
 
233
+ | | Base Model | Python | Instruct |
234
+ | --- | ----------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- |
235
+ | 7B | [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) | [codellama/CodeLlama-7b-Python-hf](https://huggingface.co/codellama/CodeLlama-7b-Python-hf) | [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) |
236
+ | 13B | [codellama/CodeLlama-13b-hf](https://huggingface.co/codellama/CodeLlama-13b-hf) | [codellama/CodeLlama-13b-Python-hf](https://huggingface.co/codellama/CodeLlama-13b-Python-hf) | [codellama/CodeLlama-13b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-13b-Instruct-hf) |
237
+ | 34B | [codellama/CodeLlama-34b-hf](https://huggingface.co/codellama/CodeLlama-34b-hf) | [codellama/CodeLlama-34b-Python-hf](https://huggingface.co/codellama/CodeLlama-34b-Python-hf) | [codellama/CodeLlama-34b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-34b-Instruct-hf) |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
238
 
 
 
 
 
 
239
 
240
+ ## Model Use
241
 
242
+ To use this model, please make sure to install transformers from `main` until the next version is released:
243
 
244
+ ```bash
245
+ pip install git+https://github.com/huggingface/transformers.git@main accelerate
246
+ ```
247
 
248
+ Model capabilities:
249
 
250
+ - [x] Code completion.
251
+ - [x] Infilling.
252
+ - [ ] Instructions / chat.
253
+ - [ ] Python specialist.
254
 
 
255
 
256
+ ```python
257
+ from transformers import AutoTokenizer
258
+ import transformers
259
+ import torch
260
 
261
+ model = "codellama/CodeLlama-13b-hf"
262
 
263
+ tokenizer = AutoTokenizer.from_pretrained(model)
264
+ pipeline = transformers.pipeline(
265
+ "text-generation",
266
+ model=model,
267
+ torch_dtype=torch.float16,
268
+ device_map="auto",
269
+ )
270
 
271
+ sequences = pipeline(
272
+ 'import socket\n\ndef ping_exponential_backoff(host: str):',
273
+ do_sample=True,
274
+ top_k=10,
275
+ temperature=0.1,
276
+ top_p=0.95,
277
+ num_return_sequences=1,
278
+ eos_token_id=tokenizer.eos_token_id,
279
+ max_length=200,
280
+ )
281
+ for seq in sequences:
282
+ print(f"Result: {seq['generated_text']}")
283
+ ```
284
 
 
285
 
286
+ ## Model Details
287
+ *Note: Use of this model is governed by the Meta license. Meta developed and publicly released the Code Llama family of large language models (LLMs).
288
 
289
+ **Model Developers** Meta
290
 
291
+ **Variations** Code Llama comes in three model sizes, and three variants:
292
 
293
+ * Code Llama: base models designed for general code synthesis and understanding
294
+ * Code Llama - Python: designed specifically for Python
295
+ * Code Llama - Instruct: for instruction following and safer deployment
296
 
 
 
 
 
 
297
  All variants are available in sizes of 7B, 13B and 34B parameters.
298
 
299
+ **This repository contains the base version of the 13B parameters model.**
300
+
301
  **Input** Models input text only.
302
 
303
+ **Output** Models generate text only.
304
 
305
+ **Model Architecture** Code Llama is an auto-regressive language model that uses an optimized transformer architecture.
306
 
307
  **Model Dates** Code Llama and its variants have been trained between January 2023 and July 2023.
308
 
309
+ **Status** This is a static model trained on an offline dataset. Future versions of Code Llama - Instruct will be released as we improve model safety with community feedback.
310
 
311
+ **License** A custom commercial license is available at: [https://ai.meta.com/resources/models-and-libraries/llama-downloads/](https://ai.meta.com/resources/models-and-libraries/llama-downloads/)
312
 
313
  **Research Paper** More information can be found in the paper "[Code Llama: Open Foundation Models for Code](https://ai.meta.com/research/publications/code-llama-open-foundation-models-for-code/)".
314
 
315
+ ## Intended Use
 
 
316
  **Intended Use Cases** Code Llama and its variants is intended for commercial and research use in English and relevant programming languages. The base model Code Llama can be adapted for a variety of code synthesis and understanding tasks, Code Llama - Python is designed specifically to handle the Python programming language, and Code Llama - Instruct is intended to be safer to use for code assistant and generation applications.
317
 
318
  **Out-of-Scope Uses** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in languages other than English. Use in any other way that is prohibited by the Acceptable Use Policy and Licensing Agreement for Code Llama and its variants.
319
 
320
+ ## Hardware and Software
321
+ **Training Factors** We used custom training libraries. The training and fine-tuning of the released models have been performed Meta’s Research Super Cluster.
 
322
 
323
  **Carbon Footprint** In aggregate, training all 9 Code Llama models required 400K GPU hours of computation on hardware of type A100-80GB (TDP of 350-400W). Estimated total emissions were 65.3 tCO2eq, 100% of which were offset by Meta’s sustainability program.
324
 
325
+ ## Training Data
326
+
327
  All experiments reported here and the released models have been trained and fine-tuned using the same data as Llama 2 with different weights (see Section 2 and Table 1 in the [research paper](https://ai.meta.com/research/publications/code-llama-open-foundation-models-for-code/) for details).
 
328
 
329
+ ## Evaluation Results
330
+
331
  See evaluations for the main models and detailed ablations in Section 3 and safety evaluations in Section 4 of the research paper.
332
 
333
+
334
+ ## Ethical Considerations and Limitations
335
+
336
  Code Llama and its variants are a new technology that carries risks with use. Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, Code Llama’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate or objectionable responses to user prompts. Therefore, before deploying any applications of Code Llama, developers should perform safety testing and tuning tailored to their specific applications of the model.
337
 
338
  Please see the Responsible Use Guide available available at [https://ai.meta.com/llama/responsible-user-guide](https://ai.meta.com/llama/responsible-user-guide).