Text Generation
Transformers
Safetensors
llama
code
granite
Eval Results
text-generation-inference
amezasor commited on
Commit
bbe82c2
1 Parent(s): f1785a4

model card revision

Browse files
Files changed (1) hide show
  1. README.md +7 -8
README.md CHANGED
@@ -270,25 +270,24 @@ Limitations
270
  Citation
271
  -->
272
 
273
- # Granite 3B Code Base
274
- <!-- ![granite](https://github.com/ibm-granite/granite-code-models/blob/main/figures/granite.png) -->
275
 
276
  ## Model Summary
277
- **Granite 3B Code Base** is a decoder-only code model designed for code generative tasks (e.g., code generation, code explanation, code fixing). It was trained from scratch on 4 trillion tokens sourced from 116 programming languages, ensuring a comprehensive understanding of programming languages and syntax.
278
 
279
  - **Developers:** IBM Research
280
  - **GitHub Repository:** [ibm-granite/granite-code-models](https://github.com/ibm-granite/granite-code-models)
281
  - **Paper:** [Granite Code Models: A Family of Open Foundation Models
282
  for Code Intelligence](https://)
283
  - **Release Date**: May 6th, 2024
284
- - **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0) license.
285
 
286
  ## Usage
287
  ### Intended use
288
- Prominent enterprise use cases of LLMs in software engineering productivity include code generation, code explanation, code fixing, generating unit tests, generating documentation, addressing technical debt issues, vulnerability detection, code translation, and more. All Granite Code Base models, including the **3B parameters model**, are able to handle these tasks as they were trained on a large amount of code data from 116 programming languages.
289
 
290
  ### Generation
291
- This is a simple example of how to use Granite Code Base 3B model.
292
 
293
  ```python
294
  import torch
@@ -333,8 +332,8 @@ for i in output:
333
  ## Infrastructure
334
  We train the Granite Code models using two of IBM's super computing clusters, namely Vela and Blue Vela, both outfitted with NVIDIA A100 and H100 GPUs respectively. These clusters provide a scalable and efficient infrastructure for training our models over thousands of GPUs.
335
 
336
- ## Limitations
337
- Large Language Models are often prone to generating incorrect information, typically referred to as hallucinations. **Granite 3B Code Base** model is not the exception in this regard. Even though this model is suited for code-related tasks as it is trained on source code from 116 programming languages, the generated code is not guaranteed to work as intended. It can be inefficient and can also contain bugs or exploits. Moreover, Granite Code Base models are **NOT** instruction-following models. Thus, commands like *"Write a function that computes the square root"* may not work well. The model is best treated as a code completion or code infilling model.
338
 
339
  ## Citation
340
  ```
 
270
  Citation
271
  -->
272
 
273
+ # Granite-3B-Code-Base
 
274
 
275
  ## Model Summary
276
+ **Granite-3B-Code-Base** is a decoder-only code model designed for code generative tasks (e.g.code generation, code explanation, code fixing, etc.). It was trained from scratch with a two-phase training strategy. In phase 1, our model is trained on 3 to 4 trillion tokens sourced from 116 programming languages, ensuring a comprehensive understanding of programming languages and syntax. In phase 2, our model is trained on 500 billion tokens with a carefully designed mixture of high-quality data from code and natural language domains to improve the models’ ability to reason and follow instructions.
277
 
278
  - **Developers:** IBM Research
279
  - **GitHub Repository:** [ibm-granite/granite-code-models](https://github.com/ibm-granite/granite-code-models)
280
  - **Paper:** [Granite Code Models: A Family of Open Foundation Models
281
  for Code Intelligence](https://)
282
  - **Release Date**: May 6th, 2024
283
+ - **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0).
284
 
285
  ## Usage
286
  ### Intended use
287
+ Prominent enterprise use cases of LLMs in software engineering productivity include code generation, code explanation, code fixing, generating unit tests, generating documentation, addressing technical debt issues, vulnerability detection, code translation, and more. All Granite Code Base models, including the **3B parameter model**, are able to handle these tasks as they were trained on a large amount of code data from 116 programming languages.
288
 
289
  ### Generation
290
+ This is a simple example of how to use **Granite-Code-Base-3B model**.
291
 
292
  ```python
293
  import torch
 
332
  ## Infrastructure
333
  We train the Granite Code models using two of IBM's super computing clusters, namely Vela and Blue Vela, both outfitted with NVIDIA A100 and H100 GPUs respectively. These clusters provide a scalable and efficient infrastructure for training our models over thousands of GPUs.
334
 
335
+ ## Ethical Considerations and Limitations
336
+ The use of Large Language Models involves risks and ethical considerations people must be aware of. Regarding code generation, caution is urged against complete reliance on specific code models for crucial decisions or impactful information as the generated code is not guaranteed to work as intended. **Granite-3B-Code-Base** model is not the exception in this regard. Even though this model is suited for multiple code-related tasks, it has not undergone any safety alignment, there it may produce problematic outputs. Additionally, it remains uncertain whether smaller models might exhibit increased susceptibility to hallucination in generation scenarios by copying source code verbatim from the training dataset due to their reduced sizes and memorization capacities. This aspect is currently an active area of research, and we anticipate more rigorous exploration, comprehension, and mitigations in this domain. Regarding ethics, a latent risk associated with all Large Language Models is their malicious utilization. We urge the community to use **Granite-3B-Code-Base** model with ethical intentions and in a responsible way. 
337
 
338
  ## Citation
339
  ```