File size: 10,325 Bytes
842533b
64765e9
99b32ae
842533b
64765e9
 
 
 
 
f1785a4
 
64765e9
 
6c5a0ae
64765e9
 
 
 
 
7cd4fec
64765e9
9b4f44d
64765e9
 
 
 
6c5a0ae
64765e9
 
 
 
 
6c5a0ae
64765e9
 
 
f0ca2c8
64765e9
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
6c5a0ae
 
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
64765e9
 
 
 
 
 
 
 
 
6c5a0ae
842533b
64765e9
c2475bd
2f97d76
9b4f44d
64765e9
 
9b4f44d
64765e9
 
 
14f1d9f
64765e9
bbe82c2
64765e9
 
 
bbe82c2
64765e9
 
9b4f44d
64765e9
 
 
 
2ab68a5
9b4f44d
64765e9
2ab68a5
 
64765e9
2ab68a5
64765e9
2ab68a5
64765e9
2ab68a5
64765e9
2ab68a5
 
64765e9
2ab68a5
64765e9
2ab68a5
64765e9
b7cb4df
64765e9
 
 
 
 
 
 
 
 
2ab68a5
64765e9
bbe82c2
70ebbec
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
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
---
pipeline_tag: text-generation
inference: false
license: apache-2.0
datasets:
- codeparrot/github-code-clean
- bigcode/starcoderdata
# - Stackexchange
# - CommonCrawl
- open-web-math/open-web-math
- math-ai/StackMathQA
# - Arxiv
# - Wikipedia
# - conceptofmind/FLAN_2022 # Original link is broken, we used IBM's filtered version
metrics:
- code_eval
library_name: transformers
tags:
- code
- granite
model-index:
- name: granite-3b-code-base-2k
  results:
  - task:
      type: text-generation
    dataset:
        type: mbpp
        name: MBPP
    metrics:
    - name: pass@1
      type: pass@1
      value: 36.0
      veriefied: false     
  - task:
      type: text-generation
    dataset:
        type: evalplus/mbppplus 
        name: MBPP+
    metrics:
    - name: pass@1
      type: pass@1
      value: 45.1
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack 
        name: HumanEvalSynthesis(Python)
    metrics:
    - name: pass@1
      type: pass@1
      value: 36.6
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name: HumanEvalSynthesis(JavaScript)
    metrics:
    - name: pass@1
      type: pass@1
      value: 37.2
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name: HumanEvalSynthesis(Java)
    metrics:
    - name: pass@1
      type: pass@1
      value: 40.9
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name: HumanEvalSynthesis(Go)
    metrics:
    - name: pass@1
      type: pass@1
      value: 26.2
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name: HumanEvalSynthesis(C++)
    metrics:
    - name: pass@1
      type: pass@1
      value: 35.4
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name: HumanEvalSynthesis(Rust)
    metrics:
    - name: pass@1
      type: pass@1
      value: 22.0
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(Python)
    metrics:
    - name: pass@1
      type: pass@1
      value: 25.0
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(JavaScript)
    metrics:
    - name: pass@1
      type: pass@1
      value: 18.9
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(Java)
    metrics:
    - name: pass@1
      type: pass@1
      value: 29.9
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(Go)
    metrics:
    - name: pass@1
      type: pass@1
      value: 17.1
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(C++)
    metrics:
    - name: pass@1
      type: pass@1
      value: 26.8
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalExplain(Rust)
    metrics:
    - name: pass@1
      type: pass@1
      value: 14.0
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(Python)
    metrics:
    - name: pass@1
      type: pass@1
      value: 18.3
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(JavaScript)
    metrics:
    - name: pass@1
      type: pass@1
      value: 23.2
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(Java)
    metrics:
    - name: pass@1
      type: pass@1
      value: 29.9
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(Go)
    metrics:
    - name: pass@1
      type: pass@1
      value: 24.4
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(C++)
    metrics:
    - name: pass@1
      type: pass@1
      value: 16.5 
      veriefied: false
  - task:
      type: text-generation
    dataset:
        type: bigcode/humanevalpack  
        name:  HumanEvalFix(Rust)
    metrics:
    - name: pass@1
      type: pass@1
      value: 3.7
      veriefied: false
---

![image/png](https://cdn-uploads.huggingface.co/production/uploads/62cd5057674cdb524450093d/1hzxoPwqkBJXshKVVe6_9.png)

# Granite-3B-Code-Base-2K

## Model Summary
**Granite-3B-Code-Base-2K** is a decoder-only code model designed for code generative tasks (e.g., code generation, code explanation, code fixing, etc.). It is trained from scratch with a two-phase training strategy. In phase 1, our model is trained on 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.

- **Developers:** IBM Research
- **GitHub Repository:** [ibm-granite/granite-code-models](https://github.com/ibm-granite/granite-code-models)
- **Paper:** [Granite Code Models: A Family of Open Foundation Models for Code Intelligence](https://arxiv.org/abs/2405.04324)
- **Release Date**: May 6th, 2024
- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0).

## Usage
### Intended use
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. 

### Generation
This is a simple example of how to use **Granite-3B-Code-Base-2K** model.

```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # or "cpu"
model_path = "ibm-granite/granite-3b-code-base-2k"
tokenizer = AutoTokenizer.from_pretrained(model_path)
# drop device_map if running on CPU
model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)
model.eval()
# change input text as desired
input_text = "def generate():"
# tokenize the text
input_tokens = tokenizer(input_text, return_tensors="pt")
# transfer tokenized inputs to the device
for i in input_tokens:
    input_tokens[i] = input_tokens[i].to(device)
# generate output tokens
output = model.generate(**input_tokens)
# decode output tokens into text
output = tokenizer.batch_decode(output)
# loop over the batch to print, in this example the batch size is 1
for i in output:
    print(i)
```

## Training Data
- **Data Collection and Filtering:** Pretraining code data is sourced from a combination of publicly available datasets (e.g., [GitHub Code Clean](https://huggingface.co/datasets/codeparrot/github-code-clean), [Starcoder data](https://huggingface.co/datasets/bigcode/starcoderdata)), and additional public code repositories and issues from GitHub. We filter raw data to retain a list of 116 programming languages. After language filtering, we also filter out low-quality code. 
- **Exact and Fuzzy Deduplication:** We adopt an aggressive deduplication strategy that includes both exact and fuzzy deduplication to remove documents having (near) identical code content.
- **HAP, PII, Malware Filtering:** We apply a HAP content filter that reduces models' likelihood of generating hateful, abusive, or profane language. We also make sure to redact Personally Identifiable Information (PII) by replacing PII content (e.g., names, email addresses, keys, passwords) with corresponding tokens (e.g., ⟨NAME⟩, ⟨EMAIL⟩, ⟨KEY⟩, ⟨PASSWORD⟩). Moreover, we scan all datasets using [ClamAV](https://www.clamav.net/) to identify and remove instances of malware in the source code.
- **Natural Language Datasets:** In addition to collecting code data for model training, we curate several publicly available high-quality natural language datasets to improve models' proficiency in language understanding and mathematical reasoning. Unlike the code data, we do not deduplicate these datasets.

## Infrastructure
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.

## Ethical Considerations and Limitations
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-2K** 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-2K** model with ethical intentions and in a responsible way.