morpheuslord commited on
Commit
e1932a8
1 Parent(s): 5ba1046

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +37 -1
README.md CHANGED
@@ -7,6 +7,42 @@ language:
7
  tags:
8
  - code
9
  pretty_name: CVE-LLm-Dataset
10
- data_source: Custom data collected from CVE database
11
  data_formats: JSONL
12
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  tags:
8
  - code
9
  pretty_name: CVE-LLm-Dataset
10
+ data_source: Custom data collected from the CVE database
11
  data_formats: JSONL
12
  ---
13
+ # CVE-llm_dataset
14
+ This dataset is intended to train an LLM model for an utterly CVE-focused input and output.
15
+
16
+ ## Data extraction:
17
+ For the data extraction, I first downloaded the CVE database from NVD lists and then loaded them using the `cve_dataset_2.py` and `cve_dataset.py` both have produce different datasets one is for llama and the other is for openai GPT.
18
+
19
+ The CVE json files are mapped in this format:
20
+ ```
21
+ cves:
22
+ |
23
+ ├─1999
24
+ | ├─0xxx
25
+ | | ├─CVE-1999-0001.json
26
+ | | └─CVE-1999-0999.json
27
+ | └─1xxx
28
+ | ├─CVE-1999-1000.json
29
+ | └─CVE-1999-1598.json
30
+ └─2023
31
+
32
+ ```
33
+ The programs traverse through these folders extract the data in the files and arrange them into usable formats for the fine-tuning process.
34
+
35
+ ## llama2 Model dataset:
36
+ The llama2 fine-tuned dataset follows this format:
37
+ ```
38
+ {
39
+ "instruction": "Explain CVE-1999-0001",
40
+ "input": "Explain the vulnerability: CVE-1999-0001",
41
+ "output": "ip_input.c in BSD-derived TCP/IP implementations allows remote attackers to cause a denial of service (crash or hang) via crafted packets.\nAffected Products: n/a\nReferences: [{'tags': ['x_refsource_CONFIRM'], 'url': 'http://www.openbsd.org/errata23.html#tcpfix'}, {'name': '5707', 'tags': ['vdb-entry', 'x_refsource_OSVDB'], 'url': 'http://www.osvdb.org/5707'}]\nCVE State: PUBLISHED"
42
+ }
43
+ ```
44
+ The instruction is what we instruct the AI to do with the data provided For example we can command the AI `To take in user input analyze it and then based on what he asks returns an answer` This is also where we can add a `role` or a `personal` to the AI.
45
+
46
+ The input is the user Input of the main query or data that must be processed by the AI. This is a crucial piece of information that the AI will process in order to provide an output.
47
+
48
+ The output is the format that we define and tell the AI to generate answers in that format or provide that answer to the question asked.