Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,19 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- dattaraj/security-attacks-MITRE
|
| 5 |
+
language:
|
| 6 |
+
- en
|
| 7 |
+
library_name: mlx
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
### Fine-tuned model for Cybersecurity SOAR usecases.
|
| 11 |
+
#### Base model - microsoft/Phi-3-mini-4k-instruct
|
| 12 |
+
#### Instruction dataset - dattaraj/security-attacks-MITRE
|
| 13 |
+
#### Platform - Apple MLX on Mac M3
|
| 14 |
+
#### Author - Dattaraj Rao
|
| 15 |
+
https://in.linkedin.com/in/dattarajrao
|
| 16 |
+
|
| 17 |
+
### Why fine-tune?
|
| 18 |
+
Fine-tuned to act like a Cybersecurity expert. Understand a potential attack scenario. map to MIRE ATT&CK guidelines and provide mitigations.
|
| 19 |
+
Base model would provide generic recommendations. While the fine-tuned model shows much better domain adoption and MITRE mapping.
|