Brandl commited on
Commit
9ddf9b9
·
verified ·
1 Parent(s): 7c36112

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +31 -0
README.md ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ datasets:
3
+ - ctitools/orkl_cleaned_10k
4
+ language:
5
+ - en
6
+ base_model:
7
+ - mistralai/Mistral-Nemo-Instruct-2407
8
+ pipeline_tag: summarization
9
+ tags:
10
+ - cyber
11
+ - cyber threat
12
+ - intelligence
13
+ metrics:
14
+ - perplexity
15
+ ---
16
+ # Overview
17
+
18
+ This model was trained via [mistral-finetune](https://github.com/mistralai/mistral-finetune/) on top
19
+ of the [orkl.eu](https://orkl.eu) dataset on 10k Cyber Threat Intelligence (CTI) reports.
20
+
21
+ This is the merged model with the mistral-nemo-instruct-2407 base model, for ease of use with other inference frameworks.
22
+
23
+ Running this is as easy as, cloning the repo and running:
24
+ ```bash
25
+ pip install mistral-inference
26
+ mistral-chat neurocti-mistral-nemo-12b-orkl-instruct-merged/ --max_tokens 1024 --temperature 0.35
27
+ ```
28
+
29
+ For details on the model, the data and training, please read [https://github.com/ctitools/neurocti/](https://github.com/ctitools/neurocti/).
30
+
31
+ For understanding how to run this model, please read: [https://github.com/ctitools/neurocti/blob/main/inference.md](https://github.com/ctitools/neurocti/blob/main/inference.md)