From 45815a4572d43cc1b99e0c0419244d0d4daeac37 Mon Sep 17 00:00:00 2001
From: morpheuslord <70637311+morpheuslord@users.noreply.github.com>
Date: Sat, 29 Jul 2023 15:46:45 +0530
Subject: [PATCH] Update README.md

---
 README.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/README.md b/README.md
index 4d4c4be..ac3fe5d 100644
--- a/README.md
+++ b/README.md
@@ -314,10 +314,10 @@ Once the API is acquired just add it to the `.env` file and you are good to go.
 Using LLama2 is one of the best offline and free options out there. It is currently under improvement I am working on a prompt that will better incorporate cybersecurity perspective into the AI.
 I have to thank **@thisserand** and his [llama2_local](https://github.com/thisserand/llama2_local) repo and also his YT video [YT_Video](https://youtu.be/WzCS8z9GqHw). They were great resources. To be frank the llama2 code is 95% his, I just yanked the code and added a Flask API functionality to it.
 
-The Accuracy of the AI in offline and outside the codes test was great and had equal accuracy to openai or bard but while in code it was facing a few issues may be because of the prompting and all. I will try and fix it.
+The Accuracy of the AI offline and outside the codes test was great and had equal accuracy to openai or bard but while in code it was facing a few issues may be because of the prompting and all. I will try and fix it.
 The speed depends on your system and the GPU and CPU configs you have. currently, it is using the `TheBloke/Llama-2-7B-Chat-GGML` model and can be changed via the `portscanner` and `dnsrecon` files.
 
-For now the llama code and scans are handeled differently. After few tests I found out llama needs to be trained a little to opparate like how I intended it to work so it needs some time. Any suggestions on how I can do that can be added in the discussions of this repo [Discussions Link](https://github.com/morpheuslord/GPT_Vuln-analyzer/discussions). For now the output wont be a devided list of all the data instead will be an explaination of the vulnerability or issues discovered by the AI.
+For now, the llama code and scans are handled differently. After a few tests, I found out llama needs to be trained a little to operate like how I intended it to work so it needs some time. Any suggestions on how I can do that can be added to the discussions of this repo [Discussions Link](https://github.com/morpheuslord/GPT_Vuln-analyzer/discussions). For now, the output won't be a divided list of all the data instead will be an explanation of the vulnerability or issues discovered by the AI.
 
 ### Output