Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,18 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
---
|
4 |
+
|
5 |
+
GPTQ 4-bit no actor version for compatibility that works in textgen-webui
|
6 |
+
|
7 |
+
Generated by using scripts from https://gitee.com/yhyu13/llama_-tools
|
8 |
+
|
9 |
+
Merged weights: https://huggingface.co/Yhyu13/oasst-rlhf-2-llama-30b-7k-steps-hf
|
10 |
+
|
11 |
+
Converted LLaMA weights: https://huggingface.co/Yhyu13/llama-30B-hf-openassitant
|
12 |
+
|
13 |
+
Delta weights: https://huggingface.co/OpenAssistant/oasst-rlhf-2-llama-30b-7k-steps-xor
|
14 |
+
|
15 |
+
---
|
16 |
+
|
17 |
+
OA has done a great jobs in RLHF their pre-trained weights. I must say it is tuned to spit out CoT step by step thinking without you actively prompting it to do so,
|
18 |
+
which is a feature that we observe on ChatGPT and GPT-4.
|