jondurbin commited on
Commit
f6de495
1 Parent(s): aa08e4a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -8,9 +8,9 @@ _Not tested yet, use if you want though!_
8
 
9
  ### Overview
10
 
11
- This is a qlora fine-tuned 13b parameter LlaMa model, using completely synthetic training data created gpt4 via https://github.com/jondurbin/airoboros
12
 
13
- This is mostly an extension of [1.2](https://huggingface.co/jondurbin/airoboros-13b-gpt4-1.2) with a few enhancements:
14
 
15
  - All coding instructions have an equivalent " PLAINFORMAT" version now.
16
  - Thousands of new orca style reasoning instructions, this time with reasoning first, then answer.
@@ -36,7 +36,7 @@ Be sure you are pulling the latest branch!
36
  Then, you can invoke it like so (after downloading the model):
37
  ```
38
  python -m fastchat.serve.cli \
39
- --model-path airoboros-13b-gpt4-1.3 \
40
  --temperature 0.5 \
41
  --max-new-tokens 2048 \
42
  --no-history
 
8
 
9
  ### Overview
10
 
11
+ This is a qlora fine-tuned 7b parameter LlaMa model, using completely synthetic training data created gpt4 via https://github.com/jondurbin/airoboros
12
 
13
+ This is mostly an extension of [1.2](https://huggingface.co/jondurbin/airoboros-7b-gpt4-1.2) with a few enhancements:
14
 
15
  - All coding instructions have an equivalent " PLAINFORMAT" version now.
16
  - Thousands of new orca style reasoning instructions, this time with reasoning first, then answer.
 
36
  Then, you can invoke it like so (after downloading the model):
37
  ```
38
  python -m fastchat.serve.cli \
39
+ --model-path airoboros-7b-gpt4-1.3 \
40
  --temperature 0.5 \
41
  --max-new-tokens 2048 \
42
  --no-history