chavinlo leaderboard-pr-bot commited on
Commit
7cf7f42
1 Parent(s): 6a571f4

Adding Evaluation Results (#18)

Browse files

- Adding Evaluation Results (9aaab9a1e944f3f8fa801c2e24f690c4e59b3b3c)


Co-authored-by: Open LLM Leaderboard PR Bot <[email protected]>

Files changed (1) hide show
  1. README.md +14 -1
README.md CHANGED
@@ -6,4 +6,17 @@ Finetuned on GPT4's responses, for 3 epochs.
6
 
7
  NO LORA
8
 
9
- Please do note that the configurations files maybe messed up, this is because of the trainer I used. I WILL NOT EDIT THEM because there are repos hat automatically fix this, changing it might break it. Generally you just need to change anything that's under the name of "LLaMa" to "Llama" NOTE THE UPPER AND LOWER CASE!!!!
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6
 
7
  NO LORA
8
 
9
+ Please do note that the configurations files maybe messed up, this is because of the trainer I used. I WILL NOT EDIT THEM because there are repos hat automatically fix this, changing it might break it. Generally you just need to change anything that's under the name of "LLaMa" to "Llama" NOTE THE UPPER AND LOWER CASE!!!!
10
+ # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
11
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_chavinlo__gpt4-x-alpaca)
12
+
13
+ | Metric | Value |
14
+ |-----------------------|---------------------------|
15
+ | Avg. | 46.78 |
16
+ | ARC (25-shot) | 52.82 |
17
+ | HellaSwag (10-shot) | 79.59 |
18
+ | MMLU (5-shot) | 48.19 |
19
+ | TruthfulQA (0-shot) | 48.88 |
20
+ | Winogrande (5-shot) | 70.17 |
21
+ | GSM8K (5-shot) | 2.81 |
22
+ | DROP (3-shot) | 24.99 |