Commit
•
a4b129d
1
Parent(s):
1d22518
Update README.md
Browse files
README.md
CHANGED
@@ -9,14 +9,14 @@ This repository contains the request files of models that have been submitted to
|
|
9 |
You can take a look at the current status of your model by finding its request file in this dataset. If your model failed, feel free to open an issue on the Open LLM Leaderboard! (We don't follow issues in this repository as often)
|
10 |
|
11 |
## Evaluation Methodology
|
12 |
-
The evaluation process involves running your models against several
|
13 |
|
14 |
-
1. AI2 Reasoning Challenge (ARC) -
|
15 |
-
2. HellaSwag - 10-shot
|
16 |
-
3. MMLU -
|
17 |
4. TruthfulQA - Propensity to Produce Falsehoods (0-shot)
|
18 |
-
5. Winogrande -
|
19 |
-
6. GSM8k -
|
20 |
|
21 |
Together, these benchmarks provide an assessment of a model's capabilities in terms of knowledge, reasoning, and some math, in various scenarios.
|
22 |
|
|
|
9 |
You can take a look at the current status of your model by finding its request file in this dataset. If your model failed, feel free to open an issue on the Open LLM Leaderboard! (We don't follow issues in this repository as often)
|
10 |
|
11 |
## Evaluation Methodology
|
12 |
+
The evaluation process involves running your models against several benchmarks from the Eleuther AI Harness, a unified framework for measuring the effectiveness of generative language models. Below is a brief overview of each benchmark:
|
13 |
|
14 |
+
1. AI2 Reasoning Challenge (ARC) - Grade-School Science Questions (25-shot)
|
15 |
+
2. HellaSwag - Commonsense Inference (10-shot)
|
16 |
+
3. MMLU - Massive Multi-Task Language Understanding, knowledge on 57 domains (5-shot)
|
17 |
4. TruthfulQA - Propensity to Produce Falsehoods (0-shot)
|
18 |
+
5. Winogrande - Adversarial Winograd Schema Challenge (5-shot)
|
19 |
+
6. GSM8k - Grade School Math Word Problems Solving Complex Mathematical Reasoning (5-shot)
|
20 |
|
21 |
Together, these benchmarks provide an assessment of a model's capabilities in terms of knowledge, reasoning, and some math, in various scenarios.
|
22 |
|