Commit
•
129a056
1
Parent(s):
880eafb
Update README.md
Browse files
README.md
CHANGED
|
@@ -15,8 +15,6 @@ tags:
|
|
| 15 |
- assistant
|
| 16 |
pipeline_tag: text-generation
|
| 17 |
widget:
|
| 18 |
-
- text: "What is your name?<|endofinstruction|>"
|
| 19 |
-
example_title: Greetings
|
| 20 |
- text: "Can you explain what is Machine Learning?<|endofinstruction|>"
|
| 21 |
example_title: Machine Learning
|
| 22 |
- text: "Do you know anything about virtue ethics?<|endofinstruction|>"
|
|
@@ -108,14 +106,14 @@ The model will output something like:
|
|
| 108 |
|
| 109 |
## Evaluation
|
| 110 |
|
| 111 |
-
| Model (OPT) | Average | [ARC](https://arxiv.org/abs/1803.05457) | [TruthfulQA](https://arxiv.org/abs/2109.07958) | [ToxiGen](https://arxiv.org/abs/2203.09509) |
|
| 112 |
-
|
| 113 |
-
| [Aira-OPT-125M](https://huggingface.co/nicholasKluge/Aira-OPT-125M) | **43.34** | **24.65** | **49.11** | **56.27** |
|
| 114 |
-
| OPT-125M | 40.29 | 22.78 | 42.88 | 55.21 |
|
| 115 |
-
| [Aira-OPT-350M](https://huggingface.co/nicholasKluge/Aira-OPT-350M) | **41.56** | **25.00** | **42.13** | **57.55** |
|
| 116 |
-
| OPT-350M | 40.62 | 23.97 | 41.00 | 56.91 |
|
| 117 |
-
| [Aira-OPT-1B3](https://huggingface.co/nicholasKluge/Aira-OPT-1B3) | **43.90** | 28.41 | **46.59** | **56.70** |
|
| 118 |
-
| OPT-1.3b | 40.91 | **29.69** | 38.68 | 54.36 |
|
| 119 |
|
| 120 |
* Evaluations were performed using the [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) (by [EleutherAI](https://www.eleuther.ai/)).
|
| 121 |
|
|
|
|
| 15 |
- assistant
|
| 16 |
pipeline_tag: text-generation
|
| 17 |
widget:
|
|
|
|
|
|
|
| 18 |
- text: "Can you explain what is Machine Learning?<|endofinstruction|>"
|
| 19 |
example_title: Machine Learning
|
| 20 |
- text: "Do you know anything about virtue ethics?<|endofinstruction|>"
|
|
|
|
| 106 |
|
| 107 |
## Evaluation
|
| 108 |
|
| 109 |
+
| Model (OPT) | Average | [ARC](https://arxiv.org/abs/1803.05457) | [TruthfulQA](https://arxiv.org/abs/2109.07958) | [ToxiGen](https://arxiv.org/abs/2203.09509) |
|
| 110 |
+
|---------------------------------------------------------------------|-----------|-----------------------------------------|------------------------------------------------|---------------------------------------------|
|
| 111 |
+
| [Aira-OPT-125M](https://huggingface.co/nicholasKluge/Aira-OPT-125M) | **43.34** | **24.65** | **49.11** | **56.27** |
|
| 112 |
+
| OPT-125M | 40.29 | 22.78 | 42.88 | 55.21 |
|
| 113 |
+
| [Aira-OPT-350M](https://huggingface.co/nicholasKluge/Aira-OPT-350M) | **41.56** | **25.00** | **42.13** | **57.55** |
|
| 114 |
+
| OPT-350M | 40.62 | 23.97 | 41.00 | 56.91 |
|
| 115 |
+
| [Aira-OPT-1B3](https://huggingface.co/nicholasKluge/Aira-OPT-1B3) | **43.90** | 28.41 | **46.59** | **56.70** |
|
| 116 |
+
| OPT-1.3b | 40.91 | **29.69** | 38.68 | 54.36 |
|
| 117 |
|
| 118 |
* Evaluations were performed using the [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) (by [EleutherAI](https://www.eleuther.ai/)).
|
| 119 |
|