leaderboard-pr-bot
commited on
Commit
•
ba9fa07
1
Parent(s):
036bb03
Adding Evaluation Results
Browse filesThis is an automated PR created with https://huggingface.co/spaces/Weyaxi/open-llm-leaderboard-results-pr
The purpose of this PR is to add evaluation results from the Open LLM Leaderboard to your model card.
If you encounter any issues, please report them to https://huggingface.co/spaces/Weyaxi/open-llm-leaderboard-results-pr/discussions
README.md
CHANGED
@@ -1,42 +1,56 @@
|
|
1 |
-
---
|
2 |
-
language: en
|
3 |
-
license: mit
|
4 |
-
---
|
5 |
-
# GPT-J 6B - Janeway
|
6 |
-
## Model Description
|
7 |
-
GPT-J 6B-Janeway is a finetune created using EleutherAI's GPT-J 6B model.
|
8 |
-
## Training data
|
9 |
-
The training data contains around 2210 ebooks, mostly in the sci-fi and fantasy genres. The dataset is based on the same dataset used by GPT-Neo-2.7B-Picard, with 20% more data in various genres.
|
10 |
-
Some parts of the dataset have been prepended using the following text: `[Genre: <genre1>,<genre2>]`
|
11 |
-
### How to use
|
12 |
-
You can use this model directly with a pipeline for text generation. This example generates a different sequence each time it's run:
|
13 |
-
```py
|
14 |
-
>>> from transformers import pipeline
|
15 |
-
>>> generator = pipeline('text-generation', model='KoboldAI/GPT-J-6B-Janeway')
|
16 |
-
>>> generator("Welcome Captain Janeway, I apologize for the delay.", do_sample=True, min_length=50)
|
17 |
-
[{'generated_text': 'Welcome Captain Janeway, I apologize for the delay."\nIt's all right," Janeway said. "I'm certain that you're doing your best to keep me informed of what\'s going on."'}]
|
18 |
-
```
|
19 |
-
### Limitations and Biases
|
20 |
-
|
21 |
-
The core functionality of GPT-J is taking a string of text and predicting the next token. While language models are widely used for tasks other than this, there are a lot of unknowns with this work. When prompting GPT-J it is important to remember that the statistically most likely next token is often not the token that produces the most "accurate" text. Never depend upon GPT-J to produce factually accurate output.
|
22 |
-
|
23 |
-
GPT-J was trained on the Pile, a dataset known to contain profanity, lewd, and otherwise abrasive language. Depending upon use case GPT-J may produce socially unacceptable text. See [Sections 5 and 6 of the Pile paper](https://arxiv.org/abs/2101.00027) for a more detailed analysis of the biases in the Pile.
|
24 |
-
|
25 |
-
As with all language models, it is hard to predict in advance how GPT-J will respond to particular prompts and offensive content may occur without warning. We recommend having a human curate or filter the outputs before releasing them, both to censor undesirable content and to improve the quality of the results.
|
26 |
-
|
27 |
-
### BibTeX entry and citation info
|
28 |
-
The model uses the following model as base:
|
29 |
-
```bibtex
|
30 |
-
@misc{gpt-j,
|
31 |
-
author = {Wang, Ben and Komatsuzaki, Aran},
|
32 |
-
title = {{GPT-J-6B: A 6 Billion Parameter Autoregressive Language Model}},
|
33 |
-
howpublished = {\url{https://github.com/kingoflolz/mesh-transformer-jax}},
|
34 |
-
year = 2021,
|
35 |
-
month = May
|
36 |
-
}
|
37 |
-
```
|
38 |
-
|
39 |
-
## Acknowledgements
|
40 |
-
|
41 |
-
This project would not have been possible without compute generously provided by Google through the
|
42 |
-
[TPU Research Cloud](https://sites.research.google/trc/), as well as the Cloud TPU team for providing early access to the [Cloud TPU VM](https://cloud.google.com/blog/products/compute/introducing-cloud-tpu-vms) Alpha.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language: en
|
3 |
+
license: mit
|
4 |
+
---
|
5 |
+
# GPT-J 6B - Janeway
|
6 |
+
## Model Description
|
7 |
+
GPT-J 6B-Janeway is a finetune created using EleutherAI's GPT-J 6B model.
|
8 |
+
## Training data
|
9 |
+
The training data contains around 2210 ebooks, mostly in the sci-fi and fantasy genres. The dataset is based on the same dataset used by GPT-Neo-2.7B-Picard, with 20% more data in various genres.
|
10 |
+
Some parts of the dataset have been prepended using the following text: `[Genre: <genre1>,<genre2>]`
|
11 |
+
### How to use
|
12 |
+
You can use this model directly with a pipeline for text generation. This example generates a different sequence each time it's run:
|
13 |
+
```py
|
14 |
+
>>> from transformers import pipeline
|
15 |
+
>>> generator = pipeline('text-generation', model='KoboldAI/GPT-J-6B-Janeway')
|
16 |
+
>>> generator("Welcome Captain Janeway, I apologize for the delay.", do_sample=True, min_length=50)
|
17 |
+
[{'generated_text': 'Welcome Captain Janeway, I apologize for the delay."\nIt's all right," Janeway said. "I'm certain that you're doing your best to keep me informed of what\'s going on."'}]
|
18 |
+
```
|
19 |
+
### Limitations and Biases
|
20 |
+
|
21 |
+
The core functionality of GPT-J is taking a string of text and predicting the next token. While language models are widely used for tasks other than this, there are a lot of unknowns with this work. When prompting GPT-J it is important to remember that the statistically most likely next token is often not the token that produces the most "accurate" text. Never depend upon GPT-J to produce factually accurate output.
|
22 |
+
|
23 |
+
GPT-J was trained on the Pile, a dataset known to contain profanity, lewd, and otherwise abrasive language. Depending upon use case GPT-J may produce socially unacceptable text. See [Sections 5 and 6 of the Pile paper](https://arxiv.org/abs/2101.00027) for a more detailed analysis of the biases in the Pile.
|
24 |
+
|
25 |
+
As with all language models, it is hard to predict in advance how GPT-J will respond to particular prompts and offensive content may occur without warning. We recommend having a human curate or filter the outputs before releasing them, both to censor undesirable content and to improve the quality of the results.
|
26 |
+
|
27 |
+
### BibTeX entry and citation info
|
28 |
+
The model uses the following model as base:
|
29 |
+
```bibtex
|
30 |
+
@misc{gpt-j,
|
31 |
+
author = {Wang, Ben and Komatsuzaki, Aran},
|
32 |
+
title = {{GPT-J-6B: A 6 Billion Parameter Autoregressive Language Model}},
|
33 |
+
howpublished = {\url{https://github.com/kingoflolz/mesh-transformer-jax}},
|
34 |
+
year = 2021,
|
35 |
+
month = May
|
36 |
+
}
|
37 |
+
```
|
38 |
+
|
39 |
+
## Acknowledgements
|
40 |
+
|
41 |
+
This project would not have been possible without compute generously provided by Google through the
|
42 |
+
[TPU Research Cloud](https://sites.research.google/trc/), as well as the Cloud TPU team for providing early access to the [Cloud TPU VM](https://cloud.google.com/blog/products/compute/introducing-cloud-tpu-vms) Alpha.
|
43 |
+
|
44 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
45 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_KoboldAI__GPT-J-6B-Janeway)
|
46 |
+
|
47 |
+
| Metric | Value |
|
48 |
+
|-----------------------|---------------------------|
|
49 |
+
| Avg. | 34.57 |
|
50 |
+
| ARC (25-shot) | 40.87 |
|
51 |
+
| HellaSwag (10-shot) | 67.11 |
|
52 |
+
| MMLU (5-shot) | 27.45 |
|
53 |
+
| TruthfulQA (0-shot) | 35.74 |
|
54 |
+
| Winogrande (5-shot) | 64.72 |
|
55 |
+
| GSM8K (5-shot) | 1.36 |
|
56 |
+
| DROP (3-shot) | 4.76 |
|