Create README.md
#2
by
vwxyzjn
- opened
README.md
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: bigscience-openrail-m
|
| 3 |
+
datasets:
|
| 4 |
+
- gsm8k
|
| 5 |
+
language:
|
| 6 |
+
- en
|
| 7 |
+
tags:
|
| 8 |
+
- trl
|
| 9 |
+
- transformers
|
| 10 |
+
- rlhf
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# starcoderbase-triviaqa
|
| 14 |
+
|
| 15 |
+
This model is baesed on https://huggingface.co/bigcode/starcoderbase and is fine-tuned on the GSM8K dataset using reinforcement learning via TRL's `TextEnvironment` (https://github.com/huggingface/trl/pull/424).
|
| 16 |
+
|
| 17 |
+
|
| 18 |
+
### Out of Scope Use
|
| 19 |
+
- Replacing human expertise
|
| 20 |
+
|
| 21 |
+
|
| 22 |
+
## Bias, Risks, and Limitations
|
| 23 |
+
- May generate answers that are incorrect or misleading.
|
| 24 |
+
- May copy answers from the training data verbatim.
|
| 25 |
+
- May generate language that is hateful or promotes discrimination ([example](https://huggingface.co/trl-lib/llama-7b-se-rl-peft/discussions/7#64376083369f6f907f5bfe4c)).
|
| 26 |
+
- May generate language that is offensive to direct or indirect users or to people or groups mentioned.
|
| 27 |
+
|
| 28 |
+
|
| 29 |
+
### Recommendations
|
| 30 |
+
- Answers should be validated through the use of external sources.
|
| 31 |
+
- Disparities between the data contributors and the direct and indirect users of the technology should inform developers in assessing what constitutes an appropriate use case.
|
| 32 |
+
- Further research is needed to attribute model generations to sources in the training data, especially in cases where the model copies answers from the training data.
|