weqweasdas
/

hh_rlhf_rm_open_llama_3b

Text Classification

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

weqweasdas commited on Dec 24, 2023

Commit

13f510a

·

1 Parent(s): 7cad203

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -42,9 +42,9 @@ We further test the generalization ability of the reward model but with another
 | Dataset  training/test     | open assistant | chatbot | hh_rlhf |
 | -------------- | -------------- | ------- | ------- |
-| open assistant | 69.5           | 61.1    | 58.7    |
 | chatbot        | 66.5           | 62.7    | 56.0    |
-| hh_rlhf        | 69.4           | 64.2    | 77.6    |
 As we can see, the reward model trained on the HH-RLHF achieves matching or even better accuracy on open assistant and chatbot datasets, even though it is not trained on them directly. Therefore, the reward model may also be used for these two datasets.

 | Dataset  training/test     | open assistant | chatbot | hh_rlhf |
 | -------------- | -------------- | ------- | ------- |
+| open assistant | **69.5**           | 61.1    | 58.7    |
 | chatbot        | 66.5           | 62.7    | 56.0    |
+| hh_rlhf        | 69.4           | **64.2**    | **77.6**    |
 As we can see, the reward model trained on the HH-RLHF achieves matching or even better accuracy on open assistant and chatbot datasets, even though it is not trained on them directly. Therefore, the reward model may also be used for these two datasets.