Update README.md
Browse files
README.md
CHANGED
@@ -55,6 +55,8 @@ It achieves the following results on the evaluation set:
|
|
55 |
|
56 |
This is just a base model/example, and there is likely even better performance with larger models.
|
57 |
|
|
|
|
|
58 |
## Training and evaluation data
|
59 |
|
60 |
See the linked dataset `pszemraj/fleece2instructions` - it is a filtered/formatted version of `tatsu-lab/alpaca` to generate instructions for arbitrary text.
|
|
|
55 |
|
56 |
This is just a base model/example, and there is likely even better performance with larger models.
|
57 |
|
58 |
+
Additionally, this was trained on a dataset of **only** instructions+outputs with the `inputs` filtered out/dropped. This means that text of *1) cookies and cream 2) chocolate chip 3) mint chip 4) oreo* will **not** get you *"Rank the following icecream flavors: oreo, mint chip, chocolate chip, cookies and cream"*
|
59 |
+
|
60 |
## Training and evaluation data
|
61 |
|
62 |
See the linked dataset `pszemraj/fleece2instructions` - it is a filtered/formatted version of `tatsu-lab/alpaca` to generate instructions for arbitrary text.
|