Update README.md
Browse files
README.md
CHANGED
|
@@ -31,6 +31,71 @@ Note that we do *not* include prompts and responses from the original datasets,
|
|
| 31 |
(e.g., `conversation_id`) to enable merging with the base data. The annotation data is
|
| 32 |
sufficient for independent local analysis with the Feedback Forensics Gradio app, even without merging.
|
| 33 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
## Annotated datsets
|
| 35 |
|
| 36 |
We provide annotations for (subsets) of the following external datasets:
|
|
|
|
| 31 |
(e.g., `conversation_id`) to enable merging with the base data. The annotation data is
|
| 32 |
sufficient for independent local analysis with the Feedback Forensics Gradio app, even without merging.
|
| 33 |
|
| 34 |
+
## Format
|
| 35 |
+
|
| 36 |
+
The data is provided in the AnnotatedPairs json data
|
| 37 |
+
format, [described here](https://github.com/rdnfn/icai?tab=readme-ov-file#using-results-for-further-analysis-the-annotated-pairs-json-format-v20).
|
| 38 |
+
See a minimal illustrative example below:
|
| 39 |
+
|
| 40 |
+
```json
|
| 41 |
+
{
|
| 42 |
+
"metadata": {
|
| 43 |
+
"version": "2.0",
|
| 44 |
+
"description": "Annotated pairs dataset with annotations from ICAI",
|
| 45 |
+
"created_at": "2025-04-02",
|
| 46 |
+
"dataset_name": "dataset",
|
| 47 |
+
"default_annotator": "d36860d4"
|
| 48 |
+
},
|
| 49 |
+
"annotators": {
|
| 50 |
+
"d36860d4": {
|
| 51 |
+
"name": "Human",
|
| 52 |
+
"description": "Human annotator from original dataset",
|
| 53 |
+
"type": "human"
|
| 54 |
+
},
|
| 55 |
+
"2f45a6d0": {
|
| 56 |
+
"description": "Select the response that evokes a sense of mystery.",
|
| 57 |
+
"type": "principle"
|
| 58 |
+
},
|
| 59 |
+
"435cef52": {
|
| 60 |
+
"description": "Select the response that features a more adventurous setting.",
|
| 61 |
+
"type": "principle"
|
| 62 |
+
}
|
| 63 |
+
},
|
| 64 |
+
"comparisons": [
|
| 65 |
+
{
|
| 66 |
+
"id": "4b49a6e7",
|
| 67 |
+
"prompt": "Write a short story about a pet.",
|
| 68 |
+
"response_a": {
|
| 69 |
+
"text": "In the heart of a bustling city, a sleek black cat named Shadow prowled the moonlit rooftops, her eyes gleaming with curiosity and mischief. She discovered a hidden garden atop an old apartment building, where she danced under the stars, chasing fireflies that glowed like tiny lanterns. As dawn painted the sky in hues of orange and pink, Shadow found her way back home, carrying the secret of the garden in her heart.",
|
| 70 |
+
"model": "Model X"
|
| 71 |
+
},
|
| 72 |
+
"response_b": {
|
| 73 |
+
"text": "Across the town, in a cozy neighborhood, a golden retriever named Buddy embarked on his daily adventure, tail wagging with uncontainable excitement. He found a lost toy under the bushes in the park, its colors faded and fabric worn, but to Buddy, it was a treasure untold. Returning home with his newfound prize, Buddy's joyful barks filled the air, reminding everyone in the house that happiness can be found in the simplest of things.",
|
| 74 |
+
"model": "Model Y"
|
| 75 |
+
},
|
| 76 |
+
"annotations": {
|
| 77 |
+
"d36860d4": {
|
| 78 |
+
"pref": "a"
|
| 79 |
+
},
|
| 80 |
+
"2f45a6d0": {
|
| 81 |
+
"pref": "a"
|
| 82 |
+
},
|
| 83 |
+
"435cef52": {
|
| 84 |
+
"pref": "a"
|
| 85 |
+
}
|
| 86 |
+
},
|
| 87 |
+
"metadata": {
|
| 88 |
+
"index": "0"
|
| 89 |
+
}
|
| 90 |
+
}
|
| 91 |
+
]
|
| 92 |
+
}
|
| 93 |
+
```
|
| 94 |
+
|
| 95 |
+
The main annotations are included under `dataset['comparisons'][i]["annotations"]`,
|
| 96 |
+
where i is the index of the annotation. `dataset['comparisons'][i]["metadata"]`
|
| 97 |
+
provides metadata to enable merging.
|
| 98 |
+
|
| 99 |
## Annotated datsets
|
| 100 |
|
| 101 |
We provide annotations for (subsets) of the following external datasets:
|