dctanner commited on
Commit
d41604b
1 Parent(s): da8c055

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -8
README.md CHANGED
@@ -39,7 +39,7 @@ A filtered and curated dataset version of https://huggingface.co/datasets/databr
39
 
40
  ## Dataset Structure
41
 
42
- We structure the dataset using the format commonly used as input into [Hugging Face Chat Templates](https://huggingface.co/docs/transformers/chat_templating):
43
 
44
  ```
45
  [
@@ -52,17 +52,13 @@ We structure the dataset using the format commonly used as input into [Hugging F
52
 
53
  ### Source Data
54
 
55
- - **Source Dataset:** https://huggingface.co/datasets/databricks/databricks-dolly-15k
56
 
57
  #### Data Collection and Processing
58
 
59
- We started with https://huggingface.co/datasets/databricks/databricks-dolly-15k.
60
 
61
- Filtering and curation was done to remove conversations with:
62
- - Duplicate or very similar responses
63
- - Responses where the AI was actually responding like a person (present in this dataset as the responses are created by humans pretending to be an AI, and no everyone followed these instructions closely)
64
- - Profanity or inappropriate responses for an AI
65
- - Very short response lengths (often below 50 or 200 characters)
66
 
67
  # License
68
 
 
39
 
40
  ## Dataset Structure
41
 
42
+ We structure the dataset using the format commonly used as input into [Hugging Face Chat Templates](https://huggingface.co/docs/transformers/chat_templating). Where present, the context field text has been appending to the instruction in OpenAI style `Text: """..."""` format.
43
 
44
  ```
45
  [
 
52
 
53
  ### Source Data
54
 
55
+ - **Source Dataset:** https://huggingface.co/datasets/argilla/databricks-dolly-15k-curated-multilingual and https://huggingface.co/datasets/databricks/databricks-dolly-15k
56
 
57
  #### Data Collection and Processing
58
 
59
+ We started with https://huggingface.co/datasets/argilla/databricks-dolly-15k-curated-multilingual (en split only) which is a manually curated version of https://huggingface.co/datasets/databricks/databricks-dolly-15k.
60
 
61
+ As well as formatting to fit HF Chat style, we removed many duplicates based on the instruction text. This ensures the dataset is diverse and not repetitive.
 
 
 
 
62
 
63
  # License
64