Spaces:
Sleeping
Sleeping
Update layout/about.md
Browse files- layout/about.md +2 -0
layout/about.md
CHANGED
|
@@ -36,5 +36,7 @@ There are many opportunities to effectively scale this body of work.
|
|
| 36 |
- Using our [fine tuned model](https://huggingface.co/tykiww/llama3-8b-meetingQA) trained both on the alpaca dataset and a *noisy* synthetic dataset of over 3000 product, technical, and academic meetings.
|
| 37 |
- Updating our RAG retrieval methods: query more than meetings, change similarity search, embeddings, receive added context, account for the freshness of information, etc.
|
| 38 |
|
|
|
|
|
|
|
| 39 |
|
| 40 |
This demo is just a peek. More to come!
|
|
|
|
| 36 |
- Using our [fine tuned model](https://huggingface.co/tykiww/llama3-8b-meetingQA) trained both on the alpaca dataset and a *noisy* synthetic dataset of over 3000 product, technical, and academic meetings.
|
| 37 |
- Updating our RAG retrieval methods: query more than meetings, change similarity search, embeddings, receive added context, account for the freshness of information, etc.
|
| 38 |
|
| 39 |
+
note: There is an Out-of-memory issue with handling larger text. Currently working on chunking out batches on better sentiment groups prior to submitting data into the pipeline.
|
| 40 |
+
|
| 41 |
|
| 42 |
This demo is just a peek. More to come!
|