| | --- |
| | title: README |
| | emoji: 🦀 |
| | colorFrom: blue |
| | colorTo: purple |
| | sdk: static |
| | pinned: false |
| | thumbnail: >- |
| | https://cdn-uploads.huggingface.co/production/uploads/62a3bb1cd0d8c2c2169f0b88/SgosK-Yi9pWDKrEDEQ_bh.png |
| | --- |
| | |
| | # 🙋🏻♂️ Welcome to Data Tonic (Alignment Lab) |
| |
|
| | Just an organisation created to share unique reasoning models and their datasets. |
| |
|
| | ### Collection : Dark Thoughts |
| |
|
| | The `Dark Thoughts` [collection](https://huggingface.co/collections/DataTonic/dark-thoughts-66f83c21ee826b5daa943c8d) is a collection of alignment experiments to produce high quality thinking and reasoning datasets to improve `reasoning` ability in on device models. It's a data-to-demo collection of outcomes for producing simple reasoning and evaluation pipelines. |
| |
|
| | ### Collection : Toxic Agent - Climate Guard |
| |
|
| | The `Toxic Agent Climate Change Denial` [collection](https://huggingface.co/collections/DataTonic/toxic-agent-climate-change-denial-67adf94015909b844c953cbf) is approx. 84K row of climate change denial messages in 7 thematic categories. |
| |
|
| |
|
| | #### Common Information Model |
| |
|
| | The dataset retained has the information model: |
| |
|
| | {coming soon} |
| |
|
| | The model retained has the following prompt format / activation : |
| |
|
| | {prompt} |
| |
|
| | The embeddings model has the following format / activation : |
| |
|
| | {coming soon} |
| |
|
| | #### Use the Datasets |
| |
|
| | {coming soon} |
| |
|
| | #### Train a Model |
| |
|
| | {coming soon} |
| |
|
| | #### Use Our Model |
