Update README.md
Browse files
README.md
CHANGED
|
@@ -18,10 +18,6 @@ tags:
|
|
| 18 |
- Reasoning
|
| 19 |
---
|
| 20 |
|
| 21 |
-
|
| 22 |
-
<img src="https://www.German-RAG.ai/wp-content/uploads/2024/12/German-RAG-ICON-TO-WORDLOGO-Animation_Loop-small-ezgif.com-video-to-gif-converter.gif" alt="German-RAG Logo" width="400" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
|
| 23 |
-
|
| 24 |
-
|
| 25 |
# German-RAG-LLAMA-3.1-8B-SFT-HESSIAN-AI
|
| 26 |
|
| 27 |
<!-- Provide a quick summary of what the model is/does. -->
|
|
@@ -200,6 +196,7 @@ It's important to note that the actual power consumption may vary depending on t
|
|
| 200 |
| Model | GPU Type | Power Consumption From GPUs |
|
| 201 |
|----------------|---------------------|-----------------------------|
|
| 202 |
| German-RAG-LLAMA-SFT | A100 ([Hessian AI supercomputer](https://hessian.ai/de/)) | 0.02016 MWh |
|
|
|
|
| 203 |
## Bias, Risks, and Limitations
|
| 204 |
|
| 205 |
Like any base language model or fine-tuned model without safety filtering, it is relatively easy for a user to prompt these models to generate harmful and generally sensitive content.
|
|
@@ -209,12 +206,6 @@ Otherwise, many facts from German-RAG-LLAMA-SFT or any LLM will often not be tru
|
|
| 209 |
|
| 210 |
|
| 211 |
|
| 212 |
-
|
| 213 |
-
## Model Card Contact
|
| 214 |
-
|
| 215 |
-
|
| 216 |
-
For errors in this model card, please contact ([[email protected]](mailto:[email protected])).
|
| 217 |
-
|
| 218 |
## The German-RAG AI Team
|
| 219 |
[Marcel Rosiak](https://de.linkedin.com/in/marcel-rosiak)
|
| 220 |
[Soumya Paul](https://de.linkedin.com/in/soumya-paul-1636a68a)
|
|
|
|
| 18 |
- Reasoning
|
| 19 |
---
|
| 20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
# German-RAG-LLAMA-3.1-8B-SFT-HESSIAN-AI
|
| 22 |
|
| 23 |
<!-- Provide a quick summary of what the model is/does. -->
|
|
|
|
| 196 |
| Model | GPU Type | Power Consumption From GPUs |
|
| 197 |
|----------------|---------------------|-----------------------------|
|
| 198 |
| German-RAG-LLAMA-SFT | A100 ([Hessian AI supercomputer](https://hessian.ai/de/)) | 0.02016 MWh |
|
| 199 |
+
|
| 200 |
## Bias, Risks, and Limitations
|
| 201 |
|
| 202 |
Like any base language model or fine-tuned model without safety filtering, it is relatively easy for a user to prompt these models to generate harmful and generally sensitive content.
|
|
|
|
| 206 |
|
| 207 |
|
| 208 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 209 |
## The German-RAG AI Team
|
| 210 |
[Marcel Rosiak](https://de.linkedin.com/in/marcel-rosiak)
|
| 211 |
[Soumya Paul](https://de.linkedin.com/in/soumya-paul-1636a68a)
|