Update README.md
Browse files
README.md
CHANGED
|
@@ -10,6 +10,36 @@ pipeline_tag: text-generation
|
|
| 10 |
inference: false
|
| 11 |
quantized_by: Suparious
|
| 12 |
---
|
| 13 |
-
#
|
| 14 |
|
| 15 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
inference: false
|
| 11 |
quantized_by: Suparious
|
| 12 |
---
|
| 13 |
+
# mlabonne/Chimera-8B AWQ
|
| 14 |
|
| 15 |
+
- Model creator: [mlabonne](https://huggingface.co/mlabonne)
|
| 16 |
+
- Original model: [Chimera-8B](https://huggingface.co/mlabonne/Chimera-8B)
|
| 17 |
+
|
| 18 |
+
## Model Summary
|
| 19 |
+
|
| 20 |
+
Dare-ties merge method.
|
| 21 |
+
|
| 22 |
+
List of all models and merging path is coming soon.
|
| 23 |
+
|
| 24 |
+
Merging the "thick"est model weights from mistral models using amazing training methods like direct preference optimization (dpo) and reinforced learning.
|
| 25 |
+
|
| 26 |
+
I have spent countless hours studying the latest research papers, attending conferences, and networking with experts in the field. I experimented with different algorithms, tactics, fine-tuned hyperparameters, optimizers,
|
| 27 |
+
and optimized code until i achieved the best possible results.
|
| 28 |
+
|
| 29 |
+
Thank you openchat 3.5 for showing me the way.
|
| 30 |
+
|
| 31 |
+
Here is my contribution.
|
| 32 |
+
|
| 33 |
+
## Prompt Template
|
| 34 |
+
|
| 35 |
+
Replace {system} with your system prompt, and {prompt} with your prompt instruction.
|
| 36 |
+
|
| 37 |
+
```
|
| 38 |
+
### System:
|
| 39 |
+
{system}
|
| 40 |
+
|
| 41 |
+
### User:
|
| 42 |
+
{prompt}
|
| 43 |
+
|
| 44 |
+
### Assistant:
|
| 45 |
+
```
|