b8dd1d9a4b2bb665f3f9a4fc901f7cf89025519e734b18f41382a4a324aa163c
Browse files
README.md
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
license_name: yi-license
|
| 4 |
+
license_link: https://huggingface.co/01-ai/Yi-34B-200K/blob/main/LICENSE
|
| 5 |
+
base_model: jondurbin/bagel-34b-v0.2
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
|
| 13 |
+
This model is a finetune of jondurbin's excellent [bagel](https://huggingface.co/jondurbin/bagel-34b-v0.2) model.
|
| 14 |
+
It has been trained with new datasets and a new technique, which we will share to the community soon.
|
| 15 |
+
This model has not utilised any form of merging.
|
| 16 |
+
|
| 17 |
+
### Evaluation Results
|
| 18 |
+
|
| 19 |
+
| Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8K |
|
| 20 |
+
| --- | --- | --- | --- | --- | --- | --- |
|
| 21 |
+
| 77.29 | 74.23 | 86.76 | 76.66 | 70.22 | 83.66 | 72.18 |
|
| 22 |
+
|
| 23 |
+
### Contamination Results
|
| 24 |
+
|
| 25 |
+
With reference model jondurbin/bagel-34b-v0.2:
|
| 26 |
+
|
| 27 |
+
| ARC | TruthfulQA | GSM8K |
|
| 28 |
+
| --- | --- | --- |
|
| 29 |
+
| 0.08| 0.38| 0.88|
|
| 30 |
+
|
| 31 |
+
***
|
| 32 |
+
|
| 33 |
+
Vanilla Quantization by [nold](https://huggingface.co/nold), Original Model [abacusai/Smaug-34B-v0.1](https://huggingface.co/abacusai/Smaug-34B-v0.1). Created using [llm-quantizer](https://github.com/Nold360/llm-quantizer) Pipeline - 465d7970507dcaac4cb50221157a68c840965774
|