Update README.md
Browse files
README.md
CHANGED
|
@@ -9,17 +9,7 @@ license: apache-2.0
|
|
| 9 |
# LexiLumin-7B
|
| 10 |
|
| 11 |
LexiLumin-7B is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
|
| 12 |
-
|
| 13 |
-
# 🏆 Open LLM Leaderboard Evaluation Results
|
| 14 |
-
| Metric |Value|
|
| 15 |
-
|---------------------------------|----:|
|
| 16 |
-
|Avg. |75.72|
|
| 17 |
-
|AI2 Reasoning Challenge (25-Shot)|72.70|
|
| 18 |
-
|HellaSwag (10-Shot) |88.28|
|
| 19 |
-
|MMLU (5-Shot) |65.08|
|
| 20 |
-
|TruthfulQA (0-shot) |73.10|
|
| 21 |
-
|Winogrande (5-shot) |83.27|
|
| 22 |
-
|GSM8k (5-shot) |71.87|
|
| 23 |
|
| 24 |
## 🧩 Configuration
|
| 25 |
|
|
@@ -33,6 +23,16 @@ merge_method: model_stock
|
|
| 33 |
base_model: CultriX/MonaTrix-v4-7B-DPO
|
| 34 |
dtype: bfloat16
|
| 35 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
|
| 37 |
## 💻 Usage
|
| 38 |
|
|
|
|
| 9 |
# LexiLumin-7B
|
| 10 |
|
| 11 |
LexiLumin-7B is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
|
| 12 |
+
This model excels in roleplaying and storytelling.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
|
| 14 |
## 🧩 Configuration
|
| 15 |
|
|
|
|
| 23 |
base_model: CultriX/MonaTrix-v4-7B-DPO
|
| 24 |
dtype: bfloat16
|
| 25 |
```
|
| 26 |
+
# 🏆 Open LLM Leaderboard Evaluation Results
|
| 27 |
+
| Metric |Value|
|
| 28 |
+
|---------------------------------|----:|
|
| 29 |
+
|Avg. |75.72|
|
| 30 |
+
|AI2 Reasoning Challenge (25-Shot)|72.70|
|
| 31 |
+
|HellaSwag (10-Shot) |88.28|
|
| 32 |
+
|MMLU (5-Shot) |65.08|
|
| 33 |
+
|TruthfulQA (0-shot) |73.10|
|
| 34 |
+
|Winogrande (5-shot) |83.27|
|
| 35 |
+
|GSM8k (5-shot) |71.87|
|
| 36 |
|
| 37 |
## 💻 Usage
|
| 38 |
|