Update README.md
Browse files
README.md
CHANGED
@@ -65,13 +65,12 @@ language:
|
|
65 |
## Model description
|
66 |
The model is the result of further post-training [microsoft/Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct). This model is designed for high performance in various instruction-following tasks and complex interactions, including multi-turn function calling and detailed conversations.
|
67 |
|
68 |
-
| Model
|
69 |
-
|
70 |
-
| Phi-3 Mini 128k Instruct
|
71 |
-
| Rubra Enhanced Phi-3 Mini 128k Instruct (June)
|
72 |
-
| Phi-3 Mini 128k Instruct
|
73 |
-
| Rubra Enhanced Phi-3 Mini 128k Instruct (April)
|
74 |
-
|
75 |
* Commit `e2ecb24bd9dae689bb30dafcf13cbbc9dbddead5` is the last commit to have the April-based Phi-3 model. The latest in main is built off the June model
|
76 |
|
77 |
## Training Data
|
|
|
65 |
## Model description
|
66 |
The model is the result of further post-training [microsoft/Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct). This model is designed for high performance in various instruction-following tasks and complex interactions, including multi-turn function calling and detailed conversations.
|
67 |
|
68 |
+
| Model | Function Calling | MMLU | GPQA | GSM-8K | MATH | MT-bench | Win | Loss | Tie | Win Rate | Loss Rate | Adjusted Win Rate |
|
69 |
+
|----------------------------------------------|------------------|-------|-------|--------|-------|----------|-----|------|-----|----------|-----------|-------------------|
|
70 |
+
| Phi-3 Mini 128k Instruct (June) | - | 69.36 | 27.01 | 83.7 | 32.92 | - | - | - | - | - | - | - |
|
71 |
+
| Rubra Enhanced Phi-3 Mini 128k Instruct (June)| - | 67.87 | 29.69 | 79.45 | 30.80 | - | - | - | - | - | - | - |
|
72 |
+
| Phi-3 Mini 128k Instruct (April) | - | 68.17 | 25.90 | 80.44 | 28.12 | 7.92 | 51 | 45 | 64 | 0.31875 | 0.28125 | **0.51875** |
|
73 |
+
| Rubra Enhanced Phi-3 Mini 128k Instruct (April)| 65.71% | 66.66 | 29.24 | 74.09 | 26.84 | 7.45 | 45 | 51 | 64 | 0.28125 | 0.31875 | 0.48125 |
|
|
|
74 |
* Commit `e2ecb24bd9dae689bb30dafcf13cbbc9dbddead5` is the last commit to have the April-based Phi-3 model. The latest in main is built off the June model
|
75 |
|
76 |
## Training Data
|