Update README.md
Browse files
README.md
CHANGED
|
@@ -136,6 +136,19 @@ Phi-4 that has been tuned to be more advanced at reasoning.
|
|
| 136 |
|
| 137 |
Unlike other Parm models we had to optimize our fine tuning process to ensure accuracy while still being able to release this model. **Training loss: 0.443800**
|
| 138 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 139 |
the model uses this prompt format: (modified phi-4 prompt)
|
| 140 |
```
|
| 141 |
{{ if .System }}<|system|>
|
|
|
|
| 136 |
|
| 137 |
Unlike other Parm models we had to optimize our fine tuning process to ensure accuracy while still being able to release this model. **Training loss: 0.443800**
|
| 138 |
|
| 139 |
+
Beats qwen/qwq at MATH & MuSR & GPQA (MuSR being a reasoning benchmark)
|
| 140 |
+
Evaluation:
|
| 141 |
+
|
| 142 |
+
|
| 143 |
+
|
| 144 |
+
|
| 145 |
+
|
| 146 |
+
|
| 147 |
+
|
| 148 |
+
|
| 149 |
+
|
| 150 |
+
|
| 151 |
+
|
| 152 |
the model uses this prompt format: (modified phi-4 prompt)
|
| 153 |
```
|
| 154 |
{{ if .System }}<|system|>
|