Updated model card numbers
#3
by
sreemanti-abacus
- opened
README.md
CHANGED
@@ -79,5 +79,26 @@ print(outputs[0]["generated_text"][len(prompt):])
|
|
79 |
|
80 |
| Model | Code Generation | Code Execution |Test Output Prediction |
|
81 |
|-------------------------------------|-----------------|----------------|-----------------------|
|
82 |
-
| **Dracarys-Llama-3.1-70B-Instruct** |
|
83 |
-
| Meta-Llama-3.1-70B-Instruct |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
79 |
|
80 |
| Model | Code Generation | Code Execution |Test Output Prediction |
|
81 |
|-------------------------------------|-----------------|----------------|-----------------------|
|
82 |
+
| **Dracarys-Llama-3.1-70B-Instruct** | **33.34** | **48.329** | **49.90** |
|
83 |
+
| Meta-Llama-3.1-70B-Instruct | 32.23 | 48.768 | 41.40 |
|
84 |
+
|
85 |
+
## Breakdown of LiveCodeBench CodeGeneration
|
86 |
+
|
87 |
+
| Model | Easy | Medium | Hard |
|
88 |
+
|-------------------------------------|-----------------|----------------|-----------------------|
|
89 |
+
| **Dracarys-Llama-3.1-70B-Instruct** | **71.89** | 17.30 | **4.23** |
|
90 |
+
| Meta-Llama-3.1-70B-Instruct | 68.4 | 17.99 | 3.57 |
|
91 |
+
|
92 |
+
## Breakdown of LiveCodeBench TestOutputPrediction
|
93 |
+
|
94 |
+
| Model | Easy | Medium | Hard |
|
95 |
+
|-------------------------------------|-----------------|----------------|-----------------------|
|
96 |
+
| **Dracarys-Llama-3.1-70B-Instruct** | **60.88** | **44.53** | **39.30** |
|
97 |
+
| Meta-Llama-3.1-70B-Instruct | 51.22 | 35.91 | 34.30 |
|
98 |
+
|
99 |
+
## LiveBench
|
100 |
+
|
101 |
+
| Model | Global Average | Coding Average | Reasoning Average| Mathematics Average | Data Analysis Average | Language Average | IF Average |
|
102 |
+
|-------------------------------------|----------------|----------------|------------------|---------------------|-----------------------|------------------|-------------|
|
103 |
+
| **Dracarys-Llama-3.1-70B-Instruct** | **48.67** | **35.23** | **44.0** | **45.68** | 48 | 41.77 | 77.37 |
|
104 |
+
| Meta-Llama-3.1-70B-Instruct | 48.44 | 32.67 | 40.67 | 45.58 | 50.29 | 42.36 | 79.08 |
|