Update README.md
Browse files
README.md
CHANGED
|
@@ -93,6 +93,7 @@ print(outputs[0]["generated_text"][-1])
|
|
| 93 |
**Dataset:**
|
| 94 |
|
| 95 |
Dataset: MetaMathQA-50k
|
|
|
|
| 96 |
The MetaMathQA-50k subset of HuggingFaceTB/smoltalk was selected for fine-tuning due to its focus on mathematical reasoning, multi-step problem-solving, and logical inference. The dataset includes:
|
| 97 |
|
| 98 |
Algebraic problems
|
|
@@ -105,8 +106,11 @@ Logical deduction problems
|
|
| 105 |
Fine-tuning was conducted using the following configuration:
|
| 106 |
|
| 107 |
Learning Rate: 2e-4
|
|
|
|
| 108 |
Epochs: 1
|
|
|
|
| 109 |
Optimizer: AdamW
|
|
|
|
| 110 |
Framework: Unsloth
|
| 111 |
|
| 112 |
**License:**
|
|
|
|
| 93 |
**Dataset:**
|
| 94 |
|
| 95 |
Dataset: MetaMathQA-50k
|
| 96 |
+
|
| 97 |
The MetaMathQA-50k subset of HuggingFaceTB/smoltalk was selected for fine-tuning due to its focus on mathematical reasoning, multi-step problem-solving, and logical inference. The dataset includes:
|
| 98 |
|
| 99 |
Algebraic problems
|
|
|
|
| 106 |
Fine-tuning was conducted using the following configuration:
|
| 107 |
|
| 108 |
Learning Rate: 2e-4
|
| 109 |
+
|
| 110 |
Epochs: 1
|
| 111 |
+
|
| 112 |
Optimizer: AdamW
|
| 113 |
+
|
| 114 |
Framework: Unsloth
|
| 115 |
|
| 116 |
**License:**
|