Update README.md

Fix training data distribution in model card

README.md

@@ -21,24 +21,22 @@ A math-focused fine-tuned version of SmolLM3-3B, optimized for step-by-step math
 
 ## Highlights
 
-🧮 **Math-First**: Trained on 15K high-quality math reasoning samples
+📚 **Math-First**: Trained on ~7K high-quality math and reasoning samples
 🧠 **Chain-of-Thought**: Supports `/think` mode for detailed reasoning
 ⚡ **Lightweight**: 3B parameters, runs on consumer GPUs
-
 ## Training Details
 
 | Parameter | Value |
 |-----------|-------|
 | Base Model | HuggingFaceTB/SmolLM3-3B |
 | Method | LoRA (r=16, alpha=32) |
-| Training Data | ~15K samples |
-| - OpenThoughts3 | 12,000 (math reasoning) |
-| - s1k | 835 (high-quality math) |
-| - Mixture of Thoughts | 2,000 (science reasoning) |
+| Training Data | ~7K samples |
+| - OpenThoughts3_1.2M_think | 5,000 (math reasoning) |
+| - s1k_1.1_think | ~1,000 (high-quality math) |
+| - smoltalk_everyday_convs | 1,000 (everyday reasoning) |
 | Epochs | 2 |
 | Learning Rate | 2e-4 (cosine) |
 | Effective Batch Size | 16 |
-
 ## Usage
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer