Update README.md

README.md

@@ -12,13 +12,7 @@ pipeline_tag: text-generation
 
 # tonyzhao123/dummy_llama4
 
-Llama 4 for small size EP debug and dist
-
-## Model Details
-
-- **Base Model**: meta-llama/Llama-3.2-1B
-- **Training Method**: Knowledge Distillation + Supervised Fine-tuning
-- **Dataset**: Custom KD Dataset (1000 samples)
+Dummy Llama 4 for small size EP debug and dist
 
 ## Usage
 
@@ -59,25 +53,6 @@ response = tokenizer.decode(outputs[0], skip_special_tokens=True)
 print(response)
 ```
 
-## Training Details
-
-- **Training Framework**: TRL (Transformers Reinforcement Learning)
-- **Optimizer**: AdamW
-- **Learning Rate**: 2e-5
-- **Batch Size**: 4
-- **Epochs**: 5
-- **Scheduler**: Cosine
-- **Precision**: FP16
-
-## Performance
-
-This model has been fine-tuned using knowledge distillation techniques to maintain performance while potentially reducing model size or improving specific capabilities.
-
-## Limitations
-
-- This is a fine-tuned model and may have inherited biases from the base model
-- Performance may vary on different types of tasks
-- Always evaluate the model on your specific use case
 
 ## Citation