Update README.md

README.md

@@ -7,7 +7,7 @@ This repository contains an Attention-Linked Memory augmented Qwen model (ALM-Qw
 
 *   **AttentionLinkedMemory (ALM)**: A custom PyTorch module for two-level attention-based retrieval from structured memory. (See `ALM.py`)
 *   **QwenGenerator**: Wraps a Hugging Face Qwen model (e.g., Qwen2.5-0.5B-Instruct or Qwen2.5-7B-Instruct) for text generation.
-*   **ALMQwenModel_HF**: The main class orchestrating the ALM retrieval and Qwen generation. (See `alm_qwen_hf.py`)
+*   **ALMQwenModel_HF**: The main class orchestrating the ALM retrieval and Qwen generation. (See `alm_qwen.py`)
 *   **Saved Weights & Config**:
     *   `alm_layer_state_dict.pth`: Trained weights for the ALM layer.
     *   `alm_qwen_hf_config.json`: Configuration for the `ALMQwenModel_HF`, including ALM parameters and paths to the Qwen components.
@@ -30,7 +30,7 @@ This repository contains an Attention-Linked Memory augmented Qwen model (ALM-Qw
 
 3.  **Load the model in Python**:
     ```python
-    from alm_qwen_hf import ALMQwenModel_HF # Make sure alm_qwen_hf.py and ALM.py are in your PYTHONPATH
+    from alm_qwen import ALMQwenModel_HF # Make sure alm_qwen_hf.py and ALM.py are in your PYTHONPATH
     import torch
 
     # Desired device
@@ -79,4 +79,3 @@ The ALM layer (`alm_layer_state_dict.pth`) might have been trained. The Qwen mod
 *   The `load_model` method in `alm_qwen_hf.py` handles the reconstruction of the composite model.
 
 ---
-*This README was auto-generated. Please update with more specific details about your model.*