Update README.md

README.md

@@ -3,7 +3,7 @@ language: en
 license: apache-2.0
 ---
 
-# Shears Model Card: shears-mpt-7b-50-gsm8k-super
+# Shears Model Card: shears-mpt-7b-50-gsm8k-super-adapter
 
 The super-adapter fine-tuned on sparsified [MPT-7B](https://huggingface.co/mosaicml/mpt-7b) with GSM8K datasets using Shears.
 
@@ -13,11 +13,11 @@ The release of the super-network is to facilitate users to apply their own searc
 
 ### Information
 
-- **Model name:** shears-mpt-7b-50-gsm8k-super
+- **Model name:** shears-mpt-7b-50-gsm8k-super-adapter
 - **Base model:** [IntelLabs/shears-mpt-7b-50-base](https://huggingface.co/IntelLabs/shears-mpt-7b-50-base)
 - **Sparsity:** 50%
 - **Subnetwork version:** Super
-- **NNCF Configuration:** [nncf_shears_mpt.json](https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning/tree/main/Shears/nncf_config/gsm8k/nncf_shears_mpt.json)
+- **NNCF Configuration:** [nncf_shears_mpt.json](https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning/blob/main/Shears/nncf_config/nncf_shears_mpt.json)
 
 ### Adapter Configuration
 
@@ -59,7 +59,7 @@ def generate_prompt(instruction):
                     """
 
 base_model = AutoModelForCausalLM.from_pretrained("IntelLabs/shears-mpt-7b-50-base", trust_remote_code=True)
-model = PeftModel.from_pretrained(base_model, "IntelLabs/shears-mpt-7b-50-gsm8k-super")
+model = PeftModel.from_pretrained(base_model, "IntelLabs/shears-mpt-7b-50-gsm8k-super-adapter")
 model.eval()
 
 non_zero_params = sum([(param.data != 0).sum().item() for _, param in model.named_parameters()])