Update README.md

README.md

@@ -22,11 +22,10 @@ The model, `Tevatron/dse-phi3-docmatix-v1.0`, is trained using the `Tevatron/doc
 
 ```python
 import torch
-from transformers import AutoProcessor, AutoModelForCausalLM, AutoConfig
+from transformers import AutoProcessor, AutoModelForCausalLM
 
-processor = AutoProcessor.from_pretrained('microsoft/Phi-3-vision-128k-instruct', trust_remote_code=True)
-config = AutoConfig.from_pretrained('microsoft/Phi-3-vision-128k-instruct', trust_remote_code=True, attn_implementation="flash_attention_2", torch_dtype=torch.bfloat16, use_cache=False)
-model = AutoModelForCausalLM.from_pretrained('Tevatron/dse-phi3-docmatix-v1.0', trust_remote_code=True, config=config, attn_implementation="flash_attention_2", torch_dtype=torch.bfloat16).to('cuda:0')
+processor = AutoProcessor.from_pretrained('Tevatron/dse-phi3-docmatix-v1.0', trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained('Tevatron/dse-phi3-docmatix-v1.0', trust_remote_code=True, attn_implementation="flash_attention_2", torch_dtype=torch.bfloat16, use_cache=False).to('cuda:0')
 
 def get_embedding(last_hidden_state: torch.Tensor, attention_mask: torch.Tensor) -> torch.Tensor:
     sequence_lengths = attention_mask.sum(dim=1) - 1