Update README.md

README.md

@@ -14,23 +14,26 @@ slim-sentiment has been fine-tuned for **sentiment analysis** function calls, ge
 `{"sentiment": ["positive"]}`
 
 
-SLIM models aspire re-imagine traditional 'hard-coded' classifiers through the use of function calls, to provide a flexible natural language generative model that can be used as decision gates and processing steps in a complex LLM-based automation workflow.  
+SLIM models are designed to provide a flexible natural language generative model that can be used for decision gates and processing steps in a complex LLM-based automation workflow.  
 
 Each slim model has a 'quantized tool' version, e.g.,  [**'slim-sentiment-tool'**](https://huggingface.co/llmware/slim-sentiment-tool).  
 
 
 ## Prompt format:
 
+`function = "classify"`  
+`params = "sentiment"`  
+
 `"<human> " + {text} + "\n" + `  
 
-`"<{function}> " + {keys} + "</{function}>"`  
+`"<{function}> " + {params} + "</{function}>"`  
 
 `+ "/n<bot>:" `
 
 
   
 <details>
-<summary><b> Transformers Script </b> </summary>
+<summary>Transformers Script </summary>
 
     model = AutoModelForCausalLM.from_pretrained("llmware/slim-sentiment")
     tokenizer = AutoTokenizer.from_pretrained("llmware/slim-sentiment")
@@ -72,7 +75,7 @@ Each slim model has a 'quantized tool' version, e.g.,  [**'slim-sentiment-tool'*
 
 
     
-<summary><b>Using as Function Call in LLMWare</b></summary>
+<summary>Using as Function Call in LLMWare</summary>
 
     from llmware.models import ModelCatalog
     slim_model = ModelCatalog().load_model("llmware/slim-sentiment")