Update README.md

README.md

@@ -3,11 +3,11 @@ license: apache-2.0
 inference: false 
 ---
 
-# DRAGON-MISTRAL-0.3
+# DRAGON-QWEN-7B
 
 <!-- Provide a quick summary of what the model is/does. -->
 
-dragon-mistral-0.3 is part of the dRAGon ("Delivering RAG On ...") model series, RAG-instruct trained on top of a Mistral 7b (0.3) base model.
+dragon-qwen-7b is part of the dRAGon ("Delivering RAG On ...") model series, RAG-instruct trained on top of a Qwen2 7b base model.
 
 DRAGON models have been fine-tuned with the specific objective of fact-based question-answering over complex business and legal documents with an emphasis on reducing hallucinations and providing short, clear answers for workflow automation.
 
@@ -15,21 +15,21 @@ DRAGON models have been fine-tuned with the specific objective of fact-based que
 ### Benchmark Tests  
 
 Evaluated against the benchmark test:   [RAG-Instruct-Benchmark-Tester](https://www.huggingface.co/datasets/llmware/rag_instruct_benchmark_tester)  
-1 Test Run (with temperature = 0.0 and sample = False) with 1 point for correct answer, 0.5 point for partial correct or blank / NF, 0.0 points for incorrect, and -1 points for hallucinations.  
-
---**Accuracy Score**:  **99.5** correct out of 100  
---Not Found Classification:  95.0%  
---Boolean:  82.5%  
---Math/Logic:  67.5%  
---Complex Questions (1-5):  4 (Above Average - multiple-choice, causal)  
---Summarization Quality (1-5):  4 (Above Average)  
+1 Test Run with sample=False & temperature=0.0 (deterministic output) - 1 point for correct answer, 0.5 point for partial correct or blank / NF, 0.0 points for incorrect, and -1 points for hallucinations.  
+
+--**Accuracy Score**:  **99.0** correct out of 100  
+--Not Found Classification:  85.0%  
+--Boolean:  100.0%  
+--Math/Logic:  92.5%  
+--Complex Questions (1-5):  5 (Best in Class)  
+--Summarization Quality (1-5):  3 (Average)  
 --Hallucinations:  No hallucinations observed in test runs.  
 
-For test run results (and good indicator of target use cases), please see the files ("core_rag_test" and "answer_sheet" in this repo).  
+For test run results (and good indicator of target use cases), please see the files ("core_rag_test" and "answer_sheet" in this repo).
 
-Please note that these test results were achieved using the 4_K_M quantized version of this model - [dragon-mistral-0.3-gguf](https://www.huggingface.co/llmware/dragon-mistral-0.3-gguf).  
+Please note that these test results were achieved using the 4_K_M quantized version of this model - [dragon-qwen-7b-gguf](https://www.huggingface.co/llmware/dragon-qwen-7b-gguf).  
 
-Note: compare results with [dragon-mistral-7b](https://www.huggingface.co/llmware/dragon-mistral-7b-v0).  
+Note: compare results with [dragon-mistral-0.3-gguf](https://www.huggingface.co/llmware/dragon-mistral-0.3-gguf).  
 
 
 ### Model Description
@@ -37,10 +37,10 @@ Note: compare results with [dragon-mistral-7b](https://www.huggingface.co/llmwar
 <!-- Provide a longer summary of what this model is. -->
 
 - **Developed by:** llmware
-- **Model type:** mistral 
+- **Model type:** Qwen
 - **Language(s) (NLP):** English
 - **License:** Apache 2.0
-- **Finetuned from model:** Mistral-7b-base-0.3
+- **Finetuned from model:** Qwen2-7b-base
 
 
 ### Direct Use
@@ -66,8 +66,8 @@ Any model can provide inaccurate or incomplete information, and should be used i
 The fastest way to get started with dRAGon is through direct import in transformers:
 
     from transformers import AutoTokenizer, AutoModelForCausalLM  
-    tokenizer = AutoTokenizer.from_pretrained("dragon-mistral-0.3")  
-    model = AutoModelForCausalLM.from_pretrained("dragon-mistral-0.3")  
+    tokenizer = AutoTokenizer.from_pretrained("dragon-qwen-7b")  
+    model = AutoModelForCausalLM.from_pretrained("dragon-qwen-7b")  
 
 Please refer to the generation_test .py files in the Files repository, which includes 200 samples and script to test the model.  The **generation_test_llmware_script.py** includes built-in llmware capabilities for fact-checking, as well as easy integration with document parsing and actual retrieval to swap out the test set for RAG workflow consisting of business documents.  
 
@@ -93,15 +93,15 @@ If you are using a HuggingFace generation script:
     inputs = tokenizer(new_prompt, return_tensors="pt")  
     start_of_output = len(inputs.input_ids[0])
 
-    #   temperature: set at 0.3 for consistency of output
+    #   temperature: set at 0.0 for consistency of output
     #   max_new_tokens:  set at 100 - may prematurely stop a few of the summaries
 
     outputs = model.generate(
             inputs.input_ids.to(device),
             eos_token_id=tokenizer.eos_token_id,
             pad_token_id=tokenizer.eos_token_id,
-            do_sample=True,
-            temperature=0.3,
+            do_sample=False,
+            temperature=0.0,
             max_new_tokens=100,
             )