bling-qwen-1.5b

BLING-QWEN-1.5B

<!-- Provide a quick summary of what the model is/does. -->

bling-qwen-1.5b is part of the BLING model series, RAG-instruct trained on top of a Qwen2 1.5b base model.

BLING 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.

Benchmark Tests

Evaluated against the benchmark test: RAG-Instruct-Benchmark-Tester
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: 93.5 correct out of 100
--Not Found Classification: 75.0%
--Boolean: 87.5%
--Math/Logic: 70%
--Complex Questions (1-5): 3 (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).

Please note that these test results were achieved using the 4_K_M quantized version of this model - bling-qwen-mini-tool.

Model Description

<!-- Provide a longer summary of what this model is. -->

• Developed by: llmware
• Model type: Qwen
• Language(s) (NLP): English
• License: Apache 2.0
• Finetuned from model: Qwen2-1.5b-base

Direct Use

<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->

BLING is designed for enterprise automation use cases, especially in knowledge-intensive industries, such as financial services, legal and regulatory industries with complex information sources.

BLING models have been trained for common RAG scenarios, specifically: question-answering, key-value extraction, and basic summarization as the core instruction types without the need for a lot of complex instruction verbiage - provide a text passage context, ask questions, and get clear fact-based responses.

Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

Any model can provide inaccurate or incomplete information, and should be used in conjunction with appropriate safeguards and fact-checking mechanisms.

How to Get Started with the Model

The fastest way to get started with dRAGon is through direct import in transformers:

from transformers import AutoTokenizer, AutoModelForCausalLM  
tokenizer = AutoTokenizer.from_pretrained("llmware/bling-qwen-1.5b")  
model = AutoModelForCausalLM.from_pretrained("llmware/bling-qwen-1.5b")  

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.

The dRAGon model was fine-tuned with a simple "<human> and <bot>" wrapper, so to get the best results, wrap inference entries as:

full_prompt = "<human>: " + my_prompt + "\n" + "<bot>:"

The BLING model was fine-tuned with closed-context samples, which assume generally that the prompt consists of two sub-parts:

1. Text Passage Context, and
2. Specific question or instruction based on the text passage

To get the best results, package "my_prompt" as follows:

my_prompt = {{text_passage}} + "\n" + {{question/instruction}}

If you are using a HuggingFace generation script:

# prepare prompt packaging used in fine-tuning process
new_prompt = "<human>: " + entries["context"] + "\n" + entries["query"] + "\n" + "<bot>:"

inputs = tokenizer(new_prompt, return_tensors="pt")  
start_of_output = len(inputs.input_ids[0])

#   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=False,
        temperature=0.0,
        max_new_tokens=100,
        )

output_only = tokenizer.decode(outputs[0][start_of_output:],skip_special_tokens=True)  

Model Card Contact

Darren Oberst & llmware team