llm-eval 0.5.7

v0.5.7

put handler.py output processor back in.

llm-eval 0.5.6

v0.5.6

removed output processor for handler.py

llm-eval 0.5.5

v0.5.5

fixing output processor.

llm-eval 0.5.4

Trying different approach to post-processing output.

llm-eval 0.5.3

Trying different approach to post-processing output.

llm-eval 0.5.2

v0.5.2

Working to simplify output.

llm-eval 0.5.1

updated transformers version dependency to support Phi-3.

llm-eval 0.5.0

fixed LLM Judge class. Added Jupyter notebook to evaluate judge results.

llm-eval 0.3.5

Fixes to evaluation function

llm-eval 0.4.4

randomize first and second participant to address LLM preference for first option.