Cannot Reproduce GPT4 Evaluations

[snova-guhae]

I get "ValueError: not enough values to unpack (expected 2, got 1)" when I try

func_signature, func_docstring = tool_conv.description.split(" - ", 1)

This is with any of the benchmarks. This is in the Benchmark_GPT4.ipynb file. My LangChain version is 0.2.6. Does this differ from the expected versioning? Thank you!

[ohsuz]

I faced the same error and solved it using langchain==0.1.0