How to set "verbose=True" for models and chains in LangChain v0.3?

[phoenixcoder-ed]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

synopsis_chain = prompt_template_synopsis | llm
    review_chain = prompt_template_review | llm 
    response = sequential_chain.invoke({"title": title})

Description

In langchain v0.1, I could set the verbose value to True in the LLMChain constructor to view the execution process, but after upgrading to v0.3 or even v0.2, I was prompted to use |, but after modifying, how do I set verbose? According to the official documentation, langchain.globals.set_verbose(True) was found to be ineffective.

System Info

python=3.10
langchain=0.3.10

[feijoes]

Perhaps you can achieve this by enabling debug mode:

import langchain

# Enable debug mode
langchain.debug = True

Alternatively, you could try setting verbose=True in prompt_template_synopsis, prompt_template_review, and during LLM creation.

[phoenixcoder-ed]

Thank you, very useful advice. I found the global and local debugging instructions in documentation of langchain based on your suggestion. It explains how to set local debugging with langchain.globals.set_debug(True).