Fix openai client logprobs calculation

[pan-x-c]

Description

As the title says

Checklist

Please check the following items before code is ready to be reviewed.

• Code has passed all tests
• Docstrings have been added/updated in Google Style
• Documentation has been updated
• Code is ready for review

[gemini-code-assist]

Summary of Changes

Hello @pan-x-c, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request addresses an issue related to the consistent calculation and capture of log probabilities from the OpenAI client, particularly when interaction history is being recorded. The changes ensure that log probabilities are always requested from the OpenAI API and that their configuration is properly propagated through the system. New test cases have been introduced to validate the correct behavior, ensuring that log probabilities are reliably present and correctly structured within the recorded experience history.

Highlights

• OpenAI Client Logprobs Enforcement: The core change ensures that logprobs=True is consistently passed to the OpenAI chat completions create method whenever history recording is enabled. This is achieved by wrapping the original create method with functools.partial.
• Logprobs Configuration Propagation: The pull request updates the TrinityChatModel initialization to correctly propagate top_logprobs settings from task rollout arguments to the underlying chat model's generation parameters.
• Enhanced Test Coverage: New assertions have been added to the vllm_test.py file to validate that extracted experience objects correctly contain log probabilities, and that their lengths align with the token counts and prompt lengths.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request aims to fix the logprobs calculation for the OpenAI client by ensuring logprobs=True is always set when history is enabled. The changes in trinity/common/models/model.py use functools.partial for this, and corresponding tests are updated. While this approach works for the cases added in the tests, it has a flaw where an explicit logprobs=False from a caller would override the intended behavior. My review includes suggestions to make this more robust. I've also pointed out some opportunities to improve maintainability by removing code duplication in tests and avoiding magic numbers.

[pan-x-c]

/unittest-module-common

[github-actions]

Summary

Tests 📝	Passed ✅	Failed ❌	Skipped ⏭️	Other ❓	Flaky 🍂	Duration ⏱️
33	33	0	0	0	0	346ms

Tests

Test Name	Status	Flaky	Duration
tests/common/config_test.py::TestConfig::test_all_examples_are_valid	✅		32ms
tests/common/config_test.py::TestConfig::test_config_flatten	✅		1ms
tests/common/config_test.py::TestConfig::test_continue_from_checkpoint_is_valid	✅		1ms
tests/common/config_test.py::TestConfig::test_default_workflow	✅		1ms
tests/common/config_test.py::TestConfig::test_load_default_config	✅		28ms
tests/common/config_test.py::TestConfig::test_max_token_len_per_gpu_set_correctly	✅		1ms
tests/common/config_test.py::TestConfig::test_update_config_from_ray_cluster	✅		1ms
tests/common/experience_test.py::TestEID::test_eid_properties	✅		1ms
tests/common/experience_test.py::TestExperience::test_action_mask_and_logprobs_type	✅		1ms
tests/common/experience_test.py::TestExperience::test_assertions	✅		1ms
tests/common/experience_test.py::TestExperience::test_dpo_experience	✅		1ms
tests/common/experience_test.py::TestExperience::test_gather	✅		1ms
tests/common/experience_test.py::TestExperience::test_hf_datasets_conversion	✅		1ms
tests/common/experience_test.py::TestExperience::test_multi_turn_experience	✅		1ms
tests/common/experience_test.py::TestExperience::test_serialize_deserialize	✅		1ms
tests/common/experience_test.py::TestExperience::test_single_turn_experience	✅		1ms
tests/common/experience_test.py::TestExperience::test_to_dict	✅		1ms
tests/common/experience_test.py::TestExperienceConversion::test_batch_conversion	✅		1ms
tests/common/experience_test.py::TestExperienceConversion::test_dpo_experience_batch_conversion	✅		1ms
tests/common/experience_test.py::TestExperienceConversion::test_experience_model_experience_conversion	✅		1ms
tests/common/experience_test.py::TestExperienceConversion::test_gather_experiences_with_custom_fields	✅		1ms
tests/common/experience_test.py::TestExperienceConversion::test_multiturn_experience_batch_converstion	✅		1ms
tests/common/vllm_test.py::ModelWrapperTest_0::test_generate	✅		56ms
tests/common/vllm_test.py::ModelWrapperTest_1::test_generate	✅		36ms
tests/common/vllm_test.py::ModelWrapperTest_2::test_generate	✅		47ms
tests/common/vllm_test.py::TestModelLen_0::test_model_len	✅		20ms
tests/common/vllm_test.py::TestModelLen_1::test_model_len	✅		21ms
tests/common/vllm_test.py::TestAPIServer::test_api	✅		25ms
tests/common/vllm_test.py::TestAsyncAPIServer::test_api_async	✅		26ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask	✅		1ms
tests/common/vllm_test.py::TestTokenizer::test_action_mask_with_tools	✅		1ms
tests/common/vllm_test.py::TestAPIServerToolCall_0_deepseek_r1::test_api_tool_calls	✅		22ms
tests/common/vllm_test.py::TestAPIServerToolCall_1::test_api_tool_calls	✅		21ms

Github Test Reporter by CTRF 💚