feat: PoC for Multi Turn agent loop#1023
Draft
bxyu-nvidia wants to merge 12 commits intocwing/multi-turn-agentfrom
Draft
feat: PoC for Multi Turn agent loop#1023bxyu-nvidia wants to merge 12 commits intocwing/multi-turn-agentfrom
bxyu-nvidia wants to merge 12 commits intocwing/multi-turn-agentfrom
Conversation
Provides information and background on how llm-as-a-judge works, when to use it, and a brief walkthrough. --------- Signed-off-by: Frankie Siino <fsiino@nvidia.com>
Adds a new resources server for RDKit-based chemistry verification tasks. Includes sandbox launcher for sandbox code execution, YAML config, example data, and tests. Made with [Cursor](https://cursor.com) --------- Signed-off-by: Dane Corneil <dcorneil@nvidia.com> Co-authored-by: Christian Munley <cmunley@nvidia.com>
mmlu-pro: https://wandb.ai/nvidia/fsiino-gym-dev/runs/mi6p08ns 83.90957446808511 mmlu-prox: https://wandb.ai/nvidia/fsiino-gym-dev/runs/fxhaochj 70.33903109674858 --------- Signed-off-by: Frankie Siino <fsiino@nvidia.com>
1. Multi-process benchmark preparation 2. Various benchmark data preparation refactors 3. Refactor Nemotron 3 Ultra to be easier to use 4. Delete key directive 5. Improve dummy model config handling 6. Print config yaml when erroring on almost servers 7. Try fix progress bar print with tqdm (not successful) 8. Improve broken pipe print and handling behavior 9. Adopt Cascade eval numpy.isclose for float comparison 10. LocalVLLMModel accepts py_executable 11. LocalVLLMModelProxy --------- Signed-off-by: Brian Yu <bxyu@nvidia.com>
… modes (#1003) - Added a new NVARC resource server that supports two agent modes: transductive (outputs grid directly) and inductive (outputs Python code). - Implemented necessary configurations and request/response models for both modes. - Included a subprocess sandbox for executing Python code safely. - Added example datasets and a .gitignore for data files. - Comprehensive unit tests for grid parsing and code execution. Signed-off-by: Elad Sarafian <esarafian@nvidia.com> dataset: https://gitlab-master.nvidia.com/fsoares/post-training-data-processing/-/issues/103 Reopening #989 for @esarafian due to branch renaming. --------- Signed-off-by: Elad Sarafian <esarafian@nvidia.com> Co-authored-by: Elad Sarafian <esarafian@nvidia.com>
Signed-off-by: Brian Yu <bxyu@nvidia.com>
Signed-off-by: Brian Yu <bxyu@nvidia.com>
|
Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
Signed-off-by: Brian Yu <bxyu@nvidia.com>
Signed-off-by: Brian Yu <bxyu@nvidia.com>
Signed-off-by: Brian Yu <bxyu@nvidia.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.