Skip to content

Add eval-mode sandbox rollouts and trajectory logging#1848

Open
YanhuiDua wants to merge 1 commit into
InternLM:agentic_branchfrom
YanhuiDua:support_agentic_eval
Open

Add eval-mode sandbox rollouts and trajectory logging#1848
YanhuiDua wants to merge 1 commit into
InternLM:agentic_branchfrom
YanhuiDua:support_agentic_eval

Conversation

@YanhuiDua
Copy link
Copy Markdown
Collaborator

  • Add TB2 eval dataloader and eval AgentInSandboxLoop config
  • Disable token/logprob/routed-expert returns for eval inference
  • Preserve text-only eval responses and tokenized response length stats

@YanhuiDua YanhuiDua force-pushed the support_agentic_eval branch from b036eeb to c7ee753 Compare May 27, 2026 10:26
 - Add TB2 eval dataloader and eval AgentInSandboxLoop config
 - Disable token/logprob/routed-expert returns for eval inference
 - Preserve text-only eval responses and tokenized response length stats
 - Separate eval replay buffer from training replay buffer
 - Add regression coverage for text-only eval trajectory saves
@YanhuiDua YanhuiDua force-pushed the support_agentic_eval branch from c7ee753 to e31d4d4 Compare May 27, 2026 13:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant