feature(xjy): Fixed the accumulate_steps, game_segment/weighted_total_loss bugs and refine prompts, compute_llm_prior, and SFT loss, and added cprofile functionality. #441
Conversation
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
…_llm_prior, and SFT loss
xiongjyu
commented
Nov 24, 2025
| llm_sft_loss = torch.tensor(0.0, device=self._cfg.device) | ||
| if self.llm_policy_cfg.enable_llm and self.llm_policy_cfg.enable_rft: | ||
| with self._profile_block(name="train_llm_rft"): | ||
| llm_rft_loss = self.compute_rft_loss( |
Collaborator
Author
There was a problem hiding this comment.
这个target_value，我理解应该是和 observation 一一对应的，没有错位吧？比如target_value第一个值代表obs中第一个状态下的结果
puyuan1996
reviewed
Nov 24, 2025
| sequence_log_probs = token_log_probs.sum(dim=-1) / (mask.sum(dim=-1) + 1e-8) | ||
|
|
||
| if self.llm_policy_cfg.rft_reward=='value': | ||
| rewards_tensor = torch.tensor(batch_values, device=self._cfg.device, dtype=torch.float32) |
Collaborator
There was a problem hiding this comment.
rewards_tensor 重命名为 advantage_tansor 吧？
…lect to cprofile.
…ed the REINFORCE-series loss computation.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.