[Feat] Heterogeneous Code Part 1: Add Model and Module Code for Chameleon Lumina#377
Open
zhhsplendid wants to merge 5 commits intoInternLM:developfrom
Open
[Feat] Heterogeneous Code Part 1: Add Model and Module Code for Chameleon Lumina#377zhhsplendid wants to merge 5 commits intoInternLM:developfrom
zhhsplendid wants to merge 5 commits intoInternLM:developfrom
Conversation
| hidden_states = self.norm(hidden_states) | ||
|
|
||
| if hasattr(self, "output"): | ||
| hidden_states = self.output(hidden_states).float() |
Collaborator
There was a problem hiding this comment.
这里应该不需要把输出额外转fp32类型，我们NaiveAMPModel里有一个参数output_to_fp32来负责做这件事
| k_norm_out = self.k_norm(k_all) | ||
| k = split_forward_gather_backward(k_norm_out, ParallelMode.TENSOR, dim=-2) | ||
|
|
||
| v = rearrange(v, "b s (h d) -> b s h d", d=self.head_dim) |
Collaborator
There was a problem hiding this comment.
qk_norm这块可能要额外区分一下，如果is_using_isp()wp并行算法的话，就不需要走gather的逻辑了，因为isp算法forward时权重是完整的，算出来的qkv head也是完整的，不需要gather
| k_norm_out = self.k_norm(k_all) | ||
| k = split_forward_gather_backward(k_norm_out, ParallelMode.TENSOR, dim=-2) | ||
|
|
||
| v = rearrange(v, "b s (h d) -> b s h d", d=self.head_dim) |
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.
Motivation
This PR and next PR series will be the code adding heterogeneous support and Chameleon model for InternEvo.
We plan to merge those PRs:
This PR is the first one: adding Chameleon Model code.
Modification
As described above, this PR is the first of a series of PRs: Adding Chameleon model code.
BC-breaking (Optional)
None
Use cases (Optional)
We will add use case after the configuration PR so that we can show the training use case.
Checklist
Before PR:
After PR: