Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/9743
Note: Links to docs will display an error until the docs builds have been completed. ❌ 71 New FailuresAs of commit 81abcf9 with merge base 90f0843 ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This PR needs a
|
|
@cccclai has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@billmguo has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
Sorry I missed one thing, the CLI argument of |
| uint8_t* ptr_in = k_cache_in_prefill[i]->mutable_data<uint8_t>() - pos; | ||
| // reset decode mode pointer since it will be updated again in update_prefill_to_kv | ||
| k_cache_in_kv[i]->set_data(ptr_in); | ||
| // in order not to override existent k_cache_out |
There was a problem hiding this comment.
Sorry, should be existent k_cache_in.
|
Yeah I print the result and it looks like following |
|
This PR is sufficient to unblock internal use case. The current main issue becomes the model quality...working internally to address it |
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
No description provided.