Support multiple prompts in the runner #9817
Merged
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/9817
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 55678ad with merge base 753a88e ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D72276104 |
cccclai
requested review from
chunit-quic,
haowhsu-quic,
shewu-quic and
winskuo-quic
haowhsu-quic
approved these changes
Apr 2, 2025
kirklandsign
approved these changes
Apr 2, 2025
cccclai
added a commit
to cccclai/executorch-1
that referenced
this pull request
Apr 2, 2025
Summary: As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
1e9a596 to
7d2dc2d
Compare
cccclai
added a commit
to cccclai/executorch-1
that referenced
this pull request
Apr 2, 2025
Summary: As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
7d2dc2d to
24b6ca4
Compare
cccclai
added a commit
to cccclai/executorch-1
that referenced
this pull request
Apr 2, 2025
Summary: As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
24b6ca4 to
e3ec4b5
Compare
|
This pull request was exported from Phabricator. Differential Revision: D72276104 |
cccclai
added a commit
to cccclai/executorch-1
that referenced
this pull request
Apr 2, 2025
Summary: Pull Request resolved: pytorch#9817 As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
e3ec4b5 to
d86ecc8
Compare
|
This pull request was exported from Phabricator. Differential Revision: D72276104 |
cccclai
added a commit
to cccclai/executorch-1
that referenced
this pull request
Apr 2, 2025
Summary: Pull Request resolved: pytorch#9817 As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
d86ecc8 to
b59f256
Compare
|
This pull request was exported from Phabricator. Differential Revision: D72276104 |
cccclai
added a commit
to cccclai/executorch-1
that referenced
this pull request
Apr 2, 2025
Summary: Pull Request resolved: pytorch#9817 As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
b59f256 to
8a19c85
Compare
cccclai
added a commit
to cccclai/executorch-1
that referenced
this pull request
Apr 2, 2025
Summary: As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
8a19c85 to
28292d0
Compare
|
This pull request was exported from Phabricator. Differential Revision: D72276104 |
cccclai
added a commit
to cccclai/executorch-1
that referenced
this pull request
Apr 2, 2025
Summary: As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
28292d0 to
92ed635
Compare
|
This pull request was exported from Phabricator. Differential Revision: D72276104 |
cccclai
added a commit
to cccclai/executorch-1
that referenced
this pull request
Apr 2, 2025
Summary: Pull Request resolved: pytorch#9817 As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
92ed635 to
40a42c6
Compare
cccclai
added a commit
to cccclai/executorch-1
that referenced
this pull request
Apr 2, 2025
Summary: As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
40a42c6 to
43bd572
Compare
|
This pull request was exported from Phabricator. Differential Revision: D72276104 |
|
Just curious - will QNN be able to give response to multiple prompts? |
Summary: As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command: ``` ./qnn_llama3_2_runner --model_path hybrid_llama_qnn.pte --tokenizer_path tiktokenizer.bin --eval_mode 1 --prompt "Once upon a time" --prompt "girl named Lily." --prompt "her toys and her favorite toy was a big," --kv_updater "ShiftPointer" --logits_scale 0.1 --output_path output.txt --num_iters 1 ``` It will be hard to use any char as delimiter, so we use `--prompt` to explicitly mark a prompt and collect them together. Reviewed By: kirklandsign Differential Revision: D72276104
cccclai
force-pushed
the
export-D72276104
branch
from
43bd572 to
55678ad
Compare
|
This pull request was exported from Phabricator. Differential Revision: D72276104 |
Yeah, currently it's the same as a fresh prompt for each inference, but we will enable multi turn conversation and previous conversation will be part of the context |
kirklandsign
pushed a commit
that referenced
this pull request
Apr 11, 2025
Differential Revision: D72276104 Pull Request resolved: #9817
This was referenced Apr 14, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Summary:
As a preparation for the multiturn conversation, we can provide multiple prompts and execute them in sequence. Example command:
It will be hard to use any char as delimiter, so we use
--promptto explicitly mark a prompt and collect them together.Differential Revision: D72276104