Skip to content

[SpecDecoding] Update MLPSpeculator CI tests to use smaller model #6714

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jul 24, 2024
Merged

[SpecDecoding] Update MLPSpeculator CI tests to use smaller model #6714

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
79c204f
[SpecDecoding] Update MLPSpeculator CI tests to use smaller model
njhill Jul 24, 2024
81dceba
Fix lower max spec tokens for the new model
njhill Jul 24, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions tests/spec_decode/e2e/test_mlp_correctness.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,14 +24,14 @@
from .conftest import run_greedy_equality_correctness_test

# main model
MAIN_MODEL = "ibm-granite/granite-3b-code-instruct"
MAIN_MODEL = "JackFram/llama-160m"

# speculative model
SPEC_MODEL = "ibm-granite/granite-3b-code-instruct-accelerator"
SPEC_MODEL = "ibm-fms/llama-160m-accelerator"

# max. number of speculative tokens: this corresponds to
# n_predict in the config.json of the speculator model.
MAX_SPEC_TOKENS = 5
MAX_SPEC_TOKENS = 3

# precision
PRECISION = "float32"
Expand Down