Unverified Commit feecba09 authored by rasmith's avatar rasmith Committed by GitHub
Browse files

[CI/Build][AMD] Use float16 in test_reset_prefix_cache_e2e to avoid accuracy issues (#29997)


Signed-off-by: default avatarRandall Smith <ransmith@amd.com>
Co-authored-by: default avatarRandall Smith <ransmith@amd.com>
parent 6038b1b0
......@@ -21,6 +21,7 @@ def test_reset_prefix_cache_e2e(monkeypatch):
max_num_batched_tokens=32,
max_model_len=2048,
compilation_config={"mode": 0},
dtype="float16",
)
engine = LLMEngine.from_engine_args(engine_args)
sampling_params = SamplingParams(
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment