[Bugfix] Fix EAGLE vocab embedding construction for Llama 70B (#19033)
Signed-off-by: Benjamin Chislett <benjamin.chislett@centml.ai>
This commit is contained in:
committed by
GitHub
parent
c8134bea15
commit
3465b87ef8
@@ -5,11 +5,11 @@ import copy
|
||||
import itertools
|
||||
|
||||
import torch
|
||||
import triton
|
||||
from weight_shapes import WEIGHT_SHAPES
|
||||
|
||||
from vllm._custom_ops import cutlass_scaled_mm as vllm_scaled_mm
|
||||
from vllm._custom_ops import scaled_fp8_quant as vllm_scaled_fp8_quant
|
||||
from vllm.triton_utils import triton
|
||||
|
||||
|
||||
@triton.testing.perf_report(
|
||||
|
||||
Reference in New Issue
Block a user