Skip to content

Add FP8 quantization ignored_layers support in llama #6592

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

Closed
wants to merge 2 commits into from
Closed

Add FP8 quantization ignored_layers support in llama #6592

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c6eb095
add fp8 ignore layer support in llama
cli99 Jul 19, 2024
d2f2117
fix format
cli99 Jul 19, 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: 5 additions & 1 deletion vllm/model_executor/layers/quantization/fp8.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,7 @@ def __init__(
self,
is_checkpoint_fp8_serialized: bool = False,
activation_scheme: str = "dynamic",
ignored_layers: Optional[List[str]] = None,
) -> None:
self.is_checkpoint_fp8_serialized = is_checkpoint_fp8_serialized
if is_checkpoint_fp8_serialized:
Expand All @@ -41,6 +42,7 @@ def __init__(
raise ValueError(
f"Unsupported activation scheme {activation_scheme}")
self.activation_scheme = activation_scheme
self.ignored_layers = ignored_layers or []

@classmethod
def get_name(cls) -> str:
Expand All @@ -63,8 +65,10 @@ def from_config(cls, config: Dict[str, Any]) -> "Fp8Config":
quant_method = cls.get_from_keys(config, ["quant_method"])
is_checkpoint_fp8_serialized = ("fp8" in quant_method)
activation_scheme = cls.get_from_keys(config, ["activation_scheme"])
ignored_layers = cls.get_from_keys(config, ["ignored_layers"])
return cls(is_checkpoint_fp8_serialized=is_checkpoint_fp8_serialized,
activation_scheme=activation_scheme)
activation_scheme=activation_scheme,
ignored_layers=ignored_layers)

def get_quant_method(
self, layer: torch.nn.Module) -> Optional["QuantizeMethodBase"]:
Expand Down
25 changes: 17 additions & 8 deletions vllm/model_executor/models/llama.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,17 +65,22 @@ def __init__(
prefix: str = "",
) -> None:
super().__init__()
ignored_gate_up_proj = all(
p in quant_config.ignored_layers
for p in [f"{prefix}.gate_proj", f"{prefix}.up_proj"])
self.gate_up_proj = MergedColumnParallelLinear(
input_size=hidden_size,
output_sizes=[intermediate_size] * 2,
bias=bias,
quant_config=quant_config,
quant_config=quant_config if not ignored_gate_up_proj else None,
prefix=f"{prefix}.gate_up_proj")
self.down_proj = RowParallelLinear(input_size=intermediate_size,
output_size=hidden_size,
bias=bias,
quant_config=quant_config,
prefix=f"{prefix}.down_proj")
ignore_down_proj = f"{prefix}.down_proj" in quant_config.ignored_layers
self.down_proj = RowParallelLinear(
input_size=intermediate_size,
output_size=hidden_size,
bias=bias,
quant_config=quant_config if not ignore_down_proj else None,
prefix=f"{prefix}.down_proj")
if hidden_act != "silu":
raise ValueError(f"Unsupported activation: {hidden_act}. "
"Only silu is supported for now.")
Expand Down Expand Up @@ -128,21 +133,25 @@ def __init__(
self.scaling = self.head_dim**-0.5
self.rope_theta = rope_theta
self.max_position_embeddings = max_position_embeddings
ignore_qkv_proj = all(
p in quant_config.ignored_layers for p in
[f"{prefix}.q_proj", f"{prefix}.k_proj", f"{prefix}.v_proj"])

self.qkv_proj = QKVParallelLinear(
hidden_size=hidden_size,
head_size=self.head_dim,
total_num_heads=self.total_num_heads,
total_num_kv_heads=self.total_num_kv_heads,
bias=bias,
quant_config=quant_config,
quant_config=quant_config if not ignore_qkv_proj else None,
prefix=f"{prefix}.qkv_proj",
)
ignore_o_proj = f"{prefix}.o_proj" in quant_config.ignored_layers
self.o_proj = RowParallelLinear(
input_size=self.total_num_heads * self.head_dim,
output_size=hidden_size,
bias=bias,
quant_config=quant_config,
quant_config=quant_config if not ignore_o_proj else None,
prefix=f"{prefix}.o_proj",
)

Expand Down