Skip to content
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

Support Qwen2-7b MLP in int4 and transpose_value_cache=True #11968

Merged
merged 3 commits into from
Sep 2, 2024
Merged

Support Qwen2-7b MLP in int4 and transpose_value_cache=True #11968

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
29d1836
Support Qwen2-7b mlp in int4
yangw1234 Aug 30, 2024
e2b8c1e
fix style
yangw1234 Aug 30, 2024
0851c35
rm verify size
plusbang Sep 2, 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
7 changes: 5 additions & 2 deletions python/llm/src/ipex_llm/transformers/npu_models/convert_mp.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,6 +65,11 @@ def optimize_llm_pre(model: torch.nn.Module, qtype):
model.llm.config.model_type = "llama"
model = model.llm

if model.config.model_type == "qwen2":
from ipex_llm.transformers.npu_models.qwen2_mp import split_mlp_down_proj
from ipex_llm.transformers.npu_models.qwen2_mp import split_mlp_forward
model.apply(split_mlp_down_proj)

# lm_head to cpu optimization
if cpu_lm_head:
# disable the optimization by default
Expand Down Expand Up @@ -134,8 +139,6 @@ def optimize_llm(
intra_pp = 2
if inter_pp is None:
inter_pp = 4 if model.config.intermediate_size == 18944 else 1
if model.config.intermediate_size == 18944:
transpose_value_cache = False

from ipex_llm.transformers.npu_models.qwen2_mp import gen_qwen2_fused_model_forward
from ipex_llm.transformers.npu_models.qwen2_mp import DecodeRunner, PrefillRunner
Expand Down
53 changes: 44 additions & 9 deletions python/llm/src/ipex_llm/transformers/npu_models/qwen2_mp.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,30 @@
from ipex_llm.transformers.npu_models.common import reshape_lm_head_input
from transformers.modeling_outputs import CausalLMOutputWithPast
from torch.nn import CrossEntropyLoss
from transformers.models.qwen2.modeling_qwen2 import Qwen2MLP


def split_mlp_down_proj(module: torch.nn.Module):
if isinstance(module, Qwen2MLP) and module.down_proj.in_features == 18944:
new_linear_0 = torch.nn.Linear(0, 0, bias=False)
new_weight_0 = torch.nn.Parameter(module.down_proj.weight[:, :9472], requires_grad=False)
new_linear_0.weight = new_weight_0
new_linear_0.in_features = new_weight_0.size(1)
new_linear_0.out_features = new_weight_0.size(0)
module.down_proj_0 = new_linear_0
new_linear_1 = torch.nn.Linear(0, 0, bias=False)
new_weight_1 = torch.nn.Parameter(module.down_proj.weight[:, 9472:], requires_grad=False)
new_linear_1.weight = new_weight_1
new_linear_1.in_features = new_weight_1.size(1)
new_linear_1.out_features = new_weight_1.size(0)
module.down_proj_1 = new_linear_1

del module.down_proj


def split_mlp_forward(self, x):
h = self.act_fn(self.gate_proj(x)) * self.up_proj(x)
return self.down_proj_0(h[:, :, :9472]) + self.down_proj_1(h[:, :, 9472:])


class LowBitQwenMultiDecoderlayer(LLMBaseNNFactory):
Expand Down Expand Up @@ -201,7 +225,7 @@ def __init__(
self.compile()
print("end compiling")

def mlp(self, hidden_states):
def mlp(self, hidden_states, seq_len):
mm1 = self.linear(
hidden_states, self.intermediate_size, self.hidden_size, bias=False, wt_dtype=self.dtype
)
Expand All @@ -211,9 +235,13 @@ def mlp(self, hidden_states):
mm1 = self.eltwise_mul(self.swish(mm1), mm2) # type: ignore[attr-defined]
if self.intermediate_size == 18944:
# for qwen2-7b
hidden_states = self.linear(
mm1, self.hidden_size, self.intermediate_size, bias=False, wt_dtype=np.int8
)
mm1_0 = self.slice(mm1, begin=[0, 0, 0], end=[1, seq_len, 9472])
mm1_1 = self.slice(mm1, begin=[0, 0, 9472], end=[1, seq_len, 18944])
hidden_states_0 = self.linear(mm1_0, self.hidden_size, 9472,
bias=False, wt_dtype=self.dtype)
hidden_states_1 = self.linear(mm1_1, self.hidden_size, 9472,
bias=False, wt_dtype=self.dtype)
hidden_states = hidden_states_0 + hidden_states_1
else:
hidden_states = self.linear(
mm1, self.hidden_size, self.intermediate_size, bias=False, wt_dtype=self.dtype
Expand Down Expand Up @@ -257,7 +285,7 @@ def build_decoder(
hidden_states = self.eltwise_add(residual, attn_output)
residual = hidden_states
hidden_states = self.layer_norm(hidden_states, post_attention_layernorm_weight)
hidden_states = self.mlp(hidden_states)
hidden_states = self.mlp(hidden_states, self.seq_len)
hidden_states = self.eltwise_add(residual, hidden_states)
hidden_states = self.convert_to_fp16(hidden_states)

Expand Down Expand Up @@ -343,9 +371,13 @@ def __init__(
)
self.backend_decoders.append(decoder)

offset = 0
for i in range(intra_stages):
start, end = self.layer_ranges[i]
self.backend_decoders[i].set_weights(self.op_id, op_parameters[start * 7:end * 7])
curr_linear_ops = len(self.backend_decoders[i].linear_ops)
curr_parameters = self.op_parameters[offset:offset + curr_linear_ops]
self.backend_decoders[i].set_weights(self.op_id, curr_parameters)
offset = offset + curr_linear_ops

def forward(
self,
Expand Down Expand Up @@ -543,7 +575,8 @@ def run_decode(
(attn_layer.o_proj.weight, attn_layer.o_proj.scale),
(mlp_layer.gate_proj.weight, mlp_layer.gate_proj.scale),
(mlp_layer.up_proj.weight, mlp_layer.up_proj.scale),
(mlp_layer.down_proj.weight, mlp_layer.down_proj.scale),
(mlp_layer.down_proj_0.weight, mlp_layer.down_proj_0.scale),
(mlp_layer.down_proj_1.weight, mlp_layer.down_proj_1.scale)
]

cached_cos = curr_layer.self_attn.rotary_emb.cos_cached.to(torch.float16)
Expand Down Expand Up @@ -814,6 +847,8 @@ def run_prefill(
transpose_value=transpose_value_cache
)
convert_forward(model, Qwen2Attention, qwen2_attention_forward)
from transformers.models.qwen2.modeling_qwen2 import Qwen2MLP
convert_forward(model, Qwen2MLP, split_mlp_forward)
deocderlayers = model.model.layers

while True:
Expand All @@ -836,7 +871,6 @@ def run_prefill(

hidden_states = layer_outputs[0]
next_decoder_cache = layer_outputs[1]

result_queue.put((hidden_states, next_decoder_cache))


Expand Down Expand Up @@ -1124,10 +1158,11 @@ def qwen2_attention_forward(
cos, sin = self.rotary_emb(value_states, seq_len=kv_seq_len)
query_states, key_states = apply_rotary_pos_emb(query_states, key_states,
cos, sin, position_ids)

cache_kwargs = {"max_seq_len": max_seq_len, "transpose": transpose_value, }

if past_key_value is not None:
if transpose_value:
value_states = value_states.transpose(-1, -2)
key_states, value_states = past_key_value.update(key_states, value_states,
self.layer_idx, cache_kwargs)

Expand Down
Loading