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

Jump to bottom

[LLM] Fix Qwen2 #8584

Merged
merged 2 commits into from
Jun 12, 2024
Merged

[LLM] Fix Qwen2 #8584

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
29da31d
fix output_router_logits #8583
DrownFish19 Jun 12, 2024
29af5ea
fix with __future__
DrownFish19 Jun 12, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
19 changes: 6 additions & 13 deletions paddlenlp/transformers/qwen2/modeling.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,7 @@
# See the License for the specific language governing permissions and
# limitations under the License.
"""Paddle Qwen2 model."""
from __future__ import annotations

import math
import warnings
Expand Down Expand Up @@ -187,11 +188,11 @@ def scaled_dot_product_attention(
else:
# [ bz, seqlen, nhead, head_dim] -> [bs, nhead, seq_len, head_dim]
query_states = paddle.transpose(query_states, [0, 2, 1, 3])
# merge with the next tranpose
# merge with the next transpose
key_states = paddle.transpose(key_states, [0, 2, 1, 3])
value_states = paddle.transpose(value_states, [0, 2, 1, 3])

# matmul and devide by sqrt(head_dim)
# matmul and divide by sqrt(head_dim)
attn_weights = paddle.matmul(query_states / math.sqrt(head_dim), key_states.transpose([0, 1, 3, 2]))

if attn_weights.shape != [bsz, num_heads, q_len, kv_seq_len]:
Expand Down Expand Up @@ -1127,7 +1128,7 @@ def forward(self, prediction_scores, masked_lm_labels):
if self.enable_parallel_cross_entropy:
if prediction_scores.shape[-1] == self.config.vocab_size:
warnings.warn(
f"enable_parallel_cross_entropy, the vocab_size should be splited: {prediction_scores.shape[-1]}, {self.config.vocab_size}"
f"enable_parallel_cross_entropy, the vocab_size should be splitted: {prediction_scores.shape[-1]}, {self.config.vocab_size}"
)
self.loss_func = paddle.nn.CrossEntropyLoss(reduction="none", ignore_index=self.ignore_index)

Expand Down Expand Up @@ -1202,14 +1203,7 @@ def get_decoder(self):
return self.qwen2

def prepare_inputs_for_generation(
self,
input_ids,
use_cache=False,
past_key_values=None,
attention_mask=None,
inputs_embeds=None,
output_router_logits=False,
**kwargs
self, input_ids, use_cache=False, past_key_values=None, attention_mask=None, inputs_embeds=None, **kwargs
):
batch_size, seq_length = input_ids.shape
position_ids = kwargs.get("position_ids", paddle.arange(seq_length).expand((batch_size, seq_length)))
Expand All @@ -1230,7 +1224,6 @@ def prepare_inputs_for_generation(
"past_key_values": past_key_values,
"use_cache": use_cache,
"attention_mask": attention_mask,
"output_router_logits": output_router_logits,
}
)
return model_inputs
Expand Down Expand Up @@ -1325,7 +1318,7 @@ def forward(
hidden_states = outputs[0]

# if labels is None，means we need full output, instead of tensor_parallel_output
# tensor_parallel_output is togather with ParallelCrossEntropy
# tensor_parallel_output is together with ParallelCrossEntropy
tensor_parallel_output = (
self.config.tensor_parallel_output and labels is not None and self.config.tensor_parallel_degree > 1
)
Expand Down