Skip to content

patch falcon h1 inference #2932

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

Merged
merged 1 commit into from
Jul 12, 2025
Merged

patch falcon h1 inference #2932

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
25 changes: 15 additions & 10 deletions unsloth/models/falcon_h1.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,6 +43,9 @@
from transformers.modeling_attn_mask_utils import (
_prepare_4d_causal_attention_mask_for_sdpa,
)
from transformers.utils import (
is_torchdynamo_compiling,
)
# For Pytorch 2.1.1
try:
from transformers.models.falcon_h1.modeling_falcon_h1 import (
Expand Down Expand Up @@ -519,7 +522,7 @@ def FalconH1Model_fast_forward_inference_custom(
attention_mask = attention_mask,
do_prefill = not hasattr(decoder_layer.self_attn, "paged_attention"),
)
attention_hidden_states = attention_hidden_states * decoder_layer.attention_out_multiplier
attention_hidden_states = attention_hidden_states * decoder_layer.attn_out_multiplier
mamba_hidden_states = decoder_layer.mamba(
hidden_states=X,
cache_params=present_key_value,
Expand Down Expand Up @@ -595,15 +598,17 @@ def _fast_prepare_inputs_for_generation(
input_ids = input_ids[:, -cache_position.shape[0] :]
elif input_ids.shape[1] != cache_position.shape[0]: # Default case (the "else", a no op, is Exception 2)
input_ids = input_ids[:, cache_position]
else:
past_key_values = FalconHybridMambaAttentionDynamicCache(
self.config,
input_ids.shape[0],
self.dtype,
devices=[
self.model.layers[i].mamba.conv1d.weight.device for i in range(self.config.num_hidden_layers)
],
)
pass
# TODO: Wire up Cache to work for inference.
# else:
# past_key_values = FalconHybridMambaAttentionDynamicCache(
# self.config,
# input_ids.shape[0],
# self.dtype,
# devices=[
# self.model.layers[i].mamba.conv1d.weight.device for i in range(self.config.num_hidden_layers)
# ],
# )

if attention_mask is not None and position_ids is None:
# create position_ids on the fly for batch generation
Expand Down