Skip to content

Change default value of attn_temperature_tuning #37501

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 2 commits into from
Apr 15, 2025
Merged

Change default value of attn_temperature_tuning #37501

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fa84128
fix: change default value of `attn_temperature_tuning`
gmlwns2000 Apr 14, 2025
0a57b45
Merge branch 'main' into feat/fix-llama4-config
gmlwns2000 Apr 14, 2025
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: 4 additions & 2 deletions src/transformers/models/llama4/configuration_llama4.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -228,7 +228,9 @@ class Llama4TextConfig(PretrainedConfig):
no_rope_layer_interval (`int`, *optional*, defaults to 4): TODO
attention_chunk_size (`int`, *optional*, defaults to 8192):
<TODO>
attn_temperature_tuning (`int`, *optional*, defaults to 4): TODO
attn_temperature_tuning (`bool`, *optional*, defaults to `True`):
Whether to dynamically scale the attention temperature for each query token based on sequence length.
Recommended for long sequences (e.g., >32k tokens) to maintain stable output results.
floor_scale (`int`, *optional*, defaults to 8192): TODO
attn_scale (`int`, *optional*, defaults to 0.1): TODO
cache_implementation (`<fill_type>`, *optional*, defaults to `"hybrid"`): <fill_docstring>
Expand Down Expand Up @@ -291,7 +293,7 @@ def __init__(
no_rope_layers=None,
no_rope_layer_interval=4,
attention_chunk_size=8192,
attn_temperature_tuning=4,
attn_temperature_tuning=True,
floor_scale=8192,
attn_scale=0.1,
cache_implementation="hybrid",
Expand Down