deepseekmirror/DeepSeek-VL2
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepSeek-VL2.git synced 2025-02-23 06:09:04 -05:00
Code Issues Actions Packages Projects Releases Wiki Activity

Update configuration_deepseek.py

Browse Source 
fix Typo   change  'gready' to ‘greedy’
This commit is contained in:
digger yu 2024-12-27 16:08:07 +08:00 committed by GitHub
parent a8341f36dd
commit d975d87778
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
deepseek_vl2/models/configuration_deepseek.py
View File

@ -34,7 +34,7 @@ class DeepseekV2Config(PretrainedConfig):
Number of routed experts, None means dense model. Number of routed experts, None means dense model.
routed_scaling_factor (`float`, *optional*, defaults to 1.0): routed_scaling_factor (`float`, *optional*, defaults to 1.0):
Scaling factor or routed experts. Scaling factor or routed experts.
topk_method (`str`, *optional*, defaults to `gready`): topk_method (`str`, *optional*, defaults to `greedy`):
Topk method used in routed gate. Topk method used in routed gate.
n_group (`int`, *optional*, defaults to None): n_group (`int`, *optional*, defaults to None):
Number of groups for routed experts. Number of groups for routed experts.
@ -132,7 +132,7 @@ class DeepseekV2Config(PretrainedConfig):
qk_rope_head_dim = 64, qk_rope_head_dim = 64,
v_head_dim = 128, v_head_dim = 128,
qk_nope_head_dim = 128, qk_nope_head_dim = 128,
topk_method = 'gready', topk_method = 'greedy',
n_group = None, n_group = None,
topk_group = None, topk_group = None,
num_experts_per_tok = None, num_experts_per_tok = None,