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

Explicitly check for OPT activation function #3278

Merged
merged 6 commits into from
Apr 21, 2023
Merged

Explicitly check for OPT activation function #3278

Changes from 1 commit
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
d1fe076
Explicitly check for OPT activation function
cmikeh2 Apr 17, 2023
ed6ef56
Merge branch 'master' into cholmes/fix-opt-act-fn
mrwyattii Apr 20, 2023
89923e5
Update opt.py
mrwyattii Apr 20, 2023
a534fe4
Update opt.py
mrwyattii Apr 20, 2023
d61a33c
formatting
mrwyattii Apr 21, 2023
30a4b86
Merge branch 'master' into cholmes/fix-opt-act-fn
jeffra Apr 21, 2023
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
Prev Previous commit
Next Next commit
formatting
  • Loading branch information
@mrwyattii
mrwyattii committed Apr 21, 2023
commit d61a33c39cbcd51ef08a2adacb29cabacbe2adf8
3 changes: 2 additions & 1 deletion deepspeed/module_inject/containers/opt.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -80,7 +80,8 @@ def __init__(self, client_module, inference=True, use_load_prefix=True):
except:
HFOPTLayerPolicy._orig_layer_class = None

if hasattr(TransformerPolicy, "hf_model_config") and hasattr(TransformerPolicy.hf_model_config, "activation_function"):
if hasattr(TransformerPolicy, "hf_model_config") and hasattr(TransformerPolicy.hf_model_config,
"activation_function"):
if TransformerPolicy.hf_model_config.activation_function == "relu":
self.mlp_act_func_type == ActivationFuncType.ReLU
elif TransformerPolicy.hf_model_config.activation_function in ["gelu", "gelu_new"]:
Expand Down