github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-08-17 15:36:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

[chat]fix sft training for bloom, gpt and opt (#3418)

Browse Source 
fix sft training for bloom, gpt and opt
This commit is contained in:
Yuanchen 2023-04-04 09:46:23 +08:00 committed by GitHub
parent 638a07a7f9
commit b09adff724
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 9 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
applications/Chat/coati/models/bloom/bloom_lm.py
View File

@ -33,3 +33,6 @@ class BLOOMLM(LM):
if checkpoint: if checkpoint:
model.gradient_checkpointing_enable() model.gradient_checkpointing_enable()
super().__init__(model, lora_rank, lora_train_bias) super().__init__(model, lora_rank, lora_train_bias)
def forward(self, input_ids, attention_mask=None, labels=None, **kwargs):
return self.model(input_ids, attention_mask=attention_mask, labels=labels, **kwargs)

3
applications/Chat/coati/models/gpt/gpt_lm.py
View File

@ -33,3 +33,6 @@ class GPTLM(LM):
if checkpoint: if checkpoint:
model.gradient_checkpointing_enable() model.gradient_checkpointing_enable()
super().__init__(model, lora_rank, lora_train_bias) super().__init__(model, lora_rank, lora_train_bias)
def forward(self, input_ids, attention_mask=None, labels=None, **kwargs):
return self.model(input_ids, attention_mask=attention_mask, labels=labels, **kwargs)

3
applications/Chat/coati/models/opt/opt_lm.py
View File

@ -33,3 +33,6 @@ class OPTLM(LM):
if checkpoint: if checkpoint:
model.gradient_checkpointing_enable() model.gradient_checkpointing_enable()
super().__init__(model, lora_rank, lora_train_bias) super().__init__(model, lora_rank, lora_train_bias)
def forward(self, input_ids, attention_mask=None, labels=None, **kwargs):
return self.model(input_ids, attention_mask=attention_mask, labels=labels, **kwargs)