ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

dtype

Browse source
This commit is contained in:
Daniel Han-Chen 2024-03-18 03:39:44 +11:00
commit bc1746839c
2 changed files with 3 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

3
unsloth/models/llama.py
View file

@ -509,6 +509,8 @@ def LlamaModel_fast_forward(
if inputs_embeds is None:
inputs_embeds = self.embed_tokens(input_ids)
inputs_embeds = inputs_embeds.to(self.config.torch_dtype)
# Normalized from Gemma
IS_GEMMA = self.config.model_type == "gemma"
train_embed_tokens = self.embed_tokens.weight.requires_grad
@ -756,7 +758,6 @@ def CausalLM_fast_forward(fast_forward_inference):
else:
logits = self.lm_head(hidden_states)
pass
print(logits.dtype)
loss = None
if labels is not None:

1
unsloth/models/mistral.py
View file

@ -230,6 +230,7 @@ def MistralForCausalLM_fast_forward(
else:
logits = self.lm_head(hidden_states)
pass
print(logits.dtype)
loss = None
if labels is not None: