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

Update llama.py

Browse source
This commit is contained in:
Daniel Han 2025-03-04 19:11:54 -08:00
commit e2c04a2d79
1 changed files with 3 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

7
unsloth/models/llama.py
View file

@ -1535,7 +1535,6 @@ class LongRopeRotaryEmbedding(torch.nn.Module):
pass
@torch.inference_mode
def unsloth_fast_generate(
self,
inputs: Optional[torch.Tensor] = None,
@ -1577,7 +1576,7 @@ def unsloth_fast_generate(
kwargs["pad_token_id"] = kwargs.pop("pad_token_id", model_eos_token_id)
# Mixed precision autocast
with torch.autocast(device_type = "cuda", dtype = dtype):
with torch.inference_mode(), torch.autocast(device_type = "cuda", dtype = dtype):
output = self._old_generate(self, inputs, *args, **kwargs)
pass
@ -1972,7 +1971,7 @@ class FastLlamaModel:
# Patch generate
if model.generate.__name__ != "unsloth_fast_generate":
model._old_generate = model.generate
model.generate = unsloth_fast_generate
model.generate = types.MethodType(unsloth_fast_generate, model)
model.generate.__doc__ = model._old_generate.__doc__
return model, tokenizer
pass
@ -2410,7 +2409,7 @@ class FastLlamaModel:
# Patch generate
if model.generate.__name__ != "unsloth_fast_generate":
model._old_generate = model.generate
model.generate = unsloth_fast_generate
model.generate = types.MethodType(unsloth_fast_generate, model)
model.generate.__doc__ = model._old_generate.__doc__
return model
pass