[Bugfix][Model] Fix Qwen2 (#3554)
This commit is contained in:
parent
b7050ca7df
commit
ea5f14e6ff
@ -20,6 +20,7 @@ MODELS = [
|
||||
"stabilityai/stablelm-3b-4e1t",
|
||||
"allenai/OLMo-1B",
|
||||
"bigcode/starcoder2-3b",
|
||||
"Qwen/Qwen1.5-0.5B",
|
||||
]
|
||||
|
||||
|
||||
|
||||
@ -349,7 +349,7 @@ class Qwen2ForCausalLM(nn.Module):
|
||||
("gate_up_proj", "gate_proj", 0),
|
||||
("gate_up_proj", "up_proj", 1),
|
||||
]
|
||||
params_dict = dict(self.named_parameters())
|
||||
params_dict = dict(self.named_parameters(remove_duplicate=False))
|
||||
for name, loaded_weight in hf_model_weights_iterator(
|
||||
model_name_or_path, cache_dir, load_format, revision):
|
||||
if "rotary_emb.inv_freq" in name:
|
||||
|
||||
Loading…
Reference in New Issue
Block a user