Unverified Commit ea5f14e6 authored by Roy's avatar Roy Committed by GitHub
Browse files

[Bugfix][Model] Fix Qwen2 (#3554)

parent b7050ca7
...@@ -20,6 +20,7 @@ MODELS = [ ...@@ -20,6 +20,7 @@ MODELS = [
"stabilityai/stablelm-3b-4e1t", "stabilityai/stablelm-3b-4e1t",
"allenai/OLMo-1B", "allenai/OLMo-1B",
"bigcode/starcoder2-3b", "bigcode/starcoder2-3b",
"Qwen/Qwen1.5-0.5B",
] ]
......
...@@ -349,7 +349,7 @@ class Qwen2ForCausalLM(nn.Module): ...@@ -349,7 +349,7 @@ class Qwen2ForCausalLM(nn.Module):
("gate_up_proj", "gate_proj", 0), ("gate_up_proj", "gate_proj", 0),
("gate_up_proj", "up_proj", 1), ("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( for name, loaded_weight in hf_model_weights_iterator(
model_name_or_path, cache_dir, load_format, revision): model_name_or_path, cache_dir, load_format, revision):
if "rotary_emb.inv_freq" in name: if "rotary_emb.inv_freq" in name:
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment