Skip to content

🧠 两小时从 0 训出你的 64M 大模型。Train your own 64M LLM from scratch in 2 hours.

[update] tie embedding

jingyaogong5704766352e5
diff --git a/model/model_minimind.py b/model/model_minimind.py
index d3518ad..e8e7f54 100755
--- a/model/model_minimind.py
+++ b/model/model_minimind.py
@@ -27,6 +27,7 @@ class MiniMindConfig(PretrainedConfig):
self.max_position_embeddings = kwargs.get("max_position_embeddings", 32768)
self.rms_norm_eps = kwargs.get("rms_norm_eps", 1e-6)
self.rope_theta = kwargs.get("rope_theta", 1e6)
+ self.tie_word_embeddings = kwargs.get("tie_word_embeddings", True)
self.inference_rope_scaling = kwargs.get("inference_rope_scaling", False)
self.rope_scaling = {
"beta_fast": 32,
@@ -233,7 +234,7 @@ class MiniMindForCausalLM(PreTrainedModel, GenerationMixin):
super().__init__(self.config)
self.model = MiniMindModel(self.config)
self.lm_head = nn.Linear(self.config.hidden_size, self.config.vocab_size, bias=False)
- self.model.embed_tokens.weight = self.lm_head.weight
+ if self.config.tie_word_embeddings: self.model.embed_tokens.weight = self.lm_head.weight
def forward(self, input_ids, attention_mask=None, past_key_values=None, use_cache=False, logits_to_keep=0, labels=None, **kwargs):
hidden_states, past_key_values, aux_loss = self.model(input_ids, attention_mask, past_key_values, use_cache, **kwargs)
diff --git a/scripts/convert_model.py b/scripts/convert_model.py
index f409229..a4162a3 100644
--- a/scripts/convert_model.py
+++ b/scripts/convert_model.py
@@ -51,7 +51,7 @@ def convert_torch2transformers(torch_path, transformers_path, dtype=torch.float1
"max_position_embeddings": lm_config.max_position_embeddings,
"rms_norm_eps": lm_config.rms_norm_eps,
"rope_theta": lm_config.rope_theta,
- "tie_word_embeddings": True
+ "tie_word_embeddings": lm_config.tie_word_embeddings
}
if not lm_config.use_moe:
qwen_config = Qwen3Config(