[CI Failure]: deepseek_mtp_main_random_bf16 can't load, causes deepseek_mtp CI Failure. #20158
Description
Name of failing test
https://github.com/vllm-project/vllm-ascend/actions/runs/15890661413/job/44812465270?pr=1128
Basic information
- Flaky test
- Can reproduce locally
- Caused by external libraries (e.g. bug in
transformers)
🧪 Describe the failing test
Acsend 910b platform
pytest ./tests/e2e/long_term/spec_decode_v0/e2e/test_mtp_correctness.py
📝 History of failing test
vllm-empty/vllm/model_executor/model_loader/default_loader.py:269: in load_weights
loaded_weights = model.load_weights(
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
self = CustomDeepSeekMTP(
(model): CustomDeepSeekMultiTokenPredictor(
(layers): ModuleDict(
(5): CustomDeepSeekMu...ogitsProcessor(vocab_size=129280, org_vocab_size=129280, scale=1.0, logits_as_input=False)
)
(sampler): Sampler()
)
weights = <generator object DefaultModelLoader.get_all_weights at 0xffff7dd9f060>
def load_weights(self, weights: Iterable[tuple[str,
torch.Tensor]]) -> set[str]:
stacked_params_mapping = [
("gate_up_proj", "gate_proj", 0),
("gate_up_proj", "up_proj", 1),
]
expert_params_mapping = FusedMoE.make_expert_params_mapping(
ckpt_gate_proj_name="gate_proj",
ckpt_down_proj_name="down_proj",
ckpt_up_proj_name="up_proj",
num_experts=self.config.n_routed_experts)
params_dict = dict(self.named_parameters())
loaded_params: set[str] = set()
for name, loaded_weight in weights:
if "rotary_emb.inv_freq" in name:
continue
spec_layer = get_spec_layer_idx_from_weight_name(self.config, name)
if spec_layer is None:
continue
name = self._rewrite_spec_layer_name(spec_layer, name)
for (param_name, weight_name, shard_id) in stacked_params_mapping:
# Skip non-stacked layers and experts (experts handled below).
if weight_name not in name:
continue
# We have mlp.experts[0].gate_proj in the checkpoint.
# Since we handle the experts below in expert_params_mapping,
# we need to skip here BEFORE we update the name, otherwise
# name will be updated to mlp.experts[0].gate_up_proj, which
# will then be updated below in expert_params_mapping
# for mlp.experts[0].gate_gate_up_proj, which breaks load.
if (("mlp.experts." in name) and name not in params_dict):
continue
name = name.replace(weight_name, param_name)
# Skip loading extra bias for GPTQ models.
if name.endswith(".bias") and name not in params_dict:
continue
param = params_dict[name]
weight_loader = param.weight_loader
weight_loader(param, loaded_weight, shard_id)
break
else:
for mapping in expert_params_mapping:
param_name, weight_name, expert_id, shard_id = mapping
if weight_name not in name:
continue
name = name.replace(weight_name, param_name)
param = params_dict[name]
weight_loader = param.weight_loader
weight_loader(param,
loaded_weight,
name,
shard_id=shard_id,
expert_id=expert_id)
break
else:
# Skip loading extra bias for GPTQ models.
if name.endswith(".bias") and name not in params_dict:
continue
# According to DeepSeek-V3 Technical Report, MTP modules
# shares embedding layer. We only load the first weights.
if (spec_layer != self.model.mtp_start_layer_idx
and ".layers" not in name):
continue
> param = params_dict[name]
E KeyError: 'model.embed_tokens.weight'CC List.
No response