[Model][Gemma4] Load Weights with AutoWeightsLoader (#55911)

Signed-off-by: Josephasafg <ajgard7@gmail.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
Asaf Gardin
2026-09-18 12:07:30 +00:00
committed by GitHub
co-authored by Harry Mellor Claude Opus 5
parent 2b8a5495a8
commit 44dd18fe0b
8 changed files with 239 additions and 518 deletions
+2 -2
View File
@@ -147,10 +147,10 @@ def test_gemma4_lora_weights_mapping():
def test_gemma4_moe_lora_weights_mapping():
mapper = Gemma4ForCausalLM.hf_to_vllm_mapper
name = (
"base_model.model.model.language_model.layers.9.moe.experts."
"base_model.model.model.language_model.layers.9.experts."
"gate_up_proj.lora_B.weight"
)
assert parse_fine_tuned_lora_name(name, mapper) == (
"model.layers.9.moe.gate_up_proj",
"model.layers.9.experts.gate_up_proj",
False,
)