26 lines
506 B
YAML
26 lines
506 B
YAML
|
model_name: "vllm-ascend/Qwen3-30B-A3B-W8A8"
|
||
|
|
model_type: "vllm"
|
||
|
|
hardware: "Atlas A2 Series"
|
||
|
|
|
||
|
|
serve:
|
||
|
|
tensor_parallel_size: 2
|
||
|
|
dtype: auto
|
||
|
|
max_model_len: 4096
|
||
|
|
gpu_memory_utilization: 0.7
|
||
|
|
trust_remote_code: false
|
||
|
|
enable_expert_parallel: true
|
||
|
|
quantization: ascend
|
||
|
|
|
||
|
|
tasks:
|
||
|
|
- name: "gsm8k"
|
||
|
|
metrics:
|
||
|
|
- name: "exact_match,strict-match"
|
||
|
|
value: 0.9
|
||
|
|
- name: "exact_match,flexible-extract"
|
||
|
|
value: 0.8
|
||
|
|
|
||
|
|
num_fewshot: 5
|
||
|
|
apply_chat_template: false
|
||
|
|
fewshot_as_multiturn: false
|
||
|
|
batch_size: "auto"
|