URIAL¶
使用非指令微调模型生成响应。
URIAL 是一个预定义的任务，它使用非指令微调模型生成响应。此任务用于根据作为输入提供的对话生成响应。
输入 & 输出列¶
graph TD
subgraph Dataset
subgraph Columns
ICOL0[instruction]
ICOL1[conversation]
end
subgraph New columns
OCOL0[generation]
OCOL1[model_name]
end
end
subgraph URIAL
StepInput[Input Columns: instruction, conversation]
StepOutput[Output Columns: generation, model_name]
end
ICOL0 --> StepInput
ICOL1 --> StepInput
StepOutput --> OCOL0
StepOutput --> OCOL1
StepInput --> StepOutput
输入¶
-
instruction (
str, 可选): 从中生成响应的指令。 -
conversation (
List[Dict[str, str]], 可选): 从中生成响应的对话（最后一条消息必须来自用户）。
输出¶
-
generation (
str): 生成的响应。 -
model_name (
str): 用于生成响应的模型的名称。
示例¶
从指令生成文本¶
from distilabel.models import vLLM
from distilabel.steps.tasks import URIAL
step = URIAL(
llm=vLLM(
model="meta-llama/Meta-Llama-3.1-8B",
generation_kwargs={"temperature": 0.7},
),
)
step.load()
results = next(
step.process(inputs=[{"instruction": "What's the most most common type of cloud?"}])
)
# [
# {
# 'instruction': "What's the most most common type of cloud?",
# 'generation': 'Clouds are classified into three main types, high, middle, and low. The most common type of cloud is the middle cloud.',
# 'distilabel_metadata': {...},
# 'model_name': 'meta-llama/Meta-Llama-3.1-8B'
# }
# ]