Wolfram/LLMFunctions | Paclet Repository

[

EXPERIMENTAL

]

LLMExampleFunction

LLMExampleFunction[{ in 1  out 1 , in 2  out 2 ,…}]
	creates an LLMFunction from few-shot examples.

LLMExampleFunction[{ in 1 , in 2 ,…}{ out 1 , out 2 ,…}]
	generates the same result

LLMExampleFunction[{header,training}]
	prefaces the prompt with header .

LLMExampleFunction[prompting,form]
	includes the interpreter form to apply to the response.

Details



Examples

(5)

Basic Examples

(2)

Create an

LLMFunction

from a small training set:

In[1]:=

fun=

LLMExampleFunction

[{{1,1}2,{1,2}3,{2,2}4}]

Out[1]=

LLMFunction

template: Input: {1, 1}Output: 2Input: {1, 2}Output: 3Input: {2, 2}Output: 4Input:

1

Output:

interpreter: String



Evaluate the function on an input:

In[2]:=

fun[{3,4}]

Out[2]=

7

Clarify the task and process the output string using an interpreter type:

In[1]:=

fun=

LLMExampleFunction

[{"what is the city with this nickname?",{"windy city""Chicago","big apple""New York"}},"City"]

Out[1]=

LLMFunction

template: what is the city with this nickname?Input: windy cityOutput: ChicagoInput: big appleOutput: New YorkInput:

1

Output:

interpreter: City



Evaluate the function on an input:

In[2]:=

fun["motor city"]

Out[2]=

Detroit

Scope

(3)



SeeAlso

LLMFunction

▪

LLMPrompt

▪

LLMTool

""