General
PromptBeginner5 minmarkdown
LLMs are increasingly used to generate questions: in therapy bots
coaching apps
0
Explore
Find agent skills by outcome
105,746 skills indexed with the new KISS metadata standard.
Showing 24 of 105,746Categories: Data & Insights, General
General
PromptBeginner5 minmarkdown
These require judgment
empathy
0
General
PromptBeginner5 minmarkdown
During testing
the final Socratic chain asked:
0
General
PromptBeginner5 minmarkdown
- **Human-final**: shows candidates
human decides
0
General
PromptBeginner5 minmarkdown
| Elenchus (examination) | Chain 5: Apply lens
find tensions |
0
General
PromptBeginner5 minmarkdown
| Human as final authority | System shows
human decides |
0
General
PromptBeginner5 minmarkdown
- A **mirror**
not a judge
0
General
PromptBeginner5 minmarkdown
- **Socratic**: it asks
it doesn't conclude
0
General
PromptBeginner5 minmarkdown
- A **discovery method**
not a scoring algorithm
0
General
PromptBeginner5 minmarkdown
- Introduces value trade-offs (money vs belonging
loss vs gain)
0
General
PromptBeginner5 minmarkdown
toward examining internal meaning
assumptions
0
General
PromptBeginner5 minmarkdown
Louise
the linguist
0
General
PromptBeginner5 minmarkdown
Traditional approaches use fixed metrics: sentiment scores
engagement rates
0
General
PromptBeginner5 minmarkdown
print(json.dumps(results
indent=2
0
General
PromptBeginner5 minmarkdown
When humanity encounters aliens
the military asks: *Are you hostile?*
0
General
PromptBeginner5 minmarkdown
prompt = render_prompt(chain[prompt]
context)
0
General
PromptBeginner5 minmarkdown
output = self.llm(prompt
chain.get(output_schema))
0
General
PromptBeginner5 minmarkdown
samples = stratified_sample(corpus
n=sample_size)
0
General
PromptBeginner5 minmarkdown
def __init__(self
llm_fn=None):
0
General
PromptBeginner5 minmarkdown
def run(self
corpus: list[dict]
0
General
PromptBeginner5 minmarkdown
result = result.replace(placeholder
value_str)
0
General
PromptBeginner5 minmarkdown
def call_llm(prompt: str
output_schema: dict = None) -> dict | str:
0
General
PromptBeginner5 minmarkdown
return {_placeholder: True
schema: output_schema}
0
General
PromptBeginner5 minmarkdown
for key
value in variables.items():
0