Hacks for LLMs, Round 2

IMPROVE YOUR AI GAME

Hacks To Become an AI Pro, Round 2

Last week, we saw our first set of hacks, which alone will have considerably boosted your AI performance.

Today, we look at more advanced hacks that are also immediately applicable.

From unexpected gifts by Anthropic or Google to how to prompt multimodal models effectively with prompt frameworks like ROCC and leading to synthetic grounding, the hack, I predict, has been the missing piece in the majority of your failed interactions with LLMs.

As some hacks are more advanced, they will include some rather technical passages you can jump if not interested (I will point them).

Let’s dive in!

Squeezing Every Bit of Performance

In the previous episode of this series, we rapidly concluded that the single best advice for talking to LLMs was to imitate how trainers talked to them. And just as I wrote that Anthropic gave the world an unexpected gift.

Copying Never Felt So Good

As trainers decide the training data points, they dictate how and what the model learns. Therefore, imitating them increases the likelihood that the model elicits the desired behavior.

Now, some of these tips are being released to the public. This week, Anthropic released the system prompt for Claude models in the web browser, giving a world of insight into how you should interact with its model Claude.

If you click the link, you’ll see that they apply many of the recommendations we gave last week, such as the use of tags to clearly separate every segment of the prompt, to the use of clear and concise instructions.

But what is the system prompt?

All LLMs have a prompt hierarchy (shown below in OpenAI’s case). This means that the model prioritizes certain parts of the prompt over others.

The system message is at the top of the list, defining the model's behavior. “Be concise” and “Act this way or that way” are examples of instructions provided there.

Source: OpenAI

If you use the model API, you can define your system prompt, and in cases such as OpenAI, you can change it on the web and desktop versions, too.

This is an absolute must because it is an overarching instruction that the LLM will apply to every interaction, considerably reducing the need for you to repeat how you want it to behave every time.

OpenAI lets you update the system instructions in the web interface.

Another tip you can follow comes from Google, which highly recommends avoiding unnecessary jargon and exploiting the use of straightforward words and sentences.

If we watch Anthropic’s system prompt again, we see the same pattern: They completely avoid fancy words such as conjunctive adverbs like “therefore” or “furthermore” that we usually use to make our writing less boring.

Leave artistry to novelists. Instruction, instruction, instruction.

Another tip not shown in Claude’s prompt but recommended by OpenAI or Meta is using capital letters to make the model more compliant.

But isn’t that a really rude way to talk…

Enjoyed the introduction? Subscribe or sign in to continue with this edition.