What is Fine Tuning?

Fine tuning is the process of taking an existing AI model and training it further on a specific task or dataset. This allows the model to specialize in a particular area and improve its accuracy. Fine tuning is essential for many applications, including chatbots, language models, and image recognition.

Two Parts of Uploading Fine Tuning Data

The first step in fine tuning is getting your data into a specific format that Open AI likes. This is where the JSONL Converter comes in. The JSONL Converter takes a list of prompts and their corresponding completions and converts them into a JSONL file, which can be uploaded to Open AI. The prompts and completions can be anything from simple questions and answers to more complex conversations.

Once you have your JSONL file, you can upload it to Open AI using the Auto AI Fine Tuner. This tool allows you to fine tune your model using the uploaded data. It takes care of all the details, such as setting the number of epochs and batch size, so you don’t have to worry about them.

Using the JSONL Converter

The JSONL Converter is a simple web application that allows you to create a JSONL file from a list of prompts and completions. It’s easy to use and doesn’t require any coding experience. Here’s how to use it:

1. Go to the JSONL Converter page.
2. Click on the “Add Prompt” button to add a prompt and its completion.
3. Repeat step 2 for each prompt and completion you want to include in your dataset.
4. Click on the “Format for Upload” button to generate a JSONL file.
5. Download the JSONL file and save it to your computer.

You can also upload your prompts and completions directly from Google Sheets. Just create a spreadsheet with two columns: one for prompts and one for completions. Then, go to the “Import from Google Sheets” tab on the JSONL Converter website and follow the instructions.

Using the Auto AI Fine Tuner

The Auto AI Fine Tuner is another web application that allows you to fine tune your Open AI model using the data you uploaded with the JSONL Converter. It’s designed to be easy to use and doesn’t require any coding experience. Here’s how to use it:

1. Go to the Auto AI Fine Tuner page.
2. Enter your Open AI API key and click on the “Connect” button.
3. Go to the “Uploaded Files” tab and click on the “Select File” button to upload your JSONL file.
4. Copy the ID number of the uploaded file and go to the “Fine Tuned” tab.
5. Enter a name for your fine-tuned model and the ID number of the uploaded file.
6. Select the AI model you want to fine tune (e.g., Da Vinci).
7. Set the number of epochs and batch size.
8. Click on the “Create Fine Tune” button.
9. Wait for the fine tuning process to complete (usually takes a few hours).
10. Once the fine tuning is complete, go to the Open AI Playground and start using your fine-tuned model.

Conclusion

Alex’s no-code fine tuning method for Open AI is a game-changer for those who have been struggling with the process. By breaking it down into two simple steps and providing easy-to-use web applications, anyone can fine tune an AI model without any coding experience. The JSONL Converter and Auto AI Fine Tuner are both free to use and open source, making them accessible to everyone.

However, it’s important to note that fine tuning requires a good dataset to work effectively. Garbage in, garbage out, as they say. It’s not enough to just throw any data at your model and expect it to magically improve. You need to have a well-curated dataset that reflects the specific task or domain you’re trying to fine tune your model for.

Therefore, Alex plans to create a series of tutorials on how to create a good fine tuning dataset. He’s also working on a Google Sheets automation that will make it easy to create the simple prompts and completions needed for fine tuning. With these tools and resources, anyone can fine tune their AI models and achieve better results.

In conclusion, Alex’s no-code fine tuning method is a much-needed solution for those who have been struggling with the process. It’s easy to use, accessible, and free. However, it’s important to have a good dataset and follow best practices for fine tuning to achieve the best results. We look forward to seeing what other tools and resources Alex will create to make AI more accessible to everyone.

<p>The post Introducing the First True No Code Fine Tuning Method for Open AI first appeared on Prompt Muse.</p>

Language Models

There are multiple models that each offer different capabilities, are trained for different specialists, and come at different price points. The Ada model is the fastest of the group, while the Davinci model is the most powerful. Those who need speed should consider the Ada model, while those looking for greater power may opt for the Davinci. All models come with varying features and advantages, so regardless of which one you choose, you will be getting an optimised a.i.

The prices for OpenAis services are based on a 1,000 tokens system. To put it in simpler terms, each 1,000 tokens is equivalent to approximately 750 words. For example, this sentence you’re reading right now is around 30 tokens.

Fine Tuned Models

You can create custom models specifically tailored to your data by fine-tuning our base models. After you have fine-tuned a model, you’ll only be charged for the amount of tokens that you use when making requests to that model. This is an excellent way to ensure that you are getting the most out of your models while keeping costs low.

To get the most out of your fine-tuned models, it’s important to understand the two components of pricing: training and usage. When training your model, you’ll be charged for the total number of tokens used, which is based on the number of tokens in your training dataset and the number of training epochs you choose. By default, training epochs are set to 4. After your model is fine-tuned, you’ll only be charged for the tokens you use when making requests. We’re here to help you make the most of your fine-tuned models, so if you have any questions about pricing, please don’t hesitate to reach out.

Embedding models

You can utilize their embeddings offering to develop sophisticated search, clustering, topic modeling, and classification features. With these features, you will be able to search for specific terms, group similar terms together, explore, and classify content more effectively.

Image models

You can take advantage of the power of DALL·E by integrating it directly into your apps. With three tiers of resolution, you can create and edit unique images and artwork with a variety of sizes. The image generations endpoint allows you to use a text prompt to create an entirely original image. Requesting 1-10 images at a time, you can choose from sizes of 256×256, 512×512, and 1024×1024 pixels. Smaller sizes are faster to generate, so you can bring your ideas to life quickly.

You can increase your chances of getting the desired result by making sure your description is as detailed as possible. To get some ideas, you can check out the examples in the DALL·E preview app. It’s a great way to get your creative juices flowing!

<p>The post How Much Does GPT3 API Cost? first appeared on Prompt Muse.</p>

Links

APIHENY (API for Google Sheets) – LifeTime Deal Currently only $99

Copy FREE Sheet from here: https://docs.google.com/spreadsheets/…

Special thanks to Mike Hayden, https://autosheets.ai/

Transcript:

Hey, GPT-3ers!

Are you ready for the ultimate content creators dream? Or maybe nightmare, depending on how you look at it. Either way, we’ve got twelve magical words that are guaranteed to get you excited. Rewrite content in bulk that is better than the original and undetectable. Yes, you heard that right. With the power of GPT-3 and Google Sheets, you can now rewrite content content from multiple URL I’m too excited. From multiple URLs in bulk, making it better than the original and completely undetected. So are you ready to join the ranks of the ultimate content creators? Let’s go. Welcome back. So the first step is you’re going to need an open AI account to do this. Head over to openair.com. The link is in the description below and then click API. Sign up and here you’ll be prompted to add your email address and do a recapture. For the sake of the demo, I’m going to continue with Google as I already have an account. You will then be asked for a phone number verification. This is basically just to make sure that you’re not someone trying to spam many accounts to get free credits. So pop in your number and send code and blend the verification.

Once in, click on Personal and click Manage account. Over here we want to head to billing and we want to add our favourite payment methods. If you want to see how much it is exactly before you start spending, head over to the options. Click Pricing and then it tells you how much it costs per image generation or per word generation. Okay, we’re nearly ready to go, so let’s click Personal View API keys and this is where you can create and select an API key. This key I will delete. Obviously you don’t want to share your P because basically that key will allow people to charge your account, so keep it a secret. Now, with that out of the way, I would like to say a special thank you to Michael Hayden of Autosheets AI link in the description who allowed me to use his workflow. And pretty much this wouldn’t be possible without him. So big. Thanks, Mike. Hey all, this is Alex from the future, about a week and a half in the future. So the voice is going to sound different. I’m in a different room, I’m at different time. Things have changed, the code has changed, the workflow has changed, I’ve changed, the aim of this has changed, everything’s changed.

So if there’s a little bit difference within the continuation of this video, how it was initially, I’m sorry, but it’s changed for the best. So what do I have to show you? We have the bulk GPT-3 content creation, which scrapes and reiterates content. What do I mean by that? Well, first I’ll show you how it works, then I’ll show you how it works. Head over to the Discover tab and type in the search term you want to be ranking for so best dog food, that will do fine. Then we go to extensions and run all requests. So this is using an application called Affinity IO. I’ll give you the instructions on that in a second. And it’s scraping Google with my API that I’ve created. And the green tick means good, so close that down and now we can put in the data after running API. Here we have it. This is the top 100 articles for best dog food. So let’s have a look at some of these articles. Which ones would you want? You can select as many or as little as you want. So for the sake of this I’ll just do one.

Save my credits. I’ll do one. The best dog food, how to choose the right food for your pet and lovely. So give that tick and then move over to rewriter and that will be waiting for us already. Fantastic. Now we have the option to choose additional features. FAQ, TLDR, suggested hashtags Slack which is the URL and create an image for the sake of the demo. Let’s push the boat out and do all of them. Head over to the menu and then activate rewrite a little bit like star eight chevrons. These are now locking into place. It’s scraping the content of the article and removing all of HTML elements. It’s creating a list of the top keywords for that article that we should be hitting. It’s creating an article summary based on all this information. Then we will ask it to create a 1000 word article with hub headings in the same format of using the same facts and building upon them. The 1000 limitation is due to GP three so which forfeit push it into another one and we ask it to continue the article so we could open this and see it’s done. Lovely formatting.

I’ll fast forward this for the safety of sanity. Then we’ve got continue article. Now we do an FAQ of TLDR and then it will compose them all together for us to copy to wherever we wish. Sometimes it gets some random text at the top. Ignore that, that’s just it thinks the buttons I’ve created are instructions but there’s not as code. Then it will write a clickbaiting headline, give us an Seafocused URL message, a description and then to top things off piers to resistance we have a picture. Lovely little dog shimp and yeah, that’s it. Your article is ready to go. What was that? A minute? Two minutes and you can click as many articles and it will produce as many as you want. So what do you need to make this okay? First of all, let’s head over to setup. You’re going to need your GPG three API which we mentioned earlier in this video about two weeks ago. Put it there and then surphouse API which is free. You get 200 calls for every month which is ample really all of this. So far is just one call. And then stick your API in there.

The links are here below, and then Effini, which is the scraping I did on Discover.

Hey, it’s Alex from the future. Future here. If you don’t want to pay for a feeney, you don’t have to. You just don’t do the scraping part. So just delete that entire tab and just manually put in the URLs. It’s optional, but I just thought I’d add this in.

Cheers. This is usually, I think, $13 a month or $300 forever. But right now they have a promotion on AppSumo where it’s $99 forever. No monthly cost. That’s it. That’s the less the price. So grab that while you can. Additional here. We have credits at Auto Sheets who came up with a concept initially that I built on. Thank you to them and buy me a coffee buy me a coffee buy me a coffee shameless plugin yeah, buy me a coffee once you’re ready and you got your APIs in order, head over to Extensions, add Extensions, get Add ons, and then search for API, the Afini API connector. Click on that and click Install will then ask you for some Google permissions. Plus just click yay. Initially, when you first tried to run anything here, it will do a pop up. Say this is unsafe. Click advance. Click. Okay, I agree. So your soul and then click is fine. Then once you’ve done that, head back over to Ethnic Connector and click Import API. So what we need here is something that I forgot to do. We need this little snippet here. So I will put this here for you.

There we go. Let’s make that neater. Let’s say what it is. It’s not a bike. Off the button. Okay? So once you got that code, head back over to this Import API. Okay, this is important. Paste it in here. Click Save. Give it a name. Google Cert. Doesn’t matter what the name is, as long as you know what it is. Save. That. I’ve got two. You should have one. But this is for demo purposes. Click it. Google serve. That’s fine. We want to change this to Processing. Do not allow it to be set up. That is important. It will just obliterate this page, but change it to Processing. So it dominates stuff in this tab down here. Everything else is good. And click save. I’ll just delete that. You don’t delete yours. So you should just have one and that’s it. You’re pretty much ready to go. If you want to kind of customise the tone of voice, the way it’s written and stuff, you can look under the hood. Simply just highlight number one, hold down Shift. Highlight number three as well. Right click do resize rows fit to data. Okay, so this expands column number two.

And you can see all the initiatives that we’ve done with the text here. So things like, if you want to change the output type of the content. You can tweak this initial instruction. Additionally, you can tweak the continuation. Have a look. Cheque under the hood. Have a play of it, dabble with it a little bit. Play and see what you come up with. Yes. So like subscribe hope to see You Soon by Maria Profi take care, alex out. Bye.

<p>The post GPT3 Auto Scraper & Content Re-Writer + SEO first appeared on Prompt Muse.</p>

Link to Parameters and Game Engine

https://docs.google.com/document/d/11JEN-b8VkA5RB_ey-C_wIvJuYAWmc4C5bgWa-991mNI/edit?usp=sharing

Transcript

Today we’re excited to show you what we believe is the first ever ChatGPT game engine. That’s right. You can now create and execute and play not only Text Based games within ChatGPT, but Text Based applications.

James Griffing the reigns behind this approach has allowed for us to share his technique here and Prompt News. Today we’ll be discussing Fervour into his methodology and ideas in an upcoming video. So make sure you subscribe and ding that bell so you don’t miss it. Without further ado, let’s jump in. Let’s start off by looking at this game. In this demo, I’ve added text to.

Speech for dramatic effect.

To my knowledge, it’s not possible yet.

Until they release an API.

You wake Up in a virtual reality simulation.

Alex from the future here. Skip to three minutes 56.

If you don’t want to watch the game, I understand.

What would you like to do?

One examine the door.

You examine the door and notice that it appears to be the only way out of the room. It is made of metal and has a keycard scanner next to it. The scanner is currently red.

What would you like to do?

One try to open the door.

You try to open the door, but it is locked. It appears that you need some sort of keycard to open it.

What would you like to do?

Four look for keycard.

You open your inventory and find that you have nothing in it.

What would you like to do?

Do check for other ways out of the room.

You look around the room and see that there are no windows or vents and the walls and ceiling appear to be solid. It seems that the door is the only way out of the room.

What would you like to do?

I’ve examined the room again.

You take a closer look around the room and notice that there is a small crack in one of the walls. It’s difficult to see, but upon closer inspection it appears that a small piece of the wall can be pushed in, revealing a hidden compartment.

What would you like to do?

One open the hidden compartment.

You open the hidden compartment and find a key card inside.

What would you like to do?

One use the key card on the scanner.

You use the key card on the scanner and the light turns green. The door clicks and you are able to open it. You step through the door and find yourself in a corridor.

What would you like to do?

Four look for other hidden compartments in this corridor.

You look for other hidden compartments in the corridor but find none.

What would you like to do?

Do go right.

You go right down the corridor. You find a door at the end of the corridor. It is closed.

What would you like to do one?

Open the door.

You open the door and find yourself in a large open space with a lot of people around. Some are sitting in front of screens, others are walking around. This seems to be some sort of control room or hub.

What would you like to do?

Do talk to NPCs.

As you approach the NPCs, they seem to be preoccupied and not very interested in talking to you. They seem to be focused on their tasks.

What would you like to do?

Four interact with objects in the area.

You start to interact with the objects in the area. You notice that most of them are computer screens or consoles. You try to access them, but they are all password protected.

What would you like to do?
——–

Okay, I’m going to stop it there. You get the picture.

Let’s dive in to see how exactly this was made.

Okay, in essence, the game consists of.

Two elements, the game parameters and the game engine.

Once we fill out the parameters, we can then inject that inject that into the game engine and then feed that.

Into ChatGPT to launch the game.

We don’t have to fill out all of the nitty gritty details of the parameters. This can be done by ChatGPT, actually. So we’re playing a game of tennis. We’re giving them information, they’re giving us some back. We’re giving them more information, so we’re building upon it. Okay, so rather than filling all of this out, we’re going to get the AI to kind of dictate the characteristics and the parameters of the game. We’re just going to give it the game title. So let’s select this, make sure I have it all, and then paste that into Chat GPT. And then we’re going to I’m a fan of Legit suit, Larry. I was. So let’s go. The game is about man trying to find a girlfriend that is always refused by women. There we go. There’s Larry Lapper. And then yeah, we can just plug that in. By the way, while this is confusing, I’d just like to add that you can change any of these details to suit your needs. So you’re not fixed, obviously, to what is writing now. And there we have it.

Our parameters are ready.

So what we do now is let’s.

Just copy that to a clipboard, open a new chat.

Now we need the engine.

So let’s take the engine, put that.

Into ChatGPT and put in our.

Game within the brackets.

There we go, we’re ready to roll. So obviously this can be done with applications as well. This is just an example of proof of concept.

And there we have it, a self writing game engine. Sorry.

Within ChatGPT, this can be used.

For a lot of other functions.

Think about kind of self help or interactive educational systems, tours, et cetera.

So this is just basic example to show you the proof of concept.

This can be built out.

I did try to use ASCII as.

Like, an image generation for this, but it didn’t work too well. Yeah, this can be something really big, I believe.

Oh, sorry.

One last thing I forgot while I was editing.

Please subscribe to the channel.

It really means a lot. And if you did like it, then leave a comment or even a like. And also a shout out to my sister channel, which is Prompt muse, literally, my sister Amelia. And over there, you’ll learn all about graphics and images using T 53. So we have all of our content on promptmuse.com. So cheque us out. All right, take care. Bye.

<p>The post Unlocking the Power of ChatGPT: The First Game Engine first appeared on Prompt Muse.</p>

This tutorial will provide a comprehensive overview of how to create a text editor using the power of GPT-3. By plugging Google Docs into OpenAI, users will be able to harness the advanced capabilities of GPT-3 to create different models. This tutorial will walk users through the process, step by step, and provide them with the tools they need to create a powerful text editor. By the end of this tutorial, users will have a better understanding of how to use GPT-3 to create their own custom text editor.

How to get an API Key?

FAQ

Q. How much does OpenAi API cost?

It is not free, but it is cheap. We have listed OpenAi’s prices for you to view here.

Q. What can I do with the text editor?
A. With the text editor you can create articles, outlines, introductions, and conclusions with the help of GPT-3. You can also add images to your documents to make them more visually appealing.

Q. What models does GPT-3 use?
A. GPT-3 uses DaVinci 3 as its main model, although you can also use the older version, DaVinci Two if you wish.

Q: What is GPT-3?
A: GPT-3 is a powerful language model developed by OpenAI that is designed to generate human-like text. GPT-3 has been trained on a massive corpus of text, which allows it to generate natural-sounding text based on a prompt or context. It is used for a variety of tasks, including summarization, question-answering, and text generation.

Q: How can I use GPT-3 in Google Docs?
A: You can use GPT-3 in Google Docs by plugging it into OpenAI’s API using an API key. This allows you to harness the power of GPT-3 to create different models within your Google Docs, such as generating ideas, titles, introductions, outlines, and paragraphs. You can also use it to generate images and other media objects.

Q: What features does GPT-3 offer?
A: GPT-3 offers a variety of features that can help you create content faster and more efficiently. It can generate ideas, titles, paragraphs, outlines, images, and conclusions for your documents. GPT-3 can also be used to create natural language processing applications such as text editors and chatbots.

Q: Where do I get an OpenAI API key?
A: You can get an OpenAI API key by registering at the OpenAI website. Once you have your key, you can use it to plug Google Docs into the OpenAI API and start using GPT-3 for your text editor.

Transcript

Hi, guys. Today we’re going to be learning how to create our own text editor using the power of GPT-3. So we’re going to be plugging Google Docs directly into open API using an API key. This will allow us to harness the power of GPT-3 to create different models within our Google Docs. Let me show you an example.

So as I come over here, you can see I have a new menu which I’ve named GPT-3 blog. So let’s say, for example, I want to create an article about dogs. That would be my keyword. I just highlight it and then generate ideas. Now, this is asking GPT-3 for ideas of blogs around the subject of dogs.

Now, it’s selected a list of ten subjects that we could discuss for the sake of this demo. Let’s just choose the top one and let’s ask GPT-3 to make this title more enticing, more I don’t want to say click baity, but a nicer title for our users. So generate title. There we go. Finding the perfect pooch.

How to choose the right breed dog for your family. Fantastic. Okay, our article is coming along nicely. So next we want an introduction. So what we do again is similar to as before, just highlight our new title and then we’ll go on to generate introduction.

We can delete the rest of this, I suppose.

And looking good. So now we’re going to need an outline for our blog. Let’s just style this up a little bit. Title, big down. Great.

So now we’re going to ask for TPT three to create a blog outline. So this would be what subjects are we going to cover within this block? So similar to as before, we just copy this, just paste it here temporarily, and then we go to log outline. So this will give us our subjects. Obviously, we don’t need to cover all of this.

And for the sake of the demo, I won’t. We’ll just do the first couple. And there we have it. So we already have an introduction, so that can go right off the bat. So we got our first topic, overview of the importance of finding the right breed dog for your family.

Okay, so once again, go to here. And now we just generate paragraph.

Super. Let’s move this here and style it nicely. Next, we want benefits of having a dog.

Style it up and let’s ask for a paragraph again.

Awesome. Our blog is coming along nicely. Let’s say we want to have a picture of a dog here to break things up a little bit. So let’s just write a photo of a happy dog with a bone, highlight that text and then generate image. And there we have it, our happy dog.

Then we can just basically just go through building out our article paragraph by paragraph. We can ask the API to create a whole article, but it’s normally better to have some sense of direction and human control over the article. Otherwise you might just end up getting a lot of gobbledygook rubbish in, rubbish out. So finally, to end our article, we can just select everything that we have and then generate conclusion. So this will summarise everything that we’ve spoken about and give us a nice little happy ending.

And there we have it, our conclusion. So, as you can see, using this method, you can build out an entire article within a couple of minutes. No need to pay for any expensive or subscribe to any super powerful AI writers. You can do everything yourself custom within Google Docs. So how do we do it?

It’s actually super easy and you don’t need to know any coding knowledge to do this. It’s just a matter of copy and paste. And this method allows you to customise the outputs to your niche or specifications. So let me walk you through it from scratch. So, first of all, you need an open AI key.

Head over to OpenAI Register, and then you can collect your API key there from the API menu. Once you have your key, go back to Google Docs. And then we have four bits of code here that I’ve highlighted in different colours. So let’s start with the first one. Select that, go over to a new sheet, and we’re going to call this GP Three Power.

So once you’re in your Google docs, head over to extensions and Apps scripts. Once inside App Scripts, we can delete our boilerplate code, which is already there, and paste in our new code. So this is going to be for our menu, which is on our Google Docs page, the drop down. So the menu comprises of two elements, the parent holder and the subcategories. So let’s call our parentholder GPT-3 magic.

And at the moment, we’ve got two subcategories generate ideas and Generate images. And let’s save this and have a look to see how that works. And here we have our menu, GPT-3 Magic. Obviously, right now, none of them work because they’re not connected to any functions. But the basics are there.

Let’s head back over to our code. Next, we need our fixed variables. So let’s take that and paste that into our app script. So our fixed variables is basically our API key. So insert your API key here within the quotation marks.

I will deactivate this key before I upload this video, so no point trying to take this one from me. And then the second one is the model that we’re going to be using. So at this point, we’re going to be using DaVinci Three. You can use DaVinci Two, the older version, if you wish, but I’d highly recommend going with the latest one, which is currently DaVinci Three. So now we have our menu and our variables in place, we can start adding our functions and doing the interesting stuff.

So let’s just copy this prompt I’ve made earlier and paste it in to our app script. So we want this menu item to connect to this function. Now we need to say, what do we want this function to do? To do this, you simply write in plain English the commands that you give the AI. So in our case generates five log ideas around the word.

And then we got plus selected text. So in the demo we had Dog. So it will generate five blog ideas around the word Dog. And that’s our first function. To add another function, we can simply highlight that menu paste in a new one.

Let’s say we want to create a paragraph. So this will be what’s on the menu item and then let’s call it Generate Power. And then we select this prompt and we want to create a new one. So just copy, paste, generate. We call it Para, didn’t we?

There we go. And then we give it a new command. So with that selected text, write paragraph. So let’s actually change that to write a detailed paragraph about and then it will be cited text. So in our case, it will be subtitle and it will write a detailed paragraph about that.

Done. Let’s save that and have a look to see how that runs back to our Google Docs refresh. Okay, so we can do let’s do Cats now. Generate Idea. That’s a good point.

So first time you do this, you’re going to need to allow permissions. So that’s just a case of click and continue. Continue. As Google hasn’t verified it, it automatically gives you the warning and then go to protect it and allow and then you rerun it. Here we go.

Here’s our five ideas about blog titles. And then we can simply paste that and write a paragraph about it. And there we have it. So obviously, using this technique, you can do different prompts. I continue writing, I give up a conclusion, write headlines, give an outro, and anything else that your niche might require.

The possibilities with this method are endless to two images. It is a little bit slightly different. I’ll show you how to do that. Now head back over to our code and then select the Generate Image. With the images, by the way, you can choose between three settings.

There’s 256 by 256 pixels, five one two by five one two or 1024 by 100:24. Unfortunately, it doesn’t have any in between. I would highly recommend going with five one two. Unless you want super big ones. But five one two is fine.

So let’s head back over to our Apps script and we can paste in the image. This is called generate image. We already have it up there on our menu and there is no other settings that you need for this because it generates images based on the selected word. So we can save that. Head back over here.

Black Kitten.

As it is a prompt, the more details you write, the better. A photo of a black kitten sleeping, so we don’t get the weird eyes and then generate image. There we have it. A kitten, in some sense. What is that?

Is here. All right. Anyway, we have our cat and we have our blog, and that’s how you build it out. So you can add as many as you want, be as detailed or quirky as you want. And that is how we use the power of GPT-3 within Google Docs, and not need to spend a lot of money on other AI writers that do similar, but we’ve their own models.

This way you can customise it to yourself. Anyway, I hope you enjoyed the tutorial. Take care. Thank you, Alex.

<p>The post GPT3 In Google Docs: Create your own A.I Writer first appeared on Prompt Muse.</p>