I watched Greg Brockman’s presentation of GPT-4 on March 16th. Then I used ChatGPT Plus to access GPT-4. With Greg’s presentation in mind, I partnered with GPT-4 which generated several applications and explanations for me. Everything went incredibly smoothly!
Here is the link to the notebook:
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"provenance": [],
"toc_visible": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"language_info": {
"name": "python"
}
},
"cells": [
{
"cell_type": "markdown",
"source": [
This file has been truncated. show original
Enjoy!
I’ll be sharing a GPT-4 notebook with the API asap.
3 Likes
I just completed a notebook that runs GPT-3, GPT-3.5 - turbo and GPT-4 with OpenAI API. It’s fun to develop with GPT-4 as a partner.
Here is the link to the notebook :Transformers-for-NLP-2nd-Edition/Exploring_GPT_4_API.ipynb at main · Denis2054/Transformers-for-NLP-2nd-Edition · GitHub
1 Like
jamwjc
March 17, 2023, 12:52am
3
can you provide a link or maybe a step by step for getting gpt3.5 trubo? I’m really new to all of this but would really like to access it.
1 Like
Step by step for gpt4 would be great as well! I am trying to decipher it now.
What is the token (character) limit on this gpt4 here?
Trying it now and it seems as if GPT4 is no longer in the list. Or did I do something wrong here?
Blockquote
0 babbage
1 davinci
2 babbage-code-search-code
3 text-similarity-babbage-001
4 text-davinci-001
5 ada
6 curie-instruct-beta
7 babbage-code-search-text
8 babbage-similarity
9 gpt-3.5-turbo
10 code-search-babbage-text-001
11 gpt-3.5-turbo-0301
12 code-cushman-001
13 code-search-babbage-code-001
14 text-ada-001
15 text-embedding-ada-002
16 text-similarity-ada-001
17 text-davinci-insert-002
18 ada-code-search-code
19 ada-similarity
20 whisper-1
21 text-davinci-003
22 code-search-ada-text-001
23 text-search-ada-query-001
24 text-curie-001
25 text-davinci-edit-001
26 davinci-search-document
27 ada-code-search-text
28 text-search-ada-doc-001
29 code-davinci-edit-001
30 davinci-instruct-beta
31 code-davinci-002
32 text-similarity-curie-001
33 code-search-ada-code-001
34 ada-search-query
35 text-search-davinci-query-001
36 curie-search-query
37 davinci-search-query
38 text-davinci-insert-001
39 babbage-search-document
40 ada-search-document
41 text-search-curie-query-001
42 text-search-babbage-doc-001
43 text-davinci-002
44 curie-search-document
45 text-search-curie-doc-001
46 babbage-search-query
47 text-babbage-001
48 text-search-davinci-doc-001
49 text-search-babbage-query-001
50 curie-similarity
51 curie
52 text-similarity-davinci-001
53 davinci-similarity
Blockquote
Sure!
To dig into ChatGPT 3.5 and 4, go straight to my open-source directory:
https://github.com/Denis2054/Transformers-for-NLP-2nd-Edition/tree/main/Bonus
In this notebook, I show how to run GPT-3, GPT-3.5-turbo, and GPT-4.
Good exploration!
Basically what you do is you go to https://colab.research.google.com/ and copy each off white code block and paste each code block. Mind you that the responses are shown on the page as well in white code blocks.
I didn’t know GTP-4 would only become available after you got invited. Thanks for this. Great value!!!
1 Like
If you have access to OpenAI GPT-4 API, then it will appear.
Go through this notebook and you will see GPT-4 in the list. I ran the API with GPT-3 GPT-3. 5-turbo (ChatGPT) and GPT-4.
{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[{"file_id":"14ge8LVgd2OBTKWQrnRRaWJN8PKiEnI-2","timestamp":1678998936682},{"file_id":"1jRompDn7UP8DXAhRCimjPcC_UdNvYAe2","timestamp":1641723899457},{"file_id":"1ARBeE-LaC9UYveZbMq3tPaYy2hZ6DC1o","timestamp":1624089926713}],"authorship_tag":"ABX9TyOFQxsnYv2/hg3YkChQxrbe"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"markdown","metadata":{"id":"BPM-W5IVxfJS"},"source":["# Getting Started with Foundation Models\n","copyright 2023 Denis Rothman, MIT License\n","\n","![image.png](data:image/png;base64,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
This file has been truncated. show original
I tried it and it works. Many thanks for your contribution. I am new to programming and this is all new to me. But I could follow it. Thank you very much!
If you go through this notebook, you will see GPT-4, GPT-3.5 turbo and GPT-3 implementation for several NLP tasks:
{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[{"file_id":"14ge8LVgd2OBTKWQrnRRaWJN8PKiEnI-2","timestamp":1678998936682},{"file_id":"1jRompDn7UP8DXAhRCimjPcC_UdNvYAe2","timestamp":1641723899457},{"file_id":"1ARBeE-LaC9UYveZbMq3tPaYy2hZ6DC1o","timestamp":1624089926713}],"authorship_tag":"ABX9TyOFQxsnYv2/hg3YkChQxrbe"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"markdown","metadata":{"id":"BPM-W5IVxfJS"},"source":["# Getting Started with Foundation Models\n","copyright 2023 Denis Rothman, MIT License\n","\n","![image.png](data:image/png;base64,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
This file has been truncated. show original
Here is a step by step implementation :
{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[{"file_id":"14ge8LVgd2OBTKWQrnRRaWJN8PKiEnI-2","timestamp":1678998936682},{"file_id":"1jRompDn7UP8DXAhRCimjPcC_UdNvYAe2","timestamp":1641723899457},{"file_id":"1ARBeE-LaC9UYveZbMq3tPaYy2hZ6DC1o","timestamp":1624089926713}],"authorship_tag":"ABX9TyOFQxsnYv2/hg3YkChQxrbe"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"markdown","metadata":{"id":"BPM-W5IVxfJS"},"source":["# Getting Started with Foundation Models\n","copyright 2023 Denis Rothman, MIT License\n","\n","![image.png](data:image/png;base64,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
This file has been truncated. show original
To gain access, sign up for ChatGPT plus, for example then you can use this notebook :
{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[{"file_id":"14ge8LVgd2OBTKWQrnRRaWJN8PKiEnI-2","timestamp":1678998936682},{"file_id":"1jRompDn7UP8DXAhRCimjPcC_UdNvYAe2","timestamp":1641723899457},{"file_id":"1ARBeE-LaC9UYveZbMq3tPaYy2hZ6DC1o","timestamp":1624089926713}],"authorship_tag":"ABX9TyOFQxsnYv2/hg3YkChQxrbe"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"markdown","metadata":{"id":"BPM-W5IVxfJS"},"source":["# Getting Started with Foundation Models\n","copyright 2023 Denis Rothman, MIT License\n","\n","![image.png](data:image/png;base64,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
This file has been truncated. show original