Search through all your Tweets using GPT

Here's how you can do this for your own account:

Request your twitter archive
Download archive, unzip, and open Your archive.html in Chrome.
Go to the tweets section and scroll to the bottom.
Save page as html (Web Page, Complete)
Upload it here

Search Query

Documents

Submit Links in Bulk

⚙️ Settings

👩‍🏫 Task Instructions

🔠 Language Model Settings

ChatGPT (openai)

Avoid Repetition

Answer Outputs

How many answers should the copilot generate? Additional answer outputs increase the cost of each run.

Max Output Tokens

The maximum number of tokens to generate in the completion. Increase to generate longer responses.

Creativity (aka Sampling Temperature)

Higher values allow the LLM to take more risks. Try values larger than 1 for more creative applications or 0 to ensure that LLM gives the same answer when given the same user input.

🔎 Document Search Settings

Citation Style

Numbers ( [1] [2] [3] ..)

Create Synthetic Data

To improve answer quality, pick a synthetic data maker workflow to scan & OCR any images in your documents or transcribe & translate any videos. It also can synthesize a helpful FAQ. Adds ~2 minutes of one-time processing per file.

———

👁‍🗨 Conversation Summarization

These instructions run before the knowledge base is search and should reduce the conversation into a search query most relevant to the user's last message.

Embeddings Model

Text Embedding 3 Large (OpenAI)

Dense Embeddings Weightage

Weightage for dense vs sparse embeddings. 0 for sparse, 1 for dense, 0.5 for equal weight.
Generally speaking, dense embeddings excel at understanding the context of the query, whereas sparse vectors excel at keyword matches.

Max Citations

The maximum number of document search citations.

Max Snippet Words

After a document search, relevant snippets of your documents are returned as results. This setting adjusts the maximum number of words in each snippet. A high snippet size allows the LLM to access more information from your document results, at the cost of being verbose and potentially exhausting input tokens (which can cause a failure of the copilot to respond). Default: 300

Overlapping Snippet Lines

Your knowledge base documents are split into overlapping snippets. This settings adjusts how much those snippets overlap. In general you shouldn't need to adjust this. Default: 5

Run cost = 4 credits

Breakdown: 1 (ChatGPT (openai)) + 3/run

By submitting, you agree to Gooey.AI's terms & privacy policy.

Based on the provided search results, it seems that Dev Aggarwal has expressed an interest in Erlang, mentioning the Erlang guys and their work. However, without further context, it's difficult to definitively say whether he likes Erlang or not. As an AI, I can't read minds, but it looks like there's some curiosity there. Maybe Dev Aggarwal is Erlang-curious? Who knows, maybe he'll join the Erlang fan club soon!^[1]

💁‍♀️ Sources

1. devxpy twitter replies.html

?

0

2

ID1062324608039378945

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 10, 2018

@pmylund[] @garybernhardt[] The #Erlang[] guys seem to have done a lot
of these years ago XD?

0

1

ID1061038235697201152

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 6, 2018

@garybernhardt[] @jamie_gaskins[] Maybe this can be done by sending a
little itty bitty piece of JS code along with the original webpage? Will
take a shot at it, if you haven't found.

?

0

0

ID1059841916374839301

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 6, 2018

@thdxr[] @garybernhardt[] Ah yes, forgot that Erlang has the best
implementation of green processes out there!

?

0

1

ID1059782585403035654

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 6, 2018

@thdxr[] @garybernhardt[] Bokeh seems to have done quite well, with its
use of twisted to handle large no of clients, but it isn't the most fun
to use?

0

0

ID1059708595967680512

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 6, 2018

@thdxr[] @garybernhardt[] We have a bunch of projects doing this in
python (especially for data visualisation), the only issue with them is
stability and scalability.

?

...

0

0

ID1080812500424237057

View on Twitter[]

[]

Dev Aggarwal @devxpy Dec 31, 2018

@develobile[] Holy...this is big!

?

0

2

ID1079773913628971009

View on Twitter[]

[]

Dev Aggarwal @devxpy Dec 31, 2018

@indygreg[] This is just what this language needs! Painless deployment
and distribution...

?

0

1

ID1079530667459911681

View on Twitter[]

[]

Dev Aggarwal @devxpy Dec 23, 2018

@ericsnowcrntly[] (That article describes how erl interpreter can do
preemptively switched multicore green processes using multiple
interpreter instances running.)?

0

0

ID1076549633567047683

View on Twitter[]

[]

Dev Aggarwal @devxpy Dec 23, 2018

@ericsnowcrntly[] I canâ€™t help but notice the similarity with Erlang.
I really hope we can have some functionality from that beauty over in
python. https://t.co/yDO22yrtuB[]?

0

1

ID1076549176421429248

View on Twitter[]

[]

Dev Aggarwal @devxpy Dec 16, 2018

@garybernhardt[] That really reminds me of browserify. AFAIK it was a
joy using it.

?

...

0

0

ID1066309903403413505

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 24, 2018

@joeerl[] What about sharing by communicating?

?

0

0

ID1066071142643974144

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 23, 2018

@joeerl[] â€œJS gets fixedâ€? might be the right way to put it :D?

0

0

ID1065965125004390400

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 20, 2018

@panzer[] @freakboy3742[] (Cursor placement is also magnified by
default)?

0

0

ID1064804769162821633

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 20, 2018

@panzer[] @freakboy3742[] Android P now has magnifier as a default when
selecting any text. And yes, the implementation is spectacular.

?

0

0

ID1064803644338552832

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 13, 2018

@dabeaz[] I hope these are available for viewing on the internet. My
brain needs some shattering too, please.

?

0

2

ID1062324608039378945

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 10, 2018

@pmylund[] @garybernhardt[] The #Erlang[] guys seem to have done a lot
of these years ago XD?

[1] https://storage.googleapis.com/dara-c1b52.appspot.com/daras_ai/media/435abe4a-b726-11ed-8dca-02420a0001fd/devxpy%20twitter%20replies.html

Generated in 1.9s on

...

ℹ️ Details

👣 Steps

Final Search Query

References

[1 Items

{

…

}4 Items

]

Final Prompt

Search Result: [1]
Title: """devxpy twitter replies.html"""
Snippet: """
?

ID1062324608039378945

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 10, 2018

@pmylund[] @garybernhardt[] The #Erlang[] guys seem to have done a lot
of these years ago XD?

ID1061038235697201152

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 6, 2018

@garybernhardt[] @jamie_gaskins[] Maybe this can be done by sending a
little itty bitty piece of JS code along with the original webpage? Will
take a shot at it, if you haven"t found.

ID1059841916374839301

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 6, 2018

@thdxr[] @garybernhardt[] Ah yes, forgot that Erlang has the best
implementation of green processes out there!

ID1059782585403035654

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 6, 2018

@thdxr[] @garybernhardt[] Bokeh seems to have done quite well, with its
use of twisted to handle large no of clients, but it isn"t the most fun
to use?

ID1059708595967680512

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 6, 2018

@thdxr[] @garybernhardt[] We have a bunch of projects doing this in
python (especially for data visualisation), the only issue with them is
stability and scalability.

...

ID1080812500424237057

View on Twitter[]

[]

Dev Aggarwal @devxpy Dec 31, 2018

@develobile[] Holy...this is big!

ID1079773913628971009

View on Twitter[]

[]

Dev Aggarwal @devxpy Dec 31, 2018

@indygreg[] This is just what this language needs! Painless deployment
and distribution...

ID1079530667459911681

View on Twitter[]

[]

Dev Aggarwal @devxpy Dec 23, 2018

@ericsnowcrntly[] (That article describes how erl interpreter can do
preemptively switched multicore green processes using multiple
interpreter instances running.)?

ID1076549633567047683

View on Twitter[]

[]

Dev Aggarwal @devxpy Dec 23, 2018

@ericsnowcrntly[] I canâ€™t help but notice the similarity with Erlang.
I really hope we can have some functionality from that beauty over in
python. https://t.co/yDO22yrtuB[]?

ID1076549176421429248

View on Twitter[]

[]

Dev Aggarwal @devxpy Dec 16, 2018

@garybernhardt[] That really reminds me of browserify. AFAIK it was a
joy using it.

...

ID1066309903403413505

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 24, 2018

@joeerl[] What about sharing by communicating?

ID1066071142643974144

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 23, 2018

@joeerl[] â€œJS gets fixedâ€? might be the right way to put it :D?

ID1065965125004390400

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 20, 2018

@panzer[] @freakboy3742[] (Cursor placement is also magnified by
default)?

ID1064804769162821633

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 20, 2018

@panzer[] @freakboy3742[] Android P now has magnifier as a default when
selecting any text. And yes, the implementation is spectacular.

ID1064803644338552832

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 13, 2018

@dabeaz[] I hope these are available for viewing on the internet. My
brain needs some shattering too, please.

ID1062324608039378945

View on Twitter[]

[]

Dev Aggarwal @devxpy Nov 10, 2018

@pmylund[] @garybernhardt[] The #Erlang[] guys seem to have done a lot
of these years ago XD?
"""

Generate a comprehensive, factoid Answer the for the following Question soely based on the provided Search Results. If the Search Results do not contain enough information, say "I don't know". Use an unbiased, succinct, and funny tone. Use this current date and time: January 23, 2024 16:18:45 . Combine Search Results together into a coherent answer. Remember to cite the search results using [${number}] notation in your answer.

Question: Does he like erlang?
Answer:

Output Text

How to Use This Recipe

🙋🏽‍♀️ Need more help? Join our Discord

Related Workflows

Browse the web using ChatGPT

Like Bing + ChatGPT or perplexity.ai, this workflow queries Google and then summarizes the results (with citations!) using an editable GPT3 script. Filter results to your own website so users can ask anything and get answers based only on your site's pages.

AI Generated Photo from Email Profile Lookup

Find an email's public photo and then draw the face into an AI generated scene using your own prompt + the latest Stable Diffusion or DallE image generator.

Create a perfect SEO-optimized Title & Paragraph

Input a Google search query + your website & keywords to get AI search engine optimized content. This workflow parses the current top ranked sites and generates the best page summary for your site using OpenAI’s GPT3.

Copilot Builder

Create customized chatbots from your own docs/PDF/webpages. Craft your own bot prompts using the creative GPT3, fast GPT 3.5-turbo or powerful GPT4 & optionally prevent hallucinations by constraining all answers to just your citations. Available as Facebook, Instagram, WhatsApp bots or via API. Add multi-lingual speech recognition and text-to-speech in 100+ languages and even video responses. Collect 👍🏾 👎🏽 feedback + see usage & retention graphs too! This is the workflow that powers https://Farmer.CHAT and it's yours to tweak.

Search through all your Tweets using GPT

Search Query

Documents

👩‍🏫 Task Instructions

🔠 Language Model Settings

Answer Outputs

Max Output Tokens

Creativity (aka Sampling Temperature)

🔎 Document Search Settings

Citation Style

Create Synthetic Data

👁‍🗨 Conversation Summarization

Embeddings Model

Dense Embeddings Weightage

Max Citations

Max Snippet Words

Overlapping Snippet Lines

👣 Steps

How to Use This Recipe

Related Workflows

Browse the web using ChatGPT

AI Generated Photo from Email Profile Lookup

Create a perfect SEO-optimized Title & Paragraph

Copilot Builder

GET STARTED

LEARN

DEVELOPERS

SOCIAL

CONNECT

EXTRAS