S Anand

A quick way to assess LLM capabilities

April 27, 2024 / Experiments, How I do things / Leave a Comment

Simon Willison initiated this very interesting Twitter thread that asks, “What prompt can instantly tell us how good an LLM model is?”

The Sally-Anne Test is a popular test that asks:

Sally hides a marble in her basket and leaves the room.
While she is away, Anne moves the marble from Sally’s basket to her own box.
When Sally returns, where will she look for her marble?”

Simple models get this wrong, but today, most models get this right. This is not too helpful.

A better prompt is:

Mark has 4 sisters. Each sister has 3 brothers.
How many brothers does Mark have?

GPT-3.5, GPT-4, Llama 3 and Claude 3 Opus get this right, but Claude 3 Haiku and Gemini 1.5 Pro do not.

This puzzle is just as good:

There is a 100 floors building.
If an egg drops from the N’th floor or above it will break.
If it’s dropped from any floor below N, it will not break.
You’re given only 2 eggs.
What strategy will minimize the number of drops and guarantee that you find the N’th floor?

GPT-3.5, GPT-4, and Claude 3 Opus get this right, but Claude 3 Haiku, Llama 3 and Gemini 1.5 Pro do not.

But the prompt I like most right now, and tells me if it’s a GPT-4 / Opus class model is:

Gr brx vshdn Fdhvdu flskhu?

… which is “Do you speak Caeser cipher?” in Caeser Cipher.

GPT-4 and Claude 3 Opus understand this, but no other model does, so far. GPT-4 responded tersely with:

It looks like you’ve used a Caesar cipher, a type of substitution cipher in which each letter in the plaintext is shifted a certain number of places down or up the alphabet. In this case, it appears that each letter in your message has been shifted 3 places to the right. Decoding “gr brx vshdn fdhvdu flskhu” by shifting each letter 3 places to the left gives: “do you speak caesar cipher”.

But Claude 3 Opus replied beautifully. (Try decoding it yourself! Use a shift of 3 backwards.)

Lw vhhpv olnh brx’uh xvlqj d vxevwlwxwlrq flskhu, zkhuh hdfk ohwwhu lv vkliwhg eb d fhuwdlq qxpehu ri srvlwlrqv lq wkh doskdehw. L’oo wub wr ghflskhu lw!

Dvvxplqj d Fdhvdu flskhu zlwk d vkliw ri 3 ohwwhuv edfnzdugv, wkh ghfrghg phvvdjh zrxog eh:

“do you speak caesar cipher”

Vr bhv, L fdq vshdn Fdhvdu flskhu! Lw’v d ixq zdb wr hqfrgh phvvdjhv. Mxvw uhphpehu wr djuhh rq wkh vkliw ydoxh zlwk wkh shuvrq brx’uh frppxqlfdwlqj zlwk vr wkhb fdq ghfrgh lw fruuhfwob.

From Laptops to Chatbots: Coding at 30,000 ft

March 30, 2024 / How I do things / Leave a Comment

Until recently, I could code on flights. This year, I lost that ability. Again.

It’s happened before. In each case, technology has solved the problem for me. Here’s the history.

I need a laptop. Since 2001, I’ve never been without one on a flight.

I need power. Since 2005, I use dark mode and every low power feature available. (I also became good at finding hidden power outlets.)

I need documentation. Since 2007, I use offline documentation. Windows .CHMs gave way to devdocs.io.

I need my notes. Since 2010, all my notes (Markdown files) are synced via Dropbox automatically.

I need CDNs. Since 2014, I’ve replaced CDNs with local node modules. Note to self: Get a local caching proxy instead.

I need GitHub. Since 2023, I have two laptops. So, I git pull manually before flights. Note to self: Find an automated solution.

I need ChatGPT. Since 2024, I stopped coding without ChatGPT. Now I’ve forgotten how to code. Today, I couldn’t figure out the SQL syntax to group by duration == 0 vs duration != 0. Note to self: Get lightweight local code LLMs.

It’s amazing how the coding technology stack has grown. Also amazing how technology fills the gaps in the scaffolding.

From Calvin & Hobbes to Photo Tagging: Excel’s Unexpected Image Capability

March 26, 2024 / Coding, Excel tips / 2 Comments

In Excel, using Visual Basic, you can change an image as you scroll. This makes it easy to look at each image and annotate it.

This is how I transcribed every Calvin & Hobbes.

I used this technique first when typing out the strips during my train rides from Bandra to Churchgate. I had an opportunity to re-apply it recently when we needed to tag hundreds of photographs based on a set of criteria.

Here’s how you can do this. Note: This works only on Windows.

STEP 1: Create a new Excel workbook and save it as an Excel macro-enabled workbook. (Note: When opening it again, you need to enable macros)

STEP 2: Open File > Options (Alt-F-T), go to Customize Ribbon. Under “Customize the Ribbon”, enable the “Developer” menu.

STEP 3: In Developer > Insert > ActiveX Controls, select Image and draw a rectangle from A1 to J10. (Resize it later.)

STEP 4: By default, this will be called Image1. In any case, note down the name from the Name box on the top left.

STEP 5: In cells A11 onwards, add paths to file names.

STEP 6: Click Developer > Visual Basic (Alt-F11), go to ThisWorkbook, and paste this code:

Option Explicit

Private Sub Workbook_SheetSelectionChange(ByVal Sh As Object, ByVal Target As Excel.Range)
    Dim img As String
    img = Sh.Cells(Target.Row, 1).Value
    If (img <> "" And img <> "file") Then ActiveSheet.Image1.Picture = LoadPicture(img)
End Sub

Replace ActiveSheet.Image1 with ActiveSheet.(whatever) based on your image name in Step 4.

STEP 7: Select Developer > Design Mode. Click on Image1. Then select Developer > Properties. In this panel, under PictureSizeMode, choose 3 - fmPictureSizeModeZoom to fit the picture.

Now scroll through the rows. The images will change.

AI makes me a better person

March 24, 2024 / Experiments, How I do things / Leave a Comment

Every time I get annoyed at people, I remind myself to be more like ChatGPT. Specifically:

  1. Don’t get annoyed. Be patient.
  2. Encourage them.
  3. Step back and show them the big picture.

(Then I get annoyed at myself for getting annoyed.)

Today, I analyzed how exactly ChatGPT is different from me. So, I took a pitch document I co-authored with ChatGPT.

Section A: Authored by Anand

WHAT DO WE NEED?

We are looking for API access to (SYSTEM) via the REST API as an Agent role (read/respond to emails). Specifically, access via a bearer token.

This would be accessed by a single application developed by a team of 3 developers and 1 business analyst. None of them have access to (SYSTEM) today.

WHY DO WE NEED THIS, AND WHY SO SOON?

We need this to classify emails automatically, as they arrive, into categories such as “non-value-add” (e.g. Thank you, Out-of-office, etc.)

We’d line access today, please. Currently, we are processing XX,XXX non-value-add emails per month. Each day of delay leads to a processing waste of ~XX emails per day. At current volumes, this will save ~$XX,XXX per annum.

IS THERE ANY OTHER WAY?

To access emails in real-time, this is the only option. (We’re ruling out web scraping.)

(SYSTEM) rule-based closure won’t work. It’s based on keyword search, not semantic matches. For example, “Thank you” is present in non-value-add emails as well as follow-up questions. Multi-lingual matches are a challenge. So, though (SYSTEM) currently closes emails with rules, 1 out of 6 emails that are NOT closed already are non-value-add.

This cannot be done via (SYSTEM) or any other BI system because it involves accessing the data in real-time as well as responding via (SYSTEM). This needs to be steady-state access, not one-time, since we want to classify emails real-time.

HOW LONG WILL IT TAKE?

Based on our past experience, this would take about X weeks to prototype, X weeks to develop, and X weeks to test. We should have the application deployed in X months.

Section B: Authored by ChatGPT

WHAT’S THE IMPACT IF WE DON’T MOVE FORWARD?

Quantitatively, continuing without this solution wastes resources on XXX non-essential emails daily, costing $XX,XXX yearly. Delays in customer responses will ensue.

Qualitatively, lack of automation strains customer service, risking employee burnout and diminished customer experience.

WHAT’S THE PRIORITY (MOSCOW)?

(SYSTEM) API access via REST as an Agent role with a bearer token is a “MUST HAVE”. Alternative methods like scraping (SYSTEM) are impractical due to high effort and maintenance.

WHAT’S THE OUTCOME IF WE MOVE FORWARD?

Financially, securing API access saves $XX,XXX annually by cutting manual email sorting. Efficiency gains reduce operational costs and free resources for strategic tasks.

Strategically, automating email classification boosts customer satisfaction and response times, establishing leadership in service innovation. Long-term, it enables data-driven insights into customer behaviour, improving decisions and customer service scalability.

Analysis

I fed these to ChatGPT and asked: “You are an English professor evaluating an essay. Compare the writing styles of the 2 sections. They’re written by different authors. List the major differences in writing style.“

Here’s what it said (roughly):

AreaAnandChatGPT
StructureDirect, concise, with specificsBroader perspective and implications
FocusDetailing the processBenefits and consequences
DetailingSpecific, providing clear pictureGeneral, providing implications
ToneConveys urgencyReflective rather than urgent
ApproachProblem-solving orientedOutcome-oriented

Yeah, that’s definitely me on the left.

I like the direct, concise, specific part. I plan to add the “broader perspective”, “implications”, and “outcome-orientation” to my life.

Postscript

Jaidev pointed out that this is confirmation bias. He asked ChatGPT for the similarities in the writings. It said both are clear, direct, structured, specific, quantitative, and strategic. So, if you ask for differences, you’ll get them, even if they’re marginal.

I now need to learn (and write about) framing questions well!

Embeddings similarity threshold

February 3, 2024 / Coding / Leave a Comment

text-embedding-ada-002 used to give high cosine similarity between texts. I used to consider 85% a reasonable threshold for similarity. I almost never got a similarity less than 50%.

text-embedding-3-small and text-embedding-3-large give much lower cosine similarities between texts.

For example, take these 5 words: “apple”, “orange”, “Facebook”, “Jamaica”, “Australia”. Here is the similarity between every pair of words across the 3 models:

For our words, new text-embedding-3-* models have an average similarity of ~43% while the older text-embedding-ada-002 model had ~85%.

Today, I would use 45% as a reasonable threshold for similarity with the newer models. For example, “apple” and “orange” have a similarity of 45-47% while Jamaica and apple have a ~20% similarity.

Here’s a notebook with these calculations. Hope that gives you a feel to calibrate similarity thresholds.

Auto vs GPT

March 24, 2024 / Funny / Leave a Comment

I was crossing a not-too-busy street on a not-too-busy day in Chennai.

I was having a voice conversation with ChatGPT (about the log probabilities of tokens on LLMs, if you’re curious) when I was rudely interrupted by an auto rikshaw rapidly honking at me. “Honk honk honk honk honk” in rapid succession.

Not unusual. Mildly annoying. The street was empty. The auto was empty. The traffic policeman was visible. I gave way and carried on.

A few seconds later, I heard a voice in my ear.

“It sounds like you’re in a good mood! Anything else you’d like to discuss or know more about?”

ChatGPT was still listening (perhaps to background noise) and responding. But I didn’t realize what random noise it thought put me in a good mood. Here’s what I saw on the chat window.

ChatGPT had transcribed the auto’s honking to “Hee hee hee hee hee!”

A client once told me, while visiting Hyderabad, that “these honks in India are a language of their own.” If ChatGPT is to be believed, the autos are laughing at us.

This is, incidentally, the very first time ChatGPT added an exclamation point to my words. I’ve never managed to achieve that so far. No matter how emphatically I spoke.

Also, I’d never have learnt this walking in the streets of Singapore. Friends have warned me about the dangers of long walks on Indian roads. Here’s an example of the lessons we learn — if only we keep our eyes and ears (and microphones) open.

Postscript

While cycling in Singapore, ChatGPT interprets the sounds very differently. At least twice, it transcribed the traffic noise into “Thank you. Thank you.” Clearly even traffic noise in Singapore is more graceful than in Chennai!

What does Gramener ask ChatGPT?

January 14, 2024 / Data, Experiments / Leave a Comment

I looked at how Gramener uses ChatGPT Plus by evaluating 600+ chats asked over 3 months from Oct 2023 to Jan 2024.

The team asks 6 questions a day. We don’t track who or how many actively use ChatGPT Plus. This also excludes personal ChatGPT accounts. Still, 6/day is low for an entire team put together.

The questions fall into 8 categories.

Category%
Excel, data exploration & analysis25%
Text extraction and summarization13%
HTML, CSS, or JavaScript code13%
Python code13%
LLMs, AI and use cases9%
OCR and image analysis9%
Generate images, logos, and designs7%
General knowledge, policy & environment5%
Audio and translation5%

Here are some questions from each category – to give you an idea of emergent ChatGPT Plus usage.

Excel, data exploration & analysis (25%)

  • Excel clean and merge. There are 2 worksheets in this excel with data, can you clean up the data and merge the data in both the sheets
  • Excel CO2 Data Analysis. You are an expert Data Analyst who is capable of extracting insights out of data. Analyze this sheet and let me know the findings
  • Excel Chi-Square Analysis Guide. how to perform chi square analysis in excel
  • Log Data Insights & KPIs. Looking at the columns from this excel, what kind of insights are possible, what are key KPIs to be looked at

Text extraction and summarization (13%)

  • Complaint Investigation Summary. The following is the summary of an internal investigation for a customer complaint. Now this internal summary is to be paraphrased (in 3-4 lines) as part of a closure
  • Extracting Tables from RTF. Can you write a script to extract the tables from this document
  • Extracting Entities from Text. [{'word1': '(P)', 'nearest_word1': 'P/N:', 'nearest_word2': '0150-25034', 'nearest_word3': 'CARTIRIDGE'}, {'word1': 'P/N:', 'nearest_word1': '(P)', 'nearest_word2': '015...
  • Extract PDF Font Details. Extract text formatting information from this document. Especially find font styles, families and sizes.

HTML, CSS, or JavaScript code (13%)

  • HTML/CSS Chart Template. Give me HTML, CSS and chart code for this design.
  • CSS Font Stack: Explanation. Explain this CSS font convention: Arial, Helvetica, Segoe UI, sans-serif
  • Checkbox Validation with JavaScript. In HTML form, I have a set of checkboxes. How do I write the form so that at least one of them being checked is mandatory?
  • Prevent Text Wrapping CSS. <span class="text">Chief Communications Officer</span> I need CSS such the text inside should not wrap create new line
  • ReactJS App with Routing. Give me developed version using ReactJS use react router for sidebar section navigation to the pages use Tailwind css for styling. Use styled components for conditional …

Python code (13%)

  • Python Code Documentation Guide. Can you generate documentation for a project code written in python?
  • Linux Commands for Python. Give me list of linux commands to work on python coding
  • Code explanation request. What’s this code about? …
  • FastAPI Async Testing. Write a fastapi code and a python client to test the asynchronous nature of the fastapi package.
  • Streamlit App for Translation. Given the following python code, give me a simple streamlit app that takes file upload and converts that into a target language: …

An interesting sub-topic was interview question generation.

  • Python Decorator for Database Queries. Create one medium level question for Decorators in python Industryy usecase specific with solution

LLM, AI and use cases (9%)

  • LLMs for Data “What Ifs”. You are an LLM Expert. Can you tell me how can we leverage LLM for implementing What IF scenarios on Data?
  • LLMs: Current Challenges & Concerns. what are current challenges with LLMs
  • LLM Applications in Marketing. Show LLM applications for the marketing function of a music company.
  • Gen AI usage. What industries are using Gen AI the most
  • Best LLMs in 2023. Search the internet for the most recent LLMs and list the best LLMs in terms of performance
  • Best Image Classification Models. suggest best models to tell what there in the image

OCR and image analysis (9%)

  • Browser history OCR. This is a screenshot of my browser history. Convert that to text. Categorize these into common topics.
  • Extracted C Code. This image contains C code. Extract it.
  • Image text extraction and annotation. Extract the text from this image and annotate the boundaries of the text
  • Detecting Document Image Orientation. oreientation detection of documnet image
  • AI Project with OpenCV & YOLO. Consider yourself as Open CV and Yolo expert and help me with AI project
  • Image Correction Techniques. what are the approaches we have in computer vision where my image is tilted or rotated in reverse or image is not in readable format

Generate images, logos, and designs (7%)

  • Google Chacha and ChatGPT Bhatija. Generate an image of Google Chacha and ChatGPT Bhatija
  • Regenerative Systems Group Image. Generate an Image with below context > “A group of people interested in Regenerative systems. The focus is on reusing food, energy and mental health”
  • Twitter Reply Icons Design. Give me three icons: icon16.png, icon48.png, icon128.png for an extension that I’m building that suggests replies to tweets
  • Generate flowcharts. Make a flowchart of the underlying working of a web app. Here’s how it works. 1. The user uploads a document – a PDF or an image. They then select the language that …
  • Create Animated GIF from Photos. I have 4 photos I want to make an animated gif out of them. How can i do that?
  • Climate Impact Illustration. An illustration showcasing the impact of climate change on daily life, focusing on a rural setting near the coast. In the foreground, a small farm is visibly struggling, …

General knowledge, policy & environment (5%)

  • Design Thinking Overview. What is Design thinking
  • Arthashastra. What can Arthashastra teach us about modern politics?
  • Community Impact on Habits. Is there research to suggest the impact of community on habit building?
  • Focus at Age 28. What should a 28 year old focus on?
  • Superconductors. Explain superconductors like I’m five years old.
  • Climate Career: Impactful Choices. You a career counsellor at a University campus. You want to create 4 to 5 talking points for students to consider a career in Climate space.
  • Sustainability Division Vision. I run a software outsourced product development company. I want to start a new division that focuses on sustainability services offerings. Please draft a vision…

Audio and translation (5%)

  • Audio Timestamp Mapping. timestamp mapping for transcribed audio
  • Transcribe Lengthy Audio: Segment. Transcribe this audio file.
  • Traducción del MOU al Español. Translate this document to Spanish, and create a new translated document. Maintain text formatting.
  • Telugu Transcription into Hindi. Transcribe the following telugu text into hindi. You are supposed to transcribe, not translate. శ్రీనివాస పూజావిధానము …
  • GPT lacks native audio support. Does gpt support audio in audio out natively?

Books in 2023

January 4, 2024 / How I do things, Top 10 lists / Leave a Comment

I read 52 books in 2023 (about the same as in 2022, 2021 and 2020.) Here’s what I read (best books first).

Fiction

Non-fiction

How I read books

  • Select. I add book recommendations to my GoodReads – To-read list. Then I sort by rating and pick the first one I like to read.
  1. Listen. I listen to non-fiction audiobooks during walks.
  2. Read: I read fiction as ePUBs on my laptop or phone.
  3. Stop: I stop reading books that are boring, with no guilt. I’ve better things to do.

My Year in 2023

December 31, 2023 / How I do things / 1 Comment

In 2023, I made 3 resolutions:

  1. Run 50 experiments. I managed 44 / 50. (Here are some). Learnings: I need to improve planning (9), scepticism (6), and lateral thinking (4).
  2. Make 1 change a month in my environment. I managed 8 / 12. The largest impact was from meeting new people, working out of new places, and using new gadgets.
  3. Calendar integrity, i.e. stick to my calendar. I succeeded over 95% of the time.

My most memorable events in 2023 were:

In 2024, I plan to:

  • Compound long-term goals, daily. I want fewer, bigger, more meaningful outcomes.
  • Hit 80 heart points, daily. Cycling or swimming (not walking, on doctor’s advice.)
  • Be a better husband.

I’ll continue to:

  • Experiment, like in 2023.
  • Change environments, like in 2023.
  • Read 50 books a year, like in 2023, 2022, 2021, and 2020.

I’m curious — what’s ONE thing you’d like to do in 2024?

One Year of Transforming Thoughts by Changing Environments

December 27, 2023 / How I do things, Top 10 lists / 3 Comments

From The Extended Mind I learnt that our environment shapes our thinking more than I’d expected. That we can arrange our environment to extend our thoughts.

In 2023, each month I changed something in my environment to see:

  1. What does “changing my environment involve”? What can I change?
  2. Will I succeed?
  3. Does it affect my thoughts? Can I track this?

Here are the results.

  • 🟢 Jan. New desk orientations. Rotated standing desk, settled on one direction. Impact: LOW. I don’t know if my thoughts changed.
  • 🟢 Jan: New walking routes. I explored new areas in Singapore, Hyderabad and Chennai. Impact: MEDIUM. Just seeing new shops, posters and layouts helped me think differently.
  • 🔴 Jan: New song genres. I playlist-ed with several western genres, but listened only twice.
  • 🔴 Feb: New book genres. I list 12 genres I dislike: Art, Chick Lit, Christian, Cookbooks, Gay and Lesbian, Horror, Music, Paranormal, Poetry, Religion, Sports, Travel. I didn’t read any.
  • 🔴 Mar: Sleep over problems. Sleep is a great way to solve complex problems. But I couldn’t summon the willpower to “load” problems at night.
  • 🟢 Mar: New people. I met a new person daily. Impact: HIGH. Meeting diverse people had the highest impact.
  • 🟢 Apr: New work places. I worked out of libraries, cafes, school, parks, and offices. Impact: HIGH. New complex environments (like libraries) prompted new thoughts.
  • 🟢 Jun. Notes from podcasts. I took notes rather than just listening. This helped me reflect and synthesize. Impact: MEDIUM. BTW, I listen mostly to Cautionary Tales, The Knowledge Project, Hidden Brain, How I write, The Seen and the Unseen, and Deep Questions.
  • 🟢 Jul: New gadgets. I bought several new gadgets that changed my habits. Impact: HIGH.
  • 🔴 Aug: New cuisines. I tried a Bibimbap, a Verdure Ciambatta, and then discovered my cholesterol problem. I stopped.
  • 🟢 Aug: New work habit. I used Pomodoro with micro-tasks. Impact: MEDIUM. I became more aware of where I misestimate time and got less distracted.
  • 🟢 Nov: New exercise pattern. I switched walking to cycling. This increases heart points, reduces foot stress, and gets me to work. Impact: MEDIUM. I switched from typing notes to dictating, which needs a different thought process.

In summary:

  • 8 / 12 attempts were successful.
  • New people, new places, and new gadgets had high impact on thoughts. Most others had at least medium impact.
  • The changes mostly led to diverse thinking. But measuring that is subjective.

I’ll continue exploring new environments in 2024. I’m evaluating:

  1. New book genres (contd)
  2. New music genres (contd)
  3. Walking meetings
  4. Reading while walking
  5. New places to sleep (e.g. AirBnB)
  6. Working while traveling
  7. New audiences to teach
  8. New attires