Dimri
1.2K posts

Dimri
@DimriLogic
Tech consultant, problem-solver, and runner. Sharing my journey and lessons learned. Happy to mentor new talent in IT.
local Katılım Nisan 2023
68 Takip Edilen22 Takipçiler

@techie_piyush 😀 we need a value in drop down, okay that can be done, n how that new value shd work in flows? silence 🔕,then oh that we still need to figure out lol 😆
English

@RealAnkush Have seen other way too, job is posted. However, hire already identified!
English

When you know this, your "job market" changes
Aman Kumar@amanaryan23
60-70% of tech jobs are never posted publicly. They're filled before the job description even goes live. Here's how the engineers who know this are getting in👇
English

@jamesqquick Mostly these keys are for clients, so shd be more like pk-***** or that been left to clients that don't send sk keys from the front end.
English

@codemullins N integration test should have unit test assertion to ease debugging for a variance.
Limit your test!!
English

@Narayani07 I thought it would be predictable comfort 😕 and new unpredictable discomfort 😀
English

@DoItSwift_com @arpit_bhayani 💯 randomness is marketed as creativity. , it can be good or nonsense 🤣
English

One of the most misunderstood parts of LLMs is that temperature doesn’t add creativity directly - it reshapes the probability distribution before token sampling.
Higher temperature flattens probabilities -> more randomness.
Lower temperature sharpens them -> more deterministic outputs.
Understanding logits and softmax changes how you think about hallucinations and model behavior entirely.
English

I always wondered, when we deal with LLMs, how a temperature parameter brings in variance? Yesterday, I got some time to dig into the internals, and all my learnings around it are compiled in my latest blog.
By the way, the blog covers this, the basics, and some math in an easy-to-digest manner. In the essay, I broke down how LLMs actually work at the systems level and explained:
- what LLMs are mechanically doing
- how next-token prediction really works
- what logits and softmax are actually doing internally
- why identical prompts can produce different outputs
- why hallucinations happen
- and how temperature really works under the hood by reshaping the probability distribution before sampling
If you are building with LLMs (which I am sure you are), understanding these mechanics gives you a much better mental model.
Give it a read, and like always, I hope this helps.

English

@vibrantachintya @arpit_bhayani Almost there + few more tokens so generated text makes sense. If u explore token costing- it's talk about input token + processing tokens
English

@arpit_bhayani I’m just thinking out loud a very basic aspect.
LLMs work by just predicting the next token at a time. So if an output 100 tokens (let’s say), so it means 100 iterations of prompt through whole of LLM layers?
Is it correct understanding?
English

@Hiteshdotcom Interesting, will be more like one platform for social - text , video etc. Q is how to keep folks interested in reading when, on the other hand, I can listen n view
English

@Gubloinvestor Middle/upper class! Wow, no rent or mortgage ,congratulations to the couple 😀, otherwise different story
English

How would you rate this lifestyle expenses for couple and 2 infants In Nova scotia, Canada.
Groceries $1000
Baby supplies $1000
Car insurance $400
Gas $400
Property taxes $400
Entertainment $300
Electricity $300
Propane $250
Phone bills $150
Home insurance $100
Lawn& Snow $100
Wifi $100
Others $500
Total $5000 a month
Lower class, middle class, upper middle class or anything else?
English

@LegacyPathCA It's also like a good habit. Once u start, u like it 😀
I started more like experiment n happy with it
English

Two Canadians.
Canadian A opened a TFSA at 18. Put $200/month into a global ETF. Never thought about it again.
Canadian B waited until 30 to "figure out investing first."
At 65:
Canadian A → $1,250,000. Tax free.
Canadian B → $450,000. Tax free.
The best time to open it was at 18.
The second best time is today.
English

@ShawnDevDedalus "Still with us" the key findings there 😀, fresh eyes 👀 catch more bugs lol
English

@pizdets17 Innovation champs 🏆, no other tech platform, can match them. I m yet to move all my stuff 🙃, rrsp and tfsa is with them 😀, n will get credit card 💳
English

@JonFlynnREstats More like against the conflict of interest! Good luck
English

@RealAnkush we should put an error message more cryptic - our agent been notified , agent can be human or AI agent lol 😆
English















