@FactswithDinesh@TheVishalKay@tehseenp There will be a time when solar capacity in day exceeds demand. This has happened in Netherlands. Goverment is preparing for that. The excess solar will be used to generate hydrogen. Not coal
I was lying in a medical facility on a remote island in the South Indian Ocean, recently shipwrecked and rescued.
On the 2nd day, the French doctor rushed in:“Is the Prime Minister of India an important man?”
I thought it a silly question—until he added: “He is going to call you.”
I thought someone was pulling a fast one. But the PM did call. He recalled meeting me at his residence and asked after my health. Between hiccups, pain and burning in my chest, I told him his words gave me the courage to return to sea.
He later spoke about it in Mann Ki Baat.
It did not end there. A few days later, INS Satpura arrived. I was airlifted, and the Commanding Officer told me the PMO had been seeking updates every 15 minutes!
Wishing the Prime Minister Shri @narendramodi a very happy (belated) birthday!
@Mentatso@LottoLabs Get quantised version. Use 0 gpu offload run it off in ram. Use gpu for context. Will be slow. But it might run. Or use airllm. Significantly slower but works
@priteshlakhani Many businesses have these bizarre rules. Ridiculously large MOQs. Won't educate client about their own products. One such owner was telling me customers should pay for consultancy if they don't know exactly what to buy.
so i built a new AI architecture called CIPHER from scratch.
not based on GPT-2 tokenizer. not a fine-tune. not a wrapper.
an actual architecture - built by reading 6 research papers, finding 6 real flaws in transformers, and fixing them one by one all with help of claude assistance.
here's what's different:
→ linear O(N) recurrence instead of O(N²) attention
→ persistent memory across sequences
→ byte-level tokenizer i built myself - no GPT-2 tokenizer bias
→ sparse connections - only fires what matters
→ runs entirely on CPU. no GPU needed.
→ and it will be smaller in size.
current model: 7.6MB. 2 million parameters.
what it learned after 100 epochs:
→ single digit math: 100%
→ Devanagari (Hindi) math: 100% - responds in Devanagari script automatically
→ ७ + ७ → १४. it figured out Hindi numbers, Hindi arithmetic, and Hindi output. nobody programmed that.
→ it invented column-by-column addition on its own for multi-digit problems
→ double digit: partially working
where i found a bug live:
some Devanagari inputs were failing. i looked closely. the model wasn't broken but my input formatting was. inconsistent spacing around operators was confusing the tokenizer before the model even saw the input. ६+६ failed. ७ + ७ worked. one space. that was the entire difference.
found it. naming it. fixing it.
that's how this works. you build, you test, you find what's wrong, you fix it honestly.
the interesting part isn't that it's perfect. it's that a 7.6MB model with a byte-level tokenizer learned Hindi arithmetic at all.
most small models fail on non-English scripts because their tokenizers were built for English. CIPHER doesn't have that problem by design.
just a curious person who some papers with claude assistant, asked the right questions, and built something that works in an interesting way.
next: fix the spacing normalizer, fix double-digit carry logic, run proper benchmarks against comparable-size models and fine-tune it on proper dataset formatting and eval tests.
well atleast i can try :) fyi i'm not an AI researcher i was just curious and asked some "why" .
now let's see if it works the way i want it to. hoping for the best.
- pxlcorp labs
@abhealthin I have been trying to renew the policy but I can't login. Tried on multiple devices. Post entering the otp nothing happens. Can't renew via entering policy no. Been trying for a week. My insurance expires tomorrow. What a terrible experience I've had.