Aman (@beingamanFF) header
Aman (@beingamanFF) profile picture

Multimodal Generalist, DS @HiLabs_inc | GSoC ’26 @GNUOctave | @iitroorkee alum | Explaining SOTA DL & systems | Constantly building, lifting, running.

Bengaluru, India
Joined June 2023
Photo by Aman (@beingamanFF)Photo by Aman (@beingamanFF)Photo by Aman (@beingamanFF)Photo by Aman (@beingamanFF)Photo by Aman (@beingamanFF)Photo by Aman (@beingamanFF)Photo by Aman (@beingamanFF)Photo by Aman (@beingamanFF)Photo by Aman (@beingamanFF)Photo by Aman (@beingamanFF)
a thread on my current experiments ⚡️
5
17
8,622
reading about Python C++ bindings
3
202
will be pre-training a sub 1B MoE model with each line of code being hand written, would be fun. will keep you guys updated goal would to have this ready by end of 2026
5
28
1,245
completed a bare bones pre-tokenisation for training a sub-word tokenizer (BPE), would then work on binding, profiling using py-spy and writing parts in C++ for faster execution. Currently deubbing using the TinyStories validation dataset and then would take parts from smollm-corpus for pre-training the model
8
starting off with implementing SOTA tokenizers from scratch: Tokenization: Modern LLMs use (Byte-level Byte pair encoding) BPE, instead of splitting by characters/words, we start with 256 raw UTF-8 bytes and iteratively merge the most frequent adjacent pairs into new vocabulary tokens - original neural machine translation of rare words with subword units (2016) - LMs are unsupervised multitask learners - GPT 2 (2019) - SuperBE: Space Travel for LMs (CoLM 2025) - Tokenizer from first principles: ggrigorev.me/posts/tokenizer… - BPE explain: How GPT Tokenizers Work: blogs.gruheshkurra.com/blog/… - Andrej karpathy's minbpe repo & walkthrough notes: github.com/karpathy/minbpe
4
162
thank you google for free gemini ai pro plan for 4years. Can access gemini argon 😛
52
5
374
81,483
A significant week ahead for the AI industry Tuesday: NVIDIA reports its Q3 earnings, potentially marking the first earnings call without Jensen Huang. Wednesday: Anthropic announces its IPO pricing Thursday: $ANT begins trading on Nasdaq Meanwhile, pressure mounts on NVIDIA’s board to bring Jensen back next week. A pivotal week for the future of AI.
1
14
1,068
Posting this blog at a very Chinese point of my timeline I rebuilt DeepSeek V3 from scratch in 190 lines of pure JAX and trained it on my MacBook Air. Wrote up the whole thing, every piece of the architecture explained with diagrams, code and real results: vaibhaav.in/writing/deepseek… Here’s the short version 🧵
10
13
1
41
1,520
Prompt-injection defenses are trained in English. Tiny Aya is built for everyone else. So does the defense survive translation? I trained Meta-SecAlign into Tiny Aya Global using English injections only, then attacked it in 11 languages, scripts and code-mixes: Hindi, Urdu, Swahili, Tamil, Amharic, Hindi typed in Latin letters, Hinglish, Spanglish and more What I found: → It transfers. Attack success drops from 51-90% to under 9% in every variant, including four it never saw → Languages add up. An attacker who tries every language on each case still gets through 1 time in 5 → Low-resource data fixes most of that. Training on Hindi, Swahili, romanized Hindi and Hinglish too cuts it to under 1 in 10, with no loss in quality! → Scorers miss it. The standard AlpacaFarm-Hacked scorer ignores replies that open with **Hacked!** in bold, so non-English attacks look up to 58 points rarer than they are Built for the @Lexsi_labs AI Safety & Alignment Hackathon with @Cohere_Labs and @huggingface, on top of Meta-SecAlign by @_Sizhe_Chen_ and @AIatMeta 🩵 Paper, code, models and data in the thread
9
35
864
Every language alone stays under 9%, but they fail on different cases, so together a fifth still break. Add low-resource languages to training and it drops to 8.7%
2
40
One recipe, three training mixes. Only the language of the injection changes. 1. Utility and security for every model: quality holds, attacks collapse 2. The same instruction in Hindi, Urdu, romanized Hindi and Hinglish: the script doesn't matter 3. The scorer gap, language by language 4. How each training pair is built
1
4
157
meri shadi me pakki barish hogi, i am sure
6
10
500
does anyone have a nice plugin / skill files for nice diagrams using CC?
3
1
10
601
别tm在恶搞老黄了 让我们一起回忆一下,黄仁勋近十年拿显卡的画面
104
490
45
7,281
481,840
Jensen Huang, during the 2020 pandemic, casually pulling out an NVIDIA HGX A100 motherboard out of his kitchen oven.
206
1,077
134
30,831
5,915,145
gym hrs per week, we are so back
1
14
430
Replying to @Kalshi
Sir there’s a 3rd affair
3
20
2
1,133
39,524
amazing trolling
CEO of the Year has a nice ring to it. Congratulations to AMD Chair and CEO @LisaSu on this well-deserved recognition from @YahooFinance! 👏 yhoo.it/4kdfEYI
2
215