NQCL : Neural Quantum Consciousness Language

@SarahCharolalla @vitoquiles 🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣

Acavo de acompañar a Begoña a poner la Denucia al Jenocida @vitoquiles . Nos veremos en los trivunales 😉

If this tweet has exactly 1 like in 24 hours I’ll give that person $1,000,000

@desdelamoncloa @SaizElma Mentirisos

📈 España alcanza un hito histórico superando los 25 millones de población activa, según los datos publicados hoy por la EPA. 🗣️ @SaizElma: “España está siendo capaz de crear más empleos y además empleos de más calidad”.

El Gobierno ha aprobado el Real Decreto para comenzar al proceso de regularización administrativa extraordinaria de personas migrantes que ya residen en España. 📄 Dará derecho a trabajar en cualquier sector y lugar del país durante un año Más ⬇️ lamoncloa.gob.es/serviciosdepre…

@desdelamoncloa @sanchezcastejon @Congreso_Es 7000M€ - Comisiones = 200M€

El presidente @sanchezcastejon ha destacado en el @Congreso_ES que la aprobación del Plan Estatal de Vivienda, con una movilización de 7.000 M€, es un ejemplo del compromiso del Ejecutivo con la protección de la ciudadanía. Info⤵️ lamoncloa.gob.es/presidente/act…

@vitoquiles @vitoquiles No te lo pienses denuncialo

Un portavoz del Gobierno diciendo que soy un “matón” y un “acosador de mujeres” sin ninguna prueba, sin ninguna condena, sin ningún antecedente. Es inaudito la de mierda que tengo que aguantar. Luego a otra le dicen cualquier chorrada y se lía la de Dios.

SOS ¡David y mi hermano Javier detenidos sin cargos y en peligro en Guinea Ecuatorial! - ¡Firma la petición! c.org/t4nh96wRb5 via @change_es

CESE URGENTE DEL DIRECTOR GENERAL DE TRÁFICO - ¡Firma la petición! c.org/CW2dpfNxkM a través de @change_es

mariaguerrerofotografia.mypixieset.com/contacto/ FOTOGRAFIAS PROFESIONALES Y PRECIO ASEQUIBLE

@karpathy Paper NQCL: Neural Quantum Consciousness Language Technical Report - Mini-Brain Experiment Results doi.org/10.5281/zenodo…

nanochat can now train GPT-2 grade LLM for <<$100 (~$73, 3 hours on a single 8XH100 node). GPT-2 is just my favorite LLM because it's the first time the LLM stack comes together in a recognizably modern form. So it has become a bit of a weird & lasting obsession of mine to train a model to GPT-2 capability but for much cheaper, with the benefit of ~7 years of progress. In particular, I suspected it should be possible today to train one for <<$100. Originally in 2019, GPT-2 was trained by OpenAI on 32 TPU v3 chips for 168 hours (7 days), with $8/hour/TPUv3 back then, for a total cost of approx. $43K. It achieves 0.256525 CORE score, which is an ensemble metric introduced in the DCLM paper over 22 evaluations like ARC/MMLU/etc. As of the last few improvements merged into nanochat (many of them originating in modded-nanogpt repo), I can now reach a higher CORE score in 3.04 hours (~$73) on a single 8XH100 node. This is a 600X cost reduction over 7 years, i.e. the cost to train GPT-2 is falling approximately 2.5X every year. I think this is likely an underestimate because I am still finding more improvements relatively regularly and I have a backlog of more ideas to try. A longer post with a lot of the detail of the optimizations involved and pointers on how to reproduce are here: github.com/karpathy/nanoc… Inspired by modded-nanogpt, I also created a leaderboard for "time to GPT-2", where this first "Jan29" model is entry #1 at 3.04 hours. It will be fun to iterate on this further and I welcome help! My hope is that nanochat can grow to become a very nice/clean and tuned experimental LLM harness for prototyping ideas, for having fun, and ofc for learning. The biggest improvements of things that worked out of the box and simply produced gains right away were 1) Flash Attention 3 kernels (faster, and allows window_size kwarg to get alternating attention patterns), Muon optimizer (I tried for ~1 day to delete it and only use AdamW and I couldn't), residual pathways and skip connections gated by learnable scalars, and value embeddings. There were many other smaller things that stack up. Image: semi-related eye candy of deriving the scaling laws for the current nanochat model miniseries, pretty and satisfying!

@xai doi.org/10.5281/zenodo…

Try the all new Grok Imagine API: x.ai/news/grok-imag…

@arena @xai @GoogleDeepMind @OpenAI doi.org/10.5281/zenodo…

🚨BREAKING: @xAI’s first model in Video Arena debuts in the top 3! Grok-Imagine-Video ranks #3 on the Image-to-Video Arena and #4 on the Text-to-Video Arena. It is close to the top-ranked @GoogleDeepMind Veo 3.1 and @OpenAI Sora 2 Pro models. Grok-Imagine-Video offers: - Text-to-video and image-to-video capabilities - Native audio generation - Up to 15-seconds video duration Congrats to @xAI on this strong launch!

@xai doi.org/10.5281/zenodo…

Understanding requires imagining. Grok Imagine lets you bring what’s in your brain to life, and now it’s available via the world’s fastest, and most powerful video API: x.ai/news/grok-imag… Try it out and let your Imagination run wild.

@xai doi.org/10.5281/zenodo…

Today, we're excited to launch the Grok Voice Agent API, empowering developers to build voice agents that speak dozens of languages, call tools, and search realtime data. x.ai/news/grok-voic…

@elonmusk doi.org/10.5281/zenodo…

Not AI