Uncertainty & Technical Advisor | Director Zenta Group | PhD(c) de la Universidad de Chile para su Doctorado en Ciencias mención Computación

Joined August 2009
127 Photos and videos
Tomas Vera retweeted
Three days ago I left autoresearch tuning nanochat for ~2 days on depth=12 model. It found ~20 changes that improved the validation loss. I tested these changes yesterday and all of them were additive and transferred to larger (depth=24) models. Stacking up all of these changes, today I measured that the leaderboard's "Time to GPT-2" drops from 2.02 hours to 1.80 hours (~11% improvement), this will be the new leaderboard entry. So yes, these are real improvements and they make an actual difference. I am mildly surprised that my very first naive attempt already worked this well on top of what I thought was already a fairly manually well-tuned project. This is a first for me because I am very used to doing the iterative optimization of neural network training manually. You come up with ideas, you implement them, you check if they work (better validation loss), you come up with new ideas based on that, you read some papers for inspiration, etc etc. This is the bread and butter of what I do daily for 2 decades. Seeing the agent do this entire workflow end-to-end and all by itself as it worked through approx. 700 changes autonomously is wild. It really looked at the sequence of results of experiments and used that to plan the next ones. It's not novel, ground-breaking "research" (yet), but all the adjustments are "real", I didn't find them manually previously, and they stack up and actually improved nanochat. Among the bigger things e.g.: - It noticed an oversight that my parameterless QKnorm didn't have a scaler multiplier attached, so my attention was too diffuse. The agent found multipliers to sharpen it, pointing to future work. - It found that the Value Embeddings really like regularization and I wasn't applying any (oops). - It found that my banded attention was too conservative (i forgot to tune it). - It found that AdamW betas were all messed up. - It tuned the weight decay schedule. - It tuned the network initialization. This is on top of all the tuning I've already done over a good amount of time. The exact commit is here, from this "round 1" of autoresearch. I am going to kick off "round 2", and in parallel I am looking at how multiple agents can collaborate to unlock parallelism. github.com/karpathy/nanochat… All LLM frontier labs will do this. It's the final boss battle. It's a lot more complex at scale of course - you don't just have a single train. py file to tune. But doing it is "just engineering" and it's going to work. You spin up a swarm of agents, you have them collaborate to tune smaller models, you promote the most promising ideas to increasingly larger scales, and humans (optionally) contribute on the edges. And more generally, *any* metric you care about that is reasonably efficient to evaluate (or that has more efficient proxy metrics such as training a smaller network) can be autoresearched by an agent swarm. It's worth thinking about whether your problem falls into this bucket too.
960
2,125
19,524
3,660,431
Tomas Vera retweeted
The hottest new programming language is English
2,292
8,968
73,183
12,727,506
Tomas Vera retweeted
25 Apr 2025
📝 New article: "Understanding the Product Knowledge Acquisition Process in Multiprovider Software Evolution Scenarios: An Exploratory Study" by @AnelisVale, @vtomasv, Carlos Vasquez, @danielperovich, @jsimmond , Sergio F. Ochoa 👉 Get your copy at authors.elsevier.com/a/1kuLT…
1
3
2
227
Tomas Vera retweeted
31 Oct 2024
1
3
6
300
Tomas Vera retweeted
¡Gracias! @vtomasv para una no code / low code geek lo que compartiste hoy es oro 🫶
3
6
535
Tomas Vera retweeted
El Tribunal de Internet de Pekín ha sentenciado que un fabricante de software ha vulnerado los derechos de una persona tras emplear una herramienta basada en #IA para replicar su voz, sin su consentimiento: globaltimes.cn/page/202404/1…
11
6
1,229
Tomas Vera retweeted
Lo que está en disputa permanentemente es el cerebro. El #cerebro es el campo de batalla del futuro. Es por esto que nosotros consideramos que Worldcoin viola la norma de #neuroderechos que está en la Constitución chilena, porque no hay sistemas de consentimiento ni claridad del uso que le darán a estos datos obtenidos desde el iris de las personas. Detalles aquí ⬇️ meganoticias.cl/nacional/443…
8
11
33
3,009
Tomas Vera retweeted
Hello Evo 131K! 🧬 > 7B biological foundation model. > StripedHyena architecture. > Trained on 300B tokens, OpenGenome. > Long-context (131K), generates > 650K seq length. > Apache 2.0 license! 🔥 Checkpoints on the Hub! ;)
4
36
223
29,566
Tomas Vera retweeted
📚Conoce a los expert@s que participaron en el libro “En Defensa de los Neuroderechos” ⬇️ Descarga en defensaneuroderechos.org #neuroderechos #neurorights #neurodireitos
12
19
1,326
Tomas Vera retweeted
Parte una nueva edición de #CongresoFuturo2024 #CF2024 Una semana, en todas las regiones del país y en distintos espacios se conversará sobre el uso y el impacto de las tecnologías emergentes y la IA en nuestra sociedad
1
5
149
Tomas Vera retweeted
¡Descubre la revolución de la #IA en nuestros Talleres de #InteligenciaArtificial en #CF2024! 🌐 Explora teoría y práctica de la IA con instituciones, universidades e investigadores líderes.
2
10
17
1,593
Tomas Vera retweeted
8 Jan 2024
¡Felicitamos a la profesora del #dccuchile, @jsimmond quien fue recientemente ascendida a Profesora Asociada de la @uchile! 👏🎉 📌Noticia en: tinyurl.com/4jzhv68v
1
21
447
Tomas Vera retweeted
3 Jan 2024
👏Felicitamos al Prof. Aidan Hogan, quien asumió como director alterno del @IMFDChile. El académico señaló que una de sus primeras metas será aumentar la participación y visibilidad del IMFD en las universidades asociadas al Instituto. 📌Noticia en 👉 tinyurl.com/aidan-hogan-dire…
1
3
10
527
Tomas Vera retweeted
🔑#DFFull | ¿Se viene el "PIX" chileno? CMF da luz verde a fintech Shinkansen para ser la segunda cámara de pagos del país df.cl/mercados/banca-fintech…
3
10
34
10,658
Tomas Vera retweeted
14 Nov 2023
🎉👏Felicitamos a la profesora Jocelyn Simmonds, reconocida como Mejor Docente de Pregrado. Esta distinción fue otorgada en el marco de las actividades del 181° aniversario de la @uchile. Noticia en: tinyurl.com/2p8r9ran
5
38
1,139
Tomas Vera retweeted
¡ES HOY! 🎉 ¡NOCHES NERD! ⚡ Recuerda, hoy a las 19:30 horas, se llevará a cabo @NochesNerd en Taller 1, General Salvo 20, Providencia. 📍Referencia: Metro estación Salvador.
1
4
6
424
El Coordinador y profesor de nuestro Diploma de postítulo en Ingeniería de Software @danielperovich, participó el 17 de marzo en el Seminario Internacional: “Oportunidades y Desafíos del Metaverso” @congresofuturo 👏 Te invitamos ver su participación en👉youtube.com/watch?v=LJK2qO27…
3
3
686
Hoy en #PECRecomienda Tomás Vera, profesor #PECDCC destaca Learning Prompting, lo último en el mundo de código abierto, que enseña a promptear en apps de IA. Puedes hablar, hacer preguntas, debatir y compartir todo lo relacionado #IA. Learning Prompting👉 learnprompting.org/
1
3
93
Tomas Vera retweeted
¿Qué oportunidades y desafíos nos ofrece la #IA? 🤔 La IA se alza como la fuerza impulsora detrás de los Metaversos, creando un paisaje de oportunidades sin igual. Desde experiencias personalizadas hasta una interacción fluida y natural que trasciende lo imaginable 🤖🌐👀
1
3
5
643