@neural_avb: Give them a bunch of money so they can do these scaling experiments upto 7B LLMs and beyond So much to learn from these…

X AI KOLs Timeline Papers

Summary

Zyphra shares their first work on continual learning for LLMs, studying whether models can learn forever from new data, and deriving a scaling law for the onset of plasticity loss in scaling experiments up to 7B parameters.

Give them a bunch of money so they can do these scaling experiments upto 7B LLMs and beyond So much to learn from these papers https://t.co/VhZvCJH0nk
Original Article
View Cached Full Text

Cached at: 06/26/26, 10:11 AM

Give them a bunch of money so they can do these scaling experiments upto 7B LLMs and beyond

So much to learn from these papers https://t.co/VhZvCJH0nk

Zyphra (@ZyphraAI): Zyphra is sharing our first work in continual learning where we study: Can LLMs learn forever from new data?

Many see continual learning as a path to AGI through recursive self-improvement (RSI).

The first obstacle is plasticity loss. We derive a scaling law for its onset 🧵

Similar Articles

@harshbhatt7585: https://x.com/harshbhatt7585/status/2063593933314113587

X AI KOLs Timeline

The author shares learnings from training a 160M parameter LLM from scratch, experimenting with architectures like multi-token prediction and hierarchical reasoning models. They emphasize the importance of fast iteration, simplifying ideas, and understanding why architectures work.

Scaling laws for neural language models

OpenAI Blog

Foundational empirical study demonstrating power-law scaling relationships between language model performance and model size, dataset size, and compute budget, with implications for optimal training allocation and sample efficiency.