Observation: Gemma 4 32B Sensory Layers Radically Changed from Base to Instruct
eliebakouch · x · 2026-07-08
@eliebakoucher shares a finding: transitioning from the base to the instruct checkpoint of Gemma 4 32B seemingly completely alters the model's early sensory layers (handling underlying parsing and grammar). He notes it makes logical sense but is fascinating to observe directly.
More from Research
- NUS builds a soft force sensor that drives actuators without electronics or power — CurieuxExplorer · 2026-07-27
- Chelsea Finn says robot RL is bottlenecked by physical rollout cost, not algorithms — ycombinator · 2026-07-27
- ICML 2026 oral paper replication scores stay middling after a stricter re-scoring — profjamesevans · 2026-07-27
- Long-running agents will need immutable event logs, this thread argues — sebpaquet · 2026-07-27
- Seed IQ navigates Doom II, prompting questions about benchmarks beyond ARC-AGI — Fit_Transition8824 · 2026-07-27
- Agentic Data Science in Practice: Agents Write Code but Answer Wrong Questions — hugobowne · 2026-07-27