Skip to content
View Gramathon's full-sized avatar

Block or report Gramathon

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Gramathon/README.md

Graham Bonomo-Jackson

Independent researcher exploring epistemic uncertainty, belief formation, and policy change in deep learning systems — with a particular interest in whether a model's observed behavior actually tracks its intended objective over time, versus fitting surface patterns that look like it does.

  • ERATO — ERATO — tracks behavioural and policy-level change across post-training checkpoints (SFT/DPO/GRPO): fixes, regressions, durability, and whether preference-margin shifts (RPMS) transfer to held-out prompts or reflect pair-specific fitting. Representation-level analysis is planned future work.
  • bayzflow — converts PyTorch models into Pyro-backed Bayesian models for uncertainty-aware prediction.

By day I build dynamical-systems/manifold-trajectory analysis for operational data; the same instinct — trajectory over snapshot, and separating real signal from second-order noise.

Pinned Loading

  1. bayzflow bayzflow Public

    Bayesian deep learning for PyTorch: convert existing neural networks into Pyro-backed Bayesian models with SVI-trained posterior guides and uncertainty-aware prediction.

    Python

  2. ERATO ERATO Public

    Empirical framework for studying alignment trajectories, behavioural transfer, policy drift, and representation change across LLM adapters and checkpoints.

    Python