What we call GPT's 'downstream behavior' is the behavior of simulacra; it is primarily through simulacra that GPT has potential to perform meaningful work.

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• GPT is behavior cloning, but it is the behavior of a universe that is cloned, not of a single demonstrator.claim0.822
  Broadening behavior cloning to universal simulation.
• GPT's ability to simulate text automata is the source of its most surprising and pivotal implications for paths to superintelligence.claim0.805
  Importance of recursive generation.
• GPT, insofar as it is inner-aligned, is a simulator which can simulate agentic and non-agentic simulacra.claim0.802
  Central thesis of the post.
• GPT does not generate rollouts during training, so there is no reason to expect that GPT will form preferences over the consequences of its output related to the text prediction objective.claim0.780
  Argues against instrumental convergence in GPT.
• GPT doesn't seem to care which agent it simulates, nor if the scene ends and the agent is effectively destroyed.claim0.776
  Illustrates the simulator-simulacra distinction.
• Park et al. 2023: Generative agents — interactive simulacra of human behaviorconcept0.774
  Paper on LLM-based simulacra of human behaviour; cited as ref 3
• Treating GPT as an unsupervised implementation of a supervised learner leads to systematic underestimation of capabilities.claim0.770
  Critique of the oracle/supervised frame.
• GPT is corrigible in a negative sense because the agent specification (prompt) is not fixed by the policy and the policy lacks direct training incentives to control its prompt.claim0.761
  GPT's corrigibility explained.