SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.12712 · 2023

Sparks of Artificial General Intelligence: Early experiments with GPT-4

Scott Lundberg, Hamid Palangi, Yuanzhi Li, Yin Lee, Harsha Nori, Sébastien Bubeck, Varun Chandrasekaran, Ece Kamar, Ronen Eldan, Eric Horvitz, Johannes Gehrke, Peter Lee, and 2 more

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 10 functions out of this paper's own repositories and ran 9 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
emrgnt-cmplxty/zero-shot-replication pwc_unofficial 6 of 6
microsoft/guidance pwc_unofficial 3 of 4
FunctionStatusWhere it lives
bytes_from Ran microsoft/guidance/guidance/_utils.py
code served (permissive licence) · get_code("033556b7164ece93")
is_equiv Ran emrgnt-cmplxty/zero-shot-replication/zero_shot_replication/core/math_helpers.py
code served (permissive licence) · get_code("19df601915034cc3")
last_boxed_only_string Ran emrgnt-cmplxty/zero-shot-replication/zero_shot_replication/core/math_helpers.py
code served (permissive licence) · get_code("a6e611557b1300a9")
load_existing_jsonl Ran emrgnt-cmplxty/zero-shot-replication/zero_shot_replication/core/utils.py
code served (permissive licence) · get_code("5879bc7b4f586cba")
load_file_or_raise Ran emrgnt-cmplxty/zero-shot-replication/zero_shot_replication/core/utils.py
code served (permissive licence) · get_code("b77de9c88ce55e2d")
load_template_class Ran microsoft/guidance/guidance/chat.py
code served (permissive licence) · get_code("c46a047369ef391b")
make_weak_bound_method Ran microsoft/guidance/guidance/_utils.py
code served (permissive licence) · get_code("a57daecf6c38000d")
prep_for_file_path Ran emrgnt-cmplxty/zero-shot-replication/zero_shot_replication/core/utils.py
code served (permissive licence) · get_code("6db67b8a56ad1b07")
remove_boxed Ran emrgnt-cmplxty/zero-shot-replication/zero_shot_replication/core/math_helpers.py
code served (permissive licence) · get_code("7ef45e32b9ecc76b")
strip_multiline_string_indents Not yet run microsoft/guidance/guidance/_utils.py
code served (permissive licence) · get_code("7229c1c8ee90348f")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Artificial intelligence (AI) researchers have been developing and refining large language models (LLMs) that exhibit remarkable capabilities across a variety of domains and tasks, challenging our understanding of learning and cognition. The latest model developed by OpenAI, , was trained using an unprecedented scale of compute and data. In this paper, we report on our investigation of an early version of GPT-4, when it was still in active development by OpenAI. We contend that (this early version of) GPT-4 is part of a new cohort of LLMs (along with ChatGPT and Google's PaLM for example) that exhibit more general intelligence than previous AI models. We discuss the rising capabilities and implications of these models. We demonstrate that, beyond its mastery of language, GPT-4 can solve novel and difficult tasks that span mathematics, coding, vision, medicine, law, psychology and more, without needing any special prompting. Moreover, in all of these tasks, GPT-4's performance is strikingly close to human-level performance, and often vastly surpasses prior models such as ChatGPT. Given the breadth and depth of GPT-4's capabilities, we believe that it could reasonably be viewed as an early (yet still incomplete) version of an artificial general intelligence (AGI) system. In our exploration of GPT-4, we put special emphasis on discovering its limitations, and we discuss the challenges ahead for advancing towards deeper and more comprehensive versions of AGI, including the possible need for pursuing a new paradigm that moves beyond next-word prediction. We conclude with reflections on societal influences of the recent technological leap and future research directions.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2303.12712")
get_code_for_paper("2303.12712")
have("2303.12712")

Connect an agent — have() is free.