Jared Kaplan

发表

Scaling Laws and Interpretability of Learning from Repeated Data

Tom B. Brown, Dario Amodei, T. Henighan, 2022, ArXiv.

Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models

Gerard de Melo, Andrew M. Dai, Christopher D. Manning, 2022, ArXiv.

Language Models (Mostly) Know What They Know

Tom B. Brown, Dario Amodei, T. Henighan, 2022, ArXiv.

In-context Learning and Induction Heads

Tom B. Brown, Dario Amodei, T. Henighan, 2022, ArXiv.

Predictability and Surprise in Large Generative Models

Tom B. Brown, Dario Amodei, T. Henighan, 2022, FAccT.

Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned

Tom B. Brown, Dario Amodei, T. Henighan, 2022, ArXiv.

Toy Models of Superposition

Dario Amodei, T. Henighan, Sam McCandlish, 2022, ArXiv.

Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned

Tom B. Brown, Dario Amodei, T. Henighan, 2022, ArXiv.

Measuring Progress on Scalable Oversight for Large Language Models

Tom B. Brown, Dario Amodei, Azalia Mirhoseini, 2022, ArXiv.

Discovering Language Model Behaviors with Model-Written Evaluations

Tom B. Brown, Dario Amodei, T. Henighan, 2022, ACL.

MAQA: A Multimodal QA Benchmark for Negation Anonymous Author(s)

Daniel M. Ziegler, Jacob Devlin, Prafulla Dhariwal, 2022 .

A General Language Assistant as a Laboratory for Alignment pdf

Dario Amodei, Tom Henighan, Tom B. Brown, 2021, ArXiv.

The Capacity for Moral Self-Correction in Large Language Models

Tom B. Brown, Dario Amodei, Azalia Mirhoseini, 2023, ArXiv.

Constitutional AI: Harmlessness from AI Feedback

Tom B. Brown, Dario Amodei, Azalia Mirhoseini, 2022, ArXiv.

Language Models (Mostly) Know What They Know

Tom B. Brown, Dario Amodei, T. Henighan, 2022, ArXiv.

Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback

Tom B. Brown, Dario Amodei, T. Henighan, 2022, ArXiv.

Evaluating Large Language Models Trained on Code pdf

Wojciech Zaremba, Lukasz Kaiser, Peter Welinder, 2021, ArXiv.

Active Learning is a Strong Baseline for Data Subset Selection

Tom B. Brown, Alec Radford, Prafulla Dhariwal, 2022 .

Towards Measuring the Representation of Subjective Global Opinions in Language Models

Sam McCandlish, Thomas Liao, Jack Clark, 2023, ArXiv.

Question Decomposition Improves the Faithfulness of Model-Generated Reasoning

Sam McCandlish, V. Chandrasekaran, Esin Durmus, 2023, ArXiv.

Towards Measuring the Representation of Subjective Global Opinions in Language Models

Sam McCandlish, Thomas Liao, Jack Clark, 2023, ArXiv.

Studying Large Language Model Generalization with Influence Functions

Benoit Steiner, Sam McCandlish, Esin Durmus, 2023, ArXiv.

Measuring Faithfulness in Chain-of-Thought Reasoning

Benoit Steiner, T. Henighan, Esin Durmus, 2023, ArXiv.

Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models

Ronan Le Bras, Stuart M. Shieber, Vinay Uday Prabhu, 2022, ArXiv.

Outlier Suppression+: Accurate quantization of large language models by equivalent and effective shifting and scaling

Michael W. Mahoney, Steven K. Esser, Prafulla Dhariwal, 2023, EMNLP.

Predictability and Surprise in Large Generative Models

Tom B. Brown, Dario Amodei, T. Henighan, 2022, FAccT.

Evaluating and Mitigating Discrimination in Language Model Decisions

Esin Durmus, Alex Tamkin, Deep Ganguli, 2023, ArXiv.

BLESS: Benchmarking Large Language Models on Sentence Simplification

R. P. Fishburne, Teven Le Scao, Leandro von Werra, 2023, EMNLP.

Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

Daniel M. Ziegler, D. Duvenaud, P. Christiano, 2024, ArXiv.

Zero-shot sampling of adversarial entities in biomedical question answering

Christopher A. Choquette-Choo, Daniel M. Ziegler, Alec Radford, 2024, 2402.10527.

Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents

Diego Perez Liebana, William H. Guss, Alec Radford, 2022 .

Cheater’s Bowl: Human vs. Computer Search Strategies for Open-Domain Question Answering

Daniel M. Ziegler, Prafulla Dhariwal, T. Henighan, 2023 .