Ronen Eldan
First author of TinyStories and co-author of Microsoft's Phi small language model reports
Ronen Eldan is first author, with Yuanzhi Li, of the 2023 TinyStories paper, written at Microsoft Research, which introduced a synthetic dataset of simple short stories and showed that language models with fewer than 10 million parameters can learn to write fluent, consistent text. Eldan's Hugging Face account publishes the TinyStories dataset under the CDLA-Sharing-1.0 license, along with the small models trained for the paper. Eldan is also a co-author of the phi-1, phi-1.5, Phi-3, and Phi-4 technical reports and a listed contributor to Phi-2. With Mark Russinovich, Eldan wrote "Who's Harry Potter?", a 2023 paper on approximate unlearning in language models, and Eldan is a co-author, listed with an OpenAI affiliation, of the 2025 paper "Early science acceleration experiments with GPT-5."124567891211
Work
Phi small language modelsOpen weights
Co-author of the phi-1 (2023), phi-1.5 (2023), Phi-3 (2024), and Phi-4 (2024) technical reports and a listed contributor on the Microsoft Research blog post announcing Phi-2. The phi-1.5 report states that the model was open-sourced, and its Hugging Face model card lists the MIT license.5678910
TinyStories (external site: huggingface.co)Open data
First author of the TinyStories paper (2023). The dataset card on Eldan's Hugging Face account lists the CDLA-Sharing-1.0 license, points to the paper, and names the models from the paper (TinyStories-1M through TinyStories-33M) published on the same account; the TinyStories-33M model card lists the MIT license.1234
Who's Harry Potter? Approximate Unlearning in LLMs (external site: arxiv.org)Published research
First author, with Mark Russinovich, of a 2023 paper proposing a technique for unlearning a subset of a language model's training data without retraining, demonstrated on Llama2-7b, with the fine-tuned model made available on Hugging Face.12
Other documented roles
Roles the sources document but that are not confirmed as current.
- Listed with an OpenAI affiliation on the November 2025 paper "Early science acceleration experiments with GPT-5", OpenAI11
- Listed with a Microsoft Research affiliation on the 2023 TinyStories and "Who's Harry Potter?" papers, Microsoft Research112
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.