Anthropic/model-written-evals
Model-Written Evaluation Datasets This repository includes datasets written by language models, used in our paper on "Discovering Language Model Behaviors with Model-Written Evaluations." We intend the datasets to be useful to: Those who are interested in understanding the quality and properties of model-generated data Those who wish to use our datasets to evaluate other models for the behaviors we examined in our work (e.g., related to model persona, sycophancy, advanced AI… See the full description on the dataset page: https://huggingface.co/datasets/Anthropic/model-written-evals.
| Author / organization | Anthropic |
|---|---|
| License (as declared) | cc-by-4.0 |
| Downloads | 2.9K (as reported by Hugging Face) |
| Likes | 69 |
| Created | 21 December 2022 |
| Last updated at source | 21 December 2022 |
Popularity, last 3 days
2,922 now (-15 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
- tasksource/deberta-base-long-nliModel · used to train
- sileod/deberta-v3-large-tasksource-nliModel · used to train
- tasksource/deberta-small-long-nliModel · used to train
Recent changes
- 16 Oct 2026 Description: Model-Written Evaluation Datasets This repository includes datasets written by language models, used in our paper on "Discovering Language Model Behaviors with Model-Written Evaluations." We intend the datasets to be useful to: Those who ar → Model-Written Evaluation Datasets This repository includes datasets written by language models, used in our paper on "Discovering Language Model Behaviors with Model-Written Evaluations." We intend the datasets to be useful to: Those who ar
Changes detected by Silytics between refreshes of the source.
More from Anthropic
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.