Ensuring AI systems' chain-of-thought reasoning faithfully reflects their actual internal computations.
Organizations
3
Grants
3
Total Funding
$258K
Cluster: Interpretability
Parent Area: Interpretability
Grants3
| Name | Recipient | Amount | Funder | Date |
|---|---|---|---|---|
| Meridian — Avoiding Encoded Reasoning in LLMs | Meridian | $245K | Coefficient Giving | 2025-06 |
| Mechanistic Interpretability research for unfaithful chain-of-thought (1 month) | Iván Arcuschin Moreno | $11K | Manifund | 2024-11-19 |
| Unprompted Unfaithful Chain of Thought Dataset Project | Robert Krzyzanowski | $2K | Manifund | 2025-01-04 |
Funding by Funder
| Funder | Grants | Total Amount |
|---|---|---|
| Coefficient Giving | 1 | $245K |
| Manifund | 2 | $13K |
Tags
interpretabilitychain-of-thoughtfaithfulness