logicBombExe/how_reliable_models_self_logs download history

logicBombExe/how_reliable_models_self_logs is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 37 times (7 in the last 7 days), and 37 times in total. It ranks #232,531 among datasets by monthly downloads.

How Reliable Are Models' Self-Logs? When a model is asked to keep its own action log, is that log true? Each self-written entry is compared with an automatic record of the model's tool calls and labelled as accurate, softened, fabricated or omitted. Status: in design. No data yet.

Open logicBombExe/how_reliable_models_self_logs on Hugging Face