{ "task": "sst2", "split": "validation", "model_path": "/home/ubuntu/master-research/not-trust-quivalent/models/sst2/distilbert-base-uncased", "metrics": { "accuracy": 0.8956422018348624 }, "loss": 0.2606525630024927, "total_examples": 872, "evaluation_time": 1.011415719985962 }