albert-large-v2-qqp / train_results.json
cdhinrichs's picture
Added a model and config
8f09873
{"eval_loss": 0.09762223809957504, "eval_accuracy": 0.9668376181131577, "eval_f1": 0.9555347548257284, "eval_combined_score": 0.9611861864694431, "eval_runtime": 2585.7201, "eval_samples_per_second": 140.714, "eval_steps_per_second": 17.589}