{"train/loss": 0.3117, "train/grad_norm": 0.5412354469299316, "train/learning_rate": 3.7096774193548386e-05, "train/epoch": 38.86, "train/global_step": 204, "_timestamp": 1711137734.6107543, "_runtime": 3941.20282125473, "_step": 77, "eval/loss": 0.5963708162307739, "eval/runtime": 9.5493, "eval/samples_per_second": 3.246, "eval/steps_per_second": 0.628} |