{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.1551724137931036, "global_step": 1500, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.07, "learning_rate": 9.760536398467434e-05, "loss": 1.0752, "step": 50 }, { "epoch": 0.14, "learning_rate": 9.521072796934867e-05, "loss": 0.7892, "step": 100 }, { "epoch": 0.14, "eval_loss": 0.6328039765357971, "eval_runtime": 15.8331, "eval_samples_per_second": 87.917, "eval_steps_per_second": 5.495, "step": 100 }, { "epoch": 0.22, "learning_rate": 9.281609195402298e-05, "loss": 0.6564, "step": 150 }, { "epoch": 0.29, "learning_rate": 9.042145593869731e-05, "loss": 0.6956, "step": 200 }, { "epoch": 0.29, "eval_loss": 0.5755433440208435, "eval_runtime": 15.7097, "eval_samples_per_second": 88.608, "eval_steps_per_second": 5.538, "step": 200 }, { "epoch": 0.36, "learning_rate": 8.802681992337166e-05, "loss": 0.6492, "step": 250 }, { "epoch": 0.43, "learning_rate": 8.563218390804599e-05, "loss": 0.6087, "step": 300 }, { "epoch": 0.43, "eval_loss": 0.547985315322876, "eval_runtime": 15.7303, "eval_samples_per_second": 88.492, "eval_steps_per_second": 5.531, "step": 300 }, { "epoch": 0.5, "learning_rate": 8.323754789272032e-05, "loss": 0.5863, "step": 350 }, { "epoch": 0.57, "learning_rate": 8.084291187739465e-05, "loss": 0.632, "step": 400 }, { "epoch": 0.57, "eval_loss": 0.5268416404724121, "eval_runtime": 15.7238, "eval_samples_per_second": 88.528, "eval_steps_per_second": 5.533, "step": 400 }, { "epoch": 0.65, "learning_rate": 7.844827586206897e-05, "loss": 0.6051, "step": 450 }, { "epoch": 0.72, "learning_rate": 7.60536398467433e-05, "loss": 0.5828, "step": 500 }, { "epoch": 0.72, "eval_loss": 0.5133854746818542, "eval_runtime": 15.7436, "eval_samples_per_second": 88.417, "eval_steps_per_second": 5.526, "step": 500 }, { "epoch": 0.79, "learning_rate": 7.365900383141763e-05, "loss": 0.5162, "step": 550 }, { "epoch": 0.86, "learning_rate": 7.126436781609196e-05, "loss": 0.535, "step": 600 }, { "epoch": 0.86, "eval_loss": 0.49968191981315613, "eval_runtime": 15.7272, "eval_samples_per_second": 88.509, "eval_steps_per_second": 5.532, "step": 600 }, { "epoch": 0.93, "learning_rate": 6.886973180076629e-05, "loss": 0.5922, "step": 650 }, { "epoch": 1.01, "learning_rate": 6.647509578544062e-05, "loss": 0.5509, "step": 700 }, { "epoch": 1.01, "eval_loss": 0.48899218440055847, "eval_runtime": 15.7366, "eval_samples_per_second": 88.456, "eval_steps_per_second": 5.529, "step": 700 }, { "epoch": 1.08, "learning_rate": 6.408045977011495e-05, "loss": 0.5053, "step": 750 }, { "epoch": 1.15, "learning_rate": 6.168582375478928e-05, "loss": 0.449, "step": 800 }, { "epoch": 1.15, "eval_loss": 0.48498210310935974, "eval_runtime": 15.7379, "eval_samples_per_second": 88.449, "eval_steps_per_second": 5.528, "step": 800 }, { "epoch": 1.22, "learning_rate": 5.92911877394636e-05, "loss": 0.4719, "step": 850 }, { "epoch": 1.29, "learning_rate": 5.689655172413794e-05, "loss": 0.4778, "step": 900 }, { "epoch": 1.29, "eval_loss": 0.48171550035476685, "eval_runtime": 15.7197, "eval_samples_per_second": 88.551, "eval_steps_per_second": 5.534, "step": 900 }, { "epoch": 1.36, "learning_rate": 5.450191570881227e-05, "loss": 0.4746, "step": 950 }, { "epoch": 1.44, "learning_rate": 5.2107279693486586e-05, "loss": 0.5006, "step": 1000 }, { "epoch": 1.44, "eval_loss": 0.4748086929321289, "eval_runtime": 15.7211, "eval_samples_per_second": 88.543, "eval_steps_per_second": 5.534, "step": 1000 }, { "epoch": 1.51, "learning_rate": 4.971264367816092e-05, "loss": 0.4911, "step": 1050 }, { "epoch": 1.58, "learning_rate": 4.7318007662835254e-05, "loss": 0.5066, "step": 1100 }, { "epoch": 1.58, "eval_loss": 0.4703415036201477, "eval_runtime": 15.7276, "eval_samples_per_second": 88.507, "eval_steps_per_second": 5.532, "step": 1100 }, { "epoch": 1.65, "learning_rate": 4.4923371647509585e-05, "loss": 0.5066, "step": 1150 }, { "epoch": 1.72, "learning_rate": 4.252873563218391e-05, "loss": 0.4803, "step": 1200 }, { "epoch": 1.72, "eval_loss": 0.466526061296463, "eval_runtime": 15.7166, "eval_samples_per_second": 88.569, "eval_steps_per_second": 5.536, "step": 1200 }, { "epoch": 1.8, "learning_rate": 4.013409961685824e-05, "loss": 0.4495, "step": 1250 }, { "epoch": 1.87, "learning_rate": 3.773946360153257e-05, "loss": 0.5075, "step": 1300 }, { "epoch": 1.87, "eval_loss": 0.46300598978996277, "eval_runtime": 15.7269, "eval_samples_per_second": 88.511, "eval_steps_per_second": 5.532, "step": 1300 }, { "epoch": 1.94, "learning_rate": 3.53448275862069e-05, "loss": 0.49, "step": 1350 }, { "epoch": 2.01, "learning_rate": 3.295019157088123e-05, "loss": 0.4738, "step": 1400 }, { "epoch": 2.01, "eval_loss": 0.46087321639060974, "eval_runtime": 15.7218, "eval_samples_per_second": 88.54, "eval_steps_per_second": 5.534, "step": 1400 }, { "epoch": 2.08, "learning_rate": 3.055555555555556e-05, "loss": 0.4432, "step": 1450 }, { "epoch": 2.16, "learning_rate": 2.8160919540229884e-05, "loss": 0.4286, "step": 1500 }, { "epoch": 2.16, "eval_loss": 0.4616571068763733, "eval_runtime": 15.7304, "eval_samples_per_second": 88.491, "eval_steps_per_second": 5.531, "step": 1500 } ], "max_steps": 2088, "num_train_epochs": 3, "total_flos": 2855841569095680.0, "trial_name": null, "trial_params": null }