{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.0, "global_step": 22128, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.05, "learning_rate": 2.9322125813449026e-05, "loss": 2.223, "step": 500 }, { "epoch": 0.09, "learning_rate": 2.8644251626898048e-05, "loss": 1.5712, "step": 1000 }, { "epoch": 0.14, "learning_rate": 2.7966377440347073e-05, "loss": 1.4365, "step": 1500 }, { "epoch": 0.18, "learning_rate": 2.7288503253796095e-05, "loss": 1.3255, "step": 2000 }, { "epoch": 0.23, "learning_rate": 2.661062906724512e-05, "loss": 1.2879, "step": 2500 }, { "epoch": 0.27, "learning_rate": 2.5932754880694143e-05, "loss": 1.2262, "step": 3000 }, { "epoch": 0.32, "learning_rate": 2.5254880694143168e-05, "loss": 1.2299, "step": 3500 }, { "epoch": 0.36, "learning_rate": 2.457700650759219e-05, "loss": 1.1777, "step": 4000 }, { "epoch": 0.41, "learning_rate": 2.3899132321041215e-05, "loss": 1.1717, "step": 4500 }, { "epoch": 0.45, "learning_rate": 2.322125813449024e-05, "loss": 1.1411, "step": 5000 }, { "epoch": 0.5, "learning_rate": 2.2543383947939266e-05, "loss": 1.1099, "step": 5500 }, { "epoch": 0.54, "learning_rate": 2.1865509761388288e-05, "loss": 1.1054, "step": 6000 }, { "epoch": 0.59, "learning_rate": 2.1187635574837313e-05, "loss": 1.0961, "step": 6500 }, { "epoch": 0.63, "learning_rate": 2.0509761388286335e-05, "loss": 1.0753, "step": 7000 }, { "epoch": 0.68, "learning_rate": 1.983188720173536e-05, "loss": 1.0311, "step": 7500 }, { "epoch": 0.72, "learning_rate": 1.9154013015184382e-05, "loss": 1.0377, "step": 8000 }, { "epoch": 0.77, "learning_rate": 1.8476138828633408e-05, "loss": 1.0191, "step": 8500 }, { "epoch": 0.81, "learning_rate": 1.779826464208243e-05, "loss": 0.9882, "step": 9000 }, { "epoch": 0.86, "learning_rate": 1.7120390455531455e-05, "loss": 0.9928, "step": 9500 }, { "epoch": 0.9, "learning_rate": 1.6442516268980477e-05, "loss": 0.9701, "step": 10000 }, { "epoch": 0.95, "learning_rate": 1.5764642082429502e-05, "loss": 0.9901, "step": 10500 }, { "epoch": 0.99, "learning_rate": 1.5086767895878524e-05, "loss": 0.9647, "step": 11000 }, { "epoch": 1.04, "learning_rate": 1.4408893709327548e-05, "loss": 0.784, "step": 11500 }, { "epoch": 1.08, "learning_rate": 1.3731019522776571e-05, "loss": 0.7568, "step": 12000 }, { "epoch": 1.13, "learning_rate": 1.3053145336225597e-05, "loss": 0.7582, "step": 12500 }, { "epoch": 1.17, "learning_rate": 1.237527114967462e-05, "loss": 0.7732, "step": 13000 }, { "epoch": 1.22, "learning_rate": 1.1697396963123646e-05, "loss": 0.7331, "step": 13500 }, { "epoch": 1.27, "learning_rate": 1.101952277657267e-05, "loss": 0.7623, "step": 14000 }, { "epoch": 1.31, "learning_rate": 1.0341648590021693e-05, "loss": 0.7403, "step": 14500 }, { "epoch": 1.36, "learning_rate": 9.663774403470717e-06, "loss": 0.7096, "step": 15000 }, { "epoch": 1.4, "learning_rate": 8.98590021691974e-06, "loss": 0.7049, "step": 15500 }, { "epoch": 1.45, "learning_rate": 8.308026030368764e-06, "loss": 0.7097, "step": 16000 }, { "epoch": 1.49, "learning_rate": 7.630151843817788e-06, "loss": 0.7169, "step": 16500 }, { "epoch": 1.54, "learning_rate": 6.952277657266811e-06, "loss": 0.7172, "step": 17000 }, { "epoch": 1.58, "learning_rate": 6.274403470715835e-06, "loss": 0.7446, "step": 17500 }, { "epoch": 1.63, "learning_rate": 5.5965292841648585e-06, "loss": 0.6982, "step": 18000 }, { "epoch": 1.67, "learning_rate": 4.918655097613883e-06, "loss": 0.6878, "step": 18500 }, { "epoch": 1.72, "learning_rate": 4.240780911062907e-06, "loss": 0.6856, "step": 19000 }, { "epoch": 1.76, "learning_rate": 3.5629067245119307e-06, "loss": 0.6951, "step": 19500 }, { "epoch": 1.81, "learning_rate": 2.8850325379609543e-06, "loss": 0.6837, "step": 20000 }, { "epoch": 1.85, "learning_rate": 2.2071583514099784e-06, "loss": 0.6826, "step": 20500 }, { "epoch": 1.9, "learning_rate": 1.529284164859002e-06, "loss": 0.6854, "step": 21000 }, { "epoch": 1.94, "learning_rate": 8.514099783080261e-07, "loss": 0.6944, "step": 21500 }, { "epoch": 1.99, "learning_rate": 1.735357917570499e-07, "loss": 0.6932, "step": 22000 }, { "epoch": 2.0, "step": 22128, "total_flos": 6.662030063085158e+16, "train_loss": 0.9526093118189042, "train_runtime": 484536.8861, "train_samples_per_second": 0.548, "train_steps_per_second": 0.046 } ], "max_steps": 22128, "num_train_epochs": 2, "total_flos": 6.662030063085158e+16, "trial_name": null, "trial_params": null }