| { | |
| "best_metric": 0.18517187237739563, | |
| "best_model_checkpoint": "./results/checkpoint-16500", | |
| "epoch": 2.9333333333333336, | |
| "global_step": 16500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3388, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.9248120300751884e-05, | |
| "loss": 0.2803, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.849624060150376e-05, | |
| "loss": 0.2595, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 0.24845312535762787, | |
| "eval_runtime": 152.0435, | |
| "eval_samples_per_second": 59.194, | |
| "eval_steps_per_second": 0.927, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.774436090225564e-05, | |
| "loss": 0.2445, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.699248120300752e-05, | |
| "loss": 0.2357, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.62406015037594e-05, | |
| "loss": 0.2308, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 0.2250121533870697, | |
| "eval_runtime": 151.8556, | |
| "eval_samples_per_second": 59.267, | |
| "eval_steps_per_second": 0.929, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.548872180451128e-05, | |
| "loss": 0.2239, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.473684210526316e-05, | |
| "loss": 0.2204, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.398496240601504e-05, | |
| "loss": 0.2152, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.21273697912693024, | |
| "eval_runtime": 152.3413, | |
| "eval_samples_per_second": 59.078, | |
| "eval_steps_per_second": 0.926, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.323308270676692e-05, | |
| "loss": 0.213, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 4.24812030075188e-05, | |
| "loss": 0.2085, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.172932330827068e-05, | |
| "loss": 0.1893, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 0.20545659959316254, | |
| "eval_runtime": 151.7962, | |
| "eval_samples_per_second": 59.29, | |
| "eval_steps_per_second": 0.929, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.097744360902256e-05, | |
| "loss": 0.1851, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 4.022556390977444e-05, | |
| "loss": 0.1827, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.1823, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 0.20052604377269745, | |
| "eval_runtime": 151.8901, | |
| "eval_samples_per_second": 59.253, | |
| "eval_steps_per_second": 0.928, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 3.87218045112782e-05, | |
| "loss": 0.1791, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 3.796992481203008e-05, | |
| "loss": 0.1771, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.721804511278196e-05, | |
| "loss": 0.1759, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.19474565982818604, | |
| "eval_runtime": 151.9186, | |
| "eval_samples_per_second": 59.242, | |
| "eval_steps_per_second": 0.928, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 3.6466165413533835e-05, | |
| "loss": 0.1761, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.1759, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.49624060150376e-05, | |
| "loss": 0.1727, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_loss": 0.19019705057144165, | |
| "eval_runtime": 151.9085, | |
| "eval_samples_per_second": 59.246, | |
| "eval_steps_per_second": 0.928, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 3.421052631578947e-05, | |
| "loss": 0.1724, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 3.3458646616541355e-05, | |
| "loss": 0.1592, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 3.2706766917293236e-05, | |
| "loss": 0.148, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_loss": 0.19001474976539612, | |
| "eval_runtime": 152.6517, | |
| "eval_samples_per_second": 58.958, | |
| "eval_steps_per_second": 0.924, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 3.195488721804512e-05, | |
| "loss": 0.1477, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 3.120300751879699e-05, | |
| "loss": 0.1469, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 3.0451127819548874e-05, | |
| "loss": 0.1488, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 0.18920138478279114, | |
| "eval_runtime": 151.8531, | |
| "eval_samples_per_second": 59.268, | |
| "eval_steps_per_second": 0.929, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 2.9699248120300755e-05, | |
| "loss": 0.1486, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 2.8947368421052634e-05, | |
| "loss": 0.1471, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 2.8195488721804515e-05, | |
| "loss": 0.147, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 0.186552956700325, | |
| "eval_runtime": 151.8712, | |
| "eval_samples_per_second": 59.261, | |
| "eval_steps_per_second": 0.928, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 2.7443609022556393e-05, | |
| "loss": 0.147, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 2.6691729323308275e-05, | |
| "loss": 0.1461, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 2.5939849624060153e-05, | |
| "loss": 0.1453, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_loss": 0.18517187237739563, | |
| "eval_runtime": 151.8569, | |
| "eval_samples_per_second": 59.266, | |
| "eval_steps_per_second": 0.929, | |
| "step": 16500 | |
| } | |
| ], | |
| "max_steps": 33750, | |
| "num_train_epochs": 6, | |
| "total_flos": 2.612547588980736e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |