| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 32952, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.878241811761841e-05, | |
| "loss": 0.8515, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.726044076464142e-05, | |
| "loss": 0.6069, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.573846341166444e-05, | |
| "loss": 0.5641, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.421648605868745e-05, | |
| "loss": 0.5586, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 9.269450870571045e-05, | |
| "loss": 0.5239, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 9.117253135273349e-05, | |
| "loss": 0.508, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 8.96505539997565e-05, | |
| "loss": 0.5077, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 8.81285766467795e-05, | |
| "loss": 0.5063, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 8.660659929380252e-05, | |
| "loss": 0.4774, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 8.508462194082553e-05, | |
| "loss": 0.4782, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 8.356264458784853e-05, | |
| "loss": 0.4743, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 8.204066723487156e-05, | |
| "loss": 0.4735, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 8.051868988189456e-05, | |
| "loss": 0.4609, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 7.899671252891757e-05, | |
| "loss": 0.448, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 7.747473517594059e-05, | |
| "loss": 0.4652, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 7.59527578229636e-05, | |
| "loss": 0.4387, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.44307804699866e-05, | |
| "loss": 0.4547, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 7.290880311700962e-05, | |
| "loss": 0.4391, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.138682576403264e-05, | |
| "loss": 0.4377, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 6.986484841105565e-05, | |
| "loss": 0.4415, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 6.834287105807866e-05, | |
| "loss": 0.4355, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7072, | |
| "eval_loss": 0.3910021185874939, | |
| "eval_runtime": 57.1361, | |
| "eval_samples_per_second": 85.533, | |
| "eval_steps_per_second": 5.356, | |
| "step": 10984 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.682089370510168e-05, | |
| "loss": 0.4319, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 6.529891635212468e-05, | |
| "loss": 0.3413, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 6.377693899914769e-05, | |
| "loss": 0.3365, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 6.225496164617071e-05, | |
| "loss": 0.3261, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 6.073298429319372e-05, | |
| "loss": 0.3228, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 5.921100694021673e-05, | |
| "loss": 0.3396, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 5.768902958723974e-05, | |
| "loss": 0.3431, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 5.616705223426275e-05, | |
| "loss": 0.3318, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 5.464507488128577e-05, | |
| "loss": 0.3363, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 5.3123097528308786e-05, | |
| "loss": 0.3303, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 5.16011201753318e-05, | |
| "loss": 0.3295, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 5.0079142822354806e-05, | |
| "loss": 0.3304, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 4.855716546937782e-05, | |
| "loss": 0.342, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 4.703518811640083e-05, | |
| "loss": 0.3274, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 4.551321076342384e-05, | |
| "loss": 0.3242, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 4.3991233410446854e-05, | |
| "loss": 0.3311, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 4.246925605746987e-05, | |
| "loss": 0.3424, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 4.094727870449288e-05, | |
| "loss": 0.3192, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 3.9425301351515894e-05, | |
| "loss": 0.3151, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.79033239985389e-05, | |
| "loss": 0.323, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 3.6381346645561915e-05, | |
| "loss": 0.334, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 3.485936929258493e-05, | |
| "loss": 0.3233, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7321, | |
| "eval_loss": 0.38333889842033386, | |
| "eval_runtime": 57.0615, | |
| "eval_samples_per_second": 85.644, | |
| "eval_steps_per_second": 5.363, | |
| "step": 21968 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.333739193960794e-05, | |
| "loss": 0.3239, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 3.1815414586630955e-05, | |
| "loss": 0.2248, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.0293437233653966e-05, | |
| "loss": 0.2207, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.8771459880676976e-05, | |
| "loss": 0.2394, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 2.724948252769999e-05, | |
| "loss": 0.2378, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 2.5727505174723e-05, | |
| "loss": 0.2318, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 2.4205527821746013e-05, | |
| "loss": 0.2357, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 2.2683550468769027e-05, | |
| "loss": 0.2325, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 2.1161573115792037e-05, | |
| "loss": 0.2309, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 1.963959576281505e-05, | |
| "loss": 0.2416, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1.8117618409838064e-05, | |
| "loss": 0.2287, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 1.6595641056861074e-05, | |
| "loss": 0.2385, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 1.5073663703884086e-05, | |
| "loss": 0.2336, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 1.35516863509071e-05, | |
| "loss": 0.2252, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 1.2029708997930111e-05, | |
| "loss": 0.2329, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 1.0507731644953125e-05, | |
| "loss": 0.2214, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 8.985754291976135e-06, | |
| "loss": 0.2335, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 7.463776938999148e-06, | |
| "loss": 0.219, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 5.94179958602216e-06, | |
| "loss": 0.2252, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 4.419822233045172e-06, | |
| "loss": 0.2381, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 2.8978448800681847e-06, | |
| "loss": 0.2359, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 1.375867527091197e-06, | |
| "loss": 0.229, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7424, | |
| "eval_loss": 0.44141659140586853, | |
| "eval_runtime": 57.0579, | |
| "eval_samples_per_second": 85.65, | |
| "eval_steps_per_second": 5.363, | |
| "step": 32952 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 32952, | |
| "total_flos": 1.6052008203583488e+17, | |
| "train_loss": 0.35377963603946533, | |
| "train_runtime": 9210.1652, | |
| "train_samples_per_second": 28.62, | |
| "train_steps_per_second": 3.578 | |
| } | |
| ], | |
| "max_steps": 32952, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.6052008203583488e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |