| { | |
| "best_metric": 0.9042719006538391, | |
| "best_model_checkpoint": "/gscratch/xlab/hallisky/rewriting/src/models/toxic/bart-base_2.5e-05_0_32_jigsaw_randmask/checkpoint-38000", | |
| "epoch": 10.552624271035823, | |
| "global_step": 38000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.4750000000000002e-05, | |
| "loss": 1.2139, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2.45e-05, | |
| "loss": 1.1274, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 0.9955187439918518, | |
| "eval_runtime": 183.7737, | |
| "eval_samples_per_second": 158.445, | |
| "eval_steps_per_second": 1.241, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 2.425e-05, | |
| "loss": 1.1115, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.0866, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 0.9784897565841675, | |
| "eval_runtime": 183.1935, | |
| "eval_samples_per_second": 158.947, | |
| "eval_steps_per_second": 1.245, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 2.375e-05, | |
| "loss": 1.0865, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 2.35e-05, | |
| "loss": 1.0748, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_loss": 0.9661561846733093, | |
| "eval_runtime": 183.3031, | |
| "eval_samples_per_second": 158.852, | |
| "eval_steps_per_second": 1.244, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 2.3250000000000003e-05, | |
| "loss": 1.073, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 1.0481, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_loss": 0.9609841704368591, | |
| "eval_runtime": 183.0018, | |
| "eval_samples_per_second": 159.113, | |
| "eval_steps_per_second": 1.246, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2.275e-05, | |
| "loss": 1.0424, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 2.25e-05, | |
| "loss": 1.0441, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_loss": 0.9524271488189697, | |
| "eval_runtime": 183.0159, | |
| "eval_samples_per_second": 159.101, | |
| "eval_steps_per_second": 1.246, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 2.2250000000000002e-05, | |
| "loss": 1.0383, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 1.0368, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_loss": 0.9480794668197632, | |
| "eval_runtime": 183.0254, | |
| "eval_samples_per_second": 159.093, | |
| "eval_steps_per_second": 1.246, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 2.175e-05, | |
| "loss": 1.0295, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 2.15e-05, | |
| "loss": 1.0294, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_loss": 0.9428065419197083, | |
| "eval_runtime": 183.0141, | |
| "eval_samples_per_second": 159.103, | |
| "eval_steps_per_second": 1.246, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2.125e-05, | |
| "loss": 1.0213, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 2.1e-05, | |
| "loss": 1.0148, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_loss": 0.9450138807296753, | |
| "eval_runtime": 182.9272, | |
| "eval_samples_per_second": 159.178, | |
| "eval_steps_per_second": 1.246, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 2.075e-05, | |
| "loss": 1.0141, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 2.05e-05, | |
| "loss": 1.0136, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 0.9394757151603699, | |
| "eval_runtime": 183.1908, | |
| "eval_samples_per_second": 158.949, | |
| "eval_steps_per_second": 1.245, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 2.025e-05, | |
| "loss": 1.0117, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0083, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "eval_loss": 0.9403573274612427, | |
| "eval_runtime": 182.9098, | |
| "eval_samples_per_second": 159.193, | |
| "eval_steps_per_second": 1.247, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.9750000000000002e-05, | |
| "loss": 1.0062, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 1.0045, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "eval_loss": 0.9311016201972961, | |
| "eval_runtime": 182.9614, | |
| "eval_samples_per_second": 159.148, | |
| "eval_steps_per_second": 1.246, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 1.925e-05, | |
| "loss": 0.9962, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.9913, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_loss": 0.9348850846290588, | |
| "eval_runtime": 182.9469, | |
| "eval_samples_per_second": 159.161, | |
| "eval_steps_per_second": 1.246, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.9914, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.9925, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "eval_loss": 0.9286701679229736, | |
| "eval_runtime": 182.7432, | |
| "eval_samples_per_second": 159.338, | |
| "eval_steps_per_second": 1.248, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 1.825e-05, | |
| "loss": 0.9957, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.9929, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "eval_loss": 0.9286787509918213, | |
| "eval_runtime": 183.092, | |
| "eval_samples_per_second": 159.035, | |
| "eval_steps_per_second": 1.245, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 1.775e-05, | |
| "loss": 0.9873, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.9825, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "eval_loss": 0.9284389615058899, | |
| "eval_runtime": 183.0692, | |
| "eval_samples_per_second": 159.055, | |
| "eval_steps_per_second": 1.245, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 1.725e-05, | |
| "loss": 0.9791, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.9761, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "eval_loss": 0.9231427311897278, | |
| "eval_runtime": 182.9694, | |
| "eval_samples_per_second": 159.141, | |
| "eval_steps_per_second": 1.246, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 1.675e-05, | |
| "loss": 0.9806, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.9788, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "eval_loss": 0.9270732998847961, | |
| "eval_runtime": 182.9924, | |
| "eval_samples_per_second": 159.121, | |
| "eval_steps_per_second": 1.246, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 1.6250000000000002e-05, | |
| "loss": 0.9771, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.9799, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.9208848476409912, | |
| "eval_runtime": 183.0859, | |
| "eval_samples_per_second": 159.04, | |
| "eval_steps_per_second": 1.245, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 1.575e-05, | |
| "loss": 0.9684, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.9655, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "eval_loss": 0.9188591837882996, | |
| "eval_runtime": 183.2317, | |
| "eval_samples_per_second": 158.914, | |
| "eval_steps_per_second": 1.244, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "learning_rate": 1.525e-05, | |
| "loss": 0.969, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.9678, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "eval_loss": 0.920549213886261, | |
| "eval_runtime": 182.9534, | |
| "eval_samples_per_second": 159.155, | |
| "eval_steps_per_second": 1.246, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 1.475e-05, | |
| "loss": 0.9672, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.9723, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "eval_loss": 0.9192747473716736, | |
| "eval_runtime": 182.9328, | |
| "eval_samples_per_second": 159.173, | |
| "eval_steps_per_second": 1.246, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 1.4249999999999999e-05, | |
| "loss": 0.9667, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.9603, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "eval_loss": 0.9202588200569153, | |
| "eval_runtime": 182.9947, | |
| "eval_samples_per_second": 159.119, | |
| "eval_steps_per_second": 1.246, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 1.3750000000000002e-05, | |
| "loss": 0.9589, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.9593, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "eval_loss": 0.9192423820495605, | |
| "eval_runtime": 183.1637, | |
| "eval_samples_per_second": 158.973, | |
| "eval_steps_per_second": 1.245, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 1.3250000000000002e-05, | |
| "loss": 0.961, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 6.66, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.9634, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 6.66, | |
| "eval_loss": 0.9148619771003723, | |
| "eval_runtime": 182.8693, | |
| "eval_samples_per_second": 159.228, | |
| "eval_steps_per_second": 1.247, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 1.2750000000000002e-05, | |
| "loss": 0.958, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.9577, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "eval_loss": 0.9157229065895081, | |
| "eval_runtime": 182.8974, | |
| "eval_samples_per_second": 159.204, | |
| "eval_steps_per_second": 1.247, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 1.225e-05, | |
| "loss": 0.9543, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.9505, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "eval_loss": 0.9146404266357422, | |
| "eval_runtime": 182.7838, | |
| "eval_samples_per_second": 159.303, | |
| "eval_steps_per_second": 1.247, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 1.175e-05, | |
| "loss": 0.9478, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.9546, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "eval_loss": 0.9135451912879944, | |
| "eval_runtime": 182.7429, | |
| "eval_samples_per_second": 159.339, | |
| "eval_steps_per_second": 1.248, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 1.125e-05, | |
| "loss": 0.9481, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.948, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "eval_loss": 0.9123603701591492, | |
| "eval_runtime": 183.1278, | |
| "eval_samples_per_second": 159.004, | |
| "eval_steps_per_second": 1.245, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 1.075e-05, | |
| "loss": 0.948, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.9461, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "eval_loss": 0.9139257073402405, | |
| "eval_runtime": 182.9341, | |
| "eval_samples_per_second": 159.172, | |
| "eval_steps_per_second": 1.246, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 8.19, | |
| "learning_rate": 1.025e-05, | |
| "loss": 0.9442, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9441, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "eval_loss": 0.9109994173049927, | |
| "eval_runtime": 451.1148, | |
| "eval_samples_per_second": 64.547, | |
| "eval_steps_per_second": 0.505, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 9.750000000000002e-06, | |
| "loss": 0.9431, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.9408, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "eval_loss": 0.911307692527771, | |
| "eval_runtime": 182.7894, | |
| "eval_samples_per_second": 159.298, | |
| "eval_steps_per_second": 1.247, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 9.25e-06, | |
| "loss": 0.9429, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 9e-06, | |
| "loss": 0.9391, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "eval_loss": 0.9080024361610413, | |
| "eval_runtime": 183.033, | |
| "eval_samples_per_second": 159.086, | |
| "eval_steps_per_second": 1.246, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 8.75e-06, | |
| "loss": 0.9427, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.942, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "eval_loss": 0.909125804901123, | |
| "eval_runtime": 183.0287, | |
| "eval_samples_per_second": 159.09, | |
| "eval_steps_per_second": 1.246, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 8.25e-06, | |
| "loss": 0.9371, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9338, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "eval_loss": 0.9082431793212891, | |
| "eval_runtime": 183.1001, | |
| "eval_samples_per_second": 159.028, | |
| "eval_steps_per_second": 1.245, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 7.75e-06, | |
| "loss": 0.9354, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.9337, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "eval_loss": 0.9084173440933228, | |
| "eval_runtime": 183.0616, | |
| "eval_samples_per_second": 159.061, | |
| "eval_steps_per_second": 1.245, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 7.25e-06, | |
| "loss": 0.9358, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.9355, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.907382071018219, | |
| "eval_runtime": 183.0579, | |
| "eval_samples_per_second": 159.064, | |
| "eval_steps_per_second": 1.246, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 6.750000000000001e-06, | |
| "loss": 0.9324, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.9323, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "eval_loss": 0.9063876271247864, | |
| "eval_runtime": 183.0015, | |
| "eval_samples_per_second": 159.113, | |
| "eval_steps_per_second": 1.246, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.9295, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 6e-06, | |
| "loss": 0.9321, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "eval_loss": 0.9042719006538391, | |
| "eval_runtime": 183.0185, | |
| "eval_samples_per_second": 159.099, | |
| "eval_steps_per_second": 1.246, | |
| "step": 38000 | |
| } | |
| ], | |
| "max_steps": 50000, | |
| "num_train_epochs": 14, | |
| "total_flos": 1.4479335456768e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |