| { | |
| "best_metric": 0.8638888888888889, | |
| "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-inaturalist/checkpoint-140", | |
| "epoch": 80.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.03333333333333333, | |
| "eval_loss": 3.200693368911743, | |
| "eval_runtime": 3.9107, | |
| "eval_samples_per_second": 184.111, | |
| "eval_steps_per_second": 1.534, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.044444444444444446, | |
| "eval_loss": 3.188920021057129, | |
| "eval_runtime": 3.5624, | |
| "eval_samples_per_second": 202.11, | |
| "eval_steps_per_second": 1.684, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_accuracy": 0.06388888888888888, | |
| "eval_loss": 3.1746575832366943, | |
| "eval_runtime": 3.358, | |
| "eval_samples_per_second": 214.412, | |
| "eval_steps_per_second": 1.787, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6863054633140564, | |
| "learning_rate": 2.5e-05, | |
| "loss": 3.1888, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.10972222222222222, | |
| "eval_loss": 3.144199848175049, | |
| "eval_runtime": 3.3224, | |
| "eval_samples_per_second": 216.708, | |
| "eval_steps_per_second": 1.806, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_accuracy": 0.14583333333333334, | |
| "eval_loss": 3.1183247566223145, | |
| "eval_runtime": 3.3659, | |
| "eval_samples_per_second": 213.913, | |
| "eval_steps_per_second": 1.783, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.21944444444444444, | |
| "eval_loss": 3.071033477783203, | |
| "eval_runtime": 3.1013, | |
| "eval_samples_per_second": 232.158, | |
| "eval_steps_per_second": 1.935, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "eval_accuracy": 0.30416666666666664, | |
| "eval_loss": 3.0330796241760254, | |
| "eval_runtime": 3.0337, | |
| "eval_samples_per_second": 237.333, | |
| "eval_steps_per_second": 1.978, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.7356524467468262, | |
| "learning_rate": 5e-05, | |
| "loss": 3.0673, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4388888888888889, | |
| "eval_loss": 2.96268367767334, | |
| "eval_runtime": 3.2316, | |
| "eval_samples_per_second": 222.8, | |
| "eval_steps_per_second": 1.857, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "eval_accuracy": 0.49444444444444446, | |
| "eval_loss": 2.9109385013580322, | |
| "eval_runtime": 3.1222, | |
| "eval_samples_per_second": 230.605, | |
| "eval_steps_per_second": 1.922, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5763888888888888, | |
| "eval_loss": 2.8359875679016113, | |
| "eval_runtime": 3.0808, | |
| "eval_samples_per_second": 233.708, | |
| "eval_steps_per_second": 1.948, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "eval_accuracy": 0.6055555555555555, | |
| "eval_loss": 2.780921459197998, | |
| "eval_runtime": 3.0798, | |
| "eval_samples_per_second": 233.781, | |
| "eval_steps_per_second": 1.948, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.76748126745224, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 2.8151, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6541666666666667, | |
| "eval_loss": 2.695770025253296, | |
| "eval_runtime": 3.3541, | |
| "eval_samples_per_second": 214.662, | |
| "eval_steps_per_second": 1.789, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "eval_accuracy": 0.6763888888888889, | |
| "eval_loss": 2.640125036239624, | |
| "eval_runtime": 3.1085, | |
| "eval_samples_per_second": 231.622, | |
| "eval_steps_per_second": 1.93, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6944444444444444, | |
| "eval_loss": 2.558335304260254, | |
| "eval_runtime": 3.0528, | |
| "eval_samples_per_second": 235.852, | |
| "eval_steps_per_second": 1.965, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 14.8, | |
| "eval_accuracy": 0.7083333333333334, | |
| "eval_loss": 2.503415822982788, | |
| "eval_runtime": 3.1925, | |
| "eval_samples_per_second": 225.525, | |
| "eval_steps_per_second": 1.879, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.8363860249519348, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 2.5143, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7347222222222223, | |
| "eval_loss": 2.4201643466949463, | |
| "eval_runtime": 3.0856, | |
| "eval_samples_per_second": 233.339, | |
| "eval_steps_per_second": 1.944, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "eval_accuracy": 0.7375, | |
| "eval_loss": 2.3662188053131104, | |
| "eval_runtime": 3.0872, | |
| "eval_samples_per_second": 233.218, | |
| "eval_steps_per_second": 1.943, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7444444444444445, | |
| "eval_loss": 2.2883973121643066, | |
| "eval_runtime": 3.1262, | |
| "eval_samples_per_second": 230.314, | |
| "eval_steps_per_second": 1.919, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 18.8, | |
| "eval_accuracy": 0.7569444444444444, | |
| "eval_loss": 2.237414598464966, | |
| "eval_runtime": 3.0834, | |
| "eval_samples_per_second": 233.51, | |
| "eval_steps_per_second": 1.946, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.8517465591430664, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 2.2236, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7777777777777778, | |
| "eval_loss": 2.1632001399993896, | |
| "eval_runtime": 3.0769, | |
| "eval_samples_per_second": 234.001, | |
| "eval_steps_per_second": 1.95, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 2.1174519062042236, | |
| "eval_runtime": 3.0592, | |
| "eval_samples_per_second": 235.356, | |
| "eval_steps_per_second": 1.961, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.7930555555555555, | |
| "eval_loss": 2.052760124206543, | |
| "eval_runtime": 3.21, | |
| "eval_samples_per_second": 224.301, | |
| "eval_steps_per_second": 1.869, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 22.8, | |
| "eval_accuracy": 0.7958333333333333, | |
| "eval_loss": 2.009880304336548, | |
| "eval_runtime": 3.0534, | |
| "eval_samples_per_second": 235.801, | |
| "eval_steps_per_second": 1.965, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.8587987422943115, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 1.9677, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8013888888888889, | |
| "eval_loss": 1.9488461017608643, | |
| "eval_runtime": 3.1017, | |
| "eval_samples_per_second": 232.129, | |
| "eval_steps_per_second": 1.934, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 24.8, | |
| "eval_accuracy": 0.8097222222222222, | |
| "eval_loss": 1.9112929105758667, | |
| "eval_runtime": 3.0366, | |
| "eval_samples_per_second": 237.104, | |
| "eval_steps_per_second": 1.976, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8138888888888889, | |
| "eval_loss": 1.8581663370132446, | |
| "eval_runtime": 3.0592, | |
| "eval_samples_per_second": 235.359, | |
| "eval_steps_per_second": 1.961, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 26.8, | |
| "eval_accuracy": 0.8138888888888889, | |
| "eval_loss": 1.8241873979568481, | |
| "eval_runtime": 3.2249, | |
| "eval_samples_per_second": 223.26, | |
| "eval_steps_per_second": 1.861, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 0.8794375061988831, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 1.7467, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8111111111111111, | |
| "eval_loss": 1.7740373611450195, | |
| "eval_runtime": 3.1535, | |
| "eval_samples_per_second": 228.319, | |
| "eval_steps_per_second": 1.903, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "eval_accuracy": 0.8055555555555556, | |
| "eval_loss": 1.7457906007766724, | |
| "eval_runtime": 3.0886, | |
| "eval_samples_per_second": 233.114, | |
| "eval_steps_per_second": 1.943, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8180555555555555, | |
| "eval_loss": 1.7013169527053833, | |
| "eval_runtime": 3.0478, | |
| "eval_samples_per_second": 236.234, | |
| "eval_steps_per_second": 1.969, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 30.8, | |
| "eval_accuracy": 0.8194444444444444, | |
| "eval_loss": 1.6714116334915161, | |
| "eval_runtime": 3.1668, | |
| "eval_samples_per_second": 227.361, | |
| "eval_steps_per_second": 1.895, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.9110932946205139, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.5765, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8263888888888888, | |
| "eval_loss": 1.631589412689209, | |
| "eval_runtime": 3.3174, | |
| "eval_samples_per_second": 217.038, | |
| "eval_steps_per_second": 1.809, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 32.8, | |
| "eval_accuracy": 0.8236111111111111, | |
| "eval_loss": 1.6082607507705688, | |
| "eval_runtime": 3.107, | |
| "eval_samples_per_second": 231.734, | |
| "eval_steps_per_second": 1.931, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8291666666666667, | |
| "eval_loss": 1.5738186836242676, | |
| "eval_runtime": 3.1316, | |
| "eval_samples_per_second": 229.912, | |
| "eval_steps_per_second": 1.916, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 34.8, | |
| "eval_accuracy": 0.8347222222222223, | |
| "eval_loss": 1.553105354309082, | |
| "eval_runtime": 3.2321, | |
| "eval_samples_per_second": 222.767, | |
| "eval_steps_per_second": 1.856, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 0.9191176891326904, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 1.4431, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8430555555555556, | |
| "eval_loss": 1.5228244066238403, | |
| "eval_runtime": 3.0649, | |
| "eval_samples_per_second": 234.921, | |
| "eval_steps_per_second": 1.958, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 36.8, | |
| "eval_accuracy": 0.8444444444444444, | |
| "eval_loss": 1.5046004056930542, | |
| "eval_runtime": 3.2047, | |
| "eval_samples_per_second": 224.669, | |
| "eval_steps_per_second": 1.872, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.8472222222222222, | |
| "eval_loss": 1.4780092239379883, | |
| "eval_runtime": 3.0681, | |
| "eval_samples_per_second": 234.673, | |
| "eval_steps_per_second": 1.956, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 38.8, | |
| "eval_accuracy": 0.8458333333333333, | |
| "eval_loss": 1.4607552289962769, | |
| "eval_runtime": 3.3287, | |
| "eval_samples_per_second": 216.303, | |
| "eval_steps_per_second": 1.803, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.896637499332428, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 1.3049, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8458333333333333, | |
| "eval_loss": 1.435728907585144, | |
| "eval_runtime": 3.1142, | |
| "eval_samples_per_second": 231.201, | |
| "eval_steps_per_second": 1.927, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 40.8, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 1.4187902212142944, | |
| "eval_runtime": 3.144, | |
| "eval_samples_per_second": 229.01, | |
| "eval_steps_per_second": 1.908, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8527777777777777, | |
| "eval_loss": 1.3949499130249023, | |
| "eval_runtime": 3.0808, | |
| "eval_samples_per_second": 233.709, | |
| "eval_steps_per_second": 1.948, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 42.8, | |
| "eval_accuracy": 0.8527777777777777, | |
| "eval_loss": 1.3807573318481445, | |
| "eval_runtime": 3.1783, | |
| "eval_samples_per_second": 226.537, | |
| "eval_steps_per_second": 1.888, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "grad_norm": 0.8933643102645874, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.2312, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8458333333333333, | |
| "eval_loss": 1.3636168241500854, | |
| "eval_runtime": 3.356, | |
| "eval_samples_per_second": 214.543, | |
| "eval_steps_per_second": 1.788, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 44.8, | |
| "eval_accuracy": 0.8486111111111111, | |
| "eval_loss": 1.3513400554656982, | |
| "eval_runtime": 3.2597, | |
| "eval_samples_per_second": 220.878, | |
| "eval_steps_per_second": 1.841, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.8527777777777777, | |
| "eval_loss": 1.3329037427902222, | |
| "eval_runtime": 3.0929, | |
| "eval_samples_per_second": 232.788, | |
| "eval_steps_per_second": 1.94, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 46.8, | |
| "eval_accuracy": 0.8527777777777777, | |
| "eval_loss": 1.319313883781433, | |
| "eval_runtime": 3.1007, | |
| "eval_samples_per_second": 232.203, | |
| "eval_steps_per_second": 1.935, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 0.9017526507377625, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 1.1368, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8527777777777777, | |
| "eval_loss": 1.3025320768356323, | |
| "eval_runtime": 3.1234, | |
| "eval_samples_per_second": 230.52, | |
| "eval_steps_per_second": 1.921, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 48.8, | |
| "eval_accuracy": 0.8541666666666666, | |
| "eval_loss": 1.2945308685302734, | |
| "eval_runtime": 3.0692, | |
| "eval_samples_per_second": 234.59, | |
| "eval_steps_per_second": 1.955, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8527777777777777, | |
| "eval_loss": 1.2819503545761108, | |
| "eval_runtime": 3.1162, | |
| "eval_samples_per_second": 231.052, | |
| "eval_steps_per_second": 1.925, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 50.8, | |
| "eval_accuracy": 0.8569444444444444, | |
| "eval_loss": 1.2704639434814453, | |
| "eval_runtime": 3.2771, | |
| "eval_samples_per_second": 219.706, | |
| "eval_steps_per_second": 1.831, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "grad_norm": 0.8930371999740601, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 1.0821, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.8583333333333333, | |
| "eval_loss": 1.2615665197372437, | |
| "eval_runtime": 3.1151, | |
| "eval_samples_per_second": 231.134, | |
| "eval_steps_per_second": 1.926, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 52.8, | |
| "eval_accuracy": 0.8555555555555555, | |
| "eval_loss": 1.2545220851898193, | |
| "eval_runtime": 3.0635, | |
| "eval_samples_per_second": 235.024, | |
| "eval_steps_per_second": 1.959, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.8541666666666666, | |
| "eval_loss": 1.2422840595245361, | |
| "eval_runtime": 3.0988, | |
| "eval_samples_per_second": 232.348, | |
| "eval_steps_per_second": 1.936, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 54.8, | |
| "eval_accuracy": 0.8597222222222223, | |
| "eval_loss": 1.233168125152588, | |
| "eval_runtime": 3.1982, | |
| "eval_samples_per_second": 225.127, | |
| "eval_steps_per_second": 1.876, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "grad_norm": 0.8568278551101685, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.0232, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.8638888888888889, | |
| "eval_loss": 1.2210274934768677, | |
| "eval_runtime": 3.355, | |
| "eval_samples_per_second": 214.604, | |
| "eval_steps_per_second": 1.788, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 56.8, | |
| "eval_accuracy": 0.8625, | |
| "eval_loss": 1.2160966396331787, | |
| "eval_runtime": 3.127, | |
| "eval_samples_per_second": 230.252, | |
| "eval_steps_per_second": 1.919, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.8569444444444444, | |
| "eval_loss": 1.209418535232544, | |
| "eval_runtime": 3.1685, | |
| "eval_samples_per_second": 227.239, | |
| "eval_steps_per_second": 1.894, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 58.8, | |
| "eval_accuracy": 0.8541666666666666, | |
| "eval_loss": 1.205717921257019, | |
| "eval_runtime": 3.1918, | |
| "eval_samples_per_second": 225.576, | |
| "eval_steps_per_second": 1.88, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 0.8556333780288696, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.9814, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 1.1972941160202026, | |
| "eval_runtime": 3.1515, | |
| "eval_samples_per_second": 228.465, | |
| "eval_steps_per_second": 1.904, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 60.8, | |
| "eval_accuracy": 0.8486111111111111, | |
| "eval_loss": 1.1918764114379883, | |
| "eval_runtime": 3.349, | |
| "eval_samples_per_second": 214.987, | |
| "eval_steps_per_second": 1.792, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.8625, | |
| "eval_loss": 1.1825212240219116, | |
| "eval_runtime": 3.1401, | |
| "eval_samples_per_second": 229.295, | |
| "eval_steps_per_second": 1.911, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 62.8, | |
| "eval_accuracy": 0.8597222222222223, | |
| "eval_loss": 1.179900050163269, | |
| "eval_runtime": 3.239, | |
| "eval_samples_per_second": 222.293, | |
| "eval_steps_per_second": 1.852, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "grad_norm": 0.827712893486023, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.9415, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.8597222222222223, | |
| "eval_loss": 1.1716293096542358, | |
| "eval_runtime": 3.1517, | |
| "eval_samples_per_second": 228.447, | |
| "eval_steps_per_second": 1.904, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 64.8, | |
| "eval_accuracy": 0.8625, | |
| "eval_loss": 1.166538119316101, | |
| "eval_runtime": 3.22, | |
| "eval_samples_per_second": 223.601, | |
| "eval_steps_per_second": 1.863, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.8638888888888889, | |
| "eval_loss": 1.1611206531524658, | |
| "eval_runtime": 3.323, | |
| "eval_samples_per_second": 216.67, | |
| "eval_steps_per_second": 1.806, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 66.8, | |
| "eval_accuracy": 0.8625, | |
| "eval_loss": 1.1600357294082642, | |
| "eval_runtime": 3.2253, | |
| "eval_samples_per_second": 223.233, | |
| "eval_steps_per_second": 1.86, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "grad_norm": 0.9616327285766602, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.9135, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.8638888888888889, | |
| "eval_loss": 1.1577341556549072, | |
| "eval_runtime": 3.1095, | |
| "eval_samples_per_second": 231.548, | |
| "eval_steps_per_second": 1.93, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 68.8, | |
| "eval_accuracy": 0.8638888888888889, | |
| "eval_loss": 1.1546900272369385, | |
| "eval_runtime": 3.2289, | |
| "eval_samples_per_second": 222.988, | |
| "eval_steps_per_second": 1.858, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.8638888888888889, | |
| "eval_loss": 1.1493217945098877, | |
| "eval_runtime": 3.103, | |
| "eval_samples_per_second": 232.033, | |
| "eval_steps_per_second": 1.934, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 70.8, | |
| "eval_accuracy": 0.8611111111111112, | |
| "eval_loss": 1.1463639736175537, | |
| "eval_runtime": 3.3191, | |
| "eval_samples_per_second": 216.926, | |
| "eval_steps_per_second": 1.808, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "grad_norm": 0.8919360637664795, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.8946, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.8555555555555555, | |
| "eval_loss": 1.1423206329345703, | |
| "eval_runtime": 3.0106, | |
| "eval_samples_per_second": 239.152, | |
| "eval_steps_per_second": 1.993, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 72.8, | |
| "eval_accuracy": 0.8611111111111112, | |
| "eval_loss": 1.1402053833007812, | |
| "eval_runtime": 3.4453, | |
| "eval_samples_per_second": 208.981, | |
| "eval_steps_per_second": 1.742, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.8583333333333333, | |
| "eval_loss": 1.1375410556793213, | |
| "eval_runtime": 3.1143, | |
| "eval_samples_per_second": 231.188, | |
| "eval_steps_per_second": 1.927, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 74.8, | |
| "eval_accuracy": 0.8597222222222223, | |
| "eval_loss": 1.1360384225845337, | |
| "eval_runtime": 3.2332, | |
| "eval_samples_per_second": 222.688, | |
| "eval_steps_per_second": 1.856, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "grad_norm": 0.9335024356842041, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.8866, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8625, | |
| "eval_loss": 1.134353756904602, | |
| "eval_runtime": 3.1331, | |
| "eval_samples_per_second": 229.808, | |
| "eval_steps_per_second": 1.915, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 76.8, | |
| "eval_accuracy": 0.8638888888888889, | |
| "eval_loss": 1.1333543062210083, | |
| "eval_runtime": 3.2199, | |
| "eval_samples_per_second": 223.611, | |
| "eval_steps_per_second": 1.863, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.8638888888888889, | |
| "eval_loss": 1.1323890686035156, | |
| "eval_runtime": 3.0815, | |
| "eval_samples_per_second": 233.653, | |
| "eval_steps_per_second": 1.947, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 78.8, | |
| "eval_accuracy": 0.8638888888888889, | |
| "eval_loss": 1.1320444345474243, | |
| "eval_runtime": 3.2638, | |
| "eval_samples_per_second": 220.604, | |
| "eval_steps_per_second": 1.838, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 0.8473469614982605, | |
| "learning_rate": 0.0, | |
| "loss": 0.8798, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.8638888888888889, | |
| "eval_loss": 1.1318904161453247, | |
| "eval_runtime": 3.2096, | |
| "eval_samples_per_second": 224.325, | |
| "eval_steps_per_second": 1.869, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "step": 200, | |
| "total_flos": 7.440697863438336e+18, | |
| "train_loss": 1.5909362745285034, | |
| "train_runtime": 1036.0381, | |
| "train_samples_per_second": 115.826, | |
| "train_steps_per_second": 0.193 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.440697863438336e+18, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |