| { | |
| "best_metric": 0.9069767441860465, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-75-fold1/checkpoint-88", | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6511627906976745, | |
| "eval_loss": 0.6476810574531555, | |
| "eval_runtime": 0.6318, | |
| "eval_samples_per_second": 68.06, | |
| "eval_steps_per_second": 3.166, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.6527612805366516, | |
| "eval_runtime": 0.6173, | |
| "eval_samples_per_second": 69.663, | |
| "eval_steps_per_second": 3.24, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.8095551133155823, | |
| "eval_runtime": 0.6043, | |
| "eval_samples_per_second": 71.154, | |
| "eval_steps_per_second": 3.309, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.7679464817047119, | |
| "eval_runtime": 0.6207, | |
| "eval_samples_per_second": 69.279, | |
| "eval_steps_per_second": 3.222, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.9488413333892822, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.5994, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.5481935739517212, | |
| "eval_runtime": 0.6133, | |
| "eval_samples_per_second": 70.117, | |
| "eval_steps_per_second": 3.261, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7441860465116279, | |
| "eval_loss": 0.4983576536178589, | |
| "eval_runtime": 0.6175, | |
| "eval_samples_per_second": 69.631, | |
| "eval_steps_per_second": 3.239, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.6156066656112671, | |
| "eval_runtime": 0.6217, | |
| "eval_samples_per_second": 69.167, | |
| "eval_steps_per_second": 3.217, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7674418604651163, | |
| "eval_loss": 0.530674397945404, | |
| "eval_runtime": 0.6286, | |
| "eval_samples_per_second": 68.405, | |
| "eval_steps_per_second": 3.182, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7674418604651163, | |
| "eval_loss": 0.40364784002304077, | |
| "eval_runtime": 0.6233, | |
| "eval_samples_per_second": 68.986, | |
| "eval_steps_per_second": 3.209, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.075669288635254, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3806, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.42405834794044495, | |
| "eval_runtime": 0.6323, | |
| "eval_samples_per_second": 68.001, | |
| "eval_steps_per_second": 3.163, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.4263027012348175, | |
| "eval_runtime": 0.6181, | |
| "eval_samples_per_second": 69.564, | |
| "eval_steps_per_second": 3.236, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7441860465116279, | |
| "eval_loss": 0.6778327226638794, | |
| "eval_runtime": 0.642, | |
| "eval_samples_per_second": 66.973, | |
| "eval_steps_per_second": 3.115, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7674418604651163, | |
| "eval_loss": 0.5884802937507629, | |
| "eval_runtime": 0.6192, | |
| "eval_samples_per_second": 69.444, | |
| "eval_steps_per_second": 3.23, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.6048120260238647, | |
| "eval_runtime": 0.6378, | |
| "eval_samples_per_second": 67.423, | |
| "eval_steps_per_second": 3.136, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.8426947593688965, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.273, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.5110020041465759, | |
| "eval_runtime": 0.6259, | |
| "eval_samples_per_second": 68.697, | |
| "eval_steps_per_second": 3.195, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7441860465116279, | |
| "eval_loss": 0.3793295919895172, | |
| "eval_runtime": 0.6397, | |
| "eval_samples_per_second": 67.215, | |
| "eval_steps_per_second": 3.126, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.3635205626487732, | |
| "eval_runtime": 0.63, | |
| "eval_samples_per_second": 68.259, | |
| "eval_steps_per_second": 3.175, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.3863283097743988, | |
| "eval_runtime": 0.6683, | |
| "eval_samples_per_second": 64.347, | |
| "eval_steps_per_second": 2.993, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.37879544496536255, | |
| "eval_runtime": 0.621, | |
| "eval_samples_per_second": 69.245, | |
| "eval_steps_per_second": 3.221, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 4.38327693939209, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.2388, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.33901217579841614, | |
| "eval_runtime": 0.6226, | |
| "eval_samples_per_second": 69.07, | |
| "eval_steps_per_second": 3.213, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.45933058857917786, | |
| "eval_runtime": 0.6161, | |
| "eval_samples_per_second": 69.798, | |
| "eval_steps_per_second": 3.246, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.344134122133255, | |
| "eval_runtime": 0.6226, | |
| "eval_samples_per_second": 69.067, | |
| "eval_steps_per_second": 3.212, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.5482814908027649, | |
| "eval_runtime": 0.6286, | |
| "eval_samples_per_second": 68.407, | |
| "eval_steps_per_second": 3.182, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.6398680210113525, | |
| "eval_runtime": 0.6347, | |
| "eval_samples_per_second": 67.748, | |
| "eval_steps_per_second": 3.151, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 1.9829111099243164, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.189, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3333226442337036, | |
| "eval_runtime": 0.6246, | |
| "eval_samples_per_second": 68.841, | |
| "eval_steps_per_second": 3.202, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3325919210910797, | |
| "eval_runtime": 0.6227, | |
| "eval_samples_per_second": 69.059, | |
| "eval_steps_per_second": 3.212, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.41504570841789246, | |
| "eval_runtime": 0.6243, | |
| "eval_samples_per_second": 68.877, | |
| "eval_steps_per_second": 3.204, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.34200432896614075, | |
| "eval_runtime": 0.6237, | |
| "eval_samples_per_second": 68.94, | |
| "eval_steps_per_second": 3.207, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.36485448479652405, | |
| "eval_runtime": 0.6179, | |
| "eval_samples_per_second": 69.595, | |
| "eval_steps_per_second": 3.237, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 2.870300531387329, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.1718, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.36506161093711853, | |
| "eval_runtime": 0.6332, | |
| "eval_samples_per_second": 67.905, | |
| "eval_steps_per_second": 3.158, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.46762949228286743, | |
| "eval_runtime": 0.6159, | |
| "eval_samples_per_second": 69.813, | |
| "eval_steps_per_second": 3.247, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3543034791946411, | |
| "eval_runtime": 0.6304, | |
| "eval_samples_per_second": 68.211, | |
| "eval_steps_per_second": 3.173, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.3209027349948883, | |
| "eval_runtime": 0.6291, | |
| "eval_samples_per_second": 68.348, | |
| "eval_steps_per_second": 3.179, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.34199726581573486, | |
| "eval_runtime": 0.6272, | |
| "eval_samples_per_second": 68.564, | |
| "eval_steps_per_second": 3.189, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 3.6730294227600098, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.1466, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.37998369336128235, | |
| "eval_runtime": 0.6314, | |
| "eval_samples_per_second": 68.108, | |
| "eval_steps_per_second": 3.168, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.6547455191612244, | |
| "eval_runtime": 0.6789, | |
| "eval_samples_per_second": 63.339, | |
| "eval_steps_per_second": 2.946, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.7674418604651163, | |
| "eval_loss": 0.9743425250053406, | |
| "eval_runtime": 0.6571, | |
| "eval_samples_per_second": 65.436, | |
| "eval_steps_per_second": 3.044, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.6677446961402893, | |
| "eval_runtime": 0.6757, | |
| "eval_samples_per_second": 63.638, | |
| "eval_steps_per_second": 2.96, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.5691245794296265, | |
| "eval_runtime": 0.6355, | |
| "eval_samples_per_second": 67.666, | |
| "eval_steps_per_second": 3.147, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 1.461414098739624, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.119, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4796452224254608, | |
| "eval_runtime": 0.6302, | |
| "eval_samples_per_second": 68.228, | |
| "eval_steps_per_second": 3.173, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.32433825731277466, | |
| "eval_runtime": 0.6208, | |
| "eval_samples_per_second": 69.265, | |
| "eval_steps_per_second": 3.222, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.29689204692840576, | |
| "eval_runtime": 0.6213, | |
| "eval_samples_per_second": 69.211, | |
| "eval_steps_per_second": 3.219, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.3637494146823883, | |
| "eval_runtime": 0.6187, | |
| "eval_samples_per_second": 69.505, | |
| "eval_steps_per_second": 3.233, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.30976200103759766, | |
| "eval_runtime": 0.6227, | |
| "eval_samples_per_second": 69.059, | |
| "eval_steps_per_second": 3.212, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 3.725754737854004, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1123, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3953772783279419, | |
| "eval_runtime": 0.6186, | |
| "eval_samples_per_second": 69.516, | |
| "eval_steps_per_second": 3.233, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3197100758552551, | |
| "eval_runtime": 0.631, | |
| "eval_samples_per_second": 68.144, | |
| "eval_steps_per_second": 3.169, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.318818062543869, | |
| "eval_runtime": 0.6195, | |
| "eval_samples_per_second": 69.405, | |
| "eval_steps_per_second": 3.228, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.30999791622161865, | |
| "eval_runtime": 0.6294, | |
| "eval_samples_per_second": 68.321, | |
| "eval_steps_per_second": 3.178, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3653392791748047, | |
| "eval_runtime": 0.6265, | |
| "eval_samples_per_second": 68.633, | |
| "eval_steps_per_second": 3.192, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 3.50057315826416, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1136, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.35266682505607605, | |
| "eval_runtime": 0.6288, | |
| "eval_samples_per_second": 68.382, | |
| "eval_steps_per_second": 3.181, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3152053952217102, | |
| "eval_runtime": 0.6272, | |
| "eval_samples_per_second": 68.564, | |
| "eval_steps_per_second": 3.189, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.32772013545036316, | |
| "eval_runtime": 0.6243, | |
| "eval_samples_per_second": 68.875, | |
| "eval_steps_per_second": 3.203, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.32207396626472473, | |
| "eval_runtime": 0.6252, | |
| "eval_samples_per_second": 68.781, | |
| "eval_steps_per_second": 3.199, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.34381797909736633, | |
| "eval_runtime": 0.6342, | |
| "eval_samples_per_second": 67.803, | |
| "eval_steps_per_second": 3.154, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 1.3971636295318604, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0858, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4682579040527344, | |
| "eval_runtime": 0.6268, | |
| "eval_samples_per_second": 68.598, | |
| "eval_steps_per_second": 3.191, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4511352479457855, | |
| "eval_runtime": 0.636, | |
| "eval_samples_per_second": 67.609, | |
| "eval_steps_per_second": 3.145, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3486385643482208, | |
| "eval_runtime": 0.636, | |
| "eval_samples_per_second": 67.605, | |
| "eval_steps_per_second": 3.144, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.35942766070365906, | |
| "eval_runtime": 0.6431, | |
| "eval_samples_per_second": 66.86, | |
| "eval_steps_per_second": 3.11, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.39140263199806213, | |
| "eval_runtime": 0.6182, | |
| "eval_samples_per_second": 69.554, | |
| "eval_steps_per_second": 3.235, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 3.632699728012085, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.084, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4256921410560608, | |
| "eval_runtime": 0.6209, | |
| "eval_samples_per_second": 69.256, | |
| "eval_steps_per_second": 3.221, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4505208134651184, | |
| "eval_runtime": 0.6326, | |
| "eval_samples_per_second": 67.972, | |
| "eval_steps_per_second": 3.162, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4038391411304474, | |
| "eval_runtime": 0.6342, | |
| "eval_samples_per_second": 67.797, | |
| "eval_steps_per_second": 3.153, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.3745191693305969, | |
| "eval_runtime": 0.6342, | |
| "eval_samples_per_second": 67.8, | |
| "eval_steps_per_second": 3.153, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.3773575723171234, | |
| "eval_runtime": 0.6457, | |
| "eval_samples_per_second": 66.599, | |
| "eval_steps_per_second": 3.098, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 1.8791605234146118, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.0938, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.37120649218559265, | |
| "eval_runtime": 0.6273, | |
| "eval_samples_per_second": 68.551, | |
| "eval_steps_per_second": 3.188, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.3736191987991333, | |
| "eval_runtime": 0.6296, | |
| "eval_samples_per_second": 68.293, | |
| "eval_steps_per_second": 3.176, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3839561641216278, | |
| "eval_runtime": 0.6195, | |
| "eval_samples_per_second": 69.415, | |
| "eval_steps_per_second": 3.229, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.39015671610832214, | |
| "eval_runtime": 0.623, | |
| "eval_samples_per_second": 69.018, | |
| "eval_steps_per_second": 3.21, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.41052982211112976, | |
| "eval_runtime": 0.6203, | |
| "eval_samples_per_second": 69.32, | |
| "eval_steps_per_second": 3.224, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 1.6623305082321167, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.055, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.44984933733940125, | |
| "eval_runtime": 0.6218, | |
| "eval_samples_per_second": 69.156, | |
| "eval_steps_per_second": 3.217, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4954254627227783, | |
| "eval_runtime": 0.6202, | |
| "eval_samples_per_second": 69.328, | |
| "eval_steps_per_second": 3.225, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.6397253274917603, | |
| "eval_runtime": 0.6291, | |
| "eval_samples_per_second": 68.351, | |
| "eval_steps_per_second": 3.179, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.6271260976791382, | |
| "eval_runtime": 0.6211, | |
| "eval_samples_per_second": 69.235, | |
| "eval_steps_per_second": 3.22, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4821247160434723, | |
| "eval_runtime": 0.6297, | |
| "eval_samples_per_second": 68.291, | |
| "eval_steps_per_second": 3.176, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 2.1010677814483643, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0755, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.36991921067237854, | |
| "eval_runtime": 0.6162, | |
| "eval_samples_per_second": 69.782, | |
| "eval_steps_per_second": 3.246, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.33030539751052856, | |
| "eval_runtime": 0.6223, | |
| "eval_samples_per_second": 69.094, | |
| "eval_steps_per_second": 3.214, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.328202486038208, | |
| "eval_runtime": 0.6165, | |
| "eval_samples_per_second": 69.745, | |
| "eval_steps_per_second": 3.244, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.31808754801750183, | |
| "eval_runtime": 0.6336, | |
| "eval_samples_per_second": 67.862, | |
| "eval_steps_per_second": 3.156, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3082694709300995, | |
| "eval_runtime": 0.6161, | |
| "eval_samples_per_second": 69.798, | |
| "eval_steps_per_second": 3.246, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.6037031412124634, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0603, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.3170202970504761, | |
| "eval_runtime": 0.615, | |
| "eval_samples_per_second": 69.918, | |
| "eval_steps_per_second": 3.252, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.33968400955200195, | |
| "eval_runtime": 0.6268, | |
| "eval_samples_per_second": 68.606, | |
| "eval_steps_per_second": 3.191, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.3537546992301941, | |
| "eval_runtime": 0.6325, | |
| "eval_samples_per_second": 67.982, | |
| "eval_steps_per_second": 3.162, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.3461407721042633, | |
| "eval_runtime": 0.6296, | |
| "eval_samples_per_second": 68.3, | |
| "eval_steps_per_second": 3.177, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.33368217945098877, | |
| "eval_runtime": 0.6261, | |
| "eval_samples_per_second": 68.682, | |
| "eval_steps_per_second": 3.195, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 2.1904349327087402, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0653, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.3330100178718567, | |
| "eval_runtime": 0.6375, | |
| "eval_samples_per_second": 67.448, | |
| "eval_steps_per_second": 3.137, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.345066636800766, | |
| "eval_runtime": 0.6175, | |
| "eval_samples_per_second": 69.64, | |
| "eval_steps_per_second": 3.239, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3611831068992615, | |
| "eval_runtime": 0.6169, | |
| "eval_samples_per_second": 69.709, | |
| "eval_steps_per_second": 3.242, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.382154643535614, | |
| "eval_runtime": 0.6328, | |
| "eval_samples_per_second": 67.955, | |
| "eval_steps_per_second": 3.161, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.38750723004341125, | |
| "eval_runtime": 0.6278, | |
| "eval_samples_per_second": 68.495, | |
| "eval_steps_per_second": 3.186, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 1.9912004470825195, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0571, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.38448840379714966, | |
| "eval_runtime": 0.6272, | |
| "eval_samples_per_second": 68.554, | |
| "eval_steps_per_second": 3.189, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3641882836818695, | |
| "eval_runtime": 0.625, | |
| "eval_samples_per_second": 68.795, | |
| "eval_steps_per_second": 3.2, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3528924584388733, | |
| "eval_runtime": 0.631, | |
| "eval_samples_per_second": 68.15, | |
| "eval_steps_per_second": 3.17, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.347101092338562, | |
| "eval_runtime": 0.6246, | |
| "eval_samples_per_second": 68.848, | |
| "eval_steps_per_second": 3.202, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.35397639870643616, | |
| "eval_runtime": 0.6263, | |
| "eval_samples_per_second": 68.662, | |
| "eval_steps_per_second": 3.194, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 1.0477629899978638, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.069, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.36087337136268616, | |
| "eval_runtime": 0.6222, | |
| "eval_samples_per_second": 69.109, | |
| "eval_steps_per_second": 3.214, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3609488904476166, | |
| "eval_runtime": 0.6221, | |
| "eval_samples_per_second": 69.119, | |
| "eval_steps_per_second": 3.215, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3633655905723572, | |
| "eval_runtime": 0.6181, | |
| "eval_samples_per_second": 69.562, | |
| "eval_steps_per_second": 3.235, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3627144694328308, | |
| "eval_runtime": 0.6291, | |
| "eval_samples_per_second": 68.35, | |
| "eval_steps_per_second": 3.179, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.36094027757644653, | |
| "eval_runtime": 0.6307, | |
| "eval_samples_per_second": 68.181, | |
| "eval_steps_per_second": 3.171, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 1.145646572113037, | |
| "learning_rate": 0.0, | |
| "loss": 0.0667, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.36037373542785645, | |
| "eval_runtime": 0.6299, | |
| "eval_samples_per_second": 68.267, | |
| "eval_steps_per_second": 3.175, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.867590382823424e+18, | |
| "train_loss": 0.15281517714262008, | |
| "train_runtime": 1123.7579, | |
| "train_samples_per_second": 21.446, | |
| "train_steps_per_second": 0.178 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.30976200103759766, | |
| "eval_runtime": 0.6605, | |
| "eval_samples_per_second": 65.101, | |
| "eval_steps_per_second": 3.028, | |
| "step": 200 | |
| } | |
| ], | |
| "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": 1.867590382823424e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |